Paleontological Institute of the Russian Academy of Sciences

The H. Rausing Foundation

Valentin A. Krassilov

ANGIOSPERM ORIGINS:
MORPHOLOGICAL AND
ECOLOGICAL ASPECTS

1997
A SOFIA & MOSCOW-BASED PUBLISHING HOUSE IN ALL FIELDS OF LIFE & ENVIRONMENTAL SCIENCES

Acad. G. Bonchev Str., Bl. 6

1113 Sofia, Bulgaria
Tel./Fax +359-2-7133460
E-mail: pensoft@main.infotel.bg

© PENSOFT Publishers
ISBN 954-642-016-6
First published 1997
All rights reserved

Editors: Dr. S. Golovatch & Dr. L. Penev

Graphic design: M. Veleva & D. Velev
Cover design: Dr D. Dohrev
Photograph on the cover by V. A. Krassilov: preflower of Eoantha zherikhinii Krassilov, 1996

Printed by the Printing-office of "Prof Marin Drinov" Academic Publishing House

Sofia, May 1997
 Contents
Introduction 1
Making of the type 2
Angiosperms and Magnoliopsida 3
Classes as adaptive types A
Classes as phyla 5
Natural classification 7
Evolving characters 9
Gametophytic characters 10
Vegetative sporophytic characters 11
Reproductive sporophytic characters 12
Embryonic characters 13
Taxonomic rank intuition 13
Prehistory 15
Progymnosperms 16
Pteridospermidae (pteridosperms) 17
Cycadidae (cycadophytes) 21
Coniferidae (coniferids) 27
Proangiospermidae (proangiosperms) 31
Selected fossil links 38
Gymnosperm phylogeny 55
General tendencies and patterns 62
Early history 66
The Cretaceous 66
Cretaceous angiosperms 68
Chloranthoids (piperoids) 71
Lauroids 72
Platanoids 73
Nymphaeids 75
Ranunculids 76
Hamamelids 85
Rosoids 87
Myricoids 90
Graminoids 91
Phylogenetic relationships of Cretaceous angiosperms 95
Environments 98
Angiosperm cradle plant communities 100
Centres of origin 108
Rise of angiosperms as a biocoenotic process 109
Reproductive biology 112
A general scheme of ecological evolution 122

in
Phylogeny 125
Morphology 126
Character sharing 128
Character polarities 131
Structural homologies 135
Leaf 135
Flower 143
Stamen 145
Pollen 149
Gynoecium 153
Ovule and dispersal 157
Double fertilization and the embryo 161
Trend types 164
Proangiosperms as a morphological pool 165
Concurrent homologies 166
The gneto-bennettite line 167
Gnetophyte to monocot trends 167
Bennettite to monocot trends 168
Gnetophyte-bennettite features in other angiosperm groups 169
Caytonialean to ranunculid trends 170
Other proangiosperm contributions 172
Concluding remarks 174
Acknowledgements 176
Plates 177
References 225
Index 259
 Introduction
It has b e c o m e traditional to open a d i s c u s s i o n of a n g i o s p e r m origins by q u o t i n g
C h a r l e s D a r w i n w h o d e s c r i b e d the p r o b l e m as an a b o m i n a b l e mystery. U s u a l l y this is
f o l l o w e d by an assertion that, a l t h o u g h m u c h has been d o n e since D a r w i n w o r r i e d
h i m s e l f with this p a r t i c u l a r p r o b l e m , the m y s t e r y still r e m a i n s . But w h a t is so m y s t e ­
rious and so a b o m i n a b l e a b o u t a n g i o s p e r m origins that, after 140 y e a r s of research
and profuse p u b l i c a t i o n , there still is n o light in the end of the t u n n e l ?
Certainly, n o historical event is fully u n d e r s t o o d . All the o r i g i n s are subjects of at
least s o m e c o n t r o v e r s y . Yet this is n o r m a l , not a b o m i n a b l e . W h a t D a r w i n s e e m s to
have had in m i n d w a s that the a d v e n t of a n g i o s p e r m s d i s a g r e e d with the theory of
gradual e v o l u t i o n by m e a n s of natural selection. T h i s m i g h t h a v e b e e n the c a s e , so
w h a t ? Is it the theory or the a n g i o s p e r m s to be b l a m e d ? A s far as the theory is c o n ­
c e r n e d , the a b o m i n a b l e m y s t e r y m a y refer to either (1) the a n g i o s p e r m s h a v i n g n o
e v o l u t i o n a r y links with any other plant g r o u p a m o n g w h i c h their a n c e s t r y c o u l d p o ­
tentially be found or (2) the a p p e a r a n c e of a n g i o s p e r m s in the fossil record being
s u d d e n rather than g r a d u a l . T h e s e are the assertions directing r e l e v a n t research to
either a reconciliation of a n g i o s p e r m origins with the D a r w i n i a n theory of e v o l u t i o n
or a falsification of the latter.
In p r a c t i c e , h o w e v e r , like any p u z z l e , the m y s t e r y of a n g i o s p e r m origins defers
clarification, for p e o p l e are p s y c h o l o g i c a l l y u n p r e p a r e d to part with it. A few w o u l d
s a c r i f i c e the D a r w i n i a n t h e o r y of e v o l u t i o n to u n r a v e l l i n g the a c t u a l h i s t o r y of
a n g i o s p e r m s . E v e n f e w e r w o u l d g i v e u p a notion of a n g i o s p e r m s as a fairy separate
type of p l a n t s defined by a set of u n i q u e c h a r a c t e r s . N e w e v i d e n c e c a s t i n g any, if but
indirect, light on the origin of a n g i o s p e r m s d o e s not m a k e the professional p e o p l e
happy. T h e y tend to suspect s o m e t h i n g w r o n g a b o u t the e v i d e n c e as an e x c u s e for
ignoring it.
I h a v e already written at s o m e length about the factual basis of a n g i o s p e r m origins
(Krassilov, 1989). H e r e I bring in s o m e fresh e v i d e n c e . H o w e v e r , this b o o k deals also
with intellectual aspects of the p r o b l e m . Fortunately, several c o m p r e h e n s i v e r e v i e w s
p u b l i s h e d in the recent years (e.g. H u g h e s , 1994) allow m e to b y p a s s s o m e traditional
t h e m a t a of the story.
 Making of the type
Although at present the designations "flowering plants" and ^'angiosperms" are used as
synonyms, historically their references are somewhat different. In the pre-Linnaean and
Linnaean times, all terrestrial plants were considered as flowering, at least potentially.
" A n g i o s p e r m s " and " g y m n o s p e r m s " first appeared in the writings of "fructists", the XVII
century system-makers who relied primarily on fruit characters. Paul Hermann among them
has distinguished (in "Horti academici Lugduni Batavi Catalogus....", 1687, and other works)
between the angiosperms with a conspicuous pericarp and the g y m n o s p e r m s without it.
The latter included some members of Ranunculaceae, Umbelliferae, Compositae, etc. (see
Linnaei, 1751). In the XIX century, the angiospermy versus g y m n o s p e r m y concept ap­
proached the present-day understanding (De Candolle, 1827), whereas "flowering plants"
became synonymous to phanerogamous plants as opposed to cryptogamous, or spore, plants,
the latter having no flowers. Cones of conifers were considered as a special type of inflo­
rescence not unlike the catkins of alders or spikes of grasses. However, because angio-
spermous plants are far more diverse than g y m n o s p e r m o u s ones, the idea of a true flower­
ing plant became associated with angiospermy, while g y m n o s p e r m y seemed an aberrant
condition associated with aberrant flowers that perhaps were not true flowers after all.
As far as the notions both of flower and angiospermy are concerned, the formal dis­
tinctive characters of the angiosperms were essentially ecological, and the angiosperm
concept was primarily of a life-form type. Its subsequent transection into the realm of
morphology was not substantiated by explicit reformulation of the basic notions and, for
many, it passed unnoticed, leaving a lot of residual definition problems. Far ahead of a
plausible interpretation of coniferous seed-scale complexes as condensed bracteate shoots
(Florin, 1938-1945), plant morphologists set their minds on the problem of g y m n o s p e r m o u s
flower: whatever g y m n o s p e r m s might have was not a flower. T h u s the flower has b e c o m e
a charismatic feature of angiosperms, a possession that m a k e s them the flowering plants
in the m o d e m sense.
T h e subordinate and subsidiary characters were treated in much the same way. Defini­
tion of the flower as an organ consisting of stamens and/or carpels (with the perianth as a
dispensable component) was not meant to discriminate between angiosperms and g y m n o ­
sperms. Flower could be used discriminatingly only by asserting an exclusive right of the
angiosperms to stamens and carpels. A distressing thought of g y m n o s p e r m s also having
clusters of stamen-like and carpel-like structures, the latter in s o m e cases actually enclos­
ing the ovules, was resolutely rejected: whatever a g y m n o s p e r m might have as its pollen
sac-bearing or ovule-bearing organ was neither stamen nor carpel. A total lack of homol­
ogy was claimed for the vessels of angiosperms and those of gnetaleans or other non-
angiosperm plant groups (e.g. Takhtajan, 1966), the double fertilization in angiosperms
and the similar p h e n o m e n a in Ephedra and conifers, etc. Creation of the type has thus
b e c o m e c o m p l e t e d , with the e n s u i n g a b o m i n a b l e mystery of its origin further sustained
by seclusive morphological interpretations.

Angiosperms and Magnoliopsida

Presently there is a strong tendency a m o n g s y s t e m - m a k e r s to r e n a m e traditional
higher taxa, such as a n g i o s p e r m s , g y m n o s p e r m s , conifers, etc. on the basis of typifica-
tion with arbitrarily chosen genera serving as types not only for their families, which is
a conventional practice, but also for orders, classes, divisions, etc. T h u s , there would be
n o A n g i o p h y t a , A n g i o s p e r m a e , D i c o t y l e d o n e s , but M a g n o l i o p h y t a , M a g n o l i a t a e ,
M a g n o l i o p s i d a (and the s a m e with the P i n o p h y t a , Pinatae, P i n o p s i d a , etc.). Of course it
w o u l d be difficult to abandon habitual n a m e s used, with n o appreciable h a r m , for over
2 0 0 years. In fact, the modern s y s t e m - m a k e r s usually use M a g n o l i o p s i d a and the like
only once, as a title, then immediately switching to d i c o t y l e d o n s (e.g. Cronquist, 1981).
T h e y are, however, fully prepared to grapple with this kind of ambiguity for the sake of
stability and uniformity of the n o m e n c l a t u r e .
But is the n o m e n c l a t u r e of h i g h e r taxa g e t t i n g m o r e stable with the a d v e n t of
M a g n o l i o p s i d a ? Suppose Magnolia will be found s y n o n y m o u s to an earlier valid n a m e .
A n o t h e r conservation petition would b e c o m e m a n d a t o r y (and the conservation list is
too long already) lest all the n a m e s from the M a g n o l i a c e a e u p w a r d s should be c h a n g e d
automatically. Suppose, too, that a nonconformist would separate Magnolia and its closest
allies from the m a i n bulk of a n g i o s p e r m s . H e / s h e w o u l d then be obliged to introduce a
n e w n a m e , e.g. Rosopsida, for the latter, with all hierarchical steps appropriately re­
n a m e d . All this w o u l d hardly add to nomenclatural stability.
A s regards uniformity, there are t w o aspects of the p r o b l e m , o n e dealing with n o ­
menclature, the other with classification. For s o m e psychologically not quite o b v i o u s
reasons, those w h o a p p r o v e of M a g n o l i o p s i d a would also elevate the ranks of higher
taxa, e.g. allotting the orders the status of classes, turning classes into divisions, etc.
W h i l e higher taxa are often thought of as arbitrary, their rank might s e e m a matter of
practical c o n v e n i e n c e alone. If so, then it will be far m o r e c o n v e n i e n t to leave t h e m as
they have been for centuries rather than to e n d e a v o u r all the c u m b e r s o m e shifts and
emendations.

Classes as adaptive types

As there are repeated attempts at restructuring the system of higher plant taxa, there
should be s o m e reasons other than c o n v e n i e n c e . T h e s e m a y be uniformity and objectiv­
ity of a higher order classification achievable on the basis of s o m e general principles.
 G o e t h e (1790), b o r r o w i n g from Spinosa ( w h o , in turn, b o r r o w e d from Plato), sought
to represent, in an essentialist's spirit, the diversity of plants g r o w i n g on earth as varia­
tions of a single t h e m e , the Urpflanze. To him the Urpflanze was all leaves implying
that all the diverse plant structures arose as various leaf modifications. T h e s e ideas
inspired De C a n d o l l e (1827) and were used as the foundation of plant systematics, the
major divisions of the plant k i n g d o m being conceived of as consisting of species cre­
ated upon a c o m m o n Bauplan. T h e y survived the period of u n b o u n d e d d o m i n a n c e of
the existential evolutionary p a r a d i g m (e.g. Stebbins, 1974) and have b e c o m e , perhaps
u n a w a r e s , revived by modern cladistics to which a taxon is valid as far as it is defined
by uniquely derived characters of a c o m m o n ancestor stamped upon all its progeny.
Classification units are categories that are either introduced for c o n v e n i e n c e or are
contrived to reflect the structure of natural diversity. Natural biological classifications
convey our current understanding of biological diversity. A l t h o u g h " n a t u r a l " is often
used evaluatively as equivalent to rational or objective, it may signify neither of these.
All that " n a t u r a l " implies in the case of a biological classification, is that biological
diversity is structured in one way or another and that our intention is to arrange our
material accordingly. T h e r e can be several w a y s of doing so, hence m o r e than one
classification is natural, though not necessarily rational or objective.
T h e units of a natural classification, w h a t e v e r the basic criteria m a y be, are intended
to approach the natural entities as closely as our current u n d e r s t a n d i n g of the latter
permits. Guided by this general principle, we can focus then on different types of natu­
ral entities. T h e m o r e easily discernible a m o n g them are those bound by o n g o i n g inter­
actions, such as the reproductive interaction in populations or the trophic interactions in
biotic c o m m u n i t i e s . With the advent of the evolutionary concept, we have b e c o m e aware
of the entities bound not so by the current as by the historical relations. T h e s e are of t w o
major types, the phylogenetic, or genealogical, and the adaptive, or ecological. Inas­
much as phylogeny is directed by adaptation, they can, to a certain degree, coincide, but
their obligatorily one to o n e c o r r e s p o n d e n c e is hardly feasible.
Since an adaptive relatedness is usually m o r e obvious than a genealogical one, adap­
tive types have been recognized by taxonomists m o r e readily than phyla. F r o m the time
of Aristotle and Hipocrates, biological classifications have been built primarily upon
adaptive types. W h a t L i n n e called natural classification (in distinction of what he him­
self w o r k e d out as an admittedly artificial classification) was a system of the most clear-
cut adaptive types. He recognized their merits as higher taxa, but found it convenient to
c o m p l e m e n t them with a parallel system of conventional taxa based on a few arbitrary
characters. Old time s y s t e m - m a k e r s , notably A n d r e a C a e s a l p i n o (in D e plantis libri,
1583) and occasionally even modern taxonomists (Hutchinson, 1969) have used arbo­
real versus h e r b a c e o u s life-forms as d i s t i n c t i v e c h a r a c t e r s for c l a s s e s of a natural
botanical classification.
Generally, life-forms reflect coenotic roles of botanical species in plant c o m m u n i ­
ties - their ecological niches. M o s t botanical species are distinct life-forms, although
s o m e of t h e m are represented by more than one life-form. A s for the higher taxa, a one
to o n e c o r r e s p o n d e n c e to the life-form categories, or adaptive types, is even m o r e prob­
lematic. Nevertheless, grasses, for e x a m p l e , constitute a well-defined t a x o n o m i c order
as well as an adaptive type. Although the arboreal versus h e r b a c e o u s habit is no longer
used as the sole basis of a classification, either one or the other prevails in a n u m b e r of
supraspecies t a x o n o m i c categories. Therefore the characters traditionally used in higher
rank classifications are not entirely independent of the life-form.
Traditionally, reproductive characters are given priority over vegetative characters.
Because, in plants at least, reproductive adaptations are m u c h m o r e sophisticated than
trophic adaptations, the reproductive structures are, as a rule, m o r e diverse m o r p h o l o g i ­
cally than vegetative structures. O t h e r w i s e there will be no o b v i o u s a d v a n t a g e of one
set of characters over another. Both the foliar and the floral characters can be obviously
adaptive or (e.g. in the case of pinnate versus palmate leaf venation or axile versus
parietal placentation) apparently non-adaptive. In the latter case they are ascribed to
c o m m o n ancestry rather than a shared adaptation. As for a n g i o s p e r m s as a w h o l e , their
typical reproductive traits, flower, a n g i o s p e r m y and even d o u b l e fertilization, are cer­
tainly adaptive which m a k e s them the most inclusive adaptive type.

Classes as phyla
It s e e m e d that evolutionary approach would dispense both with archetypes and adap­
tive types, but in practice this did not happen. T h e c h a n g e s the evolutionists introduced
into the previously existing classifications were not so impressive. T h e adaptive type-
based natural units were, with a few exceptions, reinterpreted as phyla, and even many
units of artificial classifications survived on the assumption that L i n n e had intuitively
used n o n - a d a p t i v e characters evidencing nothing else but phylogenetic relations. It was
due to the latter coincidence that s o m e X I X century taxonomists thought of phylogenetic
classification as just another attempt at artificial classification. Only later did we c o m e
to think of phylogenetic classifications as the only natural ones.
But what could be expected of the so-called phylogenetic classifications entirely
lacking any factual e v i d e n c e of phylogeny as a historical p r o c e s s ? Paleobotanical evi­
d e n c e was m e a g r e and s e e m e d u n c o n v i n c i n g . It even s e e m e d to be of no use, for the
p r o b l e m of ancestry was assigned to the area of pure logic which did not tolerate any
intrusion of historical accidents. Logically, it s e e m e d quite plausible that characters of
no o b v i o u s adaptive value might have been passively inherited and could still be traced
back to a no longer surviving ancestor. That this imaginary ancestor c a m e to suspi­
ciously r e s e m b l e the Urpflanze was taken as another e v i d e n c e of our great predeces­
s o r ' s wonderful intuition.
 Only in the m i d - 1 9 6 0 ' s , with the d e v e l o p m e n t of c o m p u t e r t e c h n i q u e s , w a s the in­
tegrity of traditional t a x o n o m i c systems put to a serious test. It w a s reasoned that the
m o r e characters are shared by a pair of t a x o n o m i c units c o m p a r a t i v e with other pairs
the m o r e genes they share and therefore the closer their p h y l o g e n e t i c relatedness. N o
criteria for selecting the characters w e r e introduced in the c o u n t i n g p r o c e d u r e - the
m o r e the better, with a perspective of biological classifications turning into a branch of
applied m a t h e m a t i c s . H o w e v e r , the p r e m i s e s of the numerical classification s e e m e d
convincing to m a t h e m a t i c i a n s alone, for in m o d e r n biology the idea of o n e to o n e cor­
r e s p o n d e n c e of genetic and morphological similarities w a s no longer a c c e p t a b l e .
A n e w ideology w a s badly n e e d e d to m a k e any u s e of n u m e r i c a l t e c h n i q u e s . A n d
eventually it w a s found in a theretofore o b s c u r e e n t o m o l o g i c a l classification ( H e n n i g ,
1950) that b e c a m e k n o w n as cladistics, a n a m e first c o i n e d by its o p p o n e n t s but later
accepted by its p r o p o n e n t s as well. It w a s p r o c l a i m e d as a m e t h o d o l o g y of g e n u i n e
p h y l o g e n e t i c s y s t e m a t i c s g o i n g to replace a false m e t h o d o l o g y of traditional evolu­
tionary s y s t e m a t i c s . It w a s thus important to u n d e r s c o r e the f u n d a m e n t a l difference
between them.
In the late 1970's, " S y s t e m a t i c Z o o l o g y " sounded m o r e like a periodical in philoso­
phy of science rather than of biology. T h e founders of a n e w p a r a d i g m of biological
classification launched an attack on traditionalists using the philosophical a r g u m e n t a ­
tion b o r r o w e d from the then popular epistemologist Karl Popper. O n e familiar with
Popperian philosophy of science might w o n d e r in which way the latter m i g h t relate to
biological classification except for the assertion of a falsifiable h y p o t h e t i c o - d e d u c t i v e
s c h e m e against the alleged inductiveness of traditional systematics. Actually, falsifica­
tion is n o invention of Popperian philosophy but it rather a trivial element of prepositional
logic an elementary course of which w o u l d greatly benefit natural scientists. P o p p e r
(1972 and e l s e w h e r e ) used falsification as a criterion of scientific universal statements
that cannot be verified for all imaginable worlds. Insofar as biological classifications
deal with restricted rather than universal statements, the Popperian theory holds t h e m as
both falsifiable and verifiable.
L e a v i n g philosophy of science in peace, one finds a real difference in that traditional
evolutionary approach seeks to establish a classification on the basis of phylogeny,
while cladists d e d u c e phylogeny from classification, to a certain extent e m a n c i p a t i n g
the latter from historical evidence. W h i l e the evolutionists look u p to paleontology for
phylogenetic e v i d e n c e , the cladists find no use of paleontological data apart from cali­
brating cladogenesis (see Smith, 1994). A n o t h e r distinction is that evolutionists rely on
passively inherited non-adaptive characters the sharing of which ( s y m p l e s i o m o r p h i e s )
indicates c o m m o n ancestry, whereas the cladists prefer the shared derived characters
(synapomorphies) as indicating phylogenetic proximity. It has passed unnoticed that,
inasmuch as newly derived features are obviously m o r e often adaptive than residual
archaic characters, the cladistic classifications are u n a w a r e s a p p r o a c h i n g a system of
adaptive types.
 Natural classification
Despite the still widely held preferences, there are no purely scientific reasons for
giving phyla priority over adaptive types as the basis of a natural classification. Such
preferences m a y better be explained by non-scientific reasons. T h e public interest in
genealogies c a m e from the Old Testament tradition and prevailed in stratified societies.
Recently, the interest has shifted to adaptational aspects of natural and h u m a n histories
brought forth by the g r o w i n g environmental concern.
The structure of a natural system relates to both its functioning and its history. The struc­
tural characters that are not ascribable to the functioning can be ascribed to some historical
accidents. But the exclusive use of the latter in natural classification is hardly justifiable.
Julian H u x l e y (1942), w h o coined the terms " c l a d e s " (phyla) and " g r a d e s " ("hori­
z o n t a l " taxa sharing an adaptive zone), claimed that both were "facts of n a t u r e " deserv­
ing special places in a natural classification. A trivial argument in favour of a phylogenetic
approach is that all descendents of a single ancestor, h o w e v e r divergent ecologically,
share c o m m o n g e n e s . But actually they might lose most of the c o m m o n genes in the
course of divergence, while acquiring n e w ones by mutation and horizontal transfers.
On the other hand, the not so closely related but adaptively c o n v e r g e n t lineages might
increase their genetic similarity by parallel mutations and systemic transformations of
their g e n o m e s . T h u s the genetic similarity and the genetic relatedness are different things.
In all higher o r g a n i s m s , m o r p h o l o g y is primarily a product of epistatic interactions
and is a poor guide to the g e n o m i c structures that can be better studied at the molecular
level. But even such conservative c o m p o u n d s as c y t o c h r o m e s s e e m to h a v e preserved
not all of the structural evidence of their evolution - the latter might have been less
straightforward than it would seem. Stellaria, for e x a m p l e , might h a v e received its
c y t o c h r o m e s C by a horizontal transfer from a bacterial symbiont (Sullivan, 1996) rather
than by a mutation event at the base of the caryophyllid clade.
But even with a m o r e precise guide from structures to histories, translation of the
latter into a classification would still be a p r o b l e m . In the retrospect, not all historical
events are equally significant. F o r a natural classification to tell us s o m e t h i n g essential
about nature, w e have to c h o s e between the m o r e important and less important histori­
cal facts as c o n v e y e d by t a x o n o m i c characters.
Selection of diagnostic characters is a crucial p r o b l e m for any t a x o n o m i c m e t h o d o l ­
ogy. It m a y primarily d e p e n d on our intention to obtain a practically useful classifica­
tion system and/or to c o n v e y s o m e natural p h e n o m e n a , such as phylogeny, evolution or
the standing diversity. Objectivity of a classification is not a thing in itself; it relates to
the intentions of the s y s t e m - m a k e r s . T h e cladistic techniques are c o m m o n l y considered
as m o r e objective than the "intuitive" evolutionary approach. U s i n g parsimony, taxo­
n o m i c c o n g r u e n c e and c o n s e n s u s analyses, o n e can construe a total e v i d e n c e p h y l o g e n y
from a n u m b e r of fundamental trees incorporating alternative views on the uniquely
derived characters versus h o m o p l a s i e s , the latter's contribution being assessed by the
retention index (Farris, 1989; Vane-Write & al., 1992). H o w e v e r the use or abuse of the
cladistic techniques depends upon what we are striving at. T h e s e techniques provide for
rationality and reproducibility of a classification, yet this qualities do not m a k e it natural.
Generally, biological classifications use both natural criteria, pertaining to their ob­
jects, the classified organisms, and artificial criteria, pertaining to their subjects, the
s y s t e m - m a k e r s . A division of the plant k i n g d o m into p o i s o n o u s and h a r m l e s s species
reflects upon h u m a n users rather than plants, hence primarily artificial. P a r s i m o n y or
c o n s e n s u s may be attractive to h u m a n intellectuals, but both are virtually m e a n i n g l e s s
in respect to phylogenetic processes in nature unless we, like the X V I I century typologists,
believe in the parsimony of providence.
A m o d e r n classification strategy p r e s u p p o s e s an analysis of large sets of characters
for which various m e t h o d s of assessment are available in the literature, notably in rela­
tion to their evolutionary conservatism, unique derivation or the likelihood of a parallel
d e v e l o p m e n t . A single character approach, if there is such n o w a d a y s , w o u l d seem ar­
chaically Linnaean, and even L i n n e himself recognized it as artificial. H o w e v e r , all
modern classifications have the Linnaean artificial classification as their starting point,
and even the end-products may not be entirely different. U p to now, n o b o d y has at­
tempted to construe an entirely new multicharacter system of higher taxa without pre­
conceived a r r a n g e m e n t s . W h a t we actually do with large sets of characters appears
rather as an evaluation of the pre-existing single-character a r r a n g e m e n t s .
In other w o r d s , w e are not m a k i n g falsifiably p a r s i m o n i o u s classifications out of the
primaeval chaos. Instead w e start with a single (few) character classification trying to
assess and perhaps falsify it by adding n e w characters and/or w e i g h t i n g p r o c e d u r e s .
Incidentally, the concept of flowering plants has been put to an integrity test by intro­
ducing such initially even not suspected of character as the d o u b l e fertilization. T h e
flowering plants seem, by the data presently available, to c o m e through the test, thereby
attesting to the flower as a well chosen single diagnostic feature. Currently, the d o u b l e
fertilization m a y even seem a m o r e objectively definable single c h a r a c t e r of the angio­
sperms than the flower, although both are related to d e v e l o p m e n t a l c o n d e n s a t i o n (of
sporophytic and gametophytic developments respectively), their consistency in the taxon
thereby d e p e n d i n g on the consistency of the latter process.
T h e use of n u m e r o u s instead of a few characters is, in plants at least, scarcely advan­
tageous in itself, for the a m o u n t of a m b i g u o u s phylogenetic e v i d e n c e tends to increase
with the n u m b e r of characters. If we include in the analysis such characters as stipules
and foliar nectaries, for e x a m p l e , w e have to bear in m i n d that in s o m e , but not all,
angiosperm species foliar nectaries are modified stipules. O t h e r w i s e pluses or m i n u s e s
in the respective c o l u m n s of the character matrix would be m i s l e a d i n g . Therefore, what
matters is the quality of phylogenetic e v i d e n c e rather than numerical aspects.
In a numerical classification, it is important to differentiate b e t w e e n the functionally
related and functionally independent characters. If the characters are functionally re-
lated, as are, for e x a m p l e , the enclosed ovules and s i p h o n o g a m i c pollen tubes, they tend
to occur together in all m e m b e r s of the adaptive type w h a t e v e r their genetic affinities.
For the latter, sharing of the characters taken from different functional s y n d r o m e s may
be m o r e indicative. O n e can a s s u m e that a coinciding phylogenetic and adaptive rela­
tedness would give a taxon m a x i m a l integrity, while their discrepancy w o u l d result in a
more loosely b o u n d taxon.
T h e taxonomic practice has convincingly showed that a single well-chosen character
may surpass dozen characters picked up at random. Angiospermy is one such character.
Nominally, it relates to the condition of ovules, but actually it lies at the core of a func­
tional s y n d r o m e including stigmas, extraovular germination of pollen grains, recognition
substances in the pollen wall, female gametophyte reduction, sperm competition, etc.
Stability of a n g i o s p e r m s as a taxon is further warranted by the c o m b i n a t i o n of the
angiospermy syndrome with the flowering syndrome as a functional integration of repro­
ductive structures. However, these two syndromes may be functionally related, thus being
not entirely convincing as evidence of monophyly. But even if they denote a grade, then
this grade is so important as a fact of nature that we can hardly afford to ignore it.
M a y b e it is for the a b o v e m e n t i o n e d reasons that the a c h i e v e m e n t s of botanical
cladistics are not too spectacular. To its credit, cladistic analysis of a n g i o s p e r m origins
claims the demonstration of a close morphological proximity of gnetaleans to angiosperms
( D o n o g h u e & D o y l e , 1989). This is a robust result scarcely d e p e n d i n g on the technique.
Actually the s a m e w a s repeatedly stated in botanical text-books during the last century
at least. W h e t h e r to accept or reject gnetaleans as possible a n g i o s p e r m ancestors is a
different matter, and the cladistic analysis has no bearing on it.

Evolving characters
Historical relations are deducible, in the first place, from chronological relations and
morphological continuity. In the classifications conveying d y n a m i c processes, the evolv­
ing characters s h o w i n g certain trends in their historic d e v e l o p m e n t m a y s e e m prefer­
able to static characters. T h e palaeontological record suggests that all characters stop to
e v o l v e sooner or later, thus having a restricted period of utility for classifications of this
kind. A general trend in character evolution can be objectively defined by the relative
frequencies of a d v a n c e d to conservative states through time. For instance, through the
Paleozoic, the share of the pollen grains with distal apertures increased relative to those
with proximal apertures. It does not matter whether the trend was linear or fluctuating,
or w h e t h e r it occurred within a single lineage or in a n u m b e r of parallel lineages.
In plants, there are at least three sets of characters, i.e. gametophytic, sporophytic veg­
etative, and sporophytic reproductive, the evolutionary rates of which are but loosely cor­
related and which, therefore, could be used alternatively in evolutionary classifications.
 Gametophytic characters
T h e higher plants show four major options of the g a m e t o p h y t e to s p o r o p h y t e relation­
ships: (1) the g a m e t o p h y t e s are free-living, while the sporophytes are e p i g a m e t o p h y t a l
throughout their life, as in b r y o p h y t e s ; (2) both g a m e t o p h y t e s and s p o r o p h y t e s are free-
living at s o m e stage of their d e v e l o p m e n t , as in h o m o s p o r o u s p t e r i d o p h y t e s ; (3) the
female g a m e t o p h y t e s are e n d o s p o r o u s , nutritionally d e p e n d e n t on the s p o r e , as in
heterosporous pteridophytes; and (4) the female g a m e t o p h y t e s are e n d o s p o r a n g i a l , nu­
tritionally d e p e n d e n t on the sporophyte, as in seed plants. W h i l e the distinctions of (1)
from (2) and of (2) - (3) from (4) are traditionally allotted the highest t a x o n o m i c rank,
those between (2) and (3) ascend no higher than the ordinal rank, although from the
point of g a m e t o p h y t e evolution they s e e m no less important. This situation can be ex­
plained by the fact that heterospory, evolving independently in a n u m b e r of free-sporing
plant lineages (thus being a typical grade feature), scarcely affects major sporophytic
characters, while the endosporangial d e v e l o p m e n t , though p e r h a p s also of polyphyletic
origin, involves not only the sporangial, but also the extrasporangial sporophytic tissues
in the nutrition and protection of the d e v e l o p i n g s p o r o p h y t e and the e n s u i n g e m b r y o .
T h e care of the latter is consequently transferred from the g a m e t o p h y t e to the sporophyte.
This results in a radical restructuring of sporophyte morphology. Since the evolutionary
strategies (3) and (4) are entirely different, the seed plants might h a v e n e v e r had any
free-sporing heterosporous pteridophytes in their ancestry ( C h a l o n e r & Hemsley, 1991).
A phylogenetic succession of the four stages of g a m e t o p h y t e evolution might have
not been linear. Recent discoveries of the massive, fairly persistent g a m e t o p h y t e s in
early land plants ( R e m y & Remy, 1980) seem to be in favour of isomorphic alteration as
a primary m o d e , at the s a m e time suggesting a derivation of the m o d e s (1) and (3) from
(2), the m o r e so that s o m e early land plant g a m e t o p h y t e s , e.g. Sciadophyton (Remy &
al., 1980), r e s e m b l e thalloid hepatics. T h e early g a m e t o p h y t i c plants m i g h t have been
variably vascularized and even the recent Psilotum still retains a genetic potential for
developing vascular strands in the gametophytes of polyploid varieties (Bierhorst, 1953).
T h u s a neotenic g a m e t o p h y t e reduction is traceable from the rather m a s s i v e g a m e t o ­
phytes of early land plants through the ephemeral free-living g a m e t o p h y t e s of geologi­
cally y o u n g e r spore plants and the g y m n o s p e r m female g a m e t o p h y t e s confined to the
ovules but still developing archegonia, to the e m b r y o sacs of a n g i o s p e r m s lacking struc­
turally definable sex organs. In the course of these events, the g a m e t o p h y t e s b e c a m e
more and m o r e dissimilar to the sporophytes of the s a m e plants. At the genetic level,
this p h e n o m e n o n is related to the gene c o m p e n s a t i o n effect, i.e. the ability to c o m p e n ­
sate a haploid gene dose by the doubled transcriptional activity ( w e l l - k n o w n in respect
to the sex-linked genes in animals with heteromorphic sex c h r o m o s o m e s ) . It is c o n c e i v ­
able that, in the course of neotenic g a m e t o p h y t e evolution, the c o m p e n s a t o r y effect w a s
lost in an increasingly larger proportion of g a m e t o p h y t i c genes, resulting in a time-
series of progressively deviating g a m e t o p h y t e to s p o r o p h y t e m o r p h o l o g i e s .
 This is perhaps the only character trend sustained throughout the history of higher
plants, thus being of crucial importance to evolutionary classification. Traditionally, the
g a m e t o p h y t i c features are given a very high t a x o n o m i c rank, though p e r h a p s in a s o m e ­
what oblique way. T h u s , early land plants, or protopteridophytes, might h a v e differed
from the subsequent spore plants, or pteridophytes, primarily in their relatively more
strongly d e v e l o p e d g a m e t o p h y t e s . Seed plants differ from spore plants primarily in the
possession of ovules that are essentially the sporophytic food-storing structures for fe­
m a l e g a m e t o p h y t e s and the e m b r y o s . Furthermore, the d o u b l e fertilization, a major
distinction of a n g i o s p e r m s (but actually occurring in g n e t o p h y t e s as well, see b e l o w ) , is
related to their archegonia-less e m b r y o sacs.
T h u s , as far as g a m e t o p h y t i c characters are concerned, the boundary between spore
plants and seed plants is fairly sharp, perhaps the sharpest of all the higher plant divi­
s i o n s , with virtually n o t r a n s i t i o n a l f o r m s . O n the o t h e r h a n d , the g y m n o s p e r m /
a n g i o s p e r m b o u n d a r y is comparatively less significant and, therefore, is of a lower
t a x o n o m i c rank. This situation is further affirmed by sporophytic characters.

Vegetative sporophytic characters

Insofar as the stem and leaf characters are the most c o n s p i c u o u s in the higher plant
sporophytes, it is tempting to use them as the first order distinctions. H o w e v e r , this
strategy has been successful in the spore plant high-rank classification alone, distin­
guishing b e t w e e n the typically u n b r a n c h e d m a c r o p h y l l o u s ferns, the helically branched
microphyllous lycopods, and the branched verticillate microphyllous sphenopsids. These
g r o u p s are rooted in different early land plants s h o w i n g respective distinctions in their
incipient form. M a c r o p h y l l s and microphylls might have arisen from b r a n c h i n g sys­
tems of different - penultimate and ultimate - orders respectively (rather than having
had entirely different origins, as postulated by s o m e plant m o r p h o l o g i s t s , e.g. Takhtajan,
1970). N o n e t h e l e s s , the distinctions in b r a n c h i n g and leaf size, as well as a correlation
b e t w e e n the t w o , remained fairly consistent throughout the history of spore plants.
T h e m a c r o p h y l l o u s u n b r a n c h e d or sparsely branched and the m i c r o p h y l l o u s copi­
ously b r a n c h e d lines occur also in seed plants, i.e. the cycad and conifer lineages in
g y m n o s p e r m s or the m o n o c o t and dicot lineages in a n g i o s p e r m s . T h e distinction is,
however, far less consistent than in spore plants, precluding a higher rank t a x o n o m i c
division on the basis of vegetational characters alone.
In seed plants, the decussate leaf a r r a n g e m e n t is usually c o n s i d e r e d as advanced
relative to the helical o n e . However, phyllotaxis is s o m e t i m e s verticillate or decussate
in seedlings c o m i n g u p helical in the adult plants. For instance, t w o closely related
genera of taxodiaceous conifers, Metasequoia and Sequoia, differ primarily in their
phyllotaxis being decussate and helical respectively, with the former acquired by Meta­
sequoia perhaps as a p e d o m o r p h i c feature.
 T h e seed plant microphylls are primarily of a phyllodic origin ( c o r r e s p o n d i n g to
petioles of macrophylls), representing a trend in leaf evolution involving different line­
ages of both g y m n o s p e r m s and a n g i o s p e r m s . This trend is, however, poorly, if at all,
correlated with other vegetational trends.
It appears that, in seed plants, the genetic constraints of vegetational characters are
more flexible, resulting in their lesser t a x o n o m i c significance than in spore plants. T h e
demarcation is again more sharp between pteridophytes and g y m n o s p e r m s than be­
tween the latter and a n g i o s p e r m s .
M o s t m o n o c o t s are herbs or p a c h y c a u l trees with p o l y a r c h roots and p h y l l o d i c
leaves, but there is a broad o v e r l a p with the dicots of s i m i l a r life-forms. A m o n g g y m ­
n o s p e r m s , a similar distinction separates c y c a d s and their allies from conifers and
their allies, the distinctions also b e i n g rooted in the different life-forms. A s far as
vegetational c h a r a c t e r s are c o n c e r n e d , the d i c o t / m o n o c o t d i v i s i o n is of a l o w e r rank
than the a n g i o s p e r m / g y m n o s p e r m division and is of the s a m e rank as the c y c a d / c o n i -
fer division in g y m n o s p e r m s .

Reproductive sporophytic characters

As their n a m e s imply, g y m n o s p e r m s and a n g i o s p e r m s differ in sporophytic repro­
ductive features related to their different ovule protection and pollination m o d e s . T h e
distinctions stood fairly well until the discovery of various extinct g r o u p s , such as the
Paleozoic pteridosperms or the M e s o z o i c caytonialeans and c z e k a n o w s k i a l e a n s with
enclosed ovules and, in the latter group, with stigmatic e x o s t o m e s (Krassilov, 1969),
making the reproductive demarcation of g y m n o s p e r m s and a n g i o s p e r m s far less dis­
tinct and much of the nature of their g a m e t o p h y t i c d e m a r c a t i o n discussed a b o v e .
T h e most c o n s p i c u o u s structure used in the seed plant classification is aggregation
of reproductive units, including both strobili and flowers. In g y m n o s p e r m s , the ovules
are borne ( I ) in branching systems, or stachyospermic ( L a m , 1950), as in early seed
plants; (2) on solitary laminar ovuliphores, or p h y l l o s p e r m i c , as in p t e r i d o s p e r m s ; (3) in
coaxially aggregated ovuliphores, or strobili, as in cycads; and (4) in coaxial aggregations
of strobili, or doubly-strobilated, as in conifers. Planation of o v u l i p h o r e s as well as
strobilation and d o u b l e strobilation are the trends sustained by parallel lineages of
g y m n o s p e r m o u s plants, with occasional reverse trends, such as destrobilation in Cycas,
Ginkgo or Taxus.
In a n g i o s p e r m s , the aggregation m o d e s have also been used as an important distinc­
tion between the ranalean and amentiferous taxa and is still evident in the separation of
m a g n o l i i d s and h a m a m e l i d s . F u r t h e r m o r e , the axial versus parietal placentation in
a n g i o s p e r m s is analogous to the phyllospermic and s t a c h y o s p e r m i c m o d e s in g y m n o ­
sperms, though no phylogenetic continuity is implied.
 Embryonic characters
The highest-rank division of a n g i o s p e r m s is based on a cotyledon condition that is
of no utility in the systematics of g y m n o s p e r m s . This situation m a y reflect a correlation
between directional evolution and t a x o n o m i c significance. T h e r e is no definite cotyle­
don trend in g y m n o s p e r m s , whereas in a n g i o s p e r m s a certain derivative a s y m m e t r y
may exist between the m o n o c o t y l e d o n o u s and dicotyledonous conditions. A single coty­
ledon m a y be t e r m i n a l , as in Lemna, or p s e u d o t e r m i n a l , as in t h e C y p e r a c e a e ,
P o t a m o g e t o n a c e a e , H y d r o c h a r i t a c e a e and, a m o n g d i c o t y l e d o n s , Claytonia (Portula-
caceae). F u r t h e r m o r e , monocotyly may result from reduction, as in p a l m s (Guignard,
1984), or from fusion, as in the " d i c o t y l e d o n o u s " N y m p h a e a c e a e . T h e multiple ways to
monocotyly suggest a trend, perhaps sustained by parallel lineages of a n g i o s p e r m s .
However, s o m e ' ' m o n o c o t y l e d o n o u s " e m b r y o s seem to have no cotyledon h o m o -
logues whatsoever, while for the graminoid e m b r y o there is no consensual s c h e m e of
homology at all. A consistent application of e m b r y o n i c characters would give the grasses
a t a x o n o m i c rank at the level of dicotyledons and m o n o c o t y l e d o n s . But, until a definite
trend is discerned, the e m b r y o n i c characters will be used pragmatically and, therefore,
inconsistently at different levels of the t a x o n o m i c hierarchy (as, for instance, the sup­
posed " h e l o b i a l " e n d o s p e r m d e v e l o p m e n t in the L e m n a c e a e , see M a h e s h w a r i , 1958).

Taxonomic rank intuition

T h e a b o v e discussion may suggest that we have to investigate t a x o n o m i c intuition
before a b a n d o n i n g it. Intuition relates to the sense of ambiguity that appears rather late
in the h u m a n cognitive d e v e l o p m e n t . Until this, cognition is irreversible and straight­
forward, tending to substitute all irregular forms, such as went, slept, feet, teeth, etc., by
the regular, although intuitively u n g r a m m a t i c a l , forms, such as goed, sleeped, foots,
tooths, etc. (Cairns & Cairns, 1976) T h e s e childish d i s a m b i g u a t i n g attempts resemble
the o n g o i n g correction of "intuitive" biological classifications on the principle of objec­
tivity and phylogenetic parsimony.
T h e e x a m p l e s discussed a b o v e seem to suggest that the higher t a x o n o m i c ranks are
not arbitrary but, instead, they reflect our perception of directedness, continuity and
consistency. A character acquires a higher t a x o n o m i c rank when there is a definite trend
in its evolution which is not totally counterbalanced by the reverse trends. Conceivably,
the m o r e lineages partake in the trend the more consistent the character and the higher
its t a x o n o m i c status. Incidentally, phyllotaxis illustrates a character the directional evo­
lution of which was exhausted rather early, at the spore plant/seed plant transition, hence­
forth of a minor t a x o n o m i c significance. In contrast, m o n o c o t y l y reflects a trend that
appeared rather late, at the g y m n o s p e r m / a n g i o s p e r m transition, henceforth acquiring a
t a x o n o m i c significance the e m b r y o n i c features never had before.
In addition, the evolution of a high-rank character has to be consequential for other
characters inducing t h e m to evolve. Heterospory in free-sporing plants is an e x a m p l e of
a character s h o w i n g a definite trend of only little c o n s e q u e n c e for other characters and
thereby of a rather m o d e s t t a x o n o m i c rank (as c o m p a r e d with the " c o n s e q u e n t i a l " shift
to endosporangial g a m e t o p h y t e nutrition in seed plants).
In a few characters, w e are able to perceive a hierarchy of gradational to abrupt
c h a n g e s as well as the degrees of consistency. In both g a m e t o p h y t i c and sporophytic
character sets, the spore plant/seed plant discontinuity is far m o r e p r o m i n e n t than the
g y m n o s p e r m / a n g i o s p e r m discontinuity, w h e r e a s the cycad/conifer and the m o n o c o t /
dicot discontinuities are of the s a m e rank and are both subordinate to the g y m n o s p e r m /
a n g i o s p e r m o n e . Such considerations, partly intuitive, underlie the traditional system of
higher taxa that is less arbitrary than it may seem. T h e a b o v e a r g u m e n t s are in favour of
a conservative system of the higher taxa:

Divisions Classes Subclasses

Pteridophyta Protopteriae
Filicinae
Lycopodiae
Articulatae

Spermophyta Gymnospermac Protospermidae

Pteridospermidae
Cycadidae
Coniferidae

Angiospermae Dicotylidae
Monocotylidae
 Prehistory
A n g i o s p e r m prehistory starts with the a p p e a r a n c e of the early D e v o n i a n seed plants
with protected (cupulate) ovules. That these plants w e r e allied with g y m n o s p e r m s rather
than with a n g i o s p e r m s is d u e primarily to the pre-existing idea of phylogenetic rela­
tions b e t w e e n these g r o u p s .
As we noticed a b o v e , the initial concept of g y m n o s p e r m y referred to fruits lacking
morphologically distinct pericarps, thus having nothing to d o with our present-day idea
of g y m n o s p e r m s . T h e latter b e c a m e distinctive by their negative characteristics as lack­
ing in such and such typical a n g i o s p e r m characters. T h e r e s e e m e d to b e also clear-cut
life-form and c o m m e r c i a l differences. T h e conifers p r o d u c e d soft w o o d and needle
leaves. T h e less familiar c y c a d s also bore c o n e s but differed in the thick, scarcely
branched stems c r o w n e d by large pinnate leaves. Ginkgo, a fairly isolated g e n u s in the
m o d e r n flora, j o i n e d g y m n o s p e r m s on account of its naked ovules and motile sperms as
in c y c a d s . T h e latter feature testified primitiveness of both these g r o u p s .
T h u s , on the p r e s e n t t i m e p l a n e , g y m n o s p e r m s a p p e a r e d well s e p a r a t e d from
a n g i o s p e r m s , with three genera of gnetaleans, or c h l a m y d o s p e r m s , as the only possible
intermediates. Interpretations of the vessels, cupules, nuclear e m b r y o sacs and d o u b l e
fertilization in gnetaleans as h o m o l o g o u s or n o n - h o m o l o g o u s to the corresponding struc­
tures in a n g i o s p e r m s were subjects of long-lasting controversies. In any case, the extant
gnetaleans formed a very thin morphological link at most.
With the addition of n u m e r o u s fossil g r o u p s , the distinctions b e t w e e n g y m n o s p e r m s
and angiosperms b e c a m e less clear-cut. Pteridosperms and caytonialeans bore their ovules
in ovary-like cupules, while bennettites had flower-like bisexual reproductive shoots. If
these plants lived today, the classification of S p e r m o p h y t a might h a v e been very differ­
ent from what w e are used to. S o m e extinct Paleozoic and M e s o z o i c seed plants went,
as it were, with g y m n o s p e r m s , though scarcely fitting the original g y m n o s p e r m con­
cept. T h u s inflated, g y m n o s p e r m s s e e m no longer clear-cut from a n g i o s p e r m s .
O n e can argue that the division of seed plants into g y m n o s p e r m s and a n g i o s p e r m s
reflects the present-day situation and is inadequate in terms of the evolutionary history.
Logically, an evolutionary classification would c o n v e y c h a n g e s in the diversity pat­
terns through geological time rather than an accidental pattern of a single time plane.
But the prospects of such a classification are still r e m o t e .
T h e r e h a v e been several attempts at a superordinal g y m n o s p e r m classification (Ar­
nold, 1948; L a m , 1950; Pant, 1959; Emberger, 1960; Chamberlain, 1 9 6 6 a ; M e y e n , 1987),
using such characters as pycnoxyly versus manoxyly, microphylly versus macrophylly,
p h y l l o s p e r m y versus s t a c h y o s p e r m y (ovules borne on leaves or axial organs), radio-
spermy versus platyspermy (pertaining to the s y m m e t r y of ovules), and p h a n e r o g a m y
versus p r e p h a n e r o g a m y , i.e. the shedding of seeds or unfertilized ovules. S o m e of these
characters are adaptive (e.g. wood structure), others scarcely seem so (e.g. seed symmetry:
in Ginkgo, there are both platyspermic and radiospermic ovules). T h e i r phylogenetic
significance is still to be learned. But the reason why most of these features have failed
as supraordinal distinctions is that they are either inconsistent within the g y m n o s p e r m
orders (e.g. seed s y m m e t r y ) or infringe the ordinal boundaries (e. g. microphylly j o i n ­
ing conifers with e p h e d r a c e a n g n e t a l e a n s ) . S i n c e a n g i o s p e r m s are defined by the
ovuliphore structure (angiospermy) c o m b i n e d with the ovuliphore a r r a n g e m e n t (flower),
it may be appropriate to apply the s a m e criteria to the n o n - a n g i o s p e r m o u s seed plants.
An adaptive type classification of the nature of the a n g i o s p e r m / g y m n o s p e r m divi­
sion can be based on the ovules being borne open ( g y m n o s p e r m s proper) or in cupules
( c h l a m y d o s p e r m s ) . T h e latter would include, in addition to the extant gnetaleans, also
several extinct groups of cupuliferous plants, such as p t e r i d o s p e r m s , c a y t o n i a l e a n s ,
bennettites and c z e k a n o w s k i a l e a n s . T h e ovuliphore a r r a n g e m e n t can be described as
non-strobilate in p t e r i d o s p e r m s , strobilate in c y c a d s and their allies (including a few
destrobilate taxa, such as Cycas and Ginkgo) and doubly-strobilate, as in coniferids.
C o m b i n i n g the ovuliphore m o r p h o l o g y with ovuliphore a r r a n g e m e n t s , in the m a n n e r of
the a n g i o s p e r m - flowering plant c o m b i n a t i o n that is m o s t closely a p p r o a c h e d by
p r o a n g i o s p e r m s , we arrive at the following s c h e m e :

Gymnospermidae Chlamidospermidae

Cycadidae Coniferidae Pteridospermidae Proangiospermidae

Glossopteri dales Cordaitales Hydrospermales Caytoniales

Vojnovskyales Walchiales Lagenostomales Bennettitales
Peltaspermales Coniferales Trigonocarpales Czekanowskiales
Nilssoniales Callistophytales Gnetales
Cycadales
Gink^oales

We shall return to this s c h e m e after briefly considering the g y m n o s p e r m orders.

Although general characteristics of the orders can be found e l s e w h e r e , it is necessary to
put forth the characters and interpretations that will serve for the analysis of phylogenetic
relations.

Progymnosperms
G y m n o s p e r m characters appeared in the Devonian aneuropterids and archaeopterids,
arboreal plants with a coniferoid stem anatomy, by virtue of which they are assigned to
p r o g y m n o s p e r m s (Beck, 1960). T h e y produced three-dimensional s y s t e m s of lateral
b r a n c h e s , and their u l t i m a t e a p p e n d a g e s c o u l d be w e b b e d and leaf-like, yet not form­
ing leaf b l a d e s . Fertile parts of the lateral branch s y s t e m s b o r e p a i r e d s p o r a n g i a . S o m e
a r c h a e o p t e r i d s w e r e h o m o s p o r o u s , the o t h e r h e t e r o s p o r o u s , w i t h m e g a s p o r a n g i a
slightly larger than, but o t h e r w i s e similar to, m i c r o s p o r a n g i a . In t h e s e p l a n t s , hete­
rospory is c o n s i d e r e d as a d e r i v e d c h a r a c t e r w h i c h , h o w e v e r , w a s scarcely correlated
with the a d v a n c e m e n t of assimilating o r g a n s . F o r e x a m p l e , in Tanaitis, a late M i d d l e
D e v o n i a n a r c h a e o p t e r i d , h e t e r o s p o r y is a s s o c i a t e d with n o n - w e b b e d u l t i m a t e b r a n -
chlets of a very p r i m i t i v e s y n t e l o m i c aspect ( K r a s s i l o v & al., 1987).
That ovules appeared in the direct d e s c e n d a n t s of p r o g y m n o s p e r m s , is yet to be
proved. First seed plants c o n t e m p o r a n e o u s with p r o g y m n o s p e r m s had ovules with lobed
integuments borne in horny cupules that looked as whorled, repeatedly forking branches.
In Hydrosperma, such c u p u l e s contained also m i c r o s p o r a n g i a ( L o n g , 1977a). Lenlogia,
a M i d d l e D e v o n i a n plant from Siberia (Krassilov & Z a k h a r o v a , 1995), had clustered
fertile b r a n c h e s quite similar to the seed-bearing c u p u l e s but actually containing paired
sporangia (Fig. 1). Its axial organs s h o w e d a primitive protostelic structure suggesting
that seeds could appear in small plants totally lacking g y m n o s p e r m o u s stem anatomy.

Pteridospermidae (pteridosperms)
T h e c o n c e p t of p t e r i d o s p e r m s , or seed ferns, is d u e to a fascinating discovery of
seeds taphonomically associated with and later found actually attached to the Carbonife­
rous fern-like fronds that w e r e for m a n y years considered as fern foliage. After the
initial distrust, the concept b e c a m e very popular a m o n g plant m o r p h o l o g i s t s and be­
c a m e inflated to e n c o m p a s s the Permian and M e s o z o i c plants with fern-like leaves but
having their seeds b o r n e in strobili, like in m o d e r n g y m n o s p e r m s . In the extant cycad
Bowenia, the leaves are bipinnate and very fern-like. Is it a p t e r i d o s p e r m ? E v e n Ginkgo
has a fair c h a n c e of being included into p t e r i d o s p e r m s ( M e y e n , 1984). Apparently,
pteridosperms have to be defined more strictly so as not to engulf all traditional orders of
gymnosperms. T h e initial idea of leaf-like ovuliphores may suffice as a demarcation crite­
rion setting pteridosperms apart form cycads which have ovulate scales that are gathered in
more or less compact strobili as well as from ginkgoaleans with the ovules typically termi­
nal on the simple or forked pedicels. This criterion brings peltasperms and other Mesozoic
"pteridosperms" closer to cycads than to the typical pteridosperms (Krassilov, 1990).
T h e r e are parallel pteridosperm classifications based on stem a n a t o m y and seed char­
acters ( S e w a r d , 1917). In the former, the t w o major o r d e r s , L y g i n o p t e r i d a l e s and
M e d u l l o s a l e s , are d i s t i n g u i s h e d by the v a s c u l a r s t r u c t u r e b e i n g e i t h e r transitional
protostelic-eustelic or polystelic (actually s e g m e n t e d protostelic). T h e i r stratigraphic
ranges overlap and they might h a v e g r o w n side by side in Carboniferous forests. H o w ­
ever, lyginopterid a n a t o m y and, especially, its c a l a m o p i t y a c e a n and b u t e o x y l o n a c e a n
Fig. 1. Devonian precursors of seed plants: (1,2) Lenlogici k rys lit ofov i chi i (Rixdcz.) Krassil. & Zakharova,
cupule-like clusters of bisporangiate synangia, from the Late Devonian of Minusinsk Basin, Siberia
(Krassilov & Zakharova, 1995), x8, (3) helical tracheids from the protostele of the same plant, SEM,
xl 000, (4) Moresnetia zalesskyi Stockmans, cupule of an early seed plant from the Devonian of Belgium
(courtesy of A.R. Ananiev), xlO.
variants appeared earlier and w e r e initially associated with primitive syntelomic as­
similating organs and the likewise primitive early seeds. T h e latter c o m b i n a t i o n deser­
ves the status of a separate order, H y d r o s p e r m a l e s , including the d i c h o t o m o u s l y dissec­
ted c u p u l e s containing several to m a n y small ovules, triangular in cross-section, with a
lobed integument and a distally free nucellus ( L o n g , 1977a; M a t t e n & al., 1980). T h e
nucellar apex w a s transformed into a special pollen-catching d e v i c e , l a g e n o s t o m e : a
broad pollen c h a m b e r was pierced by a central c o l u m n and e x t e n d e d into a funnel-
shaped salpinx. T h e c o l u m n , vascularized with spiral tracheids m i g h t h a v e functioned
as a secretory organ. Occasionally, the m e g a s p o r e m e m b r a n e s h o w s a triradiate scar
and even the r e m n a n t s of a tetrahedral tetrad, as in Lyrosperma (Pettitt, 1969). In a
closely allied but m o r e a d v a n c e d s e e d - g e n u s Gnetopsis, the ovules w e r e s o m e w h a t flat­
tened and w e r e provided with a p a p p u s of apical hairs (Gensel & S k o g , 1977).

T h e advanced lyginopterids had a Sphenopteris-iypt foliage often s h o w i n g mar­

ginal hydathodes, as well as the branching pollen organs of the Feraxotheca-Crossotheca
type, with the sporangia in erect or p e n d i n g clusters, proximally fused but separated at
maturity and p r o d u c i n g spore-like prepollen (Stubblefield & al., 1982). T h e y roughly
c o r r e s p o n d to the L a g e n o s t o m a l e s of a seed-based classification, including the radially
s y m m e t r i c a l o v u l e s , solitary in cyathiform c u p u l e s that s h o w characteristic secretory
glands. T h e o v u l e s had vascularized integuments displaying a residual lobing but, at
least in the a d v a n c e d seed-genera, forming a conical m i c r o p y l e over a c o n s p i c u o u s
l a g e n o s t o m e . T h e m e g a g a m e t o p h y t e p r o d u c e d three apical a r c h e g o n i a ( L o n g , 1977c)
and a special p r o t u b e r a n c e , or tent-pole, piercing the b o t t o m of the pollen chamber. In
the geologically y o u n g e r representatives, there w a s a tendency to smaller and consider­
ably flattened s e e d s with a r e d u c e d l a g e n o s t o m e and the c u p u l e f o r m i n g an outer
m i c r o p y l e , or e x o s t o m e .

T h e M e d u l l o s a l e s are linked to the lyginopterids through such morphologically, if

not phylogenetically, intermediate forms as Questora that had six central and multiple
peripheral vascular bundles in the leaf traces, although the leaves w e r e decussate, an
allegedly derived feature in p t e r i d o s p e r m s ( M a p e s & R o t h w e l l , 1980). S o m e m e d u l -
losaleans might have been lianas with long decurrent leaves. However, the larger Permian
species w e r e trees with the stems u p to 5 0 c m thick. In the primitive m e d u l l o s a l e a n s , the
leaf traces w e r e concentric and produced a considerable a m o u n t of secondary wood.
T h e s e "foliar s t e l a e " w e r e m u c h like the n u m e r o u s additional axial stelae p r o d u c e d at
the leaf n o d e s . In the late species, the leaf traces lost their secondary tissues, and the
additional stelae b e c a m e reduced.
Characteristic of m e d u l l o s a l e a n s w e r e the pitted tracheids with helical and reticulate
thickenings, the p h l o e m fibers, as in c y c a d s , and the s c l e r e n c h y m o u s strands in the
periderm. T h e leaves w e r e large, pinnate, with the venation pattern open or reticulate.
T h e typical pollen organs w e r e large c u p - s h a p e d or bell-shaped s y n a n g i a of multiple
tubular sporangia arranged in concentric rings (Millay & Taylor, 1979; Taylor & Millay,
 1981). Such structures might have been related to poUinivory and insect pollination.
S o m e synangia w e r e pubescent and resembled seeds ( D e l e v o r y a s , 1964), p e r h a p s en­
couraging the visits of pollinivorous insects to the latter. T h e prepollen grains were
large, with a m o n o l e t e or occasionally trilete (in Potonied) leasure, with distal folds and
a thick alveolate sexine (Monoletes: Taylor, 1978) e v o l v i n g in the direction of a m o n o -
saccus or, rarely, bisaccate.
T h e ovules assigned to the Trigonocarpales were radially s y m m e t r i c a l , with the in­
tegument free from the nucellus and with both integument and nucellus vascularized.
T w o apical a r c h e g o n i a w e r e observed in the m e g a g a m e t o p h y t e d e v e l o p i n g from a func­
tional m e g a s p o r e of a tetrahedral tetrad. In the form of the m i c r o p y l e and the pollen
chamber, with a short nucellar beak, the trigonocarps s e e m m o d e r n i z e d in respect to the
l a g e n o s t o m a l e a n s . Vascularization of the nucellus is c o m m o n l y c o n s i d e r e d primitive,
but actually the trigonocarpalean nucelli, s o m e of t h e m ribbed as i n t e g u m e n t s ( S m o o t
& Taylor, 1983/1984), could have evolved by m e a n s of fusion of the l a g e n o s p e r m a l e a n
or h y d r o s p e r m a l e a n - t y p e nucelli with their integuments. T h e trigonocarpalean integu­
ments would then be h o m o l o g o u s to the l a g e n o s t o m a l e a n - t y p e c u p u l e s .

T h e Callistophytales, a m i n o r but phylogenetically important order of Carbonifer­

ous p t e r i d o s p e r m s , w e r e represented by shrubby plants with creeping spiny stems bear­
ing adventitious roots and 3-4-pinnate leaves (Stidd & Hall, 1970; R o t h w e l l , 1980;
Taylor, 1981). T h e y r e s e m b l e lyginopterids in the vascular structure, s c l e r e n c h y m a t i c
periderm ( s p a r g a n u m ) , and glands. Their synangiate pollen organs of radially disposed
sporangia s e e m also derivable from those of lyginopterids. T h e pollen grains were,
however, of a m o r e a d v a n c e d coniferoid morphology, and the o v u l e s w e r e platysper-
mic, with lateral w i n g s . H e n c e , the Callistophytes show a m o s a i c of typically pterido-
s p e r m o u s and coniferoid characters.

T h e G i g a n t o p t e r i d a l e s w e r e c h a r a c t e r i s t i c of the P e r m i a n e a s t e r n A s i a t i c , or
Cathaysian, palaeofloristic province, but the g r o u p e x t e n d e d to the M i d d l e East and
North A m e r i c a as well. T h e 2-3-pinnate leaves s h o w e d a tendency to fusion of succes­
sively larger units until, in the end of the series, broad entire leaves w e r e formed ( A s a m a ,
1962, 1976; Li & al., 1994). Their venation w a s typically hierarchical areolate, partly
retaining the original vein pattern of the constituent p i n n a e . Both the pollen and the
ovule-bearing organs were leaf-like, with linear synangia and submarginal r o w s of ovules
respectively (Li & Yao, 1983). T h e North Amencm Spermopteris, a laminar taeniopteroid
ovuliphore with n u m e r o u s ovules aligned along and slightly p r o t r u d i n g b e y o n d the
margins (Cridland & M o r r i s , 1960), was obviously related to this Asiatic g r o u p . T h e so-
called c y c a d s from the Permian of North A m e r i c a ( M a m a y , 1976) had similar, though
s o m e w h a t reduced, ovuliphores. Unlike in c y c a d s , their nucelli w e r e m o r e strongly
cutinized than the integuments.
 Cycadidae (cycadophytes)
T h e c y c a d o p h y t e s are usually referred to as the fossil seed plants r e s e m b l i n g c y c a d s
in the m o r p h o l o g y of their leaves and/or reproductive structures. A s such, they m a y
include many, if not all, pteridosperm leaf genera recalling of Bowenia, as well as the
leaf-like o v u l i p h o r e s r e s e m b l i n g Cycas. H e r e they are defined primarily on the basis of
strobilate ovuliphores (including a few derived destrobilate forms) that m a k e t h e m dif­
ferent from p t e r i d o s p e r m s . T h e leaf-like ovuliphores of glossopterids are conceived of
as formed by fusion of the axillar reproductive shoots with their subtending leaves.

T h e Glossopteridales constituted a d o m i n a n t g r o u p of the P e r m i a n G o n d w a n a Prov­

ince extending into the Triassic. T h e y are characterized by the entire elliptical leaf blades
with reticulate or s o m e t i m e s open venation. T h e midrib was formed of separate bundles
given off as lateral veins. T h e a n a s t o m o s e s w e r e of different types, p r o d u c e d as short
side b r a n c h e s as well as by splitting and looping of the lateral veins. Atypical leaf
m o r p h o l o g i e s of glossopterid alliance were taeniopteroid, lobed, as in Pteronilssonia
(Pant & M e h r a , 1963), or ribbon-shaped with parallel venation and resembling cordaites,
as in Noeggerathiopsis.
A s for the growth forms, the opinions vary. Glossopteris leaves w e r e found attached
to Vertehraria stems or r h i z o m e s the characteristic articulate aspect of which w a s due to
the a e r e n c h y m o u s w e d g e s divided by x y l e m e ridges associated with adventitious roots
( S c h o p f , 1 9 8 2 ) . T h e s e s t r u c t u r e s s u g g e s t a m a n g r o v e h a b i t a t . H o w e v e r , n o t all
glossopterid allies had Vertebraria structure, and the leaves w e r e occasionally bunched,
as if borne in spur-shoots ( T h o m a s , 1952; Pant, 1968).
T h e pollen organs were the dichotomously branched androclades, with the sporangial
clusters terminal on the ultimate branchlets or strobilate, as in Kendostrobus (Surange
& C h a n d r a , 1975). T h e pollen grains w e r e typically saccate striate, with the exine di­
vided into parallel stripes, or taeniae. T h e ovulate structures w e r e of n u m e r o u s types,
s o m e of them (Arberia) gynoclades of the same d i c h o t o m o u s structure as the androclades
and bearing ovules on incurved branchlets. T h e m o r e advanced cupulate structures might
have been derived from this basic type (Rigby, 1978). T h e y w e r e cyathiform or discoid,
open (or closed at earlier d e v e l o p m e n t a l stages), variously lobed or w i n g e d , stalked or
sessile and axillar (Pant, 1977), or, m o r e frequently, attached to the leaf-like bracts. T h e
ovules w e r e p l a t y s p e r m i c with lateral wings or glochidial a p p e n d a g e s and with a long
m i c r o p y l e tube that w a s occasionally funnel-shaped. T h e nucellus w a s free, forming a
broad pollen chamber. T h e m e g a s p o r e m e m b r a n e w a s cutinized. T h e m e g a g a m e t o p h y t e
was cellular, with t w o apical archegonia. T h e seedlings p r o d u c e d t w o lobed cotyledons
(Pant & Nautiyal, 1987).
In distinction from the typical pteridosperms in which the ovules occurred in fertile
regions of the lateral branch systems evolving into leaf-like ovuliphores, the leaf-cupule
complexes of glossopterids appear secondary, derived by fusion of the axillar ovulate
branches to their subtending bracts. W h y such structures were so widespread a m o n g glosso­
pterids and in no other group of Permian g y m n o s p e r m s , remains enigmatic. Functionally,
they could have served as dispersal units in the same way as bracteate angiosperm fruits.

T h e Vojnovskyales were characterized by the radial a r r a n g e m e n t of bracts and ovules

in the apical z o n e of scaly short shoots (Vojnovskya) or on p e d i c e l l a t e r e c e p t a c l e s
(Gaussia). Short shoots alternated v^iih Nephropsis-iype r h o m b o i d leaves or cataphylls,
the latter s h o w i n g an open flabellate venation pattern ( N e u b u r g , 1963). O t h e r foliar
types associated with the vojnovskyalean ovulate organs w e r e lanceolate or taeniate,
with parallel veins and intercostal stomatal g r o o v e s {Rufloria) and w e r e b o r n e in termi­
nal clusters on short shoots. Zamiopteris-iypt leaves with an indistinct m i d r i b and o b ­
lique lateral veins, less c o m m o n l y with reticulate venation, m i g h t h a v e b e l o n g e d to the
s a m e alliance.
All these foliar morphologies occurred also in the Glossopteris flora of G o n d w a n a l a n d
localities. Such similarities are currently ascribed to evolutionary parallelisms. In par­
ticular, the parallel-veined leaves might share a phyllodic origin. H o w e v e r , parallelisms
extend to reproductive structures as well. T h e radial ovulate structures occurred in both
g r o u p s , w h i l e Cladostrobus, a p u t a t i v e v o j n o v s k y a l e a n p o l l e n c o n e with p e l t a t e
s p o r a n g i o p h o r e s ( M e y e n , 1984), r e s e m b l e d Kendostrohus of glossopterids ( S u r a n g e &
C h a n d r a , 1975), and the pollen grain m o r p h o t y p e s w e r e very m u c h alike, too. On the
other hand, the proclaimed cordaitalean affinities of vojnovskialeans are b a s e d on su­
perficial foliar similarities alone, the reproductive structures having had little, if any­
thing, in c o m m o n with the c o m p o u n d strobili of the cordaites.

T h e Peltaspermales, another d o m i n a n t g r o u p of the P e r m i a n and Triassic seed plants

usually linked to p t e r i d o s p e r m s , is here assigned to c y c a d i d s on the a c c o u n t of their
typically strobilate reproductive structures. T h e y are reconstructed as small trees or
shrubs with the stem u p to 5 c m thick, bearing c o n s p i c u o u s discoid g l a n d s and having a
rather thin cylinder of secondary w o o d (Harris, 1964). T h e leaves w e r e typically 2-4-
pinnate with interstitial pinnules, 1-pinnate or simple in the derived forms. T h e petioles
were typically swollen at the base, anatomically r e s e m b l i n g the m e d u l l o s a l e a n s and
s o m e transitional forms b e t w e e n the latter and lyginopterids (see a b o v e ) . T h e rachises
were forked or overtopped by the distal p i n n a e (Kerp, 1986). T h e r e w a s a tendency to
segment fusion involving the successively larger units, as in the gigantopterids, and
a c c o m p a n i e d by a transition from an open to areolate venation (a m o r e detailed account
of the latter will be given in the next chapter). T h e e n d - m e m b e r s of the m o r p h o l o g i c a l
series w e r e bilobed or dichopodially lobed flabellate leaves, as in Furcula or Scoresbya
(Harris, 1932). Characteristic of t h e m w e r e thick cuticles s h o w i n g a radial a r r a n g e m e n t
of stomatal subsidiary cells that formed an irregular petaloid structure with the stomatal
pits encircled by a thick papillate ridge (Fig. 2).
Fig. 2. Scytophylliim vuli^are (Pryn.) Dobrusk., pcltasperm leaves from the Late Triassic of Eastern Urals
(Krassilov, 1995) showing a marginal fusion of the pinnae and interstitial pinnules that are marked by a
conservative venation pattern: two leaf fragments, x2, and papillate cuticle (bottom), SEM, x40().
 T h e pollen organs were dichotomously branched androclades, with the ultimate peltate
branchlets bearing pendent sporangia in rows or radial clusters, free or synangiate. T h e
pollen grains were sulcate asaccate or with a bilobed m o n o s a c c u s , as in the callistophytes,
or bisaccate. S o m e striate m o r p h o t y p e s , notably Vittatina, m i g h t h a v e been p r o d u c e d
by peltasperms ( M e y e n , 1984 and e l s e w h e r e ) . T h e ovulate organs w e r e loose strobili of
peltate ovuliphores, with t w o to m a n y abaxial ovules encircling the stalk. T h e latter
ovule a r r a n g e m e n t w a s typical of the M e s o z o i c p e l t a s p e r m s less frequently occurring
in the P e r m i a n calliptrids that are presently considered as archaic p e l t a s p e r m s . T h e
ovules w e r e p l a t y s p e r m i c , with a long reflexed m i c r o p y l e tube, a free nucellus and a
cutinised m e g a s p o r e m e m b r a n e .

T h e C o r y s t o s p e r m a l e s w e r e widespread in the Triassic of the southern continents

producing a great diversity of growth forms (Anderson & A n d e r s o n , 1983, 1985). T h e
s t e m in t h e larger f o r m s (Rhexoxylon) shows a medullosalean vascular structure
(Archangelsky, 1968b). T h e leaves were essentially like those in peltasperms (see above),
but forking in the proximal third, in the derived forms bilobed, flabellate or entire. T h e
venation w a s typically open, the ultimate s e g m e n t s w e r e occasionally single-veined.
T h e stomata show an irregular or, less c o m m o n l y , a radial or tetracytic a r r a n g e m e n t of
p r o x i m a l l y papillate or ridged subsidiary cells. T h e pollen o r g a n s w e r e p i n n a t e or
strobilate, with forking peltate s p o r a n g i o p h o r e s bearing free sporangia in r o w s or radial
clusters. T h e pollen grains w e r e bisaccate (Taylor & al., 1984). T h e ovulate organs
were loose g y n o c l a d e s , their lateral branches forking, terminating in a pair of incurved,
indistinctly lobed cupules, the latter containing a solitary ovule with a protruding bilobed
micropyle ( T h o m a s , 1955). T h e nucellus was thinly cutinised and lacking cutinized
megaspore membrane.

T h e Pentoxylales are k n o w n from the Jurassic and L o w e r C r e t a c e o u s of India and

Australia (Rao, 1974; Harris, 1983; Drinnan & C h a m b e r s , 1986). T h e g r o u p received
its n a m e d u e to the vascular system with a ring of typically five, rarely three to nine,
p r i m a r y b u n d l e s p r o d u c i n g p y c n o x y l i c s e c o n d a r y w o o d with u n i s e r i a t e r a y s . T h e
tracheids s h o w e d hexagonal bordered pits, as in bennettites. T h e shoots w e r e dimor­
phic, with sparse or d e n s e leaf scars. T h e leaves w e r e s i m p l e taeniopteroid or pinnate
with taeniopteroid pinnules, their midribs formed of u p to nine separate b u n d l e s ; the
lateral veins w e r e simple or forked and looping, terminating in the marginal h y d a t h o d e s .
T h e leaf blades w e r e pubescent, h y p o s t o m a t i c . T h e s t o m a t a w e r e typically tetracytic or
with 6-7 radial subsidiary cells. Both pollen and ovulate o r g a n s w e r e clustered on dif­
ferent short shoots, the former delicate, simple or forked, with free sporangia, the latter
strobilate, with erect ovules in five oblique r o w s , coalescing at maturity into a fruit-like
body. T h e pollen grains w e r e m o n o s u l c a t e asaccate with a granular infrastructure and a
lamellate riexine ( O s b o r n e & al., 1991). T h e ovules s e e m to h a v e been bitegmic (?),
with a m a s s i v e sarcotesta and a cutinized sclerotesta, the latter forming a m i c r o p y l e .
T h e nucellus w a s free, the m e g a s p o r e m e m b r a n e w a s not o b s e r v e d .
 T h e Nilssoniales were ubiquitous in the M e s o z o i c , surviving to the terminal Creta­
ceous and usually assigned to c y c a d s , but they p r o d u c e d the b r a n c h i n g long shoots and
the d e c i d u o u s spur-shoots bearing clusters of simple, lobed or pinnate, rarely bipinnate,
leaves (Krassilov, 1972c; K i m u r a & Sekido, 1975). T h e leaf blade w a s thick, flat or
involute, occasionally serrate. T h e venation was typically open pinnate, with simple
lateral veins. S o m e associated foliar m o r p h o t y p e s had reticulate venation (Ctenis), or
their p i n n a e w e r e s u p p l i e d with t w o basal v e i n s g i v i n g off s u b p a r a l l e l b r a n c h e s
(Heilungia). T h e pollen c o n e s were cycad-like, with scaly s p o r a n g i o p h o r e s . T h e pollen
grains w e r e sulcate asaccate, with an alveolar infrastructure. T h e ovulate cones were
loose, pendent, with peltate ovuliphores bearing two ovules. In distinction from cycads,
even the i m m a t u r e c o n e s were open and the ovules occurred abaxially between the
involute margins of a distal pelta (Harris, 1964). T h e ovules w e r e p l a t y s p e r m i c , with
resin bodies in the sarcotesta. T h e nucellus was free apically, forming a pollen c h a m b e r
c r o w n e d with a short nucellar beak. T h e r e was a cutinized m e g a s p o r e m e m b r a n e .

T h e C y c a d a l e s include ten extant and a few Tertiary genera, w h e r e a s the pre-Terti-

ary records are problematic except, perhaps, for the Late C r e t a c e o u s Cyccis s h o w i n g
cuticular structures peculiar for the g e n u s (Krassilov, 1979). T h e stem is thick cylindri­
cal or tuberous subterranean, u n b r a n c h e d (pachycaul), a r m o u r e d with petioles of shed
leaves, with an apogeotropic coralloid root m a s s . Occasional b r a n c h i n g is d u e to w o u n d
m e r i s t e m s or adventitious buds. T h e primary body is formed by the apical m e r i s t e m as
well as by the separate lateral meristem and diffuse g r o w t h , the contribution of the
secondary tissues being less significant. T h e leaf traces are girdling the stem. T h e leaves
are p r o d u c e d in clusters (in Stangeria singly) once a year or biannually, alternating with
the c o n e s . T h e y are I-pinnate stipulate, conduplicate or involute in bud, in Bowenia 2-
pinnate fern-like and circinnate in bud (Stevenson, 1981), with the pinnules linear, lan­
ceolate, r h o m b o i d or flabellate, pubescent. T h e venation is open or a n a s t o m o s i n g , rarely
with a series of marginal loops (Stangeria). T h e e p i d e r m i s s h o w s alternating zones of
longitudinal and transverse-oblique cells (occurring in the C r e t a c e o u s species already),
the cell walls straight or sinuous. T h e stomata occur in g r o u p s or files, slightly or deeply
sunken, s o m e t i m e s with encircling cells d o m e - l i k e o v e r h a n g i n g the guard cell - sub­
sidiary cell c o m p l e x (Pant & Nautiyal, 1963).
T h e r e p r o d u c t i v e o r g a n s are s t r o b i l a t e , d i c l i n o u s , but o c c a s i o n a l l y with a few
sporangia in ovulate strobili. T h e pollen c o n e s are large, with n u m e r o u s (up to 700)
sporangia abaxial on the shortly petiolate or sessile, entire or, rarely, bilobed scales. T h e
sporangia o c c u r in d e n s e masses, grouped by 2-5 in inconspicuous sori, free or proxi­
mally fused in half-synangia (Smith, 1907), with thickened cells at the apex and along
the d e c h i s c e n c e line and with stomata, a feature shared with p e l t a s p e r m s (Antevsia)
alone. T h e pollen grains are produced in great quantities, sulcate asaccate or with rudi­
mentary sacci, transferred by insects (Tang, 1987). T h e pollen wall infrastructure is
alveolate. T h e ovulate c o n e s are very large, d e v e l o p m e n t a l l y apical on the stem, seldom
lateral, in Bowenia on short lateral branches, shifted to the leaf axils or, in Cycas, form-
ing an imbricate c o v e r over the stem apex that r e s u m e s g r o w t h after s h e d d i n g the
ovuliphores. T h e latter are scaly, with sterile a p o p h y s e s peltate or flared and lobed, in
Cycas leaf-like ( C h a m b e r l a i n , 1966b). T h e ovules are typically t w o per scale, inverted,
pendent at maturity, in Cycas from t w o to eight in t w o r o w s , r a d i o s p e r m i c . T h e integu­
ment is thick three-layered, with stomata and hairs, the latter s o m e t i m e s clustered around
the micropyle. T h e nucellus is strongly cutinized in the u p p e r third alone, but with a
delicate cuticle extending d o w n to the base. T h e pollen c h a m b e r is c r o w n e d with a
nucellar beak. T h e vascular system is d o u b l e , the peripheral in the i n t e g u m e n t and the
interior shared by the nucellus and integument (Pant & Nautiyal, 1963). T h e m e g a s p o r e
tetrad is linear. T h e r e is a c o n s p i c u o u s m e g a s p o r e m e m b r a n e .
T h e m e g a g a m e t o p h y t e remains nuclear until a 100-nucleate stage, with t w o to four
apical archegonia. T h e pollen tubes are haustorial. Both s p e r m s m a y function in fertili­
zation (Rao, 1974). P o l y e m b r y o n y is c o m m o n . T h e p r o e m b r y o r e m a i n s n u c l e a r until a
6 4 - 1 0 0 - n u c l e a t e stage. T h e c o t y l e d o n s are usually t w o (in Ceratozamia solitary),
haustorial, their apices remaining in the seed.

T h e G i n k g o a l e s were diverse in the M e s o z o i c , with the only extant - cultivated -

species, a deciduous tree with dimorphic shoots. T h e long shoots are pycnoxylic, whereas
the short shoots are m a n o x y l i c . T h e leaves are long petiolate flabellate, on the long
shoots lobed, on the short shoots notched, with short resin ducts b e t w e e n the veins. T h e
venation is flabellate d i c h o t o m o u s , arising from t w o basal veins running along the mar­
gins, mostly open, with infrequent but constantly occurring a n a s t o m o s e s (Arnott, 1959.
T h e stomata are provided with radial subsidiary cells, papillate. T h e j u v e n i l e leaves are
pubescent, shedding hairs at maturity.
T h e trees are dioecious. T h e pollen organs are strobilate, pedicellate, occurring on
the short shoots in a subapical ring peripheral to the apical sterile leaves and are encir­
cled by the glandular bracts. T h e s p o r a n g i o p h o r e s are stalked, peltate, with an apical
gland, bearing t w o or, rarely, 3-4 abaxial sporangia T h e pollen grains are sulcate asaccate
or with residual sacci (Herzfeld, 1927), and with alveolar infrastructure. T h e g y n o c l a d e s
consist of a long, slightly flattened stalk, distally e x p a n d e d and bearing t w o erect arillate
ovules. T h e y are borne in the apical zone of the short shoots intermingled with sterile
leaves (and are s o m e t i m e s described as axillar to the leaves but topologically equivalent
to them) and surrounded by t w o circles of bracts, the outer sessile, the inner petiolate.
T h e leaf petioles and the ovulate stalks are identical externally, but the latter are sup­
plied with four bundles instead of two. A n o m a l o u s o v u l i p h o r e s bear a solitary o v u l e or
several ovules on short branches, or they are leaf-like ( F a v r e - D u c h a r t r e , 1956) and
probably chimeric, arising by fusion with sterile leaves.
T h e ovules are typically platyspermic bicarinate, with t w o i n t e g u m e n t a r y bundles
along the keels, less c o m m o n l y radiospermic 3-4-keeled, with a c o r r e s p o n d i n g n u m b e r
of vascular bundles. An arillate collar develops later than the i n t e g u m e n t ( P a n k o w &
S o t h m a n , 1967). T h e nucellus is free in the apical portion alone, scarcely cutinized,
forming a broad pollen c h a m b e r with a nucellar beak. T h e m e g a s p o r e m e m b r a n e is
cutinized. T h e m e g a g a m e t o p h y t e remains nuclear until a 2 5 6 - 1 0 0 0 - n u c l e a t e stage, with
two or, rarely, four apical archegonia and with a tent-pole. Fertilization occurs before or
after shedding the ovules. T h e p r o e m b r y o remains free nuclear until a 128-256-nucle-
ate stage, but with a partial cell wall laid d o w n after the first division (Lee, 1955). T h e
cotyledons are normally two, as in c y c a d s .
T h e extinct M e s o z o i c forms were either similar to the extant species or differing in
the lanceolate parallel-veined or palmately dissected leaves with or without petioles as
well as in t h e d i s t i n c t l y s t r o b i l a t e o v u l i p h o r e s b e a r i n g e r e c t or i n c u r v e d o v u l e s
(Atchangelsky, 1965), the latter with free cutinized nucelli and the early e m b r y o g e n e s i s
accomplished before shedding the seeds (Krassilov, 1969).

Coniferidae (coniferids)
Mostly arboreal with pycnoxylic wood, heteroblastic, m i c r o p h y l l o u s , with the pol­
len and seed c o n e s , or the latter only, c o m p o u n d , doubly-strobilate, including conifers
and t w o extinct g r o u p s , cordaites and walchians.

T h e Cordaitales were represented by the mid- to late Paleozoic arboreal or shrubby

plants with simple parallel-veined leaves and with both pollen and seed c o n e s d o u b l y -
strobilate. In the " h y d r o p h i l i c ' ' g r o u p , the stems w e r e creeping, with axillar branching
and adventitious roots, eustelic. T h e primary vascular system w a s s y m p o d i a l , with col­
lateral leaf traces e x t e n d i n g through t w o or three i n t e m o d e s (Rothwell & Warner, 1984).
T h e shoots w e r e heteroblastic, with proximal needle leaves and distal parallel-veined
leaves of the Cordaites type; the leaves w e r e s c l e r e n c h y m a t o u s , h y p o s t o m a t i c , with
tetracytic stomata. T h e c o n e s w e r e e p i c o r m i c , bearing bracteate fertile shoots. T h e
s p o r a n g i o p h o r e s p r o d u c e d terminal clusters of 4-5 tubular s p o r a n g i a axillar to bracts.
T h e p o l l e n g r a i n s w e r e m o n o s a c c a t e , w i t h a distal l e p t o m a . T h e o v u l i p h o r e s w e r e
simple or branched, with terminal ovules. T h e latter w e r e erect, laterally w i n g e d , with
the i n t e g u m e n t 3 - 4 - l a y e r e d , v a s c u l a r i z e d . T h e n u c e l l u s w a s free, w i t h a v a s c u l a r
p l a t f o r m at b a s e ( S e r l i n , 1 9 8 2 a ) .
In the " m e s o p h i l i c " group, the adventitious roots were w a n t i n g , the stems s h o w i n g a
s p a r g a n u m - t y p e s c l e r e n c h y m a . T h e primary vascular system w a s formed of leaf traces
(Trivett & R o t h w e l l , 1985). T h e leaves w e r e of the Cordaites type. T h e s p o r a n g i o p h o r e s
bore four sporangia in rows, p r o d u c i n g the m o n o s a c c a t e prepollen that s h o w e d a dis­
tinct triradiate scar (Taylor & Daghlian, 1980). T h e ovules w e r e p l a t y s p e r m i c , with the
sarcotesta forming a chalazal collar. T h e m e g a s p o r e m e m b r a n e w a s p r o m i n e n t , the early
m e g a g a m e t o p h y t e stages were allegedly free nuclear ( G r o v e & R o t h w e l l , 1980).

T h e Walchiales w e r e the coniferoid mid- to late P a l e o z o i c plants with heteromor-

phic, scaly falcate or needly, pubescent leaves, with the stomata in single or double
rows. T h e pollen strobili were simple, with the sporangial clusters abaxial or whorled
on the stalk. T h e prepollen grains were saccate. T h e seed c o n e s b o r e the axillar shoots
with u p to 30 sterile scale-leaves and the terminal or subterminal abaxial ovules, the
latter platyspermic, mostly solitary, s o m e t i m e s t w o or several per shoot, originally de­
scribed as typically erect, occasionally inverted as in Walchia arnhardtii (Florin, 1938-
1945), but later proved to be typically inverted ( M a p e s & R o t h w e l l , 1984; C l e m e n t -
Westerhof, 1984; M e y e n , 1 9 8 1 , 1984) with a possible exception of Moyliostrobus. The
nucellus w a s free, forming a pollen c h a m b e r with a nucellar beak. T h e m e g a s p o r e m e m ­
brane was cutinized.
Although traditionally assigned to conifers, those plants are here c o n s i d e r e d as a
separate order on account of the prepollen and the leafy ovulate shoots that are widely
held as precursory to the coniferous seed-scales (Florin, 1938-1945), but are m o r p h o ­
logically distinct from the latter. T h e southern Paleozoic coniferids Buriadia and re­
lated genera with furcate leaves and stalked inverted ovules that are solitary on ultimate
or penultimate b r a n c h e s are remotely if at all related to the Walchiales (Pant, 1982).
Ferugliocladus represents another potential Paleozoic order with simple pollen c o n e s ,
m o n o s a c c a t e pollen and the c o m p a c t seed cones bearing erect axillar ovules ( A r c h a n ­
gelsky & C u n e o , 1987).

T h e Coniferales constitute, since the early M e s o z o i c and to this day, the most di­
verse and highly h e t e r o g e n e o u s g r o u p of g y m n o s p e r m s with 57 extant and a c o m p a r a ­
ble n u m b e r of extinct genera belonging to about 15 families. T h e p o p u l a r idea of a
conifer based on the c o m m o n Northern H e m i s p h e r e g e n e r a is that of a tall tree with
conical c r o w n , needle-leaves, simple pollen-cone, bisaccate pollen grains, c o m p o u n d
seed c o n e bearing biovulate seed-scales axillar to bracts, and w i n g e d s e e d s . All these
characters are variable, however. S o m e conifers are giant trees u p to 110 m high, others
are shrubs or even dwarf-shrubs. T h e w o o d is p y c n o x y l i c , the branch s y s t e m s are ra­
dial, less c o m m o n l y flattened, rarely forming pinnate c l a d o d e s . T h e shoots are often
heteromorphic, with the short shoots d e c i d u o u s or s o m e t i m e s persistent, s h e d d i n g their
leaves, c o m m o n l y heteroblastic, with the basal scale-leaves and the distal foliage leaves
having different epidermal structures. Phyllotaxis is helical, less often d e c u s s a t e , in the
latter case with m o r e or less prominently heteromorphic lateral and facial leaves. T h e
leaf traces are single, in the decussate leaves s o m e t i m e s d o u b l e , derived from different
axial s y m p o d i a ( N a m b u d i r i & Beck, 1968). T h e leaves are shortly petiolate or sessile,
often decurrent, s o m e t i m e s basally c o n n a t e with each other and with the axis, rarely
double, as in Sciadopitys, needly or falcate and 3-4-angular in cross-section or flat bifacial,
rarely bilaterally flattened, as in Acmopyle, occasionally forked, single-veined or with
two or m a n y parallel veins. Ontogenetically, linear leaves usually p r e c e d e scale leaves
(Laubenfels, 1953). T h e stomata are mostly perigenous, reportedly also m e s o p e r i g e n o u s ,
or the polar subsidiary cells m e s o g e n o u s (Johnson & Riding, 1981).
T h e c o n e s are n o r m a l l y d i c l i n o u s , but with s p o r a n g i a s o m e t i m e s o c c u r r i n g in seed
c o n e s (Tosh & P o w e l l , 1986). T h e pollen c o n e s are c o m p o u n d in the Triassic Voltzia-
ceae, that are often considered as a stem group, simple or rarely c o m p o u n d in the m o d ­
ern genera in which they are s o m e t i m e s gathered in inflorescence-like aggregates. T h e
s p o r a n g i o p h o r e s are peltate h y p o - or perisporangiate, with the sporangia aligned with a
stalk, p e n d e n t at maturity. T h e s p o r a n g i a are d e v e l o p m e n t a l l y e p i d e r m a l or, in the
Taxodiaceae, C u p r e s s a c e a e , Taxaceae and C e p h a l o t a x a c e a e , h y p o d e r m a l . T h e pollen
grains are sulcate or m o n o p o r a t e , with a distal papilla, rarely inaperturate (Agathis),
occasionally with a triradiate scar and prepollen-like (Cranwell, 1961). T h e pollen wall
infrastructure is typically alveolar in the saccate forms, granular in the asaccate and
some protosaccate forms (Zavada, 1985), occasionally with irregular c o l u m e l l a e (Klaus,
1979). T h e seed c o n e s are terminal or lateral on branches of different orders, occasion­
ally proliferating, usually solitary, but s o m e t i m e s gathered, like pollen c o n e s , in inflo­
rescence-like aggregates. Typically, the c o n e scale consists of a bract and an axillar
seed-scale, the latter bearing inverted ovules. T h e relative d e v e l o p m e n t of the bract in
respect to the seed-scale, the degree of their fusion and the n u m b e r of ovules are taxo­
nomic characters of familial rank, although their alternative states can c o - o c c u r in s o m e
genera, such as Athrotaxis. F o r m s with solitary o v u l i p h o r e s are usually c o n s i d e r e d
destrobilate, yet in a n u m b e r of g e n e r a they can be p r i m a r y (Sinnott, 1913). Erect
ovules, o c c u r r i n g in such forms a l o n e , can be s e c o n d a r y - p s e u d o o r t h o t r o p o u s - c h a n g ­
ing orientation during their d e v e l o p m e n t , as in Dacrydium. In the P o d o c a r p a c e a e , the
ovule is enclosed in the seed-scale derived e p i m a t i u m that s o m e t i m e s fuses with the
integument and, in Dacrycarpus, also with the bract. In Dacrydium, the e p i m a t i u m is
short, arillate and not unlike the arillus of the C e p h a l o t a x a c e a e . In Araucaria, the ovule
is i m b e d d e d in the c o n e scale, with the apex of the seed-scale forming a m e m b r a n o u s
outgrowth, or ligule. In s o m e araucarias, the winged scale with seed in it serves as a
fruit-like dispersal unit.
T h e bract - seed-scale c o m p l e x is typically supplied with t w o g r o u p s of vascular
bundles, with the xylem adaxial in the bract and inverted in the seed-scale, the latter
bundles mostly terminating in the chalaza or, rarely, e x t e n d i n g into the integument, as
in Amentotaxus that has 8-14 integumental bundles (Chen & W a n g , 1985). T h e nucellus
is fused proximally to the integument, less c o m m o n l y free. T h e m e g a s p o r e tetrad (triad)
is linear, s e l d o m tetrahedral. T h e m e g a s p o r e m e m b r a n e is variable, thick in p o d o c a r p s
(Quinn, 1966). At anthesis, the m e g a g a m e t o p h y t e is still nuclear. T h e tent-pole occurs
in the P o d o c a r p a c e a e , C e p h a l o t a x a c e a e and T a x a c e a e . In t h e f o r m e r family, the
archegonia are large, u p to 2 m m , in the latter t w o they are strongly reduced, with their
initials s o m e t i m e s directly functioning as eggs (Georgiev, 1985). In the T a x o d i a c e a e
and C u p r e s s a c e a e , and occasionally in other families, the a r c h e g o n i a form the chalazal
and/or micropylar, as well as lateral, c o m p l e x e s . T h e neck cells are reduced and, in the
latter family, the ventral canal cell tends to be lost. This tendency parallels the reduction
of the prothallial cells. Occasional fusions of the ventral canal nucleus with the egg, as
reported in Keteleeria and Pseudotsuga (Konar & Oberoj, 1969), is of interest as precur­
sory to double fertilization.
 T h e pollination m o d e varies from the o n e c o n v e n t i o n a l for g y m n o s p e r m s to the
extraovular pollen germination, as in a n g i o s p e r m s . T h e micropyle is often secretory,
trapping pollen grains that are then sucked in with pollination d r o p secreted by the
nucellus. In the Pinaceae, the micropyle or one lobe of it is often papillate, stigmoid (the
Narbenmicropyle: Kugler, 1970). Extraovular pollination occurs in the Araucariaceae
and, sporadically, in other families, correlating with erosive s i p h o n o g a m o u s pollen tubes
and with reduced micropyle that leaves the nucellar beak exposed. In Araucaria, the
pollen grains settling on the cone scale concentrate in the m e d i a n g r o o v e and are di­
rected to the ligule, germinating on the m e m b r a n o u s fringe of the latter (Fiordi, 1984).
Both sperms may function in the P o d o c a r p a c e a e and C u p r e s s a c e a e and, less regularly,
in other families, with two zygotes per a r c h e g o n i u m reported in Agathis ( E a m e s , 1913).
Cleavage polyembryony is c o m m o n . T h e p r o e m b r y o cells are tiered. T h e cell walls are
laid d o w n at various stages, in Sequoia at the first division of the zygote. T h e dicotyle­
d o n o u s condition might have been derived by fusion from the 6-8-cotyledonous one
that is considered primitive (Laubenfels, 1953). T h e cotyledons are occasionally lobed,
with t w o veins, in Agathis with many veins like in the foliage leaves.
All attempts at splitting conifers into m o r e natural groups on the basis of c o m p a r a ­
tive morphology alone are deterred by the mosaic c o m b i n a t i o n s of the putatively primi­
tive and advanced characters. Araucaria, for instance, c o m b i n e s the nearly c o m p l e t e
fusion of bracts and their axillar ovuliphores, the adaptations for extraovular pollen
germination and the fruit-like dispersal units with the largest n u m b e r of prothallial cells
(40), the primitive archegonia, and the p r e p h a n e r o g a m o u s shedding of seeds at a fairly
early stage of embryogenesis.
On the basis of fossil links, a n u m b e r of m o d e m families can b e traced back to the
Voltziaceae. T h e latter appeared in the P e r m i a n and diversified in the Triassic. T h e
Permian Pseudovoltzia had dimorphic, scaly and bifacial, leaves and the entire bracts
that were fused basally to the lobed seed-scales, the latter bearing three inverted ovules
(Schweitzer, 1963). T h e seed-scale was supplied with inverted b u n d l e s . It did not look
like a shoot and bore n o r e s e m b l a n c e to the axillar shoots of the c o n t e m p o r a n e o u s
walchias. In the Triassic voltziaceans, either the seed-scale or the bract m o r p h o l o g i ­
cally prevailed in the cone-scale c o m p l e x e s . T h e first appearing forms with needles on
spur-shoots, as well as with lanceolate multinerved leaves, were parallely trending from
seed dispersal to cone-scale dispersal, the latter assisted by the bract modified into a
multinerve wing, as in Cycadocarpidium. T h e Late Triassic - Early Jurassic transitional
forms, such as Schizolepis, Elatides, Doliostrobus, Metaia, etc., link t h e m to the m o d ­
e m Pinaceae, Taxodiaceae and Podocarpaceae (Townrow, 1967; Harris, 1979), while
the coexisting Araucariaceae and Taxaceae were as distant from the voltziacean cluster
as they are today. Their independent origin from different Paleozoic coniferoids thus
seems fairly probable. In the Jurassic, a n u m b e r of conifer lineages u n d e r w e n t parallel
modifications of their seed-scale c o m p l e x e s u n d e r the influence of s e m e n i v o r y and
zoochory. R e m a r k a b l y , flashy seeds a p p e a r e d s i m u l t a n e o u s l y in the conifers, gink­
goaleans and cycadophytes.
 Proangiospermidae (proangiosperms)
T h e g r o u p is defined primarily on the basis of the ovules enclosed in the pistil-like
cupulate structures that are strobilate, apically clustered on dwarf shoots or whorled,
flower-like, typically subtended by perianth bracts. In distinction, the p t e r i d o s p e r m s
have non-strobilate ovuliphores, while in the c y c a d o p h y t e s , the o v u l i p h o r e s are scaly
or, if cupulate, then open, enclosing the ovules at early d e v e l o p m e n t a l stages alone,
with fertilization never occurring inside the c u p u l e s .
T h e scope of the g r o u p is not yet finally determined. T h e bennettites and gnetaleans
are included on account of their derived forms, such as Baisia that is distinctly cupulate
(Krassilov & B u g d a e v a , 1982), or Eoantha having flower-like gynoecial structures
(Krassilov, 1986). W h e t h e r the w h o l e orders or the derived forms alone are proangio-
s p e r m o u s , is left an open question at the m o m e n t .

T h e Bennettitales were d o m i n a n t in the M e s o z o i c , declining through the Late Cre­

taceous u p to the eventual extinction at about the Cretaceous/Tertiary boundary. T h e
early forms w e r e leptocaul, with dichopodial lateral branch s y s t e m s . T h e geologically
y o u n g e r forms were pachycaul as well, columnar, barrel-shaped or tuberous, sparsely
branched, densely covered with multicellular r a m e n t a (Fig. 3). T h e vascular system
was eustelic, the primary x y l e m consisting of helical tracheids, the leaf traces w e r e U-
shaped, divided into collateral b u n d l e s ; the secondary x y l e m was m a n o x y l i c but denser
than in c y c a d s , consisting of tracheids with hexagonal bordered pits. T h e leaves w e r e
simple taenioid, entire or lobed, or pinnate, in the pachycaul forms pinnate, petiolate,
often swollen at the base shed at the abscission layer leaving an a r m o u r of persistent
petioles on the stem. T h e pinnae w e r e spread in o n e plane or imbricate and decurrent,
linear, l a n c e o l a t e or Ungulate, rarely flabellate ( A s h , 1976), s o m e t i m e s with basal
auricules. T h e venation was mostly open, rarely reticulate. T h e cuticle w a s thick, s h o w ­
ing sinuous, rarely straight, cell walls. T h e stomata were paracytic, d e v e l o p m e n t a l l y
m e s o g e n o u s (Krassilov, 1978a, 1984), laterocytic on the leaves, tetracytic on the floral
bracts. Vessels with p o r o u s and scalariform perforations occurred in the leaf bundles
(Krassilov, 1982b).
T h e reproductive organs w e r e flower-like, diclinous or m o n o c l i n o u s , the latter per­
haps p r o t a n d r o u s , d e v e l o p i n g in the forks of the lateral b r a n c h e s or, in the pachycaul
forms, in the leaf axils, i m m e r s e d in the r a m e n t a and supplied with vascular bundles
departing from t w o or m o r e leaf traces (Delevoryas, 1968), typically consisting of a
fleshy receptacle bearing perianth bracts, leafy s p o r a n g i o p h o r e s and a g y n o e c i u m of
stalked ovules intermingled with peltate interseminal scales. T h e receptacle w a s coni­
cal, with a m o r e or less p r o m i n e n t apical c o l u m n ending in a " c o r o n a " of scales (Harris,
1969), s o m e t i m e s e x p a n d e d in a funnel-shaped structure interpreted as a vestigial sec­
ond floral n o d e (Krassilov, 1991). T h e perianth bracts arose in t w o circles and were
densely pubescent, the outer thick and stomatiferous, the inner thin, lacking stomata.
 4«. / 0 t

Fig. 3. Cycadeoideci hikinensis Krassil., a pachycaul bennetlite from the mid-Cretaceous (Albian) of
Sikhote-Alin Mts., Far East, Russia (Krassilov, 1990): (1) stem armoured with persistent leaf petioles,
x,l/3, (2) ramenta of multicellular trichomes, x2()0, (3, 4) leaf trace, cross-section, x l 5 and 80.
perhaps attractively coloured (Harris, 1973), covering young gynoecia and sometimes
even mature gynoecia, as in the cleistocarpous flowers (Gupta, 1958). Occasional bracts
proliferated in miniature pinnate blades betraying their phyllodic origin (Harris, 1969;
Krassilov, 1982b). T h e sporangiophores were basically pinnate, circinnately coiled or
bent, with conduplicate pinnae, often proximally connate and then forming a cup-like
structure, sometimes apically connate in a massive dome-like structure that prevented
their unfolding at maturity (such structures would suggest self-pollination unless they
were shed as a whole). T h e pinnae bore bivalved synangia or the latter developed directly
on the interior wall of the cup, the proximal ones often replaced by glands, these perhaps
secretory, filling the cup with nectar and suggestive of insect pollination. The pollen grains
were large, sulcate, asaccate, with alveolar or granular infrastructure (Taylor, 1973).
T h e g y n o e c i a w e r e d e n s e l y c o v e r e d w i t h i n t e r s e m i n a l s c a l e s , their p o l y g o n a l
apophyses strongly cutinized, adpressed, forming an a r m o u r with small holes through
which the micropyles w e r e protruding. T h e o v u l e s w e r e multiple or, in the aberrant
forms like Vardekloeftia, a few within the interseminal armour, erect on a simple or
occasionally forked stalk, the latter externally like the stalk of an interseminal scale but
supplied with a single vascular b u n d l e scarcely if at all divided in the chalaza, in con­
trast to u p to six vascular bundles in the interseminal scales (Sharma, 1970), radiospermic,
s o m e w h a t flattened, ribbed, s o m e t i m e s arillate, with the arillus pedicellar, scarcely ex­
tending b e y o n d the chalaza, rarely replaced by bristles ( S e w a r d , 1917). T h e integument
was either single, poorly cutinized, or s o m e t i m e s d o u b l e , with the external cover appar­
ently formed of interseminal scales and apically similar to their apophyses (Harris, 1932).
T h e nucellus was free and thickly cutinized, less typically free in the distal third alone.
T h e m e g a s p o r e m e m b r a n e w a s not reported. T h e m e g a g a m e t o p h y t e early stages were
free-nuclear (Crepet & D e l e v o r y a s , 1972). T h e e m b r y o differentiated before shedding
the seeds, with t w o large cotyledons. T h e seeds w e r e s o m e t i m e s provided with a chalazal
or micropylar p a p p u s of long bristles (Krassilov, 1973b, 1982b).

T h e Gnetales include three extant g e n e r a that are fairly insulated, perhaps represent­
ing different orders. A m o n g them, Gnetum is m o s t a n g i o s p e r m - l i k e in the life forms
c o m p r i s i n g tropical lianas and, in a single species, small axillary branched trees. T h e
shoots are d i m o r p h i c heteroblastic. T h e lower leaves are scaly, the foliage leaves are
lauroid, entire or lobed, b o r n e in d e c u s s a t e pairs, but pinnately spread in one plane,
basally c o n n a t e and decurrent, with areolate venation as in typical angiosperms. T h e
vascular system consists of axial and cortical b u n d l e s , with the leaf traces departing
from a half of the axial bundles at each n o d e and e x t e n d i n g through t w o internodes. T h e
foliage leaves are supplied with 3-5 traces forming 5-11 separate bundles that run paral­
lel in the midrib sequentially diverging as lateral veins (Rodin, 1966). T h e tracheary
elements of the primary x y l e m with helical, annular or reticulate thickenings, those of
the secondary x y l e m with bordered pits. Characteristic of the vascular system are the
vessel e l e m e n t s with oblique pergoration plates, p o r o u s or scalariform ( M o u h a m m a d &
Sattler, 1982), and the sieve tubes with likewise o b l i q u e sieve plates, the albuminous
cells resembling the sieve e l e m e n t c o m p a n i o n cells of a n g i o s p e r m s . T h e stomata are
paracytic, reportedly perigenous or m e s o g e n o u s ( M a h e s h w a r i & Vasil, 1961).
T h e reproductive organs are produced in terminal, axillar or, s o m e t i m e s , epicormic
panicles of spicate strobili supplied with the central collateral and the peripheral con­
centric vascular b u n d l e s . T h e spikes are m o n o c l i n o u s , functionally diclinous, bearing
decussate bracts at the n o d e s . T h e bracts, except for the basal ones, are fused in the
sheathing collars. T h e sporangiophores are produced in the nodal zones in several whorls
a b o v e the collar, are followed by a whorl of ovules, and are e n v e l o p e d in the cupular
perianths - occasionally t w o per perianth - protruding at maturity, bearing apical synangia
of four or fewer sporangia. T h e pollen grains are distally leptomate, with a granular-
columellate infrastructure (Orel & K u p r i y a n o v a , 1986). T h e ovulate spikes bear a sin­
gle whorl of ovules in the nodal z o n e axillar to the collar bracts or fused with them,
proximally s o m e t i m e s replaced by sporangia, in Gnetum africanum sunken in the nodal
cushion (Pearson, 1929), supplied with the vascular b u n d l e s departing form the collar
bract traces and additionally from the axial b u n d l e s . T h e ovules are erect, radiospermic,
enclosed in the c u p u l e , or " p e r i a n t h " , that s o m e t i m e s c o n t a i n s t w o ovules or also
sporangia or even an aborted spike ( K e n g , 1965), b i t e g m i c , the inner integument form­
ing a micropyle tube with a fringe, or s o m e t i m e s with t w o fringes, or with hairs over­
hanging the outer integument. T h e integuments d e v e l o p in acropetal succession (Takaso
& B o u m a n , 1986), the inner laid d o w n as an a n n u l a r ridge, the outer as two decussate
knobs. T h e vascular bundles extend to the level of splitting of the nucellus from the
inner integument.
T h e pollen grains s o m e t i m e s g e r m i n a t e in the m i c r o p y l a r cana* that is later sealed
with papillae. T h e m e g a s p o r e tetrad is linear. T h e m e g a g a m e t o p h y t e is tetrasporic or
m o n o s p o r i c , nuclear until a 1000-1500-nucleate stage. T h e archegonial initials seem
directly functioning as egg cells that differentiate simultaneously with the entrance of
u p to four pollen tubes, both their s p e r m s functioning in fertilization, producing two to
six zygotes. T h e cell wall formation p r o c e e d s in acropetal succession and is accompa­
nied by fusion of the nuclei. T h e seeds are e n d o w e d with a fleshy sarcotesta formed of
the cupule (thus n o n - h o m o l o g o u s to those of Ginkgo and c y c a d s ) . T h e e m b r y o pro­
duces t w o cotyledons and an haustorial a p p e n d a g e , or feeder, supplied with two series
of vascular bundles.
Ephedra is an ubiquitous genus of r h i z o m a t o u s shrubs or, in a single species, small
tree, with green branchlets p r o d u c e d yearly, d e c i d u o u s in dry climates. T h e shoots are
articulate, with two, less c o m m o n l y 3-5 nodal leaves, eustelic, with a d o u b l e system of
large and small bundles, the latter disposed m o r e centrally, both coalescing in the nodal
girdle, alternately producing leaf traces, typically t w o per leaf, and the axillar bud bun­
dles a b o v e them. T h e leaves are sheathing, their free blades reduced, or the sheath
reduced, the leaves then spreading directly from the nodes (Pearson, 1929). T h e tracheary
elements are tracheids and vessels with bordered pits and helical to reticulate thickenings.
T h e vessels are tracheid-like, with n u m e r o u s separate or confluent porous perforations.
T h e stomata occur in grooves on the stem, rare on leaves, sunken, with radial subsidiary
cells. T h e strobili are spicate, functionally diclinous, axillar to leaves or terminal. T h e
pollen spikes bear 2-3 bract whorls of w h i c h the p r o x i m a l ones are sterile, the distal
supporting axillar s p o r a n g i o p h o r e s . T h e latter are enclosed in the cupules formed of
t w o bracteoles, often containing t w o or m o r e sporangiophores, their stalks often connate,
s o m e t i m e s forked, bearing 6-8 terminal s y n a n g i a with a p o r o u s aperture. In s o m e spe­
cies, the pollen spikes p r o d u c e ovules functioning as nectaries. T h e pollen grains are
ribbed inaperturate, with the tectum thin b e t w e e n the ribs. T h e pollen wall infrastruc­
ture is granular (Zavada, 1984a).
T h e ovuliferous spikes bear bracts in 2 - 3 ( 4 ) - m e m b e r e d w h o r l s , the distal whorl sup­
porting a few axillar ovules or the ovules are p s e u d o t e r m i n a l , the proximal whorl some­
times containing pollen organs. T h e ovules are either solitary or, occasionally, two per
cupule and a c c o m p a n i e d by sporangia. T h e c u p u l e is formed of 2-4 connate bracteoles,
is supplied by a respective n u m b e r of collateral b u n d l e s and is therefore bilateral and
occasionally bilobed or 3-4-angular. T h e p s e u d o t e r m i n a l c u p u l e b e c o m e s bilateral
through the loss or fusion of the primordial lobes. T h e i n t e g u m e n t (the inner seed coat)
is thin, m e m b r a n o u s , laid d o w n as an annular p r i m o r d i u m , proximally fused with the
nucellus, forming a long micropylar tube. In seed, the c u p u l e hardens, while a fleshy
sarcotesta is formed of the bracts. T h e m e g a g a m e t o p h y t e is m o n o s p o r i c , initially nu­
clear, as in Gnetum, but later forming archegonial c o m p l e x e s , as in conifers (Singh &
M a g e s h w a r i , 1962). A r c h e g o n i a p r o d u c e 3 0 - 4 0 neck cells, as in araucarias. T h e egg
cell and the ventral canal cell are not divided by a cell wall, or the cell wall is reduced.
S p e r m s c o u p l e o n e with the egg, the other with the ventral canal cell nucleus (Herfeld,
1922; Khan, 1940; M o u s s e l , 1978). T h e e m b r y o g e n e s i s starts as free-nuclear. T h e coty­
ledons are two or three.
Welwitschia is a m o n o t y p i c genus of a peculiar pachycaul life form with an obconical
stem and t w o ribbon-shaped leaves arising from the rim of a broadly c o n c a v e apex. T h e
vascular system is segmented protostelic, with the partial stelae differentiated as leaf
traces alternating with the bundles that terminate in the annular ridge of the apex. The
venation is parallel "goniopteroid", with the short side b r a n c h e s coalescing to give rise
to interstitial veins (Takeda, 1913). T h e stomata occur on both sides, are sunken, paracytic,
developmentally m e s o g e n o u s or m e s o p e r i g e n o u s . T h e reproductive shoots are dichasial,
borne on the annular ridge, their terminal branches are spicate, with empty connate bracts
at the base. T h e pollen spikes bear flower-like spikelets axillar to bracts, or glumes. The
spikelets are radial but slightly c o m p r e s s e d . T h e perianth consists of t w o decussate pairs
of bracteoles the inner of which are c o n n a t e to half their length. T h e y form an open
cupule that contains six sporangiophores arranged radially around the apical ovule, the
latter extended into a papillate corona, p e r h a p s glandular. T h e sporangiophore stalks are
connate basally, bent toward the central ovule, each bearing 3-locular synangia. T h e
pollen grains are ribbed, as in Ephedra, but with a single or several colpous apertures
between the ribs. T h e ovulate spikes bear large spoon-like bracts subtending the cupular
ovules with rudimentary l^racteoles at the base. T h e c u p u l e is supplied with two vascular
bundles forming a chalazal ring. T h e micropyle tube is long, fragile. T h e axillar c o m -
plex develops into a samara with m e m b r a n o u s lateral w i n g s . T h e m e g a g a m e t o p h y t e is
m o n o s p o r i c or tetrasporic, nuclear until about a 1000-nucleate stage, then divided into
multinuclear cells in which multiple nuclei fusions take place (Martens, 1971). T h e micro­
pylar cells form the ascending tubes that meet pollen tubes. Cell walls are laid down after
the first division of the zygote. T h e e m b r y o produces a haustorial feeder, like in Gnetum,
and two unequal cotyledons with the axillar buds that overtop the primary apex.
Typical gnetalean characters a p p e a r in the Triassic genera Dinophyton and San-
miguelia that can be considered as protognetalean. T h e y can then be traced through the
Jurassic Heerala to the even m o r e definitely gnetalean Crttactous Leongathia, Drewria
and Eoantha and the recently found pollen inflorescences of the Baisian A s s e m b l a g e .
T h e s e p r a a n g i o s p e r m o u s plants will be described in the next chapter. G n e t o p h y t e s are
thought to be represented by the diverse polyplicate " e p h e d r o i d " pollen grains of the
Ephedripites, Eqidsetosporites, Singhia, Steevesipollenites, Gnetaceaepollenites and
other types (Osbon & al., 1993). H o w e v e r such pollen grains w e r e produced also by the
p o l l e n o r g a n s of the t y p i c a l l y c y c a d a l e a n m o r p h o l o g y , such as Piroconites (Van
Konijnenburg-van Cittert, 1992). On the other hand, the typically gnetalean androclades,
such as Baisianthus, might p r o d u c e entirely different pollen grains (see b e l o w ) .

T h e Caytoniales constituted a small but widespread g r o u p of extinct M e s o z o i c plants

reconstructed as trees or shrubs with thick heteroblastic shoots bearing scaly cataphylls
and c o m p o u n d foliage leaves. T h e latter w e r e petiolate, on p r o m i n e n t leaf cushions,
abscissed at the base of the petiole (Harris, 1951a, 1964), with four, occasionally two,
leaflets that spread palmately, but not exactly in o n e p l a n e and could be separately
abscissed. T h e t w o m e d i a n leaflets w e r e s o m e w h a t larger than the lateral pair, the ar­
rangement suggesting t w o decussate pairs. T h e leaflets w e r e lanceolate, somewhat asym­
metrical, with the lamina unilateral at the base. T h e venation w a s reticulate, with the
midrib consisting of separate parallel bundles that successively diverged as lateral veins.
T h e reticulations were formed of coalescing terminal b r a n c h e s of the repeatedly forked
lateral veins or by short transverse a n a s t o m o s e s , the areoles decreasing toward the mar­
gins. T h e stomata were fully e x p o s e d , with the guard cells not sunken below the level of
anomocytic subsidiary cells. T h e pollen organ w a s a dichopodially branched androclade,
with the lateral branches forked, bearing elliptical s y n a n g i a terminally in sessile radial
clusters or laterally on short stalks twisted to o n e side. T h e synangia were radially sym­
metrical, consisting of four tubular sporangia, occasionally bilateral, with the sporangia
unequally developed (Krassilov, 1977b), adnate from end to end but splitting in the
m i d d l e at m a t u r i t y , d e h i s c i n g by an a d a x i a l slit. T h e p o l l e n g r a i n s w e r e m o n o -
protosaccate, with a bilobed saccus. T h e pollen wall infrastructure was alveolate (Zavada
& Crepet, 1986). T h e g y n o c l a d e s w e r e pinnately b r a n c h e d with the lateral ovuliphores
bent toward one side, consisting of a short stalk bearing a spherical cupule. T h e latter
was fleshy, strongly cutinized, lacking stomata, vascularized, with the tracheary ele­
ments s h o w i n g bordered pits ( R e y m a n o w n a , 1974). A slit-like e x o s t o m e occurred at
the base of the cupule, w a s subtended by a short a p p e n d a g e ("lip"), and w a s connected
to the ovules by a series of channels ( T h o m a s , 1925). T h e ovules w e r e n u m e r o u s , clus­
tered at the base of the cupule, with their m i c r o p y l e s facing the e x o s t o m e , bitegmic,
with the inner integument not reaching the m i c r o p y l e (Krassilov, 1984a). T h e nucellus
w a s free, s t r o n g l y c u t i n i z e d . T h e m e g a s p o r e m e m b r a n e w a s n o t o b s e r v e d . T h e
e m b r y o g e n e s i s w a s well advanced before the seeds w e r e shed (Krassilov, 1977b).

T h e C z e k a n o w s k i a l e s (Leptostrobales) w e r e a d o m i n a n t g r o u p of M e s o z o i c plants
in northern Asia, (formerly assigned to g i n k g o p h y t e s ) arboreal with d i m o r p h i c shoots.
The spur-shoots w e r e tuberous, d e c i d u o u s , bearing scaly cataphylls and linear foliage
leaves. T h e latter w e r e ribbon-shaped, parallel veined, bifacial, h y p o s t o m a t i c , in the
Phoenicopsis g r o u p s o m e t i m e s a m p h i s t o m a t i c , forked, single-veined, bifacial or, in the
Czekanowskia g r o u p , s o m e t i m e s triangular in cross section, with the s t o m a t a on three
facets. T h e stomata w e r e tetracytic or cyclocytic, with the subsidiary cells petalocytic,
unequal, strongly cutinized, often papillate. T h e pollen organs w e r e catkin-like strobili,
with the lateral s p o r a n g i o p h o r e s arranged spirally or p s e u d o w h o r l e d , consisting of a
slender, apically reflexed stalk bearing an adaxial s y n a n g i u m . T h e latter w a s 4-lobed,
spreading at d e h i s c e n c e , each lobe vascularized by a branch of a single pedicellar bun­
dle. T h e g y n o c l a d e s w e r e terminal on short shoots, with scale leaves at base, loose,
branching spirally, with the lateral ovuliphores distant or o v e r l a p p i n g , consisting of a
stalk and a terminal bivalved cupule (Harris, 1951b; Krassilov, 1969). T h e valves w e r e

Fig. 4. Czekanowskialean reproductive structures, from left to right: Leptostrobus gynoclade, bivalved
cupule and valve showing a stigmatic marginal flange and ovules, Ixostrobus androclade and synangium
(after Krassilov, 1969, 1982).
gaping at maturity, externally ribbed and strongly cutinized, with stomata, internally
thinly cutinized, with a broad sutural flange. In the a d v a n c e d species, the latter w a s
internally papillate, apparently stigmatic (Fig. 4 ) . T h e ovules w e r e 5-8 p e r valve, sub-
marginal, inverted, shed from the ripe c u p u l e s .

Selected fossil links

P h y l o g e n e t i c links are the forms either s h o w i n g intermediate character states in the
character polarity clines or c o m b i n i n g the characters typical of the taxa that replace
each other in a geochronological series, or c h r o n o c l i n e . T h e latter forms are d u e to
mosaic evolution, that is, the discrepant evolution rates of different functional units. A
few such fossil links are selected for further discussion of the g y m n o s p e r m p h y l o g e n y
and of the origin of p r o a n g i o s p e r m s .

Cycandra Krassilov & Delle in Krassilov & al., 1996; M i d d l e Jurassic of the C a u c a ­
sus (Plates 1,2).
T h e g e n u s is based on the pollen c o n e s that are about 2 0 c m long, c o m p a c t , bearing
spirally arranged, imbricate, peltate sporangiophores. T h e peltae of the latter are rhombic,
with a sterile marginal flange that abaxially s h o w s cyclocytic stomata and is sharply
delimited from the fertile central part, the latter containing a large adaxial s y n a n g i u m of
multiple, concentrically arranged, vertically aligned sporangia. T h e s y n a n g i u m is cov­
ered with a delicate cuticle the p r e s e n c e of which is indicative of a h y p o d e r m a l devel­
o p m e n t of the sporangia. On the surface, the s y n a n g i u m s h o w s adpressed prismatic
sporangial heads alone. T h e sporangia, exposed by maceration, are tubular, laterally
adnate and proximally fused. T h e sporangial wall consists of t w o cutinized layers, the
outer e p i d e r m a l , forming a thickened apical head traversed by a short o p e n i n g slit, the
inner endothecial, with an apical beak often filled with pollen grains that are p r o d u c e d
in great quantities, passing through the bottle-neck-like sporangial beak before disper­
sal. T h e pollen grains are anasulcate, asaccate, with alveolar infrastructure.
A l t h o u g h r e s e m b l i n g the pollen c o n e s of c y c a d s and nilssonias (Androstrohus),
Cycandra differs in the laterally adnate sporangia forming a giant s y n a n g i u m the con­
centric structure of which reminds of synangial structures occurring in the a d v a n c e d
medullosalean pteridosperms. T h e dehiscence m e c h a n i s m of the sporangia, with an apical
slit and an endothecial beak u n d e r it, is u n i q u e in c y c a d o p h y t e s s u p p o s e d l y m e d i a t i n g a
gradual dispersal of pollen grains from the long-living giant pollen c o n e s .

Semionogyna K r a s s i l o v & B u g d a e v a , 1 9 8 8 c ; L o w e r C r e t a c e o u s of T r a n s b a i k a l i a
(Plate 3).
T h e g e n u s is erected for the large spicate g y n o c l a d e s with a stout axis b e a r i n g ovules
subtended by bracts, the latter arranged spirally, spreading radially, n a r r o w lanceolate,
e x p a n d e d and decurrent basally, adhering to the ovules. Several distal bracts are empty.
Fig. 5. Semionogyna hracteata Krassil. & Bugd., a protocycadean plant from the Early Cretaceous of
Transbaikalia: ovulate spikes with juvenile and ripe ovules subtended by the bracts, with distal bracts
empty, compared to extant Cycas revoluta (right), (after Krassilov & Bugdaeva, 1988c).

Young ovules are shorter than bracts and are protected by the bract bases, protruding at
maturity, with the bracts pendent and eventually shed. T h e ovules are large, sessile,
erect, r a d i o s p e r m i c . T h e integument is thick, strongly cutinized, with stomata. T h e
nucellus is free, cutinized. T h e associated pollen c o n e s are loose, bearing spirally ar­
ranged s p o r a n g i o p h o r e s that are stalked, each with a head of the proximally fused
sporangia. T h e associated leaves are 2-pinnate, with interstitial pinnules.
In this g e n u s , the ovules are definitely c y c a d e o i d . In extant the ovules s o m e ­
times s h o w basal a p p e n d a g e s (Sabato & D e Luca, 1985) that m a y c o r r e s p o n d to ves­
tigial subtending bracts, while the " p i n n a e " of the leafy apex could be formed of e m p t y
distal bracts (Fig. 5). At the s a m e time, the seed c o n e s r e s e m b l e Sobernheimia, a rare
m o r p h o t y p e of the Permian ovuliferous organs associated with callipterids (Kerp, 1983).
In addition, the associated pollen cones and leaves are similar to those of the typical
callipterids.

Dirhopalostachys P r y n a d a ex Krassilov, 1975a; L a t e Jurassic of the B u r e y a Basin,

Far East, Russia (Plate 4 ) .
T h e g y n o c l a d e s are large, r a c e m o s e , with the lateral o v u l i p h o r e s spirally arranged,
consisting of a stalk and a pair of terminal cupules. T h e latter are c o n n a t e at base,
distally spreading at an acute angle, follicular, laterally c o m p r e s s e d , with a long, reflexed
beak and with the abaxial suture extending along the beak. T h e c u p u l e is ribbed, with
the ribs diverging from the suture, the c u p u l a r wall is pubescent, containing resin ducts,
lacking stomata. T h e beak is ribbed, abaxially pitted, with the suture m a r k e d by a m e ­
dian g r o o v e , gaping in the ripe cupules (Fig. 6). T h e ovules are solitary, filling the
Fig. 6. Dirhopalostachys rostrata Krassil., a proangiosperm from the Late Jurassic of Burya Basin, Far
East, Russia: gynoclade bearing paired cupules (right) and split cupules showing a soliltary anatropous
ovule and a rugulate beak (after Krassilov, 1975) compared with a supposedly progenitorial Beania
gynoclade (left, after Harris, 1964).
locule except in the basal portion, inverted, platyspermic. T h e m i c r o p y l e is short, indis­
tinctly lobed. T h e integument is single, glabrous, thinly cutinized. T h e nucellus is free.
T h e m e g a s p o r e m e m b r a n e was not observed. T h e shed c u p u l e s are empty, filled with
rock matrix. T h e pollen grains sticking to the c u p u l e are anasulcate, asaccate, of the
Cycadopitys type. T h e associated leaves are simple, lobed, of the Nilssonia type.
T h e paired c u p u l e s oiDirhopalostachys s e e m derivable from the peltate ovuliphores
of the Nilssoniales {Beania) by involution of the peltae bearing t w o inverted ovules
(Krassilov, 1975a). At any rate, they e v i d e n c e the possibility of the involute follicular
cupules arising in c y c a d o p h y t i c plants. W h e t h e r the pollen grains entered the ovules or
germinated on the pitted beak surface, is still to be learned, but the latter possibility
appears m o r e likely b e c a u s e the suture of the beak w a s sealed in u n r i p e c u p u l e s . Exter­
nally at least, the paired cupules r e s e m b l e the g y n o e c i a of h a m a m e l i d a n g i o s p e r m s and,
in particular, the C r e t a c e o u s Trochodendrocarpus (see b e l o w ) .

Irania Schweitzer, 1977; Early Jurassic of Iran.

T h e fertile shoot bearing proximal pollen strobili and a distal g y n o c l a d e . T h e pollen
strobili are c o m p a c t , consisting of peltate s p o r a n g i o p h o r e s with t w o or m o r e sporangia
aligned with the stalk. T h e ovuliphores are distant, stalked, bearing a terminal bilobed
cupule, the latter dehiscing into t w o valves with a flat marginal flange (Fig. 7).
Irania is interesting as an early e x a m p l e of m o n o c l i n o u s structure, although the in­
terpretation is not yet conclusive. Probably related to Irania are the Early C r e t a c e o u s
Ktalenia g y n o c l a d e s bearing bracteate cupules (Taylor & Archangelsky, 1985).

Fig. 7. Irania, a Jurassic proangiosperm: reconstruction of a fertile short shoot bearing a central gyno­
clade and peripheral androclades, and part of the gynoclade with bivalved cupules (modified from
Schweitzer, 1977).
 Baisia Krassilov in Krassilov & B u g d a e v a ,
1982; Baisian A s s e m b l a g e , Early C r e t a c e o u s
of Transbaikalia (Plates 5-7).
The genus comprises dispersed cupulate
ovules that are apical on a short rounded-tri­
angular receptacle that is persistent in dissemi-
nules, and is bearing and is m i n u t e bracts and
tufts of long bristles (Fig. 8). T h e c u p u l e is
flask-shaped, apically t h i c k e n e d in a hairy c o ­
rona, thick, utriculate, with the exterior epi­
dermal cells large, thick-walled, tabulate,
scarcely cutinized, the stone cells large, p o ­
rous, and the interior e p i d e r m a l cells thinly
cutinized, papillate. T h e ovule is solitary, erect,
basal, not filling the locule. T h e i n t e g u m e n t is
thin, cutinized, forming a short m i c r o p y l e . T h e
nucellus is free, with a low chalazal platform.
T h e pollen c h a m b e r is scarcely differentiated.
T h e pollen grains trapped in the nucellar beak
are anasulcate, asaccate, with alveolar infra­
structure.
T h e t a p h o n o m y oiBaisia that c a m e in large
Fig. 8. Baisia hirsuta Krassil., a proangio-
n u m b e r s from lacustrine deposits suggests an
spermous plant of bennettitalean affinities
of the Baisian Assemblage, Early Creta­ aquatic or water-side plant dispersing the wind
ceous of Transbaikalia, reconstruction (af­ and/or w a t e r - b o r n e " a c h e n e s " w h i l e its veg­
ter Krassilov & Bugdaeva, 1982). etative parts scarcely reached the deposition
site. T h e most probable candidate for the leaves
is ''Dicotylophyllum" pusillum (Vakhra-meev & Kotova, 1977), small, petiolate, with a
lanceolate blade and with irregular e u c a m p t - o d r o m o u s venation. A n additional speci­
men recently found by Dr. E u g e n i a B u g d a e v a (Fig. 9) is m o r e bract-like, thick with an
indistinct i m m e r s e d lateral veins, pubescent and c o m p a r a b l e to bennettitalean bracts,
except that the venation m a k e it angiosperm-like.
T h e baisian d i s s e m i n u l e s are superficially similar to c y p e r a c e o u s a c h e n e s , the anal­
ogy e x t e n d i n g not only to the tufts of bristles but also to the large t a b u l a r surface
cells. H o w e v e r , the pollination w a s o b v i o u s l y g y m n o s p e r m o u s . T h e p e r s i s t e n t hairy
receptacles w e r e c o m m o n in b e n n e t t i t e s in s o m e l i n e a g e s of w h i c h the n u m b e r of
o v u l e s w a s r e d u c e d to several or e v e n t w o , the o v u l e s s o m e t i m e s e n v e l o p e d in a
c u p u l e formed of interseminal scales. Baisia m i g h t h a v e been an e x t r e m e form with a
single c u p u l a t e o v u l e d e v e l o p i n g from the apex of the r e c e p t a c l e . T h e vestigial c o ­
rona, the structure of the o v u l e s , and the a s s o c i a t e d pollen g r a i n s are definitely in
favour of the bennettitalean affinities.
Fig. 9 Angiosperm leaves from the Baisian Assemblage, Early Cretaceous of Transbaikalia: (1)
"Dicotylophyllum"pusillum Vachr. (Vakhrameev & Kotova, 1977), x3 (courtesy of A.B. Herman), (2-4)
another specimen found by Dr. Eugenia Bugdaeva, perhaps of the same species, but more bract-like (2),
x2, with indistinct oblique lateral veins (3), x4, and round structures (hair bases?) on the coating mineral
film (4), SEM, x63().
 Dinophyton A s h , 1970, e m e n d . Krassilov & A s h , 1988; L a t e Triassic of A r i z o n a
(Plates 8, 9).
T h e shoots are copiously b r a n c h e d , heteroblastic, with scaly and n e e d l y leaves, the
latter decurrent, round in cross-section, occasionally forked or even p i n n a t e , p u b e s c e n t
and papillate, with t w o veins c o n v e r g i n g toward the apex. O c c a s i o n a l s p e c i m e n s with
involute pinnae-like leaves suggesting a transitional c a u l o m e / p h y l l o m e n a t u r e of the
leafy shoots (Fig. 10). T h e stomata o c c u r both on the shoot axis and on leaves, are
sunken, mostly tetracytic. T h e pollen c o n e s are small, with four cross-like bracts at the
base and with a few s p o r a n g i o p h o r e s e n c a s e d in bracteoles, the latter p u b e s c e n t and
with stomata as on leaves. R i p e s p o r a n g i o p h o r e s are protruding, shortly stalked, con­
cave, with a serrate m e m b r a n o u s fringe o n e cell thick. T h e y b e a r a solitary adaxial

Fig. 10. Dinophyton spinoswn Ash, a protognetalean plant from the Late Triassic of Arizona, U.S.A.: (1,
2 ) juvenile and ripe cupules with bracts, ( 3 ) ovule with a trilobed nucellus, (4) involute phyllomic struc­
ture, (5) pollen cone with bracts, (6) sporangiophore (after Krassilov & Ash, 1988).
s p o r a n g i u m or a g r o u p of three apparently fused sporangia that are basifixed, dehiscing
in t w o valves. T h e pollen grains are bisaccate. T h e ovuliphores are 4-lobed pedicellate
cupules borne on shoots of different orders, with a 4-lobed cross-like perianth formed of
two decussate pairs of bracts and fused to the base of the cupule, persistent as a propeller
in the fruit-like diaspora. T h e cupule lobes are imbricate over a slit-like e x o s t o m e , spread­
ing at maturity, pubescent, with the hairs longer and tufted on tips, cutinized, s h o w i n g
irregular-polygonal cells, lacking stomata, each receiving a single vascular bundle. T h e
propeller bracts are flat, broadly attached or slightly constricted at the base, tapering
toward the apex, less strongly pubescent and with the cuticle thicker than on the cupule,
showing rectanguloid cells arranged in longitudinal r o w s . T h e interior cuticle of the
cupule is thin with distinctly m a r k e d , large, polygonal cells. T h e ovule is basal, erect, 3 -
keeled, not filling the locule. T h e integument is scarcely cutinized, the nucellus is free at
the m i c r o p y l a r end alone, forming a broad pollen chamber. T h e m e g a s p o r e m e m b r a n e is
thick, s h o w i n g a tetrad mark.
A p h y l o g e n e t i c significance of Dinophyton lies in the c o m b i n a t i o n of the gnetalean
and p t e r i d o s p e r m o u s characters. T h e former include the d e c u s s a t e perianth bracts both
in the pollen c o n e s and seed organs and the 4-lobed bracteolate c u p u l e containing a
solitary erect ovule, w h e r e a s the structure of the latter is like in the h y d r o s p e r m a l e a n
Lyrosperma, and the leafy shoots show transitional c a u l o m e - p h y l l o m e characters, as in
the c a l a m o p y t i a c e a n foliar branch systems.

Sanmiguelia R.W. B r o w n , 1956, e m e n d . C o m e t , 1986; L a t e Triassic of Arizona.

R h i z o m a t o u s shrubs with slender aerial stems b r a n c h i n g by axillary b u d s , eustelic,
producing a relatively thick secondary w o o d . T h e tracheary e l e m e n t s with helical or
reticulate thickenings. T h e shoots are heteroblastic, with spirally arranged cataphylls
and foliage leaves, the latter u p to 3 0 c m long, s h e a t h i n g , plicate, with n u m e r o u s ,
subparallel, thick and thin veins c o n v e r g i n g toward the apex, a n a s t o m o s i n g , looping or
the thin veins c o a l e s c i n g to restore a thick vein, with resin ducts in the sheaths. T h e leaf
cuticle is thin, s h o w i n g a n o m o c y t i c or cyclocytic stomata. T h e pollen c o n e s of spirally
arranged s p o r a n g i o p h o r e s with bisporangiate s y n a n g i a dehiscing by a longitudinal slit.
The p o l l e n g r a i n s are a n a s u l c a t e , with g r a n u l a r i n f r a s t r u c t u r e . T h e s e e d organ is
paniculate, bearing terminal clusters of bracteate c u p u l e s . T h e bracts are decussate,
adnate to the c u p u l e s , h e t e r o m o r p h i c , t h e l o w e r p a i r d i s s e c t e d , t h e u p p e r p a i r e n t i r e ,
p u b e s c e n t . T h e c u p u l e is b i l o b e d a p i c a l l y a n d with t w o lateral p o c k e t - l i k e a p p e n d ­
ages, b i o v u l a t e . T h e o v u l e s are b a s a l a n a t r o p o u s , s o m e c o n t a i n i n g e m b r y o with t w o
heteromorphous cotyledons.
Sanmiguelia is r e m a r k a b l e for a m o n o c o t aspect of its plicate leaves and paniculate
inflorescences. T h e " g o n i o p t e r o i d " venation pattern, as in Welwitschia, the decussate
pericupular bracts, the pocket-like bracteoles (?), and the c u p u l e m o r p h o l o g y suggest a
gnetalean affinity which is, however, inconsistent with the allegedly a n a t r o p o u s ovules.
The alternation of large and small vascular bundles in the apparently p h y l l o d i c leaves
finds its analogy in the axial vascular system of Ephedra.
 Himierella Hoerchammer, 1933, emend. Jung, 1967; Jurassic of Central Europe (Plate 10).
T h e generic n a m e was corrected by J u n g (1967) from a grammatically incorrect
Hirmeriella: the preoccupied Cheirolepis S c h i m p e r and the substituting Cheirolepidium
Takhtajan are s y n o n y m o u s to HinnerellcL
Trees or shrubs, s o m e t i m e s preserved in situ in littoral deposits ( O l d h a m , 1976;
Francis, 1983). T h e w o o d is coniferoid, with pinoid or cupressoid pitting of tracheary

Fig. 11. ClassopoUis pollen grains from the guts of Jurassic katydid insects, SEM (Krassilov & al.,
1997): (1) proximal aspect with iriradiate scar showing exinal filaments that bind the grains in tetrad,
x3 ()()(), (2) distal aspect with rimula and pore, x3 ()()(), (3,4) infratectal structures, x6 ()()() and 7 700.
elements that are s o m e t i m e s " s e p t a t e " (Alvin & al., 1978). B r a n c h i n g is axillar. T h e
leaves are scaly, s o m e t i m e s d i m o r p h i c , spiral or decussate or verticillate, decurrent or
sheathing, s o m e t i m e s d e c i d u o u s or the short shoots (branchlets) abscissed, x e r o m o r p h i c
with a thick cuticle, often p u b e s c e n t and papillate. T h e s t o m a t a are sunken, cyclocytic.
T h e pollen c o n e s are bearing peltate s p o r a n g i o p h o r e s , with sporangia 6-8 in pendent
clusters, s o m e t i m e s synangiate (Archngelsky, 1986a). T h e pollen grains {Classopollis)
s o m e t i m e s in p e r m a n e n t tetrads, interconnected by sexinal filaments, with an equatorial
sac-like girdle, a distinct triradiate mark, an incipient subequatorial z o n o s u l c u s (rimula),
and a distal p o r o u s l e p t o m a (Taylor & Alvin, 1984), the latter s h o w i n g a m i c r o p o r o u s
m e m b r a n e and p e r h a p s aperturate (Krassilov & al., 1997a). T h e pollen wall infrastruc­
ture is columellate-rugulate, s o m e t i m e s granulate (Fig. 11). T h e seed c o n e s with bracts
subtending axillar o v u l i p h o r e s that w e r e usually considered as seed-scales but were
reinterpreted by Krassilov (1982a) as lobed cupules with adnate lateral bracteoles. T h e y
are basally attached to the bracts, but are shed separately as s a m a r a s , with the bracts
remaining on the axis (Fig. 12). T h e c u p u l e s are thickly cutinized, with stomata as on
leaves. T h e locule is lined with a thin cuticle s h o w i n g sinuous cell w a l l s . T h e o v u l e is
solitary, erect, adpressed to the locule. T h e integument is thinly cutinized. T h e nucellus
is free, c u t i n i z e d , c o n t a i n i n g a thick m e g a s p o r e m e m b r a n e or, o c c a s i o n a l l y , t w o
m e g a s p o r e s . P r o b a b l y e n t o m o p h i l o u s , with the pollen g e r m i n a t i n g on the c u p u l e .

Fig. 12. Gnetophyte seed-bearing structures: (1,2) Hirmerella, a Mesozoic gnetophyte, seed cone and a
winged samaroid cupule axillar to bract, (3,4) samaroid bracteate cupules of extant Welwitschia and the
Jurassic Heerala -Angarolepis plant (reconstruction based on dispersed samaras and bracts: Krassilov &
Bugdaeva, 1988a).
 Hirmerella is usually assigned to conifers and is actually similar to the P e r m i a n
coniferoid g e n u s Majonica (Clement-Westerhof, 1987), but the c u p u l e s s h o w an inte­
rior cuticle, thus differing from epimatia of conifers with e m b e d d e d o v u l e s . O n the
other hand, the bracteate c u p u l e s with adnate lateral bracteole-like a p p e n d a g e s that are
transformed into w i n g s in the samaras r e s e m b l e those of Welwitschia ( s o m e t i m e s also
having t w o functional m e g a s p o r e s per nucellus). T h e d e c u s s a t e s h e a t h i n g leaves and
the " s e p t a t e " tracheary e l e m e n t s are additional gnetalean characters. T h e pollen grains
are r e m a r k a b l e in s h o w i n g incipient a n g i o s p e r m o i d structures.

Heerala Krassilov & B u g d a e v a , 1988a; M i d d l e Jurassic of Siberia (Plates 11,12).

T h e genus is based on Ephedrites antiquus Heer, 1978, the large s a m a r a s with a
bilobed m e m b r a n o u s w i n g , a spindle-shaped central b o d y with c o n s p i c u o u s resin ducts,
a median suture line and a callous scar at the base. T h e associating s p o o n - l i k e bracts
with identical resin ducts might h a v e been attached at the basal scar. T h e y are petiolate
or sessile, bluntly pointed or a c u m i n a t e , s h o w i n g cyclocytic stomata, s o m e t i m e s bear­
ing a triangular a p p e n d a g e , probably the attached prophyll of an axillary ovuliferous
shoot. T h e bracts and samaras are like those of Welwitschia, also h a v i n g c o n s p i c u o u s
callous attachment scars at the base. E v e n the blunt and a c u m i n a t e bract variants have
their a n a l o g u e s in the respectively distal and proximal bracts of Welwitschia. T h e asso­
ciated pollen cones are bearing spirally arranged peltate sporangiophores with the radially
arranged sporangia adpressed to the peltae, dehiscing by an apical p o r e or a short slit.
T h e pollen grains are anasulcate asaccate. All that is presently k n o w n of the vegetative
parts are the detached dagger-shaped leaves that are basally e x p a n d e d , apparently sheath­
ing. T h e leaf blade is m e m b r a n o u s , with a thickened m e d i a n z o n e , variably supplied by
1-3 veins as in extant Ephedra chinensis (Foster & Gilford, 1974).

Drewria C r a n e & U p c h u r c h , 1987; Early C r e t a c e o u s of eastern N o r t h A m e r i c a .

B r a n c h i n g shoots with decussate sheathing leaves. T h e venation is " g o n i o p t e r o i d " .
T h e seed c o n e s are b o r n e in terminal or lateral dichasia, spicate. T h e o v u l e s are sub­
t e n d e d by a p a i r of b r a c t e o l e s . T h e associated d i s p e r s e d p o l l e n g r a i n s a r e r i b b e d ,
ephedroid.
Drewria s h o w s a r e m a r k a b l e c o m b i n a t i n of a h e r b a c e o u s l i f e - f o r m w i t h t h e
Welwitschia-Wko, leaf venation and g y n o c l a d e m o r p h o l o g y .

Leongathia Krassilov, Dilcher & D o u g l a s in Krassilov & al., 1997b; K o o n w a r r a

Fossil Bed A s s e m b l a g e , Early C r e t a c e o u s of Australia (Plates 13-15).
Slender articulate shoots with the distal i n t e m o d e s m u c h shorter than the proximal
and the distal leaf whorls c r o w d e d or even conjugate. T h e shoot axis is longitudinally
ribbed, with a transverse rugulate structure reflecting the s c l e r e n c h y m o u s fibres. T h e
nodes are typically 4-leaved, occasionally 2 - 3 - leaved, with u p to 8 leaves in the termi­
nal whorls. T h e leaves are decussate, spreading at acute angles or a l m o s t horizontally or
pendent, lanceolate, acute, about 2 m m long, c o n n a t e at the base, sheathing a b o u t 1/3 to
Fig. 13. Baisianthus. a gnetalean androclade of the Baisian Assemblage, Early Cretaceous of Transbaikalia
(Krassilov & Bugdaeva, 1997): ( 1 , 2 ) branching shoot bearing successive whorls of the distally crowded
cupulate sporangiophores on the central axis and a terminal whorl on a side branch, with linear bracts at
the nodes, x8 and 10, (3) shoot fragment with two sporangiophore whorls, x7, (4) axis with tendril-like
appendages found with androclades, possibly of the same plant, x20.
 1/2 of internode or with the sheaths reduced and then directly inserted u p o n the n o d e .
T h e stomata are densely c r o w d e d in the g r o o v e s b e t w e e n the ribs, sunken, encircled by
a thick annular ridge. T h e cortex traversed with a s c l e r e n c h y m o u s n e t w o r k . T h e vascu­
lar system consist of alternating large and small b u n d l e s . T h e leaf traces are paired,
departing from the nodal girdles. T h e tracheary e l e m e n t s are pitted, with reticulate and
helical wall thickenings. T h e pits o c c u r in t w o r o w s or occasionally c r o w d e d and s o m e
confluent forming p o r o u s perforations at the poles.
This fossil is the first evidens of the ephedroid gnetaleans existing in the Early C r e ­
taceous. It deffers from extant Ephedra in m o r e n u m e r o u s leaves per n o d e and in the
leaf to vascular bundle ratios.

Baisianthus Krassilov & B u g d a e v a , 1897; Baisian A s s e m b l a g e , Early C r e t a c e o u s of

Transbaikalia (Plates 16-19).
Branched androclades bearing flower-like whorls of bracteate c u p u l e s that are sub­
tended by bracts. T h e cupules enclose sporangiophores that are bearing synangia on the
short lateral branches and terminally. T h e synangia are 2-4-sporangiate, s h o w i n g vari­
ous degrees of sporangial fusion, the terminal sporangia s o m e t i m e s free, p r o d u c i n g
monosulcate pollen grains in tetrahedral tetrads. T h e pollen grains are fusiform, cavate,
with sexine loosely attached to the nexinal body and easily separable. T h e tracheary
elements of the axis are septate, with pits and helical thickenings that are occasionally
transformed into oblique perforation plates.
T h e s e organs are basically similar to the androclades of Gnetum, but of a peculiar
branching habit, with the fertile whorls m o r e flower-like and with the pollen grains
perhaps derived from a m o n o s a c c a t e m o r p h o t y p e . T h e associating vegetative organs
with tendril-like a p p e n d a g e s (Fig. 13) might belong to the s a m e plant.

Eoantha Krassilov, 1986; Baisian A s s e m b l a g e , Early C r e t a c e o u s of Transbaikalia

(Plates 2 0 - 2 2 ) .
Flower-like gynoecial structure of four ovuliphores s u r r o u n d e d by a perianth of lin­
ear bracts, radially spreading in the open flowers, with the floral axis protruding and
bearing apical bracts that supposedly represent! a vestigial second floral n o d e . T h e ovules
are single per the gynoecial lobes, o r t h o t r o p o u s , with the n u c e l l u s t r a p p i n g ribbed
Ephedripites-iype pollen grains in the beak and with a c o n s p i c u o u s g r a n u l a r m e g a s p o r e
m e m b r a n e . A recently found specimen (Fig. 14) is attached by a slender stalk to the axis
with the vascular bundles consisting of pitted fibres and tracheary e l e m e n t s , the latter
s h o w i n g distant pits and the helical to reticulate thickenings.
Similar vascular e l e m e n t s have been observed in the vein b u n d l e s of the associating
parallelinerve graminoid leaves that might belong to the s a m e plant. T h e leaves (Plates
2 3 , 24) show short cross-veins and longitudinal rows of intercostal s t o m a t a that are
stretched as if the blade growth continued after the stomatal pattern w a s laid d o w n .
T h e gnetalean affinities of Eoantha are evidenced by the basically decussate, al­
though spreading in o n e plane, gynoecial lobes, orthotropous ovules and characteristic
Fig. 14. Eoanthci zherikhinii Krassil., a proangiospermous gnetalean plant of the Baisian Assem­
blage, Early Cretaceous of Transbaikalia (Krassilov, 1986): recently found attached flower (Krassilov
& Bugdaeva, 1997), x l 5 .
pollen grains. At the s a m e time the g y n o c l a d e s are reduced to a single fertile n o d e , thus
b e c o m i n g definitely flower-like. L e a v e s attributed to iht Eoantha plant on the e v i d e n c e
of association and vascular similarities suggest progenitorial relations to g r a m i n o i d
m o n o c o t s . S o m e incompletely k n o w n M e s o z o i c plants, p e r h a p s related to Eoantha, are
represented by dispersed ovules with Eucommiidites pollen grains in the pollen c h a m ­
ber. T h e pollen grains are characteristic in having exinal slits parallel to the sulcus
(Pedersen & al., 1988). T h e presence of a m e g a s p o r e m e m b r a n e w a s seen as a character
inconsistent with gnetalean affinities of these seeds ( R e y m a n o w n a , 1968), but presently
Eoantha gives e v i d e n c e to the contrary.

Vitimantha Krassilov & B u g d a e v a , 1997; Baisian A s s e m b l a g e , Early C r e t a c e o u s of

Transbaikalia (Plates 25-27).
L o n g l y pedicellate flower-like structure with a perianth of four leafy bracts s h o w i n g
reticulate venation (Fig. 15) and with four elongate follicles or c u p u l e s that are margin­
ally papillate, occasionally trapping ephedroid pollen grains.
This structure is potentially of a considerable phylogenetic interest r e s e m b l i n g an
a n g i o s p e r m flower (notably of Kingdonia) with leafy r e t i c u l i n e r v e t e p a l s and the
apocarpous g y n o e c i u m of ascidiform follicles. T h e latter m i g h t h a v e stigmatic marginal
crests. T h e s e features are c o m b i n e d with the typically gnetalean pollen grains. Yet the
supposed gynoecial organs are poorly k n o w n .

Fig. 15. Vitimantha, a preflower of the Baisian Assemblage, Early Cretaceous of Transbaikalia (Krassilov
& Bugdaeva, 1997), showing four bracteate tepals with anastomosing veins and follicular structures
(polyplicate pollen grains were found stuck to the margin of the right follicle).
 Caytonia sewardiiThomdi's,, 1925; Middle Jurassic of Yorkshire, England (Plates 28-30).
This species of caytonialean ovuliferous structures was described as foliar, with the
cupules formed of involute lateral pinnae, their tips still c o n s p i c u o u s as short " l i p s "
over the e x o s t o m e , or " m o u t h " - a slit between the lip and the stalk (Harris, 1951a). I
have been able to study a few s p e c i m e n s of this species (courtesy of Professor T M .
Harris), suggesting a s o m e w h a t different interpretation. With S E M , I h a v e observed the
mouth situated on the abaxial side of the lip turned away from the stalk rather than at the
stalk to lip j u n c t i o n . This m a k e s the involute p i n n a e m o d e l i m p r o b a b l e . T h e lip appears
decurrent on the pedicel and adnate to the base of the c u p u l e in the m a n n e r of a reduced
subtending bract (Fig. 16). T h e n the e x o s t o m e could h a v e been formed along the cupule
to subtending bract j u n c t i o n at the site of their fusion meristem. T h e ovules have been
found in a tight cluster rather than scattered parietal as in the classical reconstruction,
their micropyles of bulging cells adpressed to each other and nearly at the s a m e level,
suggesting an original clustering rather than a depositional c o m p r e s s i o n . T h e micropyles
are facing the e x o s t o m e and are connected to it by a series of the so-called c h a n n e l s that
s o m e t i m e s look like funnicles a t t a c h e d n e a r the m i c r o p y l e , the o v u l e s thus b e i n g
anatropous, their funnicles modified as hollow tubes acquiring a function of c o n d u c t i n g
channels for pollen grains entering the e x o s t o m e . A few grains h a v e actually been found
stuck m i d w a y in the channels. T h e ovules are flattened, with a raphe as a m e d i a n ridge

Fig. 16. Caytonia, gynoclade of stalked cupules and a suggested prototype of the cupule (center)
with a subtending bract transformed into "lip" of the mouth (right) (after Harris, 1951a; Krassilov,
1977, 1984).
Fig. 17. Czekanowskialean reproductive structures: Leptostrobus stigmatoideus Krassil., from the Late
Jurassic of Bureya Basin, Far East, Russia (Krassilov, 1969, 1972a), detailes of stigmatic surfaces, pitted,
with papillate ridges and cones, SEM, x 120 (2) and 600 (1).
conspicuous also on the nucellus that is free and cutinized, s h o w i n g a basal platform,
apically forming a poorly differentiated pollen chamber.
An e n i g m a t i c feature of the ovules is the so-called spotted layer previously inter­
preted as consisting of eleuron cells (Harris, 1951 a). In cleared ovules, it can be seen as
a well-differentiated coat t w o cells thick, thinly cutinized, e n v e l o p i n g the nucellus and
wedging out at the level of the pollen chamber. Both topologically and m o r p h o l o g i ­
cally, it c o r r e s p o n d s to the inner integuments of bitegmic a n g i o s p e r m ovules. Its spotted
appearance can be d u e to tannin a c c u m u l a t i o n s that c o m m o n l y o c c u r in the inner in­
teguments of a n g i o s p e r m s , rather than to eleuron grains.

Leptostrobus stigmatoideus Krassilov, 1972b; Late Jurassic of the Russian Far East
(Plates 3 1 - 3 2 ) .
This species of czekanowskialeans deserves special mention on account of its stigmatic
cupules. It is based on the g y n o c l a d e s consisting of a relatively thin axis bearing loosely
spirally arranged bivalved cupules that appear oriented vertically, with their m e d i a n
plane along the axis. T h e cupules are shortly stalked, occasionally with a stout cylindri­
cal stalk bearing rudimentary scales, as in the leafy short shoots of the s a m e plant. This
is taken as e v i d e n c e of their short shoot derivation, the c u p u l a r valves then correspond­
ing to a pair of ovuliferous scales that are marginally fused, as in p a r a c a r p o u s ovaries.
In this interpretation, the marginal flange might h a v e e v o l v e d as a p r o d u c t of the fusion
meristem activity. T h e valves remain adnate to 1/3 their length, distally g a p i n g at matu­
rity. T h e i r outer wall is plicate with strong flabellate ribs, strongly cutinized, with sto­
mata, w h e r e a s the interior cuticle is delicate but fairly resistant to m a c e r a t i o n , lacking
stomata. T h e ovules, four per valve, are submarginal on interior ribs. T h e sutural crests
are smooth externally, pitted internally, microscopically with a n e t w o r k of steep ridges
bordering rounded depressions and covered with simple or bifid papillae (Fig. 17), the
latter a n a l o g o u s to the typical stigmatic papillae of a n g i o s p e r m s . T h e crests intrude the
locule as a thin ledges 1.5 m m w i d e o v e r h a n g i n g the ovules. T h i s interior ledges, also
having its a n a l o g u e s in a n g i o s p e r m s , is densely papillate, but lacks a reticulate ridge
structure of the outer flange. T h e spherical bodies sticking to the papillae could be
deformed pollen grains, but n o n e of t h e m is sufficiently well-preserved. In distinction
from the mid-Jurassic species lacking stigmatic flanges (Harris, 1951b), n o pollen grains
have ever been observed inside the cupules of Leptostrobus stigmatoideus.

Gymnosperm phylogeny
W h i l e the a b o v e g y m n o s p e r m and p r o a n g i o s p e r m divisions represent the distinctive
adaptive types, their phylogenetic relations are to be inferred on the basis of h o m o l o g y
of their respective morphological structures. Discrimination of h o m o l o g y and homoplasy
is a m u c h debated p r o b l e m of apparently no o n c e and for all solution. T h e cladistic
techniques provide a c u m u l a t i v e assessment of h o m o p l a s y at most (Farris, 1989). H o ­
mology is m o r e often inferred from p h y l o g e n y than vice versa, while p h y l o g e n y is
based on oblique reasoning that is rooted in the c o n t e m p o r a r y evolutionary philosophy.
Insofar as phylogeny primarily conveys our understanding of homology, it cannot be
more objective than the latter. T h e objective elements in it are chronological relationships
as well as morphological links, that is, the intermediate forms that show character states
midway in the polarity gradient or the mosaic forms that c o m b i n e the typical characters of
different taxonomic units (see above). A phylogenetic reconstruction would then start
with assigning the operational units - g y m n o s p e r m orders in our case - to successive
geological age groups reflecting their first appearances in the fossil record (not reflecting
the overlaps of their stratigraphic ranges). T h e relations within and between the age groups
would then be inferred on the basis of intermediate and/or mosaic forms.
M e m b e r s of a single age g r o u p are likely to h a v e i n d e p e n d e n t origins or, if con­
nected by intermediate forms, a c o m m o n origin, that is, their relations are fraternal
rather than progenitorial. O n the other hand, the m e m b e r s of successive age g r o u p s , if
connected by intermediate or m o s a i c forms, are likely to h a v e progenitorial relations.
Presently, the chronological relations between the major g y m n o s p e r m units are mostly
based on a sufficiently high stratigraphic resolution and are fairly reliable. Discovery of
linking forms m a y seem accidental but, since there w e r e intermediate habitats and dis­
crepant evolution rates for different functional units, the intermediate and m o s a i c forms
ought to be found.
A c c o r d i n g to their first appearances in the fossil record, the g y m n o s p e r m orders can
be assigned to the following age g r o u p s :
(1) L a t e D e v o n i a n - E a r l y C a r b o n i f e r o u s : H y d r o s p e r m a l e s , L a g e n o s t o m a l e s ,
Trigonocarpales (Lyginopteridales, M e d u l l o s a l e s of the axial a n a t o m y classification);
(2) M i d - C a r b o n i f e r o u s : Callistophytales, Cordaitales, W a l c h i a l e s ;
(3) Late Carboniferous - P e r m i a n : Glossopteridales, Vojnovskyales, G i g a n t o p t e r i -
dales, archaic Peltaspermales (callipterids), archaic Coniferales (voltziaceans);
(4) E a r l y M e s o z o i c : a d v a n c e d P e l t a s p e r m a l e s , C o r y s t o s p e r m a l e s , N i l s s o n i a l e s ,
Ginkgoales, Czekanowskiales, Caytoniales, Bennettitales, archaic Gnetales (proto-
gnetaleans, such as Dinophyton)',
(5) Late M e s o z o i c : Pentoxylales, C y c a d a l e s , m o d e m Coniferales, m o d e m Gnetales,
their derived p r o a n g i o s p e r m o u s forms, such as Eoantha, etc.
T h e g y m n o s p e r m s are thought to be rooted in p r o g y m n o s p e r m s of w h i c h both the
heterosporous archaeopterids and the h o m o s p o r o u s a n e u r o p h y t e s are c o n s i d e r e d as p o ­
tential ancestors (Rothwell & Erwin, 1987). Moreover, there w e r e h e r b a c e o u s plants of
a simpler axial a n a t o m y but having elaborate c u p u l e - l i k e sporangial clusters, as in
Lenlogia (Krassilov & Z a k h a r o v a , 1995). Early seed plant diversity is insufficiently
k n o w n , however, for establishing the progenitorial relations o t h e r w i s e but in a very
general form. T h e possibility of the g y m n o s p e r m a n a t o m y and seeds originating in
different lineages then c o m b i n e d by horizontal g e n e transfers in the D e v o n i a n plant
c o m m u n i t i e s cannot be excluded.
 In the age g r o u p (1), the H y d r o s p e r m a l e s with 2 - m a n y - o v u l a t e c u p u l e s , s o m e t i m e s
containing sporangia (Long, 1977a), appeared s o m e w h a t earlier than the other t w o or­
ders. That this type of cupules might have been primary is scarcely d e d u c i b l e from their
morphology: it is just a geochronological fact.
The H y d r o s p e r m a l e s merge with the L a g e n o s t o m a l e s - their respective axial organs
are assigned to a single order, Lyginopteridales, while the ovules are of the s a m e basic
type and are linked to the Trigonocarpales (Medullosales) by the forms with an inter­
mediate c a l l a m o p y t i a l e a n - m e d u l l o s a l e a n a n a t o m y , such as the L a t e M i s s i s s i p p i a n
Questora ( M a p e s & Rothwell, 1980).
Both the L a g e n o s t o m a l e s and Trigonocarpales show a tendency of a not quite syn­
chronous transition of the pollen receptive function from the nucellar apex with its
excessively developed pollen trapping and secretory structures, lagenostome and salpynx,
to the integumental micropyles, with the lobed micropyles as a transitional stage. T h e y
differ, however, in the relative d e v e l o p m e n t of the inner, integumental, and the outer,
cupular, coats of their o v u l e s , a variation that can be traced u p to the present-day
angiosperms. W h i l e in the L a g e n o s t o m a l e s the m i c r o p y l e was primarily integumental,
in the Trigonocarpales it might have been of a cupular origin, with the primary integu­
ment reduced and adnate to the nucellus, the latter thus transformed into a chimeric
structure a s s u m i n g the vascular system of integumental origin, h e n c e n o n - h o m o l o g o u s
to the nucelli of the lagenostomalean line.
T h e fraternal phylogenetic relations of L a g e n o s t o m a l e s and Trigonocarpales imply
that the ovules borne on fern-like leaves, a critical feature of p t e r i d o s p e r m s , appeared
independently in both groups in the course of their parallel evolution. S u c h parallel
developments of the fern-like foliage on the basis of syntelomic branch s y s t e m s in spore
plants and early g y m n o s p e r m s might have been urged by a d e m a n d of larger p h o t o s y n -
thetic surfaces in response to either a decreased ultraviolet radiation rate (due to oxida­
tion of a t m o s p h e r e , with the c o n c o m i t a n t l y i n c r e a s i n g w a t e r v a p o u r c o n t e n t and
stratospheric o z o n e ) or the build-up of the structural complexity, c a n o p y formation and
the ensuing competition for light in the early plant c o m m u n i t i e s , or both. In distinction
from the e x p o s e d h y d r o s p e r m a l e a n cupules, the leaf-borne cupules w e r e less effective
in wind pollination. T h e c o n c o m i t a n t d e v e l o p m e n t of the large synangiate polliniferous
structures and the secretory apical structures of the early ovules suggest pollination by
arthropods finding shelter in the foliage and not yet e n g a g e d in the leaf-cutting activi­
ties (that appeared m u c h later).

In the age g r o u p (2), the Callistophytales e m b o d i e d a m o s a i c of forms c o m b i n i n g an

essentially lyginopteroid m o r p h o l o g y of their vegetative parts and s y n a n g i a with the
platyspermic c a r d i o c a r p o i d o v u l e s (incipient p l a t y s p e r m y a p p e a r e d already in the
lagenostomalean ovules, see Stubblefield & al., 1984) and true pollen {Vesicaspora)
also r e s e m b l i n g s o m e c o r d a i t a l e a n m o r p h o t y p e s {Florinites). A m o n g cordaites,
Mesoxylon is l i k e w i s e a linking form with the e u s t e l e f o r m e d of leaf traces, as in
megaphyllous plants, and with c o m p o u n d pollen c o n e s p r o d u c i n g prepollen. T h e seeds
were, however, cardiocarpoid ( G r o v e & R o t h w e l l , 1980), like in the a d v a n c e d forms
that acquired a sympodial eustele and true pollen grains (Sterlin, 1982a). T h e Walchiales
are obviously related to cordaites, their coniferoid foliage c o r r e s p o n d i n g to the lower
leaves of the latter (the cordaitalean ribbon-shaped parallelinerve foliage leaves might
have been phyllodic). T h e y still produced prepollen, as in the primitive cordaites, and
their ovules, though borne in m o r e or less strongly flattened axillar shoots and s o m e ­
times inverted, w e r e of the s a m e cardiocarpoid type.
Neither morphological nor chronological e v i d e n c e suggests derivation of the age
group (2) g y m n o s p e r m orders from each other. T h e e v i d e n c e is rather in favour of their
c o m m o n lagenostomalean ancestry. Remarkably, these three orders h a v e in parallel de­
veloped the a n a s u l c a t e pollen on the basis of s p o r e - l i k e p r e p o l l e n , as well as the
platyspermic seeds derived from radiospermic precursors, while the c o m p o u n d strobili
have a p p e a r e d in t w o of t h e m . T r u e pollen h a v e i n d e p e n d a n t l y arised a l s o in the
Trigonocarpales, a geologically older age g r o u p (1) order that u n d e r w e n t a consider­
able modernization in the late Paleozoic. A s y n a p o m o r p h i c g r o u p based on this obvi­
ously derived character would include the a d v a n c e d m e m b e r s both of trigonocarps and
cordaites, as well as callistophytes, otherwise the most p l e s i o m o r p h i c order of the age
group (2).
A n o t h e r c o m m o n t e n d e n c y in the a g e g r o u p (2) w a s the loss of c u p u l e s e i t h e r by
reduction or by fusion with the i n t e g u m e n t , or the c u p u l e s m i g h t h a v e lost their iden­
tity in r e p l a c i n g the p r i m a r y i n t e g u m e n t s that b e c a m e i n c o r p o r a t e d in the n u c e l l i , thus
giving rise to crassinucellate ovules. In effect, the age g r o u p (2) seed plants w e r e ren­
dered true g y m n o s p e r m s replacing the prevailingly c h l a m y d o s p e r m o u s age g r o u p (1)
pteridosperms.

In the age g r o u p (3), the Permian callipterids can be c o n s i d e r e d as transitional be­

tween the callistophytes of the age g r o u p (2) and the typical Triassic p e l t a s p e r m s . T h e i r
leaves and pollen organs are quite similar to those of callistophytes, the pollen grains
are of the s a m e m o r p h o t y p e (Vesicaspora), while the seed organs of the Autunia type,
with their scaly ovuliphores bearing t w o sessile, p l a t y s p e r m i c , inverted o v u l e s , are o b ­
viously a variation of the Callospermarion t h e m e (Rothwell, 1980).
T h e s a m e can be said of a typical coniferous seed-scale. In other c h a r a c t e r s , conifers
are similar to both callistophytes (Rothwell, 1981) and archaic p e l t a s p e r m s ( M e y e n ,
1984), clustering with the latter as their sister-group. T h e in the archaic forms c o m ­
pound pollen cones ( G r a u v o g e l - S t a m m , 1978), true pollen, and laminar seed-scales are
a g a i n s t their c l o s e affinities with w a l c h i a n s , w h i l e the c o m p o u n d s t r o b i l a t i o n of
ovuliphores, shared with the latter g r o u p , as well as with cordaites, a p p e a r s a parallel
development.
At the opposite morphological pole, the peltasperms are linked to the gigantopterids
through such intermediate forms as Tinsleya, Russelites ( M a m a y , 1968, M a m a y & Watt,
1971), Protoblechnum, etc. Both peltasperms and gigantopterids s h o w a t e n d e n c y to a
consecutive fusion of the pinnules and pinnae resulting in a simple a n g i o s p e r m - l i k e leaf
with a hierarchical areolate venation (Krassilov, 1995).
T h e glossopterid stem a n a t o m y and ovulate structures are not yet u n a m b i g u o u s l y
interpreted rendering their phylogenetic position uncertain. O r the ambiguity might be
caused by polyphyletic origins. T h e protostelic stems (Vertebraria) and the dichoto-
mously branched pollen organs with terminal sporangial clusters are archaic features
found in the m o s t p r i m i t i v e l y g i n o p t e r i d s and s u g g e s t i v e of a fraternal relation to
callistophytes. T h e derived characters, notably the saccate pollen grains and platyspermic
ovules, might have evolved in parallel with the callistophytes, cordaites and peltasperms.
The leaves a p p e a r p h y l l o d i c , as in c o r d a i t e s , and s o m e leaf f o r m s {Euryphyllum,
Noeggerathiopsis) are actually quite similar to Cordaites, the similarity extending also
to vojnovskyaleans, the Angaridan counterpart of glossopterids.
Arberia, the m o s t primitive ovuliferous organ of glossopterids, is here interpreted as
a s o m e w h a t flattened, profusely branched shoot. S o m e of the derived ovuliferous struc­
tures might h a v e arisen as modifications of Arberia (Rigby, 1978). T h e i r fusion to sub­
tending bracts finds its parallel in conifers, the persistent bracts apparently assisting in
dispersal in both cases. H o w e v e r , in the a d v a n c e d glossopterids, a prevailing tendency
seems to h a v e been a shortening of the ovuliferous shoot resulting in a radial arrange­
ment of the ovules in the capitulate or c a m p a n u l a t e , s o m e t i m e s discoid, clusters. This
t e n d e n c y w a s p a r a l l e l e d by t h e v o j n o v s k y a l e a n s . R e m a r k a b l y , b o t h g r o u p s had
protosaccate taeniate pollen grains, shared also with peltasperms and s o m e early coni­
fers. A variety of P e r m i a n insects fed on such pollen (as is d o c u m e n t e d directly by the
pollen preserved in their gut contents, see Krassilov & Rasnitsyn, 1996; Rasnitsyn &
Krassilov, 1996). Pollinivorous insects might h a v e both p r o m o t e d the parallelism by
mediating horizontal g e n e transfers (see m o r e on it in the next chapters) and initiated
entomophily, the latter explaining also the capitulate ovuliferous structures of a m a s s e d
small ovules confronting a pollination vector with a brush of adpressed m i c r o p y l e s .
Characteristic of age g r o u p (3) are specialized laminar o v u l i p h o r e s - seed-scales -
developed in parallel in most of its m e m b e r s , although not necessarily h o m o l o g o u s . In
peltasperms and perhaps also in conifers, they are modified foliar o v u l i p h o r e s of the
age g r o u p (1) p t e r i d o s p e r m s , while in gigantopterids they might have been secondarily
flattened fertile shoots, partly lacking lamina in aberrant forms, such as Physmatocycas
(Gillespie & Pfeffercorn, 1986).

In the age g r o u p (4), the nilssonialeans are linked to the p e l t a s p e r m s through their
ovuliferous organs (Beania) that are basically identical to Autunia. Previously assigned
to cycads, they differ in the leptocaul life-form with d i m o r p h i c shoots, simple leaves
and the ovules abaxial on the heads of the longly stalked o v u l i p h o r e s . T h e pollen cones
are similar to those of the extant c y c a d s , but differ from those of M e s o z o i c c y c a d s , such
as Cycandra that p r o d u c e d the m u c h m o r e elaborate multisporangiate synangia. T h e
simpler b i s p o r a n g i a t e h a l f - s y n a n g i a are t h u s d e r i v e d in the m o d e r n c y c a d s . T h e
c z e k a n o w s k i a l e a n s are similar to nilssonias in the life-form characters including the
deciduous spur-shoots, while their bivalved cupules s e e m derivable from a pair of mar­
ginally fused peltasperm-type peltae.
T h e c o r y s t o s p e r m s are linked to p e n t o x y l e a n s and c y c a d s on the basis of their essen­
tially medullosalean stem a n a t o m y (Rhexoxylon), and to g i n k g o a l e a n s on the basis of
their similarly paired ovules e m b e d d e d in the short collar-like c u p u l e s conceivably re­
ducible to the " c o l l a r s " of the extant Ginkgo. In addition, the flabellate dissected leaf
morphotypes typical of the M e s o z o i c ginkgoleans occur mDicroidium, the corystosperm
foliage (Anderson & Anderson, 1983. 1985). T h e r e is also a m a r k e d s t o m a t o g r a p h i c
similarity between c o r y s t o s p e r m s and g i n k g o a l e a n s e x t e n d i n g to p e n t o x y l e a n s .
T h e cycadalean stem a n a t o m y with characteristic girdling leaf traces first appeared
in Antarcticycas ( S m o o t & al., 1985) of a c o r y s t o s p e r m - d o m i n a t e d Triassic flora. T h e
Jurassic Cycandra, a protocycad pollen cone, resembles the medullosalean pteridosperms
in having large synangia of concentrically arranged tubular s p o r a n g i a (Krassilov & al.,
1996). T h e d o u b l e v a s c u l a r s y s t e m in the o v u l e s of e x t a n t c y c a d s is still a n o t h e r
trigonocarpalean feature. At the s a m e time, the Early CvtidiCQ,ou^Semionogyna, a proto-
cycadean ovuliferous shoot with sessile bracteate ovules (Krassilov & B u g d a e v a , 1988c),
is c o m p a r a b l e to the Permian Sohernheimia of the callipterid circle ( K e r p , 1983), both
associating with peltasperm-type leaves and pollen organs. Semionogyna also s h o w s a
tendency to sterilization and planation of the distal bracts, potentially leading iodiCycas-
like leafy ovuliphore (Krassilov & B u g d a e v a , 1988c). T h u s the origin of the m o d e r n
cycads (that appeared not earlier than the m i d - C r e t a c e o u s ) r e m a i n s controversial, with
a possibility of genetic contributions from different p h y l o g e n e t i c clusters.
T h e extant Gingko is markedly different from the extant c y c a d s in the life-form and
is usually allied with coniferoid g y m n o s p e r m s rather than with c y c a d o p h y t e s . H o w ­
ever, ginkgoaleans share with the latter the general structure of both pollen and seed
strobili (with parallel destrobilation tendencies manifested in Ginkgo and Cycas), the
pollen (Shashi & U e n o , 1986) and ovule m o r p h o l o g i e s , as well as m a n y features of
reproductive biology, including the pollen tube, sperm, archegonial and e m b r y o l o g i c a l
similarities. In the Mesozoic ginkgoalean ovules {Karkenia), the nucelli were free, beaked
and strongly cutinized, thus m o r e cycad-like than in the extant species. Interpretation of
the characters shared by c y c a d s and g i n g k o as s y m p l e s i o m o r p h i e s attesting to their
primitiveness is hardly warranted, for both the m o d e m g i n k g o a l e a n s and the m o d e r n
cycads appeared rather late in the geological history of g y m n o s p e r m s . In them, the
large ovules and their correlated reproductive features are recent d e v e l o p m e n t s lacking
in their early M e s o z o i c precursors. Contrary to a widely held opinion, derived states in
plant m o r p h o l o g y are not necessarily those d u e to reduction. S o m e t i m e s they are those
due to magnification.
In the rest of age group (4), the relatedness of bennettites and gnetaleans is attested
by their basically similar ovuliferous s t m c t u r e s with o r t h o t r o p o u s ovules enclosed in
the cupules formed of bracteoles or their h o m o l o g o u s interseminal scales. T h e bennet-
titalean receptacles with an apical c o r o n a of sterile scales c o r r e s p o n d to the shortened
binodal strobili of extant Gnetum scandens (Foster & Gifford, 1974), with the distal
node reduced. T h e bennettitalean ovules, radiospermic with residual nucellar vascular
bundles, are essentially trigonocarpalean, at the s a m e time r e s e m b l i n g Gnetum in the
peculiar cutinized flange at the base of the m i c r o p y l e tube (e.g., in Bennetticarpus
crossospermum: Harris, 1932). T h e Triassic Dinophyton, a peculiar protognetalean form,
showed a 3-ribbed nucellus as in Pachytesta, a trigonocarpalean ovule. Notably, s o m e
medullosaleans had a decussate leaf a r r a n g e m e n t ( M a p e s & Rothwell, 1980) character­
istic of g n e t a l e a n s . At the s a m e t i m e , the H i r m e r e l l a c e a e , a n o t h e r p r o t o g n e t a l e a n
Mesozoic group, seems to have been be related io Majonica, a Permian coniferoid (Clem­
ent-Westerhof, 1987).
T h e Caytoniales produced c o m p o u n d leaves with GlossopterisA'ikQ leaflets occa­
sionally s h o w i n g a decussate a r r a n g e m e n t betraying their c a u l o m i c origin. T h e i r pollen
organs are essentially glossopteroid while the cupules are fused to their residual sub­
tending bracts (lips) and s e e m derivable from the bracteate glossopteridalean cupules,
perhaps by arrested d e v e l o p m e n t at a j u v e n i l e stage when s o m e of the c u p u l e s were
closed, as those described by G o u l d & D ele vo ry as (1977).
Characteristic of the age g r o u p (4) g y m n o s p e r m s , irrespective of their phylogenetic
position, are the d e c i d u o u s short shoots, the prevailingly simple, entire or pinnatifid,
often phyllodic leaves, the asaccate pollen grains, s o m e t i m e s with residual sacci or
folds betraying their origin from saccate forms, and the strobilate ovuliferous organs,
while the leafy ovuliphores, so widespread in the P a l e o z o i c , are totally lacking. T h e
latter change might have been due to the appearance of leaf-cutting insects or foliophagous
tetrapods, or both.

In the m i d - C r e t a c e o u s , the age g r o u p (4) g y m n o s p e r m s either vanished or under­

went a radical m o d e r n i z a t i o n . T h e nearly s y n c h r o n o u s a p p e a r a n c e of fleshy seeds in
cycads, g i n k g o and s o m e conifers, of fleshy cupules in caytonias (as distinct from dry
cupules of the Late Triassic forms), of fleshy receptacles in bennettites and of fruit-like
cones in p e n t o x y l e a n s might have been related to zoochory that w a s then fostered by a
rapid spread of birds and multituberculates. This factor, p e r h a p s in conjunction with the
spread of o v u l e - s u c k i n g insects, might have been impelled a better protection of ovules
either by the a r m o u r of closed peltate c o n e scales, as in c y c a d s and the c o n c o m i t a n t l y
appearing m o d e r n p i n a c e o u s , t ax od ia c eo us and c u p r e s s a c e o u s conifers, or by the newly
f o r m e d c u p u l e s t h a t r e a p p e a r e d in s e v e r a l p h y l o g e n e t i c l i n e s l e a d i n g to t h e
p r o a n g i o s p e r m o u s Leptostrobus, Caytonia, Dirhopalostachys, Baisia, Eoantha, etc. At
this stage, a n g i o s p e r m s have been added to the seed plant diversity.
 General tendencies and patterns
A few evolutionary generalizations are d e d u c i b l e from the a b o v e analysis (Fig. 18):
(1) Increase in the higher-rank diversity. A b o u t twenty orders (with t w o or three
controversial) have been added to the initial seed plant diversity in the c o u r s e of g y m ­
n o s p e r m evolution. T h e n u m b e r s of orders appearing and coexisting in the successive
age groups are as follows:

Appeared Coexisted
Late Devonian - Early Carboniferous 3
Mid-Carboniferous 3 6
Late Carboniferous - Permian 5 7
Early Mesozoic 6(9) 8(11)
Late Mesozoic (at the time of 0 (3) 8
angiosperm appearance)
Present day 0 4

T h e s e n u m b e r s seem to indicate that the higher rank n o n - a n g i o s p e r m seed plant

diversity peaked in the early M e s o z o i c , at the time of the p r o a n g i o s p e r m a p p e a r a n c e s .
T h e subsequent decrease at the e x p e n s e of p r o a n g i o s p e r m o u s orders m i g h t reflect e v o ­
lution of the latter into conventional a n g i o s p e r m s .
(2) Genetic continuity. A r e m a r k a b l e feature of seed plant p h y l o g e n y is the persist­
ence of genetic lines. Although s o m e orders ceased to exist nominally, n o n e of t h e m
disappeared without leaving d e s c e n d a n t s that in turn contributed genetic material to the
next evolutionary stage u p to the p r o a n g i o s p e r m level. This w a s a c c o m p a n i e d by the
parallel d e v e l o p m e n t s , r e a p p e a r a n c e s of the formerly e x t i n g u i s h e d c h a r a c t e r s , and
reticulations between the phylogenetic clusters. W h e t h e r the genetic continuity w a s
truncated by the appearance of a n g i o s p e r m s or the latter w e r e inheritors of all or most
of the disappearing n o n - a n g i o s p e r m o u s lineages, is yet to be learned.
(3) Nodal clusters. T h e successive radiations of g y m n o s p e r m s and p r o a n g i o s p e r m s
appear to have occurred by their lineages deriving in clusters from their stem g r o u p s
rather than by a conventional succession of d i c h o t o m i e s .
(4) P a r a l l e l i s m s . A n o t a b l e feature of a n g i o s p e r m p r e h i s t o r y is a w i d e s p r e a d
h o m o p l a s y of d e r i v e d c h a r a c t e r s s i m u l t a n e o u s l y a p p e a r i n g in the fraternal l i n e a g e s ,
such as the leafy o v u l i p h o r e s in l a g e n o s t o m a l e a n s and t r i g o n o c a r p s ; the a n a s u l c a t e
pollen and p l a t y s p e r m i c o v u l e s ( r e p l a c i n g p r e p o l l e n a n d r a d i o s p e r m i c o v u l e s in
t r i g o n o c a r p s , c a l l i s t o p h y t e s a n d c o r d a i t e s ) ; t h e c o m p o u n d s t r o b i l i in c o r d a i t e s ,
w a l c h i a n s , conifers and other P a l e o z o i c c o n i f e r o i d s ; the radial a r r a n g e m e n t of g y n o ­
ecial structures in g l o s s o p t e r i d s , v o j n o v s k y a l e a n s and b e n n e t t i t e s ; the t a e n i a t e p o l l e n
grains in g l o s s o p t e r i d s , p e l t a s p e r m s and conifers; the s p e c i a l i z e d scaly o v u l i p h o r e s -
seed-scales - in p e l t a s p e r m s , n i l s s o n i a l e a n s , conifers, e t c . ; their fusion to s u b t e n d i n g
bracts assisting in seed dispersal in g l o s s o p t e r i d s , conifers and g n e t a l e a n s ; the closed
cones of peltate c o n e scales in p e l t a s p e r m s , c y c a d s and several g r o u p s of m o d e r n
conifers; the d e s t r o b i l a t i o n in Ginkgo, Cycas, Taxus, p o d o c a r p s and o t h e r conifers as
well as in the M e s o z o i c g n e t o p h y t e s ; the s e c o n d a r y c u p u l e s in c a y t o n i a l e a n s , c z e k a ­
n o w s k i a l e a n s and o t h e r p r o a n g i o s p e r m s ; the fleshy seeds or c u p u l e s in the m o d e r n

Pi DIRHOP LEFT STIGMATOID EOANT BAIS

O
o
z CZEK CAYT GNET BENNET
<
o - CUPULATION

NILSS - DESTROBILATION

<
-BRACHYBLASTY

C/3
-TAENIAE
Q

z
o - BISTROBILATION
u
- POLLEN

- PLATYSPERMY

TRIGON
- MICROPYLE

LAGEN
1 - PHYLLOSPERMY

HYDROSPERMS
PROGYMNOSPERMS

Fig. 18. Suggested phylogenetic relations in and between the gymnosperm and angiosperm grades (left),
with horizontal dashes marking the characters (right) appearing in more than one clade.
c y c a d s , g i n k g o , Caytonia, etc.; the spread of asaccate pollen m o r p h o l o g i e s in the M e s o ­
zoic, involving nilssonialeans, cycads, g i n k g o a l e a n s , etc.
(5) Time-fashions. Of special interest a m o n g the parallelisms are those confined to
particular time planes. T h e s e are the time-specific characters, or m o r p h o l o g i c a l "fash­
ions", spreading a m o n g the coexisting groups of seed plants across the genetic b o u n d a ­
ries. A m o n g the familiar e x a m p l e s are the leafy ovuliphores s i m u l t a n e o u s l y appearing
in lagenostomalean and trigonocarpalean pteridosperms, the platyspermic ovules in most
of the d o m i n a n t late Paleozoic groups, and the taeniate pollen in the P e r m i a n . In dis­
tinction from d i a c h r o n o u s parallelisms that might reflect s o m e general evolutionary
tendencies, the time-specific fashions are induced by e n v i r o n m e n t a l factors simultane­
ously affecting different seed plant lineages. W h e r e a s s o m e time-fashions, e.g. the fleshy
disseminules in the late M e s o z o i c c y c a d s , g i n k g o s , caytonias, p o d o c a r p s , etc., are obvi­
ously adaptive, the others, such as the taeniate exine, s e e m far less so and are suggestive
of a horizontal spread of the respective mutation (see m o r e on this in the next chapters).
(6) Reversals. T h e s e may have been stipulated by the genetic continuity manifesting
itself in the reappearances of temporarily extinguished characters while the respective
genetic structures have latently persisted in the lineage. T h u s , bitegmic o v u l e s appear­
ing already in the early seed plants then faded out of the major Paleozoic g y m n o s p e r m
lineages to reappear in their late M e s o z o i c d e s c e n d a n t s .
(7) Cyclicity. Seed plant evolution has started with the a p p e a r a n c e of c u p u l a t e ovules
arising from sporangial clusters by involving a set, or a d o u b l e set, of s u r r o u n d i n g ster­
ile a p p e n d a g e s that acquired the protective function of a seed coat (or a d o u b l e coat)
stabilizing the internal e n v i r o n m e n t in which an in situ d e v e l o p m e n t of the m e g a g a -
m e t o p h y t e would take place. T h e s e c o m p o u n d structures functioned as dispersal units.
In the early g y m n o s p e r m s , the coats were still open, with prepollen settling on the
sporangial apex that developed special receptive structures. Later, the receptive func­
tion w a s also transferred to the seed coat or coats that formed an integumental m i c r o p y l e
or a cupular e x o s t o m e , or both. Consequently, the receptive structures of the nucellar
apex lost their function and were reduced. This h a p p e n e d in parallel in the g y m n o s p e r m
lineages of the second evolutionary stage and, likewise in parallel, a t e n d e n c y to the
nucellus fusion with the integument was set in. A c o m p l e t e fusion resulted in a thick­
ened and vascularized secondary sporangial wall of chimeric nature, as in trigonocarps,
protected by a thinner integument of cupular origin. T h e nucellus then p r o c e e d e d upon
the reduction track. M u c h later, the inner integument might h a v e been restored for a
special pollen-tube c o n d u c t i n g function, as in a n g i o s p e r m s . M e a n w h i l e , the reduction
of the intracupular structures w a s c o m p e n s a t e d by the addition of a n e w set of periph­
eral structures (modified ovuliphores, subtending bracts, persistent receptacles, closed
strobili) and their involvement in protection, pollination and dispersal as seed-scale
c o m p l e x e s , bract - seed-scale c o m p l e x e s , secondary c u p u l e s , etc. S u c h n e w c o m p o u n d
structures have been formed only to be e n g a g e d in the next round of integration and
reduction rendering t h e m similar to their constituent structures, as, for instance, the
c o m p o u n d strobili are h o m o m o r p h o u s to their precursory simple strobili, the c u p u l e s to
the integuments, and the inner integument - nucellus c h i m e r a e to simple nucelli, a phe­
nomenon that I have called retroconvergence (Krassilov, 1995, further illustrated in the
next chapters). T h e p r o a n g i o s p e r m s and then a n g i o s p e r m s appeared as participants of
this incessant morphological play at a certain act of its d e v e l o p m e n t .
(8) Geological-scale periodicities. Remarkably, the cyclicity of the complication-
reduction processes crudely corresponds to the periodicity of the geochronological scale.
Insofar as most early seed plants had cupulate ovules, w e can speak of the first, mid-
Paleozoic, round of a n g i o s p e r m i z a t i o n followed by g y m n o s p e r m i z a t i o n in the late
P a l e o z o i c , in turn f o l l o w e d by the s e c o n d r o u n d of a n g i o s p e r m i z a t i o n in the late
Mesozoic. A morphological distance between the c u p u l a t e structures of extinct seed
plants and a n g i o s p e r m g y n o e c i a appears m u c h shorter than b e t w e e n the latter and the
scaly o v u l i p h o r e s of extant g y m n o s p e r m s . Such a zigzag c o u r s e of seed plant evolution
is paralleled by terrestrial vertebrates a m o n g which the m a m m a l - l i k e therapsids were
replaced by the bird-like dinosaurs, in turn replaced by m a m m a l s .
T h e leafy o v u l i p h o r e s s i m u l t a n e o u s l y a p p e a r i n g in different p t e r i d o s p e r m line­
ages m i g h t h a v e attracted pollinators sheltering in the foliage. T h e y m i g h t also h a v e
assisted in seed d i s p e r s a l . At about the P e r m i a n / T r i a s s i c b o u n d a r y , they w e r e lost in
strobilation, the then p r e v a i l i n g t e n d e n c y of o v u l i p h o r e a g g r e g a t i o n a s s o c i a t e d with
seed-scales, the then p r e v a l e n t t y p e of o v u l i p h o r e s . T h e s e c h a n g e s signified n e w pol­
lination and dispersal e n v i r o n m e n t s . In the late M e s o z o i c t i m e , the o p p o s i t e t e n d e n c y
of destrobilation w a s a c c o m p a n i e d by the spread of fleshy s e e d s or c u p u l e s , s u g g e s ­
tive of z o o c h o r o u s a d a p t a t i o n s .
T h e major events of morphological evolution coincided with the m o s t p r o m i n e n t
global climatic c h a n g e s , such as the onset of the Late Paleozoic glaciation, the ultimate
Permian deglaciation and aridity, and the mid-Jurassic cooling (Krassilov, 1994). Their
effect on the terrestrial e c o s y s t e m s manifested itself in m a s s extinctions of d o m i n a n t
plants and a n i m a l s and in the prevalence of pioneer strategists in the surviving groups.
They also triggered off the widespread forestation/deforestation trends that affected the
evolution both of arboreal and herbaceous life-forms as well as the pollination and
dispersal e n v i r o n m e n t s inducing the respective c h a n g e s of r e p r o d u c t i v e strategies and
the c o r r e s p o n d i n g morphological restructuring of r e p r o d u c t i v e sphere. M o r p h o l o g i c a l
evolution was thus incorporated in the biospheric p r o c e s s e s .
 Early history
T h e history of a n g i o s p e r m s might h a v e b e g u n well before the C r e t a c e o u s , but its
major events did take place during that period. A brief o v e r v i e w that follows is based on
my recent review (Krassilov, 1985) and the literature cited therein.

The Cretaceous
Starting as occasional n e w c o m e r s into the g y m n o s p e r m - d o m i n a t e d Early C r e t a c e o u s
floras, a n g i o s p e r m s b e c a m e , on the global scale, a s u b d o m i n a n t or, locally, even a d o m i ­
nant g r o u p of terrestrial plants to the end of the period. T h e i r rise w a s paralleled by
m a m m a l s that c a m e to d o m i n a n c e a m o n g the smaller vertebrates (Lillegraven & al.,
1979) and, in the terminal C r e t a c e o u s , were m u c h m o r e diverse than the then declining
dinosaurs. Such replacements of the leading plant and animal g r o u p s , a m p l y d o c u m e n t e d
by the fossil record, signified a radical c h a n g e in the terrestrial e c o s y s t e m s . S i m u l t a n e ­
ously, the land m a s s e s stepwise acquired the familiar outlines of the present-day conti­
nents, and the earth entered the C e n o z o i c era. T h e advent of a n g i o s p e r m s w a s an inte­
gral part of this process.
The Cretaceous lasted about 70 million years, from about 130 M a to about 65 Ma. It is
divided into two nearly equal epoches about 35 million years each, corresponding to the
major periodicity of the tectonic and biotic evolution, and in turn divisible into 12
geochronological ages (stages of the rock classification), each about 5-6 million years long
which is an average age duration for the entire Phanerozoic (historically, the stages were
based on local stratigraphic divisions that, to a certain extent, but somewhat inconsistently,
reflected the global periodicities; a few ages, such as the Coniacian, are at least three times
shorter than the average, actually corresponding to the average subages). T h e Aptian to
Turonian interval, about 115-88 M a , is informally designated as the mid-Cretaceous.

Period Epoch Age group Age Ma (beginning)

Paleogene Paleocene Danian 65
Cretaceous Late Senonian Maastrichtian 77
Campanian 83
Santonian 88
Coniacian 90
mid-Cretaceous Turonian 95
Cenomanian 100
Early Albian 110
Aptian 115
 Neocomian Barremian 120
Hauterivian 125
Valanginian 129
Berriasian 134
Jurassic Late Malm Tithonian 141

S o m e s t r a t i g r a p h e r s c o n s i d e r p e r i o d s and e p o c h e s as entirely c o n v e n t i o n a l divi­

sions of the g e o c h r o n o l o g i c a l scale. M o s t P h a n e r o z o i c p e r i o d s , h o w e v e r , a r e of c o m ­
parable d u r a t i o n s and are m a r k e d by the b o u n d a r y e v e n t s that are m o r e significant in
their r a n g e s and c o n s e q u e n c e s than any intraperiod e v e n t s . T h e C r e t a c e o u s differs
from o t h e r p e r i o d s , e x c e p t p e r h a p s the P e r m i a n , in h a v i n g t h e m o s t i m p o r t a n t e v e n t s
in the m i d d l e . In effect, the L a t e C r e t a c e o u s is very different from t h e E a r l y C r e t a ­
ceous both in the p a l a e o g e o g r a p h y and biota. T h e Early C r e t a c e o u s started with the
global s c a l e rifting and faulting of c o n t i n e n t a l m a r g i n s that d e f i n e d a t e c t o n i c style of
the e p o c h . T h i c k terrestrial and paralic d e p o s i t i o n o c c u r r e d p r i m a r i l y in the m a r g i n a l
downfaulted b a s i n s , as well as in the intracratonic rift z o n e s . T h e p r e d o m i n a n t l y clastic
marginal d e p o s i t s , first studied in w e s t e r n E u r o p e , b e c a m e k n o w n as the " W e a l d e n
facies". T h e y w e r e found also in the m a r g i n a l b a s i n s of eastern N o r t h A m e r i c a , on the
Arctic c o a s t s from S p i t s b e r g e n to n o r t h e r n S i b e r i a and A l a s k a , in the A s i a t i c F a r
East, s o u t h e r n Australia, A n t a r c t i c P e n i n s u l a , e t c . T h e s e d e p o s i t s c o n t a i n the major
Early C r e t a c e o u s fossil floras d e s c r i b e d in a series of p a l a e o b o t a n i c a l m o n o g r a p h s
(Heer, 1880; S e w a r d , 1 8 9 4 - 1 8 9 5 ; F o n t a i n e , 1899; Berry, 1 9 1 1 ; W a l k o m , 1 9 1 9 ; O i s h i ,
1940; Krassilov, 1967; D o u g l a s , 1969; W a t s o n , 1969, W a t s o n & S i n c o c k , 1992, etc.).
Rich fossil plant localities o c c u r also in the lacustrine d e p o s i t s of i n t r a c o n t i n e n t a l
rifts (e.g. Krassilov, 1982b) that traversed the Asiatic land m a s s from n o r t h e r n C h i n a
across M o n g o l i a , T r a n s b a i k a l i a and A m u r l a n d to the S e a of O k h o t s k w h e r e they j o i n e d
the West Pacific m a r g i n a l fault z o n e .
T h e t e c t o n i c style c h a n g e d after the m i d - C r e t a c e o u s A u s t r o a l p i n e o r o g e n y that
affected eastern part of the T e t h y s belt, as well as the Pacific m a r g i n s w h e r e the ter­
restrial v o l c a n i c r i d g e - b a c k arc t r o u g h - island arc - t r e n c h s y s t e m s h a v e b e e n laid
d o w n . T h e s e e v e n t s in the fold belts w e r e a c c o m p a n i e d by s u b m e r g e n c e of c r a t o n i c
areas w o r l d w i d e .
In the L a t e C r e t a c e o u s , d e p o s i t i o n c o n t i n u e d , t h o u g h on a r e d u c e d scale, in m o s t
of the earlier a p p e a r i n g m a r g i n a l b a s i n s . H o w e v e r , the m a j o r d e p o s i t i o n a l foci shifted
to the s a g g i n g c o n t i n e n t a l interiors flooded by s h a l l o w s e a s . T h e richest fossil floras
recording the rise of a n g i o s p e r m s c a m e from the m a r g i n a l facies of i n t r a c o n t i n e n t a l
basins e x t e n d i n g o v e r central and northern E u r o p e , N o r t h A m e r i c a , K a z a k h s t a n , etc.
(Velenovsky, 1 8 8 9 ; K n o w l t o n , 1930; Dorf, 1942; V a k h r a m e e v , 1 9 5 2 ; K n o b l o c h &
Mai, 1986, etc.). Epicontinental seas covered most of E u r o p e , with isolated land m a s s e s ,
such as the Baltic and B o h e m i a n m a s s i v e s f o r m i n g large i s l a n d s . S e a w a y s stretched
across N o r t h A m e r i c a from the Beaufort S e a to t h e G u l f of M e x i c o , a c r o s s A s i a from
the C a s p i a n S e a to the K a r a Sea, and a c r o s s Africa from Sirt to t h e G u l f of G u i n e a
 bo?-

>

Fig. 19. Cretaceous land (stippled) and island arcs (circles) (after Krassilov, 1985).

(Fig. 19). Geographically, there were twice as m a n y continents as at present, but the
epicontinental m a r i n e passages were inconstant. In addition, large land m a s s e s e m e r g e d
in the northern North Atlantic (Tule Land), the Beringian Shelf (Beringia) and the Tasman
Sea (Tasmantis). T h e y formed broad, though probably intermittent, land bridges for
intercontinental terrestrial migrations.
T h e next global tectonic event occurred in t w o major stages in the western Tethys
belt, notably in the Z a g r o s and Indus-Tsangpo ophiolite z o n e s , c a u s i n g collision of the
Arabian and Indian terrains to the mainland Asia at about the C a m p a n i a n / M a a s t r i c h t i a n
boundary, as well as in the circum-Pacific belt causing the L a r a m i d o r o g e n y and the
e m p l a c e m e n t s of giant granitic bodies at about the Cretaceous/Tertiary boundary. T h e
latter event w a s a c c o m p a n i e d by e m e r g e n c e of cratonic areas and rifting, with a b u n d a n t
terrestrial basaltic a c c u m u l a t i o n s , such as the D e c c a n traps.

Cretaceous angiosperms
Presently it s e e m s futile to even crudely define the n u m b e r of C r e t a c e o u s a n g i o s p e r m
species, for the major fossil floras of that age have not been revised since the initial
palaeofloristic studies of the 1870's to the 1920's. M o d e m studies, with a few e x c e p ­
tions, focus mainly on selected fossils rather than on fossil floras.
T h e classical monographers tended to identify all fossil plant remains with the present-
day plants. T h e y widely applied generic n a m e s of extant genera, s o m e t i m e s modified
by adding "ites", " o p s i s " , " p h y l l u m " , " x y l o n " or "carpon", j u s t to indicate fossil m a t e ­
rial or the kind of organs. T h e mystery of a n g i o s p e r m origin w a s partly d u e to this
classification practice that produced the impression of a fairly m o d e m aspect of Creta­
ceous a n g i o s p e r m s . T h e m o r e we k n o w of the C r e t a c e o u s a n g i o s p e r m s the less con­
vincing s e e m their a s s i g n m e n t s to extant g e n e r a and even families. T h e " o p s i s " -
''phyllum" n a m e s , although valid taxonomically, scarcely indicate t a x o n o m i c affinities
implied by their etymology. C r u d e phylogenetic and palaeoecological errors resulted,
and s o m e t i m e s still elapse, from uncritical reading, e.g. of Sapindopsis as a representa­
tive of the S a p i n d a l e s .
Of all vegetative organs, a n g i o s p e r m leaves are m o s t distinctive and, in fact, the
majority of a n g i o s p e r m records are detached leaves. It was argued by Berry (1911),
however, that a n g i o s p e r m - l i k e leaf remains from the L o w e r C r e t a c e o u s , in particular,
those from the famous P o t o m a c flora, eastern North A m e r i c a , might actually belong to
gnetophytes. This suggestion may still hold for s o m e , but certainly not all, of the pres­
ently m u c h better studied P o t o m a c a n g i o s p e r m s (Hickey & D o y l e , 1977; U p c h u r c h ,
1984). Yet only typical dicotylid leaves are distinctive, w h e r e a s the scaly, needle-like,
fern-like m o r p h o t y p e s with open venation are shared with other plant g r o u p s and can
easily b e m i s i n t e r p r e t e d . F o r i n s t a n c e , the so-called Dammarophyllum, the linear
parallelinerve leaves from the Late C r e t a c e o u s of Kazakhstan (Shilin, 1986), might
belong to the s a m e plant as the Taldysaja monocotylid flowers (Krassilov & al., 1983).
They are similar to the ''Podozamites " and Krannera m o r p h o t y p e s from the C e n o m a n i a n
of B o h e m i a (Velenovsky, 1889) that are also u n d e r suspicion.
Palaeobotanists are m o r e cautious n o w in c o m p a r i n g C r e t a c e o u s fossils, especially
leaves, with extant genera. Yet recent studies of fossil reproductive structures seem to
fall into the s a m e trap as the classical "leaf p a l a e o b o t a n y " in assigning their material if
not to extant genera, then to extant families, such as the C h l o r a n t h a c e a e , Platanaceae,
Magnoliaceae, etc. T h e following review will show that in most such familial assign­
m e n t s t h e i n c o n s i s t e n t or a m b i g u o u s f e a t u r e s are u n d e r r a t e d , e.g. in a l l o c a t i n g
Nordenskioldia in the T r o c h o d e n d r a c e a e (Crane & al., 1 9 9 1 ; M a n c h e s t e r & al., 1991),
a presently m o n o t y p i c family bypassing the differences in the inflorescence m o r p h o l ­
ogy, fruit type, etc. (see b e l o w ) . It would be scarcely appropriate to e x t e n d familial
diagnoses by introducing features based on fossil material, but lacking in the extant
representatives and with no intermediate forms in sight. T h o s e palaeobotanists w h o
tend to neglect distinctions between the C r e t a c e o u s fossil and their superficially similar
modern forms are perpetuating the " a b o m i n a b l e m y s t e r y " of a n g i o s p e r m origins by
implying that a n u m b e r of the present-day families are r e c o g n i z a b l e already in the ear­
liest a n g i o s p e r m records. Actually, fossil evidence s e e m s to suggest that families of the
extant a n g i o s p e r m classification are scarcely applicable to C r e t a c e o u s a n g i o s p e r m s ,
and even the ordinal a s s i g n m e n t s are a m b i g u o u s . With this in mind, the following table
is to be considered as indicating tentative affinities at most. T h e terminal C r e t a c e o u s
(late Maastrichtian) records of m o r e obviously m o d e r n alliances are not included.
Order Plant organ Appeared
Alismatales lAlismatophyllum leaves (Berry, 1911) Albian
ILiliacidites pollen Aptian
Arales Fruits (Krassilov & Makulbekov, 1995) Senonian
Arecales Leaves, pollen (Bell, 1957; Muller, 1984) Senonian
Pa/m^jcy/oAz wood (Berry, 1916) Senonian
Annonales ITriplicarpus fruits (Velenovsky & Viniklar, 1926) Cenomanian
Ceratophyllales Fruits (Dilcher & al., 1996) Aptian
Chloranthales Androecia (Friis & al., 1986; Crepet & Nixon, 1994) Turonian
IClavatipollenites pollen (Walker & Walker, 1984) Neocomian
Cornales Fruits (Krassilov & Martinson, 1982) Turonian
Ericales Fruits (Knobloch & Mai, 1986) Turonian
y4c///zoc«/vjc flower (Friis, 1983) Senonian
Fagales Nothofagidites pollen Senonian
Graminoids Taldysaja, inflorescence (Krassilov & al., 1983) Senonian
Gerofitia inflorescence (Krassilov & Dobruskina, 1997) Turonian
Restionaceous pollen (Hochuli, 1979) Senonian
Geraniales Sarysua inflorescence (Krassilov & al., 1983) Senonian
Hamamelidales ILesqueria inflorescence (Crane & Dilcher, 1984) Cenomanian
Klikovispermum seed (Knobloch & Mai, 1986) Cenomanian
Trochodendrocarpus fruits (Krassilov, 1979) Senonian
Juglandales (Myricales) Dryophyllum leaves (Jones & Dilcher, 1990) Cenomanian
IRamonicarya fruit (Krassilov & Dobruskina, 1995) Aptian
Caryanthus fruit (Knobloch & Mai, 1986) Cenomanian
Flowers, yVorma/7c?//^5 (Friis, 1983, 1985) Cenomanian
Laurales A raliaephyllum, Magnoliaephyllum
Proteoides, etc., leaves Albian
Prisca inflorescence (Retallack & Dilcher, 1981c) Cenomanian
Mauldinia inflorescence(Drinnan & al., 1990) Cenomanian
Virginianthus (Friis «fe al.,1994) Albian
Flowers (Crepet & Nixon, 1994) Turonian
Magnoliales Seeds (Knobloch & Mai, 1986) Turonian
Pollen (Muller, 19840) Senonian
Nymphaeales ? Afropollis lAsteropolUs pollen Neocomian
Modern type pollen (Muller, 1984),
seeds (Knobloch & Mai, 1986) Senonian
Paeoniales Hyrcantha inflorescence, leaves (Krassilov & al., 1983) Albian
Platanoids IParaphyllantoxylon, Icacynoxylon wood
(Serlin, 1982b; Thyan & al., 1985) Albian
Leaves, heads (Crane & al., 1986) Albian
Sarbaya head (Krassilov & Shilin, 1995), pollen Cenomanian
Ranunculids Cissites, Vitiphyllum, Menispermites leaves Albian
Caspiocarpus inflorescence,
leaves (Vakhrameev & Krassilov, 1979) Albian
Appomattoxia fruit (Friis & al., 1995) Albian
Prototinomiscium seed (Knobloch & Mai, 1986) Turonian
Resales Asterocelastrus fruit (Krassilov & Pacltova, 1989) Cenomanian
Sapindales Insiticarpus fruit (Knobloch & Mai, 1986) Cenomanian
Pollen (Muller, 1984) Senonian
Saxifragales Scandianthus flower (Friis & Scarby, 1982 Senonian
Laramisemen seed (Knobloch & Mai, 1986) Cenomanian
Trochodendrales ICaloda inflorescence (Dilcher & Kovach, 1986) Cenomanian
IProtomonimia flower (Nishida & Nishida, 1988) Turonian
lElsemaria flower (Nishida, 1994) Senonian
Urticales Seeds (Knobloch & Mai, 1986) Cenomanian
Pollen (Muller, 1984) Turonian

T h e following supraordinal grouping of s o m e phylogenetically significant records is

occasionally different from their original a s s i g n m e n t s , reflecting the present a u t h o r ' s
opinion of their systematic affinities.

Chloranthoids (piperoids)
C h l o r a n t h o i d affinities have been suggested for the earliest dispersed a n g i o s p e r m
pollen grains Clavatipollenites hughesii (Couper, 1958; W a l k e r & Walker, 1984) re­
sembling those of the Qxi3x\iAscarina in both the external m o r p h o l o g y and exinal struc­
tures. O t h e r w i d e s p r e a d E a r l y C r e t a c e o u s p o l l e n m o r p h o t y p e s , Asteropollis and
Stephanocolpites, h a v e been in their turn c o m p a r e d with the extant c h l o r a n t h a c e a n gen­
era Chloranthus and Hediosmum (Walker & Walker, 1984). Clavatipollenites grains
have also been found stuck to small flattened unilocular o n e - s e e d e d Couperites-iypt
fruits from the L o w e r C r e t a c e o u s of North A m e r i c a and Portugal (Pedersen & al., 1 9 9 1 ;
Friis & al., 1994). T h e fruits have been c o m p a r e d to those of the C h l o r a n t h a c e a e , al­
though the fairly similar fruit m o r p h o l o g i e s occur in a n u m b e r of m o n o c o t families,
such as the Juncaceae or Cyperaceae. Pollen morphologies analogous to Clavatipollenites
and Asteropollis can be found also in the C a b o m b a c e a e ( N y m p h a e a l e s ) . At the s a m e
time, pollen grains c o m p a r a b l e with Clavatipollenites minutus h a v e b e e n o b t a i n e d from
the Albian flowers related to the C a l y c a n t h a c e a e (Friis & al., 1994).
T h e reportedly chloranthoid androecial r e m a i n s from the u p p e r Albian of M a r y l a n d ,
eastern N o r t h A m e r i c a , h a v e been described as consisting of three s t a m e n s fused at the
base and each bearing t w o opposite pairs of pollen sacs ( C r a n e & al., 1986, 1989).
However, the illustrations could also be interpreted as s h o w i n g three distinct, though
adpressed, s t a m e n s of an i n c o m p l e t e staminate flower, w h e r e a s the pollen is apparently
tricolpate and not unlike that of the C r e t a c e o u s platanoids (see b e l o w ) h a v i n g stamens
with the likewise p r o d u c e d c o n n e c t i v e a p o p h y s e s . In this particular fossil, the platanoid
affinities s e e m m o r e probable than the chloranthoid, t h o u g h these t w o g r o u p s might
h a v e been closely related (see b e l o w ) .
Chloranthistemon endressii from the S e n o n i a n of S w e d e n ( C r a n e & al., 1989) is an
androecial r e m a i n of three s t a m e n s fused by their short filaments. T h e i r m a s s i v e an­
thers consist of two bisporangiate thecae in the median stamen and a single bisporangiate
theca in each of the lateral s t a m e n s . T h e c o n n e c t i v e a p o p h y s e s are glandular. T h e
dehiscent sporangia are valvate. T h e pollen grains are spiraperturate, a m o r p h o l o g y
occurring in n o extant C h l o r a n t h a c e a e . In Chloranthistemon crossmanensis, a similarly
flattened tripartite androecial structure from the Turonian of North A m e r i c a , the pollen
aperture configuration is variable even within a pollen sac ( H e r e n d e e n & al., 1993;
Crepet & N i x o n , 1994).

Lauroids
Lauroid leaves a p p e a r a m o n g the earliest a n g i o s p e r m records and are represented by
Laurophyllum, the entire leaf blades with e u c a m p t o d r o m o u s venation. T h i s leaf genus
overlaps with Magnoliaephyllum, the latter c o m p r i s i n g larger leaves with the blade
e n d i n g in a drip-tip and decurrent on the petiole, with several strands d i s c e r n i b l e in the
midrib and with interstitial veins b e t w e e n the c a m p t o d r o m o u s lateral veins. C o m m o n
since the m i d - C r e t a c e o u s are the entire to apically dentate Daphnogene-iype cinnamo-
moid leaves ( K v a c e k & K n o b l o c h , 1967) as well as the entire to trilobed Lindera-Vike
leaves described u n d e r different n a m e s , of which the e t y m o l o g i c a l l y unfortunatey4ra//^-
ephyllum is validated by the principle of priority (Krassilov, 1973c). Eucalyptophyllum
may p r o v e to h a v e been an early lauroid form with a marginal vein and p o l y m o r p h i c
(paracytic, hemiparacytic or, occasionally, anomocytic to cyclocytic) stomata (Upchurch,
1984). In the geologically y o u n g e r forms, including the n a r r o w - l e a v e d 'Troteoides"
m o r p h o t y p e with inconspicuous secondary veins (Krassilov, 1973c), the s t o m a t a are
typically laterocytic.
Prisca Retallack & Dilcher (198 Ic) from the Late C r e t a c e o u s ( C e n o m a n i a n ) of North
A m e r i c a is a loose panicle, the terminal b r a n c h e s of w h i c h are spikes of spiral follicles
that d e v e l o p inside calyptras. T h e s e floral structures are associated with lauroid leaves
of the Magnoliaephyllum type and share with t h e m the laterocytic stomatal structure.
Similar calyptrate structures are described in the C e n o m a n i a n inflorescence Mauldinia
(Drinnan & al., 1990).
T h e well-preserved calyptroid flowers from the Turonian d e p o s i t s of the Atlantic
Coastal Plain (Crepet & N i x o n , 1994) are trimerous, with laminar s t a m e n s and whorled
to spirally disposed follicular carpels sunken in a receptacular h y p a n t h i u m . T h e adaxial
pollen sacs contain m o n o s u l c a t e pollen grains. T h e carpels are c r o w n e d with the bulg-
ing peltate or bilobed stigmas and contain winged ovules on the marginal placentae.
Between the carpels and stamens there are sterile structures described as pistillodes.
These flowers are said to share e u p o m a t o i d and calycanthoid features.

Platanoids
A n g i o s p e r m w o o d r e m a i n s first a p p e a r i n g in the A l b i a n are a s s i g n e d to the w o o d -
genera Icacynoxylon a n d Paraphyllantoxylon, the latter w i d e s p r e a d in the L a t e Creta­
ceous ( M a d e l , 1962; Serlin, 1982b; T h a y n & al., 1983, 1 9 8 5 ; T h a y n & T i d w e l l , 1984),
reaching u p to 1 m in d i a m e t e r in the T u r o n i a n ( C a h o o n , 1972). In both w o o d types,
the n a m e s imply w r o n g affinities, for they s h o w c h a r a c t e r i s t i c features of the w o o d of
extant p l a n e s as well as of the Tertiary Plataninium that differs from Icacynoxylon
pittiense in the ray structure alone (Thien & al., 1985). T h e scalariform perforation
plates of the latter taxon are p r i m i t i v e in h a v i n g u p to 3 0 bars against less than 2 0 in
the extant Platanus.
Large lobed to entire leaves with a coarse pinnate-palmate venation pattern are promi­
nent in the C r e t a c e o u s plant localities since the late Albian and are assigned either to
Palatanus or to the leaf-genera Platanites, Credneria, Protophyllum, Aspidiophylliim,
etc. (e.g. Vakhrameev, 1952). Epidermal characters w e r e studied in a n u m b e r of such
leaves (Fig. 20), yet none of the C r e t a c e o u s species s h o w e d the stomatal structures with
raised subsidiary cells characteristic of extant Platanus. S o m e m o r p h o t y p e s include the
peltate perfoliate leaves with several pairs of infrabasal veins. In the extant Platanus,
perfoliation is characteristic of stipules rather than of foliage leaves. In addition, the
Cretaceous platanoids often show marginal glands that are lacking in the plane leaves.
These o b s e r v a t i o n s s u g g e s t that p l a t a n o p h y l l s c o m p r i s e forms c o n v e r g e n t l y similar
in leaf s h a p e and venation rather than closely related. Actually, s o m e r e p r o d u c t i v e
structures a s s o c i a t e d with p l a t a n o i d leaves are related to the A l t i n g i a c e a e rather than
to the P l a t a n a c e a e (see b e l o w u n d e r h a m a m e l i d s ) , these t w o g r o u p s b e i n g superfi­
cially similar in both the leaf and inflorescence m o r p h o l o g i e s , but not necessarily
close p h y l o g e n e t i c a l l y .
The earliest platanoid records from the late Albian of M a r y l a n d , eastern North
America ( C r a n e & al., 1986), represent p e n t a m e r o u s pistillate and staminate flowers,
the latter producing pollen that was described as tricolpate, but with an exposed ectocolpus
characteristically gaping in the middle (Crane & al., 1986.), as in tricolporate forms,
(e.g. Sarbaya radiata Krassilov & Shilin, 1995),
Sarbaya is a fruiting axis from the C e n o m a n i a n - T u r o n i a n of K a z a k h s t a n , bearing
several sessile staminate heads (Plates 3 3 , 34; Fig. 21). T h e flowers are distinct, 4-
staminate, with the tepals m u c h shorter than the s t a m e n s . T h e latter s h o w a short fila­
ment, a m a s s i v e c o n n e c t i v e and the relatively narrow lateral pollen sacs. T h e pollen
grains are semitectate tricolporate with long slit-like ectocolpi slightly g a p i n g over the
porous m e s o a p e r t u r e s . T h e endoapertures are lalongate, forming a thin equatorial zone,
Fig. 20. Cuticle of a Late Cretaceous platanoid leaf found with the staminate heads Sarbaya radiata
(Krassilov & Shilin, 1995): (1) stomata in the areole of tertiary veins, SEM, x2()0, (2) stoma showing a
radially striate peristomatal flange and narrow subsidiary cells, SEM, xl 400.
Fig. 21. Sarbaya radiata Krassil. & Shilin, from the Late Cretaceous (Cenomanian) of Kazakhstan:
staminate heads and individual florets (after Krassilov & Shilin, 1995).

or endocingulum. T h e associated leaves show a platanoid venation pattern and the promi­
nent marginal glands, with the stomata flanked by a thickly cutinized, transversely stri­
ated flange. Of special interest are the pollen grains that a p p e a r tricolpate in S E M , but
show w e l l - d e v e l o p e d p o r o u s structures in transmitted light. T h e e n d o c i n g u l a t e struc­
tures r e s e m b l e those of s o m e extant rosaceans (Retisma, 1966; Van L e e u w e n & al.,
1988) indicating possible phylogenetic links of the early platanoids with the rosid group.

Nymphaeids
The leaf m o r p h o l o g i e s typical of the N y m p h a e a l e s and N e l u m b o n a l e s (Nelumbites)
appear early in the C r e t a c e o u s records, but their n y m p a h e i d affinities remain p r o b l e m ­
atic in the a b s c e n c e of substantiating anatomical or reproductive e v i d e n c e . T h e earliest
angiosperm pollen m o r p h o t y p e s Afropollis and Asteropollis are currently related to
Magnoliales, although their apertural characters o c c u r also in the C a b o m b a c e a e and
other families of n y m p h a e a l e a n affinities, while s o m e early tricolpates m i g h t have been
produced by aquatic plants related to n e l u m o l e a n s . H o w e v e r the typical n y m p h a e i d
seeds appear not earlier than the Senonian ( K n o b l o c h & M a i , 1986).
 Ranunculids
Fossil ranunculids are represented by the leaf m o r p h o t y p e s characteristic of the
present-day R a n u n c u l i d a e as well as by floral structures b o r n e in c y m o s e or r a c e m o s e
inflorescences, with n u m e r o u s non-laminar stamens, tricolpate pollen grains, apocarpous
or partly s y n c a r p o u s gynoecia of m a n y laterally c o m p r e s s e d , thin-walled, ventricidal
follicles that are spiral or whorled on the apically protruding (toroid) receptacles, and
with ventribasal, anatropous, typically bitegmic ovules. T h e s e floral features are shared
with such orders as the Illiciales and T r o c h o d e n d r a l e s , usually assigned to magnoliids
or, the latter, also to h a m a m e l i d s .
R a n u n c u l i d s in the a b o v e sense are fairly distinct from the typical m a g n o l i i d s that
are characterized by solitary flowers, laminar s t a m e n s , m o n o s u l c a t e pollen and thick-
walled dorsicidal carpels, as well as from the typical h a m a m e l i d s h a v i n g bicarpellate
gynoecia. T h e early Paeonia-Wke forms seem related to ranunculids s.l. (Krassilov &
al., 1983).
P r o m i n e n t in the early a n g i o s p e r m leaf records are small leaves g r a d i n g from the
c o m p o u n d ternate or p a l m a t e to palmately lobed to entire reniform or peltate, with
a c r o d r o m o u s or a c t i n o d r o m o u s venation. S o m e m o r p h o t y p e s of this g r o u p , such as
''Cissites'\ are obviously similar to the leaves of extant Trollius\ Acontium, Anemone
and other ranunculacean genera. T h e y are linked by intermediate leaf m o r p h o l o g i e s
with the c o m p o u n d Vitiphyllum on the one hand, and with the indistinctly lobed or
entire Menispermites, Proteaephyllum and Populophyllum on the other. T h e most c o m ­
plete series of ranunculoid m o r p h o t y p e s (though not recognized as such in the original
systematic treatments) c a m e from the Albian P o t o m a c G r o u p of M a r y l a n d , eastern North
A m e r i c a (Fontaine, 1899; Berry, 1911; Hickey & D o y l e , 1977). T h e stomatal structures
described for s o m e of t h e m ( U p c h u r c h , 1984) are a n o m o c y t i c or h e m i p a r a c y t i c , with
scarcely specialized subsidiary cells and poorly d e v e l o p e d stomatal ledges. A parallel,
though s o m e w h a t less c o m p l e t e , series in the roughly c o n t e m p o r a n e o u s deposits of
western K a z a k h s t a n is represented by Cissites, Menispermites and "Populus" (''Cerci-
diphyllum "j potomacense (Vakhrameev, 1952). T h e cuticular structure of the Late Cre­
taceous Menispermites katicie (Krassilov & al., 1988) is m u c h like in the Albian repre­
sentatives. T h e tracheary e l e m e n t s of the primary veins s h o w circular to slit-like pits in
conjunction with helical thickenings (Fig. 22). Characteristic of the C r e t a c e o u s ranun­
culids was an irregular, " d i s o r g a n i z e d " venation pattern c o n s i d e r e d as a primitive fea­
ture (Hickey, 1973). However, s o m e extant ranunculids h a v e an even m o r e archaic
open venation, as in Circaeaster.
Caspiocarpus from the middle Albian of Kazakhstan ( V a k h r a m e e v & Krassilov,
1979) is a leafy shoot bearing a terminal paniculate infructescence (Plates 3 5 , 3 6 ; Fig.
23). T h e shoot axis is thin, grooved, slightly e x p a n d e d at the leaf n o d e . T h e leaves are
opposite, petiolate, with the blade palmately lobed, of the Cissites m o r p h o t y p e . The
panicles arise at a short distance above the leaf n o d e , tripartite, with the central part
elongate, cylindrical, tapering toward a terminal loose r a c e m e and the lateral branches
Fig. 22. Menispermites katicie Krassil., an example of ranunculoid leaf morphotype from the terminal
Cretaceous - lowermost Paleocene of the Lesser Kuril Islands (Krassilov & al., 1 9 8 7 ) : ( 1 ) leaf, x2, (2)
differentiation of the upper epidermis into costal and intercostal zones, the latter with papillate cells,
S E M , xl 500, (3) tracheary elements of the midrib, S E M , xl 500.
Fig. 23. Caspiocarpus paniculiger Krassil. & Vachr., a ranunculoid angiosperm from the Early Creta­
ceous of Kazakhstan: flowering shoot with leaves, a dehiscent follicle and ovule (after Vakhrameev &
Krassilov, 1979; Krassilov, 1984).

m u c h shorter, m o r e strongly c o n d e n s e d , conical or capitate. T h e inflorescence b r a n c h e s

bear secondary paraclades in the form of spirally arranged r a c e m e s with 4-5 to 10 folli­
cles each. T h e latter are spirally arranged, imbricate, elliptical, a b o u t 1 m m long, with a
short beak, pubescent, ventricidal but also dehiscing a short d i s t a n c e dorsally, occasion­
ally preserved with m i n u t e a p p e n d a g e s at the base, p e r h a p s representing persistent peri­
anth lobes. T h e ovules are 1-3 per follicle, small, flattened, a n a t r o p o u s , b i t e g m i c , with
the outer integument t w o cells thick, forming a m i c r o p y l e , and the inner integument
basally three cells thick, w e d g i n g out a b o v e the midlength of the o v u l e . T h e nucellus is
relatively m a s s i v e , free distally, shortly beaked.
In this fossil, the thin ventricidal follicles are typical of the R a n u n c u l a c e a e , while both
ventral and dorsal dehiscences occur in Glaucidium and in the Berberidaceae. T h e ovule
morphology is likewise ranunculoid. Of the t w o integuments, the outer o n e is typically
shorter in the Ranunculaceae, but the reverse relations occur in the extant Aquilegia,
Hydrastis and in the Berberidaceae. T h e outer integument is extremely thin in Ranuncu­
lus and even thinner in Circaeaster where it is only t w o cells thick. T h e shoot habit sug-
gests a herbaceous life-form. T h e leaves are comparable with many extant ranunculids as
well as with the fossil morphotypes assigned to Cissites and Vitiphyllum.
A similar shoot from the Early C r e t a c e o u s (Aptian) K o o n w a r r a locality, southern
Australia (Taylor & Hickey, 1990), bears small, long-petiolate, r h o m b o i d leaves with a
pinnate-palmate venation. T h e leaves are attached b e l o w a r a c e m o s e bracteate inflores­
cence or infructescence of n u m e r o u s c r o w d e d follicles. A c h l o r a n t h a c e a n affinity has
been suggested on the basis of the axillary bracts or bracteoles, while the leaf c o m p a r i ­
sons included the S a u r u r a c e a e and Aristolochiaceae. Both the leaves and the infruc­
tescence r e s e m b l e those of the slightly y o u n g e r Caspiocarpus, but the details of the
follicles and ovules are as yet not available for the Australian fossil.
Archaeanthus from the C e n o m a n i a n of Kansas, North A m e r i c a (Dilcher & Crane,
1984), is a strobilate gynoecial organ borne apically on a peduncle that shows many small
stamen scars and the larger perianth scars. An elongate receptacle bears 100-130 spirally
arranged, shortly pedicellate follicles (Fig. 24) that are thin-walled, transversely striate,
ventricidal. T h e ovules are about 100 per follicle, small, anatropous. T h e bud scales are
calyptral. T h e associated leaves are petiolate, bilobed, of the Liriophyllum type.
T h o u g h originally assigned to the M a g n o l i a c e a e , Archaeanthus is in fact closer to
ranunculids on account of n u m e r o u s free follicles (in m a g n o l i i d s , the follicles are typi­
cally c o n n a t e , or, if free as in the Austrobaileyaceae, then m u c h less n u m e r o u s ) that are
thin-walled (rather than thick-walled as in magnoliids) and contain a great n u m b e r of
small ovules. F u r t h e r m o r e , the stamens are n u m e r o u s , leaving small r o u n d e d scars (in
the E u p o m a t i a c e a e alone a m o n g the magnoliids, the s t a m e n s are likewise n u m e r o u s but
at least the lower ones are laminar). T h e leaves r e s e m b l e those of Liriodendron, but
differ in the venation with acroscopically laminaless distal b r a n c h e s , as in Hellehorus.
The resin ducts in the p a r e n c h y m o u s tissues occur not only in m a g n o l i i d s , but also in
the M e n i s p e r m a c e a e .
Caloda delevoryana from the C e n o m a n i a n of K a n s a s , North A m e r i c a (Dilcher &
Kovach, 1986), is a catkin-like r a c e m e of dense, spirally arranged, pedicellate flowers
showing a whorl of m a n y distinct follicles on a slightly e x p a n d e d receptacle. Similar
structures h a v e been described from the Senonian of Sakhalin (Krassilov, 1979). Such
apparently primitive gynoecial m o r p h o l o g i e s might have e v o l v e d into a m o r e c o m p a c t
syncarpous structure, such as Elsemaria from the Senonian of H o k k a i d o , Japan (Nishida,
1994), with ten laterally adnate follicles whorled around the central c o l u m n and con­
taining u p to 4 0 p e n d u l o u s anatropous bitegmic ovules.
Protomonimia from the Late C r e t a c e o u s (Turonian) of Japan is based on the p e r m i -
neralized sectioned material (Nishida & Nishida, 1988) and is reconstructed as a glo-
bous apocarpous g y n o e c i u m of m a n y spirally arranged follicles on a w o o d y p e d u n c l e
supplied with collateral vascular bundles that contain-vessels with o b l i q u e scalariform
perforation plates. T h e follicles are thick-walled, with a n o m o c y t i c stomata, and s h o w ­
ing adaxial stigmatic crests. T h e ovules are n u m e r o u s , marginal, sessile, anatropous
bitegmic, with both the outer thick and the inner m e m b r a n o u s i n t e g u m e n t s forming a
collar-like extension around the micropyle.
 - - -• ^

Fig. 24. Angiosperms from the Cenomanian Dakota Formation, Kansas, North America (Dilcher & Crane,
1984; Retallack & Dilcher, 19Slc): (\) Archaeanthus, pistillate spike, x l , (2) Prisca, bracteolate panicle,
xl, (3) Liriophyllum, supposed leaf of the Archaeanthus plant, xl (courtesy of D.L. Dilcher)
 This fossil w a s c o m p a r e d with the M o n i m i a c e a e on the basis of the supposedly con­
cave receptacle, although s o m e cross-sections s e e m to suggest a sterile central c o l u m n .
The a n o m o c y t i c stomata is a ranunculid feature that m a k e s a lauralean affinity p r o b ­
lematic. T h e general structure of the g y n o e c i u m , as well as the carpel morphology,
seem to support a t r o c h o d e n d r a c e a n affinity that w a s indicated by N i s h i d a & Nishida
(1988) as an alternative possibility. Similar gynoecial structures are described from the
Turonian of Japan by G h a n a & K i m u r a (1987).
Nordenskioldia, widespread in the terminal Cretaceous and Paleocene (Heer, 1970), is
based on the fruits that are borne in the long, loosely spicate infructescences that may bear
proximal buds and d i s m e m b e r e d flower remains, the latter showing thin m e m b r a n o u s
tepals. T h e fruits are sessile, alternate or in pairs, c o m p o u n d , consisting of about 20 folli­
cles enclosed in the ribbed bivalved structures inteipreted as either exocarps (Manchester
& al., 1991) or calyptras (Krassilov & Fotyanova, 1995). T h e s e structures spread disk­
like and remain on the axis from which the fruits are shed (Fig. 25). T h e follicles are
radially disposed around the central column, or t6rus, laterally flattened and adpressed,
proximally connate but separable when ripe, with shortly reflexed free styles, and con­
taining a solitary seed. T h e associated leaves are simple, entire, elliptical to reniform,
palmatinerve, with two strong acrodromous laterals that branch on the outside.
Nordenskioldia has been c o m p a r e d with Trochodendron on a c c o u n t of the similarly
arranged follicles (Kryshtofovich, 1958; C r a n e & al., 1990; M a n c h e s t e r & al., 1991).
However, in Trochodendron the leaves are pinnatinerve, the inflorescence is c y m o s e
rather than spicate, the flower lacks both perianth and calyptra, and the fruit is a septicidal
capsule rather than a follicetum. T h e m o r e similar leaf and fruit structures, the latter
sometimes calyptrate and typically consisting of n u m e r o u s w h o r l e d a p o c a r p o u s folli­
cles, can be found in the extant M e n i s p e r m a c e a e , while the o n e - s e e d e d follicles spread­
ing radially on toroid receptacles are typical of the Illiciaceae. Nordenskioldia thus
appears to belong in the ranunculid circle, but represents an extinct family.
Hyrcantha from the M i d d l e Albian of Kazakhstan (Krassilov & al., 1983) is a loose
paniculate or thyrsoid inflorescence, bracteate, with the b r a n c h e s r a c e m o s e , axillar to
small bracts, the lower ones forked. T h e flowers are pedicellate, m o n o c l i n o u s , with a
calyx of small tepals, persistent and p e n d e n t at fruit (Plates 3 7 , 3 8 ; Fig. 26). T h e stamens
are n u m e r o u s , with slender filaments, shorter than carpels. T h e g y n o e c i u m is apocarpous
of 3-5 follicles, the latter are flask-shaped, thick-walled, ventricidal, c r o w n e d by a broad
sessile stigma. T h e associated leaves are c o m p o u n d ternate (Fig. 2 7 ) , with the leaflets
shortly petiolulate, elliptical or oblanceolate, microserrulate, s h o w i n g the scattered a n o ­
mocytic stomata.
S i m i l a r b r a c t e a t e inflorescences b e a r i n g m o n o c l i n o u s flowers with n u m e r o u s sta­
mens and tricarpellate a p o c a r p o u s g y n o e c i a o c c u r in Thalictrum, an isolated g e n u s of
ranunculid a l l i a n c e . H o w e v e r , in c o m b i n a t i o n with the p e r s i s t e n t c a l y x and sessile
stigma, t h e s e c h a r a c t e r s suggest a m o r e c l o s e affinity with Paeonia, formerly as­
signed to the R a n u n c u l a l e s but presently m o r e often c o n s i d e r e d as r e p r e s e n t i n g a
Fig.25. Nordenskioldia horealis Heer, a representative of the extinct angiosperm family Nordenskioldiaceae
from the Paleocene of Tsagajan, Amur Region, Far East (Krassilov, 1976, 1995; Krassiliov & Fotyanova,
1995): (1-5) fruiting spikes with open calyptras persistent on the axis, x3 and 5, (6) intlorescence axis
with flower buds, x l , (7, 9, 10) detached calyptras and fruits, x3 and 5, (8) axis of a fruit with detached
fruitlets, x5, (11) leaf, supposedly of the same plant, x 1.
Fig. 26. Hyrcantha karatscheensis (Vachr.) Krassil., a paeonialean intlorescence from the Early Creta­
ceous of Kazakhstan and individual llowers (after Vakhrameev, 1952; Krassilov & al., 1983).
Fig. 27. "Leguminosites" karatscheensis Vachr, ternate leaves found with Hyrcantha karatscheensis
(Vachr.) Krassil., a paeonialean inflorescence from the Early Cretaceous of Kazakhstan, in comparison
with leaves of the extant Paeonia caucasica (bottom left), after Krassilov & al. (1983).

m o n o t y p i c o r d e r related to t h e D i l l e n i a l e s mainly on a c c o u n t of the d e n d r o i d androecial

traces (Cronquist, 1981). T h e associated leaves h a v e their m o r p h o l o g i c a l analogues in
the extant Paeonia caucasica.
Appomattoxia is a spiny fruit from the P o t o m a c G r o u p r e s e m b l i n g Circaeaster {¥xm
& al., 1995). It is unilocular, with a solitary o v u l e that is d e s c r i b e d as pendulous
orthotropous. M o n o s u l c a t e pollen grains found in the stigmatic area m a y suggest a
p i p e r o i d - m o n o c o t y l i d , rather than ranunculid, affinity ( C r a n e & al., 1994). Superfi­
cially at least, these fruits resemble Tyrmocarpus, the spiny capsules from the Neocomian
of the T y r m a Basin, eastern Asia (Krassilov, 1977a, 1989).
 Hamamelids
As suggested a b o v e u n d e r " p l a t a n o i d s " , a n u m b e r of the plane-like leaf species, or
platanophylls, might actually belong to hamamelids that are closer to the H a m a m e l i d a c e a e
or the Altingiaceae than to the Platanaceae. A m o n g t h e m , ''Platanus" reynoldsii, a leaf
species w i d e s p r e a d in the terminal Cretaceous and in the P a l e o c e n e , associated with
Steinhauera, the altingioid pistillate heads, and with Tricolpopollianthus, the staminate
heads of indistinct flowers producing small tricolpate pollen of a h a m a m e l i d a c e a n rather
than platanacean aspect (Krassilov, 1979). Evacarpa, a n o t h e r P a l e o c e n e e x a m p l e bear­
ing on the affinities of early platanoids (Fig. 28), is a definitely hamamelidacean carpellate
head associating with platanoid leaves ( M a s l o v a & Krassilov, 1997). T h e following
fossil g e n e r a here tentatively assigned to h a m a m e l i d s h a v e been differently interpreted
in their original descriptions.
Lesqueria from the C e n o m a n i a n of N o r t h A m e r i c a ( C r a n e & Dilcher, 1984) is a
pedunculate pistillate head with a m a s s i v e c o n v e x receptacle bearing the helically ar­
ranged, laminar a p p e n d a g e s in the peripheral zones and about 2 5 0 c r o w d e d fruitlets in
the central z o n e . T h e fruitlets are elongate, keeled dorsally and containing about 2 0
ovules disposed in t w o r o w s . T h o u g h described as follicles, the fruitlets actually show
two stylodes, are distally split in the median plane, and leave r h o m b o i d areoles when
shed. T h e y are thus c o m p a r a b l e with bicarpellate h a m a m e l i d g y n o e c i a i m m e r s e d - less
deeply than in the geologically y o u n g e r forms - in the hypanthial areoles, while the
whole structure r e s e m b l e s the pistillate heads of the P a l e o c e n e Steinhauera (Kirchhei-
mer, 1957; K n o b l o c h , 1960; Krassilov, 1976) and Evacarpa ( M a s l o v a & Krassilov, in
press), as well as of the extant Altingia.
Nyssidium, Jenkinsella, Jojfrea a n d Trochodendrocarpus constitute a g r o u p of closely
related fruit m o r p h o t y p e s appearing in the L a t e C r e t a c e o u s and w i d e s p r e a d in the Early
Tertiary (Heer, 1869; K n o w l t o n , 1930; Reid & Chandler, 1933; Kryshtofovich, 1958;
Krassilov, 1976; C r a n e & S t o c k e y , 1986). T h e y are large r a c e m o s e to p a n i c u l a t e
infructescences of loose, spirally arranged, paired, adaxially dehiscent fruits. T h e latter
contain n u m e r o u s w i n g e d seeds. T h e a c c o m p a n y i n g Trochodendroides-type leaves are
longly petiolate, elliptical to reniform, dentate to crenulate to lobed, with a p a l m a t e
actinodromous to p a l i n d r o m o u s venation pattern.
S o m e variants of t h e s e p o l y m o r p h i c leaves r e s e m b l e t h o s e of the e x t a n t Cercidi-
phyllwn, p r o m p t i n g a s y s t e m a t i c affinity with the latter. T h e r e p r o d u c t i v e structures
are p r o f o u n d l y different, h o w e v e r . Cercidiphyllum, a l t h o u g h s o m e t i m e s p l a c e d in
h a m a m e l i d s , is a fairly isolated g e n u s h a v i n g p s e u d a n t h i a of four w h o r l e d , basically
decussate, b r a c t e a t e follicles that are abaxially d e h i s c e n t and p r o d u c e w i n g l e s s seeds
of irregular form, w h i l e the fossil g e n e r a differ in the r a c e m o s e i n f r u c t e s c e n c e struc­
ture, paired fruits, a d e h i s c e n c e m o d e , and the seed m o r p h o l o g y , all s u g g e s t i n g a
hamamelidalean affinity.
Fig. 28. Evacarpa polysperma N. Maslova & Krassil.. a liquidambroid pistillate head from the P
of Kamchatka, North-East Russia (Maslova & Krassilov, 1997), SEM: (1) part of the head showi
ers of paired carpels surrounded by intrafloral phyllomes; styles are shed leaving elliptical scars,
flower with opened carpels, x3().
 Rosoids
Cretaceous fossils here included in the rosoid g r o u p are characterized by the prevail­
ingly m o n o c l i n o u s d i c h l a m i d e o u s flowers that are solitary or b o r n e in c y m o s e inflores­
cences. T h e perianth is mostly 5 - m e m b e r e d , with the calyx persistent at fruit. T h e
androecium is h a p l o s t e m o n o u s , rarely d i p l o s t e m o n o u s . T h e g y n o e c i u m is syncarpous
2-5-carpellate, with free styles and with t w o to several a n a t r o p o u s ovules per locule.
Asterocelastrus Velenovsky & Viniklar, 1926, e m e n d . Krassilov & Pacltova, 1988,
from the C e n o m a n i a n of B o h e m i a , is a p e n t a m e r o u s a c t i n o m o r p h o u s flower with a
coriaceous stellate calyx, persistent at fruit (Plate 39). T h e calyx lobes are distally free
for about half their length, proximally c o n n a t e and adnate to the ovary. T h e latter is
syncarpous with free styles, forming a 5-locular, radially dehiscent c a p s u l e . T h e ovules
are basal a n a t r o p o u s , t w o per locule. In the characters of syncarpy and the calyx to
ovary fusion, it is closer to R o s a c e a e than to Celastraceae, a formerly suggested alli­
ance. A m o n g the extant R o s a c e a e , the M a l o i d e a is similar in having the e p i g y n o u s
flowers with various extents of syncarpy and synsepaly, as well as in the basal ovules.
In the M a l o i d e a , the ovaries are occasionally 5-locular (Stenvaesia: Sterling, 1965). At
the s a m e t i m e , c a p s u l a t e fruits a r e m o r e c o m m o n in the e x t a n t m e m b e r s of the
Saxifragales that differ from the fossil form in the greater n u m b e r of ovules per locule
and the insignificant perianth to g y n o e c i u m fusion. Remarkably, in Asterocelastrus,
fusion of the floral parts, an allegedly advanced feature, is m o r e e x t e n s i v e than in the
geologically y o u n g e r Paleorosa (Basinger, 1976), as well as in m a n y extant forms.
Scandianthus Friis & Scarby, 1982, from the Late C r e t a c e o u s ( S e n o n i a n ) of S w e ­
den, is similar to the a b o v e form in the p e n t a m e r o u s calyx and c a p s u l a t e fruit, but the
latter is dicarpellate l-locular with m a n y p e n d u l o u s seeds. T h i s form s h o w s a peculiar
c o m b i n a t i o n of c h a r a c t e r s o c c u r r i n g s e p a r a t e l y in different f a m i l i e s of the extant
Saxifragales, such as free sepals (Crassulaceae), a d i m e r o u s g y n o e c i u m (Saxifragaceae)
and the apical p e n d u l o u s placentae (Vahliaceae). Actinocalyx, a n o t h e r floral form from
the s a m e locality (Friis, 1985), also has a p e n t a m e r o u s perianth with distinct sepals but
with fused petals forming a floral tube. T h e a n d r o e c i u m is h a p l o s t e m o n o u s , with sta­
mens basally adnate to the floral tube and shorter than the latter. T h e g y n o e c i u m is
syncarpous, 3-locular, with m a n y anatropous ovules and d e v e l o p i n g into a loculicidal
capsule. It is c o m p a r e d with the D i a p e n s i a c e a e and Ericaceae, but is different in the free
styles and the basal, rather than axial, ovules. T h e latter characters may indicate a rosoid,
rather than dillenioid, affinity.
Sarysua Krassilov & Shilin in Krassilov & al., 1983, from the Senonian of Kazakhstan,
is a c y m o i d inflorescence of m a n y m o n o c l i n o u s d i c h l a m i d e o u s flowers (Figs. 2 9 , 30;
Plate 40) that are a c t i n o m o r p h i c , with the pedicel shorter than the g y n o e c i u m . T h e
receptacle is m a s s i v e , bearing clavate glands b e l o w the s t a m e n s . T h e calyx is persistent
at fruit, with the sepals lanceolate, caudate, reflexed. T h e a n d r o e c i u m is h a p l o s t e m o n o u s ,
of about ten stamens and s t a m i n o d e s , the latter h a m a t e . T h e g y n o e c i u m is syncarpous,
with five free styles of varying length. In the terminal flowers, the g y n o e c i a are shorter
but with longer styles. T h e ovary is 5-lobed, pubescent. T h e styles are glabrous, solid,
with the epidermis and subepidermal tissue similar to those of the stamen filaments.
T h e stigma is terminal, capitate.
C o m b i n e d , the a b o v e characters indicate affinities with the Geraniales a m o n g which
Linum is similar in the dichasial inflorescence, a c t i n o m o r p h i c flower with persistent
calyx, nectaries in the staminate disk, the 5-lobed s y n c a r p o u s ovary and at least partly

Fig. 29. Sarysua pomona Krassil. & Shilin, flowers from the Late Cretaceous of Kazakhstan (Krassilov &
al., 1983): (1) flowers crowded in a manner suggesting a single cymose inflorescence, with shorter flow­
ers in the central part, x3, (2) one of the shorter flowers, xlO.
Fig. 30. Sarysua pomona Krassil. & Shilin from the Late Cretaceous of Kazakhstan: disintegrated cymose
inflorescence (short flowers blackened) and reconstruction, individual flower and carpel with locule stip­
pled (after Krassilov & al., 1983).
free styles. In Oxalis, the styles are free and often g l a b r o u s , bearing capitate stigmas,
while in Geranium the stamens are scarcely c o n n a t e at base and s o m e of t h e m are
reduced to staminodes, the ovary is prominently 5-lobed and often p u b e s c e n t , with the
locules distally contracted and sterile.
Sarysua represents an early s y n c a r p o u s form with solid styles a n d an incipient
heterostyly. In the geranialean genera Linum and Oxalis, heterostyly is involved in a
highly elaborate breeding system (Weller, 1976). R e m a r k a b l y , style lengths in Linum
flavium are typically 6-8 m m and 8-11 m m , almost like the t w o style length m o d e s in
Sarysua. A n o t h e r interesting observation is a striking r e s e m b l a n c e b e t w e e n the stamen
filaments and the styles in diameters and the e p i d e r m a l characters.
A m o n g the C r e t a c e o u s pollen types, Aquilapollenites, a diverse g r o u p of the Late
C r e t a c e o u s dispersed pollen grains d o m i n a n t in the t e m p e r a t e z o n e , is characterized by
the short colpate apertures raised on special a p p e n d a g e s . Superficially similar pollen
morphologies occur in the Santalales (Jarzen, 1977; Wiggins, 1982; Muller, 1984), while
the assocxdXtAProteacidites, Cranwellia and Woodehousia m o r p h o t y p e s w e r e c o m p a r e d
with pollen grains of the Proteaceae, L o r a n t h a c e a e and G e r a n i a l e s respectively (Khlo-
nova, 1961; Muller, 1974, 1984). Such palynolomorphological similarities reflect scarcely
m o r e than r e m o t e affinities. For instance, Proteacidites with annulate pores is c o m p a r a ­
ble not only with the Proteaceae, but also with the O n a g r a c e a e , c o n c e i v a b l y represent­
ing a progenitorial myrtalean stock. Taken together, these pollen types are e v i d e n c e of
basic radiation of the rosoid c o m p l e x .

Myricoids
This g r o u p includes early a n g i o s p e r m s related to t w o closely allied orders, M y r i c a l e s
and Juglandales. An early a p p e a r a n c e of the j u g l a n d i o i d fruit m o r p h o l o g y is e v i d e n c e d
by Ramonicarya Krassilov & D o b r u s k i n a (1995) from the Aptian of Israel (Plate 4 1 ) , a
shortly pedicellate samaroid fruit consisting of a c o m p a r a t i v e l y large bilobed wing,
abaxially attached to a small nutlet. T h e w i n g lobes s h o w an irregular reticulate vena­
tion. On the adaxial face there is a spatulate a p p e n d a g e (winglet) o n l a p p i n g a stylar
part of the nutlet, with its t w o stigmatic arms spreading in the plane parallel to the wing.
This fruit superficially resembles Hooleya, a fossil g e n u s of Platicaryeae that has two
diverging w i n g s with irregular reticulate venation. In the Platicaryeae, however, the
wings are lateral and their primary veins are flabellate. T h e E n g e l h a r d i e a e is closer in
having involucrate fruits the involucres of which consist of a large w i n g - l i k e bract and
a m u c h smaller adaxial a p p e n d a g e , or winglet, formed of bracteoles and perianth lobes.
T h e nutlet is typically pedicellate, with the extended style bearing a b i a r m e d stigma
(Engelhardia), but the bract-wing is trilobed rather than bilobed. T h e lateral veins of
the w i n g lobes in extant Oreomunnea and related fossil forms are parallel to the midrib,
but in Alfaropsis they diverge to the margins as in the Aptian fruit. Alfaropsis is dis­
tinct, however, in a m u c h reduced winglet.
 Gurvanella and Erenia from the Early C r e t a c e o u s of M o n g o l i a (Krassilov, 1982b)
are bilocular winged fruits with persistent funnel-shaped or b i a r m e d s t i g m a s . T h e wing
is reticulate or, in the latter g e n u s , smooth, slightly oblique and avoiding the pedicel.
The latter feature is characteristic of s o m e superficially similar j u g l a n d i o i d fruits, such
as Pterocaryopsis. T h e i r attribution to j u g l a n d i o i d s r e m a i n s as yet p r o b l e m a t i c .
C o m m o n in the Cretaceous and Paleogene are the Deheya-Myricaephyllum-Dryophyllum
leaf m o r p h o t y p e s grading from the c o m p o u n d p a l m a t e to simple, with the leaflets or, in
the latter case, leaves elongate, serrate with c r a s p e d o d r o m o u s or s e m i c a m p t o d r o m o u s
lateral veins. T h e venation pattern in Dryophyllum indicates a j u g l a n d a l e a n rather than
the previously postulated fagalean affinity (Jones & Dilcher, 1990). T h e cuticle, studied
in three Debeya s p e c i e s from t h e L a t e C r e t a c e o u s of S a k h a l i n a n d Kuril I s l a n d s
(Krassilov, 1973c, 1979; Krassilov & al., 1988), is thick s h o w i n g sparse stomata and
several types of hairs and g l a n d s . T h e stomata are a m p h i c y c l i c , with both guard cells
and subsidiary cells sunken below the level of the encircling cells that form a stomatal
pit, and typically with a thick peristomatal ridge as well as the p r o m i n e n t guard cell
ledges. T h e s e leaf m o r p h o t y p e s occur in the Normapolles palynological p r o v i n c e s and
within its e c o t o n e with the Aquilapollenites province ( H e r n g r e e n & C h l o n o v a , 1981).
Some of t h e m might b e l o n g to the Normapolles-produc'ing plants.
The Normapolles-produc'mg flowers are of several types (Friis, 1983, 1985), having
in c o m m o n the bracteate perianths, inferior ovaries and basal o r t h o t r o p o u s o v u l e s .
Caryanthus a m o n g them is a small zygomorphous epigynous flower with a short bracteate
wing at fruit and with t w o bracteoles adnate to the ovary. T h e calyx consists of t w o
heteromorphous - a n a r r o w m e d i a n and a broad lateral - pairs of tepals. T h e s t a m e n s are
6-8, opposite the lateral tepals. T h e g y n o e c i u m s h o w s t w o styles persistent in fruit. In
Manningia, the flowers are a c t i n o m o r p h o u s , with 3-carpellate g y n o e c i a . Antiquocarya
is a fruit with persistent calyx of 6 tepals and with three stylodes.
Normapolles is a diverse group of brevaxonate iso- or heteropolar tri(col)porate forms
with p r o m i n e n t apical apertures. T h e g r o u p first appeared in the C e n o m a n i a n , initially
as the relatively primitive p r o b r e v a x o n a t e forms with simple apertures rapidly evolving
into m o r e elaborate types (Pacltova, 1971, 1977; Kedves, 1982, 1983; K e d v e s & Pardutz,
1983b). T h e i r infrastructure grades from granular to columellar, and they vary also in
the d e v e l o p m e n t of e n d e x i n e which is lacking in Atlanthopollis, but is apertural and
lamellate in Complexipollis. In Oculopollis, a tetrahedral nexine g e o m e t r y has been
described by M e d u s (1983). Z a v a d a & Dilcher (1986) have traced different Normapolles
subtypes of K e d v e s ' classification u p to various groups of amentifers.

Graminoids
Cretaceous plants of graminoid aspect m a y actually include progenitorial forms of
not only the Poales but also the Restionales or C y p e r a l e s or even J u n c a l e s that are not
always distinguishable as fossils. T h e g r a m i n o i d leaf m o r p h o l o g i e s a p p e a r in the Early
Cretaceous as parallelinerve leaves supposedly belonging to the gnetalean proangiosperms
(see above). On the other hand, such Late Cretaceous leaf morphotypes as Dammaro-
phyllum or Krannera might actually belong to graminoids. T h e distinctive pollen grains
of restionaceous affinities are reported from the Maastrichtian (Hochuli, 1979). However,
some early pollen morphotypes, despite their entirely different current assignments, might
actually belong to graminoid monocots. T h u s Walkeripollis tetrads occurring in the Albian
and even earlier, late Barremian to Aptian, pollen assemblages, are usually related to the

Fig. 31. Gerofitia, a cyperaceous corymbiform inflorescence from the Late Cretaceous (Turonian) of
Israel (Krassilov & Dobruskina, 1997), x3.
Magnoliales, although they are calymmate, developing a c o m m o n ectexine, a step in the
direction of cryptotetrad condition characteristic of the present day Cyperaceae. T h e fol­
lowing examples may support a rather early appearance of graminoids.
Gerofitia Krassilov & Dobruskina (1997), from the early Turonian of Israel, is a loosely
paniculate corymbiform infructescense with subopposite ascending or (proximal) pen­
dent branches subtended by the drooping inflorescence bracts and bearing the fascicles
of densely c r o w d e d , but individually discernible, spikelets (Fig. 31). T h e spikelets con-

Fig. 32. Taldysaja medusa Krassil. & Shilin, from the Late Cretaceous of Kazakhstan: ray of fasciculate
intlorescence, reconstruction showing a distal ward transition from short spikelets to solitary flowers (af­
ter Krassilov & al., 1983).
Fig. 33. Taldysaja medusa Krassil. & Shilin, a monocot inflorescence from the Late Cretaceous of
Kazakhstan (Krassilov & al., 1983): (1) fascicle rays bearing spiny bracts and three spikelets, with a
rachilla distinct in the distal spikelet, xlO, (2) bifid peduncular bract or prophyll, xlO.

sist of five to seven spirally arranged imbricate g l u m e s , the distal ones axillating the
achenes that are adaxially enveloped in the shorter scales (prophylls) forming the utricles
with pincers-shaped b e a k s .
Taldysaja K r a s s i l o v & Shilin in K r a s s i l o v & al., 1 9 8 3 , from the S e n o n i a n of
Kazakhstan, is a fasciculate pedunculate inflorescence (Plate 4 2 ; Figs. 32, 33) with a
long peduncle bearing a bifid prophyll and paired bracts. T h e radially spreading rays of
the fascicle are spiny, apically sterile, proximally bearing 1-2-flowered spikelets axillary
to the spines. T h e distal spikelets are reduced to solitary flowers (anthoids) leaving the
raised areole-like elliptical scars w h e n d e t a c h e d . T h e rachillae of the spikelets are
spinulate, with longly protruding sterile apices. T h e flowers show imbricate perianth
scales and a flask-shaped g y n o e c i u m . T h e associated leaves are taeniate parallelinerve,
of the Dammarophyllum'' type (Shilin, 1986).
Taldysaja is notable as an early graminoid with the spines of the rays and the spinules
of the rachillae interpreted as g l u m e s and glumellae respectively. T h e bifid flask-shaped
pistils are cyperoid, while the sterile tips of the rachillae projecting b e y o n d the anthoids
occur in many sedge species. At the s a m e time, the prophyll and inflorescence bract
morphologies, as well as the areolae of the rays are rather of a poalean aspect.
 Phylogenetic relationships of Cretaceous angiosperms
The macrodiversity of C r e t a c e o u s a n g i o s p e r m s increased from about four orders in
the pre-Albian records to about 21 in the Senonian records, that is, at least three-fold in
about 2 0 million years. In g y m n o s p e r m s , a c o m p a r a b l e rise of the ordinal diversity took
about 7 0 million years.
Actually, however, a larger part of the ordinal diversity is d u e to the late Albian-
Cenomanian burst of adaptive radiation when a n g i o s p e r m s first e m e r g e d as a n u m e r i ­
cally important g r o u p . F o r m s of the opposite m o r p h o c l i n e poles, such as h y p o g y n o u s
and e p i g y n o u s flowers or paracarpous and s y n c a r p o u s g y n o e c i a , are recognizable at
that stage already, while, according to the traditional criteria of primitiveness, s o m e
Cenomanian forms (e.g. Asterocelastrus a m o n g the Rosales, see a b o v e ) are even m o r e
a d v a n c e d than the e x t a n t m e m b e r s of t h e g r o u p . T h e s e d a t a s u g g e s t that floral
morphoclines are of little phylogenetic significance.
In the p r e c e d i n g 15 million years, a n g i o s p e r m s w e r e rare and the t e m p o of their
evolution w a s less impressive, with m u c h the s a m e groups represented in the N e o c o m i a n
to Aptian and early Albian records. This initial diversity might h a v e resulted either
from a burst at the yet u n r e c o r d e d low-density m o n o p h y l e t i c starting point or from
polyphyletic origins. H o w e v e r , the former possibility is unlikely b e c a u s e of the appar­
ently negative correlation between rarity and evolution rates. Furthermore, the monophyly
view a s s u m e s that a n g i o s p e r m groups having long geological records should be closer
to each other back in time, which does not s e e m to be the c a s e .
A m o n g the c o m m o n features of the first appearing a n g i o s p e r m s are herbaceous or
small w o o d y g r o w t h forms, c a u l o m i c leaves, r a c e m o s e or capitate inflorescences of
many i n c o n s p i c u o u s flowers, and small seeds. T h e s e are obviously g r a d e characters
related to either d e v e l o p m e n t a l acceleration or dispersal e n v i r o n m e n t s , both discussed
in the following chapters. Similarities scarcely extend b e y o n d these features, though.
Prominent a m o n g the early records are m o r p h o t y p e s that are related to taxonomically
isolated e x t a n t g e n e r a c o n s t i t u t i n g m o n o t y p i c or o l i g o t y p i c f a m i l i e s , such as the
Platanaceae, C h l o r a n t h a c e a e , C e r a t o p h y l l a c e a e , P a e o n i a c e a e , etc. T h e i r higher rank
assignments are, for the most part, a m b i g u o u s , for they d o not fit in any of the well-
defined angiosperm orders. Thus, the Chloranthaceae has been assigned to either Laurales
or Piperales, the latter perhaps closer but still remotely related ( D a h l g r e n , 1983). T h e
Ceratophyllaceae are likewise remotely related to N y m p h a e a l e s , but are similar also to
aquatic m o n o c o t s . O n e v i d e n c e of the fossil record, they w e r e fairly distinct m o r p h o l o ­
gically from the start, never c o m b i n i n g the diagnostic features of each other or of any
larger order. In contrast, the ranunculids were the most diversified g r o u p of the Early
Cretaceous a n g i o s p e r m s , with p o l y m o r p h i c foliar and floral characters e v i d e n c i n g their
great m o r p h o g e n i c potentials. O n the other hand, their affinities to extant taxa are less
obvious than in the case of the m o n o t y p i c g r o u p s .
 Sinse the fossil data are as yet insufficient for intergroup c o m p a r i s o n s , the latter
involve also characters of the allied extant taxa. T h e P l a t a n a c e a e are usually assigned to
h a m a m e l i d s , although sharing with the latter the general aspects of leaves and floral
heads alone while differing in the tricolporate pollen grains, h y p o g y n o u s flowers, apocar­
p o u s g y n o e c i a , o r t h o t r o p o u s o v u l e s and a c h e n o c a r p o u s fruits. T h e o r t h o t r o p o u s ovule
m o r p h o l o g y o c c u r s in m o s t of the m o n o t y p i c g r o u p s , i n c l u d i n g the early a p p e a r i n g
Ceratophyllum-Wkt forms ( D i l c h e r & al., 1996) and the p i p e r o i d line. T h e ovules
arising from the floral a p e x , as the well as the c u p u l a r g y n o e c i a d e v e l o p i n g from
a n n u l a r p r i m o r d i a e n c i r c l i n g the a p e x , as in Piperomia (Tucker, 1986), o c c u r also in
the g r a m i n o i d m o n o c o t s and m y r i c o i d d i c o t s , i n c l u d i n g the J u g l a n d a l e s that p r o d u c e
c a t k i n - l i k e i n f l o r e s c e n c e s , but are o t h e r w i s e different from the h a m a m e l i d amentifers
( T h o r n e , 1973).
T h e fossil platanoids are recognized by the p a l m a t o - p i n n a t e leaf venation, the char­
acteristic stomatal structures with raised subsidiary cells, the likewise peculiar stamens
with longly decurrent pollen sacs and protruding c o n n e c t i v e s , as well as by the one-
seeded carpels and orthotropous ovules. Characteristic of the C r e t a c e o u s platanoids are
the peltate perfoliate leaf morphotypes that are shared with the N y m p h a e a l e s and Piperales
[Piperomia) and are recorded a m o n g the first appearing a n g i o s p e r m s C*Nelumbites":
Berry, 1911). T h e tracheary elements with both scalariform of m a n y bars and porous
perforations are typical of both platanoids and p i p e r o i d s , while the latter approach
m o n o c o t s (Burger, 1977) and n y m p h a e i d s in the scattered vascular b u n d l e s and incon­
sistent d e v e l o p m e n t of vessels (in Sarcandra occurring in the roots: Carlquist, 1987).
Similarities b e t w e e n the piperoids and n u m p h a e i d s include the h e t e r o m o r p h o u s cotyle­
d o n s forming a m a s s i v e haustorium, and other e m b r y o l o g i c a l features s o m e of which
are shared also with m o n o c o t s . T h u s , typical of piperoids are tetrasporic e m b y o sacs,
while in n y m p h a e i d s the coleoptyle usually appears before the root, as in the graminoid
m o n o c o t s ( P h i l o m e n & Shah, 1985).
T h e s e c o m p a r i s o n s m a y e v i d e n c e a c e r t a i n d e g r e e of r e l a t e d n e s s b e t w e e n the
platanoids, piperoids (chloranthoids) and n y m p h a e i d s (ceratophylls) that m i g h t have
arisen as paraphyletic lineages from a n o n - a n g i o s p e r m ancestral g r o u p . T h e y could be
m a r g i n a l l y related to the g r a m i n o i d m o n o c o t s , w h i l e the m i d - C r e t a c e o u s burst of
p l a t a n o i d s m i g h t h a v e g i v e n rise to the a n c e s t r a l r o s o i d s s h a r i n g w i t h t h e m the
p e n t a m e r o u s floral structures and certain palynological characters (see a b o v e under
Sarhaja). At the s a m e time, their fossil record gives n o e v i d e n c e of their either progeni­
torial or fraternal relations with magnoliids. T h e y also s e e m fairly distant from their
c o n t e m p o r a n e o u s early ranunculids.
T h e C r e t a c e o u s ranunculids had simple or c o m p o u n d leaves, a basically dichopodial
venation pattern, the typically a n o m o c y t i c stomata, toroid flowers, n u m e r o u s apeltate
stamens, tricolpate pollen, free to partly connate ascidiform carpels, and the ventribasal
anatropous ovules. Presently ranunculids include several satellite g e n e r a that fall apart
from the major orders and are s o m e t i m e s separated as m o n o t y p i c orders, such as the
Glaucidiales (Takhtajan, 1966), while Kingdonia and Circaeaster arc s o m e t i m e s con­
sidered as the most archaic of the living a n g i o s p e r m s (Foster & Gifford, 1974).
As noted a b o v e , several taxa traditionally assigned to m a g n o l i i d s or h a m a m e l i d s are
essentially ranunculoid in their floral m o r p h o l o g i e s . T h u s , the Illiciales share with the
typical ranunculids the toroid flowers, apeltate s t a m e n s , tricolpate pollen, and the cycli­
cally disposed, laterally c o m p r e s s e d carpels. In Trochodendron, an archaic g e n u s vari­
ously placed in the M a g n o l i a l e s , H a m a m e l i d a l e s , Cercidiphyllales or T r o c h o d e n d r a l e s
(Takhtajan, 1966; Dahlgren, 1983; Endress, 1986a), the stamens are typically ranunculoid,
numerous, with distinct filaments and lacking p r o m i n e n t a p o p h y s e s . T h e floral axis is
slightly raised a b o v e the bases of the likewise r a n u n c u l o i d carpels that are cyclically
arranged, laterally c o n n a t e , With free reflexed styles, and are dorsiventricidal, as in
Glaucidium. T h e s e genera are here considered as representing a w o o d y branch of the
ranunculid cluster. Paeonia constitutes a presently m o n o t y p i c order that could be m o r e
diverse in the past, including such extinct forms as Hyrcantha that also r e s e m b l e d
Glaucidium and Thalictrum, thus strengthening the alliance of the P a e o n i a l e s with
ranunculids (Krassilov & al., 1983). On the other hand, the P a e o n i a l e s are related to
dilleniids, primarily on account of their androecial similarities. Ranunculids and dilleniids
also share phylogenetically significant palynological characters ( N o w i c k e & Skvarla,
1979). C a r y o p h y l l i d s may belong in the s a m e cluster but their fossil records are even
scantier than in the latter g r o u p .
With the exclusion of such artificial links as Trochodendron, Cercidiphyllum, etc.,
the magnoliids and h a m a m e l i d s are left well separated from each other, as well as from
the ranunculids. T h e i r shared features are primarily related to their pseudanthial flow­
ers with inner p h y l l o m e s . T h e i r g e o l o g i c a l histories are also totally different. T h e
hamamelids start in the C e n o m a n i a n {Lesqueria) and rise to d o m i n a n c e in the terminal
Cretaceous to early Paleocene, while m a g n o l i i d s , after exclusion of a m b i g u o u s pollen
records (see above), are scarcely known before the Turonian and are rare until the Eocene.
Their relationships are further discussed in the c o n c l u d i n g c h a p t e r s .
 Environments
T h e C r e t a c e o u s global e n v i r o n m e n t s were determined by extensive riffting and sub­
sequent s u b m e r g e n c e of cratonic areas. I n a s m u c h as land b i o m a s s is m a n y times larger
than marine b i o m a s s , any sea expansion causes a roughly proportionate loss not only of
the terrestrial, but also of the total b i o m a s s . In the Late C r e t a c e o u s , altogether about 60
million sq. km, or about 4 0 % of the present-day land area, were covered by epicontinental
seas. A m a r i n e transgression of such an extent would result in a drastic b i o m a s s reduc­
tion. Since b i o m a s s is a major sink for atmospheric C O 2 , the latter's concentration would
considerably increase, in turn causing a g r e e n h o u s e w a r m i n g . O n e v i d e n c e of isotopic
p a l a e o t e m p e r a t u r e s and biotic indicators, notably a m o r e than 1,500 k m shift of zonal
boundaries, the m i d - C r e t a c e o u s m e a n annual temperatures might h a v e been about 5-
6°C higher than at present.
A fairy distinct phytogeographical boundary dividing fossil plant localities with abun­
dant bennettites, b r a c h y p h y l l o u s conifers or gnetaleans ( " b r a c h y p h y l l s " ) to the south
and those d o m i n a t e d by g i n k g o a l e a n s and c z e k a n o w s k i a l e a n s to the north occurred at
about 50°N in Asia, ascending to about 60*'N in E u r o p e (Fig. 34). Key g e n e r a of these
provinces are the pachycaul Cycadeoidea and the heteroblastic ribbon-leaved Phoeni-
copsis respectively (Krassilov, 1972a). Studies in eastern and central Asia h a v e shown
that not only the d o m i n a n t s , but also the subordinate c o m p o n e n t s w e r e restricted to one

Fig. 34. Cretaceous climatic and vegetational zones (after Krassilov, 1965): vertical lines - temperate;
oblique lines - subtropical with xeromorphic vegetation; horizontal dashes - equatorial with impover­
ished xeromorphic vegetation.
or another province or they u n d e r w e n t certain c h a n g e s across their boundary. Inciden­
tally, the giant horsetails m o r e than 5 c m in d i a m e t e r occurred in the southern p r o v i n c e
alone. A m o n g ferns, the o s m u n d a c e a n Todites and the cyathacean Coniopteris with
aphlebial basal pinnules were replaced to the north by Osmunda and Dicksonia respec­
tively. Nilssonias with d e c i d u o u s dwarf shoots w e r e s u b d o m i n a n t in m a n y northern and
ecotonal fossil floras, b e c o m i n g rare to the south. A m o n g conifers, the araucarias and
podocarps w e r e mostly confined to the southern p r o v i n c e , w h e r e a s the dwarf-shoot
shedding p i n a c e a n s with Pityophyllum-iype n e e d l e leaves and Pityospermum-typQ
winged seeds were m o r e c o m m o n in the north. D e c i d u o u s n e s s of the northern province
dominants (and s u b d o m i n a n t s , including the pinaceans and nilssonias, the latter also
shedding dwarf-shoots) is amply d o c u m e n t e d by their forming leaf-mats, while their
association with coal beds suggests a seasonal s u m m e r - w e t c l i m a t e . In contrast, leaf-
mats are rather u n c o m m o n in the southern zone localities. If formed of iht Ptilophyllum-
Otozamites type bennettite leaves, they usually o c c u r in littoral facies. T h u s , the Cyca-
deoidealPhoenicopsis province boundary might have corresponded to the summer-green/
evergreen or winter-green zonal boundary. Its sharpness could hardly be explained by
the t e m p e r a t u r e differences alone, for the C r e t a c e o u s t e m p e r a t u r e gradient w a s m u c h
lower than at present, with equatorial m e a n s nearly the s a m e as at present against about
0°C at the poles (Sellwood & al., 1994). Scaly leaves, thick cuticles, sunken stomata,
pubescence, and other x e r o m o r p h i c features of the southern p r o v i n c e d o m i n a n t s may
suggest precipitations, rather than temperature, being a major differentiation factor across
the boundary.
A latitudinal differentiation within these t w o broad z o n e s w a s scarcely discernible,
with the equatorial zone apparently m u c h drier than at present. T h e present day low
latitude h u m i d z o n e is d u e to a rapid cooling of the a s c e n d i n g tropical air m a s s e s
shedding rain o v e r the equator. In the w a r m e r climate, these air m a s s e s might retain
water v a p o u r until they spread to a considerable distance from the e q u a t o r that, c o n s e ­
quently, received a lower precipitation than the extratropical z o n e s . In addition, the
low-latitude anticyclones spreading from over the latitudinal Tethys seas (like those,
but much m o r e extensive than, over the present-day M e d i t e r r a n e a n ) , m a d e tropical rain
forests unlikely. In effect, relatively m o r e b i o m a s s was c o n c e n t r a t e d in high latitudes
than in low latitudes, as e v i d e n c e d by the latitudinal distribution of the C r e t a c e o u s coal
localities (see Krassilov, 1985), with the largest reserves in northern Siberia and Alaska.
A m o n g the various feedbacks of the build-up of a t m o s p h e r i c C O 2 , the increased
productivity of terrestrial vegetation is of certain interest as a factor of early a n g i o s p e r m
evolution. Presently, a 10% higher crop yield is e x p e c t e d as a potential e c o n o m i c a l l y
positive effect of the doubling a t m o s p h e r i c CO^. T h e g r e e n h o u s e c l i m a t e w o u l d also
give a d v a n t a g e to high growth rates over drought resistance b e c a u s e of a considerably
decreasing w a t e r u p t a k e . H e n c e , a n g i o s p e r m s s e e m to h a v e a p p e a r e d in an environ­
mental situation that e n c o u r a g e d high growth rates that b e c a m e their major a d v a n t a g e
over the typically s l o w - g r o w i n g g y m n o s p e r m s ( F a v r e - D u c h r t r e , 1979).
 In the Cretaceous, large phytomass production is attested by an exceptionally high
phytophage diversity, including both invertebrate and vertebrate plant-eaters, with such
first appearing specialized forms as semenivores and frugivores. Giant herbivores, vani­
shing at the Cretaceos/Tertiary boundary and reappearing in the early E o c e n e , could be
indicative of high growth rates in the cropped plant communities. In its turn, the increased
phytophage pressure might have instigated a further developmental acceleration and pro­
lific reproduction in the cropped plant populations, as well as reciprocal interactions, in­
cluding, on the plant part, specialized entomophilous and zoochorous adaptations.

Angiosperm cradle plant communities

T h e r e are conflicting theories of tropical versus temperate, as well as upland versus
lowland, or even aquatic, a n g i o s p e r m origins that are based on palaeontological, as well
as on actualistic, evidence (see Krausel, 1956; Axelrod, 1959, 1872; Retallack & Dilcher,
1981 b; Tiffney, 1984, 1985). Tropical rain-forest is often given preference as a cradle of
a n g i o s p e r m s , although this plant formation is relatively recent, with n o C r e t a c e o u s ana­
logues yet recorded. Actually, the extant a n g i o s p e r m s that are considered primitive are
concentrated in the w a r m temperate zonal vegetation belts, such as the m o n t a n e forests
of Yunnan, China, comprising diverse magnoliacean genera, such SLsLiriodendron, Man-
glietia, Manglietiastrum, Paramichelia, Parakmeria, etc., a c c o m p a n i e d by Tetracentron,
Euptelea, and Cercidiphyllum, each representing a relict family, as well as by Semili-
quidambar and other lower h a m a m e l i d s , Davidia a m o n g the N y s s a c e a e , and the ar­
chaic open-veined ranunculid genera Circaeaster dixxd Kingdonia (Fu L i - k u o , 1992).
T h e s e forests are the remnants of a cosmopolitan M i o c e n e Arcto-Tertiary flora having
its mixed conifer - broad leaved precursors in the P a l e o g e n e , but having little in c o m ­
mon with any of the recorded Cretaceous a s s e m b l a g e s (except in the terminal Creta­
c e o u s : Krassilov, 1976, 1979; G o l o v n e v a , 1995). H e n c e , an actualistic a p p r o a c h fails in
giving any idea of the a n g i o s p e r m cradle plant c o m m u n i t y .
In p a l e o b o t a n y , plant community reconstructions are based primarily on the quantita­
tive relations of fossil plant species and the relevant taphonomic data (Krassilov, 1972c).
D o m i n a n c e is used as a major criterion, and the ensuing classification is, by necessity, far
less detailed than one achievable by the m o d e m phytosociological methods. T h e input of
plant material is controlled by sedimentation processes that, in turn, are governed by
tectonic, geomorphological and climatic conditions in each particular sedimentary basin.
Plant fossils from a single sedimentary basin constitute a local fossil flora (taphoflora)
reflecting vegetation of both the deposition site, or autochthonous, and the borders, or
allochthonous. For a plant palaeoecologist, it is essential not to lump all material as a
"fossil flora", but to deal discretely with each discernible fossil plant assemblage, the
latter conceived of as an episodic vegetation sample. Such samples usually contain both
autochthonous (hypoautochthonous) and allochthonous c o m p o n e n t s discernible by their
numerical representation and taphonomic evidence of in situ burial versus long-distance
transport, assisted by intrabasinal and interbasinal comparisons.
A b e d d i n g - p l a n e assemblage is usually d o m i n a t e d by a single or a few species that
might be d o m i n a n t in the source c o m m u n i t y or j u s t locally a b u n d a n t or depositionally
amassed. T h e i r actual status is further clarified by their frequencies in other roughly
c o n t e m p o r a n e o u s plant beds. Regional d o m i n a n t s may not be the most locally abun­
dant, but are distinguished by their high frequencies all over the basin. F o r instance,
Czekanowskia is locally abundant in a n u m b e r of Siberian M e s o z o i c basins represent­
ing a widespread but facially restricted plant c o m m u n i t y type, the C z e k a n o w s k i e t u m .
Also c o m m o n in the s a m e basins are fern-beds, Phoenicopsis-heds and Ginkgoites-
beds. Their lithological facies allow s o m e inferences of the vegetational structure. M o s t
commonly, the fern-beds occur in coal-bearing deposits w h e r e they intermingle with
the C z e k a n o w s k i e t u m in a pattern suggestive of a flood-plain or coastal s w a m p - m a r s h
mosaic, w h e r e a s the P h o e n i c o p s i e t u m is rare in the coastal facies, being most frequent
in all facies of the alluvial d o m a i n as a d o m i n a n t lowland plant c o m m u n i t y . Ginkgoites
is mostly confined to drifted plant beds, supposedly representing a slope or upland
community. Such orderly c h a n g i n g occurrences along a facies gradient are indicative of
a vegetational catena.
T h e catenic, as well as the frequently coincident serai successions can be inferred
from the vertical sequences of fossil plant a s s e m b l a g e s that replace each other follow­
ing a retreating coastal line. F o r e x a m p l e , in the U p p e r Jurassic of the B u r e y a Basin,
Russian Far East, the lower plant beds overlying the Callovian a m m o n i t e facies still
contain m a r i n e p h y t o p l a n k t o n and limulid r e m a i n s a l o n g s i d e with terrestrial plant
macrofossils, the latter representing fern marshes and the C z e k a n o w s k i e t u m as the pio­
neer plant c o m m u n i t i e s of a recently e m e r g i n g land (Krassilov, 1972b,c). Upsection,
the extinct ginkgoalean genus Pseudotorellia, the c y c a d o p h y t e Nilssonia, and the pi­
naceous Pityophyllum dwarf shoots appear in the C z e k a n o w s k i e t u m signalling an en­
richment of the serai c o m m u n i t i e s . Further on, they are replaced by a b r a c h y p h y l l o u s
conifer, Elatides, in association with Sphenobaiera, the bennettitalean leaf g e n u s Ptero-
phyllum, and occasional Phoenicopsis, the latter then raised to d o m i n a n c e in the overly­
ing alluvial facies.
Certainly, s o m e of such sequences reflect lateral, along-shore, facies migrations rather
than sere - a c o m m u n i t y build-up succession. Their interpretations h a v e to be c o r r o b o ­
rated by parallel s e q u e n c e s in the s a m e or other c o n t e m p o r a n e o u s basins. In the case of
the Jurassic successions, a remarkable parallelism has been observed between the a b o v e -
mentioned B u r e y a n sequence and those following the retreating seas in the adequately
studied sections of southern P r i m o r y e (= M a r i t i m e P r o v i n c e , Russian Far East), Ukraine
and other areas (Krassilov, 1972c). At the s a m e time, the brachyphyll and bennettite
stages w e r e m u c h m o r e diverse taxonomically in the E u r o p e a n s e q u e n c e s than in the
temperate Siberian province.
Fig. 35. Brachyphylls of xeromorphic Early Cretaceous plant assemblages of Mongolia (Krassilov, 1982):
(1-3) Brachyphyllum densiramosum Krassil., variable shoot morphologies and leaf cuticle with stomatal
pits marked by papillae, (1) x l , (2) x2, (3) SEM, x800; (4) Araucaria mongolica Krassil., spariously
branching shoot with thick scaly leaves, x2.
 M u c h the s a m e s e q u e n c e of the build-up stages have persisted in the Early Creta­
ceous. H o w e v e r , in the central Asiatic regions, including the vast sedimentary basins of
Mongolia and Transbaikalia, fern marshes were drastically reduced and s o m e t i m e s lost
as a distinct m e m b e r of the catenic and serai successions (Krassilov, 1982b). T h e y w e r e
replaced by the c o m m u n i t i e s of the later serai (and catenic) stages, m o s t l y by the
Classopollls-produc'xng hirmerellaceans and other b r a c h y p h y l l s (the term refers to the
not always distinguishable shoots of conifers and gnetaleans with helical or decussate,
adpressed, thickly cutinized, a m p h i s t o m a t i c or epistomatic scale-leaves, see Fig. 35), as
well as by bennettites of the Ptilophyllum-Otozamites g r o u p . Numerically, representa­
tion of the t w o latter groups in the coastal m u d flat - flood plain plant bed a s s e m b l a g e s
increased considerably, while their coriaceous, thickly cutinized leaves g a v e these as­
semblages a x e r o m o r p h i c aspect. In effect, their containing floras h a v e been conceived
of as xerophytic. Consequently, the floristic c h a n g e s at the J u r a s s i c / C r e t a c e o u s b o u n d ­
ary were ascribed to aridization, either regional or global (Vakhrameev, 1981).
Similar but less radical vegetational c h a n g e s , with the Weichselia-dommdXtd coastal
fern-marsh facies in the Wealden, took place simultaneously in the E u r o p e a n and M i d ­
dle Eastern - North African sectors of the southern zone, w h e r e a s in the t e m p e r a t e z o n e
the c o r r e s p o n d i n g events w e r e far less prominent, with d i c k s o n i a c e o u s fern m a r s h e s
surviving well in the Early C r e t a c e o u s . Although global climatic c h a n g e s might have
played certain roles in the a b o v e m e n t i o n e d vegetational c h a n g e s , the x e r o m o r p h i c
brachyphyll-bennettite a s s e m b l a g e s could scarcely b e taken as e v i d e n c e of that. T h e y
are constantly associated with littoral facies, including the coal-bearing paralic, as well
as the lacustrine-lagoonal calcareous deposits. Thick pinnules of Otozamites lacustris,
a characteristic species of the L o w e r C r e t a c e o u s lacustrine facies in M o n g o l i a and
Transbaikalia (Krassilov, 1982b), had x e r o m o r p h i c leaf p i n n a e with revolute margins
and with stomata sunken in thick a e r e n c h y m a t o u s spongy tissue suggesting a periodic
submergence (Fig. 36). In this case, x e r o m o r p h i s m pertains to a h e l o p h y t e , rather than
xerophyte, habitat. Upright stems of Classopollis-pvoducxng plants h a v e been found in
situ in littoral deposits (Francis, 1983), with their pollen grains c o m m o n both in the
marginal and m a r i n e facies. In addition, their pollen grains and scale leaf cuticles have
been found in the gut c o m p r e s s i o n s of the exceptionally well-preserved katydids and
other large insects from the c a l c a r e o u s deposits of the L a t e J u r a s s i c Karatau L a k e
(Krassilov & al., 1997a; Fig. 11). Being neither long-distance fliers, nor durable e n o u g h
for withstanding even a moderate-distance transportation intact, those insects conceiv­
ably inhabited the coastal vegetation d o m i n a t e d by x e r o m o r p h i c b r a c h y p h y l l s .
As is well k n o w n , x e r o m o r p h i s m is a characteristic feature of the present-day littoral
plant c o m m u n i t i e s , and perhaps it has r e m a i n e d such through the geological history. At
the same time, the climatic control of the brachyphyll-bennettite c o m m u n i t i e s is evi­
dent in their being far m o r e prominent, in terms of t a x o n o m i c diversity as well as fre­
quency of respective plant-bed a s s e m b l a g e s , in the southern (Cycadeoidea) phyto-
geographical p r o v i n c e extending into the ecotonal zone 50-60°N, but scarcely further in
Fig. 36. Otozamites lacustris Krassil. a bennettite from the Early Cretaceous of Mongolia (Krassilov,
1982b): (1) pinnule with ribbed veins, x7, (2-5) vessel-like tracheary elements of the midrib showing a
helical or reticulate radial wall thickenings and a porous perforation (2, 4) or an oblique perforation plate
at the lower end (3), x300 and (5), x600, (6) stoma sunken in aerenchymous mesophyll, SEM, xl 000.
the northern (Phoenicopsis) zone. T h e impression of a southern zone x e r o m o r p h i s m , as
well as the sharpness of the zonal boundary (see above), might be, to a considerable
extent, d u e to a climatic threshold controlling littoral vegetation that w a s poorly devel­
oped in the northern zone in c o m p a r i s o n with the southern z o n e .
For us the significance of the brachyphyll-bennettite a s s e m b l a g e s lies in their c o m ­
prising early a n g i o s p e r m pollen and macrofossil records that w e r e present neither in the
contemporaneous fern-marshes nor in the P h o e n i c o p s i e t u m and G i n k g o e t u m associa­
tions of the inland basins and slopes. This conclusion is based primarily on the study of
the L o w e r C r e t a c e o u s localities in Transbaikalia, M o n g o l i a and K a z a k h s t a n in central
Asia, as well as in the Primorye region, the Far East, where a search for early angiosperms
went alongside with palaeoecological analysis (Krassilov, 1967, 1973, 1982b, 1986;
Krassilov & B u g d a e v a , 1982, 1988; Krassilov & al., 1983, V a k h r a m e e v & Krassilov,
1979; Krassilov & M a k u l b e k o v , 1995). F o r other areas, palaeoecological data are m o r e
sketchy. H o w e v e r , even floristic descriptions reveal a leading role of b r a c h y p h y l l s and
bennettites in the W e a l d e n - t y p e floras of western E u r o p e and Atlantic coasts (Seward,
1904-1905; Berry, 191 l;Teixeira, 1948;Harris, 1953;Watson, 1982, Watson & Sincock,
1992). R e m a r k a b l y , the brachyphyll-bennettite plant beds h a v e yielded not only the
first a n g i o s p e r m s a n d a n g i o s p e r m - l i k e fossils, but a l s o the r e m a i n s of a d v a n c e d
proangiosperms of bennettitalean-gnetalean affinities. In the B a i s a locality on the Vitim
River, late N e o c o m i a n of Transbaikalia, the a n g i o s p e r m leaves ''Dicotylophyllum "
pusillum and the pollen grains Asteropollis and Tricolpites ( V a k h r a m e e v & Kotova,
1977) are a c c o m p a n i e d by the a b u n d a n t c o t t o n - g r a s s - l i k e d i s s e m i n u l e s of Baisia
(Krassilov & B u g d a e v a , 1982). Eoantha, a b r a c t e a t e p r e f l o w e r with a f o u r - l o b e d
gynoecium and with Ephedripites pollen grains in the pollen c h a m b e r s of orthotropous
ovules c a m e from the s a m e plant-bed (Krassilov, 1986). A recently found attached flower
has m a d e it possible to assign the associated graminoid leaves to the Eoantha plant.
Additional p r o a n g i o s p e r m species of gnetalean affinities are represented by the Bai­
sianthus a n d r o c l a d e s bearing bracteate nodal whorls of c u p u l a t e s p o r a n g i o p h o r e s , and
the Vitimantha preflowers with leafy reticulinerve perianth bracts (see u n d e r "Selected
fossil links" a b o v e ) .
N o less diverse are the a n g i o s p e r m o i d and p r o a n g i o s p e r m fossils in the roughly
contemporaneous localities Manlay, G u r v a n - E r e n and B o n - T s a g a n in the western Gobi,
Mongolia (Krassilov, 1982b). T h e a n g i o s p e r m o i d fruits Gurvanella and Erenia are ac­
companied there by the monocot-like Cyperacites, Graminophyllum and u n a s s i g n e d
Sparganium-Ukt and Potomageton-Wkc fruiting axes (Figs 3 7 , 38), as well as by the
pappose r e e d - m a c e - l i k e Typhaera. Their preservation is u n f a v o u r a b l e for detailed mor­
phological studies, thereby m a k i n g their a n g i o s p e r m o u s or p r o a n g i o s p e r m o u s status
uncertain. H o w e v e r , c o m p a r i s o n s with the better preserved Baisa material are rather in
favour of the latter, at least in the case of Cyperacites and Graminophyllum that resem­
ble the inflorescences of Baisianthus and the leaves of Eoantha plant respectively.
Fig. 37. Angiosperm and angiosperm-like reproductive structures from the Early Cretaceous of Mongolia
(Krassilov, 1982): ( 1 , 2 ) Gurvanella dictyoptera Krassil., winged fruit with reticulate wings, Gurvan-
Eren Assemblage, x3 and 10, (3) Erenia stenoptera Krassil., another type of winged fruit with smooth
wings and a bifid persistent stigma, same asemblage, xlO, (4) Sparganium-Wke axis bearing bracteate
heads of longly awned fruits or cupules, Manlaj Assemblage, x3, (5) "Cyperacites'\ a fasciculate inflo­
rescence with filiform bracts, same assemblage, x2.
Fig. 38. Angiosperm-like reproductive structures from the Early Cretaceous of Mongolia (Krassilov,
1982): ( 1 , 2 ) Potamogeton-Wke spike of awned fruits or cupules from the Manlaj Assemblage, x9, (3)
'"Cyperacites'\ details of spikes and filiform bracts, same locality, xlO, (4) Sparganium-Wke bracteate
head of longly awned fruits or cupules, Manlaj Assemblage, x7.
 Similar situations are described in other parts of the world. In the P o t o m a c flora of
the Atlantic coast, early a n g i o s p e r m s appeared in association with Drewria, a herba­
ceous gnetalean plant (Crane & U p c h u r c h , 1987). In the English W e a l d e n , the entry of
a n g i o s p e r m s in the pollen record w a s paralleled by the rise in the gnetoid pollen fre­
quencies ( H u g h e s & M c D o u g a l , 1987). In the K o o n w a r r a Fossil Bed, southern Aus­
tralia, a n g i o s p e r m o u s ceratophyllacean fruits (Dilcher & al., 1996) and r a c e m o s e inflo­
rescences of follicular carpels (Taylor & Hickey, 1990) are j o i n e d by the ephedroid
Leongathia ( D o u g l a s , 1969; Krassilov & al., 1996) and p e r h a p s s o m e other gnetaleans
represented by ovulate bracts and bracteate pollen c o n e s (Drinnan & C h a m b e r s , 1986).
In a small collection from the Aptian " a m p h i b i a n b e d " of M a k h t e s h - R a m o n , Israel,
a n g i o s p e r m o i d fruits are found together with Sagenopteris-iype leaves of caytonialean
p r o a n g i o s p e r m s (Krassilov & Dobruskina, 1995).
Palynological studies in northern Africa have revealed the intercorrelated frequen­
cies of the earliest angiospermoid pollen of Afropollis, Schrankipollis, and Brenneripollis
types and the p r o a n g i o s p e r m o u s Classopollis, Ephedripites, etc., through the Barremian
to C e n o m a n i a n s e q u e n c e s ( P e n n y , 1 9 9 1 ; S c h r a n k , 1 9 9 2 ; I b r a h i m , 1 9 9 6 ) , with
Classopollis c o m i n g to d o m i n a t e over the Afropollis g r o u p after the initial increase of
the latter in the late Barremian to early Aptian. T h e diverse gnetoid pollen and the
recently found macrofossils appear with the early a n g i o s p e r m pollen records in the
Aripe Basin, Brasil (Pons & al., 1992).
To s u m m a r i z e , the appearance of a n g i o s p e r m s w a s not an occasional breakthrough
against a static background, pushing other plants aside, but w a s rather a collective break­
through involving a n u m b e r of species growing side by side in the e x p a n d i n g brachyphyll
c o m m u n i t y , their parallel, although not uniformly successful, a d v a n c e c h a n n e l e d by
this c o m m u n i t y evolution.

Centres of origin
As argued in the preceding chapter, the present-day supposedly primitive angiosperms
have survived as relicts of the Arcto-Tertiary flora having little in c o m m o n with the
C r e t a c e o u s floras that c o m p r i s e d early a n g i o s p e r m s . Therefore, c o n c e n t r a t i o n s of such
relicts are scarcely indicative of a centre of a n g i o s p e r m origin. T h e latter cannot be
deduced from a phytogeographical analysis of extant vegetation. O n e can hardly rely
also on the currently earliest records of solitary a n g i o s p e r m - l i k e fossils for, in the first
place, most of such records are not entirely c o n v i n c i n g and, secondly, they d e p e n d on
c h a n c e s of preservation and collecting.
A n g i o s p e r m origins are here conceived of as a process with neither strictly definable
starting points nor deadlines. A semitectate pollen grain signals a step in the morpho­
logical evolution related to the sporophytic self-incompatibility s y n d r o m e rather than
the appearance of a n g i o s p e r m s . Such records go back in time at least to the Late Triassic
(Cornet, 1979). M o r e c o n v i n c i n g are the c o m b i n e d records of several a n g i o s p e r m traits.
Such records first appeared, after a period of single-trait signals, in the late N e o c o m i a n
to Aptian, about 120-115 M a . Close to these dates are the associate records of authentical
angiosperms and diverse p r o a n g i o s p e r m s , notably the a d v a n c e d gnetaleans, in central
Asia, M i d d l e East, Atlantic coasts, Australia, etc. (see a b o v e ) . T h e y not only testify to
the early a n g i o s p e r m appearances in these areas, but are also e v i d e n c e of the ongoing
process of angiospermization in the cradle plant c o m m u n i t i e s that are crucial for recog­
nition of geographical centres of a n g i o s p e r m origin.
In both Northern and Southern hemispheres, the critical localities occur in the ecotonal
zones of t e m p e r a t e s u m m e r - g r e e n and subtropical evergreen to winter-green vegeta­
tion, about 50°N and 40°S respectively. A n g i o s p e r m i z a t i o n might h a v e been conceiv­
ably going on all along the extent of these zones. However, most of the actual records
are confined to downfaulted grabens and s e m i g r a b e n s of the Early C r e t a c e o u s rift sys­
tems c o n c e i v a b l y c o n s t i t u t i n g m u l t i p l e a n g i o s p e r m i z a t i o n c e n t r e s . T h u s , t h e
Transbaikalian and M o n g o l i a n basins are linear depressions of the extensive rift system
striking northeast from M o n g o l i a to the Sea of O k h o t s k . This system has been initiated
by a mid-Paleozoic rifting, while s o m e of its s e g m e n t s , notably the L a k e Baikal Rift,
are still active. A m a s s i v e volcanic event close to the Jurassic/Cretaceous boundary
produced o n e of the w o r l d ' s largest basaltic provinces extending from western Gobi to
the A m u r River Basin. In the C r e t a c e o u s , the rift zones were m a r k e d by tectonic insta­
bility reflected in their sedimentary facies (Krassilov, 1985).
The lower C r e t a c e o u s lacustrine facies of the M o n g o l o - O k h o t s k i a n rift zone are
typical of stratified lakes, with the thick, finely laminated black shale s e q u e n c e s inter­
vened by p s a m m i t i c and carbonate interbeds, the latter a b o u n d i n g in fossil fish and
aquatic insect larvae remains. Large dragon-fly, may-fly and beetle (coptoclavid) lar­
vae pile u p on the bedding planes suggesting m a s s mortalities. Since all these organisms
are sensitive to p H , their m a s s deaths might have been caused by abrupt p H fluctuations
inflicted by volcanogenic acid rains affecting also the s u r r o u n d i n g wetlands and terres­
trial biota. T h e s e t a p h o n o m i c data suggest heavily impacted e n v i r o n m e n t s of the rift-
bound angiospermization centres.

Rise of angiosperms as a biocoenotic process

The clandestine period of a n g i o s p e r m evolution c a m e to an end at about the Albian/
Cenomanian boundary a b o v e which, or locally slightly b e l o w which, they b e c a m e fairly
common. Before the relatively recent discoveries of the earlier a n g i o s p e r m s , their Late
Cretaceous m a s s appearances, as if from n o w h e r e , s e e m e d mysterious. O n e of the for
the time being popular hypotheses holds that a n g i o s p e r m s could have d e s c e n d e d from
highlands w h e r e they grew for million years, p e r h a p s even since the P a l e o z o i c , without
leaving fossil traces. T h e highland hypothesis accords with the Lyellian-Darwinian doc­
trine of imperfection of the fossil record. It is true that plant c o m m u n i t i e s p r o x i m a l to
the site of deposition tend to be overrepresented in the fossil record at the e x p e n s e of the
more distant plants. However, while speaking of million-year t i m e spans, w e have to
consider the contribution of long-distance transport in the form of w i n d - b o r n e pollen
and seed rain constituting a minor, but still recognizable, c o m p o n e n t of fossil plant
a s s e m b l a g e s ( H u g h e s , 1961, 1976). Moreover, climatic fluctuations w o u l d repeatedly
inflict upland-lowland shifts of altitudinal belts bringing distant vegetation closer to the
deposition sites. Yet a n g i o s p e r m s did not appear during the major m i d - J u r a s s i c cooling
event (Krassilov, 1973a and e l s e w h e r e ) . T h e s e general c o n s i d e r a t i o n s are supported by
the late Albian and C e n o m a n i a n a n g i o s p e r m s being far m o r e n u m e r o u s and diverse in
the marginal epicontinental sea deposits (Berry, 1916; Vakhrameev, 1952) than in the
c o n t e m p o r a n e o u s deposits of intermontane depressions (Krassilov, 1989; Krassilov &
S h o r o k h o v a , 1989). Such e v i d e n c e is in favour of the coastal h y p o t h e s i s of angiosperm
dispersal (Retallack & Dilcher, 1981b) that implies s o m e trust in the fossil record. The
m i d - C r e t a c e o u s t i m e is m a r k e d by w i d e s p r e a d tectonic and v o l c a n i c e v e n t s of the
Austroalpine phase that brought about frequent sea-level fluctuations and concomitant
climatic c h a n g e s . Of the latter, the late Albian to early C e n o m a n i a n c o o l i n g (Krassilov,
1973d, 1975b) could affect, in the first place, the t h e r m o p h i l o u s b r a c h y p h y l l s and
bennettites of a n g i o s p e r m cradle c o m m u n i t i e s .
T h e m i d - C r e t a c e o u s rise of a n g i o s p e r m s w a s paralleled by a c o n c u r r e n t rise to domi­
nance of such morphologically m o d e r n taxodiacean and pinacean conifers as Sequoia,
Parataxodium (Arnold & Lowther, 1955), Pseudolarix (Krassilov, 1982b), etc. These
e v e n t s w e r e a c c o m p a n i e d by the a p p e a r a n c e of Cycas a n d o t h e r m o d e r n c y c a d s
(Krassilov, 1979; Krassilov & al., 1990). T h e situation w a s thus m o r e c o m p l i c a t e d than
a mere competition-induced replacement of g y m n o s p e r m s by angiosperms. We can rather
speak of a w h o l e s a l e floristic innovation and the a p p e a r a n c e of n e w plant c o m m u n i t i e s
c o m p r i s i n g the a n g i o s p e r m s as their d o m i n a n t or subordinate c o m p o n e n t s alongside
with the m o d e r n i z e d g y m n o s p e r m s .
Vegetation types can persist for a long time as stable systems of climax and succes-
sional c o m m u n i t i e s . Their spatial ranges depend on the tolerance of their climax species,
the successional species often having broader tolerances and potentially wider ranges.
Insofar as successional species are poor competitors, the respective serai phases are open
to new species entries by immigration or innovation. T h e early a n g i o s p e r m appearances
in the brachyphyll-bennettite communities might have been such entries. However, serai
developments are equifinal in the sense that their end-products but little depend on the
starting point. A climax phase can persist irrespective of the changes occurring in the
pioneer and successional phases, while the radical restructurings are inflicted by environ­
mental impacts that truncate the serai sequences halting their d e v e l o p m e n t at a succes­
sional stage. If such impacts are of a vegetation zone or global extent, they drive the
climax species to extinction, with the successional species then exploring their vacant
niches. Their rise to d o m i n a n c e is, thus, conjoined with functional and morphological
innovations. T h e s e are postulates of the climax cut-off model (Krassilov, 1992, 1995) as
opposed to the widely held species competitive replacement model.
The M e s o z o i c plant a s s e m b l a g e records suggest a leading role of the c o m m u n i t y
replacement and c o m m u n i t y restructuring, rather than i n t r a c o m m u n i t y competition, in
the extinctions and origins of plant higher taxa as distinct adaptive types. All plant
communities w e r e simultaneously, although differently, affected by the Jurassic/Creta­
ceous boundary and, later, the mid-Cretaceous events. T h e respective vegetational events
can be s u m m a r i z e d as follows:
(1) T h e coastal fern m a r s h e s , w o r l d - w i d e in extent in the Jurassic, b e c a m e spatially
restricted at about the Jurassic/Cretaceous b o u n d a r y but persisted to the end of Creta­
ceous, although their d o m i n a n t tree fern c o m p o n e n t w a s reduced and partly replaced by
the o s m u n d a c e a n and gleicheniacean genera. J u d g i n g by their numerical representation
in the respective plant-beds (Krassilov, 1979), a n g i o s p e r m s scarcely penetrated the fern
marshes even in the Late C r e t a c e o u s .
(2) T h e m e s o p h y t i c forests with g i n k g o a l e a n s , c z e k a n o w s k i a l e a n s and archaic coni­
fers remained the least affected at the Jurassic/Cretaceous b o u n d a r y but declined in the
mid-Cretaceous surviving as relict c o m m u n i t i e s with no or an insignificant a n g i o s p e r m
representation. Until the end of the Late C r e t a c e o u s e p o c h , their replacing c o m m u n i t i e s
of the m o d e r n i z e d taxodiacean and pinacean conifers contained a n g i o s p e r m s as a sub­
ordinate c o m p o n e n t represented by laurophylls as well as by the small-leaved Trocho-
dendroides, Araliaephyllum, Menispermites, Liriophyllum and a few other m o r p h o t y p e s
perhaps representing a successional vegetation, s o m e of t h e m c l i m b e r s .
In the flood-plain facies domain, the taxodiacean assemblages comprised the platanoid
Credneria-Protophyllum-Aspidiophyllum m o r p h o t y p e s as their c o d o m i n a n t c o m p o n e n t .
The platanoid leaves, a c c o m p a n i e d by the staminate and carpellate heads as well as by
the tricolpate-tricolporate dispersed pollen grain a s s e m b l a g e s , often formed leaf-mats
in the coarse-grained laminated p s a m m i t i c levee deposits as well as in the clayey ox­
bow lenses. T h e s e riparian mixed broadleaved platanophyll-conifer c o m m u n i t i e s were
widespread in both intracontinental and marginal basins of western E u r o p e , Kazakhstan,
West Siberia, Sakhalin, mid-western North A m e r i c a , etc.
(3) T h e x e r o m o r p h i c brachyphyll-bennettite c o m m u n i t i e s e m e r g e d as a d o m i n a n t
coastal to littoral vegetation type at about the J u r a s s i c / C r e t a c e o u s boundary, replacing
fern marshes in central Asia and, locally, in other sectors of the w a r m s u m m e r - d r y zone
and its e c o t o n e with the temperate zone. T h e widespread x e r o m o r p h i s m might have
been partly related to g r e e n h o u s e climate. Notably, in both g y m n o s p e r m s and the later
appearing a n g i o s p e r m s , xeromorphic epidermal features correlate with a low stomatal
index. Periodic penetration of xeromorphic vegetation into the t e m p e r a t e z o n e coin­
cided with transgressions of the Tethys seas over the northern land m a s s e s .
 A n g i o s p e r m s appeared as m i n o r c o m p o n e n t s of these c o m m u n i t i e s concomitantly
with their expansion and proceeded in this role for about 15 million years. T h e long
period of rarity suggests that early a n g i o s p e r m s had been competitively inferior to the
c o m m u n i t y d o m i n a n t s until the climax phase was u n d e r m i n e d by c o o l i n g and other
environmental impacts related to the global tectono-volcanic event at about the Albian/
C e n o m a n i a n b o u n d a r y . Both h i r m e r e l l a c e a n b r a c h y p h y l l s and b e n n e t t i t e s survived
through the Cretaceous and the latter even increased their frequencies in the Maastrichtian
(Krassilov, 1979). However, since the late C e n o m a n i a n , they lost their d o m i n a n t status,
thereby leaving the respective c o m m u n i t i e s undersaturated and o p e n i n g n e w perspec­
tives for the early a n g i o s p e r m s . T h e brachyphyll-bennettitalean c l i m a x cut-off w a s not
an isolated vegetational event, but co-occurred with m a s s extinctions and replacements
of the then d o m i n a n t groups in other plant c o m m u n i t i e s as well as in m a r i n e biotas.
X e r o m o r p h i c a n g i o s p e r m a s s e m b l a g e s c o m p r i s e the small to m e d i u m - s i z e d com­
pound to lobed leaf m o r p h o t y p e s of the Deheya-Dryophyllwn g r o u p as well as the nar­
row-leaved laurophylls of the Proteoides m o r p h o t y p e , and occasional p a l m leaves. The
dicotylid leaf types are x e r o m o r p h i c , often coriaceous, with thick cuticles, sunken sto­
mata and a low stomatal index (Figs. 39, 4 0 ) . T h e y associate with fossil w o o d of shrubs
or small trees (Stopes & Fujii, 1911, S h i m a k u r a , 1037; Sitholey, 1954; Prakash, 1965;
Page, 1981) and with dispersed pollen of the Normapolles group. The non-angiosperm
c o m p o n e n t s are represented by brachyphylls and cycadophytes, infrequent on the basinal
scale, but c o m m o n locally. Taxodiacean conifer r e m a i n s are less c o m m o n . A typical
e x a m p l e studied by the present a u t h o r (Krassilov, 1984b) c o m e s from the m i d d l e
C e n o m a n i a n shallow m a r i n e carbonate deposits of S e l - B u c h r a M o u n t a i n , C r i m e a , with
similar occurrences in western E u r o p e , Israel, North A m e r i c a and e l s e w h e r e within the
subtropical and w a r m - t e m p e r a t e ecotonal zones.
On e v i d e n c e of the leaf m o r p h o l o g i e s and associated fossil w o o d , the x e r o m o r p h i c
angiosperm c o m m u n i t i e s of the D e b e y o - D r y o p h y l l e t u m type are c o n c e i v e d of as coastal
shrublands or low w o o d l a n d s , their brachyphyll c o m p o n e n t suggesting a u t o c h t h o n o u s
origins from the Early C r e t a c e o u s brachyphyll-bennettite c o m m u n i t y . In the alluvial
facies d o m a i n , the x e r o m o r p h i c plant a s s e m b l a g e s have m e r g e d with the platanophyll-
d o m i n a t e d riparian ones, the latter most prominent in the northern t e m p e r a t e realm. At
about the Cretaceous/Tertiary boundary, the Debeyo-Dryophylletum, although still promi­
nent in the P a l e o g e n e , have receded before the a d v a n c i n g conifer-broadleaved forests.

Reproductive biology
It has long been suspected that plant-animal interactions might h a v e played certain,
perhaps even leading, roles in the origin of angiospermy. M a n y extant a n g i o s p e r m s
considered primitive are insect-pollinated. A m o n g them the beetle-pollination, or can-
Fig. 39. Typical angiosperms of the Late Cretaceous plant assemblages of Sakhalin (Krassilov, 1979): (1)
Debeya tikhonovichii (Kryshl.) Krassil., compound leaf of juglandacean-myricacean affinities, x l , (2)
AraliaephyUum polevoi (Krysht.), Irilobed lauraccan leaf, x l .
Fig. 40. Cuticle of a xeromorphic angiosperm Debeya from the terminal Cretaceous - lowermost Paleocene
turbiditic island arc foredeep deposits of the Lesser Kuril Islands, Far East (Krassilov & al., 1987): (1)
stoma with sunken guard cells and a prominent peristomatal ridge, SEM, xl 000, (2) well-spaced stomata
of the lower epidermis with a low stomatal index, x250.
tharophily, is characteristic of the M a g n o l i a c e a e and is often considered as primary for
all a n g i o s p e r m s . H o w e v e r , m o s t early C r e t a c e o u s a n g i o s p e r m s had inflorescences of
many m i n u t e inconspicuous flowers, as in Caspiocarpus ( V a k h r a m e e v & Krassilov,
1979), that looked anything but cantharophilous.
Extant c y c a d s and gnetaleans are k n o w n to have insect pollinators, s o m e of them
perhaps inherited from their M e s o z o i c precursors. T h e m o r p h o l o g y of bennettite pre­
flowers, b o w l - s h a p e d with glands and perhaps filled with nectar (Harris, 1969), sug­
gests insect visits for which there is s o m e direct e v i d e n c e ( D e l e v o r y a s , 1965; Crepet,
1974). T h e caytonialean p r o a n g i o s p e r m s had fleshy cupules, w h i l e their pollen w a s
found in coprolites (Harris, 1964).
A direct e v i d e n c e of pollinivory has been obtained by maceration of the c o m p r e s s e d
gut contents preserved in fossil insects of different geological a g e s . Material c a m e from
the extensive entomological collections deposited in the Paleontological Institute, M o s ­
cow and w a s studied by the present author jointly with Prof. A . P Rasnitsyn and his
collaborators in the Laboratory of A r t h r o p o d s of the s a m e institute. T h e coaly gut c o m ­
pressions of the Early C r e t a c e o u s xyelid insects from the Baisa locality, Transbaikalia,
remarkable for the diversity of proangiospermous plants (see above), have hitherto yielded
only b i s a c c a t e pollen with laterally c o n n e c t e d sacci, as in c a y t o n i a l e a n s and s o m e
peltasperms (Krassilov & Rasnitsyn, 1982). W h i l e these studies are in progress, s o m e
interesting data have c o m e from other localities.
Striate pollen grains w e r e found in the gut c o m p r e s s i o n s of extinct h y p o p e r l i d and
grylloblattid insects (Fig. 4 1 ) from the P e r m i a n of the U r a l s ( R a s n i t s y n & Krassilov,
1996). T h e s e d a t a h a v e s o m e b e a r i n g on p a r a l l e l i s m s of p o l l e n e v o l u t i o n in g y m n o ­
sperms and, by i m p l i c a t i o n , in a n g i o s p e r m s . Striate p o l l e n g r a i n s c o n s t i t u t e a d o m i ­
nant e l e m e n t of the P e r m i a n s p o r e - p o l l e n a s s e m b l a g e s and are c h a r a c t e r i z e d by the
external layer of their e x i n e being d i v i d e d , or only p r o x i m a l l y so, into parallel or
subparallel b a n d s , or t a e n i a e . T h e i r n u m b e r and pattern are d i a g n o s t i c of the n u m e r ­
ous f o r m - g e n e r a differing from each other also in the p r e s e n c e of sacci and their
configuration, with the a s a c c a t e m o r p h o t y p e s o c c a s i o n a l l y also s h o w i n g r u d i m e n t a r y
sacci. Typically, they differ from the m o d e r n saccate p o l l e n of p i n a c e o u s conifers in a
protosaccate, rather than e u s a c c a t e , infrastructure, with the i n n e r p a r t i t i o n s filling the
sacci and a t t a c h e d to the c o r p u s .
Striate pollen types w e r e produced by at least three different groups of g y m n o s p e r m s :
glossopterids, p e l t a s p e r m s , and conifers (Clement-Westerhof, 1984; M e y e n , 1978;
Zavada, 1991), the former d o m i n a n t in G o n d w a n a l a n d and the t w o latter c o m m o n to
dominant in the E u r a m e r i c a n , S u b a n a g a r i d a n and C a t h a y s i a n p r o v i n c e s of Laurasia.
Thus, the global spread of striate pollen was due to parallel d e v e l o p m e n t s . M e y e n (1978)
has suggested parallel successions of the non-striate to striate p r o t o s a c c a t e to striate
asaccate m o r p h o t y p e s in both the northern and southern p r o v i n c e s .
Aridity is generally considered as a leading factor of floristic evolution in the Permian.
In particular, the scale-leaved conifers as well as p e l t a s p e r m s with thick leaf cuticles are
Fig 41. Taeniate pollen in the gut contents of pollinivorous insects from the Early Permian of the Urals
(Rasnitsyn & Krassilov, 1996): (1) Idelopsocus diradiatiis Rasn., a hypoperlid insect with pollen clump
preserved at the end of the hindgut, x6.5, (2) pollen clump with some partly digested grains striped of the
tectum and showing an infratectal reticulum, SEM, x8()(). (3) well-preserved Luncitisporites-iype pollen
grain from the clump shown in (2), SEM, xl 500, (4) Vittcninci-iypc pollen grain from the gut of another
pollinivore, SojanideliaJloralis Rasn., from the same locality, xl 200.
conceived of as representing a xeric type of Permian vegetation distinguishable p a l y n o -
logically by the prevalence of striate pollen grains. H o w e v e r , their frequent association
with coal-bearing facies m a k e s this interpretation improbable. Glossoprterids, or at least
some of t h e m , p r o d u c e d the Vertebraria-type a e r e n c h y m a t o u s u n d e r g r o u n d organs evi­
dencing a p e a t b o g or m a n g r o v e s w a m p habitat. S i n c e t h e h e l o p h y t e , r a t h e r than
xerophyte, adaptations prevailed a m o n g the producers of striate pollen grains, aridity
seems to have n e v e r been a leading factor in their m o r p h o l o g i c a l evolution.
P o l l i n i v o r y p r o v i d e s an alterative e x p l a n a t i o n of the r e m a r k a b l e p a r a l l e l i s m of
palynological characters in the P e r m i a n g y m n o s p e r m s . Both the protosacci (as func­
tionally different from the bladders of a n e m o p h i l o u s conifers) and taeniae might h a v e
had a v o l u m e - r e g u l a t i n g ( h a r m o m e g a t h y ) f u n c t i o n , at t h e s a m e t i m e p r o v o k i n g
pollinivory by increasing the pollen m a s s . H a r m o m e g a t h y functions could h a v e been
even m o r e important for pollen transferred by large slow-flying insects than for any
wind-borne grains.
In a d d i t i o n , insects are k n o w n to transfer v a r i o u s g e n e - t r a n s d u c i n g m i c r o o r g a n ­
isms that can c o n f e r parallel g e n e t i c c h a n g e s in the target p l a n t s . T h i s m e c h a n i s m is
particularly i m p o r t a n t for p o l l e n grains that are haploid o r g a n s with the m a i n l y single
gene-controlled exinal characters.
T h e effects of the plant-insect interactions could scarcely have been restricted to pol­
len organs alone. T h e larger the pollen grains, the fewer of them could c o m p e t e for room
in the target pollen chamber, thus creating a selection pressure in favour of the larger or,
alternatively, the densely aggregated ovules, as well as stimulating adaptive modifica­
tions of their receptive structures. T h e s e processes might h a v e brought about the burst of
seed plant diversity in the Permian, as well as the subsequent burst in the M e s o z o i c .
T h i s P a l e o z o i c story is here narrated in s o m e detail b e c a u s e of its r e l e v a n c e to the
Mesozoic story. T h e M e s o z o i c x e r o m o r p h i c brachyphyll-bennettite c o m m u n i t i e s w e r e ,
in a s y n g e n e t i c s e n s e , direct d e s c e n d a n t s of the P e r m i a n c o n i f e r - p e l t a s p e r m c o m m u ­
nities, even their respective coniferid elements could have been related phylogenetically
( C l e m e n t - W e s t e r h o f & van K o n i j n e n b u r g - v a n Cittert, 1991). T h e r e c e n t d i s c o v e r y of
angiospermoid ClassopoUis pollen grains in the intestines of J u r a s s i c g r a s s h o p p e r ­
like k a t y d i d insects ( K r a s s i l o v & al., 1997a: F i g . 4 2 ) a d d s to t h e e c o l o g i c a l similarity
of the P e r m i a n and M e s o z o i c x e r o m o r p h i c c o m m u n i t i e s in several w a y s . In the first
place, they p r o v i d e e v i d e n c e of pollinivory as a m o d e of p l a n t - a n i m a l interactions
involving the d o m i n a n t plants of both c o m m u n i t i e s . F u r t h e r m o r e , b o t h the P e r m i a n
and M e s o z o i c p o l l i n i v o r e s w e r e large slow-flying insects that c o u l d h a r d l y h a v e been
transported intact o v e r a c o n s i d e r a b l e d i s t a n c e , t h u s s u g g e s t i n g p r o x i m i t y of their
habitats to the d e p o s i t i o n sites. T h e y also testify to the m o s t a d v a n c e d pollen t y p e s of
each of the e p o c h e s b e i n g related to the plant-insect i n t e r a c t i o n s , with p o s s i b l e impli­
cations for plant r e p r o d u c t i v e biology.
T h e first a p p e a r i n g a n g i o s p e r m pollen g r a i n s are r e c o g n i z e d as such by the fea­
tures of the self-incompatibility s y n d r o m e ( Z a v a d a , 1984; Z a v a d a & Taylor, 1986a)
Fig. 42. Proangiosperm pollen in the guts of pollinivorous insects (Krassilov & al., 1997): Ahoilus amplus
Gorochov, a large katydid from the Late Jurassic of Karatau, Kazakhstan, x l , with a dark intestine com­
pression filled with clumped ClassopoUis pollen grains, SEM, x83() and 1 700.
 that is c o n c e i v e d of as a form of sexual selection d i s t i n g u i s h i n g a n g i o s p e r m s from
g y m n o s p e r m s . T h e s p o r o p h y t i c self-incompatibility m e d i a t e d , by the p o l l e n - s t i g m a
interaction, m i g h t h a v e a p p e a r e d before the g a m e t o p h y t i c self-incompatibility result­
ing from the p o l l e n - s t y l a r c o n d u c t i n g tissue interactions, the latter m o d e c o r r e l a t i n g
with solid, rather than o p e n , styles recorded not earlier than the S e n o n i a n ( K r a s s i l o v
& al., 1983). H o w e v e r , a less a d v a n c e d form of pollen selection by m i c r o p y l a r struc­
tures, such as the stigmatic ( N a r b e n ) m i c r o p y l e s (Kugler, 1970), o c c u r s also in g y m ­
nosperms: p o l l e n g r a i n s are c a p t u r e d in the slime e x c r e t e d by t h e m i c r o p y l e and are
then s u c k e d in by a b i o c h e m i c a l l y different pollination d r o p e x c r e t e d by the n u c e l l u s
(Owens & B l a k e , 1984). T h i s m e c h a n i s m anticipates the t w o s u c c e s s i v e secretory
stages of a n g i o s p e r m p o l l i n a t i o n . M o r e o v e r , in the e x t a n t Pseudolarix, pollen s o m e ­
times g e r m i n a t e s in the m i c r o p y l a r canal (Allen, 1943), w h i l e in Caytonia, I have
observed pollen in the t u b e s c o n n e c t i n g the e x o s t o m e with the o v u l e s (Krassilov,
1977b). T h e infratectal structures serving for storage of r e c o g n i t i o n s u b s t a n c e s m i g h t
evolve on the basis of the h o n e y c o m b infrastructures of p r o t o s a c c a t e g y m n o s p e r m
pollen grains in relation to the m i c r o p y l a r pollen selection function. In a n g i o s p e r m s ,
this function m i g h t h a v e d e v e l o p e d primarily for p r e v e n t i n g the a b o r t i o n of g y n o e c i a
inflicted by m i s p o U i n a t i o n . J u d g i n g by the frequent o c c u r r e n c e s of a b o r t e d o v u l e s
and c u p u l e s , notably those of Caytonia, this p h e n o m e n o n actually took p l a c e in M e s o ­
zoic seed p l a n t s . In extant a n g i o s p e r m s , mispoUination s o m e t i m e s c a u s e s a seed yield
reduction, as in Ipomea aggregata ( K o h n & Waser, 1985). A s s u c c e s s i o n a l p l a n t s , the
early a n g i o s p e r m s tended to increase their seed yield. T h u s , g y n o e c i a l a b o r t i o n s had
to be p r e v e n t e d .
A d v a n c e s in the pollen selection function m a d e a n g i o s p e r m s p r e a d a p t e d to the
d e v e l o p m e n t of m o n o c l i n o u s ( " b i s e x u a l " ) r e p r o d u c t i v e o r g a n s that o c c a s i o n a l l y a p ­
pear also in conifers (Tosh & P o w e l l , 1986) but are u n s t a b l e in t h e m b e c a u s e of their
imperfect self-incompatibility s y s t e m s . E v e n in bennettites a n d g n e t o p h y t e s , the mor­
phological m o n o c l i n y usually d i g r e s s e s to the functional dicliny, with t h e a n d r o e c i a l
(in Williamsonia) or g y n o e c i a l (in the pollen p r e f l o w e r s of Welwitschia) o r g a n s trans­
formed into secretory structures. In the insect-pollinated d i c l i n o u s g y m n o s p e r m s , the
insects are often attracted to t h e o v u l e s by s w e e t p o l l i n a t i o n d r o p s . F o r a n g i o s p e r m s
lacking the pollination d r o p m e c h a n i s m , s t a m e n s are the m a j o r attraction. M o n o c l i n y
is therefore essential as a c o m p o n e n t of the a c c o m p l i s h e d a n g i o s p e r m o u s e n t o m o p h i l y
s y n d r o m e i n c l u d i n g also s p e c i a l i z e d pollen m o r p h o l o g i e s , as in Stellatipollis and re­
lated f o r m s .
T h e katydid e x a m p l e s h o w s that p r o a n g i o s p e r m o u s and, by implication, their suc­
ceeding early a n g i o s p e r m o u s plants could use unconventional pollinators, while bee­
tles and other then existing anthophilous insects w e r e feeding in the m o r e c o n s p i c u o u s
bennettite preflowers (for insects they were flowers irrespective of what plant mor-
phologists might think of them). S o m e extant pollinators of the " p r i m i t i v e " Winteraceae,
such as Sahatinca ( M i c r o p t e r i g i d a e ) , are closely related to the fossil forms (Para-
 'i ^ i

3 1

Fig. 43. Pwhlematospennum ovale Turutanova-Ketova, presumably bennettitalean pappose seeds from
the Late Jurassic of Karatau, Kazakhstan (Krassilov, 1973b), xl (2, 3), 5 ( 1 ) and 7 (4); sinuous cell walls
of the seed coat, x58 and 146 (5, 6) and pollen grains stuck to the micropylar tube, x562 (7).
sabatinca) appearing in the early N e o c o m i a n (Thien & al., 1985) prior to the advent of
angiosperms. With a decline of bennettites in the m i d - C r e t a c e o u s , their insect s y m b i o n t s
passed over to their replacing a n g i o s p e r m s . This was a turning point in the history of
terrestrial e c o s y s t e m s m a r k e d by the first a p p e a r a n c e s of the insect g r o u p s that formed
a core of the C e n o z o i c e n t o m o f a u n a s (Zherikhin, 1979). T h e a n g i o s p e r m s exploring the
temporarily vacant, after the m a s s extinction of bennettites, cantharophilous niche, have
developed large solitary flowers of m a n y parts that are characteristic of cantharophily
and are often considered primitive (e.g. Takhtajan, 1966).
As it w a s suggested e l s e w h e r e (Krassilov, 1981), the m i d - C r e t a c e o u s extinction of
giant s a u r o p o d s and their replacement by the newly appearing d i n o s a u r herbivores with
crashing dental batteries could have been related to the bennettite by a n g i o s p e r m re­
placement that, at the life-form level, m e a n t a substitution of the sparsely branched
pachycaul forms with large leaves by the copiously branched leptocaul forms with small
leaves characteristic of early a n g i o s p e r m s . T h e e n s u i n g increase in shoot m a s s at the
expense of the drastically reduced leaf m a s s required adequate dietary and m o r p h o l o g i ­
cal innovations in the larger herbivores. T h e i r indiscriminate shoot b r o w s i n g (Wing &
Tiffney, 1 9 8 7 ) , p e r h a p s i n v o l v i n g r e p r o d u c t i v e s h o o t s , m i g h t h a v e p r o m o t e d
endozoochory to a leading plant dispersal strategy. At the s a m e time, a c o m p a r a t i v e
analysis of the dental m o r p h o l o g i e s and digestion s y s t e m s indicates s e m e n i v o r y as a
primary p h y t o p h a g o u s adaptation in m a m m a l s (Agadjanian, 1996). S e m e n i v o r y w a s
obviously related to the a p p e a r a n c e , in Caytonia already, of fleshy c u p u l e s containing
many small seeds. T h e overall significance of this feeding habit could h a v e increased
with the advance of the Cretaceous multituberculates, s o m e of t h e m arboreal (Lillegraven
& al., 1979), c o m i n g to d o m i n a n c e in the small vertebrate herbivore n i c h e . T h e r e w a s a
concomitant increase in the multituberculate teeth and the smaW Spermatites - Costatheca
type seed frequencies in the fossil a s s e m b l a g e s . T h e simultaneous rise of a n g i o s p e r m s
and m a m m a l s might thus have been d u e to coevolution of s e m e n i v o r u o s (frugivory)
and e n d o z o o c h o r o u s strategies.
Epizoochory is another dispersal strategy related to the early a n g i o s p e r m - m a m m a l
interaction. P u b e s c e n c e is a characteristic feature of bennettitalean reproductive struc­
tures (Harris, 1969) that, in the derived forms, produced the p a p p o s e d i s s e m i n u l e s , as in
Probleinatospermum (Fig. 43), Typhaera or Baisia (Krassilov, 1973b, 1982b; Krassilov
& Bugdaeva, 1982). T h e s e structures, perhaps primarily a n e m o c h o r o u s , w e r e preadapted
to dispersal by sticking to furred a n i m a l s . T h e g l o c h i d - l i k e a p p e n d a g e s , as in the
ceratophyllacean fruits from the Aptian of Australia ( D o u g l a s , 1969; D i l c h e r & al.,
1996;) and the c o n t e m p o r a n e o u s fruit-like structures from eastern A s i a (Krassilov, 1967,
1977; Plate 43) point to e p i z o o c h o r y as an early appearing dispersal m e c h a n i s m .
One more aspect of the plant-phytophage interaction is related to the biochemical defense
evolution in angiosperms. In their progenitorial stocks, accelerartion might involve not
only the morphogenic, but also the biochemical processes, resulting in abbreviation of the
shikimate pathway in the biosynthesis of detergent alkaloids (Kubitzki & Gottlieb, 1984b)
and the loss of tannins and lignins. Early angiosperms might have thereby been especially
vulnerable to phytophage pressures that had kept them rare until they developed their own
arsenal of detergents, some of which had functionally reversed to attractants in the course
of subsequent biochemical evolution (Pellmyr & Thien, 1986).

A general scheme of ecological evolution

It follows from the preceding discussion that a n g i o s p e r m origins w e r e an integral
part of the M e s o z o i c plant c o m m u n i t y evolution. A n g i o s p e r m s entered the fossil record
after an e x p a n s i o n of the x e r o m o r p h i c brachyphyll-bennettite c o m m u n i t y that had re­
placed coastal fern m a r s h e s in the M e s o z o i c catenic successions. Typically, the angio­
sperms appeared in clusters of a few species representing different life-forms, s o m e of
them aquatic, and w e r e a c c o m p a n i e d by the diverse g n e t o p h y t e s and other proangio­
s p e r m o u s plants that e v o l v e d in the s a m e direction, a c q u i r i n g certain traits of the
a n g i o s p e r m y s y n d r o m e in parallel with the conventional flowering plants.
T h e palaeontological evidence of a "collective b r e a k t h r o u g h " a c c o m p l i s h e d by the
p r o a n g i o s p e r m s g r o w i n g side by side in the "cradle c o m m u n i t i e s " sheds n e w light on
parallel evolution as a result of biotic interactions, involving insects, such as pollinivorous
katydids, that might have transferred gene-transducing m i c r o o r g a n i s m s , thus spreading
mutations across the community. T h e first appearances are confined to the broad ecotone
of the w a r m evergreen or s u m m e r - g r e e n and the temperate s u m m e r - g r e e n zones at about
50°N and 40°S, with the major dispersal centres in the active rift z o n e s . T h e s e circum­
stances suggest geological impacts as a factor impelling a d e v e l o p m e n t a l acceleration
and condensation, the prevailing modes of morphological evolution in early angiosperms.
A m o n g the M e s o z o i c vegetation t y p e s , the x e r o m o r p h i c b r a c h y p h y l l - b e n n e t t i t e
a s s e m b l a g e s of the h e l o p h y t e to littoral facies d o m a i n s c o m p r i s e m o s t of the earliest
p r e - A l b i a n a n g i o s p e r m r e c o r d s . E x p a n s i o n of x e r o m o r p h i c a s s e m b l a g e s u n d e r green­
h o u s e c o n d i t i o n s g a v e i m p u l s e to e c o l o g i c a l and m o r p h o l o g i c a l i n n o v a t i o n s , angio­
s p e r m y o n e of t h e m .
In the early a n g i o s p e r m evolution, there was a "clandestine period", about 120-100
M a , of occasional records suggestive of a subordinate role the a n g i o s p e r m s played in
the brachyphyll-bennettite c o m m u n i t i e s . At this stage, the h e r b a c e o u s forms are some­
what better d o c u m e n t e d than w o o d y forms, with the fossil w o o d records exceedingly
rare and perhaps not always authentical. Typical for this stage w e r e small entire or
lobed leaf blades, an irregular venation network, the inflorescences of inconspicuous
diclinous flowers, and small to m e d i u m - s i z e d follicles and a c h e n e s .
An i m p o r t a n t feature of the early a n g i o s p e r m r e p r o d u c t i v e b i o l o g y w a s an ad­
vanced pollen selection s y s t e m as e v i d e n c e d by the exinal c h a r a c t e r s of the self-in­
c o m p a t i b i l i t y s y n d r o m e . Initially, the d e v e l o p m e n t of the latter m i g h t h a v e been re-
 lated to an increased seed yield by p r e v e n t i n g the o v u l e (carpel, flower) abortion
inflicted by m i s p o U i n a t i o n . A n g i o s p e r m s might h a v e b e e n p r e d o m i n a n t l y d i c l i n o u s
at this stage. With e l a b o r a t i o n , self-incompatibility e n a b l e d a n g i o s p e r m s to a c q u i r e
monocliny ( u n s t a b l e in g y m n o s p e r m s ) , that, in turn, p r o m o t e d e n t o m o p h i l y to a lead­
ing pollination m o d e .
Rarity d u r i n g the c l a n d e s t i n e period m i g h t h a v e been related to the d e v e l o p m e n t a l
acceleration i n v o l v i n g not only m o r p h o l o g i c a l but also b i o c h e m i c a l traits and result­
ing in the loss of certain d e f e n s e c o m p o u n d s (notably lignins and, in the m a g n o l i a n
line, tannins) r e n d e r i n g early a n g i o s p e r m s an easy prey to v a r i o u s p h y t o p h a g e s .
T h e b e n n e t t i t e and b r a c h y p h y l l m a s s e x t i n c t i o n s and the c o m p e n s a t o r y rise of
angiosperms at about 100 M a marked the end of the clandestine period. Insofar as both
these extinct g r o u p s w e r e relatively t h e r m o p h i l o u s , the late Albian-early C e n o m a n i a n
cooling might h a v e been a major factor of their decline. T h e a p p e a r a n c e at that time of
a huge cirum-Pacific volcanic belt could have contributed to extinctions either indi­
rectly, through its climatic effects, or directly, as a source of acid rains that affected both
plants and their insect s y m b i o n t s . Vast marine transgressions in cratonic areas have
caused a reductuion of terrestrial b i o m a s s that seves as a major sink of a t m o s p h e r i c
carbon dioxide. T h e ensuing g r e e n h o u s e conditions have p r o m o t e d high growth rates
giving the early a n g i o s p e r m s an a d v a n t a g e over the slow g r o w i n g g y m n o s p e r m s .
In effect, the c l i m a x p h a s e of the b r a c h y p h y l l - b e n n e t t i t e c o m m u n i t i e s w a s u n d e r ­
mined and w a s r e p l a c e d by the l i k e w i s e x e r o m o r p h i c D e b e y o - D r y o p h y l l e t u m a s s o ­
ciation with s h r u b b y d i c o t y l i d s and p a l m s , the relict b r a c h y p h y l l s still locally a b u n ­
dant. T h e s i m u l t a n e o u s r e p l a c e m e n t s of the m e s o p h y t i c arboreal c o m m u n i t i e s g a v e
rise to the t a x o d i a c e a n - l a u r o p h y l l and t a x o d i a c e a n - p l a t a n o p h y l l forest t y p e s , the lat­
ter mostly riparian. T h e m i d - C r e t a c e o u s rise of a n g i o s p e r m s w a s paralleled by the
rise of m o d e r n conifers and the a p p e a r a n c e of m o d e r n c y c a d s . It w a s , t h u s , not an
isolated e v e n t that c o u l d be a s c r i b e d to a c o m p e t i t i v e r e p l a c e m e n t of g y m n o s p e r m s
by a n g i o s p e r m s , but w a s instead related to the w h o l e s a l e r e s t r u c t u r i n g of d o m i n a n t
plant c o m m u n i t i e s .
T h e r e p l a c i n g plants c o n v e r g e d on the r e p l a c e d . T h e m i d - C r e t a c e o u s a n g i o s p e r m s
inherited the s y m b i o n t s and c o n s u m e r s of the r e p l a c e d b e n n e t t i t e s and b r a c h y p h y l l s
thereby a c q u i r i n g the bennettite-like c a n t h a r o p h i l o u s flowers c u r r e n t l y c o n s i d e r e d
primitive in the e x t a n t flowering plants.
Since m o s t of the early a n g i o s p e r m s had s m a l l e r leaves than b e n n e t t i t e s , the re­
placement resulted in a l o w e r leaf m a s s to shoot m a s s ratio. T h e larger v e r t e b r a t e
herbivores had to c o p e with the leaf m a s s reduction by a c q u i r i n g n e w nutritional
habits, such as shoot b r o w s i n g i n v o l v i n g r e p r o d u c t i v e s h o o t s and p r o m o t i n g e n d o -
zoochory. N o less i m p o r t a n t for e n d o z o o c h o r o u s dispersal a n d t h e d e v e l o p m e n t of
the respective fruit m o r p h o l o g i e s , to a certain extent a n t i c i p a t e d by Caytonia among
p r o a n g i o s p e r m s , w a s s e m e n i v o r y , a p r i m a r y p h y t o p h a g o u s a d a p t a t i o n in m u l t i t u b e r -
culates and o t h e r early m a m m a l s . T h e s e d e v e l o p m e n t s c o i n c i d e d with the a p p e a r a n c e
of a b u n d a n t small seeds in the L a t e C r e t a c e o u s microfossil plant a s s e m b l a g e s . E p i z o o -
c h o r o u s a d a p t a t i o n s , arising on the basis of p a p p o s e w i n d - b o r n e d i s s e m i n u l e s wide­
spread a m o n g the p r o a n g i o s p e r m s , c o n s t i t u t e d a n o t h e r a s p e c t of the early a n g i o s p e r m
- early m a m m a l c o e v o l u t i o n that led both g r o u p s to d o m i n a n c e in their respective
domains.
T h e next stage of a n g i o s p e r m evolution is related to the C r e t a c e o u s / T e r t i a r y bound­
ary e v e n t s . T h e m o s t clearly m a r k e d a m o n g t h e m w e r e the L a r a m i d t e c t o n i s m affect­
ing both the c i r c u m - P a c i f i c and A l p i n e belts, the c o n c o m i t a n t retreat of epicontinental
seas and the global c o o l i n g inflicting a c o m p l e t e r e s t r u c t u r i n g of the d o m i n a n t plant
c o m m u n i t i e s . T h e fern m a r s h c o m m u n i t i e s d i s a p p e a r e d , and there w a s a r e d u c t i o n of
x e r o m o r p h i c s h r u b l a n d s , the c r a d l e of a n g i o s p e r m s that c a m e to d o m i n a n c e t h e r e in
the Late C r e t a c e o u s . T h e s e highly p r o d u c t i v e and resilient vegetation t y p e s supported,
t h r o u g h the M e s o z o i c era, large d i n o s a u r h e r b i v o r e s that p e r i s h e d with t h e m .
In the a d v a n c i n g d e c i d u o u s forests, there w a s a r e p l a c e m e n t of Sequoia-VxkQ coni­
fers and laurophyll a n g i o s p e r m s , d o m i n a n t in the L a t e C r e t a c e o u s , by Metasequoia
and the b r o a d l e a v e d h a m a m e l i d g e n e r a that had p r e v i o u s l y c o n s t i t u t e d a riparian c o m ­
p o n e n t . T h e i r e m e r g e n c e as the d o m i n a n t s of d e c i d u o u s forests g a v e the latter a mod­
ern b r o a d l e a v e d aspect, but in fact these P a l e o c e n e forests w e r e d o m i n a t e d by extinct
genera, such as Trochodendroides, Steinhauera (Plate 4 4 ) a n d / o r Nordenskioldia.
T h e y soon r e c e d e d before the then a p p e a r i n g m o d e r n a m e n t i f e r s . Yet t h e s e chapters
of a n g i o s p e r m history are b e y o n d the s c o p e of this b o o k .
 Phylogeny
P h y l o g e n y is interesting not so as a guide to w h o b o r e w h o m as a reflection of
evolutionary p r o c e s s e s . T h e a n g i o s p e r m p h y l o g e n y incorporates the p r o c e s s e s of seed
plant evolution that started long before the a p p e a r a n c e of a n g i o s p e r m s . A s outlined in
the preceding chapters, the seed plant evolution is divisible into c h r o n o l o g i c a l stages,
each characterized by the prevalence of morphological characters that define succes­
sive evolutionary grades. A n g i o s p e r m s are successors of the h y d r o s p e r m , pteridosperm,
cycadophyte and p r o a n g i o s p e r m grades. T h e i r precursory character states have been
a c c u m u l a t i n g from g r a d e to g r a d e until their a l m o s t c o m p l e t e set a p p e a r e d at the
proangiosperm level.
M o n o p h y l y versus polyphyly of a n g i o s p e r m s is a m u c h debated, yet little-studied,
problem. As H u g h e s (1961) noticed, decisions are often m a d e in a d v a n c e of the study.
Indeed, the p r o b l e m has s o m e non-scientific aspects, such as the Biblical N o a h legend
of each species arising from a single pair, or an aesthetic preference of tree topography
over bush topography, or, m o r e importantly, a s u b c o n s c i o u s urge for unity i m p e d e s
satisfaction until all items are brought to one end. In effect, m o n o p h y l e t i c systems are
much more popular than polyphyletic systems, of which even the most thoroughly worked
out ones (e.g. Kuznetsov, 1936) have had little c h a n c e of survival.
T h e cladistic techniques provide a seemingly objective approach to the p r o b l e m by
numerical estimates of intergroup versus o u t g r o u p similarities. It is a s s u m e d that, for
each g r o u p , there is a c o m m o n ancestor, the p r o b l e m is only h o w d e e p in the tree it can
be found. T h e p r o b l e m thus boils d o w n to the closest c o m m o n ancestor being or not
being a m e m b e r of t h e g r o u p c o n c e r n e d . C o r r e s p o n d i n g l y , t h e g r o u p is e i t h e r
holophyletic, neatly interconnected by the alternative character states arising from each
other, or paraphyletic, with the character states arising independently in the clades that
diverged before the a p p e a r a n c e of the g r o u p . But the c o n c l u s i o n s are, to a considerable
extant, p r e d e t e r m i n e d by the basic a s s u m p t i o n s .
A p h y l o g e n y of even a single g e n u s can b e biased by the a s s u m p t i o n of m o n o p h y l y ,
while an u n p r e j u d i c e d analysis indicates p o l y p h y l y ( L a n y o n , 1994). In the c a s e of
such a d i v e r s e and long studied g r o u p as a n g i o s p e r m s , the p r o b l e m is further c o m p l i ­
cated by the a m b i g u o u s c h a r a c t e r cline polarities, p s e u d o h o m o l o g i e s , s e m a n t i c in­
consistencies, etc. If the typical c h a r a c t e r s are p r i m i t i v e , then the c o m m o n a n c e s t o r
will certainly be an a n g i o s p e r m . If, on the o t h e r h a n d , the a b e r r a n t c h a r a c t e r s are
primitive, then it m a y not b e an a n g i o s p e r m . W h a t e v e r the o u t g r o u p , the unity of
a n g i o s p e r m s can be easily c o r r o b o r a t e d on the basis of c h a r i s m a t i c c h a r a c t e r s , such
as flower, fruit, d o u b l e fertilization, e t c . F o r e x a m p l e :
 Gnetophytes Dicotyledons Monocotyledons

Flower - + +
Stamen - + +
Carpel - + -i-
Cupule +

Before taking this table seriously, w e ought to look into definitions and possible
h o m o l o g i e s of the characters used. O t h e r w i s e their p r e s e n c e or a b s e n c e d e p e n d s on
what we m e a n by t h e m .
A n o t h e r p r o b l e m is consistency of intragroup divisions. If, in the a b o v e example,
s o m e taxa nominally assigned to m o n o c o t s are, on the bulk of the e v i d e n c e , closer to
dicots or vice versa, then the morphological distance b e t w e e n the t w o w o u l d appear
smaller than it actually may be. But this is just what is constantly h a p p e n i n g to the
higher-rank a n g i o s p e r m divisions. T h e magnoliids, for e x a m p l e , are fairly distinct from
the ranunculids on the basis of their gynoecial, as well as their stamen and pollen
m o r p h o l o g i e s . But, with a questionable inclusion of the Illiciaceae and Schizandraceae
in the former g r o u p , the distinctions fade out. A n g i o s p e r m divisions s e e m not so natu­
rally interlinked as intermixed by m i s a s s i g n m e n t s .

Morphology
T h e following compilation of typical characters may serve as a m o r p h o l o g i c a l diag­
nosis of a n g i o s p e r m s :
C r e t a c e o u s - extant; life-form arboreal, leptocaul or p a c h y c a u l or brachycormic,
herbaceous, aquatic, geophytic, epiphytic, etc. Roots typically tetrarch, lateral roots
often diarch or triarch, in m o n o c o t s often polyarch. Apical m e r i s t e m stratified, with
tunica-corpus differentiation, but the latter less distinct or lacking in m o n o c o t s ; second­
ary growth by apical, in m o n o c o t s m o r e typically by lateral, m e r i s t e m . P r i m a r y vascu­
lar system of free or interconnected symposia, with vascular b u n d l e s collateral or -
m o r e often in m o n o c o t s - concentric, large and small alternating. W o o d , with a few
exceptions, p o r o u s , with vessels; sieve e l e m e n t s with c o m p a n i o n cells and with S-
p l a s m i d s . N o d e s 1-lacunar with t w o leaf traces, less often 3 - or multilacunar. Shoots
often h e t e r o m o r p h i c , s o m e t i m e s heteroblastic. L e a v e s spiral or d e c u s s a t e , less com­
monly verticillate, e x p a n d e d laterally by marginal and plate m e r i s t e m s , simple or com­
p o u n d , s o m e t i m e s s h e a t h i n g , s o m e peltate or ascidiform, bifacial or, in monocots,
unifacial, c a u l o m i c or phyllodic; stipules c o m m o n ; venation reticulate, pinnate or pal­
mate, with 3-5 orders of areoles, the ultimate ones often filled with blind veinlets, or,
more often in monocots, primary veins parallel or a c r o d r o m o u s , with c o m m i s s u r a l cross-
veins, with alternating large and small bundles and with interbundle sclerenchymous
strands; stomata anomocytic, cyclocytic, paracytic or tetracytic, with subsidiary cells
mesogenous, perigenous or mesoperigenous, other types (about 20) less c o m m o n .
Reproductive shoots with apical floral meristem typically producing cyclic aggre­
gates of androecial and/or gynoecial structures (flowers) s o m e t i m e s intermingled with
intrafloral phyllomes; flowers solitary or gregarious in inflorescences of m a n y types,
diclinous or m o n o c l i n o u s , with perianth c o m p o n e n t s heteromorphic or h o m o m o r p h i c ,
their m e m b e r s free or connate, or sometimes perianth lacking. Pollen organs, or sta­
mens, arranged in 1-2, rarely more cycles, solitary or in fascicles, supplied with solitary
or dendroid interconnecting traces respectively, rarely forked or branched or laminar
and supplied with 1-3 traces, developmentally centripetal or centrifugal, typically fila­
mentous, with anthers terminal or lateral, in laminar stamens abaxial, rarely adaxial or
marginal; anther consisting of a sterile connective, this s o m e t i m e s protruding (produced)
apically, and a bilobed synangium, 4-sporangiate, rarely of 2 or m a n y sporangia.
Pollen grains dispersed singly, less c o m m o n l y as m o n a d s or persistent tetrads, polyads
or pollinia, bi- or trinucleate, these conditions correlating with dry or wet stigma types
respectively, ana-, cata-, trichotomo-, or zonosulcate, variously colpate, porate or col-
porate, rarely inaperturate, coated with intine and exine, in aquatic forms sometimes
with intine alone, asaccate but s o m e t i m e s cavate, with airspace b e t w e e n sexine and
nexine, rarely with residual sacci; covered with trifine or pollenkit, tectate or semitectate,
with tectum reticulate or perforate (these structures correlating with s p o r o p h y t i c or
gametophytic self-incompatibility), rarely atectate; pollen wall infrastructure columellate
or granulate; lamination of nexine lacking (with exceptions) or lamellae present under
apertures alone.
G y n o e c i u m of solitary to many closed or partly open c o m p o u n d ovuliferous struc­
tures, or pistils, exposed or partly or totally sunken below and connate with perianth
and/or a n d r o e c i u m , or sunken in receptacle, provided with pollen receptive structures,
or stigmas, these raised on hollow or solid styles or sessile, s o m e t i m e s formed by sur­
rounding structures, as in receptacular hyperstigma {Tamhourissa)\ pistils consisting of
one or t w o to many ovuliphores, or carpels, these distinct, c o n n a t e or partly so, with
ovules produced basally, apically, marginally, on sutures or p r o x i m a l to t h e m or scat­
tered on walls or arising from floral axis axillary to carpels or a c a r p e l l a t e ; o v u l e s
orthotropous or anatropous or of transitional polarities, uni- or bitegmic, rarely with 3 -
4 integuments, with micropyle formed by inner or outer integument or both, sometimes
zigzagged; n o n e or seldom one of integuments, rarely both, vascularized, s o m e cutinized
and stomatiferous; nucelli sometimes beaked, with or without parietal tissue and with
archesporium more or less reduced.
Pollen tubes s i p h o n o g a m o u s , reaching the e m b r y o sacs by various routes, often dis­
charged into synergid; m e g a s p o r o c y t e single, rarely m a n y (Paeonia)\ m e g a s p o r e tet­
rads of five basic types, often linear or T-shaped; e m b r y o sac m o n o - , bi- or tetrasporic.
seldom with haustoria or protruding into style, of different developmental types, more
c o m m o n l y 8-nucleate with micropylar and chalazal groups of nuclei or, in tetrasporic
e m b r y o sacs, also with lateral groups or with chalazal nuclei lost (Oenothera) or some­
times multiplied (Sparganium)\ syngamy involving two or m o r e nuclei, seldom want­
ing, e n d o s p e r m nucleus triploid or polyploid, seldom diploid; seeds shed at different
stages of embryogenesis, dispersed separately and sometimes arillate or dispersed with
fruits, these dehiscent, septicidal or loculicidal, or indehiscent, dry or fleshy of many
types, formed of gynoecia or their parts or involving other parts of flowers or even
inflorescences. Embryogenesis cellular, seldom initially free nuclear {Paeonia), endo­
sperm development free nuclear, less c o m m o n l y cellular or first cellular then free nu­
clear. P r o e m b r y o cell pattern of several types, differing in contribution of the basal cell,
rarely irregular (Degeneria). Cotyledons one, terminal or pseudoterminal and haustorial,
or two, seldom more, sometimes heteromorphic and one haustorial (Piperomia) or coty­
ledons unidentified.

Character sharing
A n g i o s p e r m s are usually thought of as sharing m o r e characters with each other than
with any of n o n - a n g i o s p e r m o u s seed p l a n t s . T h i s m a y be d u e to the functionally
intercorrelated angiospermy syndrome comprising a great n u m b e r of discrete charac­
ters related to production, pollination and dispersion of enclosed ovules, and constitut­
ing the bulk of the c o m m o n l y analyzed characters. With the functionally not or weekly
correlated character sets the situation might have been be entirely different. A morpho­
logical analysis of the present-day angiosperms shows the following:

( I ) T h e typical or diagnostic characters are those shared by a considerable number

of angiosperm species. N o n e of such characters is shared by all species currently as­
signed to angiosperms. There are alleged angiosperms lacking a distinct stratification of
the apical meristem, with an a n o m a l o u s secondary growth, h o m o x y l i c (the Winteraceae,
Trochodendraceae, Tetracentraceae, N y m p h a e a c e a e , Ceratophyllaceae and a few Moni-
miaceae; Sarcandra (Chloranthaceae) has been considered as vesselless, but actually it
has m e t a x y l e m vessels: Carlquist, 1987; Takahashi, 1988), lacking typical sieve ele­
ment c o m p a n i o n cells (Austrohaileya: Behnke, 1986), with open venation (Kingdonia,
Circaeaster: Foster & Gifford, 1974), with the flowers " r e d u c e d " to a single sporangio­
phore (Hedyosmum) or ovuliphore (Cercidiphyllum), with sacci (Lactoris: Z a v a d a &
Taylor, 1986b), with lamellate endexine (Austrohaileya: Z a v a d a , 1985; Degeneria, some
Liliaceae, etc.) or the endexine lacking (Hedycarya, s o m e m o n o c o t s ) , with the pistils
open (Drimys, Platanus, Sassafras, etc.), with intracarpellate pollen germination (some
A n n o n a c e a e , Alismataceae, Liliaceae: Vasil & Johri, 1964; Johri, 1984), acarpellate
(Sattler, 1974a), with both integuments vascularized (Euphorbiaceae), with a massive
perisperm (Piperaceae), with m a n y chalazal e m b r y o sac nuclei (Sparaganium), lacking
a typical double fertilization (Onagraceae), with e m b r y o g e n e s i s nuclear u p to 64-128 or
even 256-nucleate stages (Paeonia: Yakovlev, 1973), etc. C o m b i n e d , all these "aber­
rant" states would give a plant profoundly different from typical a n g i o s p e r m s and hardly
assignable to angiosperms at all.

(2) M o s t of the "aberrant" angiosperm character states are typical for s o m e other
seed plant groups. A n d , on the contrary, virtually all the typical a n g i o s p e r m features
occur as aberrant in other seed plant groups, e.g. the apical tunica-corpus differentiation
(Araucariaceae, gnetaleans), vessels (in gnetaleans, previously thought of as being non­
homologous to those of angiosperms, but actually perhaps h o m o l o g o u s : M u h a m m a d &
Sattler, 1982) and in bennettites (Krassilov, 1982b), unifacial leaves (Acmopyle, Dac-
ridium), stipules (cycads), a hierarchical reticulate venation (gigantopterids, Gnetum),
mesogenous stomata (bennettites: Krassilov, 1978a; gnetaleans: M a h e s h w a r i & Vasil,
1961; s o m e pteridosperms and conifers: Johnson & Riding, 1981; R e i h m a n & Schabilon,
1985), a flower-like arrangement of the tepals as well as of the androecial and gynoecial
organs (bennettites), monocliny (hydrospermaleans, bennettites, gnetaleans, occasion­
ally conifers: Tosh & Powell, 1986), persistent tetrads (ClassopoUis), multiaperturate
pollen (Welwitschia, ClassopoUis), a lack of endexine lamination or this discernible at
the early d e v e l o p m e n t a l stages alone (Crossotheca a m o n g p t e r i d o s p e r m s , the extant
ginkgo: Rohr, 1977), closed ovuliphores, sometimes with stigmatic crests (Leptostrobus),
bitegmic ovules (Caytonia: Krassilov, 1984a), extraovular pollen germination (Arau­
cariaceae, Saxegothea, Tsuga, some Cupressaceae, probably Hirmerellaceae), archegonia-
less e m b r y o sacs (Gnetum, Welwitschia), fusion of the e m b r y o sac nuclei (gnetaleans,
some conifers: K o n a r & O b e r o i , 1969), functioning of both s p e r m s in copulation (regu­
larly in p o d o c a r p s , g n e t a l e a n s , etc.), d o u b l e fertilization involving the ventral canal
cell n u c l e u s (Ephedra), cellular e m b r y o g e n e s i s (Sequoia), fruit-like dispersal units
involving periovular structures ( p t e r i d o s p e r m s , cayionialQans, Araucaria, Welwitschia,
Hirmerella), etc.
Thus, the morphological boundary between angiosperms and other groups of seed
plants is not absolute. Notably, a n g i o s p e r m s are m o r e distinct from the present-day
gymnosperms than from the fossil non-angiospermous groups having their ovules en­
closed in the pistil-like cupules and with the androecial and gynoecial parts arranged
cyclically as in the typical flowers. A possible reason for the disappearance of these
angiosperm-like plants might have been that they have evolved into the full-fledged
angiosperms. O n e dealing with the M e s o z o i c e x a m p l e s alone would hardly d r a w a ma­
jor demarcation line between the then existing angiosperms, on the o n e hand, and all
other seed plants, on the other, and would perhaps cluster caytonias and bennettites with
angiosperms rather than with conifers.
 (3) Not a single character considered typical of angiosperms is uniform within the
group. There are at least two types of vascular sympodia, open and reticulate, the latter
more c o m m o n in herbs; four basic types of nodal anatomy; t w o basic leaf types, one
occurring in dicots, the other, supposedly phyllodic (Kaplan, 1970), in m o n o c o t s , but
the " d i c o t y l e d o n o u s " type occurring in such monocot families as Smilacaceae, Phile-
siaceae, Dioscoreaceae, Stenomeridaceae, Triliaceae and Taccaceae, mostly climbers
(Conover, 1982); about 20 stomatographic types, also n o n - r a n d o m l y distributed be­
tween dicots and monocots, the latter being more uniform in this character (Tomlinson,
1974; Rasmussen, 1983; Inamdar & al., 1986); many distinct flower types distinguished
on the basis of their having the full ("perfect") or only a partial c o m p l e m e n t of the floral
organs, a single set or two sets or no perianth organs, as well as being monoclinous or
diclinous, in the relative positions of the perianth, androecium and g y n o e c i u m on the
floral receptacle, in their morphologically recognizable or supposed pseudanthial or
euanthial origins, etc.
To this can be added two types of androecia, with solitary or fasciculate stamens;
two developmental types of pollen grains and about 16 apertural types, of which the
anasulcate one is more c o m m o n in monocots; six major placentation types; t w o basic
types of ovule polarity with modifications and two integumental types, also with modi­
fications; 11 e m b r y o sac types, five types of p r o e m b r y o cell patterns; and, eventually,
two cotyledon types, formally diagnostic of dicots and m o n o c o t s , although at least 11
dicot species have a single cotyledon. T h e caryological and biochemical characters (not
included in the above description, because only an insignificant portion of angiosperm
species have been hitherto appropriately studied) are likewise separable into distinct
types differing in the c h r o m o s o m e numbers (with at least three distinct modal levels of
n=5-8, 12-14 and 19-21), the differentiation of sex c h r o m o s o m e s , etc. (Stebbins, 1974;
Grant, 1975; Gottlieb, 1984; Kubitzki & Gottlieb, 1984a, b; Antonov, 1995). Remark­
ably, the M a g n o l i a c e a e is isolated biochemically by lacking tannins and s o m e other
c o m p o u n d s that are c o m m o n in the rest of angiosperms as well as in non-angiospermous
groups (Blagoveshchenskiy, 1975; Cronquist, 1977).

(4) T h e character states assignable to a certain morphological type are not necessar­
ily h o m o l o g o u s in all group m e m b e r s and, in fact, are rarely so. For e x a m p l e , t w o leaf
traces of a 1-lacunar node can arise form a single or, as in Austrohaileya, from different
axial sympodia; the c o m p o u n d leaves can result from a localized activity of the mar­
ginal meristem or the leaflet primordia are differentiated before the marginal meristem
(Periasamy & Muraganathan, 1985); both the paracytic and cyclocytic stomata can be
developmentally m e s o g e n o u s , perigenous or m e s o p e r i g e n o u s ; the spiral arrangement
of the flower parts is either primary (Endress, 1983) or derived from a cyclic arrange­
ment (Tucker, 1960); the developmentally fasciculate stamens arising from a c o m m o n
primordial k n o b may appear solitary in the ripe androecia (Pauze & Sattler, 1978); they
can arise either in centrifugal or centripetal sequences; the "typical" follicles often thought
of as the leafy ovuliferous stmctures can be neither leafy in their shape and vascularization
- that is, vascularized otherwise than by the canonical single dorsal and t w o ventral
traces, incidentally receiving five traces, as in the R a n u n c u l a c e a e and Berberidaceae
(Chute, 1930; Terabayashi, 1985), or two dorsal traces, as in Austrohaileya, or the ven­
tral traces departing from the cortical, rather than axial, bundles - nor ovuliferous (with
the ovules arising from the floral axis), or these structures are actually p s e u d o m o n o m e r o u s
(Vink, 1978); s o m e apparently orthotropous ovules are pseudoorthotropous, while the
superficially crassinucellar ovules can be pseudocrassinucellar (Davis, 1966; S p o m e ,
1974); the m o r p h o l o g i c a l l y similar i n t e g u m e n t s can be d e v e l o p m e n t a l l y d e r m a l or
subdermal or derived from more than one histological zone, while a u n i t e g m i c condi­
tion can result from fusion or reduction or shifting (De B o e r & B o u m a n , 1972; B o u m a n
& Calis, 1977), whereas a bitegmic condition can ensue from splitting of a single in­
tegument or the peripheral nucellar tissues in respect to the pollen-conducting function
(Heslop-Harrison & al., 1985). There are different developmental relations between the
primary coats of the ovule and the testa/tegmen histological differentiation of the ripe
spermoderm ( B o u m a n n , 1984; Schmid, 1986). A single cotyledon can result from fu­
sion (syncotyly) or an asymmetrical development and reduction (heterocotyly), while
an apical cotyledon can be developmentally lateral, as in p a l m s (Guignard, 1984) or, if
actually terminal as in Lemna, then perhaps n o n - h o m o l o g o u s to other cotyledons (Arber,
1925). T h e caryotypes are likewise liable to p s e u d o h o m o l o g y due to intricate c o m b i n a ­
tions of polyploidy and aneuploidy (Grant, 1975), w h e r e a s s o m e characteristic bio­
chemical c o m p o u n d s actually arise as the end-products of dissimilar biosynthetic path­
ways (Kubitzki, 1973). P s e u d o h o m o l o g y in a n g i o s p e r m s is a prolific source of phylo­
genetic misconceptions.

Character polarities
For m a n y characters, the extreme states can be linked by intermediate states in a
more or less continuous sequences, or morphoclines. W h e n based on c o n t e m p o r a n e o u s
forms, the morphoclines can be read in either direction. Historically, t w o c o m p e t i n g
schools of angiosperm morphology differed primarily in their giving priority to either a
simple to c o m p l e x or the opposite directions, in particular, a progression from the in­
conspicuous gregarious flowers of few parts to the conspicuous solitary flowers of many
parts, or vice versa (Wettstein, 1907; Bessey, 1897, 1915; F a m e s , 1961). T h e underly­
ing reasoning has had nothing to d o with the actual chronological trends but has been
rooted instead in philosophical doctrines, such as the essentialistic Urpflanze concept
holding that any progression is a reduction from the most perfect archetype to its less
perfect reflections in the perceptible plant diversity.
 A new approach introduced into plant morphology by the theory of evolution was
the outgroup comparisons, that is, a character state evaluation by c o m p a r i s o n with a
close, but distinct group, the pre-existing philosophical differences n o w affecting the
choice of a particular outgroup. W h i l e Ephedra was used as an outgroup by Wettstein
(1907), the then recently discovered bennettites played this role for the Hallierian school.
Although with time the relevance of both Ephedra and bennettites as outgroups w a s put
to doubt, the polarity controversies remained.
If a n g i o s p e r m s were monophyletic, then all the alternative m o r p h o t y p e s should be
derivable, directly or sequentially, from o n e of them. T h e comparative plant morphol­
ogy is governed by this creed and is, thereby, focused on sequencing the morphotypes
from the primitive (primary) to the advanced (derived), yet the criteria of advancement
are less advanced than might be expected. T h e following are c o m m o n l y used:

(a) T h e statistical, or "the c o m m o n is primary". Since the growth forms with the
nearly continuous subapical primary and lateral secondary thickening meristems occur
in not so n u m e r o u s woody monocots (Chakroun, 1983; Stevenson, 1980a, b ; Diggle &
D e M a s o n , 1983) and a few dicots, such as the Piperales, it is conceived of as "anoma­
lous" and, by implication, derived; the anatropous and the bitegmic ovules are more
widespread (the latter occurring in 152 of 341 dicot families and in 52 of 6 9 monocot
families: B o u m a n & Calls, 1977) and are, by virtue of it, primary in respect to the
orthotropous and the unitegmic ovules; the ''Polygonum'' type of e m b r y o sac develop­
ment is m o r e c o m m o n than the others and is, therefore, primary, etc. This approach
tends to dismiss rare character states as of no phylogenetic significance. It is inherited
from the typological Urpflanze concept (see above) in which all the essential characters
occur in the archetype, wliereas any departures from the norm are secondary.

(b) T h e conceptual, determined by a phylogenetic paradigm. This concept is also

typological, but the primary versus secondary character states are determined on the
basis of a preconceived idea of primitiveness rather than of any statistical considera­
tions. For instance, since all floral organs are modified sporophylls, the laminar sta­
m e n s , although relatively rare, should be primitive (despite the fact that they are highly
specialized, sometimes e n d o w e d with a capilliform connective penetrating between the
carpels and holding the stamen in a fixed position during pollination, see Canright,
1952, while their vascular anatomy is highly diverse and, in Victoria at least, rather
caulomic than phyllomic, see Heinsbroek & Heel, 1969, or superficially phyllomic with
three traces, but the latter departing from the cortical rather than axial bundles, as in
Degeneria). F o r the s a m e reason, a follicular carpel has been p r o c l a i m e d primitive
whatever the interpretation that varied in response to the involute, conduplicate, and
conduplicate-peltate ascidiform models of megasporophyll transformation successively
replacing each other (Sinnott & Bailey, 1914; Troll, 1932; Bailey & S w a m y , 1951;
B a u m , 1949; Lienfellner, 1950; Rohweder, 1967; Tucker & Gifford, 1974).
 T h e s e m o d e l s imply that a primitive placentation should be marginal or submarginal
or localized in the ventral Q u e r z o n e respectively. That such features of the follicular
carpels as their tubular growth from annular primordia (Rhoeweder, 1967), characteris­
tic of c o m p o u n d structures, or their vascularization by the double dorsal and sometimes
also double ventral traces, as in Drimys sect. Tasmannia (Tucker, 1975) are scarcely
phyllomic, passed u n n o t i c e d until s o m e typical follicles h a v e been reinterpreted as
p s e u d o m o n o m e r o u s (Leroy, 1977; Vink, 1978), that is, fairly advanced.
With a c h a n g e of the paradigm, a totally different m o r p h o t y p e can be selected as
primary or primitive. T h u s , the appearance of the telome concept replacing the p h y l l o m e
concept of a basic morphological unit has m a d e the branched stamens, as in the Euphor-
biaceae or M o n i m i a c e a e , conceivably more primitive than the laminar stamens (Wilson,
1937). Characters are often labelled primary if they occur in a paradigmatic primitive
group, such as the Magnoliaceae. This is an insecure approach because, with the primi­
tive status of the Magnoliaceae reconsidered (Bate-Smith, 1972; Boulter & al., 1972;
Martin & D o w d , 1984, 1986), all the polarities are to be reversed.

(c) T h e technological, holding that of t w o character states the primary is that of

which the other can be most readily manufactured. T h u s , inaperturate pollen grains are
primary because they provide a carte blanche upon which any apertural type can be
developed (Muller, 1970; Walker, 1974a). T h e monoclinal flowers are primary because
the diclinal flowers can be produced from t h e m by the loss of either a n d r o e c i u m or
gynoecium, and a loss is certainly easier than an acquisition. Similarly, a41 the perianthless
flowers are "reduced", even if developmental evidence of the reduction is totally lack­
ing, as in the Saururaceae (Tucker, 1975, 1981). Technologically, inferior ovaries might
arise by fusion with the receptacular and/or appendicular tissues, but, in s o m e cases at
least, developmental evidence is ambiguous (Douglas, 1944; Komar, 1972; Sattler, 1974a;
Leins & Erbar, 1985). On the s a m e ground, a succession from c o m p l e x to simple is
commonly preferred to the opposite. Phytochemical evolution is often assessed techno­
logically, with a p r e c u r s o r y c o m p o u n d of a b i o s y n t h e t i c p a t h w a y c o n s i d e r e d as
phylogenetically primitive. This approach, though fairly popular, is based on an unwar­
ranted analogy between organic evolution and h u m a n technological experience.

(d) T h e correlational, holding that primitive characters are those correlated with other
primitive characters. T h e logic of such deductions is not entirely convincing, for one
can argue that, a m o n g contemporaneous species, primitiveness of s o m e characters should
be compensated by advancement in s o m e others. In practice, the correlational assess­
ments are often a m b i g u o u s . T h u s , the 3-lacunar nodal anatomy correlates with stipules
that are primary according to Cronquist (1968), but derived a c c o r d i n g to Takhtajan
(1970) and Stebbins (1974), while the 1 -lacunar nodes correlate with simple entire leaves
and occur in the homoxylic species that are c o m m o n l y , but not unanimously, considered
as primitive. At the same time, they correlate with decussate leaves that are thought of
as derived by many morphologists (Bailey, 1956; Carlquist, 1985), but not by Stebbins
(1974). T h e fasciculate stamens might be secondary, for their d e v e l o p m e n t is, with a
few exceptions, centrifugal, an allegedly derived pattern, but there is a substantial de­
velopmental evidence to the contrary (Stebbins, 1974; Sporne, 1974).
In the correlation approach, it is difficult to avoid circularity. S i m p l e entire leaves
are primary because they are characteristic of magnoliaceans that are primitive by vir­
tue of having, a m o n g other supposedly primary characters, simple entire leaves. This
approach does not take into consideration differential rates of character evolution re­
sulting in mosaic morphology. T h e notion of magnoliaceans as the most primitive liv­
ing a n g i o s p e r m s , justified or not, d o e s not m e a n that all their c h a r a c t e r s , including
multilacunar nodal anatomy and the lack of tannins, are primary.

(e) T h e outgroup-directed polarizations in which a character state is hold primitive if

occurring in a taxon marginally related, but not assigned in the group and, on the whole,
more primitive. T h e homoxylic condition, for instance, is hold primary in angiosperms
for it is widespread in g y m n o s p e r m s . This would m e a n , according to the monophyly
concept, that g y m n o s p e r m s having vessels acquired them in parallel with angiosperms
and not as phylogenetic precursors of the latter. However, the h o m o x y l i c w o o d s are not
uniform in respect to the structure of tracheary elements and they m a y be primary in
s o m e families but secondary in the others, thus requiring a selective outgroup compari­
son. I have already mentioned that competing phylogenetic concepts are based on the
selection of either Ephedra or bennettites or Caytonia as the o u t g r o u p . But these
outgroups may not necessarily be mutually exclusive. They and s o m e others could jointly
have contributed to the angiosperm morphological diversity. This idea is, however, in­
compatible with a methodology assuming monophyly as the starting point of a morpho­
logical analysis and inevitably leading to monophyletic arrangements as its end-pro­
duct, thus making the monophyletic concept self-sustainable.
A s follows from the a b o v e a n a l y s i s , the typical a n g i o s p e r m c h a r a c t e r s are not
exclusively confined to a n g i o s p e r m s but occur, though less consistently, in the seed
plants that are not formally r e c o g n i z e d as a n g i o s p e r m o u s . O n the o t h e r hand, the
atypical c h a r a c t e r states are often w i d e s p r e a d in n o n - a n g i o s p e r m o u s g r o u p s . The
a n g i o s p e r m concept is based on the m o r e c o m m o n character states, with a tendency
to dismiss the less c o m m o n states as aberrations. H o w e v e r , " a b e r r a n t " characters tend
to o c c u r as typical in geologically older seed plant g r o u p s lending support to their
interpretation as primary (or, if secondary, then arising by reversion to the ancestral
m o r p h o l o g y ) . C o m b i n e d , the aberrant characters may give an idea of a primordial
angiosperm that may be closer to an ancestral form than is the paradigmatic angiosperm.
Such form m a y not b e c o n v e n t i o n a l l y classified as an a n g i o s p e r m , t h u s m a k i n g
a n g i o s p e r m s cladistically paraphyletic, with the typical a n g i o s p e r m c h a r a c t e r s , not
yet occurring in the c o m m o n ancestor, a p p e a r i n g in parallel in the d e s c e n d a n t lines
and, therefore, not uniquely derived.
 However, in the widely held opinion, the uniqueness of a n g i o s p e r m s is due, in the
first place, to their having flowers, stamens and carpels as well as double fertilization,
the " c h a r i s m a t i c " structures and processes wanting, at least nominally, in any other
seed plants and, therefore, exempt from outgroup comparisons. Yet what is flower ex­
cept that it is an organ of flowering plants? W e have to inquire h o w far phylogenetic
concepts are predetermined by semantics, and not only in the case of paradigmatic
angiosperm structures, but also in relation to leaves, pollen and ovules that are nomi­
nally recognized both in angiosperms and non-angiosperms.

Structural homologies
Plant organs b e l o n g i n g to different structural categories are w e a k l y if at all func­
tionally correlated and are d e v e l o p m e n t a l l y controlled by different g e n o m i c d o m a i n s
with insignificant interdomain pleiotropic effects, thus acting as i n d e p e n d e n t evolu­
tionary units. In this chapter, the morphological information on the organ h o m o l o g i e s ,
given in the preceding chapters, is briefly reviewed so as to arrive at sustainable s e m o -
phyletic m o d e l s .

Leaf
Of the various a n g i o s p e r m leaf m o r p h o l o g i e s , the t w o contrasting types, i.e. the
dorsiventrally flattened bifacial with a several-rank areolate venation and with xylem
adaxial in the vascular bundles, and the bilaterally flattened unifacial with parallel veins
and inverted bundles, are characteristic of dicots and m o n o c o t s respectively (yet the
Smilacaceae, Dioscoreaaceae, Philesiaceae, Taccaceae and a few other m o n o c o t fami­
lies, mostly climbers, have the dicot-type leaves: Conover, 1982). T h e latter type is
more c a u l o m i c , although the c a u l o m i c features are displayed also in the c o m p o u n d
leaves with leaflets developing as outgrowths of the axis (instead of developing as lobes
by the localized marginal meristem activities, see Periasamy & M u r u g a n a t h a n , 1985,
on Tahehuia\ Merril, 1986, on Fraxinus, etc.), as well as in the stipules that are linked
with leaves and epiphyllous appendages in a morphogenetic c o n t i n u u m (Sattler & Maier,
1977; Rutishauser & Sattler, 1986). Such features support the "partial shoot" (Arber,
1950), or the c a u l o m e - p h y l l o m e leaf concept (Howard, 1974; Sattler, 1974b).
In the early seed plants, leaves are initially c a u l o m i c , that is, s y n t e l o m i c . In the
Carboniferous g y m n o s p e r m s , the little modified syntelomic systems co-occurred with
differentiated shoots bearing bifacial leaves. If descended from the latter, angiosperm
leaves might be secondarily caulomic. However, even s o m e M e s o z o i c plants seem to
have retained the syntelomic leaf structures. In Dinophyton, a Triassic protognetalean
 p l a n t ( A s h , 1 9 6 8 ; K r a s s i l o v & Ash,
1988), t h e leafy s h o o t s s h o w archaic
syntelome features, such as the cylindri­
cal, occasionally forked or even pinna-
tely branched lateral a p p e n d a g e s ("lea­
ves"), the uniform trichome and stomatal
frequencies over the " l e a v e s " and "axis",
r-x P-^X \/^*TN'^ circinnate ptyxis of the whole
\ / 1^ j \ \ r shoot system.
V I s " ^ ' ^ ! M g e o l o g i c a l l y y o u n g e r gneta-
V . lean p r o a n g i o s p e r m s , the photosynthetic
function w a s transferred from the lat­
eral a p p e n d a g e s , a l m o s t l a c k i n g sto­
mata, to the axis. T h e alternation of large
and small vascular b u n d l e s in m o n o c o t -
t y p e l e a v e s is characteristic also of the
stem i n t e r n o d e s of b o t h QxiSLUt Ephedra
and of the Early CretSiCQOUs Leongathia
elegans ( K r a s s i l o v & al., 1997b) and is,
conceivably evidence of phyllomization.
T h i s m o d e l is further s u p p o r t e d by the
g r a m i n o i d Early C r e t a c e o u s leaves, pre­
sently attributed t o Eoantha or allied
gnetalean proangiosperms (Plate23),
S u c h parallelinerve b l a d e s c o u l d b e for­
Fig. 44. Bennettitalean leaves and bracts bearing on m e d by fusion of linear n o d a l leaves,
phyllodic leaf concept: Ptilophyllum leaf with a with a few of t h e m r e m a i n i n g distinct
swallen petiole (left) and Cycadolepis bracts with as auricles.
reduced blades (after Harris, 1969; Krassilov, 1982).
At the s a m e t i m e , the repeatedly sug­
gested p h y l l o d i c origin of t h e m o n o c o t
leaf t y p e ( D e C a n d o l l e , 1827; H e n s l o w , 1 9 1 1 ; Arber, 1918; K a p l a n , 1970) receives
s o m e support from t h e fossil record, but with a p r o t o t y p e a m o n g b e n n e t t i t e s rather than
dicots. In bennettites, the t e n d e n c y to a p h y l l o d i c transformation i n v o l v e s , in the first
place, their essentially parallelinerve or flabellinerve Cycadolepis-iyp& c a t a p h y l l s (Fig.
4 4 ) occasionally t e r m i n a t i n g in a residual p i n n a t e or p i n n a t i n e r v e b l a d e (Harris, 1969:
Krassilov, 1982b). O n e of the first a p p e a r i n g fossil a n g i o s p e r m - l i k e leaves, ''Dico-
tylophyllum " pusillum (Fig.9) from the Early C r e t a c e o u s of T r a n s b a i k a l i a (Vakhrameev
& K o t o v a , 1977; Krassilov & B u g d a e v a , 1997) r e s e m b l e s a b e n n e t t i t a l e a n bract with a
flabellate venation transformed into an o b l i q u e p i n n a t i n e r v e o n e .
Fig. 45. Bennettitalean stomata: Nilssoniolteris sp., from the Late Jurassic of Bureya Basin, Far East,
Russia, an aborted stoma (central in the top photograph) and normally developed syndetocheilic stomata
(Krassilov, 1978), SEM, x500.
Fig. 46. Bennettitalean stomata: Nilssoniolteris sp., from the Late Jurassic of Bureya Basin, Far East,
Russia, details of a normally developed stoma (top right) and an aborted stoma showing the derivation of
both guard cells and lateral subsidiary cells from a single meristemoid; left photographs show stomatal
patterns in the intecostal zones (Krassilov, 1978), SEM, xl 500 and 250.
 Further bennettitalean foliar features in monocots are vessel elements with scalariform
and p o r o u s perforations found in the leaf veins of the Early C r e t a c e o u s Otozamites
lacustris (Krassilov, 1982b; Fig. 36). Such vessels are characteristic of the graminoid
and palm leaves (Tomlinson & Wilder, 1984). T h e "syndetocheilic" laterocytic (tetracytic
in cataphylls) bennettitalean stomato-type was developmentally m e s o g e n o u s (as eviden­
ced by aborted stomata in fossil leaves: Krassilov, 1978a; Figs. 4 5 , 4 6 ) , thus correspond­
ing, both morphologically and developmentally, to one typical of m o n o c o t s (Ras-mussen,
1983; Tomlinson, 1974; Wilder, 1985).
The dicot-type simple bifacial leaves with reticulate venation appeared in the Paleozoic
glossopterids, at least s o m e of them producing such leaves clustered on deciduous short
shoots. Most Mesozoic seed plants of the temperate zone, in particular the proangio­
spermous czekanowskialeans, had their leaves similarly clustered (with the leaf num­
bers c o n s t a n t in the c z e k a n o w s k i a l e a n s p e c i e s ) , w h e r e a s the c a y t o n i a l e a n foliage
(Sagenopteris) consisted of what a p p e a r as long-petiolate c o m p o u n d leaves of four
leaflets that w e r e occasionally arranged in t w o d e c u s s a t e pairs, betraying derivation
from a short shoot with apically c r o w d e d decussate leaves. T h e leaflets were s o m e ­
what smaller, but nearly identical, in both their b l a d e s h a p e and v e n a t i o n , to the
Glossopteris-type leaves.
Phyllomization of short shoots resulting in the transitional c a u l o m e - p h y l l o m e mor­
phologies is evidenced by such Early Cretaceous forms as Chankanella (Krassilov, 1967)
intermediate between the brachyblasts with planate clusters of Glossopteris-hkQ simple
leaves and the c o m p o u n d Sagenopteris-hke leaves with palmate leaflets. Planation in
the leaf clusters was accompanied by fusion of the leaf (leaflet) blades that evolved in
the direction of the palmately lobed simple leaf blades (Fig. 47). Presently we have a

Fig.47 A suggested transition from the Chankcmella-iype short shoot (left, after Krassilov, 1967) to
palmate leaf.
Fig. 48. Caulomic pinnate-palmate and palmate leaves with dichopodial branching of primary veins in
proangiosperms and angiosperms: (1) Scoresbya, a Jurassic proangiosperm of caytonialean affinities, palma­
te leaf with forking segments, (2) Imania, a closely related Late Triassic leaf genus, pinnate morphotype
with forking lateral segments, (3) Sujfunophyllum, a pinnate-palmate morphotype, supposedly proangio­
spermous, although found in association with Eariy Cretaceous angiosperms, (4) Sagenopteris, a cayto­
nialean leaf, apparently palmate, but with a residual decussate pairing of the leaflets, (5) Proteophyllum,
an angiosperm leaf of Late Cretaceous (Cenomanian) age, (6) Garuya, an extant angiosperm with dicho-
podially branching cotyledons and dissimilar leaves (after Harris, 1935, 1951a; Krassilov & Shorokhova,
1970; Knobloch, 1978; Ruffle, 1980a).
 complete series of transitional forms, such as Imania with partly fused leaflets (Krassilov
& Shorokhova, 1970; Plate 45), Scoresbya (Harris, 1932) with a dichopodial branching
of primary veins that are asymmetrically laminaless (psilobasal) on o n e side, Sujfuno-
phyllum and BaikalophyHum, the geologically younger forms of the s a m e type (Krassilov,
1967; B u g d a e v a , 1983) that co-occurred with the m o r e conventional early angiosperm
leaves (Fig. 4 8 ) . C o m m o n a m o n g the latter w e r e the m o r p h o l o g i e s intermediate be­
tween the c o m p o u n d and the lobed Scoresbya-Vikc, dichopodially s e g m e n t e d leaves
with psilobasal primary veins, as in Proteophyllum, Fontainea, Manihotites, Aralio-
phyllum, Anisophyllum, etc. (Ettingshausen, 1868; L e s q u e r e u x , 1874; Velenovsky, 1889;
Newberry, 1895; Fontaine, 1899; Berry, 1911; Velenovsky & Viniklar, 1926; Knobloch,
1978). T h e y formed a distinctive g r o u p of archaic a n g i o s p e r m s (Krassilov, 1973a;
Knobloch, 1978), their leaf architecture still found in s o m e extant ranunculids and reit­
erated in the cotyledon morphology of s o m e advanced angiosperms, such as Garuya
floribunda (Ruffle, 1980a).
T h e morphological c o n t i n u u m formed by the a b o v e series of fossil links between the
proangiosperm Sagenopteris-Scoresbya and the early angiosperm P r o t e o p h y l l u m leaf
morphotypes may be of a certain phylogenetic significance. A n a s t o m o s i n g of the veins,
variable in the Sagenopteris-Scoresbya group, is similariy developed in Kingdonia, an
archaic extant ranunculid genus. In addition, the anomocytic stomata lacking stomatal
ledges are characteristic of both groups.
The incipient areolate venation appears in gigantopterids and peltasperms in the course
of marginal fusion of the pinnae p r u d u c i n g an entire b l a d e , but with a conservative
venation pattern betraying its c o m p o u n d origin. A p e l t a s p e r m species Scytophyllum
vulgare has polymorphic leaves that show all transitions from pinnate with interstitial
pinnules to lobed and, eventually, entire blades. In the latter, the lateral veins grade
from pinnately branched, marking the formerly distinct pinnules, to fasciculate to fork­
ing. Remarkably, the end products are similar to the initial state: venation of the c o m ­
pound b l a d e c o n v e r g e s on that of its c o n s t i t u e n t p i n n a e , a p r o c e s s d e s i g n a t e d as
retroconvergence and supposedly evidencing a Gestalt-holding capacity of genetic m e ­
mory (Krassilov, 1995). Veins of the marginally fused pinnules approach each other,
but are still separated by the narrow veinless stripes over the fusion meristem (derived
from the marginal meristem that normally does not p r o d u c e vascular tissues). In the
course of further transformation, however, a successive looping of the veins occurs in
the highly meristematic zone at the junction of adjacent pinnae with their interstitial
pinnule, the latter overarched by the series of distal loops (Fig. 4 9 ) . In the advanced
stage, the interstitial pinnule is m a r k e d by a fasciculate vein filling the areole. T h e
resulting pattern looks m u c h as a scalariform tertiary venation of a dicotylid leaf.
Fig. 49. Scytophyllum vulgare (Pryn.) Dobrusk., peltasperm leaves from the Late Triassic of Eastern
Urals (Krassilov, 1995) showing marginal fusion of the pinnae and interstitial pinnules that are marked
by the conservative venation pattern: (1-3) area of fusion incorporating an interstitial pinnule, its lateral
veins approaching (2) or joining (3) those of the adjacent normal pinnules, thus forming a series of
scalariform loops, with the proximal areole filled with short veins arising from the rachis, x3 (3) and 7 ( 1 ,
2): (4, 5) entire pinna formed of fused pinnules still marked by fasciculate lateral veins that are distally
transformed into dichotomous venation; branches of the adjacent veins approach each other without
actually joining, with occasional loops alone, x3 (5) and 7 (4).
 Flower
Familiar by association with weddings and funerals, flowers have escaped a precise
morphological definition. Historically, flower was not considered as a discriminating
feature: Linne said that all plant species had flowers, even if concealed of our eyes
(Linnaei, 1751). Jung, Ray, Tournefort and other XVII century botanists defined flower
in relation to fmit as the precursory of the latter. In their time, any definition of a natural
phenomenon was meant to reflect its essence, and the essence of flower revealed itself
in fruit. The "fructists" based their classification systems solely on fruit characters (re­
viewed in Linnaei, 1751), with g y m n o s p e r m s initially set apart (by Paul H e r m a n n ,
Christoph Knaut and others) as plants having "naked fruits". But, insofar as fruits are
defined in relation to the flower parts (or the whole flowers transformed into anthocarps
or diclesia), the fructist definition is circular. Linne changed it by proclaiming anther
and stigma the essentials of flower. In the thus established tradition, flower was con­
ceived of as an assembly of stamens and/or carpels, with perianth as a less essential,
even dispensable, attribute. However, since the definitions of the stamens and carpels
are in their turn far from precise, it seems better to define flower independently of its
however essential parts.
While we no longer believe that all plants have flowers and since we use the notion
of flower discriminatingly as a special feature of angiosperms separating them from
other plant groups, it seems logical to define this structure in relation to reproductive
stmctures of non-angiospermous seed plants. In the majority of angiosperms, flowers
consist of sporangiophores and/or ovuliphores assembled on the receptacles that may
bear sterile perianth m e m b e r s proximally to the fertile zone. T h e latter develops from
the floral meristem that is similar to the apical meristem of vegetative shoots but is more
active mitotically and is of a less distinct zonal structure. M o r p h o l o g i c a l transitions
between carpels, stamens and petals may reflect a morphogenetic integrity of the floral
meristem rather than their c o m m o n phyllomic origins implied by the classical morphol­
ogy. T h u s , the typical flowers are specialized shoots c o m p a r a b l e to those of Ginkgo
where the sporangiophores and ovuliphores develope apically, surrounded by cataphylls.
These latter shoots are diclinous, but the examples of similar m o n o c l i n o u s structures,
such as Irania, are also provided by the fossil record (see under "Selected fossil links"
above). Their apices are not fully fertile, however, bearing bracts or, in Ginkgo, fully
developed leaves intermingled with ovuliphores.
Bennettites had similarly organized receptacles bearing ovuliphores and the so-called
interseminal scales. T h e floral axis protrudes over the g y n o e c i u m and is crowned by
sterile remnants of the second floral node forming a "corona". In Gnetum, the fertile
organs and their subtending bracts are verticillate in the successive nodal zones, that
can be reduced to two, as in G. scandens (Gifford & Foster, 1974). In Eoantha, the
ovuliphores develope in the proximal fertile node, whereas the second n o d e is reduced
a tuft of linear bracts. In Welwitschia, the functionally polliniferous structures are flower-
Fig. 50. Preflowers, anthostrobili and primitive flowers: (1) an imaginary prototypal monoclinous floral
shoot, based on Irania and Dirhopalostachys, (2) Eoantha, a bracteate gynoecial structure with protrud­
ing floral axis bearing a corona of bracts, (3) Williamsoniella, a monoclinous bennettitalean anthostrobilus,
(4) Liquidambar, an extant pretlower with interfloral phyllomes, (5) Kingdonia, an extant anthostrobilate
flower (Krassilov, 1975; Schweitzer, 1977; Bogle, 1986; Hu & Tian, 1986).
like, weakly z y g o m o r p h o u s , with a cycle of introrse stamen-like sporangiophores. Their
central secretory organ is formed of a solitary ovule c r o w n e d by a peltate protuberance.
Although differing from the typical flowers, these structures c o r r e s p o n d to s o m e
anomalous flowers with bracts in the gynoecial zone and/or with a central sterile col­
umn - a protruding apical extension of the receptacle, or torus (Fig. 50). T h e latter
occurs in the flowers of the Winteraceae, Piperaceae, Lauraceae, Illiciaceae, Euptelea
and s o m e ranunculids (Aquilegia), while in the strobilate m a g n o l i a c e a n flowers the
vascular bundles ascend above the lateral gynoecial organs. Although strobilation may
be either primary, as in Austrohaileya (Endress, 1983), or secondary, as in the M a g ­
noliaceae (Tucker, 1960), the floral structures with sterile apices are anthostrobili rather
than typical flowers.
Sterile phyllomic a p p e n d a g e s in the fertile zone occur in Eupomatia w h e r e they
form an " i n n e r c o r o l l a " b e t w e e n the c a r p e l s a n d s t a m e n s ( t h e i r i n t e r p r e t a t i o n as
staminodes is inconvincing, since stamens tend to be sterilized on the outside of, rather
than inside, the fertile zone). T h e less conspicuous inner p h y l l o m e s constantly occur in
Austrohaileya, Degeneria, Exbucklandia, Rhodolea, Altingia, Liquidambar and s o m e
other genera. T h e y are sometimes secretory, as in Rhodolea, but are mostly functionless.
Their vascular anatomy differs from that of either pistils or stamens (Bogle, 1986).
The internally bracteate flowers might have a pseudanthial origin (Bogle, 1986) or
they could result from an incomplete fertilization of the floral apex, as in bennettites. In
the later case, they are preflowers (Krassilov, 1989, 1991) rather than true flowers. T h e
closest approach to their prototypic form occurs in the extant Cercidiphyllum the pistillate
stmctures of which consist of 2-7 decussate bracteate follicles. T h e m o r e flower-like
preflowers of primitive hamamelids and magnoliids are derivable from a similar struc­
ture, their inner phyllomes corresponding to subtending bracts of ovuliferous organs.
Thus, alongside with the typical flowers, here defined as apical aggregates of ovu­
late and/or staminate structures, there are at least t w o types of flower-like structures, the
anthstrobili, or subapical aggregates of fertile organs c r o w n e d by a sterile extension of
the floral axis, and the preflowers, or fertile organs intermixed with sterile phyllomes in
the floral zone, both considered as transitional between the flowering shoots of g y m n o ­
sperms and those of angiosperms.

Stamen
The antiquity of laminar stamens postulated by the p h y l l o m e theory has n o support
in the fossil record. T h e y do not constitute a uniform m o r p h o t y p e but are instead di­
verse morphologically, with the anthers abaxial, adaxial or lateral, superficial or e m ­
bedded, and with one (Michelia), two coalescent (Austrohaileya) or three vascular traces
derived from both axial and cortical bundles (Degeneria) and, therefore, only superfi­
cially resembling leaf traces, or arising from petaline traces, as in Nelumbo (Moseley &
Uhi, 1985) or stelar, as in Victoria (Heinsbroek & Heel, 1969). T h e y are highly special­
ized (Canright, 1972), in particular, for egg deposition by specific pollinators, such as
Thysanoptera (Endress, 1986a). In addition, the laminar stamens occur in tropical spe­
cies, whereas their relatives in temperate climates have filamentous stamens.
On developmental and vascular evidence, stamens are caulomic rather than phyllomic.
Morpologically, their caulomic origin is betrayed by their occasional branching or fork­
ing, as in Myrica, Ulmus, Carpinus, etc. (Wilson, 1937). Fusion of s t a m e n s in the
Chloranthaceae and M o n i m i a c e a e finds its analogy in gnetaleans (Pearson, 1929).
A caulomic origin is most evident in the case of fasciculate stamens that are often
considered as derived from the more c o m m o n solitary stamens. However, several lines
of evidence suggest reverse phylogenetic relationships. In the first place, the fasciculate
stamens are not rare. They are presently found not only in dilleniid families, but also in
such a phylogenetically remote g r o u p as p a l m s (Uhl, 1976). T h e typical fasciculate
androecia develope from a few primary primordia, each giving rise to 3-10 secondary
primordia. Paired stamens developing from split primordia, as in the Saururaceae (Tucker,
1985), apparently belong in the same category. However, in the ripe androecia, their
fasciculate arrangement is less evident (Pauze & Sattler, 1978). A fasciculate nature of the
apparently individual stamens is sometimes betrayed by dendroid traces, typically devel­
oped in Paeonia. In a less distinct but still recognizable form, such traces occur in the
"primitive" Degeneriaceae, Annonaceae and Papaveraceae, as well as in the Myricaceae
(Stebbins, 1974), or the vascular bundles branch in the filaments, as in palms (Uhl, 1976).
Historically, the seed plant sporangiophores seem to have been fairly conservative,
retaining a transitive c a u l o m e - p h y l l o m e morphology not only in the Paleozoic pteri­
dosperms but also in their derivative peltasperms, corystosperms, nilssonialeans, etc.,
with the laminar fern-like sporophylls occasionally appearing in a few lines, such as the
Permian gigantopterids, but short-lived in terms of geological time and scarcely proto-
typic for the angiosperm stamens.
A m o n g the possible prototypes, Meeusella is a c o m p o u n d androclade of peltaspermid
affinities, with spirally arranged lateral branches that are either simple and sterile or
bearing a subapical pair of stalked peltate sporangiophores, each with a reflexed synangial
head of 2-4 fused sporangia (Plate 4 6 ; Fig. 51). In occasional branches, the stalks are
reduced rendering the sporangial heads sessile on the penultimate axis, the scale-like
expanded apex of the latter protruding over them (Krassilov & B u g d a e v a , 1988b). In
the derivational model based on this structure, a shortening of the axis would result in a
fasciculate arrangement of the lateral fertile branch systems evolving into fasciculate
stamens, the sterile branches giving rise to staminodes. T h e paired ultimate branches
bearing synangial heads are then precursory to the thecae of a typical anther, while the
sterile scale-like apex of penultimate branch corresponds to a p r o d u c e d connective.
T h u s , a conventional filamentous stamen is conceived of as a c o n d e n s e d branching
system, each c o m p o n e n t of the anther developing from a distinct morphological unit. A
probable apotype of Meeusella would have a massive, distinctly lobed latrorse anther
Fig. 51. Meeusella proteiclada Krassil. & Bugd, a supposed prototype of fasciculate stamens from the
Early Cretaceous of Transbaikalia: branching androclade (left) and its paraclades (one sterile) showing a
transition from stalked to sessile staminate heads, each of the latter supposedly giving rise to the theca of
an angiosperm anther (bottom), with the apophysis of the connective conceivably corresponding to the
protruding paracladium apex (after Krassilov & Bugdaeva, 1988b).
Fig. 52. Prototypical staminate structures: (1) Caytonanthus, a branched caytonialean androclade, (2)
''dendroid" androecial traces of Paeonia, with the branching pattern resembling (1), (3) Weltrichia, a
bennettitalean staminate preflower with proximal synangia transformed into nectaries, (4, 5)
sporangiophores of Ixostrobus (czekanowskialeans) and Dinophyton (protognetaleans) with adaxial
sporangia, (6) Austrobeileya, stamen with adaxial pollen sacs, (7) Meeusella, paracladium of a branched
androclade, with stalked sporangial heads, presumably prototypic to bracteate stamens, as in the extant
Trochdendroides (8) or with lateral appendages, as in Monimia (9) (after Harris, 1951a; Krassilov, 1972;
Krassilov & Ash, 1988; Endress, 1986b; Krassilov & Bugdaeva, 1988b).
with a leafy, perhaps peltate connective and a distally branched vascular bundle. Such
stamens can be found in a n u m b e r of the early appearing angiosperm groups, notably
the Chloranthaceae and Platanaceae.
A similar stamen morphology is derivable from the Ixostrobus-iypt sporangiophore
(czekanowskialeans), but with the adaxial 4-lobed anther subtended by a reflexed vas­
cularized connective. Little-modified stamens of this type occur in Austrobaileya and
the A n n o n a c e a e .
Caytonanthus, a caytonialean androclade, is remarkable for its dichopodial "dendroid"
branching r e m i n d i n g of a similar vascular pattern in the androecia of dilleniids and
Paeonia (Fig. 52). This pattern was conceivably inherited from a dichopodial branch
system that b e c a m e incorporated in the receptacle (as another manifestation of vascular
conservatism, see above), its free ultimate branches emerging as indistinctly fasciculate,
perhaps forked, stamens with a slender filament and a relatively massive anther.
In Dinophyton, a Triassic protognetalean plant, the sporangiophores protrude from
bracteate cupules and are laminar, with adaxial sporangia. T h e laminar part is inter­
preted as a bracteole fused to the simple or distally branched stalk with apical sporangia
(Krassilov & Ash, 1988). This structure pertains to the origin of bracteate stamens (con­
ventionally interpreted as 1-staminate flowers), as in Hediosmum (Leroy, 1983), and
the bracteate clusters of such stamens, as in the staminate flowers of Myricaceae.

Pollen
Developmentally, angiosperm pollen grains seem different from those of gymnosperms
in the patterns of primexinal matrix and sporopollenin deposition (Meyer, 1980; Pennell
& Bell, 1986), although intermediate patterns occur in the Annonaceae (Waha, 1987).
Notwithstanding the validity of such distinctions, their applicability is restricted to extant
plants alone. Dispersed pollen grains of early angiosperms are recognized by their reticu­
late surface pattern and columellate infrastructure, the characters of the sporophytic self-
incompatibility syndrome (Walker & Walker, 1984; Zavada, 1984b) that might have ap­
peared independently in different lineages. Incidentally, both the perforate tectum and the
columellate infrastructure appear in Cornetipollis (Vasanthy & Pocock, 1986), one of the
costate morphotypes related to gnetaleans. At the same time, a n u m b e r of angiosperm
families, including the "primitive" Magnoliaceae, Degeneriaceae, Eupomatiaceae, Annona­
ceae, N y m p h a e a c e a e and Myricaceae, have a granular infrastructure that is c o m m o n also
in various n o n - a n g i o s p e r m groups. Moreover, a transitional granular-columellate, or
pseudocolumellate, infratectum occurs both in angiosperms and g y m n o s p e r m s . It has been
described in Ginkgo, some conifers, and Gnetum (Surova & Kavadze, 1988) as well as in
the fossil gnetalean pollen Equisetosporites and ClassopoUis, Such infrastructures are
characteristic of some Late Cretaceous morphotypes {Complexipollis) of the then domi­
nant Normapolles group (Kedves, 1982).
 Although a non-lamellate endexine has been used as a major distinctive feature of
angiosperm pollen morphology (Van C a m p o , 1971; Doyle & al., 1975), its practical use is
complicated by pseudohomologies (Taylor, 1982), with a lamellate foot layer sometimes
described as endexine, as well as by developmental flexibility of the nexinal structures in
general. In angiosperms, primary lamellation often disappears in ripe grains, but some­
times persists in a less strongly condensed apertural endexine or is even discernible in the
interapertural areas, as in Austrohaileya (Zavada, 1985), Degeneria, Michelia fusca and
some Liliaceae. T h e lamellate nexinal layer of gymnosperms is developmentally different
from the endexine of angiosperms (Zavada, 1984b), with a possible exception of Ginkgo
(Rohr, 1977) and some fossil forms. On the other hand, a typical endexine is often lacking
in extant monocots (Zavada, 1983) and "primitive" dicots (e.g. Hedycarya), whereas it
was irregularly developed in some early angiosperm pollen grains (Walker & Walker,
1984) and the Late Cretaceous Normapolles group (Kedves & Pardutz, 1983b).
At least o n e of the diverse apertural types occurring in a n g i o s p e r m pollen grains, the
distally sulcate (anasulcate), characteristic of the M a g n o l i a c e a e and m o n o c o t s , is also
widespread a m o n g g y m n o s p e r m s . Other apertural types pose a m o r e difficult problem,
however. In the early angiosperm pollen, such as Clavatipollenites, the anasulcate mor­
phologies are closely linked to the trichotomosulcate ones, but the s o m e t i m e s postu­
lated transition from the latter to the likewise early a p p e a r i n g tricolpate type is not
supported by any factual evidence. Inaperturate grains as an alternative starting point
for all angiosperm apertural types (Muller, 1970) disagree with the o b v i o u s relatedness
of anasulcate forms in g y m n o s p e r m s and angiosperms.
T h e prevalence of the radially disposed apertures in a n g i o s p e r m s has been related
either to a tetrad configuration (Wodehouse, 1935) or to germination on stigma (Eyde,
1971; Stebbins, 1974). Although multiple radial apertures certainly appeared before
stigmas (e.g. in the M e s o z o i c gnetaleans), the latter, especially in conjunction with
e n t o m o p h i l y , m i g h t h a v e increased the significance of the v o l u m e r e g u l a t i n g (har-
m o m e g a t h y ) structures. In addition, the pollen grain v o l u m e to surface ratio is affected
by the ploidy level (Popova, 1971), typically being high in primitive a n g i o s p e r m s . Har-
m o m e g a t h y might involve germinal apertures and other structures, such as proximal
triradiate scars, sacci and the residual folds of reduced sacci. T h e r e could b e reversals
from the h a r m o m e g a t h y to the apertural functions, with respective morphological trans­
formations conceivably giving rise to the following apertural types:
(1) Bipolar apertures that appeared, in their incipient form, in Classopollis, a wide­
spread M e s o z o i c pollen type produced by the hirmerellacean gnetophytes and showing
both an apertural triradiate scar and a distal leptome.
(2) Z o n o s u l c a t e apertures developing on the basis of m o n o s a c c a t e structures, as,
incipiently again, in Classopollis, the equatorial girdle and its bordering subequatorial
groove (rimula) of which might have been derived from a reduced annulate monosaccus
(Krassilov & al., 1997a). Similar zonosulcate structures occur in m o n o c o t s .
 (3) Additional apertures d e v e l o p i n g from the lateral sacci. In g y m n o s p e r m s , the
saccate and asaccate pollen m o r p h o t y p e s diverged already at the level of Paleozoic
pteridosperms, but saccate structures were lost in several M e s o z o i c groups, a tendency
extended to angiosperms. Residual sacci are still occasionally discernible, as an atavis­
tic feature, in the pollen grains of cycads and ginkgo (Shashi & U e n o , 1986). A m o n g
the Cretaceous pollen grains, a reduction of sacci can be traced from their vestigial
forms, as in Baculivesiculites or Microcahridites, to h a r m o m e g a t h y folds, as in Verm-
mono-colpites (Zavada, 1984a), and to additional sulcate apertures, as in Eucommiidites
(Fig. 53). A remarkable similarity between these g y m n o s p e r m pollen morphologies and
the early Tricolpites may suggest a transformation of the h a r m o m e g a t h y structures into
the transequatorial tricolpate apertural type.
(4) Polycolpate forms arising from polyplicate structures, as in Welwitschia and the
fossil forms of the Equisetosporites type in which the intercostal pseudocolpi are ro­
tated transequatorially in one of the hemispheres (Vasanthy & Pocock, 1986).
(5) Tricolporate structures derived from the tricolpate endocingulate. T h e possibility is
prompted by Sarbaya, a mid-Cretaceous angiosperm (Krassilov & Shilin, 1995), whose
tricolporate pollen grains show lalongate endoapertures forming a continuous weak zone.

Fig. 53. Pollen grains illustrating a reduction of sacci as a pathway to additional apertures: (1) Micro-
cachridites antarcticus, trisaccate grain, (2) Trichotomosulcites subgranulatus, trichotomosulcate grains,
(4) Baculivesiculites inchoatus, showing folds of reduced sacci, (5) Eucommiidites, the subequatorial slits
corresponding to reduced sacci, (6) Tricolpites sagas, tricolpate grain with polar folds (after Hughes,
1976; Zavada, 1984a; Dettmann, 1986).
Fig. 54. A suggested origin of triapreturate pollen grains from tetrads: permanent tetrads of the extant
Lactoris (1) and Cretaceous Walkeripollis (4) compared with Normapolles-iype grains (2, 3) that some­
times show a tetrahedral outline of the nexine (after Medus, 1983; Walker & al., 1983; Kedves & Hegedus,
1983; Zavada & Taylor, 1986).
or endocingulum, conceivably inherited from the Classopollis-iype equatorial girdle. T h e
ores might have arisen at intersections of the colpi with such equatorial structures.
(6) Radially aperturate structures of tetrasporic origin (Krassilov, 1989, etc.). Wal­
keripollis, Early Cretaceous permanent tetrads with rimulate pores (Walker & al., 1983)
resemble those of ClassopoUis, the distal pore (or leptome) of which is encircled by a
supposedly monosaccus-derived girdle. In the former, however, the individual grains
are c a l y m m a t e , as in the Lactoridaceae, the only extant angiosperm family having saccate
pollen grains with the reduced sacci encircling the apertures ( Z a v a d a & Taylor, 1986b).
A further integration of such tetrads and a shift of meiosis o n e step back could have
produced tetrasporic pollen grains that b e c a m e trilobed triaperturate by the loss of a
fourth m e m b e r in symmetrical readjustment of a basically tetrahedral structure (Fig.
54), the latter still evident in the tetrahedral nexine configuration of s o m e Cretaceous
triporate pollen grains of the Normapolles type ( M e d u s , 1983). T h e graminid pollen
grains with rimulate pores and the cryptotetrads characteristic of the C y p e r a c e a e might
derive from the Walkeripollis prototype.

Gynoecium
A n g i o s p e r m g y n o e c i a c o n s i s t of o v u l i f e r o u s o r g a n s , o r c a r p e l s , t h a t f o r m , indi­
vidually o r j o i n t l y , the c l o s e d o r s e m i - c l o s e d pistillate s t r u c t u r e s . A s in the c a s e of
flowers, c a r p e l s h a v e b e e n defined m o r p h o l o g i c a l l y in n o o t h e r w a y a s , circularly,
the structural u n i t s of a n g i o s p e r m g y n o e c i a . T h a t c a r p e l s a r e v a r i o u s l y m o d i f i e d
fertile l e a v e s , o r m e g a s p o r o p h y l l s , is n o t a m o r p h o l o g i c a l o b s e r v a t i o n b u t is an a s ­
sumption of the n o w o b s o l e t e p h y l l o m e ( U r p f l a n z e ) t h e o r y ( s e e a b o v e ) . T h e foliar
interpretation of c a r p e l s b e c a m e a central d o g m a of t y p o l o g i c a l m o r p h o l o g y , the
debates t u r n i n g a r o u n d the p r i m o r d i a l t y p e of leaf a n d t h e m o d e of its f o l d i n g . N e i ­
ther a n a t o m i c a l n o r d e v e l o p m e n t a l s t u d i e s h a v e c o n f i r m e d t h e foliar i n t e r p r e t a t i o n ,
however. E v e n the p a r a d i g m a t i c r a n u n c u l i d follicle d e v e l o p e s of an a n n u l a r p r i -
m o r d i u m ( R o h w e d e r , 1967) that is n e i t h e r folded n o r s h o e - s h a p e d as r e q u i r e d by,
respectively, the c o n d u p l i c a t e and the i n v o l u t e foliar m o d e l s , w h i l e t h e s u p p o s e d
" m a r g i n s " are, in fact, ribs e x t e n d i n g o v e r the v a s c u l a r b u n d l e s . In t h e typical "foliar"
m a g n o l i a c e a n c a r p e l w i t h t h r e e t r a c e s , the dorsal t r a c e arises from an axial b u n d l e ,
while t h e t w o ventral o n e s d e p a r t from the cortical b u n d l e s , a n d t h e o v u l e s are s u p ­
plied w i t h an interstitial vein a s c e n d i n g b e t w e e n t h e d o r s a l a n d v e n t r a l b u n d l e s
(Skipworth, 1 9 7 0 ; T u c k e r , 1975). In t h e s u p p o s e d l y e v e n m o r e " p r i m i t i v e " carpel
of Drimys, t h e t r a c e s are in t w o pairs ( T u c k e r , 1975) s u g g e s t i n g a b i v a l v e d p s e u d o -
mono-merous structure.
Although carpels are conceived of as, primarily, the ovule-bearing structures, in a
number of a n g i o s p e r m s the ovules are either borne laterally on the floral axis, as in
lllicium floridanum or Ochna atropurpurea (Pauze & Sattler, 1979), or are developed
from the apex of the floral shoot, as in Myrica gale (Sattler, 1974a), Piperaceae (Kanta,
1962; Tucker, 1986), Juglandaceae and cyperid m o n o c o t s (Cohen, 1970; Macdonald,
1977), and are separate from the carpels that are formed either of the lateral primordia
as the ovule-subtending structures or of the annular primordia encircling the floral apex.
In addition, the vascular supplies of the carpels and ovules in such gynoecia are discon­
nected, c o m i n g from different axial bundles. S o m e of such "acarpellate" gynoecia can
be secondary (Philipson, 1985), but the assertion that they all are secondary lacks suffi­
cient morphological evidence.
A foliar nature of carpels was once thought of to have been confirmed by the discov­
ery of pteridosperms with their leaf-like ovuliphores. Presently w e k n o w that, initially,
ovules were borne in non-foliar cupules formed of syntelomic three-dimensional branch­
ing structures. T h e bilateral cupules might have resulted from syncupuly, i.e. fusion of
two or m o r e radial cupules (Long, 1986). T h e leafy ovuliphores have appeared much
later as a result of the gynoclade phyllomization. T h e y w e r e lost from the geological
record at least 100 million years before the appearance of angiosperms, thus rendering
the phylogenetic continuity unrealistic.
A m o n g the morphological and chronological predecessors of angiosperms, Leptostrobus
had bivalved cupules with the marginal stigmatic crests intruding the locule as inner stig­
mas and overhanging the anatropous ovules that were borne on submarginal placentae. A
minimal morphological distance separates these structures from the pseudomonomerous
pistils of the Winteraceae and perhaps other magnolialean "follicles" (Fig. 55).
The Caytonia-iypo, cupule has been chosen by T h o m a s (1931) as prototypic for the
ranunculid-type follicle. T h e ovules (reinterpreted in Krassilov, 1977b) are basal, ana­
tropous, bitegmic, connected by presumably funicular tubes to a slit-like e x o s t o m e . The
latter is flanged by the free tip of subtending bract adnate to the cupule as a broad "lip".
A minor elaboration is needed to transform this structure into an "ascidiform" carpel
with the exostomal slit corresponding to the plicate region of the latter and the "lip"
developing into a broad sessile stigma, as in Paeonia and its allied C r e t a c e o u s Hyrcantha
(Krassilov & al., 1983).
Another Mesozoic proangiosperm, Dirhopalostachys, provides an example of a one-
seeded beaked cupule comparable to the follicles of Kingdonia, an isolated extant genus
remotely related to the Ranunculaceae (Hu & Tian, 1985). Superficially at least, the paired
Dirhopalostachys cupules are similar to the likewise paired follicles of Trochodendrocarpus,
a widespread Cretaceous to Paleogene fructification type (Krassilov, 1976, etc.).
At the s a m e time, the secondary cupules of M e s o z o i c bennettites and gnetaleans
were formed of interseminal scales or bracteoles connate around the o v u l e that was
lateral or apical on the floral axis. A m o n g them, the Triassic Dinophyton produced a 4-
lobed bracteate cupule that, in the early developmental stages, appeared as an annular
collar with incipient lobing, later expanding to enclose a solitary orthotropous ovule.
T h e lobes bore tufts of long hairs probably assisting in pollination (Plate 7; Fig. 10).
Fig. 55. Prototypic carpellate structures: (1) Leptostrobus, bivalved cupules with submarginal anatropous
ovules and stigmatic marginal fringes, compared to (2) Drimys, extant, pseudomonomerous bicarpellate
"follicles", (3, 4) Caytonia, gynoclade and cupule with the exostomal "lip" resembling leafy stigmas of
the extant Paeonia (5), (6) Dinophyton, 4-lobed bracteate cupule with a solitary orthotropous ovule,
comparable with (7) Myrica, extant, pistillate flower with bracteoles, (8) Dirhopalostachys, beaked cupule
with a solitary anatropous ovule, similar to (9) follicle of Kingdonia, extant (after Harris, 1951a; Foster,
1961; Krassilov, 1972, 1975, 1984; Cronquist, 1981; Krassilov & Ash, 1988).
 T h e Early Cretaceous Baisia illustrates a s o m e w h a t different cupule morphology,
utriculate, enclosing a solitary ovule that is basal on the receptacle that is covered with
bristles developing from the perianth bracts. T h e cupule appears to have been formed of
connate interseminal scales, their tips joined into a terminal corona. T h e contemporane­
ous Eoantha had a 4-valved gynoecium, each valve with a basal ovule and with the
floral apex protruding and terminating in a tuft of sterile bracts. T h e s e cupular struc­
tures relates to the 3-4-carpellate unilocular ovaries with basal ovules occurring in the
Piperaceae, M y r i c a c e a e and glumiferous m o n o c o t s . Transformations of the prototypic
4-lobed structures into the 3 - or 2-lobed ones by the loss of a lobe or a pair of lobes
occur in both the extant gnetaleans (Pearson, 1929) and the piperacean angiosperms,
the latter having 3-4-carpellate gynoecia (Tucker, 1976, 1982).
M o s t of the early angiosperms had sessile stigmas characteristic of the forms with
sporophytic self-incompatibility and morphologically traceable to the exostomal struc-

Fig. 56. Moditlcations of the second tloral node in (1) Eoantha, cupulate pretlower with a corona of
bracts, (2) Welwitschia, extant, staminate pretlower with a central peltate glandular structure, (3, 4)
Williamsonia, gynoecium with a corona of vistigial interseminal scales, (5) Baisia, cupule with an apical
corona, and presumably (6) Piperomia, extant, pistil with a bushy stigma (after Harris, 1969; Krassilov,
1972, 1986; Cronquist, 1981; Krassilov & Bugdaeva, 1982).
Fig. 57. A suggested derivation of free styles from stamens prompted by the filament to style similarity in
Sarysua, a Cretaceous angiosperm (Krassilov & al., 1973).

tures found in the M e s o z o i c proangiosperms, such as the " l i p " in Caytonia, the mar­
ginal papillate crests in Leptostrodus, the hairy cupule lobes in Dinophyton, the termi­
nal corona of scale tips in Baisia, and the supposedly h o m o l o g o u s tufts of sterile scales
crowning the floral axis in Eoantha (Fig. 56). T h e s e structures are morphologically
comparable with the foliar, decurrent, lobed, and bushy stigmas respectively.
The styles might have appeared in parallel in different angiosperm lineages in rela­
tion to gametophytic self-incompatibility and with no plausible prototypes in proan­
giosperms. However, in the Late Cretaceous Sarysua, the styles were remarkably simi­
lar to the filaments both of stamens and staminodia (Krassilov & al., 1983). Evidence of
a conjoined developmental control is provided by e x a m p l e s of the similarly branched
styles and stamens in the Euphorbiaceae, a correlation of the stylar and pollen polymor­
phisms in heterostylic species, and the occasional reversals of stigmas to anthers (Rao,
1935). T h e s e observations raise a possibility of styles originating from staiminodes
adnate to the ovary (Fig. 57). T h u s , in c o m p o u n d ovaries, the n u m b e r of styles m a y not
correspond to the n u m b e r of carpels.

Ovule and dispersal

T h e ovule is perhaps the only reproductive organ indisputably shared by g y m n o ­
sperms and a n g i o s p e r m s , thereby substantiating a morphological unity of all seed plants.
Presently there is little doubt that ovules evolved from sporangia borne not on leaves,
but rather in syntelomic branching systems. Lenlogia with paired sporangia surrounded
by the whorled sterile appendages may illustrate one of the possible prototypes (Krassilov
& Zakharova, 1995), with the nucellus arising of a pair of fused sporangia and with the
integument, or t w o successive and nearly h o m o m o r p h o u s integuments, formed simul­
taneously of the whorls of indusial a p p e n d a g e s .
In primitive g y m n o s p e r m s , the integuments are the apically open, lobed structures
serving for protection and perhaps for trapping air-borne pollen grains that settled on
top of the nucellus. Subsequently, the nucellus has developed receptive structures that
emerged over the integumental lobes. T h e cupule might protect a y o u n g ovule and was
overgrown by the integument at maturity or, alternatively, it could o v e r g r o w the integu­
ment and overtake its functions. In the latter case, the integument w a s set upon the track
of reduction and of the proximal or even wholesale fusion with the nucellus, perhaps
contributing to the apical structures of the latter (see a b o v e u n d e r pteridosperms).
Both ovules and seeds may serve as dispersal units. T h e former m o d e is considered
primitive and is s o m e t i m e s described as " p r e p h a n e r o g a m o u s " . H o w e v e r , its chrono­
logical occurrence, as well as its correlation with other phylogenetically important char­
acters are inconsistent in both g y m n o s p e r m s and a n g i o s p e r m s , s o m e of the latter also
" p r e p h a n e r o g a m o u s " . Notably, in s o m e M e s o z o i c g i n k g o a l e a n s , attached seeds con­
tained the m o r e advanced e m b r y o s than in the living g i n k g o (Krassilov, 1972b). Early
g y m n o s p e r m s have dispersed both cupulate and excupulate ovules.
A functional c h a n g e occurred with the a p p e a r a n c e , in m o r e than o n e lineage, of
micropyles as integumental, or also cupular, pollen receptive structures. This change
marked a transition from the archaic to the advanced g y m n o s p e r m s w h o s e nucelli lost
their apical receptive structures (or retained their rudiments functioning as glands). In
the advanced g y m n o s p e r m s , cone scales and even the entire seed cones serve as disper­
sal units or, less frequently, the seeds are furnished with arillate structures. Transition
from the radial to the bilateral symmetry, also parallel in different lineages, seems to
have been related to the cone scale dispersal, while the secondary seed dispersal might
restore the radial symmetry.
A n g i o s p e r m y has led to the even m o r e radical functional c h a n g e s . T h e integuments
have lost their pollen receptive and partly also protective functions, at the s a m e time
acquiring a pollen t u b e - c o n d u c t i n g function. T h e nucelli h a v e lost their nutritional
" e n d o s p e r m " function that has been passed over to the e m b r y o n i c e n d o s p e r m . Further­
more, in angiosperms, the periovular structures are consistently involved in dispersal,
forming various fruits.
Fruit dispersal may seem a decisive factor of seed evolution in a n g i o s p e r m s . How­
ever, morphologically as well as functionally, the fruit is a dispersal unit formed of
periovular structures of s o m e kind(s), such as ovaries, perianth bracts, w h o l e inflores­
cences, etc., enclosing seeds (critical c o m m e n t s on alternative definitions see above
under "flower"). In this sense, the samaras of Hirmerella or Welwitschia are likewise
fruits, the production of which being scarcely restricted to a n g i o s p e r m s . In particular,
such dispersal units are characteristic of proangiosperms that produced samaroid cupules
with persistent bracts, as in Dinophyton or Hirmerella\ fleshy cupules with m a n y seeds,
as in Caytonia, perhaps endozoochorous; dry achene-like cupules, as \nDirhopalostachys',
utriculate bristled cupules, as in Baisia', closed preflowers with persistent fleshy recep­
tacles, as in s o m e bennettites, etc. T h e dispersal m e c h a n i s m s presently restricted to
angiosperms had first appeared in proangiosperms, e. g., the parachute of coronal bris­
tles in Problematospermum, or the cotton-grass-like p a p p u s of Baisia (Plates 5, 6; Fig.
43) conceivably assisting in a n e m o c h o r o u s or epizoochorous dissemination, or both.
Yet it is difficult to decide how far the functional changes were responsible for the
morphological distinctions. In angiosperms, seeds are m o r e diverse morphologically,
especially in respect to dispersal m e c h a n i s m s , their symmetry (radial versus bilateral)
is more variable, s o m e t i m e s even within a species, and the lower seed size limit is
considerably lower than in g y m n o s p e r m s . T h e latter distinction, as well as the neotenic
gametophytes, seem to be related to developmental acceleration in the early angiosperms
in turn associated with a pioneer ecology and, alternatively or additionally, with epizoo­
chorous dispersal m o d e . Actually, the hooked appendages in the ceratophyllacean achenes
from K o o n w a r r a ( D o u g l a s , 1969; D i l c h e r & al., 1996) or t h e s p i n y c a p s u l e s of
Tyrmocarpus (Krassilov, 1977a, 1989) point to epizoochory as an early appearing dis­
persal m o d e in angiosperms.
However, the prevalence in angiosperms of anatropous bitegmic ovules is less a m e ­
nable to a functional explanation. On the ground of their occurrence in s o m e allegedly
primitive angiosperms, these features are often considered as primary for the group.
Then the orthotropous type should be secondary, whereas an unitegmic condition might
arise by fusion as well as reduction either of the outer or, m o r e often, of the inner
integument that could be developmentally retarded or shifted to the apex by endochalazal
growth ( B o u m a n , 1984; B o u m a n & Calis, 1977). T h e s e processes are inferred on the
basis of c o m p a r a t i v e m o r p h o l o g y . T h u s , in the Liliaceae h a v i n g both b i t e g m i c and
unitegmic ovules, the latter apparently lost their m n e r integument in fusion with a c o m ­
paratively massive nucellus (Sterling, 1973). Alternatively, priority is given to ortho­
tropous unitegmic ovules as m o r e c o m m o n in g y m n o s p e r m s (e.g. M e e u s e & B o u m a n ,
1974), with the anatropous condition considered as derived in relation to s i p h o n o g a m o u s
pollination m o d e (Takaso & B o u m a n , 1984). T h e bitegmic condition might then arise
by acquiring either the outer integument of an arilloid origin or the inner integument of
a nucellar origin, the latter functioning as a pollen tube-conducting structure ( D e Boer
& B o u m a n , 1972; Heslop-Harrison & al., 1985), or, in orchids, as a germination-medi­
ating structure (Lucke, 1984). Any generalizations are risky, however, in the light of the
great variability of the ovular characters in angiosperms.
Incidentally, the anatropous condition can arise either by a reversion of the funicle or
by a developmental shift of the raphe toward the micropyle, or s o m e t i m e s by inversion
after fertilization, or even by reversion of the carpel (Vijayaraghavan, 1964). Both in­
teguments can be of a dermal origin or they arise from n o n - h o m o l o g o u s - dermal and
subdermal - initials, as in most of the ranalean angiosperms. In 75 families, both integu­
ments contribute to the micropyle (sometimes zigzagged, with the outer, or exostome,
not in line with the inner). T h e micropyle is formed either of the inner or of the outer
integument alone in 38 and 4 families respectively, while in 7 9 families this character
varies (Tilton, 1980). Vascular bundles either terminate in the chalaza or extend to one
or, rarely, both integuments, as in the Euphorbiaceae. T h e cutinization of the integu­
ments and the development of stomata (in 25 families: Rugenstein & Lersten, 1981) are
likewise variable. In the Capparaceae, the outer integument occasionally encloses two
ovules. At least s o m e of these variations are paralleled by g y m n o s p e r m ovules.
As is convincingly shown in the early seed plants, the orthotropous ovules are pri­
mary in s p e r m o p h y t e s , while t w o or m o r e h o m o m o r p h o u s or h e t e r o m o r p h o u s inte­
gumental structures can arise simultaneously, the outer forming a c u p u l e that contains
one or m o r e ovules. T h e cupules enveloping solitary ovules can have their exostomes
not in line with the i n t e g u m e n t a l m i c r o p y l e s , as in t h e z i g z a g g e d m i c r o p y l e s of
angiosperms. In pteridosperms, the inner integument, or integument proper, tends to be
lost as a distinct morphological structure in fusion with the nucellus, resulting in the
massive vascularized nucelli, as in trigonocarps and cycads. However, the bitegmic (or
occasionally tritegmic) condition was retained in the gnetaleans and s o m e bennettitaleans
that advanced to the proangiospermous level with orthotropous ovules, as in Baisia,
Eoantha and Gnetum. In the latter genus, the cupule, or outer integument, occasionally
contains m o r e than one ovule (Keng, 1965) and the integuments d e v e l o p in acropetal
succession, the inner and the middle ones from the dermal, and the outer one from both
the d e r m a l and s u b d e r m a l initials. T h e y are v a s c u l a r i z e d , a n d t h e o u t e r o n e s are
stomatiferous, while the inner one is archaically lobed. Furthermore, the middle integu­
ment is developmentally retarded, whereas the inner integument is shifted to the apex
by endochalazal growth (Takaso & B o u m a n , 1986), the tendencies that can be traced to
many angiosperms.
In Baisia, a derived representative of the Early C r e t a c e o u s bennettites (Krassilov
& B u g d a e v a , 1982), the c u p u l e was utriculate and c o m p a r a b l e with the utricles of the
Late C r e t a c e o u s lemnoid m o n o c o t s (Krassilov & M a k u l b e k o v , 1995). B o t h the baisian
cupules and the lemnoid utricles enclosed a solitary o r t h o t r o p o u s o v u l e arising from
the bristled receptacle in the former, and with a funicle in line with the fruit pedicel in
the latter (Plates 4 7 , 4 8 ) . T h e a p p e a r a n c e of a n a t r o p o u s o v u l e s in several g y m n o ­
sperm lineages, such as c o r y s t o s p e r m s , nilssonialeans and conifers, correlates with
strobilation and anemophily, and is p e r h a p s e x p l i c a b l e by a e r o d y n a m i c s of the air­
borne pollen grains whirled around the c o n e axis (destrobilation is a c c o m p a n i e d by
r e a p p e a r a n c e of a secondarily orthotropous - p s e u d o o r t h o t r o p o u s - o v u l e polarity, as
in Ginkgo, Cycas and s o m e p o d o c a r p s ) . At the s a m e t i m e , the a n a t r o p o u s condition
correlates with dispersal of the seed-bearing c o n e scales (in conifers) or c u p u l e s (in
c z e k a n o w s k i a l e a n s and caytonias) that is precursory to fruit dispersal in angiosperms.
Furthermore, while the b i t e g m i c condition tends to b e lost in the e x p o s e d o v u l e s d e ­
veloping thick coats, both i n t e g u m e n t s tend to be retained as distinct, although deli­
cate, coats in the enclosed o v u l e s .
At least Caytonia has a d v a n c e d to the p r o a n g i o s p e r m o u s level with a n a t r o p o u s
bitegmic ovules (Krassilov, 1984a) that are strikingly similar to those of a mid-Creta­
ceous ranunculoid angiosperm, Caspiocarpus (Vakhrameev & Krassilov, 1979; Krassilov,
1984a) in general shape and dimensions, as well as in s h o w i n g a m e d i a n raphe trace,
and in having the integuments a few cells thick, of which the inner o n e w e d g e s out
slightly a b o v e the midlength.
T h e historically deep-rooted divergence of the orthotropous and anatropous ovule
morphotypes in p r o a n g i o s p e r m s may thus b e phylogenetically significant, for these types
are traceable to different groups of early angiosperms, such as the piperoid - juglandioid
and the ranunculoid lines respectively.

Double fertilization and the embryo

D o u b l e fertilization is often considered as a major, if not the only, diagnostic feature
of angiosperms. It has to be kept in mind, however, that a n g i o s p e r m s had been well-
established as a higher taxon long before the p h e n o m e n o n of d o u b l e fertilization was
discovered (by Navashin in 1898) and asserted by a representative s a m p l e of angiosperms
from different families that revealed a considerable variability of the process (see above).
In spite of that and of s o m e very atypical f o r m s of d o u b l e fertilization, as in the
Onagraceae, w h e r e the term is scarcely justified at all, n o n e of the k n o w n angiosperms
has yet been excluded from the group on the pretext of having no d o u b l e fertilization.
For extinct plants, there is no direct e v i d e n c e of fertilization, but, insofar as the
double fertilization associates with accelerated g a m e t o p h y t e d e v e l o p m e n t , an indirect
evidence may c o m e from the size of the ovules. Although this character is primarily
under ecological control (see above), with the early successional plants tending to have
more n u m e r o u s and smaller seeds, in g y m n o s p e r m s there is a lower seed size limit of
about 1 m m , apparently set in by the g a m e t o p h y t e d e v e l o p m e n t . In a n g i o s p e r m s , an
accelerated g a m e t o p h y t e development, with archegonia reduced to the clusters of nu­
clei, have opened a perspective of seed miniaturization d o w n to dust size, as in orchids.
Actually, the advent of angiosperms in the mid-Cretaceous is m a r k e d by the appearance
of fossil seeds well under 1 m m (Hall, 1963; Colin, 1973; B i n d a & N a m b u d i r i , 1983;
Knobloch, 1983, etc.), perhaps suggesting the a n g i o s p e r m o u s level of female g a m e t o ­
phyte r e d u c t i o n . R e m a r k a b l y , e v e n such p r o a n g i o s p e r m s as t h e c a y t o n i a l e a n s a n d
czekanowskialeans have produced smaller seeds than did the bulk of g y m n o s p e r m s .
Double fertilization is a c o m p l e x morphological as well as physiological p h e n o m ­
enon, the c o m p o n e n t s of which occur in non-angiosperm seed-plants, too. P o l y e m b r y o n y
of various kinds is widespread a m o n g g y m n o s p e r m s . Likewise, the sperm-bearing pol-
len t u b e s and the i n v o l v e m e n t of both s p e r m s in fertilization are n o t restricted to
a n g i o s p e r m s , but occur, systematically, in conifers and, sporadically, in other non-
angiospermous groups. Such steps on the way to double fertilization as (1) the appear­
ance of archegonial c o m p l e x e s and the ensuing diminution of individual archegonia, as
in Ephedra and Taxaceae, and their complete reduction in the end m e m b e r s of the not
closely related gnetalean lineages, (2) the functioning of both sperms, as in the Taxaceae,
Cephalotaxaceae, Ephedra, etc., and (3) the sexualization of the e m b r y o sac nuclei,
appearing, again independently, in Gnetum, Welwitschia and a n g i o s p e r m s , are parallel
developments taking place in different seed plant divisions.
Actually c o m m o n for all properly studied angiosperms is not so the highly variable
double fertilization process as the lack of morphologically distinct archegonia, but this
feature is shared with at least t w o extant n o n - a n g i o s p e r m o u s g e n e r a , Gnetum and
Welwitschia. T h e e m b r y o sac nuclear homology between these genera and the typical
angiosperms is difficult to decipher, however, for, typically, the t w o former have much
more n u m e r o u s nuclei, an indefinite n u m b e r of which are sexualized and can poten­
tially function as eggs, while syngamy frequently occurs both before and after fertiliza­
tion (Maheshwari & Vasil, 1 9 6 1 ; Martens, 1971). Yet in the third extant gnetophyte
genus Ephedra, the archegonia are morphologically distinct, j o i n e d in the archegonial
c o m p l e x e s and are conceivably on their way to fusion into a multinuclear structure.
Characteristically, in angiosperms the sperms are discharged in a synergid correspond­
ing to an archegonial neck cell in Ephedra. In the latter, one of the sperms reaches to
the ventral canal cell, the other to the egg that sinks deeper in the venter (Herzfeld,
1922; Khan, 1940; Moussel, 1978). A mass of cells is produced of the fertilized ventral
canal cell (Pearson, 1929). On the other hand, the nuclear configuration in a typical 8-
nucleate e m b r y o sac is suggestive of two oppositely oriented archegonia w h o s e poten­
tial neck cells differentiate as synergids and antipodal cells respectively. A potential
ventral canal cell of the micropylar c o m p l e x forms the egg, while the potential eggs
form the polar nuclei that, like archegonial eggs, sink in the c o m m o n venter and meet
there, still presenting a target for the second sperm.
In other w o r d s , what w e conventionally call the e g g cell in a n g i o s p e r m s can be
h o m o l o g o u s to the archegonial ventral canal cell, while the archegonial e g g s corre­
spond to the polar nuclei (Cocucci, 1973; Krassilov, 1989, 1991). D o u b l e fertilizations
in Ephedra and in typical angiosperms can thus prove to be developmentally homolo­
gous (Fig. 58).
At the s a m e time, the m o s t c o m m o n free-nuclear e n d o s p e r m d e v e l o p m e n t in angio­
s p e r m s is like a typical e m b r y o d e v e l o p m e n t in g y m n o s p e r m s , and e v e n the less com­
m o n type of an initially cellular and then free-nuclear e n d o s p e r m d e v e l o p m e n t may
have its parallel in Ginkgo (see a b o v e ) . In contrast, the e m b r y o n i c d e v e l o p m e n t in
a n g i o s p e r m s , with a sole exception of Paeonia, is cellular, while in g y m n o s p e r m s it is
free-nuclear, the polyploid Sequoia being a single a n o m a l y . T h u s , to all appearances,
Fig. 58. Possible homologies of angiosperm egg and polar nuclei (left) to the ventral canal cell and egg,
respectively, of a gymnosperm archegonium.

the e m b r y o of g y m n o s p e r m s is h o m o l o g o u s to the e n d o s p e r m , rather than to the e m ­

bryo, of a n g i o s p e r m s .
As for the intragroup and outgroup homologies of embryological structures, it has to
be reminded that the idea of cotyledons being not, or not always, h o m o l o g o u s in monocots
and dicots, first put forth by Arber (1925), is presently confirmed in a n u m b e r of case
studies. Dicotyledonous e m b r y o s are c o m m o n in several groups of g y m n o s p e r m s , s o m e
of them having h e t e r o m o r p h o u s cotyledons and s h o w i n g a tendency of cotyledon fu­
sion, the tendencies extending both to dicotylids and m o n o c o t y l i d s (e.g. Piperomia,
with the cotyledons heteromorphic, and one of t h e m haustorial). In s o m e m o n o c o t s , the
cotyledon b e c o m e s p s e u d o t e r m i n a l by r e p l a c i n g the p r i m a r y a p e x (e.g. in p a l m s :
Guignard, 1984), as is occasionally the case in dicots as well (Claytonia). However, in
a n u m b e r of m o n o c o t genera, c o t y l e d o n s p r o p e r s e e m totally lacking ( G e r a s i m o v a -
Navashina, 1972).
T h e grass e m b r y o alone has been e x e m p t from h o m o l o g i z a t i o n by a n o n c o m m i t t a l
descriptive n o m e n c l a t u r e of its structures. T h e a t t e m p t s at h o m o l o g i z a t i o n of, e.g.,
the grass s c u t e l l u m with h y p o c o t y l or fused c o t y l e d o n s or their parts or p r o p h y l l and
other structures of a c o n v e n t i o n a l e m b r y o (see N e g b i & Koller, 1 9 6 2 ; C o c u c c i &
Asregiano, 1978; Tsvelev, 1987) h a v e not been entirely successful yet. A c t u a l h o m o ­
logies m a y b e found outside the g r o u p ( M e s t r e & G u e d e s , 1983), p e r h a p s involving
such haustorial structures as the feeder of the gnetalean e m b r y o s .
 Trend types
T h e principles of phylogenetic analysis applied a b o v e to g y m n o s p e r m s are valid for
angiosperms as well, but g y m n o s p e r m s differ from their precursory progymnosperms
in the presence of ovule, a single charismatic character. In contrast, angiosperms are
defined by a n u m b e r of loosely c o r r e l a t e d c h a r a c t e r s that often f o r m intersecting
morphoclines.
In the following analysis, phylogenetic relations are deduced on the basis of multi­
ple h o m o l o g i e s , that is, homologies of different, independently evolving, organ sys­
tems. In plants, lacking a distinct h e r m line, evolution has been, to a considerable ex­
tent, directed by functional adaptations. Although a growth form may affect pollination
and seed dispersal, the functional correlation of the vegetative and reproductive sys­
tems is too weak for impelling their coordinated evolution. Both the vegetative and
reproductive shoot systems are integrated developmentally by m o r p h o g e n i c fields that
occasionally inflict a morphological convergence of their constituent structures, as in
the petaloid or carpelloid stamens, but developmental correlation of the vegetative and
reproductive organs does not preclude their independent evolution u n d e r different se­
lection pressures. W e have to distinguish between the general selection pressures, such
as environmental instability, inducing whole plant trends, such as d e v e l o p m e n t a l accel­
eration, and the specific selection pressures, such as a pollen vector efficiency, inducing
morphological trends in the respective structures alone. Both trend types can be "fash­
ionable" for a particular geological time interval or they can be long-lasting, traceable
through geological times and across taxonomic boundaries of successively appearing
plant groups. T h e g a m e t o p h y t e reduction trend belongs to the latter category (see above
under "gametophytic characters"), while phyllomization of leafy shoots s e e m s to have
been "fashionable" a m o n g M e s o z o i c proangiosperms, resulting in the highly flexible
c a u l o m e - p h y l l o m e systems characteristic of angiosperms as an adaptive type. Such sin­
gle character (semophyletic) trends are, by themselves, insufficient for substantiating
phylogenetic relationships that can be m o r e convincingly demonstrated by the concur­
rent trends involving different organ systems.
Since trend recognition is crucial for our analysis, the following terminology is pro­
posed for different trend types:
- Morphocline, or semophyletic trend: pertaining to a particular organ or organ system;
- C o r p o r a l , or w h o l e plant, trend: manifested in the evolution of different organ
systems, such as developmental acceleration;
- Concurrent trend: involving functionally independent organ systems;
- Chronocline: chronological succession of character states;
- Time-fashion trend: a trend involving concomitant plant lineages;
- Transgrade trend: a trend extending from one morphological grade to another across
a conventional grade boundary.
 T h e sequential stages of phylogenetic analysis then will be:
- Recognition of the corporal and semophyletic trends;
- Selecting the chronologically relevant time-fashion trends;
- Singling out the continual transgrade chronoclines;
- E n s e m b l i n g the concurrent transgrade chronoclanes as a material for phylogenetic
reconstructions.
A practical realization of this ideal scheme runs into certain difficulties related both
to morphocline and chronocline ambiguities. Trends deduced from the data of c o m ­
p a r a t i v e m o r p h o l o g y a l o n e m a y h a v e little p h y l o g e n e t i c s i g n i f i c a n c e , w h e r e a s
chronoclines are, for most part, fragmentary. In effect, phylogenetic reconstructions are
hypothetical to be verified or falsified by the m o r e convincing h o m o l o g i e s , as well as
by the m o r e c o m p l e t e transgrade chronoclines.

Proangiosperms as a morphological pool

T h e p r o a n g i o s p e r m g r a d e as defined a b o v e b e c a m e a p p a r e n t d u e to a series of
palaeobotanical discoveries (Krassilov, 1969, 1973b, 1975, 1977b, 1 9 8 2 a , b ; 1 9 8 4 , 1 9 8 6 ;
Krassilov & Bugdaeva, 1982, 1997; Schweitzer, 1977) that brought to light the seed
plants chronologically preceding or c o n t e m p o r a n e o u s with the earliest angiosperms and
showing character states typical of the latter, but lacking in s o m e critical characters
upon which the conventional recognition of angiosperms is based. A m o n g the typical
angiosperm features, the following appeared in one or m o r e p r o a n g i o s p e r m groups as
described in the preceding chapters:
- Vessels of the axial and/or foliar vascular systems in gnetophytes and bennettites;
- Paracytic stomata in bennettites;
- Graminoid leaf morphology in gnetophytes;
- C o m p o u n d - p a l m a t e to l o b a t e - p a l m a t e l e a v e s with r e t i c u l a t e v e n a t i o n in t h e
Sagenopteris-Scoresbya group (the Caytoniales and allies);
- " D e n d r o i d " androclades, anticipating fasciculate androecia, in Caytonanthus;
- Pollen grains with a protocolumellar infrastructure and with zonal, as well as p o ­
rous, protoapertures in Classopollis (the hirmerellean gnetophytes);
- C u p u l a r g y n o e c i a in Dirhopalostachys of a c y c a d o p h y t e d e s c e n t (with a soli­
tary a n a t r o p o u s o v u l e ) and Baisia of a bennettitalean d e s c e n t (with a solitary orthotro­
pous o v u l e ) ;
- Ascidiform cupules with many ovules in Caytonia',
- F o u r - m e m b e r e d c u p u l a t e g y n o e c i a with a b r a c t e a t e p e r i a n t h in Eoantha (the
gnetophytes);
- Syncupulate capsules with stigmatic crests in Leptostrobus (the Czekanowskiales), etc.
 Different proangiosperm groups are thus c o m p l e m e n t a r y in forming a morphologi­
cal pool of incipient angiosperm characters. Moreover, in the C r e t a c e o u s representa­
tives of the principle proangiosperm groups, the reproductive structures s h o w the ten­
dencies of condensation and fusion (as, e. g. in the pollen organs of Basianthus, see
a b o v e ) , with the c h i m e r i c o r g a n s d e r i v e d from the fusion m e r i s t e m s h a v i n g great
potentials for morphological innovation (e. g. in the case of areolate venation, see above
under " L e a f ) . T h e early angiosperms seem to have been likewise c o n d e n s e d and highly
meristematic plants.
But n o single proangiosperm lineage might have conceivably given rise to the basic
angiosperm diversity evidenced by the mid-Cretaceous fossil record. R a t h e r the diver­
sity of angiosperm morphologies could have arisen by a recombination of the diversity
of proangiosperm morphologies, involving both vertical and horizontal transmissions
of genetic material. T h e latter pathway is assumed for certain palynological characters
(see above) but is probably of a m o r e general significance for parallel evolution (Syvanen,
1994). T h e angiosperms as a whole might thus have evolved from the proangiosperms
as a whole, although s o m e particular lines of descent m a y still be discernible across the
p r o a n g i o s p e r m - a n g i o s p e r m grade boundary.

Concurrent homologies
A semophyletic trend is significant for the whole plant p h y l o g e n y if it is supported
by other semophyletic trends pointing in the s a m e direction. So, as the next step of our
analysis, w e shall select the concurrent h o m o l o g i e s of the functionally independent
s t r u c t u r e s . O n t h e b a s i s of the r e c e p t a c u l a r o r t h o t r o p o u s v e r s u s t h e a p p e n d i c u l a r
anatropous ovule m o r p h o l o g i e s , both p r o a n g i o s p e r m s and early a n g i o s p e r m s can be
arranged in t w o series:

Receptacular orthotropous Appendicular anatropous

Bennettites Dirhopalostachians
Gnetaleans Czekanowskialeans
Caytonialeans
Ceratophyllids Ranunculids
Platanoids Hamamelids
Piperoids Magnoliides
Myricoids Rosoids
Graminoids Alismatoids

If these series have a phylogenetic meaning, then a n u m b e r of the features shared by

their juxtaposed m e m b e r s are homoplasies. This has to be verified by the concurrent
homologies of the phyllodic versus caulomic leaves, the sulcus-derived versus sac-de­
rived pollen apertures, the bract (bracteole)-derived versus ovuliphore (seed-scale)-de-
rived cupules, etc.

The gneto-bennettite line

B o t h g n e t o p h y t e s and b e n n e t t i t e s h a v e b e e n u s e d as o u t g r o u p s in a n g i o s p e r m
p h y l o g e n y (Wettstein, 1907; A r b e r & P a r k i n , 1907). H o w e v e r , until recently, the
gnetaleans had no fossil record, w h e r e a s the a n g i o s p e r m o i d features of their extant
representatives, particularly those related to g a m e t o p h y t e reduction, could be ascribed
to parallel d e v e l o p m e n t s . T h e bennettite - angiosperm c o m p a r i s o n s have involved, in
the first place, the strobiloid flowers that, in both groups, are related to cantharophily
and, therefore, homoplastic. N e w evidence has c o m e from the recently discovered fos­
sil gnetophytes, starting with Eoantha that showed transitional angiospermoid gynoecial
(Krassilov, 1986) and foliar (Krassilov & Bugdaeva, 1997) structures. It was soon fol­
lowed by other forms of diverse vegetative and reproductive morphologies. At the same
time, angiospermoid characters have been found in a few derived bennettitalean forms,
such as Problematospermum and Baisia (Krassilov, 1973b; Krassilov & Bugdaeva, 1982).
Summarily, these findings have provided a factual basis for analyzing the gneto-bennettite
to angiosperm concurrent homologies.

Gnetophyte to monocot trends

The bifacial leaves with a several-rank areolate venation, the vessel elements with
scalariform perforation plates ( M u h a m m a d & Sattler, 1982), the sieve elements c o m ­
panion-like cells (Pearson, 1929), and the fertilization pattern anticipating double ferti­
lization (Herfeld, 1922; K e n g , 1940; M o u s s e l , 1978) are the features shared by the
extant gnetalean genera with angiosperms rather than with each other. However, these
genera are relicts of a previously much m o r e diverse group.
The Early Cretaceous gnetophytes, appearing together with the earliest angiosperm
records, were small, sedge- or grass-like plants evidencing the miniaturization and con­
densation corporal trends in their m o r e robust early M e s o z o i c precursors. T h e g a m e t o ­
phyte r e d u c t i o n s l e a d i n g to the a r c h e g o n i a - l e s s e m b r y o s a c s , as in Gnetum and
Welwitschia, were conceivably associated with an accelerated sporophyte development.
Both trends are continued in angiosperms.
The dicotylid-type leaves, as in Gnetum, are lacking a m o n g the p r e - a n g i o s p e r m
records, therefore considered as parallel d e v e l o p m e n t s , p e r h a p s a p p e a r i n g earlier in
angiosperms than in gnetophytes. Fossil gnetophytes show diverse foliar morphologies,
from the poorly differentiated caulome-phyllome systems, as in Dinophyton, to the re-
duced sheathing leaves in the Leongathia-Ephedra line, their stomatiferous stems tak­
ing u p the photosynthetic function. A similar pattern of alternate large and small vascu­
lar bundles in the shoot intemodes of these gnetophytes and in graminoid leaves sug­
gests a semophyletic relatedness, with the morphologically graminoid caulomic leaves
of Eoantha plant as a fossil link. T h e Late Cretaceous graminoid leaves assigned to
Dammarophyllum (Shilin, 1986) are associated with Taldysaja, a g r a m i n o i d floral struc­
ture (Krassilov & al., 1983).
T h e floral structures of gnetophytes are typically decussate or w h o r l e d at two or
more bracteate nodes of spicate inflorescences. In Eoantha, they are solitary pedicellate
bracteate preflowers corresponding to a single floral n o d e of a typical gnetalean spike,
with the sterile second node reduced to a tuft of bracts on a short rachylla. Protruding
rachyllae occur also in Taldysaja and extant graminoid florets. T h e granular to irregular
rugulate-columellate pollen grain infrastructures are nearly identical in the gnetophytes
and extant graminoids (Zavada, 1983), while the p o r o u s apertures of the latter have
their p r o t o t y p e in the M e s o z o i c Classopollis (the h i r m e r e l l a c e a n g n e t o p h y t e s ) , the
rimulate pollen grains with a distal cryptopore. S o m e other apertural types might have
been derived from the saccate or plicate exinal structures of extinct gnetophytes.
Developmentally, the gynoecia of graminoid m o n o c o t s are similar to the gnetalean
cupules arising by tubular growth from the annular primordia (Takaso & B o u m a n , 1986).
In Eoantha, a fossil intermediate form, the g y n o e c i u m consisted of a four-lobed cupule
bearing basal orthotropous ovules. Tetramerous gynoecia occur as a plesiomorphic fea­
ture in the extant graminoids, while the m o r e c o m m o n di- to trimerous forms might
have been derived by the loss of one or t w o m e m b e r s , a widespread m o d e of morpho­
logical transformation. In addition, there are embryological features, not available in
fossils, compliant with a phylogenetic proximity of gnetophytes and graminoids, such
as the tetrasporic e m b r y o sac, coleorhiza, sheathing cotyledons, and haustorial append­
ages of the e m b r y o n i c axis - the feeder in gnetophytes and scutellum in grasses.

Bennettite to monocot trends

C o m m o n in bennettites were phyllodic cataphylls with parallelodromous or flabellate
venation, occasionally terminating in a vestigial pinnate or entire pinnatinerve blade
(Harris, 1969; Krassilov, 1982b). T h e y suggest a leaf blade reduction trend that contin­
ued in the phyllodic leaves of monocotylids (Arber, 1918), with intermediate morpho­
logies represented by the Baisian leaves or cataphylls (Fig. 9) with oblique, irregularly
c l a d o d r o m o u s lateral veins (Krassilov & B u g d a e v a , 1997). A m e s o g e n o u s develop­
mental pattern is confirmed for the "syndetocheilic" bennettitalean stomata on the evi­
d e n c e of their aborted initials (Krassilov, 1978a; Figs 4 5 , 4 6 ) . In the bennettitalean
cataphylls, the stomata w e r e typically tetracytic, as in the graminoid leaves.
Vessels in the leaf veins o c c u r in a C r e t a c e o u s b e n n e t t i t e Otozamites lacustris
(Krassilov, 1982b; Fig. 36) and are characteristic of graminoids and p a l m s .
 Typical bennettitalean flowers resemble those of magnoliacean angiosperms in the
relatively massive receptacles with an apically protruding corona (a vestige of the sec­
ond floral node), the spiral arrangement of p o l y m e r o u s floral parts, and the dimorphic -
stomatiferous sepaloid and stomata-less petaloid - perianth bracts (Harris, 1969). H o w ­
ever, these similarities are related to cantharophilous adaptations. In the advanced ben­
nettitalean forms, there was a tendency to aggregation of preflowers alongside with
oligomerization of the floral structures. Baisia, a numerically d o m i n a n t genus of the
Baisian proangiosperm assemblage, is conceived of as an end-product of these func­
tionally correlated trends. Its gynoecial preflowers produced a solitary erect ovule apical
on the receptacle and enclosed in the utriculate cupule (Plates 5,6), the latter conceiv­
ably formed of connate interseminal scales. T h e s e structures were dispersed as bristled
disseminules with persistent receptacles (Krassilov & B u g d a e v a , 1982). Their similar­
ity to cyperacean achenes may be m o r e than superficial, for in both the bristles arise
from perianth bracts, whereas the gynoecial structures are essentially cupular, develop­
ing from the floral apices.

Gnetophyte-bennettite features in other angiosperm groups

A m o n g the groups formally assigned to dicotylids, but of obvious monocotylid alli­
ances, the Piperales shares with the gnetophytes a decussate arrangement of lateral or­
gans, the tracheary elements with both scalariform of m a n y bars and p o r o u s perfora­
tions, orthotropous ovules arising form the floral apex, gynoecial structures developing
from annular primordia encircling the apex (in Piperomia: Tucker, 1986), and the hetero­
morphous cotyledons (in Welwitschia).
Antiquity of the piperoid line is evidenced by the earliest dispersed angiosperm pol­
len morphotype Clavatipollenites at least partly assignable to the Chloranthales, an or­
der of the w i d e l y d i v e r g e n t a r c h a i c g e n e r a r e c e n t l y s e p a r a t e d f r o m the P i p e r a l e s
( D a h l g r e n , 1 9 8 3 ) . A p e c u l i a r a n d r o e c i a l f e a t u r e of b o t h e x t a n t a n d C r e t a c e o u s
chloranthaleans is fusion of the stamens forming synandria with different n u m b e r s of
sporangia on the median and lateral lobes. Fusion of ovuliphores occurs in Gnetum as
well as in the Cretaceous gnetalean plant Baisianthus (Krassilov & B u g d a e v a , 1997;
Plates 16-19). Moreover, there is a similar variation of sporangial n u m b e r s in the cen­
tral and lateral synangia.
Worth considering in this respect are two other groups of archaic angiosperms, the
Platanaceae, a monotypic family usually placed in hamamelids, and the Nymphaeales, a
heterogeneous aquatic group of ambiguous dicot/monocot affinities. Platanoids differ from
hamamelids primarily in the orthotropous ovule morphology shared with the gnetoid line
and, a m o n g nymphaeids, with Barclaya and Ceratophyllum, the latter genus ranking a m o n g
the earliest angiosperms, with a precursory in the Early Cretaceous (Aptian) Koonwarra
Assemblage (Dilcher & al., 1996). Notable for further morphological comparisons are
such nymphaealean features as the stigmatic disk developing from the floral apex and
apparently homologous to the "corona" of bennettitalean preflowers or the apogynoecial
bracts of Eoantha. or other modifications of the second floral node (see above, Fig. 56). In
the pollen grains of Nymphaea, the equatorial harmomegathy zone and the rim of the
operculum (Kupriyanova, 1976) topologically correspond, respectively, to the girdle and
the rimula of ClassopoUis, a Mesozoic gnetalean pollen morphotype.
In the Myricales, the staminate "flowers" are small strobili of bracteolate sporangio­
phores that are occasionally fused and are adnate to their supporting bracts, resembling
both the Mesozoic Dinophyton and extant Ephedra. T h e pollen wall infrastructure is
granular to shortly columellate (Zavada & Dilcher, 1986), as in Gnetum, the endexine is
lacking (in the phylogenetically close Juglandaceae it is developed as a uniform layer
scarcely thickened under the apertures, with similar endexinal structures found in mo­
nocots: Z a v a d a , 1983). T h e carpellate " f l o w e r s " are provided with t w o lateral brac­
teoles, as in Welwitschia. T h e g y n o e c i u m is essentially cupular, 2 (3-4)-carpellate, sur­
rounding a solitary ovule developed of the floral apex.
T h e fossil fruits Gurvanella (Krassilov, 1982b) and Ramonicarya (Krassilov &
Dobruskina, 1995) provide evidence of an early (Aptian) appearance of myricoids. They
resemble also the samaras of Welwitschia and of the extinct hirmerellean gnetophytes,
but are as yet insufficiently studied for being considered as semophyletic links.

Caytonialean to ranunculid trends

Phylogenetic links with angiosperms and, in particular, with ranunculids have been
postulated for caytonialeans by their discoverer already ( T h o m a s , 1931), but they were
rejected by the leading plant morphologists of the time in favour of, in the words of
Agness Arber (1950), the elegant simplicity of the classical carpel concept. It must be
recalled that ranunculid follicles were then conceived of as the paradigmatic involute
sporophylls. They are not considered as such a n y m o r e (see Rohweder, 1967).
Yet, with a few exceptions (Stebbins, 1974), the caytonialeans are not discussed as
plausible angiosperm ancestors. Their morphological revision undertaken by Harris (1951
and elsewhere) tended to discredit the earlier angiosperm c o m p a r i s o n s , while, on the
other hand, the leading system-makers, with a notable exception of Hutchinson (1959),
tended to place the ranunculid orders in the M a g n o l i i d a e and to consider t h e m as de­
scending from the magnolialean plexus rather than as a stem g r o u p (e.g. Takhtajan,
1966; Cronquist, 1981). In the present a u t h o r ' s opinion, the t a x o n o m i c distances be­
tween the ranunculids, magnoliids and h a m a m e l i d s are artificially diminished by un­
warranted assignments of s o m e ranunculid taxa to magnoliids (in a n g i o s p e r m systemat-
ics, such misassignments often create a false impression of morphological continuity,
see above). T h e magnoliids have thus been turned into a collection of archaic forms,
with any non-magnoliid group derivable from them either directly, or with a few in-
 termediaries. M o s t fossil forms have fitted in the magnoliids in the a b o v e sense, thus
strenghtening their interpretation as a stem group.
My studies of the Yorkshire material and the new caytonialean records in the Far
East ( K r a s s i l o v , 1977b, 1984a), as well as the d i s c o v e r y of t h e early C r e t a c e o u s
ranunculids (Vakhrameev & Krassilov, 1979; Krassilov & al., 1983; Krassilov, 1984a)
seem to support a phylogenetic relatedness of caytonialeans to ranunculids, with the
following concurrent trends:
(1) T h e typical caytonialean leaves (Sagenopteris) are c o m p o u n d - p a l m a t e of four
lanceolate reticulinerve leaflets that resemble simple leaves of the Paleozoic Glossopteris
and are occasionally arranged in decussate pairs betraying their origin from a short
shoot with a terminal leaf cluster, as in glossopterids. Even m o r e caulomic are some
satellite genera, such as Chankanella (Krassilov, 1967), with similar, but m o r e numer­
ous, apically clustered leaves or leaflets (Figs. 4 7 , 4 8 ) . Concurrent to phyllomization of
the short shoots, there w a s a leaf blade fusion trend that led, through a series of interme­
diate forms, to the Scoresbya-iypo, palmately lobed leaves with the primary veins a s y m m e ­
trically laminaless on the outside (Harris, 1935). T h e s e peculiar leaf m o r p h o l o g i e s are
traced to the mid-Cretaceous transitional c o m p o u n d to palmately lobed Vitiphyllum-
Cissites group, with similar leaf morphotypes occurring in Trollius, Aconitum, Anemone
and other extant ranunculacean genera. Such leaves were found attached to the shoot
axis of Caspiocarpus, an early ranunculid genus (Vakhrameev & Krassilov, 1979), while
the pinnate-temate leaves ''Leguminosites'' karatcheensis with Sagenopteris-Wkc lan­
ceolate leaflets and an irregular oblique c l a d o d r o m o u s venation c a m e in association
with Hyrcantha, a paeonid fructification (Krassilov & al., 1983).
(2) A caytonialean (Sagenopteris) leaf venation is typically reticulate with frequent
anastomoses. In the satellite genera Scoresbya, Imania, Chankanella and Sujfunophyllum,
the anastomoses are sparse or even lacking. This venation type can be traced through
the mid-Cretaceous Diplophyllum (Knobloch, 1978) to the archaic extant ranunculids
Kingdonia and Circaeaster.
(3) T h e caytonialean stomata are peculiar a m o n g the M e s o z o i c g y m n o s p e r m stoma-
totypes in the leveling of the guard cells with the unspecialized subsidiary cells, and the
lack of stomatal ledges. Prof. T. Harris, generally critical of any links between cayto­
nialeans and a n g i o s p e r m s , has admitted (Harris, 1951 and personal c o m m u n i c a t i o n ,
1975) that the caytonialean stomatal structures are definitely a n g i o s p e r m - l i k e . Such
stomata are characteristic, a m o n g angiosperms, of the ranunculid c o m p l e x including
the Cretaceous ''Leguminosites" karatcheensis (see above) and Menispermites (Up­
church, 1984; Krassilov & al., 1988).
(4) T h e dichopodial ("dendroid") branching of a caytonialean androclade Cayto­
nanthus is analogous to the branching pattern of the " d e n d r o i d " androecial traces in
Paeonia, an archaic g e n u s linked to ranunculids by the m i d - C r e t a c e o u s Hyrcantha
(Krassilov & al., 1983). Such traces are characteristic of fasciculate androecia that might
have been derived from the Caytonanthus-iype branching system, with individual sta­
mens corresponding to the terminal branchlets of the latter.
(5) T h e caytonialean pollen grains, proto-bisaccate to m o n o s a c c a t e with laterally
connected bladders (Krassilov, 1977b; Z a v a d a & Crepet, 1986) m a y seem quite differ­
ent from any apertural types occurring in ranunculids. However, a reduction trend in­
volving several M e s o z o i c seed plant lineages might have led to a transformation of
protosaccate structures into h a r m o m e g a t h y folds and, further, into transequatorial colpi
(see above, u n d e r "Pollen").
(6) T h e caytonialean cupules are basically ascidiform, with a plicate e x o s t o m e and a
transcurrent fertile zone, thus corresponding to the typical ranunculid carpel (Rohweder,
1967). Broad sessile stigmas, as in Paeonia and its precursory mid-CvQtaceous Hyrcantha,
could have been derived from a morphologically similar e x o s t o m e a p p e n d a g e ("lip") of
Caytonia, in turn derivable from a reduced subtending bract (Fig. 16). Further develop­
m e n t s in the direction of m o d e m ranunculids are e v i d e n c e d by the mid-Cretaceous
Caspiocarpus having shortly beaked dorsiventricidal follicles, as in extant Glaucidium.
An elongation of the stylar region might have been conceivably induced by a transition
from the pollination drop m e c h a n i s m to extraovular pollen germination that concur­
rently led to reduction of sacci as a pollination drop-related feature.
(7) Both in the caytonialeans and ranunculids, the ovules are anatropous, ranging
from the basal to basiventral, bitegmic with thin integuments. T h e r e w a s a size reduc­
tion from about 1 m m in the Jurassic Caytonia to about 0.5 m m in the Cretaceous
Caspiocarpus, but otherwise their ovules were quite similar, platyspermic, with a short
micropyle formed by the outer integument that was t w o to several cells thick and thinly
cutinized, while the inner integument was 2-3 cells thick at the chalaza, distally wedg­
ing out a b o v e the midlength of the ovule (Plates 2 8 , 36). T h e nucelli w e r e likewise
similar, relatively m a s s i v e , shortly b e a k e d , with a m e d i a n line reflecting the raphe
(Krassilov, 1984a). On the other hand, the ovule morphologies found in Caspiocarpus
resemble, a m o n g extant ranunculids, Circaeaster in the extremely thin integuments,
and Aquilegia, Hydrastis, etc. in the outer integument forming the micropyle.

Other proangiosperm contributions

A s suggested a b o v e , the strobiloid flowers with interfloral p h y l l o m e s , as in the
p r i m i t i v e h a m a m e l i d s (Liquidambar) and magnoliids (Degeneriaceae, Himan-
tandraceae, E u p o m a t i a c e a e , and A u s t r o b a i l e y a c e a e ) , are derivable from the short shoots
bearing terminal clusters of a n d r o c l a d e s and/or g y n o c l a d e s , s u b t e n d e d by, or inter­
mingled with, scaly cataphylls. S u c h shoots o c c u r in Irania, as well as in the czeka­
n o w s k i a l e a n and d i r h o p a l o s t a c h y a c e a n p r o a n g i o s p e r m s (Plate 4 ; F i g . 7 ) . Furthermore,
a p r o t o t y p e of the w i n t e r a c e a n p s e u d o m o n o m e r o u s follicles with sutural stigmatic
crests (Vink, 1978) can be found in the bivalved cupules of Leptostrobus stigmatoideus,
an advanced species of czekanowskialean gynoclades, in which the sutural papillate
crests intrude the locule as the "inner s t i g m a s " that onlap the sutural ovules (Krassilov,
1969, 1972a). At the s a m e time, the lateral ovuliphores of Ixostrobus, a czekanow­
skialean androclade, are c o m p a r a b l e with certain stamen m o r p h o t y p e s of the extant
magnoliids, such as Austrobaileya, on account of the adaxial anthers and the scaly ster­
ile apophyses (Krassilov, 1969, 1972a). In addition, an androclade origin of the stamens
in magnoliids (in the restricted sense, discussed above) was suggested on the basis of
their vascular anatomy.
Pairing of cupules is characteristic of the M e s o z o i c Dirhopalostachyaceae (Krassilov,
1975a). Their gynoclades are strikingly similar to the Late Cretaceous and Paleocene
hamamelid fructifications Trochodendrocarpus (Krassilov, 1976 and elsewhere) in having
a loosely racemose structure, as well as in the external morphology of the paired cupules
and follicles, both showing the ventral sutural ridges, reflexed beaks and an oblique
lateral striation (Plates 4, 44). However, a wide chronological discontinuity m a k e s di­
rect phylogenetic links between these two forms unlikely.
Such similarities may indicate a certain degree of phylogenetic relatedness or, at
least a genetic contribution of the abovementioned proangiosperm groups to the respec­
tive angiosperm stalk lines. C o n v e r g e n c e is a poor explanation in the case of characters
of an obscure adaptive m e a n i n g . Alternatively, such similarities might result from a
spurious donation of genetic material by horizontal transfers from one proangiosperm
group to another within a proangiosperm assemblage that formed a c o m m u n a l source of
angiosperm origins.
 Concluding remarks
A n g i o s p e r m origins are here traced from the Devonian, passing through a series of
g y m n o s p e r m grades that show evolutionary trends and patterns m o r e fully expressed at
the angiosperm level. Several g y m n o s p e r m lineages had acquired certain angiosperm
features, thus collectively giving rise to the proangiosperm grade.
D u e to the discovery of proangiosperms, the advent of a n g i o s p e r m s m a y seem less
enigmatic than it appeared in the Darwinian time. This does not mean that w e know a
direct ancestor(s) or the precise place and time, but these may not be strictly definable
at all. W h a t we d o k n o w is a c o m m o n source of angiosperm characters, a morphological

Fig. 5 9 . Suggested phylogenetic relations of the major proangiosperm and angiosperm groups.
pool containing these characters in their incipient form. A few morphological trends are
recognized as linking the proangiosperm and angiosperm levels. G a p s remain, but at
least we k n o w where to look for them.
In deciphering the angiosperm to proangiosperm homologies, it has been necessary
to clarify our concepts of the flower, stamen, carpel, double fertilization and other sup­
posedly typical, but as yet vaguely defined, angiosperm traits. Their origins are shown
to have been due primarily to developmental acceleration and functional integration of
the respective prototypic structures found in p r o a n g i o s p e r m s . T h e s e p r o c e s s e s have
determined the concurrent trends in the evolution of separate morphological domains.
The above-suggested phylogenetic relationships (Fig.59) are based on such concurrent
trends, with s o m e recent finds substantiating the gneto-bennettite to graminid and the
caytonialean to ranunculid affinities, the other p r o a n g i o s p e r m g r o u p s considered as
possible contributors of genetic material.
Notably, a n g i o s p e r m s appeared in the a s s e m b l a g e s containing various p r o a n g i o ­
sperms. Hence, their origins s e e m to have been the product of a c o m m u n a l breakthrough
rather than a single event. Evolution of the cradle c o m m u n i t i e s m a y shed s o m e light on
the angiospermization process occurring over the whole^their ranges, with multiple di­
versification centres. Angiospermization involved a n u m b e r of lineages trending in the
same direction. W h e t h e r the angiospermy was achieved by a single or several of the
parallel lineages, is a matter of secondary importance, although both the present-day
angiosperm m o r p h o l o g y and the early angiosperm diversity seem to suggest m o r e than
one successful contender. A n g i o s p e r m validity as a class may depend not so on a solu­
tion of this d i l e m m a as on their structural integrity that is based on the shared evolution­
ary trends and the ensuing structures of a major adaptive significance. Characters ap­
pearing in a single lineage are often lost in the course of subsequent evolution, but
parallelism ensures their consistent d e v e l o p m e n t in a large g r o u p of organisms. T h e
angiosperm story is thus of s o m e general interest for understanding the nature and the
origins of higher taxa.
 Acknowledgements
I a m grateful to m y former collaborates in the P a l e o b o t a n i c a l L a b o r a t o r y of the
Institute of Biology and Pedology, Vladivostok and to m y present collaborates in the
Paleontological Institute, M o s c o w . M y special thanks are d u e to E u g e n i a B u g d a e v a and
Natalia M a s l o v a w h o contributed some valuable material and information, to m y wife
Sophia for the e n c o u r a g e m e n t and advice, to L u d m i l a Volkova for technical help and to
photographers Claudia N o v i k o v a and A l e x a n d e r M a s i n . I also a c k n o w l e d g e the lin­
guistic corrections by the editor Dr. Sergei Golovatch. This work w a s financially sup­
ported by the Russian Foundation of Basic Research, grant N o . 9 5 - 0 4 - 1 1 8 6 3 and the
International Science Foundation, grant N o . N 6 V 0 0 0 . T h e publication costs have been
met by Pensoft Publishers and the H. Rausing Foundation.
 Plate 1
Cycandra profusa Krassil. & Delle, a giant cycadalean pollen cone from the Jurassic of Georgia, Cauca­
sus, showing medullosalean synangial structures; (I) pollen cone (the numbers refer to detached sporophylls
on the slab), xO.7, (2) sporangium, the distal part of the endothecial membrane with a beak filled with
pollen grains, x240, (3) pollen grains, x500, (4) sectioned pollen grain showing sulcus (arrow), TEM,
x5 000.
 Plate 2
Cycandra profusa Krassil.& Delle. a giant cycadalean pollen cone from the Jurassic of Georgia. Cauca­
sus, showing medullosalean synangial structures: (1) heads of sporangia as seen on the rock surface.
vSEM, x70, (2) microsporophyll with a sterile fringe and a central synangium of numerous sporangia in
concentric rings, x7. (3) files of vertical, marginally adnate sporangia, the synangial cuticle with meshes
of sporangial heads seen at the left margin, SEM. x9().
 Plate 3
Semionogyna hracleata Krassil. & Bugd., a proto-cycadean ovulate spike from the Early Cretaceous of
Transbaikalia (Krassilov & Bugdaeva, 1988c): (1) spike with immature ovules, distal bracts empty, x l . 5 ,
(2) .spike with ripe ovules, x l . 5 , (3, 4) ovules showing the outlines of a nucellus, xlO, (5) bract supporting
a juvenile ovule, xlO, (6) pollen organ of presumably the same plant. xlO.
 Plate 4
Dirhopalostachys mstrata Krassil.. a proangiosperni tVom the Late Jurassic of Bureya Basin, Far East,
Russia (Krassilov, 1975): (1) gynoclade with lateral branches bearing paired follicle-like cupules, x l , (2)
cast of a cupule showing suture, x 10, (3. 4) cupule, with the ventral wall removed in (3) to show a solitary
anatropous ovule, xlO, (5, 6) distal parts of dehiscent cupules, with beaks slightly gaping, x3 and 10.
 Plate 5
Baisia hirsuta Krassil., a proangiospermous plant of bennettitalean affinities from the Baisian Assem­
blages. Early Cretaceous of Transbaikalia (Krassilov & Bugdaeva, 1982): (I) dispersed cupule with a
persistent receptacle bearing a tuft of long bristles; apical corona is distinctly marked off, x l 2 , (2, 3)
opened cupules showing seeds, x l 2 , (4) receptacle, proximal aspect with short bracts, xl40.
 Plate 6
Baisia hirsuta Krassil., a proangiospermous plant of bennettitalean affinities from the Baisian Assem­
blages, Early Cretaceous of Transbaikalia (Krassilov & Bugdaeva. 1982): (1) dispersed cupule, xl2, (2)
detached receptacle, x l 2 , (3) ovule removed from the cupule, x l 5 , (4,5) pollen chamber with pollen,
SEM. x4()(), and light micrograph. x300.
 Pi .AXES 183

Plate 7
Baisia hirsuta Krassil.. a proangiospermous planl of bennettitalean affinities from t h e Baisian Assem­
blages, Early Cretaceous of Transbaikalia (Krassilov & Bugdaeva, 1982): (1) pollen grain from the pol­
len chamber, x600, (2) infrastructure of the pollen grain seen in the areas where the tectum was erased by
Ihe pollen chambei exudates, x l 5 000.
 Plate 8
Dinophyton spinosum Ash, a protognetalean plant from the Late Triassic of Arizona, U.S.A. (Krassilov &
Ash, 1988): (1) partly cleared ovulate cupule with a four-lobed propeller-like wing formed of decussate
bracts, x l 5 , (2) cupule opened to show the ovule, x l 2 , (3) cupule showing pubescent lobes, x l 5 , (4)
nucellus, trigonal at base, xl7.
 Plate 9
Dinophyton spinosum Ash, a protognetalean plant from the Late Triassic of Arizona, U.S.A. (Krassilov &
Ash, 1988): (I) part of pollen cupule with bract, SEM, x70, (2, 3) laminar sporangiophores with a basally
attached sporangium, SEM, xllO, (4) pollen grain, SEM, xl 000.
 Plate 10
Hirmerella sp., the ovuliferous organ of a gnelalean proangiosperni from the the Early Jurassic of
Poland (Krassilov. 1973c): (1, 2) samaras with leaf-like wing lobes, x5, (3) wing cuticle with stomata,
SEM, x600, (4) double cuticle showing intersecting ceil files of the locule and the seed coat, xl70, (5)
two nucelli inside the samara shown in (I), x l 5 , (6) Classopollis pollen grains on the papillate cuticle
of the same samara. SEM, x500.
 Plate 11
A gnetalean plant from the Middle Jurassic of Ust-Baley. East Siberia (Krassilov & Bugdaeva, 1987): (1,
2) Anf>arolepis oclorata Krassil. & Bugd., cone bracts showing an attachment scar and resin bodies, x4,
(3-6) Heerala aniiqua (Heer) Krassil., samaras showing a median ridge and resin bodies, x4 and 10.
 Plate 12
Pollen organs of presumably the same gnetalean plant that produced Angarolepis-Heerala bracts and
samaras from the Middle Jurassic of Ust-Baley, East Siberia (Krassilov & Bugdaeva, 1987): (1 )Aegianthus
sibiricus (Heer) Krassil., pollen cone, x l . 5 , (2) sporophyll heads with lateral facets, x3, (3) sporangia,
xlO, (4) pollen grain, SEM, xl 800.
 Plate 13
Leongathia. ephedroid plant from the Koonwarra Fossil Bed Assemblage. Early Cretaceous of Victoria,
Australia (Krassilov & al., 1997): shoots with decussate leaf whorls; note that terminal internodes are
much shortened and sometimes actually condensed as in (1), x l 2 and 14.
 Plate 14
Leongalhia, ephedroid plant of the Koonwarra Fossil Bed Assemblage, Early Cretaceous of Victoria.
Australia (Krassilov & al., 1997), SEM: shoots differently preserved as subcrustation (mineral film infil­
trated under the cuticle) or incrustation (surface mineral deposits) and, respectively, showing either the
pattern of vascular bundles (1, 3), x30 and 100, or a ribbed surface with transverse striation reflecting a
sclerenchymous network of the cortex (2). x38. and stomata (4), xl 800. arrow in (1. 3) marking a small
vascular bundle ascending between the larger bundles at a deeper level in the vascular system.
 PI,AXES 191

Plate 15
Leongathia. ephedroid plant of the Koonwarra Fossil Bed Assemblage, Early Cretaceous of Victoria.
Australia (Krassilov & al., 1997), SEM: (1) interrib stomatal zone, x5()(), (2-4) tracheary elements with
helical and annular thickenings and with pits, in (4) showing perforation plate at the lower end (arrow);
also discernible are solid elements of a cortical sclerenchymous network overlapping the vascular tissue,
x2 ()()() (2), 2 500 (4) and 5 000 (3).

jr
 Plate 16
Baisianthus. a gnetalean androclade of the Baisian Assemblage. Early Cretaceous of Transbaikalia
(Krassilov & Bugdaeva, 1997): branching shoots bearing whorls of sporangiophores subtended by linear
bracts; lateral branches are decussate, arising at bract axils (1), and the fertile whorls arise at some dis­
tance above the branch node, but are more crowded distally (2), xlO.
 Plate 17
Baisiaiitlnis, a gnetalean androclade of the Baisian Assemblage, Early Cretaceous of Transbaikalia
(Krassilov & Bugdaeva, 1997): tracheary elements of vascular bundles of the axes, note septate elements
with porous perforations and helical thickenings transformed into irregular scalariform perforations of
what appears as a long perforation plate (arrow). SEM; x9()() and 2 100.
 Plate 18
Baisianthus. a gnetalean androclade of the Baisian Assemblage. Early Cretaceous of Transbaikalia
(Krassilov & Bugdaeva. 1997): (1) detached sporangiophore bearing synangia on lateral branches and
terminal on the main axis, x40, (2) a pair of cupules containing sporangiophores. one of them discernible
in the upper cupule, x25, (3) sporangiophore branch with sporangia, SEM, xl30.
 Plate 19
Baisiauthii.s. a gnetalean androclade of the Baisian Assemblage, Early Cretaceous of Transbaikalia
(Krassilov & Bugdaeva, 1997), SEM: (1) synangium of three sporangia, x210, (2) pollen mass of a
sporangium, note intact tetrads at the left marin (arrow), x450.
 Plate 20
Eoantha zherikhinii Krassil.. a proangiospermous gnetalean plant of the Baisian Assemblage, Early Cre­
taceous of Transbaikalia (Krassilov, 1986): ( 1 , 2 ) opened preflowers with radially spread perianth bracts
and gynoecial lobes, each of the latter bearing a solitary orthotropuos ovule, x3 and 8, (3, 4) closed
preflower showing obliquely spreading perianth bracts, one of the gynoecial lobes with a characteristic
transverse striation, and a toral extension of the floral apex with a tuft of bracts raised above the gynoecium,
x3 and 8.
 Plate 21
Eoantha zherikhinii Krassil., a proangiospermous gnetalean plant of the Baisian Assemblage, Early Cre­
taceous of Transbaikalia (Krassilov, 1986): (1) macerated ovules showing obovate, shortly beaked nucelli
with broad pollen chambers (arrow on a clump of pollen grains in the nucellar beak), x70, (3-5) clump of
polyplicate pollen grains marked in (2) and individual grains, x300, 600 and 1 000.
 Plate 22
Eoantha zherikhinii Krassil., a proangiospermous gnetalean plant of the Baisian Assemblage. Early Cre­
taceous of Transbaikalia (Krassilov, 1986; Krassilov & Bugdaeva, 1987), SEM: (1) cleared ovule of the
specimen illustrated in Text - Fig. 14, showing a low platform with canals of vascular traces at base, x50.
(2, 3) well-preserved granular megaspore membrane of the latter ovule. x65() and 300, (4) vascular strand
of the pedicel, tracheary elements with helical thickenings and pits, xl ()()().
 Plate 23
Graminoid leaf presumably of the Eoaniha plant, the Baisian Assemblage. Early Cretaceous of
Transbaikalia (Krassilov & Bugdaeva. 1997): (I) leaf with a lacerate blade and minute appendages at the
node, x8, (2) cross vein in a cleared leaf compression, SEM, x320, (4, 5) stomatal files of intercostal
grooves, those in (3) apparently stretched by leaf elongation. SEM. x69() and 750.
 Plate 24
Graminoid leaf presumably of the Eoantha plant, the Baisian Assemblage. Early Cretaceous of Trans­
baikalia (Krassilov & Bugdaeva, 1997): va.scular tissue showing fibres with small pits and tracheary
elements with helical thickenings and the larger pits, xl 400 (1) and 3 000 (2).
 Plate 25
Yitimcmtha, a preflower of the Baisian Assemblage, Early Cretaceous of Transbaikalia (Krassilov & Bugdaeva,
1997), showing four bracteate tepals with anastomosing veins and the follicular structures (arrow), pistils or
cupules, of which one to the right is fully preserved, with an attenuating apical region, x6 and 10.
 Plate 26
Viiiiiuiiuha. preflower of the Baisian Assemblage. Early Cretaceous of Transbaikalia (Krassilov &
Bugdaeva, 1997), SEM: (1) sutural papillate fringe of follicular organ (marked by arrow in Plate 25),
x900, (2) polyplicate ephedroid pollen grains stuck to the marginal fringe, xl 500. (.3) helical tracheids
from a trace of follicular structure. x220.
 Plate 27
Vitimantlui. preflower of the Baisian Assemblage. Early Cretaceous of Transbaikalia (Krassilov &
Bugdaeva, 1997), SEM: (1) marginal papillate fringe of follicular organ (marked by arrow in Plate 25),
x3 000, (2) clump of polyplicate pollen grains stuck to the marginal fringe showing sulcus and the uncleft
interrib exine, x3 000.
 Plate 28
Caytonia sewcinlii Thomas, from the Jurassic of Yorkshire, England (Krassilov, 1978b. 1984): (I) partly
cleared cupule showing the outlines of ovules, x l 5 , (2) part of a nucellus (left) and the inner integument,
same as in (4), x l 5 0 , (3) canals leading from the exostome to the ovules, with a stuck pollen grain (ar­
row), x395, (4) nucellus with a partly preserved inner integument (outer integument removed), x70.
i Plate 29
Caytonia sewardii from the Jurassic of Yorlcshire. England (Krassilov, 1978b, 1984): (1) nucellus
showing a median ridge, x l 6 6 , (2) seed showing coat cells, xlOO, (3) micropyles of two adpressed ovules
showing bulging cells, x395.
 Plate 30
Caytoiumthiis tyrmensis Krassil., a caytonialean androclade from the basal Cretaceous of the Tyrma River.
Far East, Russia (Krassilov, 1978b): (1, 2) androclade branches bearing synangia that are clustered on
shorter branches (1) or turned to one side and pendent on longer branches (2), x8, (3) synangium, x 16, (4)
pollen grain, SEM, x3 500.
 Plate 31
Czekanowskialean reproductive structures: (1) Lcploslmhus stigmatoideiis Krassil.. from the Late Jurassic of
Bureya Basin. Far East. Russia (Krassilov. 1969. 1972a). gynoclade with lateral cupules on variably develo­
ped stalks, some appearing as short-shoots, x 1. (2. 3) cupules cleared from the rock and opened to show valves
and the pitted papillate fringes. x3. (4) pits and papillae on marginal fringes, xl 17, (5) Ixostrolnis schmidtiaims
(Heer) Krassil., pollen organ from the same locality, lateral sporangiophores with 4-lobed synangia, x8.
 Plate 3 2
Leptostrohus stiginaloideus Krassil.. czekanowskialwcan reproductive structures from the Late Jurassic
of Bureya Basin, Far East, Russia (Krassilov, 1969, 1972); (1, 2) cupule valves showing scars of the
ovules, x6, (3) interior extension of the stigmatic marginal IJange intruding the locule, SEM, xl20.
 Plate 33
Reproductive structures of Cretaceous platanoids: (I, 2, 5) Sarhaya radiala Krassil. & Shilin, a staminate
platanoid head from the Late Cretaceous (Cenomanian) of Kazakhstan (Krassilov & Shilin, 1995), sta­
men with a short filament, a recurved connective apophysis and narrow pollen sacs, SEM, x50 (1), staminate
head of distinct wedge-shaped flowers, xlO (2), and flower of four stamens (three exposed) with short
perianth lobes (left), SEM, x25 (5): (3, 4) axis bearing small pistillate heads, from the Late Cretaceous of
the Amur River Valley (Krassilov, 1976), xl and 7.
 Plate 3 4
Sarhaya radiata Krassil. & Shilin. a staminatc platanoid head from the Late Cretaceous (Cenomanian) of
Kazaichstan (Krassilov & Shilin, 1995): (1) pollen grain showing an exocolpus with a granular mem­
brane, slightly gaping in the middle over the porous mesoaperture, x7 000, (2) details of the exocolpus,
SEM, xl5 000, O) oblique polar view of a pollen grain with narrow, almost straight colpi. SEM. x4 ()()().
(4) pollen grains at different foci showing porous mesoapertuics and lalongalc endoapertures forming a
thin equatorial girdle, or endocingulum, x2 000.
 Plate 35
Ccispiocarpiis paniciiliger Krassil. & Vachr., a ranunculoid angiosperm from the Early Cretaceous of
Kazakhstan (Vakhrameev & Krassilov. 1979: Krassilov. 1984): (1) branching shoot with two panicles
terminal on both central axis and lateral branch, the central axis bearing two opposite leaves (pale and
poorly seen in the photograph, arrow on the leaf node) arising below the panicle, x l , (2) panicle with two
head-like basal branches, x5, (3-5) panicle showing short lateral racemes, dense and overlapping, bearing
small elliptical follicles, in (5) terminating in a looser raceme, xlO.
 Plate 36
Caspiocarpus pcmicidiger Krassil. & Vachr.. a lanunculoid angiosperm from the Early Cretaceous of
Kazakhstan (Vakhrameev & Kra.ssilov, 1979; Kra.ssilov, 1984): (1) two ovules from a follicle, xlOO, (2)
ovule showing two thin integuments (left), the inner one wedging out much below the micropyle, x200,
(3) follicle with small appendages (tepals?) at base. x70, (4) follicle wall with hair bases, x2 000.
 P l a t e 37
Hyrcantha karatscheensis (Vachr.) Krassil.. a paeonialean inflorescence from the Early Cretaceous of
Kazakhstan (Krassilov & al., 1983): (I) cymose inflorescence, x2, (2) flower showing filaments of poorly
preserved stamens at the base of a median carpel, x7.
 Plate 3 8
Hyrcantha karatscheensis (Vachr.) Krassil.. a paeonialean inflorescence from the Early Cretaceous of
Kazakhstan (Krassilov & al., 1983): (1) flower at a fruiting stage showing reflexed lobes of a persistent
calyx, x7, (2) internal cast of a dehiscent carpel showing transverse striation of the locule, xlO, (3) carpel
showing a broad stigma, x7, (4) casts of stomata in the impression of "Leguminosites" karatscheensis
Vachr., ternate compound leaves found with Hyrcantha llowers, SEM, x350.
 Plate 39
Astemcehislriix cretcicea Velen. & Viniklar. a rosaceous fruit from the Late Cretaceous (Cenomanian) of
Bohemia (Krassilov & Pacltova, 1989): (1) open pentamerous capsule, syncarpous and adnate to the
calyx, lobes of the latter spreading star-like outside of the capsule, x3, (2-5) details of the capsule and
anatropous ovules, x7 (2) and 15 (3-5).
 Plate 40
Krassil. & Shilin, flowers from the Late Cretaceous of Kazakhstan (Krassilov & al., 1983):
(1) longitudinally split (lower with the gynoecium much longer than the stamen (left); three of the carpels and
their free styles are exposed, with the median carpel showing the outlines of a locule, xlO, (2) styles, xl5, (3)
flower with a better preserved calyx. xlO, (4) three flowers showing folds between the lobes of syncarpous
ovaries, x3, (5) hairs on the ovary, xl3(): (6) hooked staminode and a clavate gland of the androecium, x250.
 Plate 41
Rumonkaryu nevoi Krassil. & Dobrusk.. a samaroid fruit from the Early Cretaceous (Aptian) of Makhtesh
Ramon, Israel (Krassilov & Dobruskina, 1995): (1) adaxial aspect (n - nutlet, s - stigma, w - winglet), x6.
(2) bract venation, x9, (3) winglet and nutlet, stylar end with biarmed stigma, x l 5 , (4) adaxial aspect with
a partly preserved pedicel, x5.
 Plate 4 2
TaUlysajci medusa Krassil. & Shilin. a monocol inllorescence from the Late Cretaceous of Kazakhstan
(Krassilov & al., 1983): (I) fasciculate inflorescence on a long peduncle, x2, (2) flower showing the
gynoecium with a bifid stigma, xlO, (3, 4) fascicle rays showing areolate scars of shed flowers, xlO, (5)
distal parts of fascicle rays with solitary flowers. xlO.
 Plate. 43
Tyrmocarpus spinulosus Krassil.. fruit-like structures from the terminal Jurassic - lowermost Cretaceous
of the Tyrma River, Far East, Russia: (1, 2) external view with spinules, x7, (3) split fruit-like body
showing a solitary ovule, x7, (4) fruit surface, x l 2 , (5-1) fruit wall cuticle showing a spinule (5), a scar of
detached spinule (6) and stoma (7). xl46.
 Plate 44
Dominant angiosperms of the post-crisis early Paleocene plant assemblages from the Tsagajan Formation
of the Amur Province (Krassilov. 1976): (1-3) Trochodendrocarpus arcticus Krysht., panicles of paired
carpels, x l , (4) Trochodendmides-typc leaf of supposedly the same plant, x l , (5, 6) Carinalaspermwn
bureicum Krassil., liquidambroid seeds, x3, (7-9) Tricolpopollianllnis hurejensis Krassil, staminate head,
x7, and pollen grains, x395 and 1 000.
 Plate 45
Leaf polymorphism in /mania from the Late Triassic of Primorye. Russian Far East (Krassilov & Shoro-
khova, 1970) showing transitions from pinnate to dichopodial to palmate morphologies (1-6), the latter
similar to Sagenopteris. a caytonialean leaf genus with anostomosing veins (7), x8.
 Plate 4 6
Meeusella proteicUida Krassil. & Bugd.. a supposed prototype of fasciculate stamens from the Early
Cretaceous of Transbaikalia (Krassilov & Bugdaeva, 1988b): (I) androclade bearing fertile and sterile
branches, x3, (2) paraclade with two stalked sporangial heads lateral to the apex, xlO, (3) paraclade
with sporangial heads almost sessile and with the apex protruding between them, xlO, (4) sterile
paraclade, x 10.
 Plate 47
Lemnoid fruits I'rom the Lute Cretaceous of Mongolia (Krassilov & Makulbekov. 1995) supposedly re­
taining a primitive cupular structure of the utricle with an orthotropous ovule, SEM: (1) utricle, surface
with lime-secreting glands, x70, (2) split utricle showing the ovule, x70, (3) ovule removed from the
utricle, showing the funicle and operculum at the opposite ends but not quite in line with each other. x7().
(4) proximal part of the ovule showing the funicle as a direct extension of the pedicel, x7().
 Plate 48
Lemnoid fruits from the Late Cretaceous of Mongolia (Krassilov & Makulbekov. 1995) supposedly
retaining a primitive cupular structure of the utricle with an orthotropous ovule, SEM: (1)
ampohoriform utricle with a perfect funnel-shaped stigma, x70, the latter enlarged in (2), x l 5 0 , (3)
operculum with an apical coruncle, x3()0.
 References
Agadjanian, A.K. 1996. M o d e s of phytophagy in m a m m a l s . - Paleont. Z h . ( M o s c o w ) ,
N o . 6, p. 7 2 3 - 7 2 9 .
Allen, G.S. 1943. T h e e m b r y o g e n y of Pseudotsuga taxifolia (Lamb.) Britt. - Amer. J.
Bot., vol. 30, p. 6 5 5 - 6 6 1 .
A l v i n , K . L . 1982. C h e i r o l e p i d i a c e a e : biology, s t r u c t u r e a n d p a l e o e c o l o g y . - Rev.
Palaeobot. Palynol., vol. 37, p. 7 1 - 9 8 .
Alvin, K.L., Spicer, R.A., Watson, J.A. 1978. Classopollis-conidimxng m a l e c o n e asso­
ciated with Pseudofrenelopsis. - Palaeontology, vol. 2 1 , n o . 4, p . 9 6 - 9 8 .
Anderson, J.M., Anderson, H.M. 1983. Palaeoflora of Southern Africa. M o l t e n o For­
mation (Triassic). Rotterdam: Balkema, vol. 1, 227 p.
A n d e r s o n , J.M., A n d e r s o n , H . M . 1985. Palaeoflora of S o u t h e r n Africa. R o t t e r d a m :
Balkema, 423 p.
Antonov, A . S . 1995. Is D a r w i n ' s "abominable m y s t e r y " solved? - In: Evol. B i o c h e m . &
Related Areas of P h y s i o c h e m . Biol., M o s c o w , p. 2 7 3 - 2 8 2 (in Russian).
Arber, A. 1918. T h e phyllode theory of m o n o c o t y l e d o n o u s leaf, with special reference
to anatomical evidence. - Ann. Bot. (Gr. Brit.), voi. 32, p . 4 6 5 - 5 0 1 .
Arber, A. 1925. Monocotyledons, a morphological study. Cambridge: Univ. Press, 258 p.
Arber, A. 1950. T h e natural philosophy of plant form. C a m b r i d g e : Univ. Press, 247 p .
Arber, E.A.N., Parkin, J. 1907. On the origin of a n g i o s p e r m s . - Bot. J. Linn. S o c , vol.
3 8 , p. 29-44.
Archangelsky, S. 1965. Fossil Ginkgoales from the Tico flora, Santa C r u z Province,
Argentina. - Bull. Brit. M u s . (Nat. Hist.), Geol., vol. 10, n o . 5, p . 121-137.
Archangelsky, S. 1968a. On the genus Tomaxellia (Coniferae) from the L o w e r Creta­
ceous of Patagonia and its male and female cones. - Bot. J. Linn. S o c , vol. 6 1 , no.
384, p . 153-165.
Archangelsky, S. 1968b. Studies on Triassic fossil plants from Argentina. 4 . T h e leaf
genus Dicroidium and its possible relation to Rhexoxylon stems. - Palaeontology,
vol. 11, no. 4, p. 5 0 0 - 5 1 2 .
Archangelsky, S., C u n e o , R. 1987. Ferugliocladaceae, a n e w conifer family from the
Permian of G o n d w a n a . - Rev. Palaeobot. Palynol., vol. 5 1 , nos. 1/3, p . 3-30.
Arnold, C. 1948. Classification of g y m n o s p e r m s from the viewpoint of paleobotany. -
Bot. Gaz., vol. 110, no. 1, p. 1-12.
Arnold, C.A., Lowther, J.S. 1955. A new Cretaceous conifer from northern Alaska. -
Amer. J. Bot., vol. 4 2 , p. 5 2 2 - 5 2 8 .
Arnott, H. 1959. A n a s t o m o s e s in the venation of Ginkgo biloba. - Amer. J. Bot., vol. 4 6 ,
p. 4 0 5 - 4 1 1 .
Asama, K. 1962. Evolution of Shansi flora and origin of simple leaf. - Sci. R e p . Tohoku
Univ., Ser. 2, Spec, vol., no. 5, p. 2 4 7 - 2 7 3 .
Asama, K. 1976. Gigantopteris flora in southeast Asia and its phytogeographic signifi­
cance. - In: Geology and Palaeontology of Southeastern Asia. T o k y o : Univ. Press,
p.191-207.
Ash, S.R. 1970. Dinophyton, a problematical new plant genus from the U p p e r Triassic
of the S o u t h - W e s t e m United States. - Palaeontology, vol. 13, n o . 4, p . 646-663.
Ash, S. 1976. T h e systematic position of Eoginkgoites. - Amer. J. Bot., vol. 6 3 , no. 10,
p. 1 3 2 1 - 1 3 3 1 .
Axelrod, D.I. 1952. A theory of angiosperm evolution. - Evolution, vol. 6, p . 29-60.
Axelrod, D.I. 1959. Poleward migration of early angiosperm flora. - Science, vol. 130,
p. 2 0 3 - 2 0 7 .
Axelrod, D.I. 1972. Edaphic aridity as a factor in angiosperm evolution. - Amer. Natur.,
vol. 106, no. 949, p. 311-320.
Bailey, J.W. 1956. Nodal anatomy in retrospect. - J. Arnold Arbor., vol. 3 7 , p . 269-287.
Bailey, J.W., S w a m y , B . G . L . 1951. T h e conduplicate carpel of d i c o t y l e d o n s and its
initial trends of specialization. - Amer. J. Bot., vol. 3 8 , p . 3 7 3 - 3 7 9 .
Basinger, J.F. 1976. Paleorosa similkameensis, gen. et sp. nov., permineralized flowers
(Rosaceae) from the E o c e n e of British C o l u m b i a . - C a n a d . J. Bot., vol. 54, no. 20,
p. 2 2 9 3 - 2 3 0 5 .
Basinger, J.F., Dilcher, D.L. 1984. Ancient bisexual flowers. - Science, vol. 224, no.
4648, p. 511-513.
Bate-Smith, E.C. 1972. Chemistry and phylogeny of the a n g i o s p e r m s . - Nature, vol.
236, p. 353-354.
B a u m , H. 1949. Der einheitliche Bauplan der angiospermen G y n o c i e n . - Osterr. bot.
Ztschr., Bd. 96, S. 6 4 - 8 2 .
Beck, C . B . 1960. T h e identity of Archaeopteris and Callixylon. - Brittonia, vol. 12, p.
351-368.
Behnke, H . - D . 1986. Sieve element characters and systematic position oiAustrobaileya,
Austrobaileyaceae - with c o m m e n t s to the distinction and definition of sieve cells
and sieve-tube m e m b e r s . - Plant Syst Evol., vol. 152, nos. 1/2, p . 101-121.
Beland, P., Russel, D.A. 1978. Palaeoecology of Dinosaur Provincial Park (Cretaceous),
Alberta, interpreted from the distribution of articulated vertebrate remains. - Canad.
J. Earth Sci., vol. 15, p. 1012-1024.
Bell, W.A. 1957. Flora of Upper Cretaceous group of Vancouver Island, British Colum­
bia. - M e m . Geol. Surv. Canada, vol. 2 9 3 , p. 1-84.
Bentham, G., Hooker, J.D. 1862-1883. G e n e r a plantarum. L o n d o n , vol. I-III.
Berry, E.W. 1911. Systematic paleontology of the L o w e r Cretaceous deposits of Mary­
land. Baltimore: M d . Geol. Surv., 597 p .
Berry, E.W. 1916. T h e U p p e r Cretaceous deposits of Maryland. Baltimore: M d . Geol.
Surv., p. 7 5 7 - 9 0 1 .
Bessey, C.E. 1897. Phylogeny and taxonomy of angiosperms. - Bot. Gaz., vol. 24, p .
145-178.
Bessey, C.E. 1915. T h e phylogenetic taxonomy of flowering plants. - A n n . M o . Bot.
Card., v o l . 2 , p. 109-164.
Bierhorst, D.W. 1958. Vessels in Equisetum, - Amer. J. Bot., vol. 4 5 , no. 7, p . 534-537.
Binda, P L . , N a m b u d i r i , E.W.V. 1983. Fossil seed cuticles from the U p p e r Cretaceous
W h i t e m u d beds of Alberta and Saskatchewan, C a n a d a . - C a n a d . J. Bot., vol. 6 1 , p.
2717-2728.
B l a c k m o r e , S., van Helvoort, H.A.M., Punt, W. 1984. On the terminology, origins and
functions of caveate pollen in C o m p o s i t a e . - Rev. Palaeobot. Palynol., vol. 4 3 , p.
293-301.
Blagoveshchenskiy, A . V 1975. Seed proteins and the phylogeny of A n g i o s p e r m a e . -
Byull. M o s k . obshch. ispyt. prir., Biol., t. 80, no. 4, p . 87-92 (in Russian).
Blank-Louvel, C. 1984. Le genre '"Ranunculus L . " dans le Berriasien (Cretace inf.) de
la province de Lerida (Espagne). - L a Revista Ilerida, t. 4 5 , p . 8 3 - 9 2 .
Blanc-Louvel, C , Barale, G. 1983. Montsechia vidali (Zeiller) Teixeira 1954. Nouvelles
observations et reflexions sur son attribution systematique. - A n n . Paleontol., t. 69,
p. 151-174.
Bogle, A.L. 1986. T h e floral morphology and vascular anatomy of the H a m a m e l i d a c e a e :
subfamily Liquidambroideae. - Ann. M o . Bot. Gard., vol. 7 3 , no. 2, p . 325-347.
Boltenhagen, B . 1982. A m e n d e m e n t de Sofrepites legouxae Jard. 1967 pollen Albien
du G a b o n . - Rev. Micropaleontol., t. 2 5 , fasc. 3, p. 150-152.
Bose, M . N . 1966. A revision of Rajmahalia, - Palaeobotanist, vol. 14, n o s . 1/2, p . 85-88.
Boulter, D., B a m s h a w , G.A.M., T h o m p s o n , E.W. 1972. A phylogeny of higher plants
based on the a m i n o acid sequences of c y t o c h r o m e " c " and its biological implica­
tions. - Proc. Roy. S o c . L o n d o n , B , vol. 181, p. 4 4 1 - 4 5 5 .
B o u m a n , F. 1984. T h e o v u l e . E m b r y o l o g y of a n g i o s p e r m s . Berlin e t c . : S p r i n g e r , p .
123-157.
B o u m a n , F , Calis, J.L.M. 1977. Integumentary shifting - a third way to unitegmy. -
Berichte Dtsch. bot. Ges., Bd. 90, S. 15-28.
Brenner, G.J. 1967. Early angiosperm pollen differentiation in the Albian to C e n o m a n i a n
deposits of D e l a w a r e (U.S.A.). - Rev. Palaeobot. Palynol., vol. 1, nos. 1/3, p . 2 1 9 -
227.
B r e n n e r , G.J. 1 9 7 6 . M i d d l e C r e t a c e o u s floral p r o v i n c e s a n d e a r l y m i g r a t i o n s of
a n g i o s p e r m s . Origin and early evolution of a n g i o s p e r m s . N.Y.: C o l u m b i a Univ.
Press, p . 2 3 - 4 7 .
B r o w n , R.W. 1956. Palmlike plants from the Dolores F o r m a t i o n (Triassic) in south­
western C o l o r a d o . - U . S . Geol. Surv. Profess. Pap., vol. 2 7 4 H , p . 2 0 5 - 2 0 9 .
Burger, W . C . 1977. T h e Piperales and the m o n o c o t s . Alternate hypotheses for the origin
of m o n o c o t y l e d o n o u s flowers. - Bot. Rev., vol. 4 3 , p . 3 4 5 - 3 9 3 .
Cairns, H., Cairns, C. 1976. Psycholinguistics, a cognitive view of language. N.Y.: Holt.
C a h o o n , E.J. 1972. Paraphyllanthoxylon alabamense - a n e w s p e c i e s of fossil
dicotyledonous wood. - Amer. J. Bot., vol. 59, p. 5-11.
Canright, J.E. 1952. T h e comparative morphology and relationships of the Magnoliaceae.
1. Trend of specialization in the stamens. - Amer. J. Bot., vol. 39, no. 7, p. 484-497.
Carlquist, S. 1969. Toward acceptable evolutionary interpretation of floral anatomy. -
Phytomorphology, vol. 19, p. 3 3 2 - 3 6 2 .
Carlquist, S. 1987. Presence of vessels in wood of Sarcandra (Chloranthaceae): com­
ments on vessel origins in angiosperms. - Amer. J. Bot., vol. 174, p. 1765-1771.
Chadefaud, M . 1966. Sur le sac embryonnaire des angiospermes. - Bull. Soc. bot. Fr., t.
113, p. 3 7 5 - 3 7 8 .
C h a k r o u n , S. 1983. L e s f o r m a t i o n s s e c o n d a i r e s d e s m o n o c o t y l e d o n s : conceptions
actuelles. - Ann. Biol., t. 22, fasc. 2, p. 187-199.
Chaloner, W.G. 1976. T h e evolution of adaptive features of fossil exines. - In: Fergusson,
I.K. & Muller, J. (eds.). T h e evolutionary significance of the exine. London.- Aca­
demic Press, p. 1-14.
Chaloner, W.G., Hemsley, A.R. 1991. Heterospory: cul-de-sac or p a t h w a y to the seed?
- In: B l a c k m o r e , S. & Barnes, S.H. (eds.). Pollen and spores: patterns of diversifi­
cation. Oxford: Univ. Press, p. 151-167.
Chaloner, W.G., Hill, A.J., Lacey, W.S. 1977. First Devonian platyspermic seed and its
implications in g y m n o s p e r m evolution. - Nature, vol. 2 6 5 , p . 2 3 3 - 2 3 5 .
Chamberlain, C.J. 1966a. G y m n o s p e r m s . Structure and evolution. N.Y.: Dover, 4 8 4 p.
Chamberlain, C.J. 1966b. T h e living cycads. N.Y.: Hafner, 172 p.
Chandler, M.E.J., Axelrod, D.J. 1961. An early angiosperm fruit from California. - Amer.
J. Sci., vol. 259, p. 4 4 1 - 4 4 6 .
Chen Z u - k e n g , W a n g Fu-xiog. 1985. T h e ovule structure and d e v e l o p m e n t of male and
female gametophytes in Amentotaxus. - Acta Bot. Sin., vol. 27, p. 19-25.
Chute, H.W. 1930. T h e morphology and anatomy of the achene. - Amer. J. Bot., vol. 17,
no. 8, p. 703-720.
Clement-Westerhof, J.A. 1974. In situ pollen from g y m n o s p e r m o u s cone of the Upper
Permian of the Italian Alps - a preliminary account. - Rev. Palaeobot. Palynol., vol.
17, p . 6 5 - 7 3 .
Clement-Westerhof, J.A. 1984. Aspects of Permian palaeobotany and palynology, 4. -
Rev. Palaeobot. Palynol., vol. 4 1 , p. 51-166.
Clement-Westerhof, J.A. 1987. Aspects of Permian palaeobotany and palynology, 7. -
Rev. Palaeobot. Palynol., vol. 52, p. 3 7 5 - 4 0 2 .
C l e m e n t - W e s t e r h o f , J.A., van K o n i j n e n b u r g - v a n Cittert, J . H . A . 1 9 9 1 . Hirmeriella
muensteri: n e w data on the fertile o r g a n s leading to a revised c o n c e p t of the
Cheirolepidiaceae. - Rev. Palaeobot. Palynol., vol. 6 8 , p. 147-179.
Cocucci, A.E. 1973. S o m e suggestions on the evolution of gametophytes of higher plants.
- Phytomorphology, vol. 2 3 , nos. 1/2, p. 109-124.
C o c u c c i , A.E., A s t e g i a n o , M.F. 1978. Interpretacion del e m b r i o n d e las P o a c e a s . -
Kurtziana, t. 11, p. 4 1 - 5 4 .
C o h e n , L.I. 1970. D e v e l o p m e n t of t h e p i s t i l l a t e f l o w e r in t h e d w a r f m i s t l e t o e s ,
Arceiithohium. - Amer. J. Bot., vol. 57, no. 4, p. 4 7 7 - 4 8 5 .
Colin, J.-P. 1973. Microfossiles vegetaux dans le C e n o m a n i e n et le Turonien de Dor-
d o g n e (S.O. France). - Palaeontographica B , Bd. 143, S. 106-119.
Conover, M . H . 1982. T h e vegetative morphology of the reticulate-veined Liliiflorae. -
Telopea, vol. 2, no. 4, p. 4 0 1 - 4 1 2 .
C o m e t , B . 1979. Angiosperm-like pollen with tectate-columellate wall structure from
the U p p e r Triassic (and Jurassic) of the N e w a r k Supergroup, U.S.A. - Palynology,
v o l . 3 , p. 281-282.
C o r n e t , B . 1986. T h e leaf venation and r e p r o d u c t i v e structures of a L a t e Triassic
angiosperm, Sanmiguellia lewisii. - Evol. Theory, vol. 7, no. 5, p. 2 3 1 - 3 0 9 .
Couper, R.A. 1958. British Mesozoic microspores and pollen grains. - Palaeontographica
B, Bd. 103, S. 75-179.
Crane, P R . , Dilcher, D.L. 1984. Lesqueria: an early angiosperm fruiting axis from the
mid-Cretaceous. - Ann. M o . Bot. Gard., vol. 7 1 , no. 2, p. 3 8 4 - 4 0 2 .
Crane, P R . , Friis, E.M., Pedersen, K.R. 1986. L o w e r Cretaceous angiosperm flowers:
fossil evidence on early radiation of dicotyledons. - Science, vol. 2 3 2 , p . 852-854.
Crane, P R . , Friis, E.M., Pedersen, K.R. 1989. Reproductive structure and function in
Cretaceous Chloranthaceae. - Plant Syst. Evol., vol. 165, p. 211-226.
Crane, P R . , Stockey, R.A. 1985. Growth and reproductive biology of Jojfrea speirsii
gen. et sp. nov., a Cercidiphyllum-Wkt plant from the Late Paleocene of Alberta,
C a n a d a . - Canad. J. Bot., vol. 6 3 , p. 340-364.
Crane, P R . , Stockey, R.A. 1986. Morphology and d e v e l o p m e n t of pistillate inflores­
cences in extant and fossil Cercidiphyllaceae. - Ann. M o . Bot. Gard., vol. 7 3 , p.
382-393.
Crane, P R . , Upchurch, R. (Jr.) 1987. Drewria potomacensis gen. et sp. nov., an early
Cretaceous m e m b e r of Gnetales from the P o t o m a c G r o u p of Virginia. - Amer. J.
Bot., v o l . 7 4 , p. 1722-1736.
Cranwell, L . M . 1961. Coniferous pollen types of the Southern H e m i s p h e r e . 1. Aberra­
tion in Acmopyle and Pbdacarpus dacrydioides. - J. Arnold Arbor., vol. 4 2 , no. 4,
p. 4 1 6 - 4 2 3 .
Crepet, W.L. 1974. Investigations of North American cycadeoids: the reproductive bio­
logy of Cycadeoidea. - Palaeontographica B, Bd. 148, S. 144-169.
Crepet, W.L., Delevoryas, T. 1972. Investigations of North A m e r i c a n cycadeoids: early
ovule ontogeny. - Amer. J. Bot., vol. 59, p. 2 0 9 - 2 1 5 .
Crepet, W.I., Nixon. K.C. 1994. Flowers of Turonian M a g n o l i i d a e and their implica­
tions. - Plant Syst. Evol., Suppl., vol. 8, p. 73-92.
Cridland, A.A., Morris, J.E. 1960. Spermopteris, a new genus of pteridosperms from
the U p p e r Pennsylvanian Series of Kansas. - Amer. J. Bot., vol. 4 7 , p . 8 5 5 - 8 5 9 .
Croizat, L. 1964. T h o u g h t s on high systematics, phylogeny and floral m o r p h o g e n y with
a note on the origin of angiosperms. - Candollea, vol. 19, p. 18-96.
Cronquist, A. 1968. T h e evolution and classification of the flowering plants. Boston:
Houghton Mifflin Co., 396 p .
C r o n q u i s t , A. 1974. T h o u g h t of an origin of the m o n o c o t y l e d o n s . - B . S a h n i Inst.
Palaeobot. Spec. Publ., n o . 1, p. 19-24.
C r o n q u i s t , A . 1977. O n the t a x o n o m i c a l significance of s e c o n d a r y m e t a b o l i t e s in
angiosperms. - Plant Syst. Evol., Suppl., vol. 1, p. 179-189.
Cronquist, A. 1981. An integrated system of classification of flowering plants. N.Y.:
C o l u m b i a Univ. Press, 1262 p .
Dahlgren, R. 1983. General aspects of angiosperm evolution and macrosystematics. -
Nord. J. Bot., v o l . 3 , p. 119-149.
Daghlian, C P . 1981. A review of the fossil record of m o n o c o t y l e d o n s . - Bot. Rev., vol.
47, p. 517-555.
Darwin, Ch. R. 1859. On the origin of species. L o n d o n : Murray, 4 9 0 p .
Davis, G.L. 1966. Systematic embryology of the angiosperms. N . Y.: Wiley.
De Boer, R., B o u m a n , F. 1972. Integumentary studies in the Polycarpicae. 2. Magnolia
stellata and Magnolia virginiana. - Acta Bot. Neerl., vol. 2 1 , n o . 6, p . 617-629.
De Candolle, A.R 1827. Organographie vegetale. Paris: Detreville, 1.1, 558 p., t. II, 304 p.
Delevoryas, T. 1955. T h e Medullosae - structure and relationships. - Palaeontographica
B , Bd. 9 7 , S. 114-167.
Delevoryas, T. 1964. A probable pteridosperm microsporangiate fructification from the
Pennsylvanian of Illinois. - Paleontology, vol. 7, p . 6 0 - 6 3 .
Delevoryas, T. 1965 (1966). Investigations of North A m e r i c a n c y c a d e o i d s : microspo­
rangiate structures and phylogenetic implications. - Palaeobotanist, vol. 14, nos. 1/
3, p . 89-99.
Delevoryas, T. 1968. S o m e aspects of cycadeoid evolution. - Bot. J. Linn. S o c , vol. 6 1 ,
p. 137-146.
Dettmann, M . E . 1963. U p p e r M e s o z o i c microfloras from southern Australia. - Proc.
Roy. Soc. Victoria, vol. 77, p. 1-148.
Dettmann, M . E . 1973. A n g i o s p e r m o u s pollen from Albian to Turonian sediments of
eastern Australia. - Spec. Publ. Geol. S o c . Austral., vol. 4, p . 3-34.
Diggle, P.K., de M a s o n , D.A. 1983. T h e relationship between the primary thickening
meristem and the secondary thickening meristem in Yucca whipplei Torr., 2. - Amer.
J. Bot., vol. 70, no. 8, p. 1205-1216.
Dilcher, D.L. 1974. A p p r o a c h e s to the identification of angiosperm leaf remains. - Bot.
Rev., v o l . 4 0 , p. 1-157.
Dilcher, D . L . 1979. Early a n g i o s p e r m r e p r o d u c t i o n : an i n t r o d u c t o r y report. - Rev.
Palaeobot. Palynol., vol. 27, p. 2 9 1 - 3 2 8 .
Dilcher, D.L., Crane, P.R. 1984. Archaeanthus: an early angiosperm from the Cenoma­
nian of the western interior of North America. - A n n . M o . Bot. Gard., vol. 7 1 , no.
2, p. 3 5 1 - 3 8 3 .
Dilcher, D.L., Crepet, W.L., Beeker, C D . , Reynolds, H. 1976. R e p r o d u c t i v e and vegeta­
tive morphology of a Cretaceous angiosperm. - Science, vol. 2 7 , p. 8 5 4 - 8 5 6 .
Dilcher, D.L., Kovach, W.L. 1986. Early angiosperm reproduction: Caloda delevoryana
gen. et sp. nov., a new fructification from the D a k o t a Formation ( C e n o m a n i a n ) of
Kansas. - Amer. J. Bot., vol. 7 3 , p. 1230-1237.
Dilcher, D.L., Krassilov, V.A., Douglas, J.G. 1996. A n g i o s p e r m evolution: fruits with
affinities to Ceratophyllales from the L o w e r Cretaceous. - Abstr. 5th Conf. Int.
Org. Paleobot., Santa Barbara, Calif., p. 2 3 .
D o n o g h u e , M.J., D o y l e , J.A. 1989. P h y l o g e n e t i c a n a l y s i s of a n g i o s p e r m s and the
relationships of H a m a m e l i d i d a e . - In: C r a n e , P. R. & B l a c k m o r e , S. (eds.). E v o ­
lution, s y s t e m a t i c s and fossil history of the H a m a m e l i d i d a e , vol. 1. O x f o r d :
C l a r e n d o n , p. 17-45.
Dorf, E. 1942. Flora of the L a n c e Formation in its type locality, Niobrara County, W y o ­
m i n g . - Carnegie Inst. Wash. Publ., no. 580, p. 7 9 - 1 5 9 .
Douglas, G.E. 1944. T h e inferior ovary. - Bot. Rev., vol. 10, p . 125-186.
Douglas, J.G. 1969. T h e M e s o z o i c floras of Victoria, 1-2. - M e m . Geol. Surv. V i c , vol.
2 8 , p. 1-310.
Doyle, J.A. 1973. Fossil evidence on early evolution of the m o n o c o t y l e d o n . - Quart.
Rev. Biol., v o l . 4 8 , p, 3 9 9 - 4 1 3 .
Doyle, J.A., D o n o g h u e , M.J. 1986. Seed plant phylogeny and the origin of angiosperms:
an experimental cladistic approach. - Bot. Rev., vol. 52, no. 4, p. 3 2 1 - 4 3 1 .
Doyle, J.A., Hickey, L.J. 1972. Coordinated evolution in P o t o m a c G r o u p angiosperm
pollen and leaves. - Amer. J. Bot., vol. 59, p . 6 6 0 .
Doyle, J.A., Van C a m p o , M., Lugardon, B . 1975. Observations on exine structure of
Eucommiidites and lower Cretaceous angiosperm pollen. - Pollen et Spores, vol.
17, p . 4 2 9 - 4 8 6 .
Drinnan, A.N., C h a m b e r s , T.C. 1986. Flora of the L o w e r Cretaceous K o o n w a r r a Fossil
B e d ( K o r u m b u r r a G r o u p ) , South G i p p s l a n d , Victoria. - M e m . A s s o c . Austral.
Palaeont., vol. 3, p. 1-77.
Drinnan, A.N., C r a n e , P R . , Friis, E.M., Pedersen, K.R. 1990. L a u r a c e o u s flowers from
the P o t o m a c G r o u p (mid-Cretaceous) of eastern North America. - Bot. Gaz., vol.
151, n o . 3, p . 370-384.
F a m e s , A.J. 1913. T h e morphology of Agathis australis. - Ann. Bot. (Gr. Brit.), vol. 27,
p. 1-38.
F a m e s , A.J. 1961. M o r p h o l o g y of the angiosperms. N.Y.: M c G r a w Hill, 5 1 8 p.
Eiten, L.T. 1976. Inflorescence units in the Cyperaceae. - Ann. M o . Bot. Gard., vol. 6 3 ,
p. 81-112.
Emberger, L. 1954. Sur les Ginkgoales et quelques rapprochements avec d'autres groupes
systematiques. - Svensk. Bot. Tidskr., Bd. 4 8 , H. 2, S. 3 6 1 - 3 6 7 .
Emberger, L. 1960. Les vegetaux vasculaires. Traite de botanique systematique. Paris:
M a s o n , 1539 p.
Endress, P K . 1983. T h e early floral development of Austrohaileya. - Bot. J b . (Syst.),
Bd. 103, H. 4, S. 4 8 1 - 4 9 7 .
Endress, P.K. 1986a. Reproductive structures and phylogenetic significance of extant
primitive angiosperms. - Plant Syst. Evol., vol. 152, nos. 1/2, p . 1-28.
Endress, P.K. 1986b. Floral structure, systematics, and phylogeny in Trochodendrales. -
Ann. M o . Bot. Gard., vol. 7 3 , no. 2, p. 297-324.
Endress, P.K. 1987. T h e Chloranthaceae: reproductive structures and phylogenetic po­
sition. - Bot. Jb. (Syst.), Bd. 109, S. 153-226.
Endress, P.K., 1978. Bllitenontogenese, Blutenabgrenzung und systematische Stellung
der perianthlosen H a m a m e l i d o i d e a e . - Bot. Jb. (Syst.), Bd. 100, S. 2 4 9 - 3 1 7 .
Engler, A., Prantl, K. (eds.) 1 8 9 7 - 1 9 1 5 . Die naturlichen Pflanzenfamilien. Leipzig:
Engelmann.
Ettingshausen, C. 1868. Die fossile Flora der alteren Braunkohlenformation der Wetterau.
- Sitz.-Ber. Kgl. Akad. Wiss. Wien, Bd. 57, S. 8 0 7 - 8 9 3 .
Eyde, R.H. 1971. Evolutionary morphology: distinguishing ancestral structure from de­
rived structure in flowering plants. - Taxon, vol. 20, no. 1, p . 6 3 - 7 3 .
Faegri, K., van der Pijl, L. 1966. T h e principles of pollination ecology. P e r g a m o n Press,
248 p.
Farris, J.S. 1989. T h e retention index and homoplasy excess ratio. - Syst. Zool., vol. 38,
p. 4 0 6 - 4 0 7 .
Favre-Duchartre, M . 1956. Contribution a Tetude d e la reproduction c h e z le Ginkgo
biloba. - Rev. Cytol. Biol. Veget., t. 17, fasc. 1/2, p. 1-214.
Favre-Duchartre, M . 1965. A propos des gametophytes femelles et des archegones des
plantes ovulees. - A n n . Sci. Nat., Bot. & Biol. Veget., t. 6, fasc. 1, p. 157-182.
Favre-Duchartre, M . 1979. Time relations and sexual reproduction in Cichorium and
other angiosperms as c o m p a r e d with Archegoniates. - Phytomorphology, vol. 29,
no. 2, p. 166-178.
Fiordi, A . C . 1984. F i n e structure of the stigmatic m i c r o p y l e of Larix leptolepis. -
Phytomorphology, vol. 34, nos. 1/4, p. 110-117.
Florin, R. 1931. Untersuchungen zur Stammesgeschichte der Coniferales u n d Cordaitales,
L - Kgl. Svensk. Vetensk.-Akad. Handl., Bd. 10, S. 1-588.
Florin, R. 1936. Die fossilen Ginkgophyten von Franz-Joseph L a n d nebst Erorterungen
uber vermeintliche Cordaitales mesozoischen Alters. 1 , 2 . - Palaeontographica B,
Bd. 8 1 , S. 7 1 - 7 1 3 ; Bd. 82, S. 1-72.
Florin, R. 1938-1945. Die Koniferen des Oberkarbons und des unteren P e r m s , 1-8. -
Palaeontographica B , Bd. 85, S. 1-729.
Fontaine, W . M . 1899. T h e P o t o m a c or y o u n g e r M e s o z o i c flora. - U . S . G e o l . Surv.
Monogr., vol. 15, p. 1-377.
Foster, A.S., Gifford, E . M . (Jr.) 1974. C o m p a r a t i v e m o r p h o l o g y of vascular plants. San
Francisco: Freeman, 751 p.
Francis, J.E. 1983. T h e dominant conifer of the Jurassic Purbeck Formation, England. -
Palaeontology, vol. 26, p . 277-294.
Friis, E . M . 1983. Upper Cretaceous (Senonian) floral structures of Juglandalean affin­
ity containing Normapolles pollen. - Rev. Palaeobot. Palynol., vol. 39, p. 161-188.
Friis, E . M . 1985. Structure and function in Late Cretaceous angiosperm flowers. - Kgl.
Dansk. Videnskab. Selskab. Biol. Skr., vol. 2 5 , p. 1-37.
Friis, E.M., Crane, P.R., Pedersen, K.R. 1986. Floral evidence for Cretaceous chloranthoid
angiosperms. - Nature, vol. 320, no. 6 0 5 8 , p. 163-164.
Friis, E.M., Crane, P R . , Pedersen, K.R. 1988. Reproductive structures of Cretaceous
Platanaceae. - Biol. Skr., vol. 3 1 , p. 1-53.
Friis, E.M., Pedersen, K.R., Crane, P R . 1994. A n g i o s p e r m floral structures from the
Early Cretaceous of Portugal. - Plant Syst. Evol., Suppl., vol. 8, p . 31-50.
Friis, E.M., Pedersen, K.R., Crane, P R . 1995. Apomattoxia ancistrophoragtn. et sp.
nov., a n e w Early C r e t a c e o u s plant with similarities to Circaeaster and extant
Magnoliidae. - Amer. J. Bot., vol. 82, no. 7, p. 9 3 3 - 9 4 3 .
Friis, E.M., Skarby, A. 1982. Scandianthus gen. nov., angiosperm flowers of Saxifragalean
affinity from the U p p e r Cretaceous of Southern S w e d e n . - A n n . Bot. (Gr. Brit.),
N . S . , vol. 50, p. 5 6 9 - 5 8 3 .
Fu Li-kuo (ed.) 1992. C h i n a Plant Red Data Book. Rare and endangered plants, vol. 1.
Beijing, N e w York: Science Press, 741 p.
G a o Rui-qi 1982. Evolutionary trends of Cretaceous a n g i o s p e r m o u s pollen from the
Songliao Basin, N E China. - Acta Palaeontol. Sin., vol. 2 1 , no. 2, p . 2 1 7 - 2 2 4 .
Gensel, P.G., Skog, J.E. 1977. T w o early Mississippian seeds from the Price Formation
of southwestern Virginia. - Brittonia, vol. 29, no. 3, p . 3 3 2 - 3 5 1 .
G i l l e s p i e , W.H., Pfefferkorn, H.W. 1986. Taeniopteroid l a m i n a on Physmatocycas
megasporophylls (Cycadales) from the L o w e r Permian of Kansas, U.S.A. - Rev.
Palaeobot. Palynol. vol. 4 9 , nos. 1/2, p. 99-116.
Gillespie, W.H., Rothwell, G.W., Scheckler, S.E. 1981. T h e earliest seeds. - Nature, vol.
2 9 3 , no. 5 8 3 2 , p. 4 6 2 - 4 6 4 .
Goethe, J.W. von. 1790. Versuch die Metamorphose der Pflanzen zu Erklaren. Gotha, 86 S.
Georgiev, G.N. 1985. M o r p h o g e n i c and cytoembryological studies in Taxus baccata L.
I. D e v e l o p m e n t of m a l e reproductive structures. - A n n u . Sofia Univ., 1981-1982,
vol. 7 5 , p . 21-30 (in Bulgarian).
Gerasimova-Navashina, E.N. 1968. D o u b l e fertilization in a n g i o s p e r m s and s o m e theo­
retical aspects. - Materials of All-Union Conf. on Plant Embryol., Kiev: N a u k o v a
D u m k a , p. 4 4 - 4 9 (in Russian).
Goldberg, A. 1986. Classification, evolution and phylogeny of the families of Dicotyle­
d o n s . - Smiths. Contrib. Bot., no. 58, p. 1-314.
Golovneva, L . B . 1995. Environmental c h a n g e and the patterns of floral evolution dur­
ing the Cretaceous-Tertiary transition in northeastern Asia. - Paleonotol. z h u m .
( M o s c o w ) , vol. 29, no. 2A, p. 36-49.
Gottlieb, L . D . 1984. Genetics and morphological evolution in plants. - Amer. Natur.,
vol. 123, no. 5, p . 6 8 1 - 7 0 9 .
Gould, R.E., Delevoryas, T. 1977. T h e biology of Glossopteris: evidence from pertrified
seed-bearing and pollen-bearing organs. - Alcheringa, vol. 1, p . 3 8 7 - 3 9 9 .
Grant, V. 1975. Genetics of flowering plants. N.Y., London: Columbia Univ. Press, 514 p.
Grauvogel-Stamm, L. 1978. La flore du ges a Voltzia (Bundsandstein superieur) des Vosges
du nord (France) - M e m . Univ. L. Pasteur Strasb., Geol. Sci., no. 50, p. 1-226.
Grimsky, N., Hohn, T., Davies, J.W., Hohn, B . 1987. Agrobacterium-mediaied delivery
of infectious maize streak virus into maize plants. - Nature, vol. 3 2 5 , no. 7000, p.
177-179.
Grove, G.C., Rothwell, G.W. 1980. Mitrospermum vinculum sp. nov., a cardiocarpalean
ovule from the U p p e r Pennsylvanian of O h i o . - Amer. J. Bot., vol. 6 7 , p . 1051-
1058.
Guedes, M., Dupuy, P. 1983. F r o m decussation to distichy, with s o m e c o m m e n t s on
current theories of phyllotaxy. - Bot. J. Linn. S o c , vol. 87, no. 1, p. 1-12.
Guignard, J.-L. 1984. T h e development of cotyledon and shoot apex in monocotyledons.
- C a n a d . J. Bot., vol. 62, p. 1316-1318.
Gupta, K.M. 1958. Williamsonia fructifications from the Jurassic of Rajmahal Hills - their
preservation and plan of construction. - J. Palaeont. Soc. Ind., vol. 3, p. 230-232.
Hall, J.M. 1963. M e g a s p o r e s and other fossils in the D a k o t a Formation (Cenomanian)
of Iowa (U.S.A.). - Pollen et Spores, vol. 5, p . 4 2 5 - 4 4 3 .
Hallier, H. 1912. L'origine et la systeme phyletique des a n g i o s p e r m e s . - Arch. Neer.
Sci., vol. l , p . 146-234.
Harris, T.M. 1932. T h e fossil flora of Scoresby Sound, East Greenland. Caytoniales and
Bennettitales. - M e d d . Groenl., vol. 85, no. 5, p. 1033.
Harris, T.M. 1946. A coprolite of Caytonia pollen. - Ann. M a g . Nat. Hist., Sen, vol. 12,
p. 3 5 7 - 3 7 8 .
Harris, T.M. 1951a. T h e relationships of the Caytoniales. - Phytomorphology, vol. 1,
nos. 1/2, p. 29-39.
Harris, T.M. 1951b. T h e fructification of Czekanowskia and its allies. - Phil. Trans. Roy.
Soc. L o n d o n , vol. 2 3 5 B , p. 4 8 3 - 5 0 8 .
Harris, T.M. 1953. Conifers of the Taxodiaceae from the Wealden Formation of Bel­
gium. - M e m . Inst. Roy. Sci. Nat. Belg., t. 126.
H a r r i s , T . M . 1964. T h e Y o r k s h i r e J u r a s s i c flora. 2. C a y t o n i a l e s , C y c a d a l e s and
Pterridosperms. L o n d o n : Brit. M u s . (Nat. Hist.), 191 p.
Harris, T.M. 1969. T h e Yorkshire Jurassic flora. 3. Bennettitales. L o n d o n : Brit. Mus.
(Nat. Hist.), 186 p.
Harris, T.M. 1979. T h e Yorkshire Jurassic flora. 5. Coniferales. L o n d o n : Brit. Mus.
(Nat. Hist.), 166 p.
Harris, T.M. 1983. Fossils from N e w Zealand ascribed to the Pentoxylon plant. - Phyta,
D . D . Pant C o m m e m o r i a l vol., p. 9 1 - 1 0 3 .
Harris, T.M., Millington, W., Miller, J. 1974. T h e Yorkshire Jurassic flora. 4. Ginkgoales,
Czekanowskiales. London: Brit. M u s . (Nat. Hist.), 150 p .
Hecker, R.F. 1948. T h e Karatau locality of fauna and flora of Jurassic age. - Trudy
Paleont. inst. Akad. nauk S S S R ( M o s c o w ) , t. 15 (1) (in Russian).
Heel, W.A. von. 1981. A S E M investigation of the development of free carpels. - Blumea,
vol. 27, p.499-522.
Heer, O. 1869. Contributions to the fossil flora of North Greenland. - Phil. Trans. Roy.
Soc. L o n d o n , vol. 2 B , p. 39-56.
Heer, O. 1876. Beitrage zur Jura-Flora Ostsibiriens und des A m u r l a n d e s . - M e m . Acad.
I m p . Sci. St.-Petersbourg, t. 22, p. 1-122.
Heer, O. 1878. Beitrage zur fossilen Flora Sibiriens und des A m u r l a n d e s . - M e m . Acad.
I m p . Sci. St.-Petersbourg, ser. 2 5 , t. 7, p. 1-58.
Heer, O. 1880. Nachtrage zur fossilien Flora Gronlands. - Flora fossilis arctica, Zurich,
Bd. 6. S. 1-224.
Heinsbroek, P. G., van Heel, W.A. 1969. Note on the bearing of the pattern of vascular
bundles on the morphology of the stamens of Victoria amazonica (Poepp.) S o w e r b .
- Proc. Kon. Nederl. Akad. Wetensch., C , vol. 72, p. 4 3 1 - 4 4 4 .
Hennig, W. 1950. Grundztige einer Theorie der phylogenetischen Systematik. Berlin:
Deutscherzentralverlag (English edition: 1966, Univ. Illinis Press).
Henslow, G. 1911. T h e origin of m o n o c o t y l e d o n s from dicotyledons through self-adap­
tation to a moist or aquatic habitat. - Ann. Bot. (Gr. Brit.), vol. 26, p . 7 1 7 - 7 4 4 .
Herendeen, P.S., Crepet, W.L., Nixon, K.C. 1993. Chloranthus-Wkt stamen from the
U p p e r Cretaceous of N e w Jersey. - Amer. J. Bot., vol. 80, p . 8 6 5 - 8 7 1 .
H e m g r e e n , G.E.W., Chlonova, A.F. 1981. Cretaceous microfloral provinces. - Pollen et
Spores, vol. 2 3 , p . 4 4 1 - 5 5 5 .
Herzfeld, S. 1922. Ephedra campylopoda Mey. M o r p h o l o g i c der weiblichen Bliite und
Befruchtungsvorgang. - Denkschr. Akad. Wiss. Wien, Bd. 9 8 , S. 1-26.
Herzfeld, S. 1927. Beitrage zur Kenntnis von Ginkgo. - J b . Wiss. Bot., Bd. 6 6 , H. 5, S.
814-818.
Heslop-Harrison, J. 1976. T h e adaptive significance of the exine. - Linn. Soc. S y m p .
Sen, L o n d o n : A c a d e m i c Press, no. 1, p. 27-37.
Heslop-Harrison, Y., Heslop-Harrison, J., Reger, B.J. 1985. T h e pollen-stigma interac­
tion in the grasses. 7. Pollen-tube guidance and the regulation of tube n u m b e r in
Zea mays L. - Acta Bot. Neerl., vol. 34, no. 2, p. 193-211.
Hesse, M . 1984. PoUenkitt is lacking in Gnetatae: Ephedra and Welwitschia, further proof
for its restriction to the angiosperms. - Plant Syst. Evol., vol. 144, no. 1, p. 9-16.
Hickey, L.J. 1973. Classification of the architecture of dicotyledonous leaves. - Amer. J.
Bot., vol. 60, p. 17-33.
Hickey, L.J., Doyle J.A. 1977. Early Cretaceous fossil evidence for a n g i o s p e r m evolu­
tion. - Bot. Rev., vol. 4 3 , no. 1, p.3-104.
Hill R. 1980. T h r e e n e w E o c e n e cycads from eastern Australia. - Austral. J. Bot. vol. 28,
p. 105-122.
Hochuli, P A . 1979. Ursprung und Verbreitung der Restionaceen. - Vierteiljahrschr. naturf.
G e s . Zurich, Bd. 124, S. 109-131.
H o l m e s , W . B . 1974. On s o m e fructifications of the Glossopteridales from the U p p e r
Permian of N.S.W. - Proc. Linn. Soc. N.S.W., vol. 9 8 , no. 3, p. 1 3 1 - 1 4 1 .
Horhammer, L. 1933. U b e r die Coniferen-Gattung CheirolepisSchimper undHirmeriella
nov. gen. aus d e m Rhat - Lias von Franken. - Bibl. Bot., Bd. 27, S. 1-33.
Howard, R.A. 1974. T h e stem-leaf continuum of the Dicotyledoneae. - J. Arnold Ar­
bor., v o l . 5 5 , no. 2, p. 125-181.
Hu Zheng-hai, Tian Lan-xin. 1985. Studies on morphology of Kingdonia unifora F.
Balfour et W.W. Smith: 3. T h e morphology and a n a t o m y of flowers, fruits and
seeds. - Acta Phytotaxon. Sin., vol. 2 3 , p. 170-178.
Hughes, N.F. 1961a. Fossil evidence and angiosperm ancestry. - Sci. Progr., vol. 49, p.
84-102.
Hughes, N.F. 1961b. Further interpretation of Eucommiidites E r d t m a n , 1948. - Palaeon­
tology, vol. 4, p. 2 9 2 - 2 9 9 .
Hughes, N.F. 1976. P a l e o b i o l o g y of angiosperm origins. Cambridge: Univ. Press, 242 p.
Hughes, N . F 1994. T h e e n i g m a of angiosperm origins. C a m b r i d g e : Univ. Press, 303 p.
H u g h e s , N.F., M c D o u g a l l , A . B . 1987. Record of angiospermid pollen entry into the
English Early Cretaceous succession. - Rev. Palaeobot. Palynol., vol. 50, no. 3, p.
255-272.
Hutchinson, J. 1959. T h e families of flowering plants. 2nd edition. Oxford: Clarendon
Press, 9 0 3 p.
Hutchinson, J. 1969. Evolution and phylogeny of flowering plants. Facts and theory.
London, N.Y.: A c a d e m i c Press, 717 p.
Ibrahim, M.I.A. 1996. Aptian-Turonian palynology of the Ghazalat-I Well (GTX-I),
Qattara Depression, Egypt. - Rev. Paleobot Palynol, vol. 9 4 , p. 137-168.
Inamdar, J.A., M o h a n , J.S.S., Bagavathi, S.R. 1986. Stomatal classification - a review.
- Feddes Rep., vol. 97, nos. 3/4, p. 147-160.
Johnson, R.W., Riding, R.T. 1981. Structure and ontogeny of the stomatal complex in
Pinus strohus L. and P hanksiana L a m b . - Amer. J. Bot., vol. 6 8 , no. 2, p . 260-268.
Johri, B . M . ( e d . ) . l 9 8 4 . Embryology of angiosperms. Berlin etc.: Springer, 8 3 0 p.
Johri, B.M., A m b e g a o k a r , K . B . 1984. S o m e unusual features in the e m b r y o l o g y of
angiosperms. - Proc. Ind. Acad. Sci., Plant. Sci., vol. 9 3 , n o . 3, p. 4 1 3 - 4 2 7 .
J o n e s , J.H., Dilcher, D . L . 1986. A study of the ''Dryophyllum'' leaf f o r m s from the
P a l e o g e n e of s o u t h e a s t e r n N o r t h A m e r i c a . - P a l a e o n t o g r a p h i c a B , B d . 2 0 8 , S.
53-80.
Jung, W. 1967. F i n e neue Rekonstruktion der Fruchtzapfens von Cheirolepis muensteri
Schimper. - Neues Jb. Geol. Palaont. Mh., H. 2, S. 111-114.
Kanta, K. 1962. Morphology and embryology of Piper nigrum L. - Phytomorphology,
vol. 12, no. 3, p. 2 0 7 - 2 2 1 .
Kaplan, D.R. 1970. C o m p a r a t i v e foliar h i s t o g e n e s i s in Acorus calamus and its bearing
on the p h y l l o d e t h e o r y of m o n o c o t y l e d o n o u s l e a v e s . - A m e r . J. Bot., vol. 5 7 , p.
331-361.
K e d v e s , M . J . 1982. U l t r a s t r u c t u r a l i n v e s t i g a t i o n of t h e e a r l y Normapolles taxa
Complexiopollis and Limaipollenites. - Palynology, vol. 6, p. 149-159.
Kedves, M . 1983. L a stratification de Texine et la m o r p h o l o g i c des Normapolles. -
Physio-Geo, t. 6, p. 53-67.
Kedves, M.J., Pardutz, A. 1983a. Scanning electron microscopy of s o m e selected recent
amentiflorae pollens. 2. - Acta Biol. Szeged., t. 29, fasc. 1/4, p . 6 7 - 7 6 .
K e d v e s , M.J., P a r d u t z , A . 1983b. Electron m i c r o s c o p e i n v e s t i g a t i o n s of the early
Normapolles pollen genus Atlantopollis. - Palynology, vol. 7, p. 153-169.
Keng, H. 1965. Report on an abnormal ovulate strobilus of Gnetum gnemon L. - Gard.
Bull. Singapore, vol. 2 1 , no. 2, p. 213-214.
Kerp, J.H.F. 1983. Aspects of Permian palaeobotany and palynology. 1, Sohernheimia
Jonkeri nov. gen. sp., a new fossil plant of cycadalean affinity from the W a d e m e r
G r u p p e of S o b e m h e i m . - Rev. Palaeobot. Palynol., vol. 38, nos. 3/4, p. 173-184.
Kerp, J.H.F. 1986. On some interesting fructifications from the L o w e r Permian of the
Saar-Nahe Area (Western G e r m a n y ) . - Cour. Forsch.-Inst. Senckenberg, Bd. 86, S.
73-87.
Kerp, H. 1996. Post-Variscan Late Paleozoic Northern H e m i s p h e r e g y m n o s p e r m s : the
onset to the M e s o z o i c . - Rev. Palaeobot. Palynol., vol. 90. p. 2 6 3 - 2 8 5 .
Khan, R. 1940. A note on "double fertilization" in Ephedra foliata. - Curr. Sci., vol. 9,
p. 323-324.
Khlonova, A.F. 1961. Spores and pollen from the upper part of the Late Cretaceous in
the eastern part of the West-Siberian Plain. Novosibirsk: Siberian Branch, Akad.
nauk S S S R , 138 p. (in Russian).
K h l o n o v a , A.F. 1974. P a l y n o l o g y of the C r e t a c e o u s in S i b e r i a and the F a r East.
Novosibirsk: Nauka, 166 p. (in Russian).
Kimura, T., Sekido, S. 1975. Nilssoniocladus n. gen. (Nilssoniaceae n. fam.) newly
found from the early L o w e r Cretaceous of Japan. - Palaeontographica B , Bd. 153,
H. 1/3, S. 111-118.
Kirchheimer, F. 1957. Die L a u b g e w a c h s e der Braunkohlenzeit. Halle; K n a p p , 672 S.
Klaus, W. 1979. Zur entwicklungsgeschichtlichen B e d e u t u n g triadischer angiospermider
Pollenapertur und Strukturanlagen. - Beitr. Palaontol. Osterr., Bd. 6, S. 135-177.
Knobloch, E. 1960. Paleogenni flora z C e s k e h o C h l o u m k u u Karlovych Var. - S b o m .
ustr. list. Geol., sv. 27, s. 101-155.
Knobloch, E. 1978. On s o m e primitive angiosperm leaves from the U p p e r Cretaceous
of the B o h e m i a n Massif. - Palaeontographica B , Bd. 166, S. 8 3 - 9 8 .
Knobloch, E. 1983. Die Gattung Costatheca Hall in der mitteleuropaischen Kreide. -
Paleontol. S b o m . Geol. Ved, sv. 24, s. 9 5 - 1 0 5 .
Knobloch, E., Mai, D.H. 1986. M o n o g r a p h i c der Friichte und S a m e n in der Kreide von
Mitteleuropa. - Rozpr. ustred. ustavu Geol., sv. 4 7 , s. 1-219.
Knoll, F. 1948. Bau, Entwicklung und morphologische B e d e u t u n g unifazialer Verlau-
ferspitzen an M o n o k o t y l e n b l a t t e m . - Osterr. Bot. Ztschr., Bd. 9 5 , S. 163-193.
Knowlton, F.H. 1930. T h e flora of the D e n v e r and associated formations of Colorado. -
U . S . Geol. Surv. Profess. Pap., vol. 155, p. 1-142.
Kohn, J.R., Waser, N . M . 1985. T h e effect of Delphinium nelsonii pollen on seed set in
Ipomopsis aggregata, a competitor for h u m m i n g b i r d pollination. - Amer. J. Bot.,
v o l . 7 2 , no. 7, p. 1144-1148.
Komar, G.A. 1972. On the nature of the lower ovary. - Bot. Z h u m . (Leningrad), vol. 57,
p. 5 9 5 - 6 0 4 (in Russian).
Konar, R., Oberoi, Y.R 1969. Recent work on reproductive structure of living conifers
and taxads - a review. - Bot. Rev., vol. 3 5 , n o . 2, p. 89-117.
Krassilov, V.A. 1967. Early Cretaceous flora of Southern Primorye and its stratigraphic
significance. M o s c o w : Nauka, 264 p. (in Russian).
Krassilov, V.A. 1969. Approach to the classification of M e s o z o i c " G i n k g o a l e a n " plants
from Siberia. - Palaeobotanist, vol. 18, p . 12-19.
Krassilov, V.A. 1972a. Phytogeographical classification of M e s o z o i c floras and their
bearing on continental drift. - Nature, vol. 237, p. 4 9 .
Krassilov, V.A. 1972b. M e s o z o i c flora of B u r e y a River. G i n k g o a l e s and Czekanow­
skiales. Moscow, Nauka, 152 p. (in Russian).
Krassilov, V.A. 1972c. Paleoecology of terrestrial plants. Vladivostok: Far East Branch,
Akad. nauk S S S R , 208 p. (in Russian; English translation: 1975, N.Y.: Wiley).
Krassilov, V A . 1973a. Mesozoic plants and the problem of angiosperm ancestry. - Lethaia,
v o l . 6 , p. 163-178.
Krassilov, V A . 1973b. T h e Jurassic disseminules with pappus and their bearing on the
problem of angiosperm ancestry. - Geophytology, vol. 3, p. 1-4.
Krassilov, V.A. 1973c. Cuticular structures of C r e t a c e o u s a n g i o s p e r m s from the Far
East of the U S S R . - Palaeontographica B , Bd. 142, S. 105-116.
Krassilov, V.A. 1973d. Climatic changes in Eastern Asia as indicated by fossil floras. 1.
Early Cretaceous. - Palaeogeogr., Palaeoclimatol., Palaeoecol., vol. 13, p. 261-273.
Krassilov, V A . 1975a. Dirhopalostachyaceae - a new family of p r o a n g i o s p e r m s and its
bearing on the problem of angiosperm ancestry. - Palaeontographica B , Bd. 153, S,
100-110.
Krassilov, V.A. 1975b. Climatic changes in Eastern Asia as indicated by fossil floras. 2.
Late Cretaceous and Danian. - Palaeogeogr., Palaeoclimatol., Palaeoecol., vol. 17,
p. 157-172.
Krassilov, V.A. 1976. T h e Tsagajan flora of the A m u r Province. M o s c o w : N a u k a , 92 p.
(in Russian).
Krassilov, V.A. 1977a. The origin of angiosperms. - Bot. Rev., vol. 4 3 , no. 1, p. 143-176.
Krassilov, V.A. 1977b. C o n t r i b u t i o n s to the k n o w l e d g e of the C a y t o n i a l e s . - Rev.
Palaeobot. Palynol., vol. 24, p. 155-178.
Krassilov, V.A. 1978. Bennettitalean stomata. - Palaeobotanist, vol. 2 5 , p . 179-184.
Krassilov, V.A. 1979. Cretaceous flora of Sakhalin. M o s c o w : Nauka, 182 p. (in Russian).
Krassilov, V.A. 1981. C h a n g e s of M e s o z o i c vegetation and the extinction of dinosaurs.
- Palaeogeogr., Palaeoclimatol., Palaeoecol., vol. 34, p. 2 0 7 - 2 2 4 .
Krassilov, V.A. 1982a. O n the ovuliferous organ of Hirmerella. - Phyta, D . D . Pant
C o m m e m o r i a l vol., p. 141-144.
Krassilov, V.A. 1982b. Early Cretaceous flora of M o n g o l i a . - Palaeontographica B , Bd.
181, S. 1-43.
Krassilov, V.A. 1984a. N e w p a l e o b o t a n i c a l d a t a on origin and early e v o l u t i o n of
angiospermy. - Ann. M o . Bot. Gard., vol. 7 1 , p . 5 7 7 - 5 9 2 .
Krassilov, V.A. 1984b. Albian-Cenomanian flora of the Kacha-Bodrak Intershed, Cri­
mea. - Byull. M o s k . obshch. ispyt. prir., Geol., t. 59, no. 4, p . 104-112 (in Russian).
Krassilov, V.A. 1985. Cretaceous Period. Evolution of earth crust and biosphere. M o s ­
c o w : N a u k a , 2 4 0 p . (in Russian).
Krassilov, V.A., 1986. N e w floral structure from the L o w e r Cretaceous of L a k e Baikal
area. - Rev. Palaeobot. Palynol., vol. 4 7 , p. 9-16.
Krassilov, V A. 1989a. Origin and early evolution of angiosperms. M o s c o w : Nauka,
2 6 3 p. (in Russian).
Krassilov, V.A. 1989b. Correlation of the Cretaceous volcanogenic strata in the Soviet
Far East - problems and perspectives. - In: Krassilov, V.A. (ed.). Volcanic Creta­
ceous of the Far East. Vladivostok: Far East Branch, Akad. nauk S S S R , p. 6-13 (in
Russian).
Krassilov, V A . 1990. Fossil links r e c o n s i d e r e d . - P r o c . 3rd l O P Conf., M e l b o u r n e ,
p. 7-10.
Krassilov, V.A. 1991. T h e origin of angiosperms: new and old p r o b l e m s . - T R E E , vol. 6,
p. 2 1 5 - 2 2 0 .
Krassilov, V.A. 1994. Global climatic changes as a factor of biosphere evolution. - In:
E c o s y s t e m restructurings and the evolution of the biosphere, 1. M o s c o w : Nedra, p.
2 8 5 - 2 9 4 (in Russian).
Krassilov, V.A. 1995. Scytophyllum and the origin of angiosperm leaf characters. - Paleont.
Zh. ( M o s c o w ) , vol. 29, N o I A, p . 63-73
Krassilov, V.A., Ash, S. 1988. On Dinophyton - p r o t o g n e t a l e a n M e s o z o i c plant. -
Palaeontographica B , Bd. 2 0 8 , S. 3 3 - 3 8 .
Krassilov, V.A., Blokhina, N.I., Markevich, V S . 1987. T h e C r e t a c e o u s - P a l e o g e n e of
the Lesser Kuril Islands. Vladivostok: Far East Branch, Akad. nauk S S S R , 140 p.
(in Russian).
Krassilov, V.A., B u g d a e v a , E . V 1982. Achene-like fossils from the L o w e r Cretaceous
of the Lake Baikal area. - Rev. Palaeobot. Palynol., vol. 36, p. 2 7 9 - 2 9 5 .
Krassilov, V.A., B u g d a e v a , E . V 1988a. Gnetalean plants from the Jurassic of Ust-Balej,
East Siberia. - Rev. Palaeobot. Palynol., vol. 5 3 , p. 3 5 9 - 3 7 4 .
Krassilov, V.A., B u g d a e v a , E.V. 1988b. Meeusella and the origin of stamens. - Lethaia,
vol. 2 1 , p. 4 2 5 - 4 3 1 .
Krassilov, V.A., Bugdaeva, E.V. 1988c. Protocycadopsid pteridosperms from the L o w e r
Cretaceous of Transbaikalia and the origin of cycads. - Palaeontographica B , Bd.
2 0 8 , S. 2 7 - 3 2 .
K r a s s i l o v , V.A., B u g d a e v a , E.V. 1997. B a i s i a n a s s e m b l a g e of p r o a n g i o s p e r m s . -
Palaeontographica B (in press).
Krassilov, V.A., B u r a g o , V.I. 1981. N e w interpretation of Gaussia (Vojnovskyales). -
Rev. Palaeobot. Palynol., vol. 32, p. 2 2 7 - 2 3 7 .
Krassilov, V.A., Delle, G.V., Vladimirova, H.V. 1996. A new Jurassic pollen c o n e from
Georgia and its bearing on cycad phylogeny. - P a l a e o n t o g r a p h i c a B , 1996, Bd.
238,8.71-75.
Krassilov, V.A., Dilcher, D.L., Douglas, J.G., 1996. Ephedroid plant from the Lower
Cretaceous of Koonwarra, Australia. - Abstr. 5th Conf. Int. Org. Paleobot., Santa
Barbara, Calif., p. 54.
Krassilov, V.A., Dilcher, D.L., Douglas, J.G. 1997. N e w ephedroid plant from the Lower
Cretaceous of Koonwarra, Australia. - Alcheringa (in press).
Krassilov, V.A., Dobruskina, I.A. 1995. Angiosperm fruit from the L o w e r Cretaceous
of Israel and origins in rift valleys. - Paleont. Zh. ( M o s c o w ) , vol. 2 9 , N o . 1 A, p.
110-115.
Krassilov. V.A., Fotyanova, L.I. 1995. A new hamamelid fructification from the Paleogene
of K a m c h a t k a , with c o m m e n t s on Trochodendrocarpus and Nordenskioldia. -
Paleonotol. Zh. ( M o s c o w ) , vol. 29, no. 1 A, p. 8 7 - 9 5 .
Krassilov, V.A., G o l o v n e v a , L.B., Nessov, L.A. 1990. A c y c a d o p h y t e from the Late
Cretaceous dinosaur locality in northern Koryakia. - In: Krassilov, V.A. (ed.). Con­
tinental Cretaceous of the U S S R . Vladivostok: Far East Branch, A k a d . nauk SSSR,
p. 213-215 (in Russian).
Krassilov, V.A., Makulbekov, N . M . 1995. Maastrichtian aquatic plants from Mongolia.
- Paleonotol. Zh. ( M o s c o w ) , vol. 29, no. 2A, p. 119-140.
Krassilov, V.A., Martinson, H.H. 1982. Fruits from the Late Cretaceous of Mongolia. -
Paleontol Zh. ( M o s c o w ) , no. 1, p. 113-121 (in Russian).
K r a s s i l o v , V.A., P a c l t o v a , B . 1989. Asterocelastrus cretacea, a mid-Cretaceous
angiosperm from Bohemia. - Rev. Paleobot. Palynol., vol. 6 0 , p . 1-6.
Krassilov, V.A., Rasnitsyn, A.P. 1982. Unique find: pollen in the guts of Early Creta­
ceous xyelotomid insects. - Paleont. Zh. ( M o s c o w ) , no. 4, p. 83 -96.
Krassilov, V A . , Rasnitsyn, A.P. 1997. Pollen in the guts of Permian insects: first evi­
dence of pollinivory and its evolutionary significance. - Lethaia (in press).
Krassilov, V.A., Raskatova, M.G., Istschenko, A.A. 1987. A n e w Archaeopteridalean
plant from the Devonian of Pavlovsk, U.S.S.R. - Rev. Paleobot. Palynol., vol. 53,
p. 163-173.
Krassilov, V.A., Shilin, P.V 1995. N e w staminoid heads from the mid-Cretaceous of
Kazakhstan. - Rev. Palaeobot. Palynol., vol. 85, p . 2 0 7 - 2 1 1 .
K r a s s i l o v , V.A., S h i l i n , P . V , V a c h r a m e e v , V.A. 1 9 8 3 . C r e t a c e o u s f l o w e r s from
Kazakhstan. - Rev. Palaeobot. Palynol., vol. 4 0 , p. 9 1 - 1 1 3 .
Krassilov. V.A., Shorokhova, S.A. 1970. N e w Triassic plants from the Iman River basin
(Primorye) and s o m e problems of M e s o z o i c pteridophyll morphology. - In: Triassic
invertebrates and plants of the eastern U S S R . Vladivistok: Far East Branch, Akad.
nauk S S S R , p. 98-121 (in Russian).
Krassilov, V.A., Shorokhova, S.A. 1988. Phytostratigraphy of the Alchan Formation in
northern Primorye. In: Krassilov, V.A. (ed.). Volcanic Cretaceous of the Far East.
Vladivostok: Far East Branch, Akad. nauk S S S R , p. 4 3 - 4 9 (in Russian).
Krassilov, V.A., Z a k h a r o v a , T.V. 1995. MoresnetiaAikQ p l a n t s from t h e U p p e r D e ­
v o n i a n of M i n u s i n s k B a s i n , Siberia. - P a l e o n o t o l . z h u r n . ( M o s c o w ) , vol. 2 9 ,
no. lA, p. 35-43.
Krassilov. V.A., Zherikhin, V.V., Rasnitsyn, A.P. 1997. Classopollis in the guts of Jurassic
insects. - Palaeontology, (in press).
Krausel, R. 1956. Zur Geschichte der A n g i o s p e r m e n . - Bot. M a g . (Tokyo), vol. 69, no.
822, p . 537-544.
Kryshtofovich, A . N . 1958. Fossil floras of Penzhin Gulf, Tastakh L a k e , and Rarytkin
R a n g e . - Trudy Bot. inst. Akad. nauk S S S R , ser. 8, Paleobotany, t. 3 , p. 73-121 (in
Russian).
Kubitzki, K. 1973. P r o b l e m e der GroBsystematik der Blutenpflanzen. - Berichte Dtsch.
bot. Ges., Bd. 85, S. 2 5 9 - 2 7 7 .
Kubitzki, K., Gottlieb, O.R. 1984a. M i c r o m o l e c u l a r patterns and the evolution and major
classification of angiosperms. - Taxon, vol. 3 3 , no. 3, p . 3 7 5 - 3 9 1 .
Kubitzki, K., Gottlieb, O.R. 1984b. Phytochemical aspects of a n g i o s p e r m origin and
evolution. - Acta Bot. Neerl., vol. 3 3 , no. 4, p. 4 5 7 - 4 6 8 .
Kugler, H. 1970. Blutenokologie. Jena: G. Fischer, 345 S.
Kupriyanova, L.A. 1976. Pollen morphology in the genus Nymphaea in the European
part of the U S S R . - Bot. Z h u m . (Leningrad), vol. 6 1 , p . 1558-1563 (in Russian).
Kuznetsov, N.L 1936. Introduction to the systematics of flowering plants, 2nd edition.
Leningrad: A k a d nauk S S S R , 4 5 6 p. (in Russian).
Kvacek, Z., Knobloch, E. 1967. Zur Nomenklatur der Gattung Daphnogene Ung. und die
neue Art Daphnogene pannonica sp. n. - Vestn. Ustred. geol., sv. 4 2 , s. 201-210.
L a m , H.J. 1950. Stachyospory and phyllospory as factors in the natural system of the
C o r m o p h y t a . - Svenk. Bot. Tidskr., Bd. 4 4 , H. 4, S. 5 1 7 - 5 3 4 .
Lang, W.H. 1900. Studies in the d e v e l o p m e n t and morphology of C y c a d e a n sporangia.
2. T h e ovule of Stangeria paradoxa, - Ann. Bot. (Gr. Brit.), vol. 14, p . 2 8 1 - 3 0 6 .
Lanyon, S.M. 1994. Polyphyly of the blackbird genus Agelaius and the importance for
assumptions of m o n o p h y l y in comparative studies. - Evolution, vol. 4 8 , no. 3, p.
679-693.
Laubenfels, D.J. de. 1953. T h e extemal morphology of coniferous leaves. - P h y t o m o r p h o -
logy, vol. 3, nos. 1/2, p. 1-19.
Lee, C.L. 1955. Fertilization in Ginkgo biloba. - Bot. Gaz., vol. 117, no. 2, p . 7 9 - 1 0 0 .
Leinfellner, W. 1950. Der Bauplan des synkarpen G y n o z e u m s . - Osterr. bot. Ztschr., Bd.
97, S. 4 0 3 - 4 3 6 .
Leins, P., Erbar, C. 1985. Fin Beitrag zur Blutenentwicklung der Aristolochiaceen, einer
Vermittelgmppe zu den M o n o k o t y l e n . - Bot. J b . (Syst.), B d . 107, S. 3 4 3 - 3 6 8 .
Leroy, J.-F. 1983. T h e origin of angiopsperms: an unrecognized ancestral dicotyledon,
Hedyosmum (Chloranthales), with a strobiloid flower is living today. - Taxon, vol.
32, no. 2, p. 169-175.
Lesquereux, L. 1874. Contributions to the fossil flora of the W e s t e m Territories, I. T h e
Cretaceous flora. - U . S . Geol. Geogr. Surv. Territ. Rept., vol. 6, p . 1-136.
Lesquereux, L. 1876. Palaeontology. - U . S . Geol. Geogr. Surv. Territ. Rept., vol. 8, p.
271-366.
L e s q u e r e u x , L. 1 8 8 3 . C o n t r i b u t i o n s to the fossil flora of the W e s t e r n Territories, 3.
T h e C r e t a c e o u s and Tertiary floras. - U . S . G e o l . G e o g r . Surv. Territ. Rept., vol.
8, p . 1-283.
Li, H., Tian, B , Taylor, E.L., Taylor, T.N. 1994. Foliar a n a t o m y of Gigantonoclea
guizouensis (Gigantopteridales) from the U p p e r P e r m i a n of G u i z h o u Province,
China. - Amer. J. Bot., vol. 2 1 , no. 6, p. 6 7 8 - 6 8 9 .
Li X i n g x u e , Yao Zhaoqi 1983. Fructifications of gigantopterids from South China. -
Palaeontographica B , Bd. 185, nos. 1/3, S. 11-26.
Lillegraven, J.A., K i e l a n - J a w o r o w s k a , Z., C l e m e n s , W . A . (Jr.) 1979. M e s o z o i c mam­
mals: the first two-thirds of m a m m a l i a n history. B e r k e l e y : Univ. California Press,
311 p .
Linnaei, C. 1751. Philosophia botanica. Stockholm.
Linnel, T. 1933. Z u r M o r p h o l o g i c und Systematik triassischer C y c a d o p h y t e n . 2. Uber
Scytophyllum B o m e m . , eine wenig bekannte Cycadophytengattung aus d e m Keuper
- Svensk. Bot. Tidskr., Bd. 27, H. 3, S. 3 1 0 - 3 3 1 .
Long, A . G . 1966. S o m e L o w e r Carboniferous fructifications from B e r w i c k s h i r e , to­
gether with a theoretical account of the evolution of ovules, cupules, and carpels. -
Trans. Roy. Soc. Edinburgh, vol. 66, p. 3 4 5 - 3 7 5 .
Long, A . G . 1969. Eurystoma trigona sp. nov., a pteridosperm ovule borne on a frond of
Alcicarnopteris Kidston. - Trans. Roy. Soc. Edinburgh, vol. 6 8 , p . 171-182.
Long, A.G. 1977a. S o m e L o w e r Carboniferous pteridpsperm cupules bearing ovules
and microsporangia. - Trans. Roy. Soc. Edinburgh, vol. 70, nos. 1/3, p. 1-11.
L o n g , A . G . 1977b. L o w e r C a r b o n i f e r o u s p t e r i d o s p e r m c u p u l e s and the origin of
angiosperms. - Trans. Roy. Soc. Edinburgh, vol. 70, nos. 1/3, p. 13-35.
Long, A . G . 1986. T h e cupule-carpel theory. A defence. - Trans. Roy. S o c . Edinburgh,
v o l . 4 4 , no. 2, p. 2 8 1 - 2 8 5 .
L u c k e , E. 1984. S a m e n s t r u k t u r u n d S a m e n k e i m u n g e u r o p a i s c h e r O r c h i d e e n nach
V E Y R E T sowie weitere Untersuchungen. - Orchidee, Bd. 3 5 , H. 4, S. 153-158.
Macdonald, A . D . 1977. Myricaceae: floral hypothesis for Gale and Comptonia. - Canad.
J. Bot., v o l . 5 5 , p. 2 6 3 6 - 2 6 5 1 .
Madel, E. 1962. Die fossilen Euphorbiaceen-Holzer mit besondere Beriicksichtigung
neuer F u n d e aus der Oberkreide Siid-Afrikas. - Senckenberg. lethaea, Bd. 4 3 , H. 4,
S. 2 8 3 - 3 2 1 .
M a h e s h w a r i , S. C. 1958. Spirodela polyrhizn, the link b e t w e e n the aroids and the
d u c k w e e d s . - Nature, vol. 181, p. 1745-1746.
M a h e s h w a r i , P., Singh, H. 1967. T h e female g a m e t o p h y t e of g y m n o s p e r m s . - Biol. Rev.
C a m b r i d g e Phil. S o c , vol. 4 2 , no. 1, p. 88-130.
M a h e s h w a r i , P., Vasil, V. 1961. Gnetum. N e w Delhi: C o u n c . Sci. & Industr. Res., 142 p.
Maley, J. 1996. T h e African rain forest - main characteristics of changes in vegetation
and climate from the U p p e r Cretaceous to the Quaternary. - In: Alexander, I.J.,
Swaine, M . D . & Watling, R. (eds.). Essays on the ecology of the G u i n e a - C o n g o
rain forest. - Proc. Roy. Soc. Edinburgh, B (Biol. Sci.), vol. 104, p . 31-74.
M a m a y , S.H. 1966. Tinsleya, a n e w g e n u s of seedbearing callipterid plants from the
P e r m i a n of North-Central Texas. - U . S . G e o l . Surv. Profess. P a p . , vol. 5 2 3 , pt. E,
p. 1-15.
M a m a y , S.H. 1968. Russelites, new genus, a problematical plant from the L o w e r Permian
of Texas. - U . S . Geol. Surv. Profess. Pap., vol. 5 9 3 , pt. I, P, p . 1-15.
M a m a y , S.H. 1976. Paleozoic origin of cycads. - U . S . Geol. Surv. Profess. Pap., vol.
934, p. 1-48.
M a m a y , S.H., Watt, A . D . 1971. An ovuliferous cailipteroid plant from the Hermit Shale
( L o w e r Permian) of the Grand C a n y o n , Arizona. - U . S . Geol. Surv. Profess. Pap.,
vol. 750, p t . C , p. 4 8 - 5 1 .
Manchester, S.R. 1986. Vegetative and reproductive m o r p h o l o g y of an extinct plane
tree (Platanaceae) from the E o c e n e of w e s t e m North America. - Bot. Gaz., vol.
147, p. 2 0 0 - 2 2 6 .
M a p e s , G., Rothwell, G.W. 1980. Questora amplecta gen. et sp. n., a stmcturally simple
medullosan stem from the U p p e r Mississippian of Arkansas. - Amer. J. Bot., vol.
67., p . 6 3 6 - 6 4 7 .
M a p e s , G., Rothwell, G.W. 1984. Permineralized ovulate cones of Lebachia from late
Palaeozoic limestones of Kansas. - Palaeontology, vol. 27, p . 69-94.
Martens, P 1971. Les gnetophytes. - H a n d b . Pflanzenanat., Bd. 12, S. 1-295.
Martin, P.G., D o w d , J.M. 1984. T h e study of plant p h y l o g e n y using a m i n o acid se­
quences of ribulose-1,5-bisphosphate carboxylasae. 5. Magnoliaceae, Polygonaceae
and the concept of primitiveness. - Austral. J. Bot., vol. 32, p . 3 0 1 - 3 0 9 .
Martin, P.G., D o w d , J.M. 1986. A phylogenetic tree for s o m e m o n o c o t y l e d o n s and g y m ­
n o s p e r m s derived from protein sequences. - Taxon, vol. 3 5 , no. 3, p . 4 6 9 - 4 7 5 .
Maslova, N . P , Krassilov, V.A. 1997. N e w hamamelid infmctescence from the Paleocene
of Western Kamchatka, Russia. - Rev, Palaeobot. Palynol. (in press).
M a t t e n , L . C . , L a c e y , W.s., L y c e s , R . C . 1980. S t u d i e s on the c u p u l a t e seed g e n u s
Hydrosperma L o n g from Berwickshire and east Lothian in Scotland and County
Kerry in Ireland. - Bot. J. Linn. S o c , vol. 8 1 , p. 2 4 9 - 2 7 3 .
M e d u s , J. 1 9 8 3 . M o r p h o l o g i c , fonction et e v o l u t i o n d e q u e l q u e s g r a i n s d e pollen
Normapolles du Cretacee superieur. - Lethaia, vol. 16, p. 2 6 5 - 2 7 1 .
M e e u s e , A.D.J. 1963. T h e so-called " m e g a s p o r o p h y l l " of Cycas - a morphological mis­
conception. Its bearing on the phylogeny and the classification of the Cycadophyta.
- Acta Bot. N e e d . , vol. 12, no. 1, p . 119-128.
M e e u s e , A.D.J., B o u m a n , F. 1974. T h e inner integument - its probable origin and ho­
mology. - Acta Bot. Neerl., vol. 24, p. 2 3 7 - 2 4 9 .
Mehra, P.N. 1950. Occurrence of hermaphrodite flowers and the d e v e l o p m e n t of female
gametophyte in Ephedra intermedia Shrenk et May. - Ann. Bot. (Gr. Brit.), N.S.,
vol. 14, p. 165-180.
Melville, R. 1963. A new theory of the a n g i o s p e r m flower. - K e w Bull., vol. 17, no. 1,
p. 1-63.
Melville, R. 1976. T h e terminology of leaf architecture. - Taxon, vol. 2 5 , nos. 5/6, p.
549-561.
Merril, E.K. 1986. Heteroblastic seedling of green ash. 2. Early d e v e l o p m e n t of simple
and c o m p o u n d leaves. - C a n a d . J. Bot., vol. 64, p. 2 6 5 0 - 2 6 0 1 .
Mestre, J . - C , G u e d e s , M . 1983. Nature et signification phylogenetique du cotyledon. -
Bull. Soc. bot. Fr., t. 130, p. 7-22.
M e y e n , S. 1981. S o m e true and alleged Permotriassic conifers of Siberia and the Rus­
sian platform and their alliance. - Palaeobotanist, vol. 28/29, p . 161-176.
Meyen, S.V. 1982. Reproductive organs of g y m n o s p e r m s and their evolution. - Z h u m .
obshch. biol. ( M o s c o w ) , vol. 4 3 , p. 303-323 (in Russian).
M e y e n , S.V. 1984. Basic features of g y m n o s p e r m systematics and phylogeny as evi­
denced by fossil record. - Bot. Rev., vol. 50, no. 1, p. 1-111.
Meyen, S.V. 1987. Fundamentals of palaeobotany. London, N.Y.: Chapman & Hall, 432 p.
Meyer, N . R . 1980. Probable directions of evolution in the sporoderm of g y m n o s p e r m s
and s o m e angiosperms. - In: Systematics and evolution of higher plants. Lenin­
grad: Nauka, p. 82-92 (in Russian).
Millay, M.A., Taylor, T.N. 1979. Paleozoic seed fem pollen organs. - Bot. Rev., vol. 4 5 ,
p. 3 0 1 - 3 7 5 .
Moeliono, B . M . 1970. Cauline or carpellary placentation a m o n g dicotyledons. 1. The
cauline ovules of Centrosperms. Assen: Van G o r c u m - P r a k k e , 2 9 2 p .
Mosley, M.F., Uhl, N.W. 1985. Morphological studies of the N y m p h a e a c e a e sensu lato.
15. T h e anatomy of the flower of Nelumbo. - Bot. J b . (Syst.), Bd. 106, S. 6 1 - 9 8 .
Moussel, B . 1978. Double fertilization in the genus Ephedra. - P h y t o m o r p h o l o g y , vol.
28, no. 3, p . 336-344.
M u h a m m a d , A.F., Sattler, R. 1982. Vessel s t m c t u r e of Gnetum a n d the origin of
angiosperms. - Amer. J. Bot., vol. 6 9 , no. 6, p . 1 0 0 4 - 1 0 2 1 .
Muller, H. 1966. Palynological investigations of Cretaceous sediments in northeastem
Brazil. - Proc. 2nd West African Micropaleontol. Colloq., Leiden, p . 123-136.
Muller, J. 1970. Palynological evidence on early differentiation of a n g i o s p e r m s . - Biol.
Rev. C a m b r i g e Phil. S o c , vol. 4 5 , p. 4 1 7 - 4 5 0 .
Muller, J. 1974. A comparison of southeast Asian with E u r o p e a n fossil angiosperm
pollen floras. - S y m p . Orig. & Phytogeogr. A n g i o s p e r m s , L u c k n o w : B . Sahni Inst.
Palaeobot., no. 1, p. 4 9 - 5 6 .
Muller, J. 1984. Significance of fossil pollen for angiosperm history. - A n n . M o . Bot.
Gard., vol., 7 1 , no. 2, p. 4 1 9 - 4 4 3 .
Nambudiri, K.K., Beck, C . B . 1968. A comparative study of the primary vascular sys­
tem of conifers. - Amer. J. Bot., vol. 5 5 , no. 4, p. 4 4 7 - 4 7 2 .
N e g b i , M . , Koller, D . 1962. H o m o l o g i e s in the g r a s s e m b r y o . A r e e v a l u a t i o n . -
Phytomorphology, vol. 12, no. 12, p. 2 8 9 - 2 9 6 .
N e u b u r g , M . F . 1 9 6 3 . F a m i l y V o j n o v s k y a c e a e N e u b u r g . - I n : F u n d a m e n t a l s of
paleontology. G y m n o s p e r m s and angiosperms. M o s c o w : Gostekhizdat, 301 p. (in
Russian).
N e u m a y e r , H. 1924. D i e G e s c h i c h t e der B l u t e . - A b h . zool.-bot. G e s . W i e n , B d . 14,
S. 1-112.
Newberry, J.S. 1895. T h e flora of the A m b o y Clays. - U . S . Geol. Surv. Monogr., no. 26,
p. 1-137.
Nishida, H. 1985. A structurally preserved magnolialean fructification from mid-Creta­
ceous of Japan. - Nature, vol. 318, no. 6 0 4 1 , p. 58-59.
Nishida, H. 1994. Elsemaria, a new Late Cretaceous a n g i o s p e r m fructification from
H o k k a i d o , Japan. - Plant Syst. Evol., Suppl., vol. 8, p . 123-136.
Nishida, H., Nishida, M . 1988. Protomonimia kasai-nakajhongii gen. et sp. nov.: a
permineralized magnolialean fructification from the mid-Cretaceous of Japan. -
Bot. M a g . (Tokyo), vol. 101, p. 397-426.
Nishida, M . 1974. Diagnostic characters between vesselless a n g i o s p e r m o u s w o o d s and
cycadean w o o d s . - S y m p . M o r p h o l . & Stratigr. Palaeobot., L u c k n o w : B . Sahni
Inst. Palaeobot., no. 2, p. 4 7 - 4 9 .
Nishida, M., Nishida, H. 1985. Petrified w o o d s from the Upper Cretaceous of Q u i n q u i n a
Island. A report of the botanical survey to Bolivia and southern Chile. Chiba: C h i b a
Univ., p. 27-34.
N o w i c k e , J.W., Skvarla, J.J. 1979. Pollen morphology: the potential influence in higher
order systematics. - Ann. M o . Bot. Gard., vol. 66, no. 4, p . 6 3 3 - 7 0 0 .
G h a n a , T , K i m u r a , T. 1987. Preliminary notes on the multicarpelous female flower
with conduplicate carpels from the U p p e r Cretaceous of H o k k a i d o , Japan. - Proc.
Jap. Acad., vol. 6 3 B , p . 175-178.
Ohri, D., K h o s h o o , T.N. 1986. G e n o m e size in g y m n o s p e r m s . - Plant Syst. Evol., vol.
153, p. 119-132.
Oishi, S. 1940. T h e M e s o z o i c floras of Japan. - J. Fac. Sci. H o k k a i d o Univ., vol. 5, p.
123-480.
O l d h a m , T.C.B. 1976. Flora of the Wealden plant debris beds of England. - Palaeontol­
ogy, vol. 19, p. 4 3 7 - 5 0 2 .
Orel, L.I., Kupriyanova, L.A., Golubeva, E.A. 1987. Ultrastructure of acetyl-resistant
cell walls of tapetal cells and pollen grains of Gnetum africanum (Gnetaceae). -
Bot. Z h u m . (Leningrad), vol. 7 1 , p. 7 5 0 - 7 5 9 (in Russian).
O s b o r n , J . M . , T a y l o r , T . N . 1 9 9 3 . P o l l e n m o r p h o l o g y a n d u l t r a s t r u c t u r e of t h e
Corystospermales: permineralized in situ grains from the Triassic of Antarctica. -
Rev. Paleobot. Palynol., vol. 7 9 , p . 205-220.
O s b o m , J.M., Taylor, T.N., de Lima, M.R. 1993. T h e ultrastmcture of ephedroid pollen
with gnetalean affinities from the L o w e r Cretaceous of Brazil. - Rev. Paleobot.
Palynol., v o l . 7 7 , p. 171-184.
O s b o r n , J . M . , Taylor, T.N., C r a n e , P.R. 1 9 9 1 . T h e ultrastructure of Sahnia pollen
(Pentoxylales). - Amer. J. Bot., vol. 7 8 , no. 11, p. 1560-1569.
O w e n s , J.N., Blake, M . D . 1984. T h e pollination m e c h a n i s m of Sitka spruce (Picea
sitchensis). - Canad. J. Bot., vol. 62, no. 6, p. 1136-1148.
O y a m a , T , M a t s u o , H. 1964. Notes on palmaean leaf from the Oarai flora (Upper Cre­
taceous), Oarai M a c h i , Ibaraki Prefecture, Japan. - Trans. Palaeont. S o c . Jap., N.S.,
vol. 5 5 , p . 241-246.
Pacltova, B . 1971. Palynological study of A n g i o s p e r m a e from the P e r u c Formation
(?Albian-Lower C e n o m a n i a n ) of B o h e m i a . - Sborn. Geol. Ved, sv. 13, s. 105-141.
Pacltova, B . 1977. Cretaceous angiosperms of B o h e m i a - central E u r o p e . - Bot. Rev.,
vol. 4 3 , no. l , p . 128-142.
Pankow, H., S o t h m a n n , E. 1967. Histogenetische Untersuchungen an den weiblichen
Bluten von Ginkgo biloba L. - Berichte Dtsch. bot. Ges., Bd. 80, H. 4, S. 265-272.
Pant, D . D . 1959. T h e classification of g y m n o s p e r m o u s plants. - Palaeobotanist, vol. 6,
p. 65-70.
Pant, D . D . 1968. On the stem and attachment of Glossopteris leaves. - Phytomorphology,
vol. 17, p. 351-359.
Pant, D . D . 1977. T h e plant of Glossopteris. - J. Ind. Bot. S o c , vol. 5 6 A , p . 1-23.
Pant, D . D . 1982. T h e lower G o n d w a n a g y m n o s p e r m s and their relationships. - Rev.
Palaeobot. Palynol., vol. 37, p. 55-70.
Pant, D.D., Kidwai, P.F. 1978. T h e origin and evolution of flowering plants. - J. Ind.
Bot. S o c , vol. 56A, p. 242-274.
Pant, D . D . , M e h r a , B . 1 9 6 3 . O n a c y c a d o p h y t e leaf, Pteronilssonia gopalii g e n . et
sp. nov. from the L o w e r G o n d w a n a of India. - P a l a e o n t o g r a p h i c a B , B d . 113,
S. 1 2 6 - 1 3 4 .
Pant, D . D . , Nautiyal, D . D . 1963. Cuticle and epidermis of recent C y c a d a l e s leaves,
sporangia and seeds. - Senckenberg. biol., Bd. 4 4 , H. 4, S. 2 5 7 - 3 3 6 .
Pant, D.D., Nautiyal, D . D . 1987. Diphyllopteris verticillata Srivastava, the probable
seedling of Glossopteris from the Paleozoic of India. - Rev. Palaeobot. Palynol.,
vol. 5 1 , nos. 1/3, p . 31-36.
Pauze, F , Sattler, R. 1978. L'androcee centripete d'Ochna atropurpurea. - C a n a d . J.
Bot., vol. 56, no. 20, p. 2 5 0 0 - 2 5 1 1 .
Pauze, F , Sattler, R. 1979. La placentation axillaire chez Ochna atropurpurea. - Canad.
J. Bot., vol. 57, no. 2, p . 100-107.
Pearson, H.H.W. 1929. Gnetales. C a m b r i d g e : Univ. Press, 194 p .
Pedersen, K.R., Friis, E . M . 1986. Caytonanthus pollen from the L o w e r and M i d d l e
Jurassic. - Geoskrifter, vol. 24, p. 2 5 5 - 2 6 6 .
Pedersen, K.R., Friis, E.M., Crane, P.R. 1988. Eucommiidites pollen in situ in Creta­
ceous reproductive structures. - Abstr. 7th Intern. Palynol. Congr., Brisbane, p. 128.
Pellmyr, O., Thien, L . B . 1986. Insect reproduction and floral fragrances: keys to the
evolution of the angiosperms? - Taxon, vol. 3 5 , no. 1, p . 7 6 - 8 5 .
P e n n e l l , R.I., B e l l , P.R. 1986. M i c r o s p o r o g e n e s i s in Tiaxus baccata L . T h e forma­
tion of the tetrad and d e v e l o p m e n t of m i c r o s p o r e s . - A n n . B o t . (Gr. Brit.), vol.
57, p. 545-555.
Penny, J.H.J. 1991. Early Cretaceous angiosperm pollen from the borehole M e r s a Mat-
ruch I, Northwest desert, Egypt. - Palaeontographica B , Bd. 2 2 2 , S. 3 1 - 8 8 .
Periasamy, K., Muraganathan, E.A. 1985. Ontogeny of palmately c o m p o u n d leaves in
angiosperms: 1. Tebebuia pentaphlla H e n s e . - Proc. Ind. A c a d . Sci., Plant Sci.,
vol. 9 5 , no. 6, p. 4 2 9 - 4 3 6 .
Pettitt, J.M. 1969. Pteridophytic features in s o m e L o w e r Carboniferous seed megaspores.
- Bot. J. Linn. S o c , vol. 62, p. 2 3 3 - 2 3 9 .
Pettitt, J.M., Chaloner, W.G. 1964. T h e ultrastructure of M e s o z o i c pollen ClassopoUis.
- Pollen et Spores, vol. 6, p. 611-620.
Philipson, W.R. 1985. Is the grass g y n o e c i u m monocarpellary? - Amer. J. Bot., vol. 72,
no. 2, p. 1 9 5 4 - 1 9 6 1 .
Philomen, P.A., Shah, C.K. 1985. Unusual germination and seedling d e v e l o p m e n t in
t w o m o n o c o t y l e d o n o u s dicotyledons. - Proc. Ind. Acad. Sci., Plant. Sci., vol. 9 5 ,
no. 4, p. 2 2 1 - 2 2 5 .
Pigg, K.B., Stockey, R.A. 1991. Platanaceous plants from the P a l e o c e n e of Alberta,
C a n a d a . - Rev. Palaeobot. Palynol., vol. 70, p. 125-146.
P l a y f o r d , G., R i g b y , J.F., A r c h i b a l d , D . C . 1 9 8 2 . M i d d l e T r i a s s i c flora f r o m the
M o o l a y e m b e r Formation, B o w e n Basin, Q u e e n s l a n d . - Geol. Surv. Q u e e n s l a n d
Publ., v o l . 3 8 0 , p. 1-52.
Pocock, S.A.J., Vasanthy, G. 1988. Cornetipollis reticulata, a new pollen with angio­
spermid features from upper Triassic ( C a m i a n ) sediments of Arizona (U.S.A.) with
notes on Equisetosporites. - Rev. Palaeobot. Palynol., vol. 5 5 , p . 3 3 7 - 3 5 6 .
Pocock, S.A.J., Vasanthy, G., Venkatachala, B.S. 1990. Pollen of Circumpolles - an
e n i g m a of m o r p h o t r e n d s s h o w i n g e v o l u t i o n a r y a d a p t a t i o n . - R e v . P a l a e o b o t .
Palynol., vol. 6 5 , p. 179-193.
Popova, I.S. 1971. Morphological types of pollen grains in tetraploid forms of winter
wheat. - Izvestiya Sibirsk. otd. Akad. nauk S S S R , t. 3, no. 15, p . 6 2 - 6 4 (in Russian).
P o p p e r , K. 1972. O b j e c t i v e k n o w l e d g e : an e v o l u t i o n a r y a p p r o a c h . O x f o r d : Univ.
Press, 380 p.
Potonie, R., Schweitzer, H.I. 1960. Der Pollen von Ulmannia frumentaria. - Palaont.
Ztschr., Bd. 34, H. 1,S. 27-39.
Prakash, U. 1965. A survey of fossil dicotyledonous w o o d s from India and the Far East.
- J. Paleont., vol. 39, no. 5, p. 815-827.
Quinn, C.J. 1966. G a m e t o p h y t e development and e m b r y o g e n y in the Podocarpaceae. 3.
Dacrydium bidwilUi. - Phytomorphology, vol. 16, no. 1, p. 8 1 - 9 1 .
Rao, A.R. 1935. On an abnormal flower of Convolvulus pluricaulis C h o i s . - Curr. Sci.,
vol. 4, p. 4 1 4 - 4 1 5 .
Rao, L.N. 1974. Cycas beddomei Dyer. - Proc. Ind. Acad. Sci., vol. 7 9 , p. 59-66.
R a s m u s s e n , H. 1983. Stomatal development in families of Liliales. - Bot. J b . (Syst.),
Bd. 104, H. 2, S. 261-287.
Rasnitsyn, A.R, Krassilov, V.A. 1996. Pollen in the gut c o m p r e s s i o n s of fossil insects as
evidence of coevolution. - Paleont. Zh. ( M o s c o w ) , no. 6, p . 7 1 6 - 1 2 2 .
Ratter, J.A., Milne, C. 1973 C h r o m o s o m e n u m b e r s of s o m e primitive angiosperms. -
Notes Roy. Bot. Gard. Edinburgh, vol. 32, no. 3, p. 4 2 3 - 4 2 8 .
Reid, E., Chandler, M.E.J. 1933. T h e L o n d o n Clay flora. L o n d o n : Brit. M u s . (Nat. Hist.),
561 p.
Reihman, M.A., Schabilon, J.T. 1985. Stomatal structure of Alethopteris ullivantii and
Neuropteris scheuchzeri, Pennsylvanian pteridosperm foliage. - Amer. J. Bot., vol.
72, no. 9, p. 1392-1396.
R e m y , W., R e m y , R. 1980. D e v o n i a n g a m e t o p h y t e s with a n a t o m i c a l l y p r e s e r v e d
gametangia. - Science, vol. 2 0 8 , p. 2 9 5 - 2 9 6 .
R e m y , W., R e m y , R., H a s s , H., S c h u l t k a , S., F r a n z m a y e r , F. 1980. Sciadophyton
Steinmann - ein G a m e t o p h y t aus den Siegen. - A r g u m e n t a Palaeobotanica, H. 5, S.
73-94.
Retallack, G., Dilcher, D.L. 1981 a. Arguments for a glossopterid ancestry of angiosperms.
- Paleobiology, vol. 7, no. 1, p. 54-67.
Retallack, G., Dilcher, D.L. 1981b. A coastal hypothesis for the dispersal and rise to
d o m i n a n c e of flowering plants. - Paleobotany, paleoecology and evolution. N.Y.:
Praeger, vol. 2, p. 27-77.
Retallack, G., Dilcher, D.L. 1981c. Early angiosperm reproduction: Prisca reynoldsii
g e n . et s p . n o v . f r o m m i d - C r e t a c e o u s c o a s t a l d e p o s i t s , K a n s a s , U . S . A . -
Palaeontographica B , Bd. 179, S. 103-137.
Retisma, T. 1966. Pollen morphology of s o m e European R o s a c e a e . - Acta Bot. Neerl.,
vol. 15, p . 290-307.
R e y m a n o w n a , M . 1968. On seeds containing Eucommiidites troedsonii pollen from the
Jurassic of Groec, Poland. - Bot. J. Linn. S o c , vol. 6 1 , no. 384, p. 147-152.
R e y m a n o w n a , M . 1974. On anatomy and morphology of Caytonia. - S y m p . Morphol.
Stratigr. & Palaeobot., L u c k n p w : B . Sahni Inst. Palaeobot., n o . 2, p. 5 0 - 5 7 .
Rigby, J.E. 1962. Occurrence of Glossopteridae fructifications at Baralaba, Queensland
and their evolution. - Nature, vol. 195, no. 4 8 3 7 , p . 196-198.
Rigby, J.E. 1978. Permian glossopterid and other cycadopsid fructifications from Queens­
land. - Geol. Surv. Queensl. Publ., vol. 367, p . 3-27.
R o d i n , R.J. 1966. Leaf structure and e v o l u t i o n in A m e r i c a n s p e c i e s of Gnetum. -
Phytomorphology, vol. 16, p. 56-58. '
Rohr, R. 1977. Etude comparee de la formation de Texine au cours d e la microsporogenese
chez u n e g y m n o s p e r m e {Taxus baccata) et u n e p r e p h a n e r o g a m e {Ginkgo biloba).
- Cytologia, vol. 4 2 , p. 157-167.
Rohweder, O. 1967. Karpellbau und Synkarpie bei Ranunculaceen. - Berichte Schweiz.
bot. Ges., Bd. 77, S. 3 7 6 - 4 2 5 .
R o m e r o , E.J., Archangelsky, S. 1986. Early Cretaceous angiosperm leaves from south-
e m South America. - Science, vol. 234, no. 4 7 8 3 , p. 1580-1582.
Rothwell, G.W. 1980. T h e Callistophytaceae ( P t e r i d o s p e r m o p s i d a ) . 2. R e p r o d u c t i v e
features. - Palaeontographica B , Bd. 173, S. 85-106.
Rothwell, G.W. 1982a. Cordaianthus duquesnensis sp. n., anatomically preserved ovu­
late cone from the U p p e r Pennsylvanian of O h i o . - Amer. j . Bot., vol. 6 9 , no. 2, p.
239-247.
Rothwell, G.W. 1982b. N e w interpretation of the earliest conifers. - Rev. Palaeobot.
Palynol., vol. 37, p. 7-28.
Rothwell, G.W., Erwin, D . M . 1987. Origin of seed plants: an aneurophyte/seed fem link
elaborated. - Amer. J. Bot., vol. 74, no. 6, p . 9 7 0 - 9 7 3 .
Rothwell, G.W., Wamer, S. 1984. Cordaixylon dumusum n. sp. (Cordaites). 1. Vegeta­
tive stmctures. - Bot. Gaz., vol. 145, no. 2, p. 2 7 5 - 2 9 1 .
Roy C h o w d h u r y , C.^ 1962. T h e e m b r y o g e n y of conifers: A review. - Phytomorphology,
vol. 12, p . 3 1 3 - 3 3 8 .
Ruffle, L. 1980a. W a c h s t u m - M o d u s und Blatt-Morphologie bei altertiimlichen Fagales
und H a m a m e l i d a l e s der Kreide und Gegenwart. 100 Jahre Arboretum, 1879-1979.
Berlin, S. 3 2 9 - 3 4 1 .
Ruffle, L. 1980b. M e r k m a l s - E n k a p s i s und Homologie-Kriterien bei alteren Angiosper­
men und ihre Herkunft. - Schr. geol. Wiss. (Berlin), Bd. 16, S. 4 1 7 - 4 3 9 .
Rugenstein, S.R., Lersten, N.R. 1981. Stomata on seeds and fmits of Bauhinia (Legu-
minosae: Caesalpinioideae). - Amer. J. Bot., vol. 6 8 , n o . 6, p . 8 7 3 - 8 7 6 .
Rutishauser, R., Sattler, R. 1985. Complementarity and heuristic value of contrasting
m o d e l s in stmctural botany. 1. General considerations. - Bot. J b . (Syst.), Bd. 107,
H. 1/4, S. 4 1 5 - 4 5 5 .
Rutishauser, R., Sattler, R. 1986. Architecture and d e v e l o p m e n t of the phyllode-stipule
whorls of Acacia longipedunculata: controversial interpretation and c o n t i n u u m
approach. - C a n a d . J. Bot., vol. 64, p. 1987-2019.
Sabato, S., D e Luca, P. 1985. Evolutionary trends in Dion (Zamiaceae). - Amer. J. Bot.,
vol. 72, no. 9, p. 1353-1363.
S a n w a l , M . 1962. M o r p h o l o g y and e m b r y o g e n y of Gnetum gnemon L. - P h y t o m o r ­
phology, vol. 12, p . 243-264.
Sattler, R. 1974a. A new approach to gynoecial morphology. - Phytomorphology, vol.
24, nos. 1/2, p . 23-34.
Sattler, R.A. 1974b. A n e w conception of the shoot of higher plants. - J. Theor. Biol.,
vol. 4 7 , p. 3 6 7 - 3 8 2 .
Sattler, R. 1978. " F u s i o n " and "continuity" in floral morphology. - N o t e s Roy. Bot.
Gard. Edinburgh, vol. 36, p. 3 9 7 - 4 0 5 .
Sattler, R., Maier, U. 1977. D e v e l o p m e n t of the epiphyllous a p p e n d a g e s of Begonia
hispida van cucullifera: implications for comparative morphology. - C a n a d . J. Bot.,
vol. 5 5 , no. 4, p. 4 1 1 - 4 2 5 .
Schenk, A. 1867. Die fossile Flora der Grenzschichten des Keupers und Lias Frankens
5-9. Wiesbaden, 231 S.
S c h m i d , R. 1 9 8 6 . O n C o r n e r i a n a n d o t h e r t e r m i n o l o g y of a n g i o s p e r m o u s and
g y m n o s p e r m o u s seed coats: historical perspectives and terminological r e c o m m e n ­
dations. - Taxon, vol. 3 5 , p. 4 7 6 - 4 9 1 .
Schopf, J.M. 1976. Morphologic interpretation of fertile structures of glossopetrid gym­
nosperms. - Rev. Palaeobot. Palynol., vol. 2 1 , p. 25-64.
Schopf, J.M. 1982. F o r m s and facies of Vertebraria in relation to G o n d w a n a coal. -
Antarct. Res. Sen, vol. 36, no. 3, p. 37-62.
Schrank, F. Scanning electron and light microscope investigations of angiosperm pollen
from the L o w e r Cretaceous of Egypt. - Pollen et Spores, vol. 2 5 , p . 2 1 2 - 2 4 2 .
Schultze-Motel, W. 1959. Entwicklungsgeschichtliche und vergleichend-morphologische
Untersuchungen im Bliitenbereich der Cyperaceae. - Bot. Jb. (Syst.), Bd. 7 8 , S.
129-170.
S c h w e i t z e r H.-J. 1963. D e r weibliche Zapfen von Pseudovoltzia liebeana u n d seine
B e d e u t u n g fur die P h y l o g e n i e der Koniferen. - P a l a e o n t o g r a p h i c a B , Bd. 113, S.
1-29.
Schweitzer H.-J. 1977. Die Ratische Zwitterbliite Irania hermaphroditica nov. spec,
und ihre B e d e u t u n g fiir die Phylogenie der A n g i o s p e r m e n . - Palaeontographica B,
Bd. 161, S. 9 8 - 1 4 5 .
Scott, A.C., Taylor, T.N. 1983. Plant/animal interactions during the U p p e r Carbonifer­
ous. - Bot. Rev., vol. 4 9 , p. 259-307.
Scott, R.A., B a r g h o o m , E.S. 1958. Phytocrene microcarpa - a n e w species of Icacinaceae
based on Cretaceous fruits from Kreischerville, N e w York. - Palaeobotanist, vol.
6, p. 25-28.
Scott, R.A., B a r g h o o m , E.S., Leopold, E. 1960. H o w old are a n g i o s p e r m s ? - Amer. J.
Sci., vol. 258A, p. 284-299.
Sellwood, B.W., Price, G D., Valdes, P J . 1994. Colder estimates of Cretaceous tem­
peratures. - Nature, vol. 370, pp. 4 5 3 - 4 5 6 .
Serlin, B.S. 1982a. On the stmcture of Cardiocarpus tritolopus, a cordaite seed from
the L o w e r - M i d d l e Pennsylvanian of Kentucky. - Rev. Palaeobot. Palynol., vol. 36,
p. 2 9 7 - 3 0 4 .
Serlin, B.S. 1982b. An Early Cretaceous fossil flora from northwest Texas: its composi­
tion and implications. - Palaeontographica B , Bd. 182, H. 1/3, S. 5 2 - 8 6 .
Seward, A . C . 1894-1895. Catalogue of the M e s o z o i c plants in the D e p a r t m e n t of Geol­
ogy, British M u s e u m Natural History. T h e Wealden flora. L o n d o n : Brit. M u s . (Nat.
Hist.), pt. I , 3 8 4 p . ; p t . 2, 259 p .
S e w a r d , A . C . 1917. Fossil plants. 3 . P t e r i d o s p e r m e a e , C y c a d o f i l i c e s , C o r d a i t a l e s ,
Cycadophyta. C a m b r i d g e : Univ. Press, 6 5 6 p.
Seward, A . C . 1926. T h e Cretaceous plant-bearing rocks of w e s t e m Greenland. - Phil.
Trans. Roy. Soc. London, vol. 2 I 5 B , p. 57-174.
Seward, A . C , Conway, V.M. 1935. Additional Cretaceous plants from western Green­
land. - Kgl. Svenk. Vetensk.-Akad. Handl., Bd. 15, S. 1-51.
Sharma, B.D. 1970. On the structure of seeds of Williamsonia collected from the Mid­
dle Jurassic rocks of Amarjola in the Rajmahal Hills, India. - A n n . Bot. (Gr. Brit.),
vol. 34, p. 1071-1078.
Shashi, N., U e n o , J. 1986. Pollen morphology of Ginkgo biloba and Cycas revoluta. -
C a n a d . J. Bot., vol. 64, p. 3 0 7 5 - 3 0 7 8 .
Shimakura, M . 1937. Studies on the fossil w o o d s from Japan and adjacent lands, 2. -
Sci. R e p . Tohoku Imp. Univ. Sendai. Ser. 2, vol. 19, no. 1, p. 1-73.
S i n g h , H., M a h e s h w a r i , K. 1962. A c o n t r i b u t i o n to the e m b r y o l o g y of Ephedra
gerardiana Well. - Phytomorphology, vol. 12, no. 4, p. 3 6 1 - 3 7 2 .
Singh, H., Oberoi, Y.P. 1962. A contribution to the life history of Biota orientalis Endl.
- Phytomorphology, vol. 12, no. 4, p. 3 7 3 - 3 9 3 .
Sinnott, E.W. 1913. T h e morphology and reproductive structures in the Podocarpaceae.
- A n n . Bot. (Gr. Brit.), vol. 27, p . 39-82.
Sinnott, E.W., Bailey, I.W. 1914. Investigations on the phylogeny of the angiosperms. 4.
T h e origin and dispersal of herbaceous angiosperms. - Ann. Bot. (Gr. Brit.), vol.
2 8 , p. 547-600.
Sitholey, R . V 1954. T h e Mesozoic and Tertiary floras of India - a review. - Palaeobotanist,
vol. 3, p. 55-69.
S k i p w o r t h , J.P. 1970. D e v e l o p m e n t of floral v a s c u l a t u r e in t h e M a g n o l i a c e a e . -
Phytomorphology, vol. 20, p. 2 2 8 - 2 3 6 .
Smith, A.W. 1994. Systematics and the fossil record: d o c u m e n t i n g evolutionary pat­
terns. Blackwell, 223 p.
Smith, E.G. 1907. Morphology of the trunk and development of the m i c r o s p o r a n g i u m of
c y c a d s . - Bot. Gaz., vol. 4 3 , no. 3, p. 187-204.
Smoot, E.L., Taylor, T.N. 1983/1984. Studies on Carboniferous seeds: the relationship
of the integument and nucellus in Pachytesta gigantea. - Rev. Palaeobot. Palynol.,
v o l . 4 0 , p. 165-176.
Smoot, E.L., Taylor, T.N., Delevoryas, T. 1985. Structurally preserved fossil plants from
Antarctica. 1. Antarcticycas, gen. nov., a Triassic cycad stem from the B e a r d m o r e
Glacier area. - Amer. J. Bot., vol. 72, p. 1410-1423.
Sporne, K.R. 1974. T h e morphology of angiosperms. L o n d o n : Hutchinson, 207 p.
Srivastava, S.K. 1981. Evolution of U p p e r Cretaceous phytogeoprovinces and their pol­
len flora. - Rev. Palaeobot. Palynol., vol. 35, p. 155-173.
Stebbins, G.L. 1974. Flowering plants. Evolution above species level. C a m b r i d g e (Mass.):
Harvard Univ. Press, 397 p .
Sterling, C. 1965. C o m p a r a t i v e morphology of the carpel in the Rosaceae, IV. - Amer. J.
Bot., vol. 52, p. 9 3 8 - 9 4 6 .
Sterling, C. 1973. C o m p a r a t i v e morphology of the carpel in the Liliaceae: Colchicae
(Androcymbium). - Bot. J. Linn. S o c , vol. 67, p. 149-156.
Stevenson, D.W. 1980a. Radial growth in Beaucamea recurvata. - Amer. J. Bot., vol.
67, no. 4, p . 4 7 6 - 4 8 9 .
Stevenson, D.W. 1980b. Radial growth in the Cycadales. - Amer. J. Bot., vol. 6 7 , no. 4,
p. 4 6 5 - 4 7 5 .
S t e v e n s o n , D . W . 1 9 8 1 . O b s e r v a t i o n s on p t y x i s , p h e n o l o g y , a n d t r i c h o m e s in the
Cycadales. - Amer. J. Bot., vol. 6 8 , no. 8, p. 1104-1114.
S t e w a r t , W., D e l e v o r y a s , T. 1952. B a s e s for d e t e r m i n i n g r e l a t i o n s h i p s a m o n g the
Medullosaceae. - Amer. J. Bot., vol. 39, no. 7, p. 5 0 5 - 5 1 6 .
Stidd, B.M., Hall, J.W. 1970. Callandrium callistophytoides, gen. et. sp. nov., the prob­
able pollen-bearing organ of the seed fern, Callistophyton. - Amer. J. Bot., vol. 57,
no. 4, p. 3 9 4 - 4 0 3 .
Stopes, M . C . 1912. Petrifications of the earliest European a n g i o s p e r m s . - Phil. Trans.
Roy. Soc. L o n d o n , vol. 2 0 3 B , p. 75-100.
Stopes, M . C . 1915. Catalogue of the Mesozoic plants in the British M u s e u m . T h e Cre­
taceous flora. L o n d o n : Brit. M u s . (Nat. Hist.), pt. 2: L o w e r G r e e n s a n d (Aptian)
plants of Britain, 360 p.
Stopes, M . C , Fujii, K. 1911. Studies on the structure and affinities of Cretaceous plants.
- Phil. Trans. Roy. Soc. L o n d o n , vol. 2 0 1 B , p. 1-90.
Stubblefield, S . P , Rothwell, G.W., Taylor, T.N. 1984. Conostoma williamsonii n. sp., a
lagenostomalean ovule from the Pennsylvanian of the Illinois Basin. - Canad. J.
Bot., vol. 62, no. I, p. 9 6 - 1 0 1 .
Stubblefield, S . P , Taylor, T.N., Daghlian, C P 1982. C o m p r e s s e d plants from the Lower
Pennsylvanian of Kentucky (U.S.A.). I. Crossotheca kentuckiensis n. sp. - Rev.
Palaeobot. Palynol., vol. 36, p. 197-204.
Surange, K.R., Chandra, S. 1975. Morphology of the g y m n o s p e r m o u s fructifications of
the Glossopteris flora and their relationships. - Palaeontographica B , Bd. 149, S.
153-180.
Surova, T.D., K a v a d z e , E.V. 1988. S p o r o d e r m ultrastructure in s o m e g y m n o s p e r m s
(Metasequoia, Cunninghamia, Sciadopitys). - Bot. Z h u m . (Leningrad), vol.73, p.
34-44 (in Russian).
S w a m y , B . G . L . , P e r i a s a m y , K. 1 9 6 4 . T h e c o n c e p t of c o n d u p l i c a t e c a r p e l . -
Phytomorphology, vol. 14, p. 319-327.
Syvanen, M . 1994. Horizontal g e n e transfer: evidence and possible c o n s e q u e n c e s . -
Annu. Rev. Genet., vol. 28, p. 2 3 7 - 2 6 1 .
Takaso, T , B o u m a n , F. 1986. Ovule and seed ontogeny in Gnetum gnemon L. - Bot.
M a g . (Tokyo), vol. 99, p. 241-266.
Takeda, H. 1913. S o m e points in the anatomy of the leaf of Welwitschia mirabilis. -
Ann. Bot. (Gr. Brit.), vol. 27, p. 347-357.
Takhtajan, A . L . 1966. System and phylogeny of flowering plants. M o s c o w , Leningrad:
Nauka, 611 p. (in Russian).
Takhtajan, A.L. 1970. Origins and spread of flowering plants. Leningrad: N a u k a , 145 p.
(in Russian).
Takhtajan, A.L. 1980. Outline of the classification of flowering plants. - Bot. Rev., vol.
4 6 , p. 2 2 5 - 3 5 9 .
Tamura, M . 1963. Morphology, ecology and phylogeny of the R a n u n c u l a c e a e . 1. - Sci.
R e p . O s a k a Univ., vol. 11, p. 115-126.
Tang, W. 1987. Insect pollination in the cycad Zamia pumila (Zamiaceae). - Amer. J.
Bot., vol. 74, no. l , p . 90-99.
Taylor, D.W., Hickey, L.J. 1990. An Aptian plant with attached leaves and flowers:
implications for angiosperm origin. - Science, vol. 247, p. 7 0 2 - 7 0 4 .
Taylor, T.N. 1965. Paleozoic seed studies: a m o n o g r a p h of the A m e r i c a n species of
Pachytesta. - Palaeontographica B , Bd. 117, S. 1-46.
Taylor, T.N. 1973. A consideration of the morphology, ultrastructure and multicellular
microgametophyte of Cycadeoidea dacotensis pollen. - Rev. Palaeobot. Palynol.,
vol. 16, p. 157-165.
Taylor, T.N. 1978. T h e ultrastructure and r e p r o d u c t i v e s i g n i f i c a n c e of Monoletes
(Pteridospermales) pollen. - Canad. J. Bot., vol. 56, no. 24, p . 3 1 0 5 - 3 1 1 8 .
Taylor, T.N. 1981. Palaeobotany. An introduction to fossil plant biology. N.Y. etc.:
M c G r a w - H i l l , 589 p.
Taylor, T.N. 1982. Ultrastructural studies of Paleozoic seed fem pollen: sporoderm de­
velopment. - Rev. Palaeobot. Palynol., vol. 37, p. 2 9 - 5 3 .
Taylor, T.N., Alvin, K.L. 1984. Ultrastmcture and d e v e l o p m e n t of M e s o z o i c pollen:
ClassopoUis. - Amer. J. Bot., vol. 7 1 , no. 4, p. 5 7 5 - 5 8 7 .
Taylor, T.N., A r c h a n g e l s k y , S. 1985. T h e C r e t a c e o u s p t e r i d o s p e r m s Ruflorinia and
Ktalenia and implications on cupule and carpel evolution. - Amer. J. Bot., vol. 72,
p. 1842-1853.
Taylor, T.N., D a g h l i a n , C P . 1980. T h e m o r p h o l o g y and u l t r a s t m c t u r e of Gothania
(Cordaitales) pollen. - Amer. J. Bot., vol. 29, p. 1-14.
T a y l o r , T . N . , M i l l a y , M . A . 1 9 8 1 . M o r p h o l o g i c a l v a r i a b i l i t y of P e n n s y l v a n i a n
lyginopteroid seed fems. - Rev. Palaeobot. Palynol., vol. 32, p . 2 7 - 6 2 .
Teixeira, C. 1948. Flora mesozoica portuguesa. - Port. Serv. Geol., pt. 1, p . 1-118.
T e r a b a y a s h i , S. 1 9 8 5 . T h e c o m p a r a t i v e floral a n a t o m y a n d s y s t e m a t i c s of t h e
Berberidaceae. - M e m . Fac. Sci. Kyoto Univ., Ser. Biol., vol. 10, p. 73-90.
T h a y n , G.F., Tidwell, W.D. 1984. A review of the genus Paraphyllanthoxylon. - Rev.
Palaeobot. Palynol., vol. 4 3 , p. 3 2 1 - 3 3 5 .
Thayn, G.F., Tidwell, W.D., Stokes, W.L. 1983. Flora of the L o w e r Cretaceous Cedar
M o u n t a i n Formation of Utah and Colorado. 1. Paraphyllantoxylon utahense. - Gr.
Basin Natur., vol. 4 3 , no. 3, p. 394-402.
Thayn, G.F., Tidwell, W.D., Stokes, W.L. 1985. Flora of the L o w e r Cretaceous Cedar
M o u n t a i n Formation of Utah and Colorado. 3, Icacinoxylon pittiense n. sp. - Amer.
J. Bot., vol. 72, no. 2, p. 175-180.
T h i e n , L . B . , B e r n h a r d t , P., Pellmyr, O. et al. 1985. T h e pollination of Zygogynum
(Winteraceae) by a moth, Sabatinca (Micropterigidae): an ancient association? -
Science, vol. 227, no. 4 6 8 6 , p. 540-542.
Thoday, M.G., Berridge, E.M. 1912. T h e anatomy and m o r p h o l o g y of inflorescences
and flowers in Ephedra. - Ann. Bot. (Gr. Brit.), vol. 26, p. 9 5 3 - 9 7 9 .
T h o m a s , H.H. 1925. The Caytoniales, a new group of angiospermous plants from the
Jurassic rocks of Yorkshire. - Phil. Trans. Roy. Soc. London, vol. 2 1 3 B , p. 299-363.
T h o m a s , H.H. 1931. T h e early evolution of the angiosperms. - A n n . Bot. (Gr. Brit.), vol.
4 5 , no. 180.
T h o m a s , H.H. 1932/1933. T h e old morphology and the new. - Proc. Linn. S o c . London,
pt. l , p . 17-32.
T h o m a s , H.H. 1933. On some pteridospermous plants from the M e s o z o i c rocks of South
Africa. - Phil. Trans. Roy. Soc. L o n d o n , vol. 2 2 2 B , p. 193-265.
Thomas, H.H. 1952. Glossopteris with whorled leaves. - Palaebotanist, vol. 1, p. 435-438.
T h o m a s , H.H. 1955. Mesozoic pteridosperms. - Phytomorphology, vol. 5, p. 177-185.
T h o m a s , H.H., Bancroft, N . 1913. On the cuticles of some recent and fossil cycadean
fronds. - Trans. Linn. Soc. L o n d o n , Bot., vol. 8, p. 155-204.
Thorne, R.F. 1973. T h e "Amentiferae" or H a m a m e l i d a e as an artificial g r o u p : a sum­
mary statement. - Brittonia, vol. 25, no. 4, p. 3 9 5 - 4 0 5 .
Thorne, R.F. 1976. A phylogenetic classification of angiosperms. - Evolutionary Biol.,
N.Y.: P l e n u m Press, vol. 9, p. 35-100.
Tiffney, B.H. 1977. Dicotyledonous angiosperm flower from the U p p e r Cretaceous of
M a r t h a ' s Vineyard, Massachusetts. - Nature, vol. 2 6 5 , p. 136-137.
Tiffney, B.H. 1984. Seed size, dispersal syndromes, and the rise of angiosperms: evi­
dence and hypothesis. - Ann. M o . Bot. Gard., vol. 7 1 , no. 2, p. 5 5 1 - 5 7 6 .
Tilton, V.R. 1980. T h e nucellar epidermis and micropyle of Ornithogalum caudatum
(Liliaceae) with a review of these structures in other taxa. - C a n a d . J. Bot., vol. 58,
no. 17, p. 1872-1884.
Tomlinson, P B . 1974. Development of stomatal c o m p l e x as t a x o n o m i c character in the
m o n o c o t y l e d o n s . - Taxon, vol. 2 3 , p. 109-128.
T o m l i n s o n , P . B . , W i l d e r , G . J . 1 9 8 4 . S y s t e m a t i c a n a t o m y of C y c l a n t h a c e a e
(Monocotyledoneae) - an overview. - Bot. Gaz., vol. 145, no. 4, p. 5 3 5 - 5 4 9 .
Tosh, K.J., Powell, G.R. 1986. Proliferated, bisporangiate, and other atypical cones
occurring on young plantation growth of Larix laricina. - C a n a d . J. Bot., vol. 64,
no. 3, p. 4 6 9 - 4 7 5 .
Townrow, J.A. 1965. A new m e m b e r of Corystospermaceae T h o m a s . - A n n . Bot. (Gr.
Brit.), vol. 29, no. 115, p. 4 9 5 - 5 1 1 .
Townrow, J.A. 1967. On Rissikia and Matoia, podocarpaceous conifers from the Lower
Mesozoic of southern lands. - Pap. Roy. Soc. Tasmania, vol. 101, p. 103-125.
T r e v i s a n , L. 1980. Ultrastructural notes and c o n s i d e r a t i o n s on Ephedripites,
Eucommiidites and Monosulcites pollen grains from L o w e r Cretaceous sediments
of southern Tuscany (Italy). - Pollen et Spores, vol. 22, n o . 1, p. 85-132.
Trivett, M . L . , R o t h w e l l , G.W. 1985. M o r p h o l o g y , s y s t e m a t i c s and p a l e o e c o l o g y of
Paleozoic fossil plants: Mesoxylon priapi, sp. nov. (Cordaitales). - Syst. Bot., vol.
10, no. 2, p. 2 0 5 - 2 2 3 .
Troll, W. 1932. Morphologic der schildformigen Blatter. - Planta, Bd. 17, S. 153-314.
Troll, W. 1939. D i e m o r p h o l o g i s c h e N a t u r d e r K a r p e l l e . - C h r o n i c a Bot., B d . 5, S.
38-41.
Tsvelev, N . N . 1977. On the primary life form of angiosperms. - Byull. M o s k . obshch.
ispyt. prir., Biol., t. 82, p. 79-88 (in Russian).
Tucker, S.C. 1960. Ontogeny of the floral apex of Michelia fuscata. - Amer. J. Bot., vol.
4 7 , p. 2 6 6 - 2 7 7 .
Tucker, S.C. 1975. Floral development in Saururus cernuus (Saururaceae). 1. Floral
initiation and stamen development. - Amer. J. Bot., vol. 6 9 , no. 9, p. 9 9 3 - 1 0 0 7 .
Tucker, S.C. 1976. Floral development in Saururus cernuus (Saururaceae). 2. Carpel
initiation and floral vasculature. - Amer. J. Bot., vol. 6 3 , no. 2, p . 2 8 9 - 3 0 1 .
T u c k e r , S.C. 1 9 8 1 . I n f l o r e s c e n c e and floral d e v e l o p m e n t in Houttuynia cordata
(Saururaceae). - Amer. J. Bot., vol. 6 8 , no. 8, F 1017-1032.
Tucker, S.C. 1982. Inflorescence and floral d e v e l o p m e n t in the Piperaceae. 2. Floral
ontogeny of Piper. - Amer. J. Bot., vol. 69, no. 9, p . 1 3 8 4 - 1 4 0 1 .
Tucker, S.C. 1985. Initiation and development of inflorescence and flower in Anemopsis
californica (Saururaceae). - Amer. J. Bot., vol. 72, no. 1, p . 2 0 - 3 1 .
Tucker, S.C. 1986. Inflorescence and floral d e v e l o p m e n t in the Piperaceae. 1. Piperomia.
- Amer. J. Bot., vol. 67, no. 5, p. 6 8 6 - 7 0 2 .
Tucker, S . C , Gifford, E . M . 1974. C a r p e l vascularization of Drimys lanceolata. -
Phytomorphology, vol. 14, no. 2, p. 197-203.
Uhl, N.W. 1972. Inflorescence and flower structure in Nypa fruticans (Palmae). - Amer.
J. Bot., vol, 59, no. 7, p. 7 2 9 - 7 4 3 .
Uhl, N.W. 1976. Developmental studies in P t y c h o s p e r m a e (Palmae). 2, T h e staminate
and pistillate flowers. - Amer. J. Bot., vol. 6 3 , no. 1, p. 9 7 - 1 0 9 .
Upchurch, G.R. (Jr.) 1984. Cuticle evolution in Early Cretaceous a n g i o s p e r m s from the
P o t o m a c G r o u p of Virginia and Maryland. - Ann. M o . Bot. Gard., vol. 7 1 , no. 2, p.
522-550.
Vakhrameev, V A . 1952. Stratigraphy and fossil flora of Western Kazakhstan. Moscow,
Leningrad: Akad. nauk S S S R , 340 p. (in Russian).
Vakhrameev, V.A. 1981. Pollen of ClassopoUis: indicator of Jurassic and Cretaceous
climates. - Palaeobotanist, vol. 28/29, p. 301-307.
Vakhrameev, V.A., Kotova, I.Z. 1977. Ancient angiosperms and their associated plants
from the L o w e r Cretaceous of Transbaikalia. - Paleont. Z h . ( M o s c o w ) , no. 4, p.
101-109 (in Russian).
Vakhrameev, V.A., Krassilov, V.A. 1979. Reproductive structures of a n g i o s p e r m s from
the Albian of Kazakhstan. Paleont. Zh. ( M o s c o w ) , no. 1, p. 121-128 (in Russian).
Van C a m p o , M . 1971. Precisions nouvelles sur les structures c o m p a r e e s des pollens de
g y m n o s p e r m e s et d ' a n g i o s p e r m e s . - C. r. Hebd. seances Acad, sci., t. 2 7 2 D , p.
2071-2074.
Van Leeuven, P., Punt, W., Hoen, P.P., 1988. P o l y g o n a c e a e . - T h e N o r t h w e s t Europaean
Pollen Flora, v o l . 4 3 , p. 8 1 - 1 5 1 .
Vane-Wright, R.I.S., Shuiz, S., Boppre, M . 1992. T h e cladistics of Amauris batterflies:
congruence, consensus and total evidence. - Cladistics, vol. 8. p . 125-138.
Vasanthy, G., Pocock, S.A.J. 1986. Radial through rotated symmetry of striate pollen of
the Acanthaceae. - Canad. J. Bot., vol. 64, p. 3 0 5 0 - 3 0 5 8 .
Vasil, I.K., Johri, M . M . 1964. T h e style, stigma and pollen tube, 1. - Phytomorphology,
vol. 14, n o . 3 , p. 352-369.
Vaudois, B . 1981. Archegone et sac embryonnaire (developpement d ' u n e conception
recente). - Ann. Biol., t. 20, fasc. 2, p . 7-9.
Velenovsky, J. 1889. Kvetena C e s k e g o C e n o m a n u . - Kgl. B o h m . G e s . W i s s . Abh., Bd.
3, S. 1-75.
Velenovsky, J., Viniklar, L. 1926. Flora cretacea B o h e m i a e , 1. Praha, 57 p .
Vijayaraghavan, M.R. 1964. Morphology and e m b r y o l o g y of a vesselless dicotyledon -
Sarcandra irvingbailleyi Swamy, and systematic position of the Chloranthaceae. -
Phytomorphology, vol. 4 2 . no. 3, p. 4 2 9 - 4 4 1 .
Vink, W. 1978. T h e Winteraceae of the Old World. 3. Notes on the ovary of Takhtajania.
- Blumea, vol. 24, p. 5 2 1 - 5 2 5 .
Vishnu-Mittre, 1953. A male flower of the Pentoxyleae with remarks on the structure of
the female cones of the group. - Palaeobotanist, vol. 2, p . 7 5 - 8 4 .
Van Konijnenburg-van Cittert, J.H.A. 1971. In situ g y m n o s p e r m pollen from the Mid­
dle Jurassic of Yorkshire. - Acta Bot. Neerl., vol. 20, p. 1-97.
Waha, M . 1985. Ultrastruktur und systematische B e d e u t u n g des Pollens bei Bocageopsis,
Ephedranthus, Malmea und Unonopsis ( A n n o n a c e a e ) . - Plant. Syst. Evol., vol.
150, no. 3/4, p . 165-177.
W a h a , M . 1 9 8 7 . S p o r o d e r m d e v e l o p m e n t of p o l l e n t e t r a d s in Asimina triloba
(Annonaceae). - Pollen et Spores, vol. 29, p. 31-44.
Walker, J.W. 1972. C h r o m o s o m e number, phylogeny, phytogeography of the A n n o n a c e a e
and their bearing on the (original) basic c h r o m o s o m e n u m b e r s of angiosperms. -
Taxon, vol. 2 1 , no. 1, p. 5 7 - 6 5 .
Walker, J.W. 1974a. Evolution of exine structure in the pollen of primitive a n g i o s p e r m s .
- Amer. J. Bot., vol. 6 1 , no. 8, p. 891-902.
Walker, J.W. 1974b. Aperture evolution in the pollen of primitive a n g i o s p e r m s . - Amer.
J. Bot., vol. 6 1 , no. 10, p. 1112-1136.
Walker, J.W., Brenner, G.J., Walker, A.G. 1983. Winteraceous pollen in the L o w e r Cre­
taceous of Israel: early evidence of a magnolialean angiosperm family. - Science,
vol. 220, no. 4 6 0 3 , p. 1273-1275.
Walker, J.W., Walker, A.G. 1984. Ultrastructure of L o w e r Cretaceous angiosperm pol­
len and the origin and early evolution of flowering plants. - A n n . M o . Bot. Gard.,
vol. 7 1 , no. 2, p. 4 6 4 - 5 2 1 .
Walkom, A . B . 1919. Mesozoic floras of Queensland. T h e floras of the B a r r u m and Styx
River Series. - Queensland Geol. Surv., vol. 2 6 3 , p . 1-77.
Walton, J., Wilson, J.A.R. 1932. On the structure of Vertebraria. - Proc. Roy. S o c . Edin­
burgh., vol. 52, no. 2, p. 200-207.
Ward, J.V. 1986. Early Cretaceous angiosperm pollen from the C h e y e n n e and K i o w a
Formation (Albian) of Kansas, U.S.A. - Palaeontographica B , Bd. 2 0 2 , S. 1-81.
Ward, L.F. 1905. Status of the M e s o z o i c floras of the United States. - U . S . Geol. Surv.
Monogr., vol. 4 8 , pt. 1, p . 1-119.
Weller, S.G. 1976. Breeding system p o l y m o r p h i s m in a heterostylous species. - Evolu­
tion, vol. 30, no. 3, p . 4 4 2 - 4 5 4 .
Watson, J. A revision of the English Wealden Flora, 1. Charales - G i n k g o a l e s . - Bull.
Brit. M u s . (Nat. Hist.), Geol., vol. 17, p . 2 0 7 - 2 5 4 .
Watson, J. 1982. T h e Cheirolepidiaceae: a short review. - Phyta, D . D . Pant C o m m e m o r i a l
vol., p. 2 6 5 - 2 7 3 .
Watson, J., Sincock, C.A. 1992. Bennettitales of the English Wealden. L o n d o n : Palaeont.
Soc. Publ., no. 3 8 8 , p. 1-228.
Wettstein, R. 1907. Die Entwicklung der Bliite der angiospermen Pflanzen aus derjenigen
der G y m n o s p e r m e n . - Wissen fur alle, H. 4 5 , S. 7 0 5 - 7 0 8 .
W e t t s t e i n , R. 1 9 3 5 . H a n d b u c h d e r s y s t e m a t i s c h e n B o t a n i k . 4 . A u s g . L e i p z i g , W i e n ,
1152 S.
W i e l a n d , G.R. 1906. A m e r i c a n fossil c y c a d s . - C a r n e g i e Inst. W a s h . P u b l . , vol. 3 4 ,
p . 1-296.
Wiggins, V.D. 1982. Expressipollis striatus n. sp. to Anacolosidites striatus n. sp. An
U p p e r Cretaceous e x a m p l e of suggested pollen aperture evolution. - Grana, vol.
2 1 , p. 3 9 - 4 9 .
Wilder, G.J. 1 9 8 5 . A n a t o m y of n o n c o s t a l p o r t i o n of l a m i n a in t h e C y c l a n t h a c e a e
( M o n o c o t y l e d o n e a e ) , 1. Epidermis. - Bot. Gaz., vol. 146, n o . 1, p . 8 2 - 1 0 5 .
Wilson, C.L. 1937. T h e phylogeny of the stamen. - Amer. J. Bot., vol. 2 4 , p . 6 8 6 - 6 9 9 .
W i n g , S.L., Tiffney, B.H. 1987. T h e reciprocal interaction of a n g i o s p e r m evolution and
tetrapod herbivory. - Rev. Palaeobot. Palynol., vol. 50, p . 179-210.
W o d e h o u s e , R . P 1935. Pollen grains, their structure, identification and significance in
science and medicine. N.Y.: M c G r a w - H i l l , 5 7 4 p.
Worsdell, W.C. 1900a. T h e structure of the female "flower" in Coniferae. A historical
study. - A n n . Bot. (Gr. Brit.), vol. 14, p. 3 9 - 8 2 .
W u X i a n g - w u , H e Yuan-liang, Mei S h e n g - w u 1986. Discovery of Ephedrites from the
L o w e r Jurassic X i a o m e i g o u Formation, Qinghai. - A c t a Palaeobot. Palynol. Sinica,
vol. 1, 19-28.
Yakovlev, M . S . 1973.The main types of e m b r y o n i c d e v e l o p m e n t in higher plants. - In:
E m b r y o l o g y of higher plants. Kishinev: Shtiintsa, p. 16-26 (in Russian).
Z a v a d a , M . 1983. C o m p a r a t i v e morphology of m o n o c o t pollen and evolution trends of
apertures and wall structures. - Bot. Rev., vol. 4 9 , n o . 4, p . 3 3 1 - 3 7 9 .
Zavada, M . S . 1984a. A n g i o s p e r m origins and evolution based on dispersed fossil pol­
len ultrastructure. - Ann. M o . Bot. Gard., vol. 7 1 , n o . 2, p. 4 4 4 - 4 6 3 .
Zavada, M . S . 1984b. T h e relation between pollen exine sculpturing and self-incompat­
ibility m e c h a n i s m s . - Plant Syst. Evol., vol. 147, nos. 1/2, p . 6 3 - 7 8 .
Zavada, M . S . 1991. T h e ultrastructure of pollen found in the dispersed sporangia of
Arberiella (Glossopteridaceae). - Bot. Gaz., vol. 152, p. 248 - 2 5 5 .
Zavada, M . S . , Crepet, W.L. 1986. Pollen grain wall structure of Caytonanthus arberi
(Caytoniales). - Plant Syst. Evol., vol. 153, nos. 3/4, p. 2 5 9 - 2 6 4 .
Zavada, M . S . , Dilcher, D.L. 1986. Comparative pollen morphology and its relationship
to phylogeny of pollen in H a m a m e l i d a e . - Ann. M o . Bot. Gard., vol. 7 3 , no. 2, p.
348-381.
Zavada, M . S . , Taylor, T.N. 1986a. T h e role of selfincompatibility and sexual selection
in the g y m n o s p e r m - a n g i o s p e r m transition: a hypothesis. - Amer. Natur., vol. 128,
no. 4, p. 5 3 8 - 5 4 8 .
Zavada, M . S . , Taylor, T.N. 1986b. Pollen morphology of Lactoridaceae. - Plant Syst.
Evol., vol. 154, p. 31-39.
Zherikhin, V.V. 1979. Evolution and replacements of Cretaceous and C e n o z o i c faunistic
assemblages. M o s c o w : Nauka, 198 p. (in Russian).
Zhou Zhiyan ( C h o w Tseyan) 1995. On s o m e C r e t a c e o u s pseudofrenelopsids with a
brief review of cheirolepidiaceous conifers in China. - Rev. Palaeobot. Palynol.,
vol. 84, p . 4 1 9 - 4 3 8 .
Z i m m e r m a n n , W. 1930. Die Phylogenie der Pflanzen. Jena: G. Fischer, 4 5 2 S.
Z i m m e r m a n n , W. 1952. Main results of the "telome theory". - Palaeobotanist, vol. 1, p.
456-470.
 Index

aberrant 2, 3 3 , 59, 125, 129, 134 Aptian 66, 70, 7 9 , 9 0 , 92, 9 5 , 1 0 8 , 1 0 9 ,

acarpellate 154, 176 121,169,170,217
acceleration 9 5 , 100, 122, 123, 159, 164, Aquilapollenites 90, 91
175 Aquilegia 7 8 , 145, 172
acid rains 109, 123 Arales 7 0
Acmopyle 2 8 , 129 AraliaephyUum 7 0 , 7 2 , 111, 113
Acontium 76 A raliophyllum 167
Actinocalyx 70, 87 Araucaria 29, 30, 3 5 , 129
adaptive type(s) 3, 4, 5, 6, 7, 9, 16, 5 5 , Araucariaceae 30, 129
111, 164 Arberia 2 1 , 59
Afropollis 70, 7 5 , 108 Archaeanthus 79, 80
Agathis 29, 3 0 archaeopterids 16, 17, 56
Albian 32, 66, 70, 7 1 , 7 3 , 76, 8 1 , 92, 9 5 , archegonial c o m p l e x e s 35, 162
109, 110, 112, 122, 123 archetype(s) 5, 131, 132
Alfaropsis 90 Arcto-Tertiary 100, 108
Alismataceae 128 Arecales 70
Alismatales 7 0 areolate venation 3 3 , 5 9 , 135, 141, 167
Alismatophyllum 70 aridity 6 5 , 115, 117
Altingia 85, 145 aridization 103
Altingiaceae 7 3 , 85 arillate 26, 29, 3 3 , 128, 158
amentiferous 12 artificial classification 4, 8
amentifers 9 1 , 96, 124 Ascarina 71
Amentotaxus 29 ascidiform 5 2 , 9 6 , 126, 132, 154, 165,
Anclrostrobus 38 172
Anemone 76, 171 Aspidiophyllum 7 3 , 111
aneurophytes 56 Asterocelastrus 71,87,95
angiospermization 6 5 , 109 Asteropollis 70, 7 1 , 7 5 , 105
angiospermy s y n d r o m e 9, 122, 128 Athrotaxis 29
Anisophyllum 141 Atlanthopollis 91
A n n o n a c e a e 128, 146, 149 Austrobaileya 128, 130, 131, 145, 149,
Annonales 7 0 150, 173
anthostrobili 144, 145 Austrobaileyaceae 79, 172
Antiquocarya 91 Autunia 58, 59
Appomattoxia 7 1 , 84
B
Baculivesiculites 151, 173 C a b o m b a c e a e 7 1 , 75
Baikalophyllum 141 calamopityacean 17
Baisia 3 1 , 4 2 , 6 1 , 105, 121, 156, 157, 159, calliptrids 2 4
160,165,167,169 Callistophytales 1 6 , 2 0 , 5 6 , 5 7
Baisian 36, 4 2 , 4 3 , 4 9 , 50, 5 1 , 52, 160, Callospermarion 58
168, 169, 181, 182, 183, 192, 193, Caloda 7 1 , 7 9
194, 195, 196, 197, 198, 199, 200, Calycanthaceae 71
202, 203 calyptra(s) 72, 8 1 , 82
Baisianthus 4 9 , 50, 105, 169, 192, 193, calyptrate 72, 81
194, 195 cantharophily 115, 121, 167
Barclaya 169 Capparaceae 160
Barremian 6 7 , 9 2 , 108 Carboniferous 17, 20, 56, 62, 135
Beania 4 0 , 4 1 , 59 Caryanthus 7 0 , 91
Bennetticarpus 61 caryophyllids 97
bennettitalean(s) 4 2 , 6 1 , 101, 105, 112, Caspiocarpus l\,ie, 7 8 , 7 9 , 115, 161,
120, 121, 136, 137, 138, 139, 144, 171,172
148, 160, 165, 167, 168, 169, 170, catena 111
181, 182, 183 catenic 101, 103, 122
Bennettitales 1 6 , 3 1 , 5 6 Cathaysian 20, 115
bennettite(s) 15, 16, 24, 3 1 , 32, 4 2 , 6 0 , caulome(ic) 4 4 , 4 5 , 6 1 , 9 5 , 126, 132,
6 1 , 6 2 , 9 8 , 9 9 , 101, 103, 104, 105, 135, 139, 140, 146, 164, 167, 168,
110, 112, 115, 117, 119, 121, 122, 171
123, 129, 132, 134, 136, 143, 145, Caytonanthus 148, 171, 176
154, 159, 160, 165, 166, 167, 168, Caytonia 5 3 , 6 1 , 64, 119, 121, 123, 129,
169, 175 134, 154, 155, 157, 159, 161, 165,
Berberidaceae 7 8 , 131 172, 204
biochemical 122, 123, 130, 131 caytonialean(s) 12, 15, 16, 5 3 , 6 3 , 108,
biomass 9 8 , 9 9 115, 129, 139, 140, 148, 149, 161,
bitegmic 24, 34, 37, 5 5 , 64, 76, 78, 7 9 , 170, 171, 172, 1 7 5 , 2 0 6 , 221
127, 129, 131, 132, 154, 159, 161, Caytoniales 16, 36, 56, 6 1 , 165
172 Celastraceae 87
Bowenia 17,21,25 C e n o m a n i a n 66, 6 9 , 70, 7 1 , 72, 7 3 , 7 5 ,
brachyphyll(s) 9 8 , 101, 102, 103, 105, 79, 80, 8 5 , 87, 9 1 , 9 5 , 9 7 , 108, 109,
108, 110, 111, 112, 117, 122, 123 110, 112, 123, 1 4 0 , 2 0 9 , 2 1 0 , 2 1 5
brachyphyll-bennettite 103, 105, 110, Cephalotaxaceae 2 9 , 162
111, 112, 117, 1 2 2 , 1 2 3 Ceratophyllaceae 9 5 , 128
BrenneripoUis 108 Ceratophyllales 7 0
Buriadia 28 ceratophylls 9 6
Ceratophyllum 96, 169
Ceratozamia 26
Cercidiphyllales 97 coniferoid(s) 1 6 , 2 0 , 2 7 , 3 0 , 4 6 , 4 8 , 5 8 , 6 0 ,
Cercidiphyllum 76, 8 5 , 9 7 , 100, 128, 145 61,62
Chankanella 139, 176 conifer(s) 2, 3 , 11, 12, 14, 15, 16, 2 7 , 2 8 ,
Cheirolepidium 46 30, 3 5 , 4 8 , 5 8 , 5 9 , 6 1 , 62, 6 3 , 9 8 , 99,
Cheirolepis 46 100, 101, 103, 110, 111, 112, 115,
chimerae 65 117,119,123,124,129,149,160,162
chimeric 26, 57, 64 Coniopteris 99
c h l a m y d o s p e r m s 15, 16 cooling 6 5 , 110, 112, 123, 124
Chloranthaceae 6 9 , 7 1 , 72, 95, 128, 146, cordaitalean 22, 57, 5 8
149 Cordaitales 1 6 , 2 7 , 5 6
Chloranthales 70, 169 Cordaites 27
Chloranthistemon 72 cordaites 2 1 , 22, 2 7 , 57, 5 8 , 5 9 , 62
chloranthoid(s) 7 1 , 7 2 , 9 6 Cornales 7 0
Chloranthus 71 Cornetipollis 149
chronocline 38, 164, 165, 176 corona 3 1 , 3 5 , 4 2 , 6 1 , 143, 144, 156,
Circaeaster 76, 7 8 , 84, 9 7 , 100, 128, 157, 169, 170, 181
171, 172 Corystospermales 24, 56
Cissites 7 1 , 7 6 , 7 9 , 171 corystosperm(s) 60, 146, 160
cladistic(s) 4 , 6 , 7 , 9 , 5 5 , 125 Couperites 71
Cladostrobus 22 Cranwellia 90
clandestine period 1 0 9 , 1 2 2 , 1 2 3 Credneria 7 3 , 111
ClassopoUis 4 6 , 4 7 , 103, 108, 117, 118, Cretaceous 24, 2 5 , 3 1 , 32, 36, 3 8 , 39, 4 1 ,
129, 149, 150, 153, 165, 168, 170, 4 2 , 4 3 , 4 8 , 4 9 , 50, 5 1 , 52, 6 0 , 6 1 , 66,
186 67, 6 8 , 6 9 , 7 1 , 72, 7 3 , 7 4 , 7 5 , 76, 77,
Clavatipollenites 7 0 , 7 1 , 150, 169 7 8 , 7 9 , 8 1 , 8 3 , 84, 8 5 , 87, 8 8 , 89, 90,
Claytonia 13, 163 9 1 , 9 2 , 9 3 , 9 4 , 95, 96, 97, 98, 99,
climate 99, 111 100, 102, 103, 104, 105, 106, 107,
climatic 6 5 , 9 8 , 100, 103, 105, 110, 123 108, 109, 110, 111, 112, 113, 114,
climax 110, I I I , 112, 123 115, 121, 123, 124, 126, 136, 139,
C 0 2 98,99 140, 147, 149, 150, 1 5 1 , 153, 154,
community(ies) 4 , 5 6 , 5 7 , 100, 101, 103, 156, 157, 160, 161, 166, 167, 168,
108, 109, 110, 111, 112, 117, 122, 169, 171, 172, 173, 179, 181, 182,
123, 124 183, 189, 190, 191, 192, 193, 194,
Complexipollis 9 1 , 149 195, 196, 197, 198, 1 9 9 , 2 0 0 , 2 0 2 ,
Concurrent homologies 166 2 0 3 , 206, 2 0 9 , 210, 2 1 1 , 2 1 2 , 2 1 3 ,
concurrent homologies 166, 167 214, 2 1 5 , 2 1 6 , 217, 2 1 8 , 2 1 9 , 2 2 2 ,
concurrent trends 176 2 2 3 , 224
conifer-peltasperm 117 cross-veins 50, 126
Coniferidae 1 6 , 2 7 Crossotheca 19, 129
coniferids 1 6 , 2 7 , 2 8 Ctenis 25
Cupressaceae 2 9 , 30, 129
Cycadeoidea 32, 9 8 , 99, 103 diclinous 2 5 , 2 8 , 3 1 , 3 4 , 3 5 , 119, 1 2 2 , 1 2 3 ,
Cycadidae 1 6 , 2 1 127, 130, 143
cycadids 22 dicliny 119
cycadophyte(s) 2 1 , 30, 3 1 , 3 8 , 60, 101, Dicotylophyllum 4 2 , 4 3 , 105, 136
112,125,165 Dicroidium 60
Cycadopitys 41 dilleniids 9 7 , 1 4 9
cycad(s) 11, 12, 14, 15, 16, 17, 19, 20, Dinophyton 36, 4 4 , 4 5 , 56, 6 1 , 135, 148,
2 1 , 2 5 , 27, 3 1 , 34, 38, 5 9 , 60, 6 1 , 6 3 , 149, 154, 155, 157, 159, 167, 170,
64, n o , 115, 123, 129, 151, 160 184, 185
Cycandra 3 8 , 5 9 , 6 0 , 177, 178 dinosaurs 6 5 , 6 6
Cycas 12, 16, 2 1 , 2 5 , 26, 39, 60, 6 3 , 110, Dioon 39
160 Diplophyllum 171
cycle(s) 127, 145 Dirhopalostachyaceae 173
Cyclicity 64 dirhopalostachyacean 172
cyclicity 65 Dirhopalostachys 3 9 , 4 0 , 4 1 , 6 1 , 144,
Cyperaceae 13, 7 1 , 9 3 , 153 154, 155, 159, 165
Cyperacites 105, 106, 107 diversity 4, 7, 15, 24, 56, 6 1 , 6 2 , 9 5 , 100,
Cyperales 91 103, 115, 117, 131, 134, 166, 175
Czekanowskia 101 Doliostrobus 30
czekanowskialean(s) 12, 16, 37, 54, 5 5 , dominant 2 1 , 22, 3 1 , 37, 64, 6 5 , 66, 90,
60, 6 3 , 98, 111, 139, 148, 149, 160, 101, 110, 111, 112, 115, 117, 123,
161, 172, 173 124, 149, 1 6 9 , 2 2 0
Czekanowskiales 1 6 , 3 7 , 5 6 , 1 6 5 double fertilization 2, 5, 8, 11, 15, 29,
Czekanowskietum 111 125, 129, 135, 161, 162, 1 6 7 , 1 7 5
Drewria 3 6 , 4 8 , 108
D Drimys 128, 133, 153, 155
Dacrycarpus 29 Dryophyllum 70, 9 1 , 112
Dacrydium 29
Dammarophyllum 6 9 , 92, 94, 168
Daphnogene 72 ecological niches 4
Davidia 100 ecosystems 6 5 , 6 6 , 121
Deheya 9 1 , 112, 113, 114 ecotonal 99, 103, 109, 112
Debeya-Dryophyllum 112 ecotone 9 1 , 111, 122
deciduousness 99 Elatides 30, 111
Degeneria 128, 132, 145, 150 endexine 9 1 , 128, 129, 150, 170
Degeneriaceae 146, 149, 172 endocingulate 7 5 , 151
deglaciation 65 endocingulus 7 5 , 153, 2 1 0
destrobilation 12, 60, 6 3 , 6 5 , 160 Engelhardia 90
Devonian 15, 16, 17, 18, 56, 62, 174 entomophily 5 9 , 119, 150
Diapensiaceae 87 environments 6 5 , 9 5 , 9 8 , 109
Dicksonia 99
Eoantha 3 1 , 36, 50, 5 1 , 52, 56, 6 1 , 105, fructists 2, 143
136, 143, 144, 1 5 6 , 1 5 7 , 1 6 0 , 165, 167, frugivores 100
168, 170, 196, 197, 1 9 9 , 2 0 0 Furcula 22
Ephedra 2, 34, 35, 4 5 , 4 8 , 50, 129, 132,
134, 136, 162, 168, 170 G
Ephedripites 3 6 , 5 0 , 105, 108 gametophyte(s) 9, 10, 11, 1 4 , 2 7 , 159,
Ephedrites 48 161, 164, 167
ephedroid 4 8 , 50, 52, 108, 189, 190, 191, gametophytic 8, 9, 10, 11, 12, 14, 119,
202 127, 157, 164
epicontinental sea 110 Garuya 140, 141
Equisetosporites 36, 171, 173 Gaussia 22
Erenia 9 1 , 105, 106 gene-transducing 117, 122
Ericaceae 87 Geraniales 70, 88, 9 0
Ericales 7 0 Geranium 90
Euca lyptophyl him 12 Gerofitia 70,92,93
Eucommiidites 151, 173 Gigantopteridales 20, 56
Euphorbiaceae 129, 133, 157, 160 gigantopterids 22, 58, 59, 129, 141, 146
Eupomatiaceae 79, 149, 172 ginkgo 12, 15, 16, 17, 34, 60, 6 1 , 6 3 , 64,
Euptelea 100, 145 129, 143, 149, 150, 151, 158, 160,
Euryphyllum 59 162
Evacarpa 85, 86 ginkgoaleans 17, 30, 60, 64, 9 8 , 101,
evolutionary systematics 6 111, 158
Exhucklandia 169 Ginkgoales 16, 26, 56
exostomes 12, 160 Ginkgoetum 105
Exozoochory 123 Ginkgoites 101
exozoochory 123 glaciation 65
extinctions(s) 3 1 , 6 5 , 111, 112, 121, 123 Glaucidiales 97
Extraovular pollination 30 Glaucidium 78, 97, 172
glossopterid(s) 2 1 , 22, 59, 62, 6 3 , 115,
F 139, 171
Fagales 7 0 glossopteridalean 61
fasciculate stamens 130, 134, 146, 147, Glossopteridales 1 6 , 2 1 , 5 6
222 Glossopteris 2 1 , 22, 6 1 , 139, 171
feeder 34, 36, 163, 168 gnetalean(s) 2, 9, 15, 16, 3 1 , 36, 4 5 , 4 8 ,
Feraxotheca 19 49,50,51,52,60, 61,63, 92,98,
fern-marshes 105 103, 105, 108, 109, 115, 129, 136,
Ferugliocladus 28 146, 149, 150, 154, 156, 160, 162,
Florinites 57 163, 166, 167, 168, 169, 170, 186,
foliophagous 61 187, 188, 192, 193, 194, 195, 196,
Fontainea 141 197, 198
Fraxinus 135 Gnetales 16,33,56
gnetophyte(s) 1 1 , 6 3 , 6 9 , 119, 122, 126, horizontal g e n e transfers 7, 5 6 , 5 9
150,162,165,167, 168,169,170 Hydrastis 7 8 , 172
Gnetopsis 19 Hydrosperma 11
Gnetum 3 3 , 3 4 , 3 5 , 36, 50, 6 1 , 129, 143, Hydrospermales 1 6 , 1 9 , 5 6 , 5 7
149,160, 162,167,169,170 Hyrcantha 70, 8 1 , 8 3 , 84, 9 7 , 154, 171,
G o n d w a n a ( l a n d ) 2 1 , 2 2 , 115 172,214
goniopteroid 3 5 , 4 5 , 4 8
grade(s) 7, 9, 1 0 , 6 3 , 9 1 , 9 5 , 125, 164, I
165,166,174 Icacynoxylon 70, 73
graminoid(s) 1 3 , 5 0 , 5 2 , 7 0 , 9 1 , 9 2 , 9 3 , 9 4 , Illiciales 7 6 , 9 7
9 6 , 1 0 5 , 1 3 6 , 1 3 9 , 1 6 5 , 1 6 6 , 168, 199, Illicium 153
200 Imania 140, 141, 1 7 1 , 2 2 1
Graminophyllum 105 infrastructure 2 4 , 2 5 , 2 6 , 2 9 , 3 3 , 34, 3 5 ,
greenhouse 9 8 , 9 9 , 111, 122 36, 3 8 , 4 2 , 4 5 , 4 7 , 9 1 , 115, 127, 149,
growth rates 99, 100 165, 170, 183
Gurvanella 9 1 , 105, 106, 170 Insiticarpus 71
gymnospermization 65 integuments 17, 19, 20, 34, 5 5 , 5 8 , 6 5 ,
78, 79, 127, 129, 1 3 1 , 158, 159, 160,
H 161, 172, 212
Hamamelidales 70, 97 interseminal scales 3 1 , 3 3 , 4 2 , 6 0 , 143,
h a m a m e l i d s 12, 7 3 , 7 6 , 85, 96, 9 7 , 100, 154, 156, 169
145, 169, 170, 172 Irania 4 1 , 143, 144, 172
harmomegathy 117, 150, 151, 170, 172 Ixostrobus 3 7 , 148, 149, 173, 207
Hediosmum 7 1 , 149
Hedycarya 128, 150
Heerala 3 6 , 4 8 , 187, 188 Jenkinsella 85
Heilungia 25 Joffrea 85
Hellehorus 79 Juglandales 70, 9 0 , 9 6
helophyte 103, 117, 122 juglandioids 91
heterosporous 1 0 , 1 7 , 5 6 Juncaceae 71
heterospory 10, 14, 17 Juncales 91
heterostyly 9 0 Jurassic 24, 30, 36, 38, 39, 4 0 , 4 1 , 4 6 ,
Himantandraceae 172 4 8 , 5 3 , 54, 5 5 , 60, 6 5 , 6 7 , 101, 103,
Hirmerella 4 6 , 4 8 , 129, 158, 159, 186 109, 110, 111, 117, 118, 120, 137,
hirmerellacean 1 1 2 , 1 5 0 , 1 6 8 138, 140, 172, 177, 178, 180, 186,
homology(ies) 2, 1 3 , 5 5 , 5 6 , 126, 1 3 5 , 1 6 2 , 187, 188, 2 0 4 , 2 0 5 , 2 0 7 , 2 0 8 , 2 1 9
163,164,165,166,167
K
homomorphous 6 5 , 1 5 8 , 1 6 0
Karkenia 60
homoplasy 5 5 , 56, 6 2
katydid 4 6 , 1 1 7 , 1 1 8 , 1 1 9
homoxylic 1 2 8 , 1 3 3 , 1 3 4
Kendostrobus 2 1 , 22
Hooleya 90
Keteleeria 29
Kingdonia 5 2 , 9 7 , 100, 128, 141, 144, Lyginopteridales 1 7 , 5 6 , 5 7
154,155,171 Lyrosperma 19
Klikovispermum 70
K o o n w a r r a 79, 108, 159, 169, 189, 190, M
191 macrophyllous 11
Krannera 69, 92 Magnolia 3
Magnoliaceae 3, 69, 79, 115, 130, 133,
145, 149, 150
Lactoridaceae 153 Magnoliaephyllum 70, 72
Lactoris 128, 152 Magnoliales 70, 7 5 , 9 3 , 97
lacustrine 4 2 , 67, 103, 109 magnoliids 12, 76, 79, 96, 97, 126, 145,
lagenostomalean(s) 20, 57, 5 8 , 62, 64 170, 171, 172, 173
Lagenostomales 1 6 , 1 9 , 5 6 , 5 7 Magnoliopsida 3
lagenostome 19, 57 Maloidea 87
laminar stamens 72, 76, 127, 132, 133, Manglietiastrum 110
145, 146 m a n g r o v e 2 1 , 117
land bridges 6 8 Manihotites 141
Laramisemen 11 Manningia 91
Laurales 7 0 , 9 5 manoxylic 26, 31
lauroid(s) 3 3 , 72 marginal meristem 130, 135
leaf-mats 111 Mauldinia 70, 72
Lemna 13, 131 medullosaleans 1 9 , 2 2 , 6 1
lemnoid 1 7 6 , 2 2 4 Medullosales 17, 1 9 , 5 6 , 5 7
Lenlogia 17, 1 8 , 5 6 , 157 Meeusella 146, 147, 148
Leongathia 36, 108, 168, 189, 190, 191 m e g a g a m e t o p h y t e 19, 20, 2 1 , 26, 2 7 , 29,
leptoma 27, 4 7 3 3 , 34, 3 5 , 36, 64
Leptostrobus 37, 54, 5 5 , 6 1 , 129, 154, megaspore m e m b r a n e 19, 2 1 , 24, 2 5 , 26,
155, 165, 173, 207 2 7 , 2 8 , 2 9 , 3 3 , 3 7 , 4 1 , 4 5 , 4 7 , 5 0 , 198
Lesqueria 70, 85, 97 M e n i s p e r m a c e a e 79, 81
life-form 2, 5, 15, 5 9 , 60, 79, 121, 126 Menispermites 7 1 , 76, 7 7 , 111, 171
ligule 2 9 , 30 Mesoxylon 57
Liliaceae 128, 150, 159 Metaia 30
Liliacidites 70 Metasequoia 11, 124
links 1, 30, 3 8 , 56, 7 5 , 97, 141, 143, 170, M icrocachridites 151
171 microphyllous 11,27
Linum 8 8 , 9 0 micropyle 1 9 , 2 0 , 2 1 , 2 4 , 2 6 , 3 0 , 3 4 , 3 5 ,
Liriodendron 79, 100 3 7 , 4 1 , 4 2 , 5 3 , 5 7 , 6 1 , 6 4 , 7 8 , 7 9 , 119,
Liriophyllum 79, 80, I I I 127,159,160,162,172,212
littoral 4 6 , 9 9 , 103, 105, 111, 122 mid-Cretaceous 32, 60, 6 1 , 66, 6 7 , 72, 96,
Loranthaceae 9 0 98,110,111,121,123,151,161,166,
lyginopterid(s) 1 7 , 1 9 , 2 0 , 2 2 , 5 9 171, 172
266 V. A. KRASSimv

miniaturization 161 N y m p h a e i d s 75
misassignments 126, 170 nymphaeids 96, 171
Monimiaceae 8 1 , 128, 133, 146 Nyssaceae 100
monoclinous 3 1 , 3 4 , 4 1 , 8 1 , 87, 119, 127, Nyssidium 85
130, 143, 144
monocliny 119, 123, 129 O
monocotyly 13 Ochm 153
Monoletes 20 Oculopollis 91
morphocline(s) 9 5 , 131, 164, 165 Onagraceae 90, 129, 161
morphological pool 165, 166, 174 Oreomunnea 90
mosaic evolution 38 orogeny 6 7 , 6 8
Moyliostrohus 28 Osmunda 99
multituberculates 6 1 , 121, 123 Otozamites 99, 103, 104, 139
Myricaephyllum 91 outgroup 150, 151, 160, 163, 164, 176
Myricales 70, 90, 170 Oxalis 9 0
myricoids 90, 166, 170 ozone 57

N
natural classification 4, 7 pachycaul 12, 2 5 , 3 1 , 32, 3 5 , 9 8 , 121,
Nelumhites 7 5 , 96 126
Neocomian 70, 84, 9 5 , 105, 109, 121 Pachy testa 61
neotenic 10, 159 Paeonia 76, 8 1 , 84, 97, 127, 128, 129,
Nephropsis 22 146, 148, 149, 154, 155, 162, 171,
Nilssonia 4 1 , 101 172
nilssonialeans 5 9 , 6 3 , 6 4 , 146, 160 Paeoniales 7 0 , 9 7 , 9 7 - 1 7 6
Nilssoniales 1 6 , 2 5 , 4 1 , 5 6 Paleocene 66, 77, 8 1 , 82, 86, 97, 114,
Noeggerat/iiopsis 2 1 , 59 220
Nordenskioldia 69, 8 1 , 82, 124 Paleorosa 87
Normapolles 7 0 , 9 1 , 112, 149, 150, 152, Palmoxylon 70
153 Papaveraceae 170
Nothofagidites 70 pappus 1 9 , 3 3 , 159
nucellar 19, 20, 2 5 , 26, 2 8 , 30, 4 2 , 57, Parakmeria 100
6 1 , 6 4 , 131, 159, 197 parallel development 8, 58
nucellus(i) 19, 20, 2 1 , 2 4 , 25, 26, 27, 28, parallel evolution 57, 122, 166
29, 3 0 , 3 3 , 3 4 , 3 5 , 3 7 , 3 9 , 4 1 , 4 2 , 4 4 , parallelism(s) 2 2 , 5 9 , 6 2 , 6 4 , 1 0 1 , 1 1 5 , 117
4 5 , 4 7 , 4 8 , 50, 5 5 , 5 7 , 5 8 , 6 0 , 6 1 , 6 4 , Paramichelia 100
65,78,119,127,158,159,160,172, paraphyletic 151, 164
179, 184, 186, 1 9 7 , 2 0 4 , 2 0 5 Paraphyllantoxylon 70, 73
nympaheid 75 Parasahatinca 119
Nymphaea 176 peltasperm(s) 17, 2 3 , 24, 2 5 , 5 8 , 59, 60,
N y m p h a e a l e s 7 0 , 7 1 , 7 5 , 9 5 , 9 6 , 169 62,63,115,117,141,142,146
Peltaspermales 16, 22, 56 platyspermic 1 6 , 2 0 , 2 1 , 2 4 , 2 5 , 2 6 , 2 7 , 2 8 ,
Pentoxylales 24, 56 4 1 , 5 7 , 5 8 , 5 9 , 6 2 , 6 4 , 172
pentoxyleans 6 0 , 61 platyspermy 15, 57
periodicities 6 5 , 6 6 ploidy 150
Permian 17, 1 9 , 2 0 , 2 1 , 2 2 , 2 4 , 3 0 , 3 9 , 4 8 , Podocarpaceae 29, 30, 6 3
5 6 , 5 8 , 5 9 , 6 0 , 6 1 , 6 2 , 6 4 , 6 5 , 6 7 , 115, Podozamites 69
116,117, 146 pollination 12, 20, 30, 3 3 , 4 2 , 5 7 , 59, 64,
Phoenicopsis 9 8 , 9 9 , 101, 105 6 5 , 112, 119, 128, 132, 154, 159,
phyllodic 12, 22, 3 3 , 4 5 , 5 8 , 59, 6 1 , 126, 164, 172
130, 136, 167, 168 pollinators 6 5 , 115, 119, 146
phyllome 4 4 , 4 5 , 8 6 , 133, 135, 139, 145, pollinivorous 2 0 , 5 9 , 1 1 6 , 118, 122
146, 153, 164, 167, 172 pollinivory 2 0 , 1 1 5 , 1 1 7
phyllomic 4 4 , 132, 133, 143, 145, 146 polyembryony 2 6 , 3 0 , 161
phyllomization 136,139, 1 5 4 , 1 6 4 , 1 7 1 polyphyletic 1 0 , 5 9 , 9 5 , 125
phyllospermy 15 polyploid 1 0 , 1 2 8 , 1 6 2
phylogenetic 4 , 5 , 6 , 7 , 8 , 9 , 10, 12, 13, 15, Popperian 6
16,38,45,52,55,56,57,59,60,61, Populophyllum 76
6 2 , 6 3 , 6 9 , 75, 9 5 , 131, 132, 134, 135, Populus 76
141, 146, 164, 165, 166, 168, 170, 171, Potomageton 105
173, 174,175 Potoniea 20
phylogeny 4 . 5 , 6 , 7 , 3 8 , 5 5 , 5 6 , 6 2 , 125, preflower(s) 52, 105, 115, 119, 144, 145,
166,167 148, 156, 1 5 9 , 1 6 8 , 169, 170, 196,202,
phytogeographical 9 8 , 103, 108 203
p h y t o m a s s 110 prepollen 1 9 , 2 7 , 2 8 , 2 9 , 5 7 , 5 8 , 6 2 , 6 4
Pinaceae 30 Prisca 70, 72, 80
pioneer 6 5 , 101, 110, 159 proangiosperm(s) 1 6 , 3 1 , 3 8 , 4 0 , 4 1 , 5 5 ,
piperoids 7 1 , 9 6 62,63,65,92,105,108,109,115,118,
Piperomia 9 6 , 128, 156, 163, 169 122, 123, 124, 125, 136, 140, 141, 154.
Pirroconites 36 157, 1 5 8 , 1 5 9 , 1 6 1 , 1 6 4 , 1 6 5 , 1 6 6 , 1 6 9 ,
Pityophyllum 9 9 , 101 172, 173,174, 175
Pityospermum 99 Proangiospermidae 16, 31
placentation 5, 12, 130, 133 proangiospermous 3 1 , 3 6 , 4 2 , 5 1 , 5 6 , 6 1 ,
Platanaceae 6 9 , 7 3 , 8 5 , 9 5 , 96, 149, 169 62, 105, 108,115, 119, 122, 139,140,
Plataninium 73 160, 161, 181, 182, 183, 196, 197, 198
Platanites 73 Problematospermum 120, 121, 159, 167
platanoid(s) 7 0 , 7 2 , 7 3 , 7 4 , 7 6 , 8 5 , 8 7 , 9 6 , p r o g y m n o s p e r m s 16, 1 7 , 5 6 , 164
9 7 , 111, 1 6 8 , 1 6 9 , 2 0 9 , 2 1 0 Proteaceae 9 0
platanophyll-conifer 111 Proteaephyllum 76
platanophylls 7 3 , 8 5 Proteoides 70, 7 2 , 112
Platanus 7 3 , 8 5 , 128 Proteophyllum 1 4 0 , 1 6 7
protognetalean(s) 3 6 , 4 4 , 5 6 , 6 1 , 135, 148, reduction 9, 10, 13, 29, 58, 60, 6 4 , 6 5 , 98,
149,184,185 123,124, 131, 1 3 3 , 1 5 1 , 1 5 8 , 1 5 9 , 1 6 1 ,
Protohlechnum 58 162,164,167,168,172
Protomonimia 71,79 Restionales 91
Protophyllum 7 3 , 111 retroconvergence 6 5 , 141
protopteridophytes 11 reversals 64, 157
protosaccate 29, 36, 59, 115, 119, 172 Rhexoxylon 24, 6 0
Prototinomiscium 11 Rhodolea 145
province 2 0 , 2 1 , 9 1 , 9 9 , 101, 103 rift(ing) 6 7 , 6 8 , 1 0 9 , 1 2 2
pseudanthial 97, 145 rimula 4 6 , 4 7 , 150, 170
pseudohomologies 1 2 5 , 1 5 0 Rosaceae 87
Pseudolarix 110, 119 Rosales 7 1 , 9 5
p s e u d o m o n o m e r o u s 131, 133, 154, 155, rosoids 87, 96
172 Rufloria 22
pseudoorthotropous 29, 160 Russelites 58
Pseudotorellia 111
Pseudotsuga 29
Pseudovoltzia 30 Sagenopteris 108, 139, 140, 141, 165,
Psilotum 10 171,221
pteridosperms 12, 16, 15, 17, 19, 20, 2 1 , Sanmiguelia 36, 4 5
2 2 , 3 1 , 3 8 , 5 7 , 5 8 , 5 9 , 6 0 , 6 4 , 129,146, Sapindales 6 9 , 7 1
151,154,158,160 Sapindopsis 69
Pterocaryopsis 91 Sarbaya 70, 7 3 , 74, 7 5 , 151, 2 0 9
Pteronilssonia 21 Sarcandra 96, 128
Pterophyllum 101 Sarysua 7 0 , 8 7 , 8 8 , 8 9 , 9 0 , 157
Ptilophyllum 99, 103, 136 Sassafras 128
pycnoxylic 24, 26, 27, 28 Saururaceae 79, 133, 146
Saxegothea 129
Q Saxifragales 7 1 , 8 7
Scandianthus 7 1 , 87
Questora 19, 57 Schizolepis 30
Schrankipollis 108
R
Sciadophyton 10
radiospermic 16, 26, 3 3 , 34, 39, 58, 6 1 , 6 2
Sciadopitys 28
radiospermy 15
Scoresbya 22, 140, 141, 165, 171
rain-forest 110
Scytophyllum 2 3 , 142
Ramonicarya 70, 9 0 , 170
seaways 67
ranalean 12, 160
self-incompatibility 108, 117, 119, 122,
Ranunculaceae 2, 7 8 , 131, 154
123,127,149,156,157
Ranunculids 7 1 , 76, 97
semenivores 100
ranunculids 76, 79, 9 5 , 96, 97, 126, 141,
semenivory 30, 121, 125
145, 170, 171, 172
Semiliquidambar 110 syncupulate 165
Semionogyna 3 8 , 39, 60, 179 syncupuly 154
semophyletic trends 176 synsepaly 87
Senonian 6 6 , 7 0 , 7 1 , 7 2 , 7 5 , 7 9 , 8 7 , 9 4 , 9 5 ,
syntelomic 1 7 , 1 9 , 5 7 , 1 3 5 , 1 5 4 , 1 5 7
119
Sequoia 1 1 , 3 0 , 110, 124, 129, 162 T
serai 101, 103, 110
Tabebuia 164
sexualization 162
taeniate 22, 59, 62, 64, 94, 116
siphonogamous 30, 127, 159
Taldysaja 6 9 , 7 0 , 9 3 , 9 4 , 168
Smilacaceae 130, 135
Tambourissa 127
Sobernheimia 39, 60 Tanaitis 17
Sparganium 105, 106, 107, 128 taphonomic 1 0 0 , 1 0 1 , 1 0 9
sparganum 20, 27 Taxaceae 2 9 , 30, 162
S p e r m o p h y t a 14, 15 Taxodiaceae 29, 30
Spermopteris 20 taxodiacean-laurophyll 123
Sphenobaiera 101 Taxus 12
Sphenopteris 19 telome 161
stachyospermy 15 Tethys 6 7 , 6 8 , 9 9 , 111
Stangeria 25 Tetracentron 110
Steinhauera 85, 124 tetrad(s) 19, 20, 26, 29, 3 4 , 4 5 , 4 7 , 50, 92,
Stellatipollis 119 127,129,150,152,153,195
Stenvaesia 87 Thalictrum 81,97
Stephanocolpites 11 therapsids 65
stigma 8 1 , 8 8 , 9 0 , 106, 119, 127, 143, 150, time-fashion trends 165
154, 1 5 6 , 2 1 4 , 2 1 7 , 2 1 8 , 2 2 4 time-fashions 6 4
stigmatic 12, 37, 3 8 , 52, 54, 5 5 , 79, 84, Tinsleya 58
90,119,129,154,155,165,169,173, Todites 99
208 toroid flowers 9 8 , 9 9
stomata 2 4 , 2 5 , 2 6 , 2 7 , 28, 3 1 , 34, 3 5 , 36, transgrade trend 164
38, 3 9 , 4 4 , 4 5 , 4 7 , 4 8 , 5 0 , 5 5 , 7 2 , 7 4 , transgression(s) 9 8 , 111, 123
75,79,81,91,96,99,103,112,114, trend(s) 9, 11, 1 2 , 1 3 , 1 4 , 6 5 , 131, 164,
127,129,130,136,137,138,139,141, 165,166,167,168,169,170,171,172,
160,165,168,169,171,186,190,214 174,175
stomatal 2 2 , 5 0 , 7 2 , 7 3 , 7 6 , 9 1 , 9 6 , 111, Tricolpites 105, 151
112, 114, 136, 138, 141, 171, 191, Tricolpopollianthus 85, 220
199 trigonocarpalean 2 0 , 6 0 , 6 1 , 6 4
striate pollen 115, 117 Trigonocarpales 1 6 , 2 0 , 5 6 , 5 7 , 5 8
strobilation 1 2 , 5 8 , 6 5 , 1 4 5 , 1 6 0 Triplicarpus 70
strobiloid flowers 176 Trochdendraceae 69
Sujfunophyllum 140, 167, 176 Trochodendrales 7 6 , 9 7
syncotyly 131 Trochodendrocarpus 4 1 , 70, 8 5 , 154,
173,220
Trochodendroides 85, 111, 1 2 4 , 2 2 0 X
Trochodendron 8 1 , 97 xeromorphic 4 7 , 9 8 , 9 9 , 102, 112, 114,
Trollius 16, n\ 1 ] 5 j 2 3 , 1 2 6 , 1 2 9 , 130
T^uga 129 xeromorphism 103, 105
Turonian 6 6 , 7 0 , 7 1 , 7 2 , 7 3 , 7 9 , 8 1 , 9 2 , 9 3 ,
97 Z
Typhaera 105, 121 Zamiopteris 22
Tyrmocarpus 84, 159 zoochorous 6 5 , 100
zoochory 30, 61
u
Urpflanze 4, 5, 131, 132, 153
Urticales 71

V
Vardekloeffia 33
ventral canal cell 29, 3 5 , 129, 162, 163
ventral canal nucleus 29
Verrumonocolpites 151
Vertebraria 2 1 , 59, 117
vessels 2, 1 5 , 3 1 , 3 4 , 7 9 , 9 6 , 126, 128,
1 2 9 , 1 3 4 , 1 3 9 , 1 6 5 , 1 6 7 , 168
Victoria 132, 146, 189, 190, 191
Virginianthus 70
Vitimantha 52, 1 0 5 , 2 0 2 , 2 0 3
Vitiphyllum 7 1 , 7 6 , 79, 171
Vittatina 24, 116
Vojnovskya 22
vojnovskyalean(s) 2 2 , 5 9 , 6 2
Vojnovskyales 16, 22, 56
Voltziaceae 2 8 , 3 0
voltziacean(s) 30, 5 6

W
Walchia 28
Walchiales 1 6 , 2 7 , 2 8 , 5 6 , 5 8
Walkeripollis 92, 152, 153
Wealden 6 7 , 103, 105, 108
Welwitschia 3 5 , 4 5 , 4 8 , 119, 129, 143,
151,156, 1 5 8 , 1 6 2 , 1 6 7 , 1 6 9 , 170
Williamsonia 119, 156
Winteraceae 128
Woodehousia 90