How Plants May Benefit from the Animals That Eat Them
Author(s): Denis F. Owen
Source: Oikos, Vol. 35, No. 2 (Oct., 1980), pp. 230-235
Published by: Wiley on behalf of Nordic Society Oikos
Stable URL: http://www.jstor.org/stable/3544430
Accessed: 26-06-2016 01:28 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/3544430?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
Nordic Society Oikos, Wiley are collaborating with JSTOR to digitize, preserve and extend access to Oikos
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms
OIKOS 35: 230-235. Copenhagen 1980
How plants may benefit from the animals that eat them
Denis F. Owen
Owen, D. F. 1980. How plants may benefit from the animals that eat them. - Oikos
35: 230-235.
It is suggested that the mutually beneficial relationships between flowers and their
pollinators and between fruits and fruit-eaters can be extended to cover a much wider
range of plant-consumer interactions. The ability to photosynthesize is by itself of
limited value to a plant if growth and reproduction are restricted by the availability of
nutrients. Hence natural selection should favour adaptations that increase the rate of
supply of a scarce resource. By enlisting the "help" of a consumer a plant may
facilitate cycling and increase the supply of a nutrient in short supply; in particular the
deposition of sugary honeydew by aphids may increase the rate of nitrogen fixation
beneath the plant by providing an energy source for free-living nitrogen-fixing bac-
teria. Consumers may also affect the growth form of a plant in a beneficial way;
grasses, for example, survive best when cropped and are probably totally co-evolved
with grazers: one would not be possible without the other.
A speculative paper such as this almost certainly contains errors of interpretation, but
even if only some of the suggestions are correct the common viewpoint that plants
defend themselves against consumers will have to be radically changed.
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms
Introduction Five propositions
Land plants can be eaten without being killed. The Before indulging in further speculation I must state the
death of an individual as a result of being chewed or case as precisely as possible:
sucked is unusual and when it does occur the plant is (1) The ability to photosynthesize is by itself of li-
usually growing in "unnatural" circumstances or is one mited value to a plant if growth and subsequent repro-
that has been genetically altered by selective breeding duction are limited by the availability of nutrients or
by man to an extent that it is no longer able to effec- water. Plant productivity, it seems, is more often re-
tively defend itself against plant-feeders. Nearly all the stricted by the supply of nutrients than by energy input
reported plagues of destructive caterpillars and bugs are (Witkamp 1971), and water scarcity is well-known to
associated with trees and plants growing in plantations inhibit plant growth.
or fields; the problem of pests has arisen partly because (2) This being so, natural selection should favour
of the way the crop is grown (as a monoculture, for adaptations that increase the rate of supply of a re-
example), and partly because its chemical defences have source of limited availability.
been bred out. Wild cabbage growing on sea cliffs in (3) There are restrictions to a plant's ability to re-
southern England is rarely damaged by aphids and lease its own "excess" nutrients or water, but by enlist-
caterpillars, but cultivars of the same species growing in ing the "help" of a consumer there are better prospects.
fields and gardens suffer badly from these insects. If the consumer excretes sugars, water or other sub-
Few would dispute that many plants have evolved stances obtained from the plant, or if it causes prema-
strategies that entice consumers in a way that benefit ture shedding of leaves or other plant parts, these re-
both the plant and the consumer. The provision of sugary sources may be mobilized beneath the individual plant
nectar as an insect or bird attractant and the consequent from which they came and subsequently make available
pollination of the flower is a well-known example of nutrients in short supply - to the benefit of the plant.
co-evolution between the eaten and the eater. Similarly, (4) A consumer may also affect the growth form of a
brightly coloured fruits are an invitation to the eater plant. Damage to or removal of the growing point may
which results in effective dispersal of the undigested induce branching which in turn determines overall
seeds. The fact that some fruits are poisonous - and structure to an extent that the plant may be better sup-
perhaps also warningly coloured - to some animals is no ported, its photosynthetic parts better positioned, and it
more than a refinement to the arrangement, ensuring may be able to produce more seed from the increased
that the wrong fruit-eater is put off and that the right number of branches.
one, which is not affected by the poison, gets the fruit. (5) Some plants, grasses in particular, survive best
The germination rate of seeds is often better if they when they are cropped by consumers. They have
have first passed through the gut of an animal. In the evolved adaptations that permit or encourage high
Ivory Coast the seeds of 37 species of forest trees ger- levels of consumption, and in the absence of consumers
minate more rapidly after the fruit has been eaten and they survive less well, or even disappear. I shall argue
the seeds defaecated by elephants; the seeds also get a that grasses and grazers are totally co-evolved and one
good start because they sprout from a pile of elephant would not be possible without the other: their
droppings (Alexandre 1978). mutualism is as intimate as that between a flower and a
This paper is a reappriaisal and to some extent a revi- bee.
sion of an earlier one (Owen and Wiegert 1976) in
which it was suggested that the well-known mutually
Aphids and honeydew
beneficial relationships between flowers and their pol-
linators could be extended to include a variety of other Aphids pierce plant tissues with their stylets and imbibe
types of plant eating by animals. If the suggestions made sugar and amino acids in solution. Much of the sugar is
in the earlier paper and those added in this one are excreted as honeydew, a substantial proportion of it,
correct, much of our current thinking about the ways in more than 40% in some instances (Dighton 1978,
which plants "defend" themselves against consumers Michel 1942), being first converted into the trisac-
will have to be radically altered: the relationship be- charide melezitose, a sugar known only from the hon-
tween eater and eaten will have to be envisaged as eydew of aphids and other Homoptera (Bacon and Dic-
mutualism in which plants not only tolerate and reg- kinson 1957). Other trisaccharides, for example fruc-
ulate, but also attract the animals that eat them. The tomaltose, also appear in aphid honeydew (Gray and
original hypothesis that consumers facilitate nutrient Fraenkel 1953). Everything suggests that aphids must
cycling received support, particularly from models necessarily take on large quantities of sugar some of
suggested by Stenseth (1978), who concluded that which, however, they convert to unique trisaccharides
short-lived, successional plants (so-called r-strategists) before it is excreted. The amount of honeydew depo-
are more likely to benefit than long-lived trees. sited is enormous: up to 1 kg m-2 yr-1 beneath a 14 m
high lime tree (Tilia) which at the height of summer may
support over a million aphids (Llewellyn 1972).
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms
This deposited sugar is utilized by a wide range of factor to growth. Cercopid bugs extract amino acids in
organisms and it was earlier suggested that among them dilute solution from the growing points of plants and
are free-living nitrogen-fixing bacteria which increase in excrete large amounts of excess water. Members of the
numbers beneath an aphid-infested plant and hence African genus Ptyelus are gregarious on both forest and
make available more nitrogen for the plant (Owen and savanna trees and in the dry season deposit large quan-
Wiegert 1976, Owen 1978a). In effect a plant releases tities of water on the parched soil beneath trees. The
"surplus" sugar by enlisting the "help" of aphids and so-called "raining trees" are familiar to travellers in
thus increases its own rate of supply of nitrogen, which Africa - when one is discovered it is at first astonishing
in many environments may be in short supply and a to see what appears to be a shower of rain coming from
limiting factor to growth and reproduction. a tree in an otherwise dry environment. Earlier it was
The conversion of plant sugars to melezitose and suggested that the release of water onto the soil surface
other trisaccharides is especially interesting as it must enables roots to take up nutrients in solution and is
be, for the aphids, energy-demanding. An aphid is particularly beneficial in the tropics where soil nutrients
shortlived and is itself unlikely to benefit from what it is are often scarce and where there is intense competition
doing because by the time the beneficial effects to the to gain access to them (Owen and Wiegert 1976). I am
plant of an increase in available nitrogen are apparent not aware of anyone following up this suggestion with
the aphid would be dead. But aphids are parthenogene- an experimental design; this would not be difficult as
tic throughout most of the summer and therefore large Ptyelus is widespread, abundant, and large and con-
numbers are genetically identical; thousands may in spicuous. Testing the hypothesis that the water released
evolutionary terms constitute one gigantic, longlived in- by bugs increases nutrient uptake from the soil would be
dividual. The repeated suggestion that there is indeed an excellent project for someone in an African univer-
genetic variability in aphid clones has recently been sity - I have personally seen the bugs on trees on the
re-assessed and the conclusion reached that if such var- campuses of two such universities, but this was before I
iability occurs it is most unusual (Blackman 1979). began to think about their possible effect on the supply
The addition of sugar to soil is known to increase the of nutrients to trees.
amount of nitrogen fixed (Jones 1974), and to increase
the density of Azotobacter cells (Delwiche and Wijler
1956). What is not known is whether trisaccharides,
especially melezitose, synthesized by aphids play a spe-
cial role in nitrogen fixation. Experiments now being
Loss of leaves and other plant parts
conducted indicate that the addition of sugars present in
aphid honeydew does increase nitrogen fixation as Monk (1966) suggested that year-round leaf fall in cer-
measured by the acetylene reduction method (Petelle tain evergreen trees ensures that small amounts of nut-
1980). Different types of sugars stimulate the rate of rients are made available to a tree throughout the year;
nitrogen fixation in different ways: fructose stimulates this might be especially important in dry. sterile soils.
fixation most, glucose least, and sucrose and melezitose Owen and Wiegert (1976) and Owen (1978b) agree,
in between with about the same effect. adding that consumers can spread out the season of leaf
My own prediction is that aphid honeydew deposited fall, citing as an example the effects of the larvae of the
on soil beneath a plant can substantially increase the leaf-miner, Phytomyza ilicis, which extend the "nor-
rate of nitrogen fixation to the benefit of the plant. I mal" seasonal pattern of leaf fall in the holly, hlex
further predict that the synthesis of melezitose by aquifolium.
aphids is an additional adaptation that promotes nit- There is now a considerable body of evidence that the
rogen fixation to an extent that both the plant and the presence of a consumer stimulates premature shedding
aphid, viewed as a longlived evolutionary individual, of leaves and other plant parts. Weevils, Conotrachelus
benefit. Melezitose by itself may not be particularly im- affinis, that bore into the nuts of the hickory, Carya
portant in this arrangement; more likely it is the mixture glabra, cause the nuts to drop prematurely (Boucher
of sugars present in honeydew that has the optimal ef- and Sork 1979 - who cite further references to this
fect on nitrogen fixation. If I am right, an aphid can be phenomenon). By shedding nuts (or leaves) containing
thought of as a necessary "part" of a plant; it is able to insects, a plant could theoretically regulate the numbers
release surplus sugar enabling a better supply of nit- of its own consumer; early drop of infected nuts also
rogen which, in many places, is the main limiting factor saves energy and avoids wastage as such nuts cannot
to growth. germinate.
Most shed leaves and other plant parts decompose
beneath the individual plant from which they came and,
following Monk (1966), Owen and Wiegert (1976)
The release of water
suggest that by extending the season of decomposition
Provided a tree has a long tap root that penetrates to the consumers benefit the plant by making available a more
water table, the supply of water will not be a limiting constant and reliable source of nutrients to its roots.
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms
Effect of consumers on plant shape and structure Grass and grazers
Boscher (1979) describes how the larvae of the leek Leaf formation in grasses continues after cropping be-
moth, Acrolepiopsis assectella, feeding on the leaves and cause the meristem is positioned close to the ground
stems of the leek, Allium porrum (but not on the flow- where it is out of reach of most grazers. Young leaves
ers which are toxic (Arnault 1975)), not only have no that have had their tips removed by grazers soon resume
ill-effect on the plant but can induce an increase in the growth. The leaf tip is the oldest and the base the
number of fertile fruits. Larvae feeding on leaves youngest part of a grass plant. Most grasses are also
stimulate the plants to produce extra flowering stems. capable of vegetative as well as sexual reproduction. In
Off the Florida coast, the wood-boring marine no other group of plants are there such effective devices
isopod, Sphaeroma terebrans, feeds in the roots of the for recovery after defoliation.
red mangrove, Rhizophora mangle, and causes branch- Grasses flourish only where there are grazers. If graz-
ing in the aerial prop roots; the mangrove benefits by ers are excluded grasses do less well and may disappear
being better supported (Simberloff et al. 1978). altogether and be replaced by dicotyledons. Compared
These are just two examples of what I believe to be to other plants, grass is remarkably palatable and higher
such a widespread phenomenon that it has been largely levels of consumption can be sustained that in any other
overlooked. Browsing mammals often feed on the terrestrial plants (Wiegert and Owen 1971).
growing points of trees and shrubs and induce lateral The family Gramineae, which contains about 600
branching. The distinctive shapes of many trees in the genera and 7500 species, is probably relatively recent in
African savanna are the result of browsing by mammals evolutionary origin. Grasslands first developed during
such as giraffe, black rhinoceros, and many antelopes. the Miocene, about 25 million years ago, and at the
Many insects feed chiefly at the growing points or on same time there was a spectacular radiation of grazing
new leaves. In some plants, perhaps the majority, the mammals. It is often implied and sometimes stated
leaves become less palatable as they get older, de- categorically that grasslands are climatic and/or edaphic
veloping a tough texture and elaborating chemical com- climax communities, although there is also the suspicion
pounds that act as insect deterrents. Feeny (1970) in- that many contemporary "natural" grasslands may have
terprets the concentration in spring of moth caterpillars been at least in part created by man by grazing domestic
feeding on oak (Quercus) to the fact that the leaves cattle and by burning. Textbooks of vertebrate evolu-
contain more protein and less carbohydrate and have tion (e.g. Young 1975) envisage forest mammals mov-
not yet developed tannins as a defence against insect ing to the grassy plains of the Miocene and then evolv-
attack. I agree, but is it also possible that by attacking ing adaptations for grazing.
new leaves, especially at the growing point, lateral My view is that grasslands are climax communities
branching is induced to the benefit of the tree? In other maintained by grazing and that climate and soil-type are
words the palatability of young oak leaves may be as relatively unimportant. Grasslands persist in a state of
important a strategy as the unpalatability of older "permanent succession", and the grass plant is as well
leaves: leaf-eating larvae determine the growth form of adapted to grazing as the grazer is adapted to cropping
the tree; the tree acquires its characteristic shape, and and digesting grass. Miocene mammals did not move
maximizes its ability to photosynthesize and produce from the forest and evolve into grazers; instead the
seed. mammals created the grasses and grasslands, just as
I therefore suggest that plants and the insects and grasses and grasslands created the mammals.
other animals that attack their new leaves and growing
points are co-evolved to an extent that what we now
recognize as a distinctive plant shape and structure
Conclusion
would not have been possible to evolve without con-
sumers. There is a considerable literature that has a There is on the island of Madagascar an orchid with
marginal bearing on this topic and, moreover, common such a peculiar floral structure that it could only be
experience shows that simulated browsing induces pollinated by a large moth with a very long tongue. In
branching and the production of more flowers. Garden- 1891, before the moth was discovered, Alfred Russel
ers will be familiar with the practice of constantly prun- Wallace, who with Charles Darwin first put forward the
ing rose bushes and other shrubs, not only to extend the theory of evolution by natural selection, predicted the
flowering season but also to increase the number of moth's existence after considering the structure of the
flowers produced. Here again there is need for experi- orchid. He wrote, "That such a moth exists in Madagas-
ment. All that initially needs to be done is to space out car may be safely predicted, and naturalists who visit
at regular intervals young bushes with a single flowering that island should search for it with as much confidence
stem and then after the first growing season cut off the as astronomers searched for the planet Neptune". Wal-
tops of every second bush and record the subsequent lace was right: the moth, Xanthopan morgani praedicta,
success of the bushes in terms of flower and/or seed with a 25 cm long tongue, was eventually found as the
production. pollinator of the orchid.
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms
I think that it should be possible to find a plant where the grass benefits from being grazed we have a neat
much the same sort of prediction can be made about the co-evolutionary relationship between it and the grazer.
nature of its consumer. It might be that the search Nonsense? Possibly, but in trying to develop a
should be concentrated on grasses as it is these, more hypothesis like the one contained in this paper it is in-
than other plants, that tolerate high levels of consump- evitable that I should become too enthusiastic and that
tion and seem to be so closely adapted to the animals some of my guesses are wrong. Does anyone have an
that eat them. What we want to find is a species that alternative idea as to why a grass should synthesize vit-
simply cannot survive in the absence of a specific con- amin D3?
sumer.
Giant tortoises, Geochelone gigantea, used to occur Acknowledgements - I thank C. Gibson and S. A. Renvoize for
on many islands in the western Indian Ocean but were information about Aldabra and its vegetation, D. J. Duthie,
exterminated by man, except on Aldabra atoll where R. E. Yorke and A. G. Smith for much helpful discussion, and
especially R. G. Wiegert who from the beginning has shared
there is still a large population (Bourn 1976). The tor-
my faith in the hypothesis. Finally, I thank S. Ulfstrand and the
toises are grazers which also browse. Renvoize (1971, other organizers of the Uppsala Conference for what was a
1975 and in litt.) states that there are 176 species of most rewarding week of speculation.
plants on Aldabra, 17 of which are endemic. Among
them are 23 species of grass and 9 species of sedge
(which are related to grasses and share many of their References
characteristics); three of the grasses and two of the
Alexandre, D.-Y. 1978. Le r6le disseminateur des el6phants
sedges are endemic. Thus about a third of Aldabra's
en Fort de Tai, C6te - D'Ivoire. - Terre Vie 32: 47-72.
endemic plants are grasses and sedges. Other coral Arnault, C. 1975. Influence des organes floraux d'Allium
atolls in the region have very few endemic species porrum L. sur le developpement de la teigne du poireau
which, it is believed, is because they are low-level reefs (Acrolepia assectella Zell.). - C. R. Acad. Sc., 280:
2487-2488.
and hence unable to support a rich flora (S. A. Re-
Bacon, J. S. D. and Dickinson, B. 1957. The origin of
nvoize in litt.). But it might also be because there are no melezitose: a biochemical relationship between the lime
longer any tortoises. There is thus the intriguing possi- tree (Tilia spp.) and an aphid (Eucallipterus tiliae L.). -
bility that the endemic grasses and sedges of Aldabra Biochem. J. 66: 289-297.
Blackman, R. L. 1979. Stability and variation in aphid clonal
not only persist because they continue to be grazed by
lineages. - Biol. J. Linn. Soc., 11: 259-277.
tortoises but also that their evolution is the result of Boscher, J. 1979. Modified reproduction strategy of leek Al-
grazing selection by tortoises. lium porrum in response to a phytophagous insect, Ac-
In a paper as speculative as this it is easy to become rolepiopsis assectella. - Oikos 33: 451-456.
Boucher, D. H. and Sork, V. L. 1979. Early drop of nuts in
carried away with an idea and to seek all manner of
response to insect infestation. - Oikos 33: 440-443.
evidence, however flimsy, which seems to add support Bourn, D. M. 1976. The giant tortoise population of Aldabra
to the idea. Because the paper is speculative and be- (Cryptodria: Testudinidae). Part 1. Preliminary results. -
cause it is a summary of a lecture delivered at a confer- Zool. Afr. 11: 275-284.
Delwiche, C. C. and Wijler, J. 1956. Non-symbiotic nitrogen
ence where the participants were deliberately chosen
fixation in soil. - Plant-Soil 7: 113-129.
because of their willingness to speculate, I trust I will be Dighton, J. 1978. In vitro experiments simulating the possible
forgiven for putting forward one more suggestion fates of aphid honeydew sugars in soil. - Soil Biol.
which, if nothing else, will indicate what I think about Biochem. 10: 53-57.
Feeny, P. 1970. Seasonal changes in oak leaf tannins and nut-
plant-consumer relationships.
rients as a cause of spring feeding by winter moth caterpil-
Cholecaliferol (vitamin D3) is involved in the regula- lars. - Ecology 51: 565-581.
tion of phosphorus and calcium metabolism in mammals Gray, H. E. and Fraenkel, G. 1953. Fructomaltose, a recently
and other vertebrates. It is acquired by eating or by discovered trisaccharide isolated from honeydew. - Sci-
ence 118: 304-305.
synthesis in the presence of ultraviolet radiation from
Jones, K. 1974. Nitrogen fixation in a salt marsh. - J. Ecol. 62:
sunlight. Vitamin D3 also occurs in two species of Sol- 553-565.
anaceae and in the grass, Trisetum flavescens. In the Llewellyn, M. 1972. The effects of the lime aphid, Eucallip-
grass the remarkable discovery has just been made that terus tiliae L. (Aphididae) on the growth of the lime Tilia x
vulgaris Hayne. 1. Energy requirements of the aphid
it is formed only in the presence of ultraviolet radiation,
populations. - J. appl. Ecol. 9: 261-282.
suggesting a similar mode of synthesis in plants and Michel, E. 1942. Beitrage zur Kenntnis von Lachnus
mammals (Zucker et al. 1980). The discoverers were (Pterochlorus) roboris L., einer wichtigen honigtauerze-
unable to suggest why a grass might need to synthesize ugerin an der Eiche. - Z. angew. Ent. 29: 243-281.
Monk, C. D. 1966. An ecological significance for evergreen-
vitamin D3.
ness. - Ecology 47: 504-505.
To me, the explanation is obvious: it is an attractant Owen, D. F. 1978a. Why do aphids synthesize melezitose? -
for grazers which require vitamin D3, serving the same Oikos 31: 264-267.
sort of function as nectar does for a flower. Imagine a - 1978b. The effect of a consumer, Phytomyza ilicis. on
seasonal leaf-fall in the holly, flex aquifolium. - Oikos 31:
grazer short of vitamin D3. In all probability it locates
268-271.
specific grasses by smell and so all that is required is for - and Wiegert, R. G. 1976. Do consumers maximize plant
the grass to produce a characteristic aroma. Provided fitness? - Oikos 27: 488-492.
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms
Petelle, M. 1980. Aphids and melezitose: a test of Owen's Wiegert, R. G. and Owen, D. F. 1971. Trophic structure, av-
1978 hypothesis. - Oikos 35: 127-128. ailable resources and population density in terrestrial vs
Renvoize, S. A. 1971. The origin and distribution of the flora aquatic ecosystems. - J. Theor. Biol. 30: 69-81.
of Aldabra. - Phil. Trans. Roy. Soc. Lond. B. 260: Witkamp, M. 1971. Soils as components of ecosystems. - Ann.
227-236. Rev. Ecol. Syst. 2: 85-110.
- 1975. A floristic analysis of the western Indian Ocean coral Young, J. Z. 1975. The life of mammals. - Clarendon Press,
islands. - Kew Bull. 30: 133-152. Oxford.
Simberloff, D., Brown, B. J. and Lowrie, S. 1978. Isopod and Zucker, H., Stark, H. and Rambeck, W. A. 1980. Light-de-
insect root borers may benefit Florida mangroves. - Sci- pendent synthesis of cholecaliferol in a green plant. - Na-
ence 201: 630-632. ture, Lond. 283: 68-69.
Stenseth, N. C. 1978. Do grazers maximize individual plant
fitness? - Oikos 31: 299-306.
This content downloaded from 128.192.114.19 on Sun, 26 Jun 2016 01:28:59 UTC
All use subject to http://about.jstor.org/terms