To Make as
Perfectly as Possible
Roubo on Marquetry
Donald C. Williams
Michele Pietryka-Pagn
Philippe Lafargue
l o s t a rt pre s s fo rt mitchell
2013
Andr-Jacob Roubo
To Make as Perfectly as Possible: Roubo on Marquetry
A translation of LArt du menuisier
Published by Lost Art Press LLC in 2013
26 Greenbriar Ave., Fort Mitchell, KY 41017, USA
Web: http://lostartpress.com
Copyright 2013 by Donald C. Williams
ISBN: 978-0-9850777-5-4
FO R EWO R D
W
he 18th century in Europe has been recognized by historians as the Age of Enlightenment.
The major feature of this movement was the adoption of reason and science as the basis for
knowledge. Philosophers, historians, academics and encyclopedistes worked to compile treatises
on diverse aspects of life, including the trades, with this new perspective in mind.
In France, under the direction of Jean-Baptiste Colbert and support of Louis XIV, the Acade-
mie Royale was created with the purpose of publishing these treatises and maintaining an official
library of information. These publications, beginning in 1709 and numbering nearly 30 by mid-century, were known as the descriptions des arts et metiers. Essentially, every specific skill and trade
known at that time was analyzed and documented in great detail, including woodworking and furniture making.
One of the most recognized publications to include a description of the various trades was pub-
lished as a subscription series in 1756 by two authors: Denis Diderot and Jean le Rond dAlembert.
They published lEncyclopdie ou Dictionnaire Raisonn des Sciences des Arts et des Metiers in 1756, with
engravings to illustrate various aspects of the topics, such as woodworking, listed in alphabetical
order. Neither man had any specific training in the trades, as Diderot was a philosopher, art critic
and writer, and dAlembert was a mathematician, mechanician, physicist, philosopher and music
theorist. Their goal was not to discuss any specific topic in extensive detail, but to analyze, in an alphabetical format, all the important terms of the day.
It is interesting to compare the Encyclopdie of Diderot with the extensive treatise on woodwork-
ing published at about the same time by Andr-Jacob Roubo. Diderots work appeared independently of the Royal Academy, and it involved 160 authors who contributed entries to the 17 volumes of text
and 11 volumes of illustrations. Roubos work was by a single author, accepted by the Royal Academy, and it focused entirely on diverse aspects of woodworking in four volumes, with excellent illustrations, drawn by Roubo personally.
The first indication that LArt du menuisier represented a major contribution to the advancement
of knowledge appears as a two-page commentary printed at the beginning of the first volume. Immediately after the title page is the Extrait des registres de l'Acadmie Royale des Sciences, dated Dec.
17, 1768. These two pages serve to certify that the contents of A.-J. Roubos work had been reviewed
and accepted by the Royal Academy in France, unlike that of Diderot and dAlembert.
This certification is highly significant, as Mr. Roubo was not a philosopher, or other academic,
but trained as a furniture maker, and was referred to by the Academy as a Compagnon Menuisier.
In fact, he was documenting, in exhaustive detail and accuracy, the diverse facets of woodworking
that he had spent his life pursuing as a profession. He had been trained in the traditional methods
of furniture making as a member of the Compagnon, which was a historic system of learning that
v
vi
relied on transmission of knowledge from master to apprentice, working together at the bench.
Roubo was the highly educated son and grandson of a family of master woodworkers, and had
spent his life in the trade. He knew how to identify the wide variety of woods available and which
wood was appropriate for a specific trade or function. He used the tools of the trade on a regular basis
and could clearly discuss their individual features and function. He was trained in geometry, design
W
and construction, not just by reading other books, but more importantly by direct instruction from
other masters working in different fields.
In fact, the Royal Academy preface, which endorses Roubos contribution to the knowledge
of woodworking, specifically mentions that his work is written with a unique perspective, clarity
of language, and supported by the excellent engravings, thus making it accessible to the ordinary
worker. Que l'Acadmie seroit satisfaite si dans tous les Arts il se trouvoit des Ouvriers capables de
rendre aussi bien les connoissances quils ont acquises par un long exercice! [May the Academy be
satisfied if in all the Arts, Workers were to be found capable of rendering rather well the knowledge
that they have acquired by a long practice.]
The title, LArt du menuisier, is important to translate properly, as Art is intended to mean
knowledge, or understanding, and specifically of a trade, or metier. The Menuisier term is the
title of a trade of woodworking that included everything that the carpenter did not do. For example,
the carpenter would frame the house, but the menuisier would be involved in unique aspects of architecture, like building a dome or staircase, or complicated wall paneling and trim, as well as a specialist in making all types of solid-wood furniture. The only type of furniture that was not made by
the menuisier was the furniture that involved veneering, ranging from plain veneer to very complicated marquetry surfaces. These pieces were made by an bniste, and the famous Andr-Charles
Boulle, named first bniste to the king in 1672, established the standard for this trade, which continues to modern times. It is remarkable that Roubo, a Compagnon Menuisier, was able to communicate in accurate detail such complicated and specific aspects, not only of the trade of menuisier,
but also that of the bniste.
From the day of its first publication, LArt du menuisier has been accepted as a trusted point of ref-
erence for the way things are made out of wood. This work is included in many libraries, and noted
as a legal and professional benchmark that has stood the test of time. Until now, it has never been
translated into English, and the significance of this first English edition cannot be overestimated. For
the very first time in more than two centuries, the knowledge and understanding of woodworking
from an 18th-century French perspective can be enjoyed by those who do not speak French. These
volumes will immediately be recognized as a valuable reference for the modern woodworker who is
interested in understanding exactly how his craft was practiced so many years ago.
Respectfully submitted:
W. Patrick Edwards
Patrice Lejeune
The American School
of French Marquetry
Translators Note
W
Once Don Williams got wind of what I was doing for this colleague, he asked if I would be inter-
ested in translating a chapter of Andr-J. Roubo for him. As I was a novice adult student of French,
having returned to studying the language at night school after only three years of instruction in high
school, I took this as a challenge. It would be a way to improve my reading and understanding of
French sentence structure how much different could 18th-century French be, anyway?
As with any new project, the first part was the worst, even if Don had not asked me to translate
his very favorite topic first: tools. I didnt know what these things were in English, never mind describing the parts of them in French. Because I was so unsure of myself in the beginning, I tried out a
paragraph or two with three different native French speakers. I would produce my translation, then
give the original French text to them and ask them to translate it. Without fail, they had quite a bit of
difficulty, and they left out entire phrases they deemed extraneous to the meaning of a sentence or a
paragraph. This, however, gave me courage to keep on doing what I was doing because I always got
the essence of the paragraph correct.
My technique was as follows: I read each paragraph about five times. During the first read-
through, I changed all the long s characters to s, because as you probably know, in the 18th century, the s could look much like the f. On the second reading, I circled all the words that I didnt
know, which was a considerable number, given all the technical vocabulary involved in describing
parts of tools, what they do and how they must be exactly the right size or they dont work properly.
With the third reading, I substituted the word I had found in the dictionary into the sentence to
see if the sentence sounded correct. This in itself is a challenge, because French has fewer words
to describe something than English. The English language is much more precise for example, we
might have 15 different words to express the emotion happy. French has fewer words, so context
is important. The fourth reading, I cut the original sentence into several, changing or eliminating
conjunctions as necessary. In 18th-century French, each paragraph might consist of one very long
sentence. So the trick was to cut the paragraph into several sentences, yet retain the meaning that
Roubo wanted to convey. The fifth reading followed the typing and was a check to see if the entire
paragraph made sense. After a couple years and a couple of chapters of doing this, I was able to skip
writing out the new paragraphs on lined paper, and just read and type directly.
There were times when I would translate a sentence and write into parentheses: (Don, is this
vii
viii
making any sense at all?). At other times I would write: (Philippe, HELP!!!). Believe it or not, once
you get into the rhythm of this kind of work, and the vocabulary starts to be retained in your brain, the
process starts to flow, and you really are into some sort of groove, or rainure, rayure, cannelure
or orniere, as Roubo would say.
I was working with seven dictionaries set out around myself, two of them dating back to the 18th
W
century. They were a steal on eBay, because, after all, who would be interested in buying French dictionaries from that period? One dictionary was even a French-French Larousse, which I was proud
to say I could use my French had improved such that I could look up a French word, and understand
the explanation also in French!
I think one of the most painful and difficult phrases encountered was in the tools chapter, when
Roubo was describing the workbench. He went on and on concerning the bote a rappel. Well
bote means box, so he was describing a box having some purpose. Rappeller is the verb meaning
to call again, to call back or to call to order; to restore to life; to remind or call to mind; to recover; to
summon up courage; to recall; to quote; to reference; to distribute; to bring together; to call off; to
draw back a piece of something. Nowhere did I ever find that the bote a rappel referred to the vise
attached to a workbench! It was in these kinds of situations that I had to ask Don to please look at
the engraved plate in question, inspect the particular diagram on this plate and tell me what Roubo
was describing.
The footnotes were the most interesting part for me because it was there that Roubo told us about
his own childhood, growing up poor, going to work with his father at age 11, and having a grandfather who lived into his 80s, and when he died he wasnt sick. It was also where he commented on
French society at large and the place of carpenters and cabinetmakers within that society, which
was rapidly losing the appreciation for the highly skilled work that he and his workers accomplished
within that society.
Its been great fun, and a huge character-building experience for me. Don and I both understood
that someone like Philippe Lafargue was an absolute requirement for this project; but in the meantime, I was propelled forward on this project by the thought that I was unlocking a little bit of new
information with each sentence translated.
Au revoir, and Bonne chance!
Michele Pietryka-Pagn
PR EFACE
W
n 1975, I was working as a finisher and restorer at the renowned Schindler & Son shop in Palm
Beach, Fla. The old man, Pop Schindler, had started the company on the cusp of our first
Great Depression and had somehow managed to keep the doors open, in great part thanks to his
incredible depth of knowledge and skill as a traditionally trained Swiss apprentice.
One day an old-money Palm Beach client (Ambassador Something-or-other) pulled up with
boxes full of parts for what looked like just another old piece of junk to put back together. It was, in
fact, a simple (for him) tulipwood parquetry secretaire by Jean-Henri Riesener (1734-1806), successor to bniste du roi Jean-Franois Oeben, and cabinetmaker to King Louis XVI. After weeks of
our painstaking reconstruction of the former pile of debris, as the piece was finished and awaiting
delivery, Pop made a remark that puzzled me.
Then he told me about LArt du menuisier. Pop did not own a copy, but the shops most important
patron (a renowned collector of French decorative arts) did, he said. Someday when we were over at
the estate together he would ask to show it to me. That day never came, and I did not see the complete
LArt du menuisier with my own eyes until almost 10 years later.
The mid-1980s provided fertile soil for growing the organism that became To Make as Perfectly as
Possible.
Philippe Lafargue entered my studio in 1985 as a post-graduate Fellow from cole Boulle. During
his tenure with me he whetted my appetite further by translating the Table of Contents of LArt du
menuisier and described some of the sections on marquetry.
At about that same time I met and collaborated with Michele Pagn, a textile conservator who
was developing a collections preservation strategy for the heritage artifacts under the jurisdiction
of the U.S. Department of State. That collaboration would eventually come to fruition in the book
Saving Stuff (Fireside/Simon and Schuster, 2005).
So my passion for the scholarship compiled by Andr-Jacob Roubo festered, waxed and waned
for nearly four decades. On one hand I was hampered by a near total lack of foreign-language skills,
but still I was drawn to the unmistakable appeal of the images, which are themselves a great education. Over the years bits and pieces of translated text appeared in disparate places, invariably fanning
the flames of my interest.
Rob Tarules popularization of the Roubo workbench was a critical inspiration, and I recall
spending time with him at a conference where we were both presenters as he walked me through
designing, building and using the tool and the processes it engendered.
By 2008, Michele had completed two decades of globetrotting with her U.S. Navy officer husix
band and they had re-settled in Washington, D.C., and we resumed collaborating on many interesting projects. On our way back from the lunchroom one day she stopped to talk to another colleague
and the conversation revolved around some translation Michele was doing of a historic French treatise. I had not known until then that Michele did translations from arcane technical French.
To say the least my ears perked up and a light went on in my head. I was starting to work on a monoW
graph about historic finishing and wanted to know about some odd tools represented in Roubos plate
296. Could she help explain what they were used for?
After looking at the text plate in the gigantic volume, she said, Sure. Our work began in earnest
in mid 2008, and five years later the first of our volumes is making its debut.
Eventually we pitched an idea to Chris Schwarz and he graciously agreed to partner with us in
bringing Roubo to the modern woodworker. Rather than translating, annotating and interpreting
the whole of LArt du menuisier, we decided to focus on, well, the parts that interested me: marquetry,
finishing and furniture making.
Our working regimen began with Michele translating Roubo as verbatim as possible, making no
alterations to the original syntax unless that rendered the English incomprehensible. We decided
that working from a transliteration like this was the best way to capture both the information and the
flavor of the original. To assist her I frequently had to identify every tool in every engraved plate so
she would know what to call them.
Then the text came for me to edit, enhance and reconfigure as necessary to make it readable to an
artisan of the 21st century. I did not rewrite Roubo, but merely modified it enough to make it comprehensible and read smoothly, and I inserted annotations explaining some of Roubos text. It took
us more than two years to get through the first draft.
It was immediately apparent that Roubo assumed a large body of knowledge already in possession
of the reader, and as a result left many gaps in his narrative. To make our volume as useful to as many
people as possible, the original text was augmented extensively to make it more understandable. In
addition, I began to select passages needing further visual presentations through photographic essays.
During the subsequent rounds of editing, the manuscript went back and forth numerous times to
make sure we were getting it right, with each iteration taking weeks to complete. Only when we were
both satisfied with the English text was the manuscript sent to Philippe, who had trained as a traditional craftsman the cole Boulle in Paris. He was, and remains, well-versed in the arcane jargon of
ancient French cabinetmaking, which is fortunate since some of the phrases Roubo used were simply untranslatable otherwise. Philippe reviewed the text from the perspective of a native Frenchman
and historical craftsman to make sure the new English version would meet with Roubos approval.
But this was not the end of the labors. I realized that my immersion in the text had dulled my
perspective on whether or not something would make sense to you, the reader. To alleviate these
concerns I recruited three long-suffering friends whose woodworking, communication and critical-thinking skills were beyond question. Never-ending thanks are due to Robert Klein, Michael
Mascelli and Martin OBrien for their diligent review of the tens of thousands of words and hundreds of illustrations.
These readers had an entirely different responsibility than Michele (transliterating historical
text into English), me (editing and annotating it to make sure it was comprehendible in 21st-century
English), and Philippe (fealty to French craft traditions). Their charge was to ensure that To Make as
Preface
xi
Perfectly as Possible conformed to the original goal of being more than a simple recitation of historical
materials and techniques. It also was to serve as a contemporary guide for todays artisans wishing to
employ the techniques of 250 years ago.
These readers task, then, was to provide critical feedback from the artisans perspective on
whether or not the manuscript accomplished that goal. In some cases it was as simple as determining
W
whether some passage of text actually made sense; in others it was to suggest additional or different
illustrations for the photo essays demonstrating the techniques.
When Chris and I first spoke of this project, I stated the teams goal as, ... to let the reader practi-
cally experience the sounds of the saws and fragrance of the wood shavings and glue pot in the shops
where Roubo worked.
I am grateful to the folks at Lost Art Press who have contributed so much to seeing this through
the press: editor Christopher Schwarz; copy editors Megan Fitzpatrick, Mary Hashman and Linda
Watts; and John Hoffman who has handled the subscriptions and distribution of the copies.
Finally, let me make a comment about the numerous tools constructed and exercises photo-
graphed for this volume. These are not intended to be slavish repetitions of that which Roubo describes, but rather to provide examples of how a modern craftsman might go about working in this
traditional manner. Most of these photo essays are neither overly detailed nor complete, but are intended to help point you in the right direction to find your own path for working creatively, making
things as perfectly as possible.
Donald C. Williams
A K ey to the T ext
W
hile reading To Make as Perfectly as Possible, here are some helpful tips about the structure
of the book and some unfamiliar terms used in this published English translation.
The pages are set up much like the original 18th-century edition. As in the original printing, the
margins of the pages contain references to page numbers and plates. Page numbers refer to the pages
in the French work; they allow readers to find passages in the original and compare our English version with Roubos French.
The plate numbers also duplicate the structure of the original. In the French text, the plates were
printed in a separate section of the book, so the plate numbers helped the 18th-century reader locate
the illustrations that matched the text. We have included the plate numbers because Roubo also used
them as a way to jump to a new topic.
At times you will find different kinds of parenthetical comments in the text. Comments that are
surrounded by parentheses are written by Roubo. Comments in brackets are from the translators.
Footnotes in the text are from the original French edition and are Roubos words.
Modern readers might also be bemused by the number of words in italics and the structure of the
individual sentences. The italicized words are reproduced from the original. The goal of this translation was to reproduce the structure and feel of the original French work as much as possible not
to edit Roubo into 21st-century English. As such, you will find some sentences that have an elaborate
structure. You will quickly develop an ear for the way Roubo writes.
You also will find essays inserted in the book that were written by Donald Williams. These essays, which explore a tool or process in detail, are set in a slightly smaller typeface. These essays are
accompanied by color photos and, of course, the voice changes to one of a 21st-century woodworker
and researcher.
Instead of converting all of Roubos measurements to U.S. Customary Units (or metric), we de-
cided to use his original terms. As such, you will find the units of thumbs and lines. A thumb is
just slightly more than our modern inch 1.066". The thumb is further divided into 12 lines. Each
line is equivalent to .088" today. The French foot is 12.44".
Christopher Schwarz
Editor
xii
he woods appropriate for cabinetry are of two types: those that serve the construction of the casework and those that serve as dressing up the casework. The woods
Chapter 10
appropriate for frames are oak, pine, holly and all other soft woods. They should be
as dry as possible, as I will explain when discussing the construction of the frames
Page 766
There are two types of wood used for the veneers on cabinetry, namely those of color, most of
which come to us from the Indies [Tropics], and the woods of France, of which some are of beautiful
enough color, but of which the greatest number need to be tinted to be employed for these sorts of
works.
The woods of the Indies are preferable to those of France, not only by their great number, but also
for their beautiful colors and their good qualities that make them well suited to receive a polish. That
is why I am going to begin with a description of the woods from the Indies, because they are the most
complicated and the most interesting, given that it is those that are the most in use at present.
The woods from the Indies appropriate for cabinetry are in great number and are very different
from each other, whether by their colors, their different given names, the appropriate area for each of
them or the different countries where they grow. One can generally consider them by reason of their
colors as having five different species that we knew formerly under the name of ebony, that is, black
ebony properly said, the red ebony, the violet, the yellow and the greens; if, however, one can give
this name to the woods where the color is more a dirty yellow, or olive brown, than green.
This manner of considering the different types of wood from the Indies is altogether natural. As
each of these woods is not exactly the color red or violet, etc., but more nuanced in these different colors (at least for the most part), I have preferred to follow the alphabetical order in the description that
I am going to make, that being the most used, and one that follows those which are already written on
this matter. *
* In working on the description of these woods from the In-
piece of wood what is the real name and from what country
almost all the authors who have written on this subject do not
to the names, the quality or the color of the woods. The mer-
the woods that belong to me, and those of the Natural His-
Page 767
Section I
and
D e sc r i p t i o n o f t h e W o o ds o f t h e I n d i e s ,
T h e i r Q ua l i t i e s , R e l at i v e t o C a b i n e t m a k i n g
W
efore making the description of each type of wood of the Indies, I have made the following
Page 767
table so that with one glance one can know their names, their colors, their qualities strong or
soft, or aromatic, and the country where they grow, which will help much in the knowledge and accuracy of the description of these same woods.
No. 1. Acaja [Anacardium occidentale, cashew]: is grown in Ceylon, and has a fruit similar to that
of prunes; its wood has a rough bark and is soft like elderberry [Sambucus nigra]; it is red and light like
the cork [in reference to Quercus suber], and is only fit for small pieces.
2. Acajou or acajous [Swietenia mahoganii, true mahogany]: This tree is a type of walnut that grows
in Malabar, and is originally from the American Islands and from Brazil; in Cayenne, Jamaica, the
mahogany wood properly said, is soft with an odor a little strong without being absolutely bad, of a
reddish color, of a grain sometimes so twisted [interlocked] and edge grain that one cannot work it
except with rasps. This type of mahogany is called Apple Cashew [Anacardium occidentale]; it is less
odoriferous than the other, seasons quicker and will brown [oxidize] quickly.
The mahogany of Cayenne is grown high and wide, and is appropriate for making into planks;
its color is reddish, sometimes veined or marbled with yellow and white. This wood has a good odor,
polishes well and is sometimes named Cedar of San Domingo.
The mahogany of Jamaica is of a brown color with a slightly reddish stripe with deep brown fol-
lowing the concentric layers of the tree; this produces very beautiful accidents in the area of knots,
where the concentric layers follow the variations. In general, the grain of mahogany is fine, its pores
a little open, especially at the end grain, that is to say, [at] the extremity of the longitudinal fibers,
which make this wood not as solid and even a bit subject to rottenness related to the humidity, which
is introduced in the opening of these pores.
3. Aloe: [We had difficulty identifying the proper Latin binomial for this.] This wood is very rare
and is called by Dioscorides as Agalochum. There are three types of aloe: the Calambac, which grows
in Cochinchine, is very soft and of many colors, [and] of a very pleasant odor. This wood is extremely
tory Department of the Kings Garden, which has put me
least from gross mistakes. This effort has for its object simply
Of Foreign Woods
resinous, which makes it so it melts more than burns, and one makes no use of it except as a perfume.
The second type of aloe is called Eagle Wood or Agalochum Sauvage; it grows in Cochinchine,
Cambaye and Sumatra; its wood is compact and heavy, pierced with many cavities, is a reddish color
and is of a very nice odor.
The third type of aloe and the most common is called Calamboure or Calambour, and comes in
W
large logs from the islands of Solor and Timor; it is a tree that resembles the olive tree [Olea europaea];
its wood is light, porous and resinous of a greenish color, leaning toward reddish.
4. Amaranthe [Peltogyne spp.], called by the English Mahogany; this type of violet wood is sold to us
by the Dutch, and is called sometimes Wood from China, albeit falsely, because this wood grows only
in Guyana in South America. This tree grows very large, and its wood is of a grain fine and tight; its
color, before being worked, is veined grey and brilliant, as if it were silver; once this wood is polished,
Page 771
its color changes and becomes a beautiful purple-brown, which with time, becomes almost black.
This change is caused by the evaporation of a whitish, resinous substance, which is found deep within the longitudinal fibers, and which appear at the end of a piece, as if an infinity of little white points
follow the concentric rings of the tree. The Amaranthe wood is of moderate hardness, is worked very
well and is greatly used presently, where we use it commonly with rosewood, on which it contrasts a
little too much, as I will explain when speaking of the way to mix the woods.
5. Amourette is a heavy, hard and compact wood, of a yellowish color, a bit reddish, and veined
with reddish-brown; I suspect that this could be the same thing as Benoit fin, which grows in the
Antilles, and which becomes very tall and very large.
6. Star Anise or Anil a Ltoile is a tree grown in China, in the East Indies, the Philippine Islands and in Siberia, from where we bring it in large logs. This wood is of a grayish color and of an
odor a bit like anise [black licorice], and is little used in cabinetry, although we use maple tinted grey.
7. Asphalatum, named by the ancients Rhodium Lignum, or Wood from Rhodes. We do not know
exactly if it is the wood known under the name Rhodes Wood or of Rose, or even if it was an Aloe or
Agalochum.
8. Brasilwood [Cselpina echinata]: This wood comes originally from Brazil, province of the Meridional Americas; it takes different names according to the other places where it grows. That which
is called Brasil of Fernambouc is the best. There is also the brasilwood properly called that of Lamon,
or San Marthe; and the Bresfillet, which grows in the Antilles and which is the least esteemed of all.
The brasil tree is of two species: namely the large one is called Sapan, and the little one is called Sapanbimas. Brasilwood has no pith, is often twisted and comes hugely large. As it has a lot of sapwood,
it loses half and even two-thirds of its size. For this wood to be good, it must be compact, strong and
very dry, and its color be on grain [this expression means, in general, parallel to the grain lines].
When it is split, the grey that it appears becomes a red leaning toward yellow, and it has a taste a little like sugar. It also grows in the East Indies: like Japan, where it has a pith, in Siam, on the coast of
Malabar, and in the two peninsulas of the Gange. This wood cannot grow outside of the Torrid Zone.
The Indian Sapan referred to here is a different variety than that of America, but it is used equally in
woodworking and dyeing.
9. Cinnamon [Cinnamomum zeylanicum], which is called incorrectly Sassafras. This is a tree that
only grows on the island of Ceylon, and which grows to only 4 fathoms [24'] tall. Its wood is hardgrained, white and has no odor; there are in old trunks of these trees nodes or knobs appropriate
for cabinetry.
Page 772
Alphabetic Table of Foreign wood known under the name Wood of the Indies,
and which are Appropriate for Cabinetry;
Page 768
Names of Wood
Country or Origin
Color
Quality
1.
Red
Tender, soft
2. Mahogany
Rosey
3. Aloe, Agalochum
Many colors
Rust
Solid
Aloe, or Calambour
Greenish
Soft
4. Amaranth
Violet brown
Hard
5. Amourette
Red Brown
Hard
6. Star Anise
China, Philippines,
West Indies
Grey
7.
Cashew
Asphalate,
or Rhodes Wood
Soft
8. Brasilwood
Solid
9. Cinnamon or Sassafras
Hard
10. Cayenne
Solid
Aroma
Aromatic
Aromatic
Aromatic
11. Cedar
Syria and America
Reddish and veiny
Solid, incorruptible Strong, soft
12. Cedar
Asia, America, Siberia,
whitishred
moist
Fragrant,
Europe
as above
13. Snakewood
China, East Indies, and
Redbrown
hard
Guyana, West Indies
touched with Black
14. Candlewood
American Islands
yellow red
solid
Aroma of lemon,
nutmeg or vanilla
15. Citron
Asia and Middle-East
White, veined
hard and
Rot-free
16. Copal
17. Coralwood
Spotted red
Solid
Porous
18. Cypresswood
Asia
Yellowish, striped
Hard,
Rot-free
19. Ebony
Very hard
Black
Black and white,
spotted
Hard
Madagascar, Africa
Hard
Hard
Of Foreign Woods
Their colors, their Qualities, hard or soft, or fragrant, and the name of the
country where they grow.
Names of Wood
Country or Origin
Color
Quality
Porous
25. Fereol
Isle of Cayenne
Solid
26. Ironwood
American Islands
Very hard
27. Fuset
Veined yellow
Soft
Very hard
White veined
With black
Very hard
31. Jacaranda
West Indies
Yellow, gold,
Veined, wavy
Solid
33. Lapire
West Indies
Solid
34. Nutmeg
West Indies
Moist
West Indies
Brown grey
Very hard
36. Olive
Syria and Mid-East
Yellowbrown,
striped
Hard
37. Orange
China, Europe
Solid
38. Platane
White
Solid
Wavy
Solid
40. Asphalate
Jamaica, America
41. Rosewood
42. Bloodwood
Antilles, America
Solid
Deep red
Hard
43. Sandalwood
white
Solid
Red mixed w/
yellow & Brown
Hard
Red veined w/
Yellow
Solid
47. Violet
East Indies
Veiny white &
Solid
Violet, striped
48. Palissander,
Isle of St. Lucie, or Alousie,
GreyBrown,
Porous
Called St. Lucie
West Indies
veiny
Page 769
Aroma
Strong
odor
Of a good
odor
Bad aroma
Of a good and
penetrating odor
Rose aroma
Odor of
Citron
Odor as above,
Less strong
Violette aroma,
very light
Nice aroma
stronger than
Violet
One grows in Florida as a rather beautiful tree, which is called Pavane, or Cinnamon Wood, be-
cause of its beautiful odor; we state that it is the same thing as Sassafras.
There is still another type of yellowish wood leaning toward brown, named Carabaccium, the
odor of which smells like the clove tree, which could well be the same as the Pavane from America.
10. Cayenne. There are two types of Cayenne woods: one veined with yellow and reddish, of
W
which the grain is fine and tight, the other being a brown-red veined and grey on the edges. Both of
these two types of wood are interspersed with small cavities filled with a type of gum or resin that
evaporates in the air. The gum follows the longitudinal fibers of the wood, which appears at the end of
a piece to contain an infinite number of small irregular pipes, which does not prevent this wood from
being polished very well.
11. Cedar [Cedrus libani]: This is one of the most beautiful and largest trees in the world. The largest used to grow on the mountain in Lebanon, where they were 120 to 130 feet in height. They do not
grow there presently except in the environs of Biblos and Tripoli in Syria. They grow well enough in
America, and in the Isles of Cyprus and Candie, but they are less beautiful than those of Asia Minor.
It still grows in Siberia, but this is the least beautiful of all, and it has no odor.
Cedar wood is of two types: namely the red and the white. The red, which is the most beautiful,
is a solid wood, mildly close-grain, of a reddish color tending toward the yellow, and where the separations of concentric layers make a brown-red tending toward violet. This type of wood, when split,
follows one of the stripes and appears striped like Japan [pine] wood, and is of a very good odor, similar to that of musk.
12. White Cedar is not striped like the red, and is of a color altogether more ginger than white and
is softer and lighter than the red cedar, and has about the same odor. In general, the cedar, however
tender, takes a polish well enough, and passes as being incorruptible [totally rot-resistant], which was
appreciated by the Romans who made good use of it for their furnishings. Seneca, the songster of the
poor, had 500 identical tables of Cedar wood supported on feet of ivory.
13. Snakewood [Lignum sinense], in Holland, Letterhout or Wood of Letters, because of the
Page 773
letters or marks one can make from the spots with which it is covered. This wood is hard, heavy and
extremely dense, taking polish well. It is of a red-brown color, marked with little brown spots or especially black, which start at the center of the tree, following the direction of the stripes. This wood
is resinous and of a very fine grain. However strong, it splits easily along the grain, which renders
it difficult to work. Whats more, its color blackens upon aging. Although this wood is called Wood
from China in France, it is highly doubtful that it comes from this country. On the contrary, we are
almost sure that it grows only in the country of Guyana, of Meridional [South] America. What is
certain is that in Guyana there grows a type of wood that is lustrous, hard and of a red color spotted
with black. Sometimes the base of this wood is yellow, but in one or the other case, it rarely grows
to 4 thumbs in diameter.
14. Lemon [Xymalos monospora], is thus named because of its odor and its color. It grows in the
American islands, at the edge of the sea. The Americans named it Candlewood, because they used
it to light the night. Its resinous wood is heavy, compact, of a strong odor leaning toward that of citron. Its grain is closed [tight] and extremely straight, and takes polish well. It is available in trunks
weighing about 1,000 pounds. That is how one can distinguish Santal citron, which it resembles,
but of which the logs weigh no more than 100 pounds. Otherwise, citrine is less heavy than the
lemonwood, of a softer aroma and more agreeable and of a better taste than the former, which we
Of Foreign Woods
also name Cocoa Wood and Jasmine. We believe that it is the same as the rosewood from Guyana.
15. Citron Tree [Citrus medica]. This tree is not large, its wood is white and without odor. It is
originally from Asia, from where it was transported to Greece and from there to Italy. It is now
grown in Spain, in Portugal, in the Piedmont of Italy and Provence. The wood of Citronnier was
very rare and highly esteemed in Rome at the end of the Republic. Cicero had a table of it that cost
W
2,000 ecus. Asinius Pollio had one of 30,000 libra aurea. Pliny says that one had to be a very great
nobleman to make use of this wood, whose beauty consisted in the variety of waves and nodules in
the roots. This wood is not used at the present.
16. Copal [Hymena Verrucosa], a tree from which comes the balm of Copahu. Its wood is of a
deep red, flecked with touches of vivid red, of a hardness almost equal to oak, and used for dyeing.
We wonder if this wood is the same as the Fernambouc, of which it has the same odor. The Copaiba
is grown in the forests of Brazil, in the island of Maragnan and in the Antilles.**
17. Coral [Ethryna spp.]. This tree is grown in the Windy Islands, in America. Its wood is of a
pale red color, striped with veins of a coral red mixed with brown, following the concentric layers of the tree. This wood is medium heavy and very porous. It has very open fibers, filled with a
cinder or gum that is more or less dark, along the veins where it is found. It has the fault of being
very difficult to work, because all the concentric layers are of a different inclination [slant], which
produces reverses [grain switchbacks], which are inevitable. They sometimes sell this wood as red
Santal. But the latter is much more compact than the wood of Coral, and is of a much deeper red.
18. Cypress [Cupressus spp.], a tree of an average size. It is originally from the mountains of Crete
and the Islands of the Archipelago. Its wood is compact, solid and the color yellow. When it is cut at
end, its concentric layers are distinguished easily from that of pine. It is not heavy and is not subject
to rot or splitting. Worms never are found there. It almost always has as pleasant an odor as cedar,
and takes polish very well. Almost all the former authors agree this wood is rot-resistant. What is
certain is that the doors of the old St. Peters Church in Rome, made at the time of the Emperor
Constantine and which were made of this wood, have lasted close to 1,200 years. There is another
type of Cypress in America, which is called White Cedar, which is not known in France.
19. Ebony. Little is known about the tree of ebony. Its wood is very hard and heavy, however
less so than Ironwood. There are four types of Ebony, namely the black, the red, the green and the
white.
Black ebony [Diospyros melanida], which is the most common, comes from Madagascar, where
the inhabitants named it Hasonmainthi, that is to say, black wood. The most beautiful black ebony
comes from the Island of Maurice, from which it takes its name. For ebony to be beautiful, it must
be without any knots, be of a much closed [tight] grain, and of a lustrous color. When working
it, the shavings rise up well without breaking as happens sometimes to a type of ebony such that
we say the wood must be burned, and where the shavings are like sawdust. Ebony has the fault of
being sometimes spotted with rosy [reddish] veins, especially that which comes from the Island of
** I have seen in the Natural History Office in the Kings
Page 774
10
Maurice, which diminishes the value because the beauty of this wood consists of its perfectly black
and lustrous color.
In general, ebony is one of the most beautiful woods that one can use in cabinetry, as much with
regards to its full and compact quality, which makes it easy to work, as by its beautiful black color,
which augments the polish that this wood takes perfectly well. It is almost exclusively used for this
W
part of woodworking [veneer work], and to which it has given its name [bnisterie]. Because other
woods of color have become more common, one makes less use of Ebony, and we do not use it much
Page 775
20. There is a black and white ebony, named Ebony of Portugal, which is little used, of which I do
ed, positioned on a brown background, such that one senses (although with a bit of difficulty) the
shape of concentric layers. The end grain of this wood cannot be better compared than with rush.
There is not, or very little, difference in its vertical stripes, whether they be cut on the radial plane
or parallel to the concentric layers in the tangential cut.
By this, this wood differs from all the others, which, while they are striped at the end grain it is on
the grain split on the radial (that is to say, following the stripes of the tree), and veined while it is split
parallel to the concentric layers. Since Epi de Bl has very open pores and its stripes [rays?] are of unequal density, it is a little difficult to polish. One must take care to fill the openings of these pores with
Of Foreign Woods
11
a mastic prepared as I will indicate when speaking of polishing woods. This observation is general for
all woods where the pores are open like that of which I am speaking.
25. Fereol. This wood grows in Cayenne, and carries the name of he who discovered it. It is also
named marbled wood; the base of this wood is white and veined or spotted with red. There is, at the Office of Natural History in the Kings Garden, some Fereol wood where the grain is very fine, and the
W
background is a deep yellow color, with narrow stripes of the color brown leaning toward violet. It is
perhaps a nuance in this species; as to the rest, this wood is beautiful and works very well.
26. Fert [Caesalpina ferrea, Ironwood] is a tree that grows in the American Islands, the size of a man
at the trunk. Its wood is extremely hard, of a fauve color, brown leaning toward the black, especially
at the heart of the wood, which is extraordinarily hard. However, the concentric layers are easily distinguished, although the grain can be, at least, as closed [tight] as that of black ebony. The color of this
wood is generally unimpressive, and one can hardly use it in marquetry except for representing terraces or other objects where its color is suitable.
27. Fuset. This is a bush that comes from Jamaica and the south of France. Its wood is of a beautiful
veined yellow, but it is not very solid.
28. Guaiacum or Holy Wood, is grown in America, in the Islands of San Domingo and Port of Peace.
Its wood is solid, compact and resinous, of a green color, striped following the concentric layers [bold
grain pattern with a dramatic difference between the grain rings], which is alternately green, pale or
yellow and deep black [that is] more intense as they approach the center. It cannot be split except parallel to the concentric layers, which sometimes separate by themselves. The ligneous fibers are so irregular that one cannot separate them except with a saw. When one works this wood it makes a strong
odor that is not, however, disagreeable. The logs of this wood, which are brought to France, weigh up
to 500 pounds, and their crosswise cut is not round like most other trees, but it is elongated in the form
of a pear.
There are still other types of Holy Wood, a bit different from this one, among others, one that is as
hard, but the color is yellow, a little like boxwood and sometimes veined in green or black. This latter
type of Holy Wood is more appropriate for cabinetry than the first one because its wood, although
very hard, is easier to work.
29. White Gum Tree. This tree is grown in Guadeloupe. Its wood is white, veined with grey, leaning sometimes toward black. This wood is sometimes hard and difficult to work, and sometimes soft
and porous, although of a fine and closed [tight] grain.
30. Indigo or Laurier Aromatique [Haematoxylum campechianum, Logwood], is commonly called
Campeachy Wood. This is the heart of a very large tree that grows in America, in the Bay of Campeche,
from where it takes its name in the Jamaican islands, on Saint Croix, in Martinique and Grenada.
This wood is so heavy it does not float on water. It is compact and of a relatively fine grain. Its grains
intermix one with the other, which makes it a bit difficult to work, and which does not prevent it,
however, from taking a polish very well. The color of this wood is of a brilliant red and even transparent, or better said, glossy in a deep yellow. This color changes with the weather, or when this wood
has been too long in water. Then it becomes brown and sometimes of a black-grey, which has fooled
many of those who have described it so, some having given it the color red, and others a brown or
even a violet. This wood has a bit of a strong odor without being disagreeable, and it used for black
and violet dye. Indigo wood or Campeachy wood is the same as aromatic laurel [Sweet Bay?] or the
Jamaican Pear tree, or a tree that has the seed of four spices [Allspice?].
The real Campeachy wood is known by its cut, which is made with cuts of the hatchet by the
Page 777
12
Spanish who made a great commerce of it. There is some Campeachy wood that is of a deep brown,
spotted with black at a distance just about equal, but it is very rare currently.
31. Jacaranda [Dalbergia nigra, Brazilia rosewood], a large tree that grows in the Far East Indies.
There are two types, one white and the other black, and both marbled and very hard, but only the
black is fragrant. There is in the Office of Natural History of the Kings Garden, some jacaranda
W
wood, of which the background is yellowish striped with brown, violet and a little bit like the ear of
wheat but with a much more fine and closed [tight] grain. This is perhaps a nuance of this species.
32. Yellow [Tulipfera liriodendan?]. The tree that produces this wood becomes very large. It grows
in the Antilles and on the Island of Tobago. We also call it Fustoc and Clairembourg. It is solid,
without being absolutely hard and heavy. It works well and polishes well, although without being
porous. Its color is of a beautiful deep yellow, which approaches that of gold. We also call this wood
Yellow Satin. It grows in North America as a large and tall tree, which is named the Tulip Tree, which
is thought as being the same yellow wood.
33. Lapire, a tall tree that grows in Cayenne, of which the heart is mixed red and yellow, and which
has a very nice odor.
34. Nutmeg [Myristica spp.]. This tree, which bears mace and nutmeg, grows in the East Indies
and is the height of a pear tree. Its wood is soft and its bark ashen. It is not used much but is good to
work, however.
35. Pheasant Eye or Pheasant Wood [According to the French word it could be Partridge Eye]. I
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
The dozen or so cousins of the family used by French cabinetmakers include two comparatively less-
er-known species (to modern woodworkers): Dalbergia frutescens (tulipwood), Dalbergia cearensis (kingwood), along with the more widely known pair of Dalbergia nigra (Brazilian rosewood) and Dalbergia latifolia (Indian rosewood). Even a casual browsing of the furniture of Jean-Franois Oeben, Bernard II van
Risamburgh, et al., reveals the use of these siblings as an almost-universal constant throughout the trade. The
vibrancy of the grain, sometimes overwhelming to the modern eye, made it a natural canvas for the exuberant
taste of the ancients.
In addition to their appearance, the dalbergias are renowned as hard, waxy woods perfect for taking the
mirror-like finish desired by the French court. Their density and toughness are confirmed by the performance
of my favorite mallet head for the past two decades, a sold block of tulipwood that remains completely intact
and virtually unscarred by heavy use in chopping countless mortises of both architectural and furniture scale.
I own one small log of tulipwood, and I use bits of it after only a long deliberation.
For generally lighter-toned pieces, tulipwood was seemingly a tailor-made material. Freshly cut, the wood
has a bright yellow early wood background and a scarlet late wood grain pattern (although these descrip-
Of Foreign Woods
13
do not know where this wood comes from. It is perhaps a type of ironwood that grows in China,
and which is so hard that one uses it to make anchors for vessels. Pheasant wood is very hard and
very heavy. Its color is more obscure than that of the iron wood, and although very dense, you can
perceive along its longitudinal fibers very fine pores filled with an ash or whitish gum, which only
appears in the end grain as almost imperceptible little white dots. Its usage is a bit the same as that
W
of the ironwood.
36. Olivewood [Olea europaea]. This tree is originally from Syria and grows in the Mideast. It is
hard, or better said, closed, resinous and in general a bit solid and twisted. Its color is yellowish and
striped with brown following the concentric layers, which makes it wavy or veined on its vertical
faces according to the way it was cut. However it may be, this wood is most beautiful used as end
grain. Its humps [burl] or grain swirls are also highly sought after for the variety of shapes that they
represent. This wood has the defect of unrolling, that is to say, the concentric layers detach from each
other, which means one can only use them in small parts.
37. Orangewood. The tree is of average size, originally from China, from where the Portuguese
brought it back as a seed. Its wood is rather compact, of a yellow color and white toward the center.
38. Platane or Plane Tree, comes from Asia and North America. Its wood is white, rather compact,
pliant [flexible] and of a closed [tight] texture. It resembles beech wood. It holds the middle between
beech and maple, of which species it could be grouped. It can be used for cabinetry in its natural
color, but is still tinted in diverse colors. (See following entry with the maple description.)
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
tors are not literal in tropical hardwoods as the growing pattern is not seasonal as with temperate woods).
With time, such brightness mellows, but it remains striking nevertheless.
Kingwood, equally unknown to modern artisans, is the negative iteration of tulipwood. Rather than scar-
let-on-yellow, kingwood is a much more subdued presentation of something akin to my favorite ultra-dark
chocolate on a plum sauce background. Like tulipwood, it is able to withstand nearly brutal abuse, and it
requires edge tools of razor sharpness to be worked well. My few boards of kingwood are consumed with
parsimonious attentiveness.
The two rosewoods, Brazilian and Indian, possess the same nearly black dark grain pattern, although
presented against a dark orange background in the former case and against a dark brown, sometimes reddish-brown in the latter. Both of these rosewoods are somewhat less physically robust than their siblings listed above (they splinter and split more easily), but can provide large expanses of strongly linear grain lines,
especially in the case of Indian rosewood.
Taking dalbergia out of the equation makes contemplation of what the world of historic veneered furni-
Page 778
14
39. Puant [Stinking Tree]. [It] is the size of the oak, which grows at the Cape of Good Hope. It is
of a beautiful nuanced grain, and regardless of the bad smell, we make use of it because it loses its bad
odor over time. This wood is little used in France.
40. Rhodes. The ancients were not in agreement on the nature of this wood, as I said in speaking
about Asphalatae. They also named this wood Candia Wood, apparently because it grew on this
W
island. We presently know two types of wood from Rhodes or of rose: one that comes from Jamaica
and the other from the Antilles Islands. The first type comes in the size of the thigh of a man. Its bark
is rough and brown and is studded with spines. Its wood is solid, white and has much pith, and is of a
very penetrating odor.
41. The second type of wood from Rhodes is the most common, and is known under the name
of Rosewood or Marbled Wood [Dalbergia frutescens, probably tulipwood]. This tree grows high and
straight. Its wood is close-grained [tight], without being hard. Its color is that of a dead leaf, or better
Page 779
said, it is mixed yellow and rose [pink, reddish], and a red-violet, positioned in about equal layers,
which follow the concentric layers of the tree. When splitting this wood in different directions, it
is striped, veined or even marbled as much as there is sinuousness along its grain, which gives it the
name Marbled Wood. It is one of the most beautiful woods that comes to us from the Indies, and is one
of which we make the most use. This wood is subject to being hollow in the center, and loses over
time the liveliness of its red color, which becomes pale due to the evaporation of the resinous substance, which is more abundant in the red veins than in the yellow. This is contrary to violet wood,
which darkens in air, because the resinous substance is more abundant in the lighter layers than in
the brown ones. This observation is almost general for all refined woods, where the color is bright
and the pores are open. Rosewood smells when one works it a light odor of a rose, and its knots are
hardly detrimental to the excellence of the work. On the contrary, when it is used with taste, they
augment its beauty.
There is in the Office of Natural History in the Kings Garden, some rosewood of which the color
is grey, veined with brown. I do not know if it is the same as that which I just described, from which
it differs only by its color.
42. Redwood or Bloodwood [Brosimum paraense]. This is a tree that grows in [Central] America,
close to the gulf of Nicaragua. Its wood is hard and of a very beautiful red, and serves as a dye.
43. Santal or Sandalwood [Santalum spp.]. There are three types of sandalwood: named the citrous,
the white and the red. The yellow, or citrous, grows in China in the Siam Kingdom [Thailand], and
in the islands of Solor and Timor, to the height of a walnut tree. Its wood is average hard and heavy.
Its fibers are straight, of a pale reddish color or yellowish, leaning toward citron. Its taste is aromatic,
a little bitter, which fills the mouth without being disagreeable. It produces an odor that approaches
that of musk and of rose.
44. White Sandalwood grows in the same countries, and is similar to that of citrous Sandalwood,
from which it differs only in color. We believe it is the same tree of which the heart is yellow and the
ends are white. The Sandal Tree is also named Sarcanda.
45. Red Sandalwood, also named Pantagna, is grown in the island of Tanasserin, and on the
coast of Coromandel. Its wood is compact and heavy. Its grain is sometimes straight, sometimes
wavy. This wood is brought separately from its bark and from its ligneous exterior (that is to say, its
exterior parts) and so it has an exterior of a reddish brown, almost black, and the interior of a deep
red, mixed with a bit of yellow. This wood has no odor, and it is said that the wood from Caliatour is
the same thing as this.
Of Foreign Woods
15
46. Satinwood [Chloroxylon swietenia]. This tree grows in the Antilles Islands. Its wood is solid,
hard, resinous and very porous. Its ligneous fibers are filled with an ash or shiny gum, which appears
as little dots at the end grain. Satinwood is more nuanced or wavy than striped. It comes in many
colors, from deep red veined with yellow, to deep red mixed with olive green. Whatever color it is,
Page 780
this wood always has an iridescent appearance, which is its principal beauty. Because this wood is
W
not regularly veined on its end grain, one can make it appear striped by quartersawn cutting. One
can also make it appear wavy or even blazing by splitting it in concentric layers. This wood polishes
well and is highly used at present.
47. Violet wood [Dalbergia cereansis, probably Kingwood]. This wood comes from the East Indies, from where the Dutch brought it. It is similar to Rosewood, at least in form and position of
its concentric layers. Its dominant color is violet brown, almost black, striped with white veins,
which change [oxidize] with time for the reasons that I said when speaking of Rosewood, which is
even more compact than the violet. The latter has the defect of being hollow at the core, to having
vicious knots, and being difficult to work, because the grain of the concentric layers are in different
directions. This wood is of a nice violet odor, which, added to its color, gives it the name Violet.
I suspect that one could give this name to a number of Jacarandas, which are really more similar,
than the Palixander wood, a type of violet wood, named by different authors Jacaranda.
48. There is another type of Violet Wood, named Palissander or Palixander [Dalbergia baronii,
Rosewood], that comes from the West Indies in large logs of 7 to 8 feet long, by 12 to 15 thumbs in
diameter. This wood is less beautiful than the violet, more porous, with a grain almost as large
as ordinary oak. Its color is of a deep grey, more brown than violet, scattered with veins of a reddish-white, all positioned along the concentric layers. The one with the most of this type of vein is
the most desired. But in general, the color of this wood is unimpressive and disagreeable. The wood
of Palissander gives off a very nice odor, stronger than that of Violet-wood and the more it is heated,
the more it smells good. This is only natural, as this wood is very resinous. The aromatic gum that
fills the pores evaporates even more easily by the action of rubbing, as these pores are more opened
[along the] grain line than the end [grain].
The wood of Palissander is also named Wood of San Lucia. I do not know if this is because it
grows on the island of San Lucia or San Alouzie, one of the Antilles Islands, or because the odor of
Palissander is a bit like that of the wood of San Lucia, properly called: tree that grows in Lorraine
and is of the genre Ceresier [the cherry family].
There you have it, at least a bit closer to the detail of all the Woods from the Indies [Tropics]
relative to cabinetry. I have endeavored to describe the best that it has been possible for me, so that
one can, with knowledge [born of interest], make choices from one to the other as one judges appropriate. It is still good to be mindful of the fact that it has not been possible for me to see all these
woods in various degrees of aging, nor to even see large pieces of many of them. The descriptions
may not be as exact as one would like, which is what I would have wished myself. The same subjects sometimes vary in their different parts, which has surely resulted in many names to woods
that could come from the same type of tree, as I have indicated above. One must also pay attention
that the woods change in color, not only upon aging but when working them, and that the polish
browns them a lot, at least for the most part. That is why, before putting to work the woods which
one wishes to match in color, whether between them, or to match a given color, one must work and
burnish the samples, in order not to be mistaken concerning their real colors, at least those [colors]
that are left.
Page 781
16
In the description that I am going to make of Woods of France, I will only treat those that are appropriate for cabinetry, whether for veneer or inlay, in order not to repeat what I have already said on the
W
subject of other woods at the beginning of the first part of my work, page 22 and following on [which
is not included in this volume].
Color
Quality
49. Alizie
White
Hard
50. Aulne
Reddish
Soft
51. Boxwood
Yellow
Very hard
52. Cherry
Rosy, veiny
Solid
53. Hornbeam
White
Very hard
54. Serviceberry
Reddish
Very hard
greenish
Solid
56. Barberry
Yellow
Solid
57. Maple
Solid
Yellowish &
Greenish, striped
Aroma
Hard
59. Ash
Solid
60. Prickwood
Pale yellow
Hard
61. Holly
white
Hard
62. Yew
Reddish
Hard
Striped, reddish
Firm
64. Mulberry
soft
65. Walnut
Veiny black
solid
66. Pear
Reddish
very solid
67. Apple
white
Solid
68. Plum
Redwhite and
Rusty, veined
Solid
Rusty grey
Solid
Whitish
Hard
71. Elder
Yellow
Hard
Of French Woods
17
49. Alizie sorbus ariar [Rubia spp., madder root tree]. This tree is average in size, its wood is hard
and full, its grain fine and closed. Its color is white, sometimes rust, and almost always blackish to-
Page 783
ward the heartwood. This wood is very good at being stained and produces no nuances [can be evenly
stained], although the concentric layers are apparent at the end grain. Azerolier is similar to Alizier, and one can use it as well.
50. Aulne/Aune [Alnus glutinosa, alder]. Aquatic tree, very tall; its wood is light and a little soft.
The grain is fine; it is the color of rust more than red. It takes dyeing well, but it does not polish very
well because it is too soft.
51. Boxwood [Buxus spp.], bush of average size. There are two sorts of boxwood, namely, that of
France and that of Spain. The boxwood of France is heavy, hard, compact and of a fine and closed
grain. Its color is yellow and it polishes very well. The Boxwood of Spain is a bit similar to that of
France, except that it is a bit less hard and its color less deep. Whats more, the concentric layers of
the Boxwood of Spain are obvious enough to form lines or waves along the grain line. The nodes or
bumps and the roots of the Boxwood of Provence are highly sought and work very well because they
are not subject to cracking [or splitting].
52. Cherry [Prunus cerasus], fruit tree, originally from Asia, from where it was brought to Europe
by Lucullus, on the return from the war against Mithridate [an enemy of the Romans, from the Bosphorous regions]. The wood of the cherry is medium-hard, solid enough, although the grain is a little
coarse, and its concentric layers are very obvious. Its color is reddish-grey, deeper toward the heartwood than at the outer region [sapwood]. This wood works well, and takes a polish easily. One uses
it commonly for cabinetry, for making little works of solid wood, and sometimes only for making
veneer like the wood from the Indies [Tropics].
53. Hornbeam [Carpinus betulus], yoke-elm is a forest tree of average size. Its wood is highly pliant
and difficult to work. It splits badly because its grain is interwoven. The color of this wood is white.
We do not make great use of this wood because it is subject to rot, it works badly and it is subject to
great instability. That is why we prefer to work Holly [Ilex aquifolium].
54. Service Berry or Cormier [Sorbus domestica], fruit tree of average size. Its wood is, after the
boxwood, the hardest and most solid of the Woods of France. There are two types of Service Berry
by color: one that is a white/rust, and the other that is reddish and is the most highly esteemed. In
general, the wood of Service Berry is very solid, its grain is fine, but short, which makes it split [or
crack] easily. It is worked rather easily, and it takes polish well. Although this wood is hard, its concentric layers are distinguished easily on the end grain, and even along quartersawn grain. The Service Berry has the defect of being unstable, and it is subject to being wormy or damaged by insects
when it becomes old. The color improves upon aging, and it becomes almost brown when polished
with oil.
55. Bean Trefoil or Alpine Ebony, tree of average size. The wood of this tree has lots of sapwood,
under which one finds the best wood, which is solid and very pliant. See what I have said in speaking
of green Ebony, page 775.
56. Barberry [Berberis spp.], bush of which the wood is solid, yellow in color and which is easily
worked. Its small size means one can only use it for small pieces and for veneer, or for inlay. It is also
useful for dyes.
57. Maple [Acer spp.], tree of average size. Its wood is solid enough, however light, deep-toned
and bright, approaching the quality of Beechwood, and has the merit of not being unstable. The
Page 784
18
color of Maple is white with a bit of red. This wood is sometimes veined and wavy, that is to say, its
fibers are serpentine instead of being straight, which produces waves of a very great beauty. They are
not obvious except when quartersawn, and more or less relative to the size of the fibers. The burrs
and nodes of this wood [birds-eye grain] are also very sought after for the figures they represent.
They also polish very well. See a bit later, the article on the difference in woods according to their
W
different tints, plate 277, Figs. 15 and 16 [not shown in this volume]. There are many different species
of Maple, of this the principal ones are the Platane Maple [Platanus occidentalus], which comes from
America, which I spoke of above. The Sycamore Maple is the least esteemed of all, because it is less
solid, white in color and rarely wavy, which puts it in the class of White Woods.
There are still a great number of Maples that have little difference among them, but the knowl-
edge of which is of little use to woodworkers, given that these differences consist, for the most part,
only in their size, or in the shape of their flowers, or of their leaves. Their description cannot nor
should not enter into this work, as I said at the beginning of the description of the wood from the
Indies. See the note, page 767.
58. False Acacia. This tree, originally from America, is twisted and thorny. Its color is brilliant
and striped, alternating green-yellow and brown, leaning toward the green in following the direction
of the concentric layers. This wood is hard; however, it does not polish very well because its fibers
are a bit interwoven, such that one always finds wood on the end grain that produces a type of nap
or hair (so say the Workers), difficult to eliminate perfectly. It also has the defect of rotting when
moist.
59. Ash [Fraxinus spp.], a tall tree of which the wood is solid and pliant enough. Its color is white
and striped with yellow at the separation of the concentric layers. It is subject to insects, and we
make little use of it in cabinetry. Because of its color and its small stripes one could use it advantageously in small pieces.
Page 785
60. Prickwood, a bush of which the wood is hard and full, and the color is pale yellow. One can
use it in cabinetry, especially when it is well dried and on grain.
61. Holly [Ilex spp.], large bush. Its wood is hard, pliant and of a fine and close grain, white in
color, sometimes brown toward the center. This wood is greatly used in cabinetry because it takes
dye well, and it has a fine and pliant grain. One can cut it into small parts as one judges appropriate.
It also serves to make string inlay, which I will talk about in the proper place.
62. Yew [Taxus baccata]. This tree is of average size, although it has been seen at 20 feet in diameter. Its wood is ordinarily twisted, very hard and pliant, of a beautiful red color mixed with yellow
and brown. Although this wood is very solid, its concentric layers are easily distinguished and are
ordinarily marked by a little string deeper than the color of the wood, which becomes deeper just up
to the closest layer to the center of the tree, which produces beautiful nuances along the grain. The
wood of yew is, of all the woods of France, that which by its color and hardness most approaches the
woods of the Indies [Tropics]. We make little use of it in cabinetry. I do not know for what reason.
There is some yew that is soft and extremely porous, but it cannot be of any use for the cabinetry
that I am addressing.
63. Wild Cherry Tree [Prunus avium]. This type of wild cherry is a large fruit tree, of which the
wood is closed and firm, and of a rather fine grain. Its color is reddish, striped with very fine yellow
veins. This wood, when it is good, polishes well, and we make the same use of it as the Cherry. See
Cherry, page 783.
Of French Woods
19
64. Mulberry Tree [Morus spp.], tree of two species, the black and the white. The black mulberry
wood from Europe is white on the edges and yellow in the heartwood, which blackens upon aging.
The black Mulberry of Europe is more solid than that of white mulberry from Asia. It is very durable,
it resists water, and whats more, it is not subject to vermin. They say it chases away bugs. If this were
true, it would be very good to use it to make bedsteads. This wood is used little in cabinetry, however,
W
and other than the defect of its color, it could be very well used.
65. Walnut [Juglans spp.], fruit tree. I made the description of this wood in the first part of my
work, which is why I am not going to extend myself much on the qualities of this wood. Although little
used presently, it is one of the most beautiful that one could use in cabinetry, as much for the grain as
for its warts, which are commonly named burl, or Roots of Grenoble. Other than the walnut of France,
of which the most beautiful comes from Dauphine and from Auvergne, there is still the Walnut of Virginia, which is cultivated now in Bourgogne, which is blackish and veined, but has the defect of being
porous and brittle. The Walnut of Virginia is a little tree of which the wood is pliant, compact, very
hard and of the color white. The walnut of Louisiana, named also Pacanier, is similar to ash, at least
in appearance. Of all the different walnuts the black is that which is the most sought after for cabinetry, because of its beautiful nuances. It is too bad that the walnuts are subject to rot and to vermin.
65. Reed. This is an aquatic bush, of which the wood is very slender, pliant, soft and white, and
of which the grain lines are extremely fine. This wood serves cabinetmakers for making string work
when holly is not available. [Numerical typographical error in the original volume.]
66. Pear [Pyrus, spp.], fruit tree, of which the wood is solid, compact and medium-heavy, of a fine
and close grain, and a reddish color. This wood polishes perfectly well, and takes well the black dye,
such that it is used as a substitute for ebony. It is too bad that this wood has the defect of being unstable
because it is one of the most beautiful woods that one can use for small pieces. Pear trees sometimes
grow very large we have seen them in England of 6 feet in diameter.
67. Apple Tree [Malus spp.], fruit tree of which the wood is less hard than that of the pear, pliant and
very soft; its color is white. In general, this wood closely approaches that of the Service Berry tree.
68. Plum Tree [Prunus spp.], little fruit tree, of which the wood, although solid, is soft and light
(and sometimes hard, according to the different subjects) and the color is a grey veined with red. This
wood is very nice to see, as much for the concentric layers as for the radial, which are small and shiny.
This wood has short little tissues, which does not prevent it from taking polish well. It is too bad that
its red color fades easily, but one can remedy this by varnishing it.
69. San Lucia or Padus, a species of wild Cherry, which grows in Lorraine and in the environs of
Verona, Italy. This wood is compact, of a reddish-grey and pleasant to look at. It has a very nice odor
that increases as it ages. One must choose it when dry, without nodes or sapwood. The wood of San
Lucia is also named Mahaleb, which is an error, because the wood of the Mahaleb tree, or Cherry of
woodlands, however similar to the wood of San Lucia for its color, is much harder and has no odor.
70. Sauvegeons [Wild Stock]. We name some Pears and Apple trees with this name when they
are not grafted. Their wood is not as beautiful as those of the pears and apples properly named; it is
normally harder and more difficult to work. See above, the article on Pear & Apple.
71. Elder Tree [Sambucus spp.], large bush, of which the wood of the trunk and large branches is
solid, hard and very pliant, yellow in color, and similar to that of Boxwood. This wood spoils with difficulty, and is very proper to use in cabinetry, because of its excellence of performance, and its beautiful
color.
Page 786
20
Page 787
After the details that I just made for the Woods of the Indies [Tropics] as well as the Woods of
France, one can judge how many of the former are preferable to the latter. This is true not only because of their number, but also for their beautiful colors, the fineness of their grain, the nice odor that
most of them render, and their almost stable qualities, which are not encountered except in a very
small number of our Woods from France. We impart color by means of dyes, which I will speak of
W
later, but one could never have all the good qualities of the Woods of the Indies [Tropics]. Whats
more, the color of stained woods diminishes with time. This could not be otherwise, as long as the
means of reducing the salt in the wood has not been found, the evaporation of which causes the stain
to evaporate as well. It is too bad that in the colors of the Woods of the Indies [Tropics], we are missing
two essential colors: namely blue and green. These colors are indispensable for truly representing a
landscape of many other objects in a composition requiring these two colors, and where one can only
procure it by means of a dye. These two colors being absent, the Woods from the Indies [Tropics]
furnish almost all the others as lively or subtle as one could desire. It is why in the following paragraph, before making the description of the different nuances of the woods, I am going to provide a
table in which I have put in order and in one list all the woods, whether from the Indies [Tropics] or
from France including their different colors. I arrange the latter among themselves so that with the
glance of the eye one could choose that species of wood that one judges appropriate. This will form
the palette of the cabinetmaker or painter in wood. I have paid attention to mark each type of wood
by the number that it has in the alphabetical table, and in the article of its description, to which one
could refer, so as not to mix together the hard and the soft woods, at least as much as it is possible.
Colors in general, and the Woods from the Indies and from France, with
Regards to their different colors and their nuances
The art of marquetry is also named painting in wood, as I said at the beginning of this part of my
work. That is because of the designs that the cabinetmakers execute by the mixing of woods, whether
tinted or natural, which requires of workers a great knowledge of designs of all types, but yet a perfect knowledge of colors with which they can execute them.
I am not going to speak here of colors with regards to their properties and their relation, neither
following the definitions given by the scientists. I am going to limit myself to speak relative to the objects that I work on.
There are five sorts of simple or primitive colors, which being mixed together, produce all the
others. The simple colors are blue, yellow, red, tawny and black. Wood is found of the four latter
colors, more or less deep. There is only the first, that is to say the blue, which is missing.
Page 788
The composed colors consist of purple, crimson, amaranth and violet, which come from the mix
of blue and red. Green, of all sorts, comes from the mixture of blue and yellow. The mix of red and
yellow gives the yellow color of gold, the color of marigold, orange, pomegranate, etc. Red mixed with
tawny gives the color of cinnamon, chestnut and musk. The yellow mixed with tawny gives all the
colors of dead leaves. Of all the composed colors, there is only the green and the orange, and the colors
close by, which one cannot find in wood, whether of the Indies [Tropics] or of France, and for which
one must have recourse to staining, together with grey, which dyes I will give the composition [of] afterward, and the manner of making use of them.
21
After this general knowledge of touching on colors, cabinetmakers should apply themselves to
know perfectly the woods that make these different colors, in order not to be embarrassed when representing with precision all sorts of designs, whether flowers, landscapes, etc.
List of Woods of the Indies and from France, according to their color
Red. The woods of this color are: Cashew, #1; Amourette, #5; Brasilwood, #8; Snakewood, #13;
Coral, #17; Red Ebony, #21; Campeachy, #30; Bloodwood, #42; Red Sandalwood, #45.
Woods of a reddish color, deeply nuanced: Mahogany, #2; Aloes or Eagle Wood, #3; Cayenne
Wood, #10; Cedar, #11; Copaiba, #16; Aulne, #50; Service Berry, #54; Yew, #62; Mulberry, #63;
Pear, #66; Plum, #68; Wild Cherry #69.
Yellow. The woods of this color are: Citron, #14; Fuset, #27; Yellow Wood, #32; Cinnamon
Wood, #9; Yellow Citrin, #43; Boxwood, #51; Barberry, #56; Prickwood, #60; Elder, #71; Ear of
Wheat, #24; Lapire, #33; Rose, # 41; and Satinwood, #46 [tulip poplar].
Woods of a yellowish color deeply nuanced are: Cypress, #18; Olivewood, #36; Orange, #37;
Page 789
The woods white veined or wavy are: Star Anise, #6; White Cedar, #12; Citron, #15; Portu-
guese Ebony, #20; Fereol, #25; White Gum, #29; Jacaranda, #31; Cherry, #52; Maple, #57; Ash,
#59; Walnut, #65; and Wild Stock, #70.
All the woods named above are not exactly the color red, yellow, etc., but as they are nuanced,
and each of them more or less deep in one spot as in another, it is very easy to match the nuances of
possible colors. Except for blue and a true green, colors are found in all sorts of wood, as much the
French as those from the Indies [Tropics], but more so in the latter than in the others.
If a perfect knowledge of the different colors of wood is essential to a cabinetmaker, he must also
distinguish these same woods by means of their nuances, or better said, by the different shapes that
Plate 277
the tints of the fibers represent, in order not to use them without choice nor knowledge of their char-
Different Sorts of
acter.
Woods, with regards to the conformation of tints of their fibers, can be considered as making four
distinct species, one from the other. They are: those of which the concentric layers are alternately
tinted in diverse colors but of a large and irregular manner, as you can see in Figs. 1, 2 and 3. The first
one represents a piece of wood of which the concentric layers are tinted at unequal distances, which
produces similar stripes on the grain line, Fig. 2, split according to the direction of the stripes of the
Positioning according
to Their Hue
22
tree. If on the contrary, one splits them parallel to the concentric layers, like in Fig. 3, this wood is
only a single color more or less dark, according to which the split is made in a vein more or less light,
which makes these sorts of wood not normally used except on the quarter-round cut, as in Fig. 2, or
cut diagonally, as indicated with line AB, same figure.
The second type of wood, with regard to their grain patterns, are those of which the concentric
W
layers, although distinguished by color at the end grain, like in Fig. 4, produce no stripes along the
grain, but simply singed veins or spots, like those in Figs. 5 and 6. These sorts of woods are very nice
when they are well chosen and used with discernment, by reason of the size they will occupy, and a
Plate 277
comparison being made with that of their nuances or their spots, which are always more abundant on
the radial cut than on the concentric layer.
Page 790
The third type of wood is those of which the end is veined irregularly in all ways, like Fig. 7. These
species of woods are most likely being used on end grain or diagonally, as I observed in Figs. 8 and 9.
As to the grain line, it is hardly an effect except on the quartersawn, where the colors must be vivid,
which is quite rare in these sorts of wood.
The fourth type of wood is that where the concentric layers are regular and alternating in various
colors, like that of Fig. 10. These sorts of wood are those where one uses with the best advantage,
because not only are they beautiful on end grain, but also along the grain line, whether they are split
parallel to the concentric layers, as in Fig. 11, or according to the direction of the rays, like in Fig. 12. In
the first case, they present a wavy surface, where the spots or singes [area of lightness or of disorder,
representing a flame] are more or less large according to the split being made closer to the circumference of the tree. In the second case, that is to say, when the split is made on grain, as in Fig. 12, the
wood presents stripes almost regular, which are more or less perfect according to the split being directly made when in the center of the tree.
These four types of differences, which concern the tints of the wood, are those that are the most
striking, because there is an infinite number that are but variations between those which they resemble in some areas.
There are still some woods where the crosswise cut, or end grain, instead of being striped fol-
lowing the concentric layers, represent an infinity of points of different colors, placed a bit like rush
[reed], and always following the concentric layers of the tree. These sorts of wood make a rather good
effect on end grain, as one can see in Fig. 14. As to their grain, it is always striped, in whatever direction one is splitting it, except the stripes that most approach the circumference of the tree, and which
parallel it, are a bit larger than the others. See Fig. 13.
What I just said about the different tints of wood can hardly be applied except to the woods from
the Indies [Tropics], because a very small number of Woods of France are striped or veined, as one
can see above. Among the latter, maple, although without color, is one of the most beautiful woods
that one could use for cabinetry [veneering], especially when one tints in a grey or some other color.
This wood differs from all the others, whether from the Indies [Tropics] or from France, by the conformation of its ligneous fibers, which, instead of being straight, go in a snakelike pattern with each
concentric layer, as I explained above, page 784. It has the advantage of making a very beautiful effect,
in whatever manner it is split. If it is cut on the radial plane, as in Fig. 15, it represents a wavy surface
that seems to be in relief. If, on the contrary, it is split parallel to the concentric layers, its surface becomes moir [watermark], like the fabric. See Fig. 16.
The woods, such as I just described, are for the most part split straight along their length. When
Plate 277 . Different Sorts of Woods and Their Positioning according to Their Hue
24
Page 791
one wishes to augment their nuances, and consequently their beauty, one splits them a bit at a slant,
which makes them appropriate to use for in-fill of panels, whether of a heart, a rose, etc., as I will
explain when speaking of the way to position veneer woods.
These woods are used also on end grain. In this case, they must be of a small diameter, so that
It is good to observe that in making a description of the colors and character of woods, whether
from the Indies [Tropics] or from France, I have described them as they are put to work. For the
most part these woods change in color when worked, and especially when polishing them, which
Plate 277
saturates them, without counting that these same woods also change upon aging. That is why it is
necessary that cabinetmakers join theory with knowledge of the woods, with much experience
touching on the effect of these same woods, in order not to be surprised by the changes, which
necessarily take place.
In the absence of experience, which one cannot acquire except with much time, young cabinet-
makers should have samples of all the different woods. They should have three or four samples of
each type, so they can burn them [so as to learn about their characteristics], plane them and polish
them in different ways. By this manner you can acquire much knowledge in a bit of time. This is
the method that served me, and which, added to other knowledge that I have on this subject, was a
great help to me in the part I am addressing.
In general, whatever species be the wood that one is using for cabinetry, one must take care to
choose them very dry, without nodes or chinks (with the exception of the nodes of certain woods,
which one may use advantageously, as I said in some places). One must, however, avoid having the
wood too dry, because it tends to be susceptible to insects or vermin infestation, it works with difficulty and does not take glue well. Whats more, when the wood is too dry, it loses part of its strength
and its beauty. As to the choice of woods, which one puts together, I will speak of this when it is a
matter of veneering woods. All I am going to say presently is that one must avoid putting together
woods of uneven density, because it is not easy to finish work where woods both hard and soft are
used together, which I believe is very easy to understand.
The tinting [dyeing or staining] of woods is of great importance for cabinetmakers, because it is with
its help that one can give to woods the different colors, which are necessary for representing all sorts
of objects, such as fruits, flowers, animals, etc. However, cabinetmakers always make a great secret
of the composition of their dyes in order to preserve exclusivity, and not to increase the number of
workers in their trade. From that circumstance comes the fact that most of the compositions that the
ancient cabinetmakers used have not been passed on to us, or are presently badly imitated. Those
being used presently are defective, or even if they are good, cannot be perfected given that those who
possess them hide the process. They keep this information secret not only from their colleagues but
even from those for whom the theory could be useful in perfecting the composition of their dyes.
This would be much more advantageous than the enjoyment of maintaining a secret, which is not a
big thing, but which, even when it is perfectly well known to us, leaves us still to regret the loss of the
25
method of Jean de Veronne, who tinted woods with boiling dyes and oils that penetrated them. This
would be a very helpful thing to know, the research of which would be a worthy undertaking for some
of our scholars. It is highly wished that one could find the means to use the chemicals having a good
tint in the dyeing of woods, because their colors would be more durable. Sadly, the colored parts of
most of these chemicals are too thick to penetrate the interior of woods, which is absolutely necesW
sary, so that when working with tinted woods they are all found to be of uniform color throughout
their entire thickness and the surface.
That is why in the description of the woods, without the means to which I would like to know the
procedure to accomplish the perfect tint, I will explain only ordinary procedures to cabinetmakers,
to which I will add some of my own experiences, which is still a long ways from attaining the perfection of which this part can be capable. ***
The five primitive colors are, as I said above, blue, yellow, red, taupe [brown] and black. Each of
these colors is given by different chemicals, which, when mixed together, give the second or composite colors.
The blue appropriate for tinting woods is made with indigo, diluted in oil of vitriol [also known
large to penetrate into the interior of the wood, and that they
colors that you have given me for samples, and I cast my eyes
on the products of the dyes that you let me have, seeing that
all these shades were well dyed into their interiors, I judged
that all these colors being made were with chemicals of a false
tint. One cannot desire anything else, if not to make the same
dient, leave but only some very solid colors on the vegetable
have not taken, except for the rather dull colors, which are
nounce this method. I did have success with what you used
solid, for cottons and twines. For that I used the same prepa-
These woods thus prepared have rather well taken the dye,
but the color is not always accepted except on the surface, the
Page 793
26
Red is made from the boiling of wool, or even a concoction of Brasilwood mixed with alum.
Black is made with the wood of the Indies, the gall nuts and iron sulfate.
Before entering into the detail of the composition of different stains, I am going to give a general
idea of the chemicals of which they are composed, so that the cabinetmakers may be less subject to
W
Indigo is a type of ash of a deep blue, provided by the leaves of a plant that grows in the Americas
and Indostan, and which they sell in little pieces. For it to be good, it must be medium-hard, so that it
floats on water, so it is inflammable and of a beautiful blue or deep violet color. Its interior should be
strewn with little silver-colored spangles, and appear reddish when rubbed with a fingernail. Indigo
is preferred over all other chemicals for staining woods because it is a powder of extremely fine and
granular pieces, which are easily introduced into the pores of the woods.
Oil of vitriol or sulfuric acid is the final spirit that one gets from vitriol. This acidic liquor should
be very concentrated and be absolutely free of all aqueous parts to be of a beautiful blue color, as I will
speak more of later.
Barberry is a little bush of which the fruits, and the bark of the roots are stained in yellow. That
from Candie [island off the coast of Crete] has a very yellow wood, and passes for the best.
Woad [this cannot be woad that produces a blue dye] is a rather common plant in France. One
boils it in water to extract a yellow liquid, which mixed with a bit of alum, tints very well. Dyers prefer
that one, which is the most spare [meaning thinnest] and of a rosy color.
One also dyes in yellow with the yellow wood of which I spoke above page 777.
Saffron is a plant that grows in France, especially in Gatinois [western part of France]. It is the
pistil of the saffron flower, which gives these little reddish filaments, or better said, orange, which
they sell under the name of saffron, which gives a dye of a golden yellow. For saffron to be good, it
should be fresh, of a pungent odor, of a brilliant color and when touched it should seem oily and
should stick to the hands.
Alum is a fossil salt and mineral, which is used much in dyeing, whether to set up the materials to
be stained or whether for fixing the colors [as a mordant], which it retains all the particles by its astringent quality. The best is that of Rome, which is white in color, and is transparent, a bit like crystal.
Liquor decanted from boiled wool is sold by the wool merchants. In boiling this wool, one gets a
decoction of the color rose, which is more or less deep, according to how much water is used to scour
the wool, proportional with its quantity.
I spoke up above of Brasilwood, page 771. I will content myself to say here that the decoction of this
although very sought after, has not had success rather marked
There, Monsieur, you have all that I could do. You see
icals for a good tint, at least we have not found the secret to
ing all these tinctures cold, and leaving the woods there until
27
wood gives off a clear red color, tending toward the orange, and that one deepens its color by adding
a bit of alum. Brasilwood from Fernambouc is the best, and they sell it all chopped up at the spice
merchants, who sell it by the pound.
The husk of walnuts is nothing more than the first wrapping of these nuts, which one takes off
before they are perfectly mature, and which one boils in water to extract a brownish or taupey tint.
Page 795
Indian Wood, of which I spoke on page 777, gives off a concoction of a deep red, which one stains
Nut gall is a type of excretion that is found on the tender shoots of a type of oak named Rouvre.
The most highly esteemed nut gall comes from the Levant [the name given to the countries on the
eastern coast of the Mediterranean]. The best ones are those that are the heaviest, and where the
surface is thorny. There are both green and black ones, both of which work equally to stain in black.
Ferrous sulfate is a type of vitriol that is found in copper mines. It is the most powerful of the
acids, it corrodes iron and copper, and it etches the soft parts with an infinite number of small holes,
into which the dye is introduced. Ferrous sulfate is also named Roman vitriol or English vitriol, according to whether it comes from one or the other countries. We make some in France that is, they
say, as good as the others. The color of ferrous sulfate [known also as green vitriol] is of a light green:
it should be neat and shiny.
Verdigris also works well as a wood dye. It is the green rust scraped from copper sheets. For it to
be good, it should be dry, pure, of a deep green and filled with white spots.
There you have a bit of a description of the ingredients commonly used for staining/dyeing woods.
Cabinetmakers ordinarily use a stoneware butter pot for putting the wood into the dye, which is
very convenient because the shape of this vase enables one to put in rather large pieces, without the
need of having a very great quantity of dye.
It is very easy to know when the interior of the wood is penetrated, given that you only have to cut
a small piece of the wood about 2 to 3 lines from its end. When the pieces that you want to dye cannot
be cut like this, you put with them another piece of the same quality, with which you test the degree
of penetration of the other pieces.
Page 796
28
these chemicals is 2 liters [in this case the French word refers to litron, which is about 79 percent of
a modern liter, so 2 modern liters is a much larger quantity] of barberry, 6 sols [a French penny] of
yellow earth, and 4 sols of saffron.
A concoction of woad gives a very beautiful yellow of a good tint, and you soak the wood as nor-
mal. When this concoction is added to a bit of verdigris, you have a sulfurous yellow color. Saffron
infused in grain alcohol gives a very beautiful golden yellow.
The Way to dye in R ed
Red is normally made with brasilwood, which one boils with 6 sols of alum for each pound of wood.
This red is a false tint because it is more orange than red. You can substitute the boiling-liqueur from
wool, which gives a very beautiful red, leaning toward rose, which one makes deeper by passing the
pieces that you have stained into the liqueur of the mixture of Brasilwood mixed with alum. This
makes a very beautiful red, more or less deep, depending on whether you leave the pieces of wood
more or less a long time in the dye bath of Brasilwood.
Dyeing with decanting liqueur is done very easily. One only needs to boil some wool dyed to this
effect, just until it makes a beautiful red concoction. Avoid boiling too much, because the wool will
take back the color that it discharged at first.
The proportion of the liquor of wool to be decanted is 1 pound to 4 pints of water for the first
decanting, to which one can add a second, even a third, until the wool renders no more color. The
Page 797
concoction of Brasilwood without alum gives a yellowish red, which is sometimes attractive, and is
named Capucine.
The concoction of Indian Wood is very red, but it makes a blackish stain, which makes a very
beautiful violet when mixed with alum from Rome, as I will speak of it later.
How to dye Taupe [Brown], Black and Grey
Taupe dye is made with a concoction of walnut husk, which can be more or less strong, as you judge
appropriate, always adding to it a bit of alum.
An attractive black is made by staining the wood first in a concoction of wood of India (or
Campeachy, which is the same thing). When this first application is dry, you dip the wood in a concoction of gall nut in which you have put some ferrous sulfate, or vitriol of Rome. Sometimes one
only makes a single dye of these various ingredients, of which the proportion should be 1 part nut
gall, 1 part vitriol and 6 parts of Campeachy, all boiled together, into which you dip the wood until it
is penetrated.
A grey tint is made with a concoction of nut gall, into which you dissolve some green vitriol
[ferrous sulfate] in smaller quantity than for the black stain. The more ferrous sulfate-cuprous there
is, the deeper grey it will be. The normal proportion is one part of ferrous sulfate for two parts of
nut gall.
29
One can make a very beautiful apple green in staining first the wood in ordinary blue, and then
dipping it in a concoction of woad, and that with more or less time according to whether one wants to
have a green more or less strong.
Violet is made with a concoction of Campeachy, to which one has mixed some alum from Rome.
One can have violet more or less deep by staining first the woods in rose and then in the blue, which
will give a clear violet.
If, on the contrary, one wishes to have a brown-red leaning toward violet, one stains the wood first
One can obtain composite dye of all nuances imaginable by tinting the wood in a primary color
Page 798
then in another one more or less dark, so that the stain that results from these two colors reflects more
or less of each other. This is very possible to do because one is the master to strengthen or weaken
the primary colors as one judges appropriate, whether by reason of what the form of the object requires, or even by reason of the different quality of wood, which takes the dye more or less well, or
strengthens or weakens the color. This has to be highly considered, and it requires much attention
and experience on the part of cabinetmakers.
In general, all the dyes of which I just spoke are applied in cold baths. It is not that many of them
cannot be used hot, but it is that because it takes a considerable amount of time for the same dye to
penetrate into the interior of wood, it is not possible to use them hot. Whats more, cold dyed wood
has much more vibrancy than when used with a hot bath.
There it is, a bit of the details of staining [dyeing] wood, at least those that most cabinetmakers
use, or which I myself have employed in the attempts that I have made. These have succeeded rather
well, but they have not been followed by a long enough time to be assured of the success of my attempts. It would be highly wished that those who are currently making use of these dyes, or who will
be using them later, apply themselves to perfect them which, I believe, is not absolutely impossible.
Having done this, they would be rather good citizens to not make a mystery of their discoveries, but
only succeed by rendering them public.
Cabinetmakers dye not only their woods for veneer to use them in the place of the natural color
of the woods. They also use these same dyes to accentuate various parts of their works while they
are being worked. As such, these dyes, like the red of Brasilwood, the violet of the Campeachy, the
black, etc., are used hot, which is very easy to do because it is sufficient for only the exterior of the
woods being dyed. Other than these dyes, woodworkers in furniture sometimes use a type of yellow
color for bedsteads, which is composed of yellow ochre and common varnish, or of this same ochre
and the very clear English glue, sometimes they even put it in only water, which is of little use.
Before finishing the dyeing of wood, I believe I ought to give a least-costly method of dyeing white
Page 799
30
helps a lot and at the same time gives a red water, which not only stains the surface of the wood, but
penetrates the interior 3 to 4 lines deep. In staining the wood with this dye, one must take care that all
the pieces be of the same species, and about equal in density if one wishes that they be of equal color
throughout. This observation is general for all water-based stains, which have no palpable thickness
nor even appearance [they leave no residue or any evident change in appearance], which requires the
W
cabinetmaker to make a choice of wood of equal color and a density as I mentioned before. This demands a lot of experience and attention on the part of the cabinetmakers. And with the exception of
the way to compose and use dyes, it is hardly possible to give theoretical rules on this part, for which
success is not often due to anything but experience, which is not acquired except with a lot of time,
attention and work.
S e c t i o n II
On
t h e s aw i n g o f
fo r
Wood
a ppropr i at e
Cabinetmak ing
S the wood that one uses for cabinetmaking is for the most part very expensive, because it costs
Plate 278
roughly 10 sols up to 30 sols, and sometimes even one crown per pound, according to the differ-
ent types of wood, we have great interest in using these woods sparingly; that is why instead of making
furniture or other pieces of cabinetry in solid wood, we have tried to execute splitting [sawing] wood
Explanation
into laminates, or very thin sheets, that one applies on the furniture cases made of ordinary wood.
It is not the carpenter-cabinetmakers who split [saw] their wood, but the workers [sawyers] who
do only this work, and who saw not only for the cabinetmakers, but also for the musical instrument
makers, and generally all those who use thin wood. These workers or sawyers are paid by the pound,
that is to say, according to the weight of the piece of wood that they use, including the waste-wood and
sawdust, rendering the wood close to two-thirds more expensive, which makes a piece made in this
manner very important.
Veneer wood is split [sawn] at about a thickness of 1 line at most [1/12" to 1/14"]; when one wants
to spare it, one makes from 10 to 11 leaves from a thumb-thickness [inch], which is worthless because
even before the veneer is polished, it has left only a half-blade of thickness [1/24" to 1/32"], which is
then reduced almost to nothing when the piece is finished; it is absolutely necessary to avoid making
veneers this thin, although that is used a lot at the present. When one wants to cut up a piece of wood
to make a veneer, one begins by choosing the side of the log that allows for the easiest sawing, the goal
being to orient the wood for the best advantage, and to yield the largest sheets of the veneer; then one
puts the piece of wood in the vise, and with a standing saw [a saw to be used while standing, and a vise
designed to facilitate that action], one saws it to a thickness that one judges appropriate (which I am
going to explain, after having provided the description of the bench or vise with a standing saw, and
of the saw appropriate for this task).
Page 800
The saw appropriate for cutting wood from India, which we name also the saw with vise, Figs. 1
and 2 [to increase or decrease the tension on the blade] is a little bit similar to the saw for cutting used
by the woodworking builders [often known in the modern era as a frame saw]. It is composed of two
verticals and of two crosswise or crossbeam elements, of which the ends project out and are round-
Plate 278 . The Way to Split Veneer Wood, and Its Explanation
32
ed, so that the two sawyers can hold the saw easily. The middle of these crosspieces is convex on the
outside, in order to give them more strength, and that they not bend while one increases the tension
on the saw blade.
the saw, or, better said, of its chassis. These frames of iron, represented
by Figs. 4, 5, 8 and 9, are made of iron plate, and the largest possible, so
that the saw cannot turn easily, and one tightens a nut to that above, for
putting there a screw a b, Figs. 4 and 5, which serves to control the tension of the saw blade.
On the outside of the cross-members one insets a steel contact plate at-
tached with some screws, which prevents the pressure of the screw of the
frame to not ruin anything nor to make any holes. See Fig. 3.
The blade of the saw, as I just said, is 4 thumbs [inches] size at least, taper-
ing barely toward the back [away from the teeth]. We do not put a set on these
sorts of saws, because that would eat up the wood excessively with an unnecessarily wide kerf, and one takes great care that the teeth be perfectly straight on
the horizontal, and that their teeth be also perfectly equal in height, so that they
grab all equally, and that they do not chatter, resulting in uneven thickness of
the wood, which is also to be feared, which ruins so many sheets of veneer. The
teeth of these saws should be spaced equally, about 5 to 6 lines from one tooth
to the next one at least, and should be positioned in such a way that the bottom
[what we now call the tip] of each tooth is level with one another, because being
so arranged, they are less subject to become dull, which would happen unfailingly if they were made ordinarily, as is seen that almost all wood from India is
hard, and consequently causes more resistance to the teeth of the saw. See Figs. 6 and 7, which represent one part of the saw blade viewed from the front and side, half-size.
Page 8 0 1
The standing saw vise, represented in Fig. 11, is one type of small
bench, about 3 to 3.5 feet long, by 2 feet high, at the base of which one puts
the vise, which serves to hold in place the piece that one wishes to saw.
is good that the brace [the jaw] A, Fig. 11, have about 6 thumbs [inches]
thickness, as well as the top of the bench, in which the screws enter,
which to be good, should have at least 2.5 to 3 thumbs [inches] in thickness, and the threads be long enough so that when there is a piece of wood
8 to 10 thumbs [inches] thickness placed in the vise, there remains at least
enough length of the screw in the bench, as observed in this figure. As
this bench is very short, and is subject to vibration by the movement of
the saw, one loads stones on the bottom shelf to make it more solid; but I believe it would be better to
make the legs of the bench long enough to be anchored to the floor of the shop, then one makes a hole
in front of the bench to set in the piece of wood to be sawn in order to not extend upwards more than 3
On Sawing Veneer
33
feet above the top of the vise, locating it thus both for the comfort of the sawyers and for maximizing
the yield of the piece being sawn. Not all the standing saw vises are part of an overall bench, such as
the one represented here, in Figs. 10 and 11; this is why ordinary vises attached to a little bench are less
solid than making them as I propose here.
When one wishes to saw with the vise, one begins by placing the piece to saw in the vise, of which
W
the screws tighten with an iron lever, that one removes after being worked, so that it is not in the way;
then, with an ordinary saw, one begins to mark all the lines to be sawn on the end of the workpiece, just
up to 2 to 3 lines deep [3/16"], then one uses the frame saw, Fig. 1, which is guided horizontally by two
men, observing the advantageous slight incline on the side of the tooth rake, and of the lifting up of
Plate 278
does
not
When one saws with a vise, one begins with the outside edge of the log, so that the first sheets
sawn bend away from the log and facilitate the passage of the saw, which could not be the case if one
sawed in the middle; as one does when one saws large pieces of wood being used by carpenters or by
ordinary woodworkers, given that the frame saw blade is very thin, and that it has no set. Sawyers at a
vise do not lay out or mark a line on the side of the piece that they wish to saw; but after having begun
on the end with an ordinary saw, they continue the rest by eye, which they do very well, for the most
part; they are very sure to saw their veneers not only very straight, but still perfectly of equal thickness, as well. See Fig. 11, which represents the cut of the bench or upright vise saw, and a piece of wood
sawn into sheets just up to the middle.
To finish what this looks like at the cutting of wood appropriate to the cabinetmaker I have
represented in Fig. 12, a saw named the carving saw, which serves to cut up not only hard wood,
whether wood with the grain or
cross-grain, or standing wood,
but also coral, ivory and motherof-pearl. The framework of these
sorts of saws is all iron, of which
the upper branch is widened on
the outside, so that one can adapt the blade and set it as one judges appropriate, which is done in
the following manner.
After having pierced a hole in the blade of the saw, b, corresponding with that of the lower arm
of the frame of the saw, you put this one [arm], and the one that is opposite, in a vise or other thing
capable of bending them [squeezing them together], in a manner that they tend to meet one against
Page 8 0 2
34
the other, and tightens them as much as is judged appropriate, to give the saw all the tension necessary; then the blade of the saw, being stopped at point b, one makes it enter in the upper arm of the
frame, and one traces the place for the hole at point a, which one pierces to place there a peg; this
being done, one again bends the arms of the frame, just until it gives liberty to pass the peg below,
and which serves to hold the peg in place, as one can see in this figure.
W
The blades for these sorts of saws are very thin, and one does not give them a set, so they have a
very narrow kerf and lose less material, and they pass easily; one thins them on the back [away from
the teeth], which one does with a file that one passes down the length just until they are thinned
Plate 278
enough as one judges appropriate; then one rubs them with sand to remove the unevenness that the
filing could have made; this operation is called demaigrir [thinning], a workers term.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
On Sawing Veneer
n plate 2 78 , Roubo illustrates methods for re-sawing lumber, or more precisely, sawing veneer from
solid stock.
I use the phrase more precisely self-consciously for reasons you will learn in a moment. A short session
of hand re-sawing veneer instantly demonstrated the level of hand skill required to do what I treated as being
akin to breathing air, so natural as to not need a lot of explanation. Like much of Roubo, attempting to replicate the work is a challenging and humbling experience.
The practice of re-sawing as described in Roubo, and also illustrated in Denis Diderots Encyclopdie, re-
quires a substantial dedication of space with its sawing bench, a 4'-long frame saw and two sturdy lads, one at
either end. In the near future I expect to have all the space I need for a 4' saw, but for now that is not practical.
My day-to-day personal workshop is not capacious perhaps it could even be described as cramped so I
pursued a scaled-down version of the same general concept.
The frame saw I built holds a 28" x 2" x 3 tpi rip-tooth blade I bought at a tool store. The frame itself is 8/4
hard maple with stout but unglued mortise-and-tenon joints. Long, straight and robust tenons are the key to
keeping the saw frame straight and planar when under tension and in use. With good joinery, the mere tightening of the saw blade holds everything square and true. The hardware used to affix the blade to the frame
was made with bar stock, screws and bolts from the hardware store. Using wrenches to tighten the tensioning
bolt, I can get the blade so tight it sounds like a piano string when plucked.
incomprehensible.
On Sawing Veneer
35
When the carving saws serve to saw ivory or other hard materials, it is necessary that their blades
be harder than for ordinary wood; so one uses saws of tempered steel, or of springs from pendulums
[clock springs], which are appropriate for this use, and which one thins on the grindstone, their hardness making them impervious to the teeth of a file.
The teeth of these saws are normally inclined; and to these ends when sawing ivory, one files
W
them diagonally from both sides, in order that they present an acute point [double rake], which divides the material without splitting it.
Carving saws are of different sizes, according to different needs. The one represented here is 1/4
Plate 278
There are other saws, named English saws, where the blade tension is kept by the means of a
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
steady. The shelf underneath the bench serves as a perfect platform for stacking it with stones, bricks or an
anvil or two.
For clamping screws I re-used (actually, cannibalized) two vintage leg-vise units I picked up at an auc-
tion a few summers ago. Because I do not yet possess the tap of the size required to shoot threads all the way
through the width of the top, I merely drilled holes to accommodate the screws and affixed the thread nuts
to the face of the bench. It was sufficient. A pair of wedging plates to hold a round log in place while sawing
completed the bench, and the sawing could begin.
Working alone at first, I placed some old-growth antique cypress lumber into the jaws and started saw-
ing. Remarkable! Re-sawing a 6" piece by myself, I averaged almost 11/2" per minute. Effortlessly. Eight inch
old-growth mahogany you ask? Like butter. With the weight of the 8/4 maple frame pulling the sharp teeth
through the wood, all I had to do was steer and keep it moving steadily back and forth.
The only negative in my report is that the saw is so precise that it has no forgiveness in its heart, and so it
amplifies any errors on the part of the sawyer. Keep to the line and everything is glorious! Wander a little and
the workpiece is wasted. There is just no way to recover from a misdirection, an unfortunate feature of this
technique that I found disheartening, because almost every other technique of hand sawing allows for some
recovery from a wandering saw.
To use this simple and powerful tool effectively requires a level of hand skill that comes only with diligent
practice. A second sawyer is necessary for precise work, as two sets of hands and eyes are required to control
the saw and follow the desired cut line on both the front and back sides.
The fact that cabinetmakers of the past cut acres of 18"-wide veneers less than 1/12" thick is a tangible testi-
monial to the virtues of an apprentice system that nurtured craftsmanship and demanded perfection.
tension.
the floor. The only way I could get the bench to stop
36
screw [like a modern hacksaw where the blade is affixed to a threaded rod with a pin that tightens via
a butterfly nut]. I will mention these when treating the Tools of the Woodworker-Cabinetmakers,
for which the description is going to be the object of the following paragraph.
basics of woodworking construction, as per example the workbench and sharpening tools; however
there are many more that are very different from the ones of the ordinary woodworkers and require
a particular explanation, and as follows I will give a short description of the lathe and its tools in all
Boite a Rappel
relativity to the cabinetmaker, as well as other tools and instruments, which their use is necessary in
Page 803
Because cabinetmakers are making a lot of small works, they need to be more cautious in their
construction than for other types of woodworking; this is why we have created a workbench more
convenient than the one used regularly [for joiners].
This bench, represented in Fig. 1, is named the German bench (being that it was invented in Ger-
many, or, more likely, by German cabinetmakers, which are great in numbers in Paris).This bench,
as I say, is made, like all others, with four feet and at the extreme end of which is placed a fully housed
[or boxed] end vise serving to hold the wood in place on the bench, of such length that it fits whatever pieces are being worked, and without having a need for a clamp. This is done by means of two
hooks or small strips of iron [bench dogs], Figs. 1a and 1b, of which one is placed in the vise housing [or
box], and the other on the bench, and which one can change the place of as one judges appropriate,
as I will explain later.
The end vise, which is the most essential part of the bench of which I am speaking, is 14 to 15
thumbs in length, with 3.5 thumbs of width, having a thickness equal to that of the bench, which is
originally 4 thumbs. This vise box, represented full scale in Figs. 5, 6, 7, 8, 10 and 12 is hollow inside,
for the passage of the screw and its nut, and is composed of four pieces or sides (A, B, C, D; Fig. 7)
of a head piece E, (Fig. 5), and another strong piece or head, F (Figs. 6 and 8), across which passes
the iron hook [bench dog], as you can see in Fig. 6, which represents the box open in front, with the
mortise for the hook, indicated by the perforated lines a, b, c, d, and the Fig. 8, which represents the
longitudinal profile of this same box, and of the mortise for placing the hook. The piece F, presents
itself on the end of the grain [the end of the screw abuts the end grain], Fig. 6, and it is constructed in
a manner where it is notched in a cavity below the bench, where it passes in the form of a line, about
8 to 10 thumbs in length, and is retained in place by a rod G (Figs. 6 and 7), through which it enters a
notch. This rod, G, serves to support the bottom of the box at the location of the hook, and to relieve
the tab or tongue of the back piece, which without, the line of piece F would support the weight of the
vise all by itself.
The tab/tongue at the back of piece B (Fig. 7), of which I just spoke, should have little height, to
be able to manage the force of the flange that it restrains, with which one augments the strength with
countersunk screws [we believe Roubo actually means piece D instead of B here], which one puts
there at a remote distance from each other, as one can see in this figure, which represents the sidewise
cut-away view of the box, and of one part of the bench. Look at Fig. 10, which represents the side of
the bench ready to receive the box, with bar GG, which supports it.
Plate 279 . The German Bench with Its Explanation of the Boite a Rappel
38
Page 804
The head of the box E (Figs. 5, 6, and 8) is joined to the piece below, and above, with tongue and
groove, and is held there with a screw, or at least by glued pegs; the piece up above, A, has a notch
above the main piece F (Fig. 8), and there is likewise stopped, and that piece below [C] is assembled
by mortise and tenon, as one can see in this figure.
Piece B at the back, Fig. 7, is attached with screws at the head of the box and on the principal piece
W
[F], as one could see in Fig. 12, which represents the box viewed from the back; the piece at the front,
which serves as a door, is attached similarly with screws, which cannot be held in place until one has
attached a nut.
Plate 279
The screw that serves to move this bolt is made of iron, as well as its nut, which is the most solid;
however, when one wants to economize, one makes the screw in wood, and the nut in iron, coated
with lead mixed with antimony, as I will explain later.
Whether the screw is made of wood or of iron, it is necessary that it be the whole length of the box,
the
bench.
same way, and 8 to 9 lines of diameter are sufficient. While it is of wood, it is necessary, to be solid, that it be at least 15 lines of diameter.
Screw nuts of iron are normally made, and one makes a long tab of about 6 thumbs, which fits
into the mortise through the thickness of the bench, with which one
holds it by means of two screw bolts, which one distances one from
the other as much as possible, in order that the nut will be less subject
to being wobbled. See Fig. 7, where the nut is represented from the
side view with these two bolts; and Fig. 9, which represents this same
nut viewed from above.
above and the other below, which serve to retain the box and to relieve
the grooves of the back piece, as I have shown, Fig. 7.
39
When the screws are made in wood, the nuts are made the same as those above, at least as to the
exterior, and one makes them large enough so that the hole that one makes in the middle is about 3 to
4 lines larger than the size of the screw. One flares [widens] the hole on two sides, and one pierces the
holes on four sides of the nut, so that the materials that trickle out when one makes the screw nut hold
there solidly. When the nut is thus done, one places it in the bench, and one secures it with bolts, obW
serving that the grip of the screw is found very tight with regards to the box; then one takes one piece
of screw that is identical to the one to be used, and one coats it about a half-line of thickness with very
fine clay, ground up with glue, which is necessary for the set of the screw, and to prevent the materials
being burned by the heat of the screw. When this coating is dry, one places the dummy screw in the
Plate 279
nut, and one coats it with clay at the outside edge to prevent the materials from escaping; then one
flows the materials between the nut and the dummy screw, which one removes when the material is
cooled, and the nut is made. Look at Figs. 6, 7, 8, where the nut is represented from the front, the side
and the profile, and where one sees the manner by which it is positioned, also the materials that form
the nut, which is composed of lead and antimony, of which the correct quantity of lead is about one
to two. [Casting the nut in place with molten pewter.]
The hooks or latches of iron [bench dogs], which serve the bench as I just described, have about
10 lines squared of bulk, and a length equal to the thickness of the bench, plus 9
to 10 lines, which it is necessary to make level, which makes about 5 thumbs of
length; and you fit there one or two springs on the sides, so that they stay in place
at the desired height that you judge appropriate. See Figs. 2 and 3.
The hook [bench dog], which is positioned on the bench in whichever mor-
tise one judges appropriate, in the holes that are drilled 4 thumbs of distance, one from the other, and
at 18 lines from the front edge of the bench, so that their midsection meets just with the middle point
of the screw, see Fig. 5.
The holes where the hooks are placed are drilled on a slant, which contrasts to that of the boxed
vise [as] one can see in Fig. 6, so that when one compresses the
wood between the hooks, the force of the pressure does not
release the hook, and so that piece ef, Fig. 1, which is placed
between them, does not get loose.
place one vise on a side of this sort of bench, and for more
convenience, one places a second, adapted with a movable
post, which straddles the front crosspiece at the base of the
bench, and a groove and a tongue in the upper portion. This
second vise moves from one side of the bench to the other,
as one judges appropriate, and can be removed very easily,
given that the tongue on the crosspiece below is eliminated close to the foot on the front of the bench.
These vises can be separated as much as one judges appropriate; and to hold them equally open
from bottom to top, one adds at the base of their side beams a small board, Fig.
4, which passes across the vise foot, and which is drilled with many holes, in
which one places an iron peg, which stops them at a convenient distance.
In spite of the vises of which I am speaking, one places in the thickness of the benchtop little
vises, mounted with iron screws, of which the jaws are a size equal to the thickness of the table of
Page 805
40
the bench. Look at Fig. 11, where I have represented this vise in profile, with its screw and nut, and
where I have indicated by dotted lines, the placement of the hooks [bench dogs], the slot for the nut
or screw assemblies, which need only three-fourths the thickness of
the table, which is most proper, and works better than to make them
pass through, which is the custom.
W
Page 806
Plate 279
bench, one can place some wood, tapped into the thickness of the
bench, at the extremity of which one notices a hub, to pass a crank, enabling them to move. Sometimes these screws do not have any head, and are [instead] tapped down the entire length; the type
that at one end screws into the bench, and at the other into a framework, which presses against the
side brace, which amounts to be the same thing as when they have heads; the exception being when
it happens that the framework is positioned perpendicularly, it surpasses the top of the side brace of
the press, which is rather inconvenient.
After benches, vises are the greatest tools of the cabinetmaker. They are of two sorts, namely,
those of Figs. 1 and 3, in which the movement is made horizontally, and of which the screws have
holes to receive iron rods, serving to move them. These vises are composed of two twin vertical
supports, AB and CD, which are of 5 to 6 thumbs in size, by 3 to 4 thumbs of thickness, because of
their lengths [in order to resist bending], which varies between 2 to 4 feet, that is to say, in that of
AB, the screws are tapped in [the rear jaw], instead of their entering completely [being tapped] in
the other [front jaw]. The length of screws of these vises should be around two-thirds the length
of the former [vises mentioned in the previous section], and 2 to 3 thumbs diameter. One should
take care that their heads are bound with an iron ring in order to prevent their splitting while one
forces them in making them move. See Figs. 1 and 3. One makes use of these vises on the bench, in
order to saw while upright, to work on the piece, or to glue the work. In one or the other of these
different cases, one secures the bench vise with two clamps so that they will be held in a fixed and
unvaried manner.
The other type of vise, Figs. 2 and 4, works vertically [and is more like a press] and is composed,
like the first one, of two vertical supports and two screws, with the exception that the screws are
inserted and pegged in the bottom beam, which is larger than the other [beam], so that it can present
more of a footing to the vise. The top beam moves by itself and is stopped by two frames a, b and c, d,
Figs. 2 and 4, in[to] which the screws are tapped. These types of vises are of different sizes, as are the
first, and serve for gluing as much for solid wood as for veneer work.
After vises, the greatest tool and one of the most necessary is the grindstone, which serves to
sharpen the tools, and which is a much better way [to sharpen] than what can be done with a sandstone, like all the other joiners do.
The grindstone (Figs. 5, 6, 7) is composed of a base of about 16 to 17 thumbs high, by 2 to 2.5 feet
42
long, and 9 to 10 thumbs wide. On this base is placed a trough of oak wood, of a size sufficient to contain a stone of about 2 feet in diameter, and 2 to 3 thumbs thickness. This trough should be well assembled, and its bottom should be about 2 to 3 thumbs thick, so that it will be hollowed below to receive
Page 807
the base, on which one secures it with some screws. One end of the trough should be elevated above
the grindstone drilled with a hole in which one sets a piece of wood (e and f Fig. 6), which serves to
W
work mother-of-pearl, about which I will speak at the appropriate time. At the other end of the stone
is a board serving to prop up the hand of whoever is sharpening. This board is supported by two iron
bands, attached to two sides of the trough with two screws g, in a manner so that they can turn freely.
Plate 280
The other end h is attached with two other bands of flat iron, at the lower extremity of which one
makes a type of groove of 5 to 6 lines in size, by 3 to 4 thumbs in length, which serves to pass a screw
through the side of the trough, where the screw nut i, which has a projecting head, being iron, holds
in place the iron bar, and consequently the board with which it is attached, which one lowers or raises
as one judges proper, according to the diameter of the stone.
The interior of the trough is normally fitted with very thin lead [sheet] to better hold water, which
is better than doing it totally with bare [wood]. When they are well assembled, and they have served
for a while, they hold water rather well, which, in filling the pores of the wood, they absorb the waste
from the grindstone, which makes a type of adhesive.
The grindstone is of a sandstone that comes from Languedoc, Auvergne or Champagne; but
those that come from Langres are the highest valued; their color is of a bleached-grey/white-grey.
For them to be good, it is necessary that their grain is fine, level and open, because when it is too tight,
that is a sign that the stone is hard.
The stone is mounted on an axle to which it is fastened, blocked with wedges of oak. This axle
is positioned on collars of copper, in which are applied grooves, where some type of straps or rings
are fitted, put on two sides
of the tree-fork, which prevent it from slipping to either side, and consequently
keeps the stone in the middle of the trough.
The collars of copper
have two parts, the bottom
one is attached to the sides
of the trough; the one on
top is removable, and sets
down upon the axle where
it presses by means of a screw, which is tapped in an iron tie, which is screwed to the trough wall.
The grindstone turns by foot or by hand, according to what is appropriate; in either case, the
crank handle works nonstop, because when one turns it by foot, there is a cord attached on the
step, or pedal. See Figs. 5, 6 and 7, where one is represented in the top view from the end; the other
is represented viewed from the side. The last is the cross-section view, which is sufficient to give
all the information necessary for this type of tool, which, however useful to joiners, is most often
made by locksmith-machinists and others. This is why those who would like to acquire a perfect
knowledge of all types of grindstones could refer to the Arts of the Cutlery and of the Turner, where
they are amply described.
43
Although I say that cabinetmakers sharpen their tools on the wheel, it is not that they do not also
use a sandstone like the other woodworkers; but that the wheel is more in use, both for sharpening
well as for doing it with more diligence.
When the tools are sharpened, whether on the sandstone or on the grindstone, in order to give
them a sharp and fine edge, one passes them over an oiled stone, which renders them ready to cut
W
These stones come from Levant; the best are those that are of a blonde color, of a tight grain,
uniform and very even. This stone has the fault of having little white veins, some lengthwise, some
crosswise, and so many warts that prevent a really good sharpening. That is why it is necessary to
Plate 280
choose those which are the most even, or at least, which will have the smallest veins, and in the least
quantity.
As these stones are not of a very large size, and are otherwise fragile, one encases them in a piece
One uses oil to sharpen tools on these stones, and it is good that they always be able to soak it in,
and that they are always very neat and clean, because the dust that settles forms a film that ruins them,
and prevents the tools from working on the stone.
The oilstones are expensive; those that are passably good cost from 4 francs up to 100 sous per
pound; but also, when they are found to be good, this is a big advantage that diminishes their costliness a great deal.
The sharpening tools of cabinetmakers, like clamps and bench hold-down clamps, fillisters, var-
lopes [varlopes are planes to dress the edge or flat of long boards] rebate-planes, etc., are the same as
those of other woodworkers. It is only the planes that are a little different, although cabinetmakers
also make use of ordinary [joiners] planes, for the construction of their frames and other works of
less importance.
Plate 281
Various Sorts of
Planes Appropriate for
Cabinetmakers
The planes of cabinetmakers [veneerworkers] (Figs. 1 and 2) differ from ordinary planes by the
Page 809
45
Whenever these planes be upright, one hardly makes use of them with ordinary iron, whether
upright or perpendicular. Because when it concerns planing flat veneer, whether in loose sheets or
glued down, it is necessary therefore to use a toothing iron, which will not make any splinters, so that
it will only scour the wood, whether on grain [along the grain] or cross-grain.
The toothing irons represented in Fig. 3, are the same form and size as ordinary irons, with the
W
The channels of the toothing iron are carved on the face in a triangular form, and between each
of them there is a little flat ridge, which is sharp [at the cutting edge], given that the iron being sharpened, each of these ridges forms a type of tooth of a squared form, which increase in the thickness
Plate 281
There are some notched irons of which the teeth are more or less large, according to the different
needs. See Fig. 3 where I have represented one made with great skill, in which the teeth are of an average size, and where more than half are smaller, and half are larger, of which one makes use according
to the diverse sorts of works, as I will speak later.
The bodies of the planes of which I speak do not have anything different from those of other
woodworkers: there are some straight ones, like in Figs. 1 and 2, and some curved on the base, in
which one puts smooth irons, or irons with teeth, as the case requires.
Joining planes, for making miters for cabinetmakers, are constructed in the same manner as other
planes, where they differ only in their length, which vary from 10 to 16 thumbs.
There are low-angle planes with an iron base, Fig. 4, which are appropriate for working very
hard wood or [working] against the grain, with end grain and with soft metals. These sorts of planes
differ from others only by the slant of their throat [and bed], which is extremely inclined, and by the
position of their iron, which is turned facing upside-down, as one can see in this figure, which is represented in this cross-section, so that one can see the slant of the throat, and the position of the iron,
which should be level with the bottom of the iron sole, against which it is pressed.
The throat [escapement] of the low-angle plane, as it is represented here, becomes a bit longer,
and shortens by consequence of the force of the tool; it is why one makes them with two touch-holes,
one that serves to place the iron and the wedge, and the other indicated by lines e, f, g, h, by which the
shavings exit. This manner of laying out the touch-holes is the least utilized; however, it is the best.
The iron sole is held to the body of wood by means of six screws, which pass through the sole and
are tapped into the thickness of the body, as one can see in Fig. 5, which represents the cross-section
at the location of two screws. See also Fig. 6, which represents this same plane viewed from below,
with the screw end.
What I just said about the planes with iron sole could also apply to other planes, which it would
be good to construct in the same fashion, especially those used to work end grain, or those woods that
are extremely coarse, of which the rubbing ruins in little time the bottom of the plane, which therefore cannot work as it should, at least until one resurfaces the plane bottom properly.
It is this difficulty which has made us create iron planes, as much for the metals as for hard wood.
These tools, represented by Figs. 8, 9, 10 and 11, are constructed in the form of empty boxes from the
top, in which one adds a wooden wedge A, Fig. 9, which serves to retain the back of the iron; whenever one wishes that this iron be set perpendicularly, as it is in Figs. 8 and 9, [or] placing it at an incline,
as indicated by dotted lines BC; in this case one fits the slanted wooden wedge according to how
one judges appropriate to give [an angle to] the iron of the tool. Look at Fig. 7, which represents this
wedge viewed in profile and from above.
Page 810
46
The iron for these sorts of planes is held in place by means of another wedge D, Fig. 9, which rubs
against the iron pin E, which holds both sides or flanges of the plane.
the pin E cannot work anymore, and one places another pin
F, which enters through one side of the plane, and as a square
W
on the other side, so that one can take it out when one judges
appropriate, and which does not turn when in place.
Plate 281
As for the construction of these sorts of tools, I am going to make no mention, given that it is the
Cabinetmakers use saws similar to those of building carpenters. That is why I will not speak any
more of them [carpenters saws]. However, it is good to note that cabinetmakers make more use of
turning saws, quite another thing, for both sawing lengthwise or widthwise. That is why they have
them [compass or bow saws] in different sizes, to make use of them as need be. This method is very
good for sawing wood that is not very thick; but I do not know why they use turning saws to saw end
grain or joinery, [tenons], etc., given that one ordinary saw is better than one turning saw for end
grain, especially when it is necessary that these last be cut with precision.
Page 811
I will not speak here either of the saw for carving or for marquetry, because it would be better to
do that while treating the question of how to use them. It would be the same with many other tools, of
which I will not make mention except that I describe the different works of the cabinetmaker, or better said, the diverse operations of the cabinetmaker, which will avoid all type of repetition, and will
be altogether more natural when describing the tool. I will give the manner of using them according
to different cases; that is why I make this description of other tools in the following chapters, leaving
me only to speak of the construction of framing, which will be the object of the following section.
S e c t i o n III
T h e F r a m e s [C a s e s ] a ppropr i at e t o r e c e i v e
Ve n e e rwo r k , a n d h ow t o pr e pa r e t h e m
a n d co n s t ru c t t h e m
HE preparation and construction of frames [cases] appropriate to receive veneerwork, al-
Plate 282
The Way of
Preparing Frames
to Receive
Veneerwork
though neglected by the majority of ordinary workers, are however of a very great consequence.
All the success of the final work is reflected in the degree of perfection with which the frames are
treated. However well made they are, they cannot subsist for long if the frames that carry the veneering are not built with all the necessary care.
Three things contribute to the perfection of the frames to receive the veneerwork: first, the choice
of material; second, the general preparation; and third, good construction. I said at the beginning of
this part, page 766, that the appropriate woods for construction of frames are soft oak, pine, linden
and all other soft and dry woods. However, you must pay attention that only the first, that is to say,
the oak, that one [should] use for works of any consequence. The others, however light and little
47
troublesome they are, are not solid enough to make excellent work. That is why one should never use
another wood than oak when the frames [cases] are appropriate for joined works, such as bureaus,
secretaries, etc. We have only tables and other works of this nature that can reasonably be made of
white wood. The cabinet workers [veneerworkers] do not make the frames [cases] themselves, but
they are made cheaply by other woodworkers who do nothing but that, and who use invariably other
W
sorts of wood, which they always find to be good, provided they are not expensive. They make frames
Page 812
with wood for boats, that is, oak or pine, [or] staves from barrels, of linden, poplar, chestnut and other
bad woods. To poor quality woods, they add very bad workmanship; the most part of their frames
[cases] being made with little joinery, and then only poor butt joins. When the work is curved in
plan, they make the panels flat-joined and glued onto battens as much on grain as cross grain, which
makes a very bad piece, which is destroyed in very little time.**** That is why I do not know how to
exhort strongly enough that woodworkers/cabinetmakers make their frames of good oak, a bit soft
and very dry, so they take glue better, and they are less obvious [do not telegraph through the veneer]
as a result.
After the choice of wood appropriate for making frames, the cabinetmakers should take care of
their overall preparations. They should determine the shapes after having very exactly designed the
whole work, in plan as well as elevation, so they do not arrive at having a shape or section unbroken.
This sometimes happens when you are working for the frames to be finished to finally determine the
shape and size of the sections that are to be veneered.
When designing, whether the plan or the elevations of the work, and consequently of the frames,
When these same doors or drawers will overlap, that is to say, they will put the frame on the pro-
for the most part, do not wish to sell work as totally finished
they work, for the most part, only for the merchants who pay
for their works only what they wish. The prevailing display
for the very rich people who have the means to pay well.
out having the means to pay what they are worth. This has
Plate 282
49
jection of the work, as in Fig. 2, and that this body will not add any thickness, it is necessary to make
a double rebate at the back, which bears the weight of opening, and consequently relieves the [pressure on the] veneer from outside.
When the doors or drawers are completely on view, as in Figs. 1 and 5, one must always make such
rebates, so that the front piece of the frames, Fig. 1, always be even, and so the opening part always is
W
I do not know why cabinetmakers do not use this custom, being content, for the most part, to
Page 813
place the base of the drawers on the battens. For as little as these are disturbed, or are worked, as with
the backs of the drawers, the front of the work does not match, which makes a very bad effect.
Plate 282
As to the forms and the construction of frames, I am not going to speak but very succinctly, given
that I treated this part at length when I made the description of woodworking for furniture, which
you can review. All that I can recommend here is that you must construct these frames with all the
precaution and firmness possible, so that they do not make a [bad] showing when they are used.
That is why it is good, after having planed the wood and made the joints, to let them dry in a place
where they will not be too exposed to heat or humidity. You must pay the same attention to the panels, that they be well dried before and after having glued them. You then assemble the whole thing,
and you glue the joinery so as not to use any pegs, at least in those places covered with veneer, because
if they come out of their place, they will lift [the veneer], which you could prevent, by gluing the pegs.
[He must be referring to assembly with trunnels and glue, as opposed to simply pinning or nailing the
structure].
Because the panels for frames of which I just spoke should be leveled, it is good that they be thick,
that is to say they must have 9 lines of thickness at least, so [they] do they not bend while being worked
from above.
All the doors, tabletops and other works of this nature should be joined all round, that is to say,
they be composed of a panel and a frame, assembled along the grain-line. Even when tables and other
works are of a certain size, you would do well to use instead of a panel [slab], to use an assembled parquet [like flooring], so that they do not bend/warp in any way.
You must also avoid, when you construct frames, that any end grain appear because the glue does
not take as well in this direction of wood, as along the grain line. You should take care, as much as is
possible, that they be set up before being veneered on the outside. That is why it sometimes happens
that for the interior of frames [cases] to be veneered, you begin by making the interior veneer work
and polishing it, then you continue to assemble the frame like normally.
When the frames interiors are thus finished, you must take great care that in re-planing them,
they be well trimmed in all directions, and that there be no type of unevenness, because if any remains, they will pierce [distort] the veneer when you go to finish it.
When you re-plane frames, you must use a low-angle plane with teeth, with the largest iron notch-
es, which you pass over in all directions, as much for trimming the work as for striping [grooving] it
in all directions so that the glue attaches better. When there are projections to the work, you finish
the returning parts first [the edges perpendicular to the front plane], at least as much as possible, and
you work where you find the joints, which make the executions easier, as I observed in Figs. 1 and 2.
If, however, you cannot do the joints where there are projections, without doing damage to the wood
by thinning and weakening the wood, you must then make the joints farther from the projections in
order to save the width and the solidity of the wood, as I observed in Fig. 3. In this case, since the plane
Page 814
50
or notched [toothed] jointing-plane cannot pass across the piece, you put the panel in place before
gluing it within the frame, with which it matches piece to piece. Then you shave the panel separately
with the toothed plane, and you glue it to the frame, over the recessed part of which you pass some
sort of wood scraper, so as to stripe it a bit like the notched [toothed] plane.
When the frames are all finished, it is good to let them dry for some time before covering them
W
with veneer, because by not being covered right after being finished, you are always able to repair
them, supposing that they need something, although you have taken all the precautions that I have
recommended above.
Plate 282
Plate 282
Chapter 11
the art of marquetry, nevertheless demands no less attention. Whats more, the
procedures that serve this first type of woodworking are a little like the same as for
the two other types. That is why it is essential to make an exact description of this
Page 815
technique, which being once made will shorten greatly the necessary description of the two other
techniques and will aid in their understanding.
Before beginning the veneering process, it is necessary that the carcases be entirely completed
and the patterns finalized in order to trace directly on the carcase, the main outline of the composition one wishes to make, as I just said above; all that remains is to make the choice of woods that
one should use.
Two reasons should determine the choice of wood for veneering: they are economy and ap-
propriateness. The first of these two reasons normally has no place other than for a work of little
consequence which does not cost much and obliges one therefore to use inexpensive, common
woods. In beautiful works one should not spare any expense in order to have wood of one color or
a particular quality. The second reason, that of suitability to the composition, is the most essential
because it is important not to put together woods of opposing colors, or of unequal hardness, which
require different types of work, whether in their planing or polishing.
A too-great disparity of color is also a very large defect because the different compositions con-
trast too much with one another. I do not pretend, however, that one should put together only
woods of a very similar color (which would produce a monochromatic effect almost as disagreeable
to see as a great disparity of color), only to not put together woods of which the colors, however
different one from the other, should not be too much in opposition to the other, like black with
white, red with yellow, etc. It is not that one cannot use woods of opposing colors in the same composition; but if one does it, one should put a medium color between these woods, which makes a
nuanced transition between the opposing colors so that the transfer from one to the other is not
so disturbing to the eye. This observation should take place only for large compositions in veneer
work; because for the small compositions like applied banding and stringing, it is good that they are
distinguished by their opposing color to that of the veneer background, so that they become more
apparent. [In this passage Roubo is presenting some fundamental training in balanced composition.
It is not simply the oxidation of age that yields a subtle surface and composition that we so admire.
51
Page 816
52
By following Roubos guidance, the craftsman is prevented from creating garish compositions.]
If the too-large difference in color of wood makes a bad effect on the face [front] of the work, it is
equally true for the side of the cabinet, when one finds a thickness of wood of one color, and a surface
of different thickness on another, like those of G, I, H, Fig. 4, because the veneered area appears to
be of three parts and of different colors. Cabinetmakers do not ordinarily pay a lot of atW
tention to this inconvenience, being content only to put all apparent thicknesses on the
side of the work, which lets the default of which I speak remain, as one can see in Fig.
1, where, on the sides A & B, one sees the thickness of the veneer from the front side of
Plate 282
the work. That is why I believe that it is better, while one has protruding corners to be
veneered with woods of different colors, to miter the joints, as those in E and F, Fig. 1,
which works very well, and has no difficulty other than its execution, which becomes
a bit more complicated than the ordinary, without however being impossible, because
it concerns only dressing the joint of the piece that is glued first with a rebate plane of which the bottom, and by consequence the iron, is at an angle to the
sole. A rebate plane set up as previously described can
absolutely suffice; however, it is good to have two with
opposing blade angles in order to make use as needed; it
is even good to have one on the side, so that in the case
where the two sides were to be glued first, as in E and F,
Fig. 1, one can trim them without one damaging the other
[Roubo does not show this plane or give its formal name,
so we are left to guess as to exactly what the tool looks
like]. Other than the case where the woods of the protruding corners are of different colors, it is not
necessary to make miter joints; it is sufficient that the edge of the veneer be on the less obvious side
of the piece, as I have shown in all the figures on this plate.
Veneer on returns [such as A and F, Fig. 1] is glued first, and then one glues the one from the pro-
truding sides; and when it happens that there are some [cutaway] canted sides, as in Fig. 1, it is good to
make some cuts as in angle D, because when one glues the side piece against the piece of the [cutaway
or canted] area, as in angle C, the slant that one is required to give to the side piece creates an edge
subject to damage when one is finishing the work. It is necessary, however, to observe that when one
makes a cut like D, it is necessary to make it of equal thickness, the first piece being glued and nearly
finished before gluing the other, so that on finishing the work, angle D is always in the middle of the
joint. [Roubo is clearly teaching the reader that there is a definite order to gluing down veneers on
complex angular surfaces.]
When it happens that the carcase structure has a little projection, but that this projection is more
considerable than the thickness of the veneer, it is necessary then to glue the veneer on the back
side, and to finish it on its edge; next one veneers to the side of the projection a little section of the
same wood that one would be veneering on the front piece, which being well joined with this last
part, appears to be but one and the same piece.
Page 817
There are occasions where changing the veneer colors is advantageous, as in the case of a shelf,
represented in Fig. 7. This variation is available two ways: to glue the veneer of the edges horizontally last in such a way that it borders the edge of the shelf, as I just showed in this figure, sides G
& I. The second manner of positioning the veneer on the shelves of which I speak, side H and L,
54
same figure, is, in contrast to the first; that is to say, that it is necessary to glue the veneer from the
top and bottom last, in such a way that it overlaps the sides of the work. In one or the other case,
it is certain that the difference in the veneer color works very
well; but it can only be in the case of a small shelf or other small
works of this type.
W
Figs. 11 & 12, and the veneers are of different colors, it is good
to make the joints mitered, as in Fig. 11, unless one wishes to set
Plate 282
askew the wood of the sides M & N, Fig. 12, and let the thickness
of these two sides serve as a border.
When rounded corners are veneered, as in this figure and
that of Fig. 6, one uses thicker veneer than ordinarily because
of the rounded angle, and one glues it first, [also] while it will be
placed between two pieces, as in Fig. 12, and the last when it forms a sharp edge on
a piece, as in Fig. 6.
While one does not ordinarily veneer anything but flat panels, one can, how-
ever, very well make some protrusions with mouldings in [a] different color
wood, which will work very well and will not require but a bit of care on the part
of the worker; that is why I believe that it is very easy to make and more commonly
done than one thinks. See Figs. 8, 9, and 10, where
I have traced various sorts of mouldings, of which
the sections are made of wood of different colors,
and of which the construction and disposition can
be applied to all sorts of works, of whatever nature they might be.
Section I
The
d i v e r s e s o rt s o f
s o m e de t a i l a n d t h e
Veneer Wood
Arrangement
in general:
o f wo o d v e n e e r
Plate 283
The Way to Cut Out
Compositions
bly of diverse compositions of which it is comprised. These compositions are of two types: large
and small.
The large compositions are ordinarily made of large pieces of veneer wood, often of the same spe-
cies, of which the joints and the position of their grain oppose one another, forming different figures
Page 818
in either the overall composition of the work or in the various pieces of which it is comprised.
Small compositions differ from the first not only in their size, but also by the different woods that
are used; the smaller compositions are often surrounded by banding and stringwork that separate
them from the rest of the piece, as we will see later.
In general, the divisions, large as well as small, are formed by straight lines or circular lines, or
finally a composition of one or the other, which makes the work more or less difficult to do [because
the lines are so crisp and any errors are immediately apparent], whether one makes use of a little bit
56
of the same procedures for executing the different compositions. That is why I am going to begin
with the first technique of compositions composed of straight lines, as being the easiest, for which I
will give some general examples, applications for practice being inexhaustible [in their combinations
and varieties], not to give some specific precepts and details on this method, which being a matter of
preference depends absolutely on the taste and creativity of the artist who composes the design of
W
the work, whatever its nature. But before entering in [to] the details of these different compositions
it is good to say something about cutting up and the general arrangement of wood veneer because
this knowledge is absolutely necessary [for the craftsman] to be able to decide on the form and size of
Plate 283
these compositions according to the species of wood that one wants to use.
After the appropriate wood veneer has been sawn in the manner that I discussed previously,
page 801, one replanes them with a toothing plane as much for giving them an equal thickness as for
giving them a grooved texture, so that they are in a better condition to absorb the glue with which
one uses to adhere the veneer. During this operation, one needs to take great care to keep together
the sheets that are produced, one on top of the other, because their grain and their nuances match
perfectly, it is good, in tailoring them for size and form for different compartments, that similar
pieces be placed side by side, which makes a very beautiful effect, as the grain and generally all the
nuances that meet up in a sheet of veneer are thus doubled by being placed on each side of a join
[book matched], as one can see in Figs. 1, 2 and 6 of plate 284.
On the matter of how the wood veneer is split [sawn], this observation is general: If they are
made diagonally or in layers, and if they form rosettes, as in Fig. 1 of the same plate, it is certain that
the nuances of two sheets, thus split, cannot be perfectly equal as they are produced one on top of
the other. Given that the different concentric layers of the tree are of an unequal diameter, they offer
differing figures according to how they are cut more or less close to the center or the circumference
of the tree.
Page 819
If the woods are split parallel to their grain, it is still the same thing not only with regards to dif-
ferent diameters of the concentric layers of the piece of wood, but also with regards to the different
irregularities that are encountered in the same piece, like neighboring knots and other irregularities
that give various configurations to the fibers of the wood. These may be visible in one place of the
piece but not found one-half thumb farther away. This observation is essential, especially for the
compositions where it is necessary that the same patterns of the wood are repeated in numerous
elements, like the points of diamonds, stars, rosettes, etc.
I know very well that it is not possible to have, in the same piece, more than two sheets that are
perfectly the same; but as these sheets are very thin the difference from the first (beginning at the
heart of the log) up to the 8th or 10th one is very slight, especially when the piece of split [sawn]
wood is of a very large diameter and the grain of the wood formed by the concentric rings are very
close from one to another. If, on the other hand, the piece of sawn wood is found to have very opened
grain, and to be at the same time of a small diameter, as in Fig. 1, one can hardly have but two or
three sheets similar on each side, which means that one is obliged to use the others in the pieces
that require less regularity. See Figs. 3, 6 and 7, which represent the elevations of sheets ab, cd, and
ef of Fig. 1, which are all different one from the other.*
* It is not exactly true that the concentric rings from what-
wood [that is] even the most regular. If, therefore, I have rep-
57
When the wood is as I have just said, and it is necessary to have a number of similar sheets, as in
Fig. 3, or those of Figs. 6 or 7, since one cannot take but four or six identical from both sides and on the
thickness of the tree, one is obliged to take them one from the end of the other. This presents many
difficulties, given that it
is very rare to find a piece
vantages that are found in the cutting of veneer wood [in any
Page 820
58
being on that of il, which, finally, makes nothing at all [meaning there is nothing usable left].
This matter of having sheets perfectly similar among themselves is very good, imposing no diffi-
culty in the execution, and is not subject to uneven changes. If one wishes to make the cuts from the
center of the piece, as indicated by the dotted lines, the form of the leaf Fig. 5 will always be the same,
and will redouble against [or create a mirror image of] line l2, which is only natural given that this
W
last one passes by the center of the tree. There are occasions where this last manner of splitting [sawing] sheets is very good, because it makes elliptical figures very elongated, which makes a very good
effect in certain lopsided compositions where one can use them as appropriate.
Plate 283
The wood of a hard quality and a small diameter is used sometimes as end woods [oyster cut
veneer ovals], not for cutting them as horizontal slices, but by angling their cut a bit, as for example,
kerf AB, Fig. 3 [plate 283], which gives them more solidity, changing their form only a bit, which becomes a bit more elongated on the length AB than the ordinary diameter of the piece [square to the
main axis], but as I just said, this difference is a very minor thing.
When, on the contrary, the obliqueness of this slice is considerable, as that of CD, the figure
changes considerably and round though it be, cut horizontally or almost, becomes an ellipse as I have
represented in Fig. 2, which is taken on the diagonal CD, Fig. 3.
When one makes slices or laminates of wood on end grain or horizontally, which is the same, one
should take care to make them thicker than the ordinary sheets, of one-half the thickness of the last
[he is almost certainly referring to making the sheets thicker by an additional one-half or one-third];
and of only a third when the branches are taken obliquely, which is only natural. The wood, even
though very flexible, is always less strong on end grain than along the grain.
The cutting of wood veneer, such as I just succinctly described, requires great attention and prac-
tice on the part of the worker to make them appropriately, to choose the pieces of a size and an appropriate quality according to different needs, and also requires much experience, which is not acquired
but with time and practice.
Plate 284
Different Ways
to Position Veneer
Woods
Page 821
As to the arrangement of wood veneers, that depends mostly on the taste and the wishes of those
who use them rather than any certain rule. However, one can consider four different manners of
arranging [laying out]
veneer woods; namely, the first and most
simple, which is to put
them on grain that is
parallel or perpendicular [to the grain], as in
Figs. 3 and 4, side GH.
The second is to
put the grain of the wood vertical and horizontal, creating a joint on the diagonal ab, Fig. 5, which
we call miter joints, or, the points of diamonds, and one should take care in gluing woods in this
manner, that their grains be similar as much as possible, and to put the light or dark woods together,
as I have made in this figure.
One glues wood also in the form of a lozenge, as in Fig. 2, side EF, which comes back to the same
thing as the points of diamonds, with the exception that the grain of pieces, which form the lozenge,
are not and cannot even be parallel with the sides of the work, which form voids CD, which one
59
refills with other veneers, whether wood on end or wood on grain, or even the same direction as the
lozenges, which is sometimes practiced, especially when
this last one is separated by a flat band or stringwork, as in
this figure.
their grain diagonally, so their perpendicular and horizontal joints cut the grain obliquely, as in Fig. 6, side M & N.
When the wood is thus positioned, one puts the middle of
the sheet on the diagonal bc, and one takes what remains
Plate 284
of
the cutting of
wood veneers.
In general,
whatever manner is used to lay out wood veneers one should take great
care that they be nuanced equally on each opposing side,
and that these nuances correspond with each other as
much as possible, which will add much to the perfection
of the work. One should also have the same attention for
the edge bands, OO, PP, QQ, RR, Figs. 3, 4, 5 and 6, whether these last be placed on grain, as in Fig. 3, or at an angle or mitered, as in Figs. 4 and 5, or finally on end
grain, as in Fig. 6, which depends on the taste of the worker, or better said, on the manner in which
the large sheets of veneer are placed, the grain of which the edge bands should be opposite.
The edge bands are ordinarily separated from the panels by simple strings of holly or willow (or
Page 822
60
other species of light-colored wood along the grain), or better, with flat bands SS or seder bands
[edge bands with cross grain in the center flanked by strings on grain] as the cabinetmakers say, with
two strings on each side, as I have shown in Figs. 2, 3, 4, 5 and 6 of this plate.
The edge bands are ordinarily face grain, at least if they are not of a considerable size, like 46
lines, because although one puts them sometimes on end grain, especially when they are of a nuW
anced grain and they form diverse contours; but the manner most typical is to make them from uniform wood color and place the wood on grain as I have said before.
Plate 284
The placement of wood on end grain is a bit the same as that of veneer in heart, Fig. 1, because
whatever manner in which one arranges them they cannot form anything but rosettes, or better said,
many concentric circles one to the other, or a totally different circle, for which one orients to the
center the seams of the pieces that make the composition, observing always to mirror the patterns,
whether regular or irregular, with one facing the other.
This should be the same thing for the position of knots or gnarls [burl], of whatever species they
be if the pieces are not large enough to make an entire panel; then, one sees no regularity, at least,
one never finds in the same work [a way] to achieve two equal panels; so that it is necessary to fill in
symmetrically, placing the various irregularities of the veneers one opposite the other.
There you have it, in general, all that one can say about the placement of wood veneers: the taste,
the experience and the various occasions for using them. Those are the surest rules that one can
consult in order to understand well this part of woodworking, which as I have already said, depends
greatly on the taste of the worker, current style, and use received in each century.
The different divisions of pattern in veneer work are of two types, as I said above, namely, large ones
and small ones. Both are made by the symmetrical arrangement of joints and grain lines in the veneer,
Sections Appropriate
or by various figures that one gives to friezes, frames or flat banding that one adds to these last ones,
that is to say, to the compositions formed simply by the joints and grain lines, which in simple works
Work
Page 823
Figure 1 represents a composition of a squared form, the most simple, surrounded by a uniform
frieze and flat banding, which separates this last with the panel in the middle. The wood of this panel
is positioned in two different manners, with joints diagonal [on the right side of the panel] and in opposition with the direction of the friezes [on the left side of the panel], which are placed whether on
grain, or on end grain, or even diagonally, all according to the interior placement of the panel.
which the infill is made diagonally, and in two parts on its width. The wood of the interior of this
figure is positioned in two manners: namely, diagonally and with squared joints, as those of A & B,
and with squared joints diagonally, as those of C, D, E, F, of a manner that the grain line is found in
the middle of each piece, which works rather well. The flat banding that surrounds the friezes of this
figure, and which separates them from the middle panel, are all accompanied by stringwork, which I
could not mark here because of the small size of these elements, which would have been required [to
be] as large as the actual execution for the stringwork to be perceptible [in this figure].
Figure 3 represents a composition squared without friezes, with squared ears in the four corners,
62
which produces 12 angles [the illustration shows more than 12], which wrap around the joints that are
angled in the center of the piece, and according to which one adjusts the veneer wood in the direction
that one judges appropriate, observing nevertheless to do this symmetrically, as I have done in this
figure where the veneer wood is placed in three different manners, so that one can choose which one
is appropriate.
the angles from which start the joints that separate the veneer,
Plate 285
rosette with eight sides, surrounded by a flat band and stringwork. This composition
makes a very beautiful effect and requires much care
to produce all the perfection
with which it may be accomplished.
The compositions that I just presented here, while small in number, give a general idea of the
manner of placing the seams and the grain lines in all possible cases, which constitutes the first part
of composition. It is necessary, however, to observe that when the work to be veneered is more of a
rectangular form (instead of being square, as in the figures of this Plate), it is always necessary that
the joints or the wood grain adapt to the angles in the work, without being concerned whether it is
squared or not. **
Page 824
It is hardly possible to give the precepts touching the second type of composition, that is to say, those
that are formed by the different shapes that one gives to friezes and to flat bands that surround the panels,
given that they depend on the general form of the work and the elaboration which one wants to give, and
still more than that, of the creativity of the artist. I will content myself here therefore to give some examples of compositions, both the interior of the panels and the friezes that surround them. I will emphasize those patterns that are the most in use, based on which one can invent others, or make use of those
here, as one judges appropriate, the choice of these different compositions being purely a matter of taste.
** This observation, however simple, is very essential, given
less elongated.
63
In making the description [above] of the sections of the first type of veneer work, I supposed they
are of a uniform wood, that is to say, of the same color and quality; to the contrary, those of which I
am going to speak now are made not only of a wood of a color and quality that is opposite, but also in
which one makes the choice of nuance, to make mismatched sections less apparent.
Figure 1 represents the type of section the most simple of all, and is named en chiquier checkW
ered. It is composed of many squares of wood of different colors, placed alternately side by side,
so that a white square is found surrounded by four black squares, and a black square is surrounded
by four white squares. This sort of composition, although very simple, requires much attention so
that all the joins of each piece align and meet perfectly. It is also necessary to observe that when one
Plate 286
makes this sort of composition that the number of squares that comprise them be unequal so that
Different Sorts of
one finds four of the same color on angle, which could not be if the number of squares were equal,
Sections Appropriate
as one can see in this figure [this translated text is correct, though confusing]. If it happens that the
piece that one would cover with this sort of composition were rectangular, little different from a
square, one would divide the two sides of the piece in an equal number of parts, and always in odd
number, in order to have the same number of squares on each side; and that these same squares for
filling be rectangular, matching with that of each piece to
be covered.
which I speak, it is unimportant provided that the infill squares be of a color distinct from each other
and of all the friezes and flat bands that surround them. If it happens that one is obliged to make the
friezes or flat bands of the same color as that of one of the two types of squares, it will be necessary
to separate them with stringwork extended to the outside edge of the work, as I have shown in the
different diagrams of this plate.
Figure 2 represents another section composed, as with the preceding, of squares of different col-
ors placed alternatively side by side. This second composition differs from the first in that the squares
of which it is composed are placed on an angle, which gives them the name lozenge pattern. This
composition works well enough, and is convenient that one can put the squares of which it is composed in an even or odd number, according to how one judges appropriate, observing all the while
that a quarter of a square be found in an angle, so that all the squares of the same type touch their
points on the flat band or the stringwork of the outer edge, and that the others be all cut in the middle
of their length, as I have shown in this figure.
When the space to be covered is found to be a little longer in one direction than the other, one al-
ways divides the squares of infill into an equal number on each side, which then will change the form,
which, formerly squared, now becomes a lozenge, whether on the length or on the width; and one
will always make the divisions with great care, whether of squares or of lozenges. The reason for the
divisions is the size of the space to be covered, so that one never finds an irregular cut [presumably
meaning that the divisions should result in whole-number parts, not a fraction. You cannot have twothirds of a half of a lozenge when everything else is a full half; does not look good], which is a very
Page 825
65
great defect, and is absolutely to be avoided in all sorts of divisions regardless of the space, as I have
shown in the different divisions represented in this plate.
The divisions of which I just spoke are normally employed in sections of an average size, but
when the sections to be covered become too large one makes no use of these divisions but rather
surrounds them with flat bands and stringwork, which makes types of frames as in Fig. 3, and which
Page 826
gives the option of making for larger squares or lozenges, as much as one judges appropriate.
When one makes these sorts of compositions, their division is made from the inside of flat bands,
which serve as frames, without having any regard to the size of the stringwork, so that the angle of
this last touches the node of the flat band that surrounded the work, as I have observed in this figure.
As these compositions are large, it is necessary to take care that they are very regular, whether they
make a perfectly squared or lozenge form, which is equal, provided that one finds at the four corners
two half-squares, as in this figure, or even only a quarter of a square supposing that the division ends
at line AB, same figure.
The interior squares of these compositions may be filled in different manners, as I have shown
here. Sometimes one inlays rosettes or other ornaments, which makes a very pretty effect, as I will
explain later.
Figure 4 represents a composition with dice or cubes, placed on a background of whatever color;
these dice or cubes are hexagons, placed side by side, in a manner such that their points touch each
other, as you can see in this figure.
Each of these hexagons, or figures with six sides, is composed of three lozenges of any colors as-
sembled together to make the dice or cubes appear in relief. Lozenge C (which is the daylight side)
is an example of the shape in question and is made in rosewood. Lozenge D, which is the top of the
cube, is of grey or yellow wood. Lozenge E, which is the shade side, is of violet wood. The remaining space [unmarked but primarily horizontal] is of some other wood that one judges appropriate,
provided that it differs in the color of wood that forms the cubes. The cubes should not only differ in
color from that of the bottom, but also each lozenge comprising the cube should all be different from
each other. One accomplishes this by choosing pieces darker in color from one side to the other, or
even by passing them over hot sand, as I will teach later.
Figure 5 represents another section, which does not differ from that of which I just spoke, except
that it does not have any remaining space or background like the last one. To the contrary, all the dice
or cubes fit one inside the other without leaving any void space, which works quite well. However, it
is good to observe when making this last type of section, to make a space or background between the
cubes on top and on the bottom, as I have shown in this figure, which works much better than to see
the ends of cubes cut up, as one does ordinarily, and which I have indicated by line FG.
In general, whether the sections of which I am speaking are with a background as in Fig. 4, or
without a background, as in Fig. 5, it is necessary to take great care when making the section that a
whole number of cubes is found on the length, and that the uppermost end of these same cubes reach
the banding or stringwork that surrounds them, as I have shown here. This is very easy to do since it
is only necessary to adjust the proportions of the cubes according to the need, it not being absolutely
necessary that the hexagon of the cubes be perfectly regular. Whatever way it can be done is the better way, and is so much easier to do when the three lozenges that compose the hexagon are of a similar
shape, which does not ordinarily happen when the hexagon is of an irregular shape.
If one does not wish to make dice or projecting cubes, as in Fig. 5, one could make sections of
Plate 286
66
cubes to fill the lozenges in a unified wood, which does not work badly when the joints are well made,
as one can see in this figure. [This is in fact my favorite manner of preparing a composition such as
this. I find the subtlety much more to my taste, especially when using a wood with a fine grain pattern
with a noticeable difference from early wood to late wood, such as bald cypress on the radial plane.]
Page 827
Figure 6 represents a section with mixed stars, which is a section that is very complex in appearW
Plate 286
These sorts of sections can be made with a projecting appearance, or be filled with segments of
the same wood, which is equal for the form and disposition of the joinery, which is always given by
the parallel lines, horizontal and perpendicular, and [rather than being comprised of lozenges] by
equilateral triangles, of which the tops are opposite one another. Inspecting this illustration alone is
by itself better than all the explanations that one can give.
These sorts of sections can also be made entirely of a uniform wood, the stars as well as the back-
ground, or even the stars of one color more or less darker than the background, which is made in the
same technique.
Page 828
There are still many more sections appropriate for application to cabinetry panels, the details of
which I will not go into here beyond the small number of examples that I just gave in this panel, the
explanation that I just made being sufficient to aid in the regular tracing of such sections as one would
like. It is important to always make the sections for the space and not the space for the sections, as
67
happens very often. It is also critical to avoid above all to infill the panels with sections that are
either too large or too small, making it necessary to mutilate their frames or flat bandings on the edges
of the panel. This happens only by the negligence of the workers, who once they have made their design for a section and the appropriate tools have been adjusted for the different pieces, do not want to
change them. This might be due to laziness or inability, or, which is more accurate, by the impossible
W
situation they are in when the merchants pay only half the amount necessary for it to be well made.
Cabinetry panels are ordinarily surrounded by friezes, whether of the same wood oriented in
different directions, or in compositions, which is the same thing; in one or the other case, one sometimes puts banding
of
different
col-
Plate 287
Sections Appropriate
ors, surrounded by
stringwork, as I have
Whether the bandings are simple, as in Figs. 2 and 3, or they are doubled, as in Figure 1, it is always
necessary that they be surrounded by stringwork, which separates them from the rest of the work,
which is a general rule in all cases. This stringwork is ordinarily white; however, one can make them
of other colors, which is not important, provided that their color makes a break with the woods that
separate them, and that they be of a wood very flexible [pliable] and along the grain, so as to be able to
work them very easily, as I will teach in a moment. See Fig. 4, which represents a banding with its two
strings, which are glued there, showing as much of the face as the side.
When the colors of the frieze are very different from that of the panels, it happens that the
The second way to fill in the interior of friezes is much more complicated than the first, because
Page 828
69
the space of these last ones being ordinarily limited, the parts that compose the composition of which
they are filled in can only be very small. This makes their perfect execution very difficult, especially
since one normally puts Greek keys or broken bands there, which are comprised of a large number of
different small pieces, as one can see in Figs. 8, 9, 10, 11 and 12.
The Greek keys or broken bands, represented in these different figures, are more or less comW
posed according to the size of the friezes, and are traced in the same manner, as I am going to explain.
When one wishes to trace this sort of ornament, it is necessary first of all, after having traced the
middle of the frieze, as line ab, Fig. 8, to divide the width of the frieze in as many equal parts as the
composition requires, seven being the number in this figure [the filled being equal to the empty; in
Plate 287
other words, there is balance between the positive and negative space]. This being done, one traces
as many parallel lines as there are points in the given division; then one traces these same spaces or
divisions perpendicularly, observing that one finds one in the middle of the work, as in this figure;
after which one determines the shape of the broken bands, to which one makes as many turns as are
necessary to fill in the length of the frieze, observing that at the end, one has made an entire revolution, or at least a happy ending, without having seemed to have been cut, as I had to do in Fig. 8, side
G; in that of Fig. 9, side H; that of Fig. 10, side I, and that of Fig. 11, side L.
One inconvenient observation is that if the width of the frieze is bordered, its length cannot be
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
The patterns and planing templates would have been created much in the same manner as die-makers
used to do before CNC (computer numerical control) machines. The investment in the creation of such tools
would have encouraged their use as widely as possible within the scope of the ateliers work.
to make them.
70
made until after dividing this same width in as many pieces as one judges appropriate, as one could
see above. If, on the contrary, it is the length of the frieze that is given, as happens ordinarily, one
cannot determine the width until after having made the choice of composition that one wants to use,
and of the number of turns that half of the length of the frieze could contain, which will give a number
of whichever parts, on which one divides the middle of the length of the frieze, observing still to put
W
one of these divisions in the middle of the length. The division of the length of the frieze once made, one will easily
have the width, since the division is already made, repeated
Plate 287
Page 830
squared sections, separated one from the others, as is found sometimes, especially in the corners of friezes, where they can take the place of rosettes or
other ornaments.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
At this point, in the absence of further direction from Roubo, I judged the most expedient means of pro-
gressing was to replicate the border pattern layout on the background veneer, which was already adhered to
the substrate. Repeating the same technique as excavating the channels in the pattern block, voids for each
component could be cut into the background and the border pattern pieces set in place with hot hide glue.
Roubo does intimate that the banding and the stringing were two separate elements that required distinct
steps for execution. In this exercise, I decided to pursue the technique of tarsia a toppo (employed for almost
two centuries by the time of Roubo) whereby a loaf or block of inlay is pre-assembled, and from which
strips are then cleaved to serve as the banded inlay.
For the relatively simple running border illustrated, what follows are the cutting/planing patterns that
were needed for these elements, and the process I used for replicating them. As to the exact size and length
of each of the elements of the pattern, I noticed that in each case the spacing of the pattern is based on the
width of the banding. Moreover, I think that the actual layout of the running border was executed only after
takes.
71
In general, all parts that compose these compositions of which I am speaking should be fitted by
miters in all their angles; and one should take great care that they align well one with the other in all
directions, the tightness of their fit making the principal beauty of these sorts of compositions, and
of all others in general.
Friezes are filled also with circular compositions, such as round interlaced ribbons, or ovals, simW
ple or double, which are always surrounded by stringwork and are inlaid on a solid background,
whether of plain wood or in a harmonious manner. See Figs. 1 and 2.
Plate 288
Circular Sections
for Friezes and
Sometimes in-
Panels
relays
of
I am not going to
*** There are occasions where the interlacings are filled with
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
the banding was fabricated, and the primary layout tool (beyond a square and miter square) was the banding
itself, or a template mimicking the banding precisely. The following photo series explains the process far
better than words can.
It was only after this pattern had been established that the individual elements for the pattern could be
Using mahogany, although any durable hardwood will suffice, I made a block and cut channels into it to
securely hold the banded inlay while shooting its ends. The channels must be the same width as the inlay
banding so that the banding will not wiggle when being planed. Because this simple pattern involves only
a small number of elements, and each of these has a 90 or 45 terminus, the starting point is to cut channels
across the block at perfect 90 and 45 orientations. The block was now ready to allow the cutting and trimming of long pieces of banding.
Fig. 3.7 Once the edges have been cut for the chan-
73
extend myself further on these sorts of compositions given that however different as to the form,
those of which I just spoke above are subject to the same care when tracing them. That is to say, it is
necessary that they be of a size relative to the place they are going to occupy, so that they are not being
cut by the frieze-frame.
There are occasions where one cannot put a frieze around the cabinetry panels, but simply an
W
open background in a contoured shape, which is then separated from the ground of the panel by a
single string, as is represented in Fig. 5.
Sometimes this background passes straight, as in Fig. 6, and the panel is surrounded, or better
Plate 288
said, ornamented by a flat band in a contoured shape, which makes a nice effect especially when the
contours are soft and graceful.
When the cabinetry panels become very large, one sometimes puts in the middle a contoured
band, or even one makes a little panel, separating the largest portion by a single stringwork, as in Figs.
5 and 6.
Or even instead of a flat band, one uses a type of ribbon, as in Fig. 7, which works very well when
In general, for circular compositions to work well it is necessary that their contours be soft, flow-
ing, and without any type of unevenness. That is why it is good, after seeing them traced by hand
as perfectly as possible, to lay them out with a compass to make them the most symmetrical, so that
the contour of one side is identical to the opposite side. When one traces the arc with a compass,
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
On examining the nature of this border pattern, there are 11 elements in each corner. (See drawing.) Fortu-
nately, the artisans of the 18th century were just as interested in standardization and efficiency as we are! Of
these 11 elements, Nos. 1 and 11 are connectors between the panels corners and must be left until the final assembly step in order to ensure that their longitudinal dimensions are perfect. The other nine elements can be
cut, end trimmed and essentially mass produced quickly and easily because the pattern reveals that elements
Nos. 2 and 6 are identical, as are elements Nos. 5, 8 and 9, and elements Nos. 4 and 7. With that determination,
channels must be created to fit each identical element group, along with individual channels for elements Nos.
3 and 10.
The easiest method for cutting the channels is to follow a mitering guide for cutting the first shoulder with
a fine backsaw or knife, laying a piece of the banding against the cut line to establish the exact width, then
moving the miter gauge to the location of the second shoulder, and then cutting the second edge also with the
fine backsaw or knife. The channel can be excavated with chisels, router planes or rebating planes. I cut all the
1
3
10
6
7
11
74
it is necessary to take great care that the centers of the smallest portions of circles be placed on the
Page 831
largest radius [I believe what Roubo is saying is that when two opposing curves intersect, such as
with a serpentine form, when laying out the inlays as in
Figs. 6, 7 and 8 that it is critical to match up the inner arc
of one curve with the outer arc of the connecting curve.
W
And that the line between the two compass points establishing the two curves is to be perfectly perpendicular to
the two curves at their connection. This is best shown in
Plate 288
Fig. 6], and that the opposing centers be a straight line with
their points for meeting or attaching (which is the same
thing), just as I have observed in Figs. 5 & 6, on which I
have marked the different centers of curves so that their
radii, which I have indicated by punctuated [dotted] lines,
so that for as little as one wishes to call attention to various
operations traced on these figures, one will be well situated to trace with the compass all the curves
required, without having need of an explanation more extended than this one.
There you have just a little bit of all that it is possible to say about the theory of compositions,
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
channels across the full width of the block, and then back-filled them to create the precise lengths required for
each small element. Using additional scrap pieces of banding as the back-fill makes the most sense, and with a
dab of hot glue, is nearly instantaneous.
In practicing with the planing template block, I found it to be exceedingly useful for trimming the ends
of banding where the elements are all long-grained. However, if the center element in particular is crossgrained, as was often the case, a slightly different technique served me better. Building off of the techniques
Roubo presented, I created a cutting (or slicing) template that allowed for accurate and rapid fitting of pieces
using the shoulder knife or any other sharp edge tool.
The practice Roubo described and that I replicated here provides a means to produce reasonably complex
borders in a short time, once the template blocks have been created. The first set of corner banding took me
about three hours to complete, including layout, fabricating the block and photography (I am a slow photographer). The second corner took me about 10 minutes, about four or five minutes for the cutting and fitting
of the banding elements, with the layout and excavation of the background taking another five or 10 minutes.
With well-made templates and razor-sharp tools, I expect that the skilled journeymen of Roubos day could
produce even the most intricate interlocking patterns very quickly.
75
about which I am not going to extend myself any further given that this matter is almost inexhaustible, and that the little that I have said, however general, can be applied to all possible cases. All that
I can recommend to workers is that they exercise great exactness in tracing their compositions in
consideration of the space, as I said above, the importance of which I cannot overstate. They should
also make the design of their work [in full size], and trace it large, and even to color the principal
W
When the work is to be ornamented with bronze mounts, it is necessary to include them on the
plan, while the surface is straight, or even to make the models in relief in order to design the ornaments above, following the contours or the projections of the work. These precautions are long and
Plate 288
costly, truthfully, but also the only way one can make the work in question with all the perfection
possible. Furthermore, as I have already said and I repeat again, this type of woodworking is not
meant for everyone. I believe that making cabinets as well as they can be, and selling them for what
they are worth when well made, will be a sure way of restricting this branch of current luxury that
could do no wrong to the craftsman. This will not be any harm to the workers, since by cutting back
on part of their work they will augment the price of what remains to be made, which becomes better
for them. The worker, once assured of being well paid, will apply himself diligently to his work and
will become in a short time more knowledgeable and creative, so that the Art of Cabinetry, which
seems now reduced to a type of routine more or less content, will be reborn, that is to say, from its
ashes, and one will see works capable of not only equaling, but even of surpassing, those of preceding
centuries.
The manner of cutting & adjusting the pieces so they are straight,
and the proper Tools
After having summarily discussed the different compositions appropriate for veneering (which are
for this type of woodworking, the profiles and contours of other parts for this art, already described),
it is now time to pass from theory to practice in showing the manner to execute these same compositions whether composed of straight parts or circular parts. I will include the detail of many tools
needed for the execution of this part of veneering, as I just announced before on page 811.
The compositions of marquetry are, as one could see above, composed of an infinite variety of
wood veneer pieces of different forms and sizes (however similar, as the workers say, each one has its
own individuality), that is to say, that in a composition comprised of 100 different pieces, there are,
for example, 10 of one form, six of another, 20 of another, etc., which should all be perfectly equal
among them, whether in size, direction of grain, or for the nuances in color, etc.
Before entering into this detail, I am going to give the manner of preparing the stringwork and
their thickness, so that I do not repeat [myself] while I address the method of using them.
Stringwork, which one makes use of today is of two types: namely, those of willow (or wick-
er), which is the lesser of the two, and which basketmakers sell already prepared; and those of holly
wood, which the cabinetmakers prepare themselves in the following manner.
One begins by choosing a sheet of holly, split [sawn] into ordinary wood veneer, of which the
grain is as straight as possible and which one dresses on the edge by following the grain of the wood.
This being done, one splits it with a trusquin or splitting gauge by means of a jig named the wood
splitter. This jig possesses a trough, AB, represented in Fig. 1, [which] is nothing other than the end
Page 832
Plate 289
The Way to Trim
Straight Pieces and the
Way to Inlay Them
76
This excavated area is not made down the whole length of the jig, but one leaves a shoulder [to
act as a stop] across the upper end of side A, against which can be pressed [propped or wedged] the
W
Page 833
end of the sheet. The width of the fence at the side of the excavated area, against which one applies
the sheet, is arbitrary; it is sufficient that it be well trued and exactly parallel from the inside to the
outside of the fence; because otherwise the stringwork will be of an unequal width, which is only
Plate 289
natural, since the cutting gauge D has a fixed beam [and] cannot be anything but parallel, which, I
believe, needs no demonstration.
The gauge that one uses to split the veneer sheets is hardly different in form from other marking
gauges with long points, if only that the point of this one here should be of a good tempered steel
of average hardness, and sharpened into a cats tongue [splitting blade] (as the workers say), so
that it enters more easily through the thickness of the veneer sheet, which one splits twice; that is
to say, that after having given a stroke on one side, one flips it over lengthwise to repeat the cut on
the other side.
As to the manner of holding the splitting gauge and the sheet on the jig bed, look at Fig. 1, of
which an inspection is better than a long explanation, which besides, seems to me to be useless here.
When the stringwork is thus split, one makes them of equal thickness with a tool called the
string pull. This tool, represented in Figs. 3, 4 and 5, is comprised of a principal piece EF [a block
of wood] containing a blade; in the middle of the thickness and the width of the block is placed an
iron piece L, which is parallel to the vertical faces. Above the body of the string pull is placed a
handle or lever, which serves to press the stringwork against the iron, as I will describe later. This
handle or lever is fixed at one of the ends at point a to a rectangular pin bc, to which it is attached
via a knuckle joint and a riveted pin to form a hinged pivoting arm; and this same rectangular pin is
set into a deep mortise in the body of the tool.
At the other end of the lever, which is rounded into a handle, is assembled a little tenon a, which
enters around 68 lines in thickness above the shaft of the wood-pull, to hold the handle in place
when one makes use of the tool.
The space between the iron and the wood-pull should be open with an equal thickness the whole
width of the iron, less about one line on each side, which is necessary for it [the iron] to be buried, so
that it does not make an obstruction [a space where the workpiece can be wedged next to the blade],
or shavings can accumulate and get stuck.
The surfaces in front and behind the blade should present a surface a little relieved away from the
blade over the width of the tool, and be trimmed with an iron plate, so that the continuous shavings of
the wood which one passes above do not remain in place like a temporary shim and cause the iron of
the tool to take more wood in those places than in others, thus creating unevenness in the stringwork,
which one should take great care to avoid.
It is necessary to have the same attention to the underside of the lever, which is also good to finish
with an iron plate and not in copper, because copper becomes hot and will stick to the stringwork,
and could blacken it.
The trimmer/string pull is placed sometimes on the bench, where one secures it with one or
two clamps. But ordinarily, one puts it in the twin-screw face vise, in which one secures it by means
of two shafts, I, M, which are assembled and protrude below the trimmer/string pull. These shafts
Plate 28 9 . The Way to Trim Straight Pieces and the Way to Inlay Them
78
are not only necessary for holding the tool in place, but also for procuring room for the projecting
iron, as one can see in Fig. 4, which represents a side view of the trimmer/
string pull with one part of the bench and the brace vise. When the trimmer is
placed on the bench, shafts I, M are removed and one puts the iron there so it
does not protrude from the thickness of the tool below, which is inconvenient,
W
especially when one wishes to adjust the iron. That is why it is better to secure
it with shafts, such as I represent here. [The shafts help elevate the body of the
trimmer away from the bench so that each time you need to adjust the
Plate 289
iron you do not need to remove the trimmer from the vise.]
Page 834
I said above that pin bc, Fig. 3, to which is attached lever GH, is
of parquetry. It is worth remembering that mastering the vocabulary of certainty unleashes the creativity of risk.
Many of these tools and accessories for making parquetry are familiar, at least in
concept, to todays most accomplished artisans. But when I think back to points in
my own craft history, earlier encounters with them would have yielded a quizzical re-
sponse. In truth, there are a couple of them that still make me wonder how they actually
width.
improved productivity.
79
could stop it with a pressure screw or with a key, such as the shaft of a marking gauge. This would
provide the means to raise or to lower the lever according to the thickness that one would like to make
the stringwork, or any other piece that one judges appropriate to set the thickness using this tool.
In making pin bc thus movable, one would need to extend the other end of the lever de, which
one cannot hold in place given that it is necessary that it remains movable all the time, but which one
W
will secure the fall with a screw stop or a key placed below; and one will take care that this pin de, be
set in such a manner that it follows an axis taken from the center of a of the lever, so that one can raise
it as much as it is necessary.
When one wants to use a trimmer/string pull, one holds the handle of the lever in the left hand;
Plate 289
then with the right hand, one takes a split string, which one passes
between the blade and the lever, pulling the string toward oneself, and
one pushes on the lever as much as is necessary for the iron of the tool
to bite into the string, which one returns end over end while the first
end has had [its] thickness taken, which is easy to know when the iron
shaves no more, and the lever bears on the shaft of the tool. See Fig. 5.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Here is a brief recitation of some of the tools and templates employed in the workmanlike production of
parquetry.
Veneer Shooting Board and Plane. The first and in some cases the most critical step for composing
parquetry is the truing of the veneer stock. After sawing and flattening it with toothing planes, the parquetry
stock was prepped and trued to be cut precisely and uniformly in subsequent steps.
A modified shooting board provides the basis for this task, as illustrated in Plate 289, Figure 2. For mine, I
simply prepared two pieces of 12mm-thick Baltic birch plywood of equal length, with one piece slightly wider
than the other.
Gluing them together with both ends and one edge aligned yields a near-perfect tool. The plywood is about
as stable as anything you can find, and the offset between the two pieces along one edge provides a shelf for the
miter plane used to trim the edge of the veneer.
for tool making is that often the very same tools for
80
great care to choose the grain of the wood to be straight so that the iron does not make a splinter, or at
least does not tear the stringing while it is being made; one can partly remedy this by placing the iron
upright [to scrape the wood]; [this] does not eliminate the choice of the wood grain since it [choosing
straight wood] requires only a little bit of attention on the part of the worker.
Long-grain flat banding and other thin parts of veneer are split by the gauge, as with the stringW
work; and one uses a fence of wood [a wooden edge, scarfing edge, notched edge],
to set the width, represented in Fig. 2, in which is made a rebate that is equal to
the width of the desired banding. Which means that for a different width, it is
Plate 289
necessary to enlarge or reduce the rebate, or one needs [to make] all the different
widths one can use, which becomes a little awkward, especially when one has a
great number of pieces of different widths.
One could remedy this inconvenience by not making the rebate on the width
of the board but by placing on top a movable fence, a little like a wood engravers
Page 835
[or draughtsmans] parallel, which, consequently, can only move parallel, and
would be kept in place by means of a pressure screw. Because this tool could
become a little complicated and costly, workers would not make much use of it,
always preferring simple and inexpensive tools.
Of whatever manner one makes these sorts of fences of wood, it is always necessary that they be
of a hard and stable wood, of which the grain is straight, and even a little inclined in the direction the
tool is pushing, to be less subject to tearing the wood since when the pieces are of the correct width,
the iron of the plane or of the miter plane bears against the board, which one must avoid shaving,
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Because modern commercial veneer is extremely thin (and because it is sliced as opposed to sawn), there
is a tendency for the sheet to pucker when shooting its edge, unless a long caul is placed on top of the workpiece. Virtually any straight, stiff material will suffice, whether it be a piece of flat metal or a wooden bar held
in place by pressing it on top of the veneer with your off hand while working the edge with the plane. Sawn
veneer tends to be much more forgiving as it is both thicker and stiffer than peeled veneer, so it might not always need pressure from above.
For trimming the edges of the veneer any well-functioning wood plane with a fine throat opening should
suffice. My favorite for this was a bubinga-bodied wood plane kit, but the sides began to get chewed up through
use. To alleviate this problem and add some much desired heft to the plane, I laminated 1/8"-thick brass plates
to either side.
My first step was to cut a purely decorative ogee on each end, and then rough cut some brass plate to fit the
sides of the plane body. I epoxied the sides in place, then filed and trimmed the sides to match the body. All told
it took me about three hours during two nights.
In use, the plane rides the shelf perfectly, cleaning the edge of a veneer sheet with a stroke or two. For
cutting huge quantities of stock precisely parallel, such as would be necessary when preparing the strip stock
for creating large expanses of cubic hexagon parquetry, a second shooting board with a fixed fence would be
used after the initial ripping was accomplished with a slitting table. By this practice a nearly endless supply of
perfectly dimensioned stock could be prepared.
Special shooting templates were no doubt used when truing numerous strips used for cutting parquetry.
In that case, a shooting board with a fixed fence would guarantee that the pieces of veneer stock would all be
exactly the same width, a critical ingredient for producing tight, intricate parquetry.
81
because one will diminish its width, whether in whole or in part. See Fig. 2, which represents a fence
of wood for trimming pieces that are wider, with the manner of making use of it. The side view of
this same trimmer, and that of piece NO, represented by side P.
If it happens that one has long pieces to fit for which the width needs to be tapered from one
end to the other, one uses the same tools for splitting them and for setting the width, observing to
W
make the shoulder of the rebate for splitting and the shoulder of that to be made wider, tapering
from one end to the other to the same degree the workpiece must have.
When the unevenness of the pieces is more considerable, it is necessary to have two fences of
wood to set the desired width of the wood, of which the grain should be in opposite direction one
Plate 289
from the other, so as to be able to readjust the pieces by following the grain line, to avoid tearing so the
joint will always be sharp and fine.
When the pieces being trimmed are wider, one cuts them to a convenient length, whether square-
ly or mitered or whatever other angle is necessary, which is made with an ordinary saw, after having traced them with calipers or shapes appropriate to each
of them; then one fits them on their fences of wood, according to the taper that
they should have.
The rebates in Figs. 6, 7, 8 and 9 are of wood pieces around 3 thumbs thick-
ness, on the side of which is placed a mortise where one enters the end of a
bench dog, which holds them on the bench, as represented in Fig. 6.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Cutting Stringing or Long Stock. Using the off-side of the shooting board I fabricated a near-perfect
replica of the slitting table Roubo illustrates in Plate 289, Figure 1. A strip of 1/8"-thick aircraft-grade plywood
provided a simple edge fence, although any clear, straight wood could be used.
The rationale behind the slitting board setup is that the block of the slitting gauge will ride against the
fence edge of the board rather than against the edge of the veneer, resulting in cleaner cuts.
Any marking gauge with a cutting knife will work, but I prefer to have a substantially wider head for this
type of work. Recycling a vintage orphaned head from a panel gauge, I fabricated a number of new beams to
use with the block, so that it could be used for both slitting and excavating.
In the former I fit a simple heavy-bodied cutting knife blade into the beam and used it to slice off numerous
Additional cutters of nearly any configuration could be used with the head, a beam and the shooting board.
for me.
edge.
stock.
82
The top of these rebates is trimmed to about 2 lines of depth (which is necessary for containing
the pieces one wants to fit), and [there are] various kinds, according to the different shapes that the
pieces should have, whether the form is squared, or simply at a right angle, or whatever taper [gradient]. When the pieces make an oblong form, after having made them the right width, as I have said
above, one puts them in rebate R, Fig. 7, of which the width should be equal to that of the piece whose
W
Page 836
end one wishes to trim, which one adjusts then with a plane or a miter-angled plane, of which the
iron is placed perpendicularly, taking care to press the piece well against the side of the trimmer [this
could be called a trimming block] opposite the plane and to hardly take any wood at this time, so as to
Plate 289
avoid splinters, which are very easy to make with veneer end grain no matter what precautions one
takes.
When the pieces are of a shape exactly squared, one makes them a little larger than necessary, and
makes their correct width by bundles (which is much preferable to doing it one by one), and when
they are cut lengthwise as close as possible, one fits them in the rebate S or T, according to their size,
which are trimmed to be able to contain the piece before they are fitted [to the piece].
In trimming these sorts of pieces, [with] one of their sides along the grain being perfectly dressed,
one begins by [next] finishing the two end-grain sides by pushing against the side of [the] string
[banding] that has not had any width removed, and by which one finishes the last side along the grain
to even up the splinters that were made in trimming the end-grain sides.
When the pieces are cut by miter, or any other angle, one trims them first from one end of the
rebate U, Fig. 8, of which the width is always equal to that of the piece; then one cuts their [rough]
length and one trims them in [the] other rebate X, where the length and the shape are exactly similar
to those of the piece, whether it be cut parallel in the slant, as in rebate X, or squarely on end, as in Fig.
8g, or in the opposite direction, as from gh, which is the same.
If the grain of the piece to be trimmed is found to be in the opposite direction of the grain line
of the rebate, as from
I to L, Fig. 8, side Y, it
will be necessary then
to make a rebate in re-
verse, as I have shown here, so that the wood, of which the direction is found from I to L, is always cut
along the grain.
As much as one has pieces of the same width, and of a different shape and length, it is necessary
to make rebates in the trimming block, as I have represented in Fig. 9; and one should not make use
of these, except after having perfected these rebates one after the other, and make sure that all the
pieces that have been trimmed are well-fitted in the shape and size necessary for the composition to
be perfectly well made.
Whatever shape the pieces to be fitted may be, one always uses
the same method, at least for the straight parts, whether their faces
be parallel, as those that compose a hexagon, Fig. 14, or whether
they make an irregular form, like those that compose Fig. 17.
When one has compositions where the flat bands are cut in little lengths, as in the case of
broken bands or others, one glues the stringwork to both sides, as
in Fig. 13, after which one cuts them lengthwise, as usual. As to the
stringwork separated into flat bands that are very thin, one does
not cut them with a saw, and one does not trim them with a plane in a wood trimmer; but one
83
cuts them and adjusts with a chisel, or even with a little knife of a size represented in Fig. 16.
Page 837
This piece of wood is notched on two sides of its thickness, about 34 lines, which is more than
is necessary for pressing against the piece of veneer that one wishes to trim,
which one puts against the heel of the wood to be trimmed of which the ends
are trimmed in shape of a pied de biche [claw foot] to be able to hold it in place
while in use.
The heels of the wood to be trimmed are cut squarely on one side, and on
a miter on the other, as in A, B, C, D; and one should observe that the mitered
rebates are made in opposite directions one to the other, to serve in the case
where the wood grain is found disposed in this manner, just as I demonstrated
it above, in explaining Fig. 8 of this plate.
When one wishes to make use of a wood trimmer in the manner I just de-
scribed, one places a peg in a hole of the bench, against which one presses one
of the ends of the wood to trim to keep it fixed in place, pressing the other end
against ones chest. Then one takes in the left hand the piece that one wants
to trim, and one places it on the wood trimmer, when one holds it firmly in place against the heel,
then one trims it with the plane that one holds in the right hand, just until it takes on the wood to be
trimmed, which one should take great care not to touch with the plane or the miter plane. See Fig.
15, which represents a cabinetmaker making use of a
wood trimmer, and on the bench of which is placed a
trimmer or wood fitter, of the type of which is represented in Figs. 7, 8 and 9.
I said above, on page 823, speaking of the positioning of veneer wood, that when the areas to be
covered are found to be lopsided or oblong (which are equal), it is always necessary that the diagonal joints, or the grain of the fill-pieces, are laid [to match] according to the angles of the piece to
be covered [the ground], which in turn gives different angles to the cuts of the fill-pieces. In this
case, it is necessary when one has a large number of pieces of a similar cut to make a special wood
trimmer of which the heels [ends] will be angled precisely according to the different angles of the
pieces to be trimmed, so that they are supported while one trims them.
It is also necessary that the jut of the heel be relieved below the surface, so that the veneer piece
remains in place while being trimmed and does not pop up, which happens sometimes, especially
when the piece is too long for the workmans flat hand to hold securely in place. See Fig. 11, which
represents a wood trimmer viewed from the side, and the end of this same wood, side E, same figure.
Plate 289
84
Page 838
After what I just said about the manner of cutting and trimming straight pieces, it is very easy to
imagine that this part of veneering, however simple in appearance, demands much in precision and
attention on the part of the worker. This is especially true when the designs are comprised of many
small parts, which should all be cut and trimmed before being glued, which is very difficult to do well.
As to large pieces of veneer, they impose fewer problems because not only are they fewer in number,
W
but also because they are trimmed one after the other, as they are glued, as one will see later.
The difficulty in trimming the different pieces that make up little compositions has created var-
ious methods of dealing with them to render them the most perfect possible and to accelerate their
Plate 289
execution. Some are cut and trimmed all in one piece [perhaps piece by piece or bulk trimming], and
they are glued one after the other whether all the pieces are of a similar form or not, which renders
the work perfect enough but very time consuming and very difficult to execute.
Other workmen, to avoid these difficulties, have preferred (especially when the compositions
offer regular figures, like hexagons and stars) to stick [create] the pieces of each figure first by sections, then [put them] all together such that they make a total module, which they continue, and then
glue to other veneers (or they inlay in solid wood). This last manner of execution of small compositions is the fastest and the most convenient. That is why I am going to be self-serving in treating the
manner of trimming straight compositions, both small and large, which I will make the most succinct
as possible. I am content to give some general examples that are applicable to all sorts of works, and
the number of my examples and their application is [what is] not essential, but rather the manner of
their implementations, especially when applicable to many cases, as happens in this part of veneering; apart from the knowledge of wood or other materials and of different compositions, almost all
the theory of this art consists of various ways of execution applied to different works.****
Plate 290
Follow-on the
Trimming Straight
Pieces and the Way to
Inlay Them
When compositions are comprised of small parts whose assembly forms regular shapes, as with
hexagons, stars, etc., one glues them one to the other before gluing or inlaying them, which is done in
the following manner.
One begins first by trimming and adjusting all these pieces according to the size and shape that is
correct; then, if it is, for example, a hexagon, as in Fig. 1, one glues together
the two parts, AB, Fig. 2, and one lets them dry, after which one adjusts
the third piece C, which one glues next, and one lets it also dry.
Page 838
When the three pieces that compose a hexagon are thus glued, one
trims the last one [which could be piece C or the completed hexagon piece]
in a trimming block, so as to give the most perfect possible shape, supposing that the first pieces which one has trimmed leaves too much wood, which should be avoided as
much as is possible.
In trimming hexagons, or any other assembled figures, it is good to take guard against pieces
that are of uneven size, in order to take out as much wood as is necessary, so that the angles of
**** What I propose here seems to be contradictory to what I
the form, whether to the use of these works on which you are
another work, since you always use the same ways of making
85
the figure are very exact at the meeting joint for each piece, as I have observed in Figs. 1 and 2.
[For a hexagon comprised of three identical diamond shapes, it is critical to make sure the size
of each diamond lozenge is exactly the same, and that the angles are precisely 60 and 120. Roubo
leaves us to wonder about the cutting techniques and tools for rendering hexagonal elements. For my
own use, I employ the cutting jig.]
The more complicated the shapes are, and the more difficult their execution becomes, one still
uses the same means for making them; stars, for example, are in this group because not only are they
comprised of many pieces, but also because it is not possible to re-shape them once they are glued all
together, which requires much precision in adjusting these individual pieces.
Plate 290
The stars like those represented in Fig. 3 are those of which the form is the
most simple since these have but four points: D, E, F, G, which are each made of
two pieces, and should be equal distance one to the other, such that the distance
DE, be equal to that of EF, that of FG equal to that of DG, etc., which is general for all stars of a regular form.
Whatever be the number of points in a star, they are always made the same way, that is to say, that
after having prepared and trimmed the pieces that should be used, like those of H, I, L, M, Fig. 4, one
glues them together two by two, as in N, O, same figure. When
the glue is dry, one trims and adjusts them again, if that is necessary; then one glues these same points two by two, (or three by
three, if the star is composed of six points), as in PQ, Fig. 5, and
one touches-up again, if it is necessary; after which one finishes
the star by joining parts P & Q together.
Whatever shape these different compositions, one always uses the same method for executing
them; that is to say, one forms, as much as possible, regular parts comprised of many pieces joined and
glued together, as I just explained above. This being done one applies them side by side, as in Fig. 6,
observing that in their assemblage they present straight lines, whether horizontal, perpendicular or
diagonal. Any lines that exist in such a design serve to orient the composition, which can be positioned
by straight strands in
one or another direction, as one can observe
in this figure.
When the infill forms large pieces, as in Fig. 8, one begins by situating piece A in the middle,
Fig. 9, to which one adjusts and glues B, C, of which one trims the extremities a, b, when the glue
is dry; then one adjusts hexagon D, and when it is glued, one joins the two other pieces E, F; the
rest is as usual.
Page 840
86
Infill of large compositions, like those of the interior of figures in Plate 285, are not trimmed in
advance like those of which I just spoke. On the contrary, one does not cut and fit them in place until
after the friezes that surround them are glued. Those pieces are cut and fitted piece by piece, so that
one glues them after proper fitting, using the manner that corresponds to each piece of work, as I will
explain here later.
The pieces of infill of which I just spoke above are not only applied on a single unified base [as a
completed marquetry panel overlaid on the substrate], but also are inlaid in prominent woodworking, with which they are made level with the surrounding surface, which requires nothing different
Plate 290
to the manner of preparing them for their application. As to the manner of preparing the substrate
that receives the inlays, it is very simple, since it concerns nothing but excavating in these same substrates the space necessary to receive the pieces of inlay. When they are comprised of straight parts,
as in the figures of this plate, it is very easy to do since they can be excavated, or at least begun, with a
grooving plane, and then finished with chisel and a guimbarde [router plane], like the places on sites
G, G, G, of Fig. 10, which means they are always done perfectly straight, of a width and depth equal
everywhere; as to not having a grooving plane, one can always use a marking/slitting gauge that cuts
marks to the depth that one needs, and always parallel between them.
Page 841
When the places destined to receive the inlay, although comprised of straight lines, do not pres-
ent parallel sides between them where one can use a marking/slitting gauge as for example star I, I,
Fig. 10, one cuts them and excavates them with chisels, and then finishes them with a router plane,
which is the most ordinary way to do these sorts of inlays. No matter how much care one takes in cutting with chisels, it is difficult to do without any splits and be straight for the entire length.
That is why I believe that it is better, especially when the parts to inlay are of a certain size and
made of wood difficult to work, it would be better I say, to cut them on the sides with a veneer saw,
which one can pass against an iron ruler, or at least of wood really strong, applied on the work along
the line of the area to be cut, which is very easy to do, and will render the work much more perfect,
than by following the ordinary manner.
When the sides of the area are ready to receive the inlay, having been cut with a saw, one com-
pletes cleaning the excavation with chisels and the router plane, as represented in Fig. 12.
The router plane, Figs. 11 and 13, is nothing other than a piece of wood, of 23 thumbs of width, by
12 thumbs thickness, according to its length (which varies from one foot to a foot-and-a-half, according to different needs), in the middle of which is a hole, a little slanted, and of a size capable to contain
an iron a, Fig. 13, of 34 lines of thickness, and a wedge b, for holding it in place. This iron should be
grooved like those of toothing planes, and exceed [extend out] the bottom of the router plane body by
the depth of the rebate [cavity] that one wishes to make.
This tool is held with two hands, and one makes it work by pushing [it] before oneself, as from
cd, observing, when one begins to work, to ease up a bit from behind so that the iron does not take
up too much wood, only as much as one judges appropriate. When there are very small spaces to excavate, one can avoid the router plane to do this; or if one judges it appropriate, one makes little router
planes, which can pass over all, and are easier to use than the one I just made a description of. Whatever shape and size router plane is made, it is always necessary that the iron be grooved/toothed, so
that the bottom of the excavated area is a bit rough and takes the glue better.
After having presented here the manner of cutting and trimming straight pieces, I am now going
to address those pieces that are circular, or likely with contours. Since they cannot be trimmed by a
Plate 29 0 . Follow-on the Trimming Straight Pieces and the Way to Inlay Them
88
plane like the first ones, this makes them a little more difficult to execute perfectly, as we are going to
see next.
The manner of cutting curved pieces, and the tools that are appropriate
W
Pieces of veneer with uneven shapes are cut with a saw and inlaid in the normal way with the trimming knife and other tools, of which I will give their description when speaking of the inlay of these
Plate 291
Plans and Elevations
pieces.
Two sorts of tools are necessary for cutting pieces of veneer: namely the mule [called a don-
for a Mule
key in the English tradition] or vise, and the cutting saw, otherwise called the marquetry saw. The
mule, Figs. 1, 2, 3, 6 and 9, is a type of little bench, supported by three feet, of 1516 thumbs high, which
is about two-and-a-half feet in length, by one foot wide at its largest point.
On the upper part of this bench about 6 thumbs from the end is placed a wooden vise B, Fig. 2,
which is nothing other than a piece of wood of 34 thumbs wide by 2 thumbs thickness, and 1215
thumbs high, taken from the top of the bench on which it is assembled using a double tenon, and is
secured there by means of a wedge as in Figs. 2 and 3, and Figs. 7 and 8, which represent the face view
and side view [which are] double the size of elevations 1, 2 and 3.
The vise is split [sawn] in its thickest part into an opening of about 34 lines of width or more,
and one takes care that it be a little narrower at the top than at the bottom, so that it tightens better at
the opening top. A slot should extend just 2 thumbs from the attachment of the vise; and one should
observe that the flange or jaw of the vise, which is on the side of the worker, should be thinner than
the other side, so that it bends more easily when one wishes to make use of it. The other flange of the
vise, that is to say, the thicker one, is applied against a buttress c, Fig. 2, which is assembled there by
mortise and tenon into the top of the bench. Sometimes we do not use a mortise in the flange of the
vises, but simply a notch of 68 lines deep, coming from nothing at the base, in which the buttress
enters completely, as I have noted in Figs. 7 and 8.
In the middle of the upper section of the bench is placed a little riser a, Fig. 2, of about 5 thumbs
back from the front of the vise, on top of which is assembled a levered [hinged] arm D, the other end
of which comes to abut against the recess at the front of the vise in order to make it bend and by this
means hold solidly the sheets [of veneer] that one places within for cutting them up. The lever [arm]
enters the notch on the recess of the vise, so that it does not wander to the right nor to the left, and it
pushes down by means of a cord from below, which is attached at points a, a in Figs. 1 and 2, which,
passing across the bench, becomes attached at the foot pedal FG, on which the worker puts his foot
to tighten the jaw of the vise, as I will explain in speaking of the manner of cutting. See Fig. 1, which
represents the mule from the end of the lever, and Figs. 4 and 5, which represent this same lever front
and side views.
Page 843
The bench on the vise side needs a shelf to hold the little pieces that the worker puts there [the
pieces that have been cut out]. The top of the same bench must be hollowed enough so that the
worker can be seated comfortably.
We also use another type of mule or bench, which differs from that which I just spoke of, in that
it is placed on a table of 16 to 18 thumbs high, where the middle is hollowed to place the worker who is
working seated on a tabouret, and consequently is surrounded on three sides of the table on which he
90
puts his work on the right and the left. As these sorts of mules are little in usage at present (although
very convenient when you have many pieces to cut out), I am content to make a short description,
without giving any figures, which does not seem to me to be necessary, given that the essential part
of this mule, that is to say, the bite [the vise], is constructed the same way as what I just described.
There are other small vises or mules that you hold on the bench with a clamp that do not differ
W
from the one here, except that they are smaller, they have no feet and the plank of which the bench
is assembled is notched to accept a small vertical post from which the lever extends. You make use
of this mule the same way as the one with feet; that is to say, you always close the jaw of the bench by
Plate 291
means of cord aa, on which you set a foot, and you substitute a little platform in place of the step.
There is still another type of vise or wooden mule, Figures 10, 11 and 13, where the jaws open paral-
lel by means of a wooden screw held by a collar [the English word is garter] in one of the jaws, such
that it holds to the latter [rear jaw], whether opening or closing the vise.
This vise is held on the bench by means of a clamp, and it is convenient not only for cutting and
filing pieces, but also for gluing small pieces that could not resist the shock of a clamp. As to the
way that the nut of this bench is held in jaw I, I am not going to speak of that here because it is the
same thing as that of the end vise of the German bench, which I described at the beginning of this
part, page 803.
Page 844
The marquetry saw is a tool most necessary and most complicated used by cabinetmakers. It is
composed of a frame of iron, Fig. 4, ABCD, with two jaws, E and F, which serve to hold the blade
of the saw and of an arm or handle GH, of which part H holds the screw for tightening the blade as
Plate 292
much as is necessary.
The frame of a marquetry saw is made of flat iron of 68 lines in width by one-and-a-half lines
thickness at most. The upper part of this frame is flattened at the end, and is pierced by a square hole
through which passes the end of pin-clamp [pin vise] E, which is threaded to receive a nut that holds
it fixed on the frame; in which it cannot rotate since the hole at the top of the frame is squared, and the
end of the jaws is of the same form, see Fig. 1, which represents the frame of the saw viewed from above
with its nut, Fig. 5. See Fig. 2, which represents this same saw cut sideways down its entire length, and
Fig. 3, which represents the front view.
Plate 29 2 . Marquetry Saw and Its Explanation and the Way to Use It
92
The lower part of the frame, Fig. 4, is terminated by a socket I, which is pierced by a square hole
up above, similar to that of the upper section, and in which [the] dowel part G of the arm is fixed to
stay. See Figs. 6 and 7, which represent the end of the arm of the frame, and its shaft [socket] viewed
sideways, and this same socket viewed from below.
The jaws E and F, Fig. 4, are both composed of two parts each, which are separated in half of the
W
thickness by the jaw, and are retained together by means of a screw, as one can see in Figs. 2, 3 and 4.
These jaws are the kind that when tightening these screws, one locks in place the blade of the saw
ab, Figs. 3 and 4, which because of its small width, cannot be otherwise.
Plate 292
The two sets of jaws are, as I just said similar, on either side of the saw. But they differ in that
the one up above has
a very short pin, instead of the bottom
that has one that
penetrates the full
length of the handle,
with the lower part of
which is fastened by
means of a nut. Other
than this difference in
length they are perfectly equal, having
both a square pin to
prevent their turning
in the frame, [and]
both are threaded equally, as one can see in Fig. 2 and in Fig. 13, which represents the jaw below in all
its length, with the total shape of the handle indicated only by a tracing.
The interior of these jaws should be trimmed [textured] like a file, in order to better hold the saw-
blade. It is also good that the mobile part of the jaw enters [sits] in a rebate at the base of the fixed jaw,
so that it is not subject to turning while one tightens the screw. See Figs. 8, 15 and 16, which represent
a large jaw as executed, the mobile part of this jaw viewed from inside, and the screw appropriate to
tighten them both.
Page 845
The handle of a marquetry saw is, as I just said above, comprised of two parts: the one G, Figs.
2, 4 and 12, which is fixed to the iron frame, and the other, H, Figs. 2, 4 and 11, which is mobile, and
enters in the first, which is shaped as to receive the squared part of the pin of the jaw and the pin, or
coupling bolt, of the mobile part of the arm. This last one, side H, Figs. 2 and 11, is also pierced along
its whole length, to let pass the jaw-screw, which is fastened by means of a nut c, Figs. 9 and 11, of the
manner that when turning part H, Figs. 2 and 11, one raises or lowers the lower jaw of the saw, which
one tightens or slackens [the blade] by this means, as one judges appropriate.
Woodworkers/Cabinetmakers have some cutting saws of different sizes, according to the work
that they wish to work. That which is represented here is the most ordinary size, but whether they be
large or small, their shapes and construction are always the same. It is always necessary that the horizontal branches of their frames open a bit in the front, so that when you handle the saw, they cannot
turn inwards.
93
The blade of the cutting saw, such as that represented here, has teeth in the same line as the frame,
like ordinary saws. However cabinetmakers use them differently, that is to say, that instead of viewing the side of the blade as in Fig. 4, the teeth are found on the front, such that they saw horizontally.
The frame is always perpendicular, at least a bit so, such that the hand is relieved, and the tracing that
the saw makes is always parallel with the top of the jaw of the vise. *****
W
The blades of cut-out saws are made with some pieces of watch springs split into different widths,
from a half-line, up to a line-and-a-half at most, by 5 to 6 thumbs in length. This makes the largest saws
having hardly more than 6 thumbs distance between their arms, which you extend in length as much
as you judge appropriate, that is to say, as much as you want, in keeping the firmness convenient.
Plate 292
The teeth of these sorts of saws are very fine, almost straight, like those of press saws, and that so
As to the way of using the cut-out saw and the mule, it is very easy, because after having traced the
Look at Fig. 14, which represents a worker occupied with cutting out a piece and is placed it on the
There are cabinetry-woodworkers who dont straddle themselves on the mule, but only sit on the
side, which requires them to place the vise obliquely to be able to work. This method is little convenient, given they are always in a constrained situation. Thats why one should make little use of this.
Before cutting out the contoured pieces it is necessary to trace them according to the design of the
work. Sometimes one glues the design down to the face of the veneer, so to better follow the contours.
Then one cuts the tracings out as I just described above, observing to follow these contours with all
the precision possible, always beginning right to left, and working so as not to pull out the saw before
the whole piece is cut around the perimeter, at least as much as is possible.
When a piece is emptied [cut] in the middle [of the work], one drills a hole big enough to fit the
me believe that the latter could use one of the other way, and
Page 846
94
blade, which one detaches from within the jaws, at the top and at the bottom, and one reattaches later.
[As a general rule the detachment of both ends is necessary only for fairly large blades. With modern
ultra-thin blades, detaching one end or the other is usually all that is needed.]
When the distance from the outside to the inside of the piece is not very considerable, wood-
workers/cabinetmakers do not bother to detach their sawblades, but after having made the exterior
W
contour, they cut across the solid piece by following the wood grain. By this method they can cut
interior lines without laying down the saw, which makes no other inconvenience than to weaken the
piece a little. The joint does not show once it is glued.
Plate 293
What I just said about the way to cut out curved pieces composes almost all the theory of this
part of cabinetry. However, when the shape of pieces is perfectly regular, as with rounds, arcs of
circles, etc., it seems to me that the method of cutting by saw is not perfect enough because however
skillfully they can be sawn, it is hardly possible not to encounter some unevenness. That being less
Use Them
the case with curved pieces in the form of a circle, one would very well make them obvious in pieces
of an irregular form [or when the curve is part of a circle, no unevenness is permissible; when the
curve is irregular, some unevenness is permissible]. That is why I believe not in using the saw, but
on the contrary to use a trammel or beam compass, as in Figs. 1 and 3. [A trammel or beam compass
consists of a pair of heads on a square beam. Usually it is the pin head that is movable; it is the
point placed on the center of the arc. The other head is fixed on the end of the beam and does the
cutting.] In the cutting head of this tool one puts a blade similar to that of a trimming knife, or even
a little saw, with which one cuts out the pieces much more regularly than with a trimming knife
given that the beam compass with pin, once adjusted, will [make] all the cut pieces perfectly similar
and without any unevenness.
Page 847
The trammel compass can be made of wood, as Figs. 1, 2 and 3, or in iron, as in Figs. 4 & 5, especial-
ly when they are destined to cut little pieces. In one or the other case, they are always composed of a
shaft AB, Fig. 1, at the end of which is placed a touch-hole [mortise for the iron], for placing there an
ordinary iron, which one holds there by means of a wedge, as in this figure, or even with a turnscrew,
as in Fig. 5, which works better than a wedge, truthfully, but also which is not practical except when
this tool is of iron or copper.
The bar of a beam compass passes through the head of a box C, Fig. 1, affixed with a point below,
and which one fixes on the shaft by means of a turnscrew D, which works better than putting a plug
there, as do the carpenters, because the strikes that one needs to give the plug to tighten it always
moves the box slightly, changing the radius distance, which one must avoid. What is more, in order
to make it stronger and more stable for all type of trimmings, it is good to make the hole or mortise of
the box, Fig. 3, and by consequence, the bar of the beam, narrower at the bottom than the top, so that
being pressed by the screw, the bar always meets perfectly within, especially on the sides.
For the compass with an iron bar, the screw of the moving head/box can be placed on an angle, as
in Fig. 4, which is more solid than when placed below or on the side because the pressure of the screw
is made on four sides of the bar by means of an angled pressure plate placed in the interior of the box,
on which pressure from the screw is applied.
As to the manner of making use of beam compasses, it is very easy, because it concerns nothing
other than placing the point of the box at the center of the piece to be cut, and to adjust the distance
between the box with the compass point and the cutting head to yield the desired arc radius, as from
EF, Figs. 6 and 10. Then after having secured the box, one takes the left hand and presses down,
Plate 29 3 . Tools Appropriate for Cutting Out Circular Pieces and the Way to Use Them
96
so it does not leave the centerpoint, and the right hand moves the head of the gauge just until the
piece is cut.
The beam compasses are not only appropriate for cutting pieces of veneer, but also for cutting
pieces of woodwork, whether foundation or veneer, for making inlays, which are not normally cut
out except with trimming knives, Figs. 11 and 12. Using a trimming knife well is a difficult process
W
because it is hardly possible to follow perfectly a contour of a certain length with this knife without
making any variances other than the unevenness of the wood grain pulling away the knife, regardless
of the care that one can take to prevent from turning. That is why it is necessary to have no regard for
Plate 293
the antiquity of this usage, and not to use a trimming knife for inlays. [In other words, in this case the
old way is not the best way.] As much as is possible, one should use a beam compass with pin.
Page 848
If it happens that one is using this compass, in order that the point of the box will not make a hole
in the center place, one could glue a little piece of thin wood, which would receive the compass point,
which would remove all type of difficulty except for the requiring of more time. I do not deny the usefulness of a beam compass with pin when used to advantage that the trimming knife does not have,
especially when a curve is composed of many arcs of circles, all of which have different centers. But
this use of time should be counted as no little thing, especially in the case of the type of woodworking
of concern right now, which cannot be done with too much care and precision.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
We all have our favorite shop knives; mine is a Swiss chip-carving tip that gets used in many ways. And
like you I have tuned it exactly to my preferences. Yet, more and more I find myself reaching for the shoulder
knife that I made at about the time this book project began.
One of the issues of knife work is balancing the power and control integral to its use. Typically one of the
limitations is the amount of force you can bring to the cutting tip, and the precise control you can exert on it.
97
One fits the beam compass cutting head with not only a type of blade, like in Fig. 7, but also saws,
between themselves.
Plate 293
386, other than by the length and the shape of its arm, which is
about 18 thumbs long. The manner of using this tool is that one
takes the two hands, a little above the cutting blade and one puts the other end of the arm on a shoulder to give leverage for pressing down, after which one directs the handle of the knife as one drives it
into the wood, holding it a little bent and pulling it toward your body.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
The determining factor is often the amount of handle you can grab comfortably. In fact, that is why my favorite knife has a small blade but a comparatively large handle. Still, I am limited to having only one hand on the
handle. A shoulder knife overcomes that because the handle extends all the way from the knife tip to, well,
your shoulder. You can obtain great power and control because it allows you to grab its handle firmly with
both hands and leverage it off your shoulder.
The shoulder knife has practically disappeared from the woodworkers tool kit, and to my knowledge
only one company supplies them commercially. Making one is fairly straightforward. Although it is a simple
tool, mastering it is not so.
The first step in making a shoulder knife is to make a pattern so it fits exactly your upper bodys dimensions
and posture relative to the work surface. You can make a template from something as simple and disposable
as heavy cardboard. A good starting point is to simply grab a yardstick tip in your hands, drape the stick over
your shoulder and make note of the measurement from the work surface to your shoulder. Mark this out on
the cardboard, then draw an arc to mimic the curve of your shoulder. Cut this out and compare it to your own
body. Revise it until the match is the one you want. I made perhaps a half-dozen patterns until I got what I
wanted, and then I cut that pattern out of three or four layers of cardboard and bonded them together to make
it sturdy enough so I could get a good feel for its shape and fit. Just to make sure, I made a final pattern out of a
piece of 6/4 softwood.
I selected a piece of scrap walnut to make my first knife, and a slab of ancient oak for the second, which is a
few inches longer than the first. I used disparate methods for building each.
98
The background areas ready to receive the circular pieces of inlay are cut out in the same manner
as for the straight pieces, that is to say, after having cut and dug them out with a chisel, one finishes
them with the router plane, at least as much as can be done. Then one glues the pieces to be inlaid,
straight or curved, as I will instruct after having discussed the manner of veneering beginning with
the most simple works, and gradually reaching those that are the most complicated, as I always do in
W
my work.
Plate 293
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
I made the walnut knife from two pieces of 3/4" stock laminated together to make setting the blade much
easier even though the final thickness was 11/8". I traced my pattern on both pieces and cut out the shape
of the handle. Using a knife and chisel, I excavated a void matching the shaft of a Swiss blade purchased at a
woodworking store on the two inner faces that were to be glued together in the final assembly. When the fit
was perfect, I glued the whole package together with hide glue, with the knife blade embedded in and protruding about 1" from the long handle.
For the oak-handled knife, I started with a 6/4 slab, then traced and cut out the shape I wanted. When
I was satisfied with the overall shape, I sliced it lengthwise on the band saw. Recycling an old chip-carving
blade, I excavated a pocket for the knife haft, then temporarily tack-glued the two pieces back together to
shape the handle. (This is unlike the first knife when I assembled the knife and then shaped it.)
With spokeshaves and files I shaped the handle to my preference, inserted the blade and glued the whole
thing back together with hot hide glue. After shaping the business ends and adding compression-fitting brass
ferrules, I coated both handles with shellac and wax, made the leather blade guards and called them complete.
My skill at using the shoulder knife is growing, but it is not yet to the degree where it is second-nature. But
classical marqueteurs probably used it about the way we would use a scalpel for cutting filigree in paper.
One of the main differences between the manner of creating marquetry between the way I did it for de-
cades and the way that Roubo practiced the art has to do with the assembling the compositional elements
into the background. I had previously always sawn them together in fairly typical tarsia a encastro technique,
and frankly it is still the practice where I feel the most comfort. But for Roubo and his contemporaries, the
elements were often set into the background by scribing the elements outline into the background with a
shoulder knife after the background had been glued to the substrate.
This is in great measure the definition of David Pyes workmanship of risk. Careful examination of
enough old pieces of marquetry will indeed reveal instances where the knife got away from the marqueter.
99
S e c t i o n II
The
m a n n e r o f g lu i n g a n d v e n e e r i n g
M a rqu e t ry
HEN the different pieces of veneer are arranged, as I have instructed above, now comes the
gluing of pieces in place, which the cabinetmakers call veneering, an operation to which they
bring much care because it is on their degree of perfection that all the strength of their work depends.
Before presenting the techniques of veneering, I am going to detail the tools appropriate for
Plate 294
The Way to Veneer
and the Tools
Appropriate for
Veneering
tate the knowledge of which I will speak when talking about the
different veneers.
There are two sorts of hammers for veneering, one, Fig. 4, which has the straight peen, and which
serves to veneer the work when the surface is flat and uniform. The other, Fig. 5, of which the peen is
curved, serves to veneer the work when the surface is concave.
There are heating irons where the bottom is curved to work on hol-
The hand clamps are of two kinds: the ones that are 6 thumbs to one
Page 848
100
foot long at the opening, and even more, which are constructed of wood. The others, which are much
smaller, are made of iron or copper.
The first of these types of presses [hand clamps], represented in Fig. 1, is composed of three pieces
of wood.
Namely part AB, is intersected by the two arms of the clamp, which are assembled by a douW
ble-fork mortise and tenon [mortise and tenon with two flanking bridles], which passes crosswise to
the main section AB. See Fig. 7, which represents one of these arms of the clamp, in cross-section,
with its assemblies, the end of which a hole is pierced [threaded] by the screw.
Plate 294
The other arm of the clamp is similar to this, except that it is not pierced [threaded] (for the screw)
like the first, and it is a little hollowed on the bottom, that is to say, inside the press, so that when
one closes the screw between this and the work, it grips and remains square, which cannot be if not
hollowed on the bottom, because however well-made the clamp and arms, they still have some kind
Page 849
of bending effect. The screw of this kind of clamp should be about 8 lines thick and terminate below
by a flathead [thumbscrew], which serves to turn it. However it works better, instead of this head, to
leave there a handle of about 3 thumbs long, of a flattened form, with which one turns more quickly
and conveniently.
The little iron clamps, named happes [from the verb happer, to grip violently], do not differ
from those of which I just spoke except by the size or the manner by which they are made. Their
forms and uses are absolutely the same, since they both serve the same function of applying the parts
of veneer for which the glue has not set, as indicated by Fig. 2, where I have represented a piece of
veneer held by the hand clamp.
Other tools appropriate for veneering are the farriers points, or, better said, the tips of flat nails
used by farriers [those who shoe horses], made of malleable iron, of which I will speak later; a sponge,
as in Fig. 6, for moistening the veneer both before and after being glued, and a pot of glue, made of
copper, with double bottoms, which is to say two pots or vases, which fit one inside the other, the
outer one of which one fills with water, the heat of which also maintains that of the glue, which is
placed in the other one. One calls this heating the glue in a double boiler, which is very convenient
because not only is the glue maintained to a degree of heat almost always the same, but also because
the glue does not attach to the pot and does not burn as happens to ordinary glue pots.
There are cabinetmakers who have little stoves of about 2 feet in height on which they heat their
glue, which is very convenient since they do not have the bother of stooping as often as when the glue
is before the fire where it heats less equally than on a stove, on which one can leave the first pot, in
which the water remains, and remove the pot that contains the glue, as one judges appropriate.
One should only use the English glue for veneering, at least when one wishes to do good work.
And one should take care that it is always hot and a little thick without, however, being too thick,
because it would create a layer under the veneer and would not all squeeze out when one uses the
veneering hammer.
If too-thick glue is not good for working on veneer, that which is too watery is even less, since it
does not stick at all and the veneer cannot really take hold, regardless of how much care one takes, and
whatever diligence one uses.
When one wishes to veneer, one begins first with the perimeter pieces of the work, which one
adjusts first the length as well as the width. Then one puts many nails along the length of the edge
against which the veneer will be joined, both along the length as on the ends, so that the veneer will
not move when applied, as one can see in Fig. 10.
Plate 29 4 . The Way to Veneer and the Tools Appropriate for Veneering
102
Page 851
Then one moulds the piece, that is to say, that one presses with the hammer head the side that one
wishes to glue, so as to hollow it and by consequence carry the adhesion better to the ends. That being
done, one moistens the piece with a sponge with tepid water on the face [of the veneer], or better with
very clear glue. For this glue, of whatever consistency, fills in the wood pores better, which one cannot do with water. This step should be strongly considered, especially if one does not thus moisten
W
the pieces of veneer on the facing side [show face] as to prevent their buckling on this side while one
spreads the glue underneath the veneer, which will not fail to happen unless one takes this precaution. The sheets of veneer
Plate 294
When the face of the veneer is thus moistened, one heats the side to be glued down, and one
moistens there if it is still necessary. Then one coats it with glue and one sticks it in the proper place
as quickly as is possible. One then takes the veneer hammer [and] presses the peen strongly on the
surface while pushing it before oneself, and while agitating from right to left without ceasing to apply
the most pressure that is possible so that the glue is introduced in the pores of the substrate and veneer as much as is necessary for it to be glued down, without which excess glue remains between the
two layers. If any excess remains, it will make a strange mass that will self-destruct later, whether by
too much heat or humidity. That is why one always begins by veneering a cabinet from the end, and
one moves measure by measure, always pushing the glue before oneself, which not being retained
exits by both sides of the piece, if isolated, as in Fig. 10. If one of the sides of the piece to be veneered is
adjoining another side that is veneered, one begins by veneering at the joint, always beginning at the
end of the piece, and pushing the glue ahead of oneself from/toward the side of the piece that is not
veneered.
Once the two sides of the piece are engaged, one veneers them in the ordinary way, observing to
put a wedge at the end, between the piece of veneer and the frame, so as to leave a passage for the glue.
One does not remove this wedge except precisely at the instant when it is necessary to veneer the end
of the piece.
When one has veneered a piece, one probes it, that is to say, one taps the surface with the head of a
hammer to find out by the sound that is made if it is adhered well overall. This is easy to learn because
when veneer is well adhered it makes a sound full and round. But when the veneer is poorly adhered
103
it makes a dead sound, which shows the place that is not well adhered, and over which one should
pass the veneering hammer, always agitating in various directions.
When one has veneered a border piece, such as that of CD, Fig. 10, one veneers that which is
opposite, such as that of EF, same figure. This being done, one veneers the two others, which one
Page 852
does not trim to fit ordinarily until the first two sides of the border have been veneered. Whatever
W
care one takes, almost always there is some dislocation, which one cannot correct once the glue has
taken hold.
Gradually, as one veneers the pieces one takes a chisel or even a wooden spatula to remove the
glue that sits on the surface in order not to lose it, or not to hinder the veneering of the rest of the
Plate 294
work.
When the pieces of the edge of whatever kind of work are finished, it is good to let them dry one
day at least, after which one adjusts the veneer pieces to the center of the panel, which one always veneers in the same manner as above, except when they are very large, one veneers them in parts, that
is sizes of 6 thumbs by 12 thumbs. This requires the use of a heating iron in order to heat up the glue
to the appropriate temperature.
The heating iron is held in the left hand, and one passes it by pressing above, from the right, as
much as one judges appropriate, to heat the glue without burning the piece to be veneered.
Instead of using a veneering hammer, one would do better to make use of a simple piece of wood,
or the sleeve of another tool for holding the heating iron because its heat puts anyone who uses it in
danger of burning their hands, and which in my opinion, is not really necessary. Look at Fig. 11, which
shows a worker busy heating a large piece of veneer, edged by straight banding, which is the type of
veneer the most simple.
{See Errata to find location of Fig. 11.} [ Roubos corrections acknowledge Fig. 11 is missing but
does not supply the illustration.]
When the cabinetry is more complex and there is flat banding around the friezes, as for example
in Fig. 1, one begins by veneering those in the middle of the work, which is done in the ordinary way,
that is, one pushes against the points, as in section AB.
Plate 295
However, when these flat bands are narrow they are subject
Follow-On to the
Cabinetmakers
that in place of this, one would do better to put a straightedge, like that of CD, against which one can push the flat
banding all along its length.
The infilling of the friezes are done after all the flat band-
ings are dry. One cannot veneer them immediately after having veneered the flat bands of the interior, like the frieze in side E, which would allow for more ease in addressing their exterior joinery,
but it would result in difficulty in veneering the exterior flat bandings, which one can barely join well
Page 853
104
against infill F. That is why it is better not to fill the friezes except when all the flat bandings, interior
as well as exterior, are glued and dried.
When the friezes are filled in by compartment, like part G, Fig. 1, one veneers the pieces of these
compositions one after the other, and always after the flat bandings are positioned. This requires
much care; even though each of the pieces of these compositions are prepared in advance, as I have
W
instructed in speaking of the manner of adjusting straight pieces, see page 835, etc.
The veneering of these pieces in the middle of this figure is always done in the normal way, that
is, positioned diagonally, as in Figs. H and I, beginning by adjusting one piece like that of HI, which
Plate 295
one veneers and one fixes, as one judges appropriate, by claw nails [veneer pins], Figs. 4 and 5. This
first piece thus glued, one then adjusts a second by its side, which one veneers and fixes in the same
manner as the first, then a third, and finally a fourth, which makes the infill a panel, at least at the
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
More creative woodworkers might resort to applying either hide glue or polyvinyl acetate (white or yel-
low glue) as the adhesive and letting it dry, then using a heated iron to tack things down. These strategies begin
to approach the traditional and nearly foolproof technique of hammer veneering.
Many artisans interested in historical craft have tried hammer veneering, and even more have observed it.
Without going into a recitation of the process, nor an exhaustive troubleshooting discourse (hint: The biggest two problems are No. 1: Not preparing the substrate or veneer well enough, as described by Roubo, and
you do, the glue will chill and gel too quickly, plac-
105
corners, because it rarely happens that each piece is of a size sufficient for filling in entirely. There
always remains some voids, like those of L and M, which one fills in with other pieces of which the
colors and veins match those of the corners as perfectly as possible, in a manner such that when the
work is finished it appears to be nothing but a single piece.
The more these pieces of infill of a panel have corners, or, as the cabinetmakers say, sharp angles,
W
the more one should take care that these areas are well veneered, because it takes only a little air to
get below for the pieces to lift right away. One remedies this by fixing them with a point that curves
down, which is in my view not very good, because this point thus curved does not work but in one
place, and it is subject to spoiling the joint, and even to crushing the tip of the piece. That is why I
Plate 295
believe that it would be better to use a claw-foot nail, Figs. 4 and 5. They differ from ordinary nails
with their little head or raised foot, which can hold the veneer without ruining anything, neither on
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
No. 2: Using the glue when it is too viscous) some fundamentals about the materials science involved might be
useful. The question that keeps coming up is a simple one, which fortunately has a pretty simple answer: why
does hammer veneering work?
The answer lies in the nature of the hot hide glue utilized as the adhesive. More properly, it lies in the pro-
cesses by which the hot liquid adhesive becomes a rigid, strong glue layer.
Hot hide glue is generally a variety of bovine proteins dissolved in water. Protein solutions such as this
undergo a non-intuitive process when going from liquid to solid. When the temperature of the hot hide glue
diminishes, at a given point of cooling the solution no longer remains liquid even if no drying has occurred
through evaporation. That temperature is known as the gel point where the solution turns into, well, a gel.
(The exact numerical value of the temperature where this occurs depends on a number of things, most importantly the gram strength and concentration of the glue.) At a craft level this is when the glue sets. But that is
only one half of the equation.
At the gel point, the glue becomes solid or at least non-liquid, but it has virtually no cohesion (stick
together-ness) nor gluing adhesion (stick to-each-other-ness). But it has enough, and it is pretty sticky (a phenomenon known as specific adhesion as opposed to mechanical adhesion, which relies on the interlocking of the components). The full hardening of the glue comes only as the water evaporates, leaving behind a
thin, epoxy-strong layer of adhesive. To reach this point, all the excess water that was used to soak the glue
off (the glue is still very soft at this point) and start
all over.
106
top nor on the side. Look at Figs. 4 and 5, which represent this type of nail in its natural size viewed
from the front, from above and sideways.
The veneering of shaped [curved] pieces is done in the same manner as that of straight pieces,
that is to say, that after having glued the flat bands, the straight as well as the curved, as in Fig. 2, one
Page 854
adjusts the pieces of infill, and one veneers them in the normal way.
W
Since pieces of infill cannot all be traced with a compass, one is obliged to take their curves from
that of the flat bandings already positioned, which is done in two different manners, both of them
reverting to almost the same thing, as I am going to explain.
Plate 295
One takes one or many sheets of paper, which one applies to the flat bandings that are in place, on
which one takes care to hold them really flat. Then, with the sleeve [ferrule] of almost any tool (provided they are round and smooth), one rubs on the paper while pushing a little in the middle of the
part to be veneered, and following exactly all the contours of the flat banding, on which one makes a
fold, or better said, a track that gives exactly the contour of the empty space, and one makes the tracing with lead or red chalk, for fear that it will disappear when flattening the paper. See Fig. 9.
After having thus taken the tracing of the piece, one flattens and glues it on a sheet of veneer.
When it is dry, one cuts out this sheet with an ordinary marquetry saw. It is necessary to observe
when gluing the tracing on the sheet not to moisten the tracing except on the space of the pattern so
that it keeps its size and will not shrink while drying.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
during its preparation must evaporate for the glue layer to reach a moisture content thats at equilibrium with
its environment.
One of the quickest ways to provoke the cooling of any molten material is to reduce the cross-section of its
mass. In other words, make it thinner so it can disperse its heat more quickly into the surrounding material.
That is precisely what the hammer veneering technique accomplishes cooling the molten hide glue to the
ambient temperature (below the gel point).
In hammer veneering the critical process involves quickly slathering on a comparatively thick layer of mol-
ten glue to the underside of the veneer and to the top side of the substrate, placing the two in contact with each
other, and laying down the veneer by pressing the cross peen of the hammer over the surface with considerable force. (Some back of the envelope calculations indicate that when I hammer veneer, the downward
force of the hammer edge approaches 500 pounds per square inch). The hammer acts like a squeegee, forcing out the excess molten glue. The nearly immediate effect is to reduce the cross-section of the molten mass,
causing the room temperature to cool it in a fraction of a second and induce almost immediate gel formation.
While the remaining glue layer is not yet very strong, it is strong enough to hold everything in place, even
if the substrate is curved. Remember, at this point the glue is neither liquid nor fully solid, it is a sticky gel. The
really good news is that it will become increasingly hard and strong in a relatively short time as the moisture
departs the system. Less good is that the glue layer remains water-sensitive, even once fully dried.
While nearly perfect, this system can be improved by manipulating the ingredients. Adding a little glyc-
erin (about 5 percent) to the hot glue solution increases the specific adhesion (it is stickier), making the
immediate gel-tack more robust. You can overcome the glues vulnerability to moisture through the incorporation of protein cross-linkers to the system. I accomplish this by brushing a 5 percent to 10 percent solution
of photographic gelatin hardener to both of the dry gluing surfaces then allowing them to dry. This allows
the cross-linking agent to leach into the liquid glue when it is applied. Depending on how much of the agent is
used, the result can range from a glue layer that is moisture-resistant to one that is impervious to it. In effect,
you can accomplish what is in working and performance effects an adhesive with the simultaneous benefits of hot hide glue and two-part epoxy.
108
When there are many pieces of a similar form to adjust, to practice the most diligence, one makes
but one tracing, which one pricks with a pin on all its contours. Then, after having positioned it on
the sheet, one rubs it with a little sachet full of lead dust or of pulverized red chalk, which passing
over the holes of the tracing, leaves on the blank sheet a little punctuated pattern, which one connects
later with a pencil.
This manner of taking the shape of curved pieces is rather inferior because it is hardly possible
to make all the shapes of the flat bandings perfectly similar, and that when they are, it is not possible
to glue them without bending them a little whether inside or outside. That is why I believe that it is
Plate 295
better to take a tracing for each piece, or even to use the following manner of taking a shape of curved
pieces, which is easier and simpler than the first.
After the flat bandings are perfectly dry, one rubs them on the ridge with chalk. Then one takes
a sheet of paper that one cuts and trims from one end, according to the direction that is appropriate.
This being done, one puts it in its place by the end that has been
fitted. One makes it go over the curved flat bandings; and with
the veneer hammer one strikes [presses] lightly on the sheet,
following as much as is possible the interior ridge and the contour of the flat
bands, which being rubbed with chalk, as I just said, transfers onto the veneer
a little white line that gives the exact contour of the piece that one normally cuts
out with a marquetry saw. See Figs. 6 and 10, which are cut out like this.
Page 855
I just said that one trims the pieces first from an end, and that is the good way to do it, because if
there is something to be retouched, you are the master of it. However, there are some cabinetmakers
who take the contours by both ends (which I suppose are both curved) at the same time, and who, by
the habit they have, are almost sure to never miss anything.
Since Fig. 2 is very small, I could not represent there the strings around the flat bandings, whether
straight or curved, even if it is always good to place some.
These strings are glued to vertical edges before being trimmed, especially when they are narrow,
like the friezes of broken bars and others. As to those that are curved,
you do not glue the strings except after they are veneered, and you let
them follow the contour of these same flat bands, using nails, Fig. 8, at
equal distance from each other, especially in the hollowed part, as I have
observed in Fig. 7.
In general, when one veneers cabinetry, it is necessary to avoid doing it in too-humid locations, or
[when] exposed to too much sunlight, because one will dry the glue too quickly, which will prevent
it from adhering. And, one way or another, [it] will allow for air passage, which will lift the veneer. It
is also necessary to take care to cover the veneered parts with some moist linen to prevent them from
drying too promptly on the exterior, and if it does not lift in the area of the joints, one should take
great care to cover them with glue, even after having veneered them.
However much care one takes, it sometimes happens that the veneer lifts. Thus, one can restick
[it] with glue by heating it again with the heating iron. And if it too is lifted to lay back down with the
veneer hammer, one can restrain it with headed nails, when it is possible to place them, or with hand
clamps, or even with a screw press or even a go stick, that is to say, with a flexed wooden stick of
which one end is put against the floor of the shop and the other against a wedge placed on the work
such that the rule straightens it between the work and the floor, or, more likely, the ceiling of the shop.
When the work is curved in design, one veneers it the same way as when it is straight [flat], ob-
109
serving, while one handles the length or the contour of the pieces, to bend them according to the
design of the work. When the design is considerably arched in an S, as for example Fig. 15, it is good
to fix the edge of the veneer with a hand clamp, on the side of the round and to hold the curved part
by means of a stick N, or a wedge O, and a cushion [wadding] P, placed between this wedge and
the work, to better hold the curve. See Fig. 15, which represents the side view
W
of one part of the veneer, thus positioned, and Fig. 17, which represents this
same veneer developed along a straight line.
What I have just said regarding curved veneer is no good unless the curve
Plate 295
or even, if they are, one can only fear that they will not
become straight again. That is why in the case of one
part being very curved and a little stretched, it is necessary to curve the pieces before gluing them, which
is done in the following manner.
resented by Fig. 11 (setting aside those of two veneers, QRS and TUX), one begins by trimming all
these pieces along their length, like that of Fig. 14, [which is] the veneer QRS or that of Fig. 16, [which
is] that of TUX.
to wet them so they bend more easily. Cabinetmakers use for this
operation the first piece of iron that they have in their hands,
which makes them not only burn themselves, but also to badly
curve their pieces, whether by heating them more in one place
than another, or whether by making them bend unequally. That
is why I believe it is better that they make use of a tool made expressly for this work. This tool is represented in Fig. 13, and I will
call it the moulding iron [bending iron]. (See previous diagram
w/ Figs. 12, 13, 15 and 17.)
When the work is convex, like side QRS, Fig. 11, one must take care that the pieces of veneer are a
Page 856
110
little more curved than the space, so that they hold well at the ends, which is essential for all sorts of
veneer, and especially in the case here. By the same reason, when the work is concave, like side TUX
(same figure), it is necessary that the veneer be less curved than its space, so there will be no difficulty.
When the spaces are absolutely too curved to apply bent veneer,
at least with any firmness, one uses a wood of an appropriate thickW
Cabinetry contains almost all the theory one can teach on this subject, and is applicable to other
types of cabinetry of which I am going to speak afterwards, at least with some differences, which I
Page 857
There only remains, for completing this chapter, that which concerns the first thing of cabinetry,
namely teaching the manner of finishing after having veneered the cabinet, and the different ways of
polishing all sorts of works, addressing all the knowledge necessary for the practitioner to gain thorough knowledge of all that concerns finishing, which is a bit the same for all similar works, and will
be the subject of the following section.
S e c t i o n III
The
way t o f i n i s h
Ve n e e r Wo r k ,
and some
d i f f e r e n t t y pe s o f p o l i s h
HEN the veneer is sufficiently dry, it is time to finish it, that is to say, to level all parts, after
which one polishes it, as I will explain here. [In the English language craft lexicon, the word
Plate 296
The Way to Polish
Cabinetry and the
polish can mean three distinct processes. First is the smoothing of the surface with scrapers and
abrasives. In the time of Roubo, these abrasives would include sandpaper and shagreen sharkskin
or ray skin pumice, or vegetable reeds and fibers. The second usage of the word polish involves
Appropriate
the application of a resinous varnish or some other film-forming finish applied to the prepared surface. Finally comes the application of a maintenance coating, such as wax furniture polish or something similar whose purpose is to impart a degree of high gloss, usually temporary. In addition Roubo
uses the word polisher to mean a variety of tools, including abrasives, burnishers and applicators.
Since Roubo is not explicit in his syntax, I will attempt to distinguish amongst these many concepts
in the edited translation and make the distinction readily apparent to the reader.]
Before re-planing a veneered surface, it is necessary to begin by removing the glue that remains
on the surface, to lift [it] off with a chisel or other tools. Then one re-planes with a toothing plane
using a very light touch to avoid any tear-out of the veneer or wood fibers. One should take care to
push the plane diagonally across the wood grain, especially when encountering veneer seams like
those of ab and cb, Fig. 1, so as not to tear the veneer when the wood grain is changing direction at
the joint.
When the end of a piece of wood is perpendicular to another, it needs to have the same attention,
and push the plane from the end grain wood to the face grain, and always diagonally, as from de, Fig.
1, or from cf, which is not only the same thing but is also what one should necessarily do so as not to
Plate 29 6 . The Way to Polish Cabinetry and the Tools That are Appropriate
112
hollow the veneer more in one direction than the other and that it be perfectly flat in all directions.
It is also necessary to pay attention when the grain of the pieces are at right angles and consequently
their joints are mitered, to approach these same joints from the lowest directional angle possible and
to push the plane from gh, so to invert the wood grain on itself [to make sure you are working with
the grain in both directions] and that they serve as pressure points on one to another. This could not
W
be possible if one pushed the plane from hg, because the wood grain being taken from its greatest
length, will not have any more support and would fray at the point of the joint, especially when the
pieces of wood are soft and porous.
Plate 296
When the veneer composition is comprised of a great number of pieces, it is hardly possible to
follow exactly the method which I give here. However, one
Page 858
should deviate as little as possible. When the pieces are absolutely too small to do this, or the veneer is contoured, as in Fig.
2, one pushes the plane by turning from right to left and to avoid
approaching any piece crosswise, at least as much as is possible. In general, it is necessary to take great care not to push the
plane crosswise to the grain against wood placed diagonally or
perpendicularly, as from il, Fig. 2, because it is almost impossible to do it without splitting the pieces
at the point of joinery, or at least without making these joints larger.
As the veneer becomes smoother, one should retract the iron of the plane just until it does not bite
any more at all, and one should even change to another plane or [another] plane iron with increasingly fine grooves on the toothing blade so that the last one used should be, more accurately described,
not as a plane but rather as a type of scraper.
While one begins to re-plane the veneer, one should take great care to rub some wax on the bot-
tom of the plane so that if the glue which [is] remaining beneath the veneer becomes overheated by
the rubbing [friction] of the plane it does attach to the plane sole as little as possible. It is necessary to
re-plane the veneer until there remains no type of irregularity over the entire surface due to pieces
that are uneven in thickness and because of any marks that the first plane may have left.
The veneer being thus re-planed, one scrapes it and abrades and finishes it as I am going to in-
struct, after having made the description of the tools and ingredients one should use for polishing
cabinetry.
The proper tools for finishing and polishing cabinetry in general are the scrapers of all kinds, soft
English files, the pumice stone, dog-fish skin [sharkskin, shagreen], horsetail rushes and polishers,
[which are] simple and elaborate wood sticks for burnishing and applying polish and rubbing cloths.
Other ingredients are wax, lac, colophony, rottenstone, charcoal, olive oil and whiting.
The scrapers in Figs. 3, 4 and 5 are composed of a shaped wooden block of 3 thumbs height by about
113
horizontally, one rubs it flat on both sides with a steel burnisher [to form a wire edge, which is the
point of cutting for the scraper], which one presses hard from above in order to flatten the edges. This
being done, one passes this same piece of steel horizontally, working hard to keep it as straight as possible, which gives it a fine wire edge on both sides, so that the scraper can remove a thin and uniform
shaving, which achieves the removal of all the minute unevenness that the last toothing plane has left.
W
The scrapers are straight as in Fig. 3, or curved as in Fig. 5, which does nothing to change the way
Page 859
they are sharpened, which remains the same, as well as the method of using them, as one will see
later.
There is still another type of scraper, which does not differ from those of which I just spoke, ex-
cept that it does not cut, but on the contrary its edges are a bit rounded rather than the sharp wire edge
of the previous tools. This scraper is called the wax scraper because it serves to remove the superfluous wax after it is applied with the polishing bundle (see below).
After the scrapers, cabinetmakers sometimes use soft English files [rifflers], of which the fine
and soft size is very correct for smoothing the wood and preparing it to receive the finish. In order
to use these files comfortably, it is good that their tongue be curved, so that their sleeve [tang] is a bit
elevated above the work.
Pumice stone is another type of stone of a light and porous substance, little compacted, and filled
with an infinite number of cavities. This stone is rough to the touch, and for making use of it it is
necessary to smooth one side on a sandstone, then on a piece of smooth wood before using it on a
piece of work. It is necessary to choose the pumice stone that has the most homogeneous grain and
in which one will not meet up with hard veins, which resist more to rubbing than the rest of the
stone and which scratch the work, which is to be avoided.
ers do not use it except for the most fine parts, like shark fins that are called dog-ear
skin, which have the finest nap of all skin and which, by consequence, scratch the work
less. See Fig. 6.
The wax of cabinetmakers used for polishing is ordinarily of the yellow beeswax.
For the polish of common works like armoires, commodes of beech and walnut wood,
etc., one uses one-third tallow with the two parts wax, and in beautiful works one uses the beautiful
white wax [filtered and/or bleached beeswax], although this is not the custom.
Lac is a type of gum or wax of red color, which comes from the East Indies, and which serves
to polish the woods of color. [He is almost certainly referring to shellac or seedlac and the wax contained therein.]
Colophony [pitch] is a type of gum or resin of brown or black color, which is done [dissolved]
with fine turpentine, cooked in [a] water [bath] until it becomes solid. This gum melted with smoked
blacking [soot], serves to polish black woods, and to infill the lines of engraving.
Plate 296
114
The polisher, Figs. 8 and 9, is a sheaf of ordinary grass or straw, about 4 thumbs long by about 2
thumbs in diameter. This sheaf is bound tightly along its length. Before making use of it, one soaks
it in molten wax, which one
lets cool, after which one rubs
Page 860
Plate 296
The wood burnishers for polishing, Figs. 10, 11 and 12, are of small pieces of walnut or another
wood of a fine and closed grain, without being too hard, of about 6 thumbs of length at least, which
are of various sizes and shapes and thin to a bevel at the end. These woods serve to polish the work, or
better said, to extend the wax into the creases and tight parts in which the polishers cannot go, such
as fillet work and other little pieces where it is necessary to keep the sharp edges.
Sometimes the wood burnishers are wrapped with beaver or buffalo skins to work different pol-
Tripoli [rottenstone] is a type of chalk or soft stone of a reddish-white, and rough to the touch
[gritty], however uniform it is. One uses it for abrasion polishing by reducing the grain size to a very
fine powder by passing it through a sifter and mixing it with water, oil, tallow or vinegar, according to
the different materials that one is polishing. Good tripoli comes from Britain.
The charcoal used for polishing is that of beech from the saplings or shoots that one should choose
equally, and without any cleft or nodes [knots]. [If wished,] one also polishes the [leftover] nodes and
the roots with indigo and vinegar, which makes beautiful nuances and shapes.
One uses olive oil for a polish by mixing it with charcoal or rottenstone. However, for polishing
with oil, which is properly speaking, more of a dye than a polish, one uses linseed oil to which is
added alkanet, a type of root of a dark red on top, which tints to a beautiful vermilion color.
I will not speak here of whiting because it is not only very well known, but also because it is not
used for finishing except on certain occasions, and for removing oily fingerprints from the work.
Once the veneer is replaned, as I have explained above, one uses the scraper, which is held by
both hands and applied by holding it with the whole hand [pressed up against your
palm] while pushing it away from your body, or conversely holding it with your
fingertips as close to the blade as possible, while at the same time pulling toward
yourself. One makes use of this in two ways: one way or the other, it is necessary
that it be inclined to the direction where one is pushing or pulling it, similar to the
direction of an angle of 45. In whatever manner the pieces of veneer are placed, it
is always necessary to scrape them with the grain, observing always that the blade
of the scraper be positioned obliquely to the wood grain, as indicated by lines ab,
Page 861
cd and ef, Fig. 14, so as to better cut the wood grain. This being done, the wood
fibers are shaved more easily, given that they are not taken all together in such a
manner that the grain, which is under the cut [edge] of the scraper [and] finding some resistance and
115
being pressed by the grain on its side, is removed without bending, which would happen certainly if
one would hold the scraper perpendicular to the wood grain, especially in soft wood.
When the piece is thus scraped in one direction, one scrapes the other direction by positioning
the blade of the scraper according to the direction of the lines gc, be and dh. Afterward, one makes
the final pass of the scraper according to the direction of the wood grain in order to achieve the reW
moval of the grain that could be twisted from one side to the other. It is not necessary to push much
in making this final cut with the scraper so as not to streak the wood by removing the soft grain and
the hard grain unequally. This never fails to happen if one pushes too hard because the wood is of an
uneven density along its length, the scraper bites more in the tender areas than on the hard and forms
Plate 296
there some streaks, which one should take great care to avoid. [Actually, the reality is just the opposite, but the result is the same.]
It is partly for this reason, even more than the difficulty of cutting the grain of the wood, which
requires placing the blade of the scraper obliquely to these same grain lines, so it works at the same
time the hard and the soft parts of the wood, so that being supported by the hard parts, it does not bite
more in the soft parts.
This observation is very essential, and one cannot pay too much attention, especially when the
pieces of veneer are layered or cut in layers, and they present alternatively some hard parts and some
soft parts along the length as well as width, which makes them very difficult to scrape and to render
their surface as flat as glass, like a well-prepared work should be.
All pieces of a veneer are thus scraped separately, and when they are positioned as in Fig. 15, or in
a completely different manner, it is necessary to take great care that the scraper does not damage the
veneers at the seams where they are joined, but on the contrary that it take the last bits obliquely or
perpendicularly to themselves, as the case requires.
It is especially at the meeting of many joints, like at point i, Fig. 15, that one must take great precau-
tion so as not to damage the seams nor the wood grain. In this case, one should turn
the scraper a little to avoid damaging one or the other. That is why practice teaches
even more than the most perfect theory.
When the veneer is scraped, one passes a shagreen to remove the rest of the
bits that the scraper could not lift. Great care is needed that this skin be very soft
and smooth so it does not scratch the work. It is also necessary to pay attention to
passing the shagreen in the same direction as the scraper, and to avoid pressing as
much on the soft spots as on the hard so as not to make streaks on the work.
One passes the shagreen while looking at the work horizontally and in various
directions. One should perceive no more bits raised above the surface, and the soft
parts of this same surface should seem to have no more cavities. So [then] one passes the horsetail to
remove the little streaks that the shagreen has made, observing the same precautions as the last, beginning by pressing lightly and in reducing the pressure of the hand bit by bit, so that when finishing,
the horsetail does almost nothing but pass lightly over the surface of the wood. The horsetail is held
with the fingers for polishing plain work, and for the stringwork it is good to wrap strings of brass or
other [around the horsetail], in order to hold it straight, so as not to spoil the ridges of the work.******
****** There are cabinetmakers who, in place of the shark-
skin, pass over their work a fine and soft file, which dresses
their length. Thats why we hardly ever use them except for
Page 862
116
In the hollowed parts, like the throats and other mouldings, where fingers could not enter to press
the horsetail, one takes a wood burnisher of a shape and size necessary to be able to fit the space. Once
the work is horsetailed, and by consequence perfectly uniform, it has a surface [that is] flat [and] dull,
which absolutely changes the color of the wood, which returns and even is augmented by means of
the polish that is made in different manners, as I will describe here.
W
There are cabinetmakers who, instead of abrasion polishing with a rush, use cork, which is not
good, because the cork heats up too much with rubbing, which makes the wax almost into a molten
infusion and exposes the veneer to becoming unglued.
When the wax is well spread with the straw bundle polisher on the flat parts as well as the cavi-
ties (which are done using the wood burnishers), one removes the excess wax with the wax scraper,
which one passes over the work in order to remove it, which one then buffs/rubs next by spreading
and removing with an ordinary cotton rag or of serge, which works better because, without being too
rough, this has more bite than a regular cotton sheeting.
This operation is the last in ordinary polishing, after which, when it is as well done as any others,
When one polishes wood that is porous, or of a reddish color, like violet-ebony, amaranth or oth-
ers, when one has spread the wax over these same woods one then spreads some gum-lac in a powder
form [shellac flour], which being spread with the straw bundle polisher fills in the wood cavities
without making any stains at the same time as it augments the color.
One can also make use of powdered colophony [pitch] for dark woods, or even mix with colored
wax in powder form, similar to the color of the wood, to prevent the wax from making a tint, which,
however well-spread it is, must not be obvious. What is more, if you fear the inconvenience of yellow
wax, it is necessary to use the white, which does not stain the wood. As to the method of mixing gumlac, colophony or another ingredient with the wax for polishing the veneer, there is certainly benefit
to doing this because it is often used to mask the defects of the work, especially to the eyes of those
who do not have a perfect knowledge and understanding of this type of work.
The polish of which I just spoke is that which one makes the most use [of] for veneer because it is
used totally dry, which is necessary to avoid degluing the pieces. There are still other types of polish
that are the common polish, water-based polish, oil-based polish, whether with tripoli or charcoal.
or finally for assembled cabinetry works. It would be high-
ly desired that one could use it for all sorts of work, which
117
Common polish is not, properly speaking, in the line of veneer work since it is used normally for
the biggest pieces, like armoires, commodes, etc. However, since cabinetmakers use it sometimes for
their most common works, I believed I had to speak here of more than just part of furnishings, so as
to assemble this information in a single article that concerns all the different polishes.
Common polishing is done without any special preparation. While the work is finished and
W
scraped the best possible, without however taking all the precautions of which I spoke up above,
one rubs it with wax that is normally combined with one third tallow, and one spreads this wax with
a brush that is a bit stiff, then one wipes the work with a serge cloth. To more correctly spread the
wax and make it fill the grain better, one sometimes uses a type of sheet-metal pan into which one
Plate 296
puts burning coals, and which one passes over the work as close as possible so as to better heat the
wax. Instead of this pan, one can use a piece of very
hot iron much like illustrated in Plate 294, Figs. 89,
even
almost
red-hot,
If instead of using mixed wax, one uses good wax, and even white wax, one could, in following
this method, make a polish admirable for both its durability and its beauty, understanding always that
one would not use this except on a piece made only from solid wood, in order that the action of the
red iron, which one passes over it, would not cause any harm. I have polished works of walnut wood
in this manner, which became as beautiful and lustrous as glass.
The water-based abrasive polish is used in the following manner: After the work has been scraped,
one takes a piece of well-smoothed pumice stone, which one soaks lightly in water. With this one
rubs the work in all possible directions, observing always to finish along the grain. One continues
the operation just until the work is perfectly smooth, which cannot be done except by repeating the
operation different times, and after having let the work dry thoroughly each time.
When the work is abraded by pumice, one rubs it with horsetail next and finishes it with ordinary
wax, or even one uses oil with a cloth or a sponge, which provides more a color than a wax coating.
This tint, which is composed of linseed oil and alkanet [a dye root], gives a brownish color to the
wood, and which becomes glossy over time.
In general, the water-based polish with pumice is hardly appropriate except for works of little
consequence, and which are made of opposing-grain or knotty wood, such as rustic walnut, twisted
mahogany and other similar woods. What is more, it still leaves a type of deposit in the pores of the
wood, which gives a bad effect and which is a real pain to get out. One cannot make use of this polish
except for works made entirely out of solid wood, moisture being contrary to works of veneer.
One also uses pumice stone dry for polishing white woods, like pine, poplar and linden, which
While one does not polish cabinetry with ordinary wax, and when the work is of enough conse-
quence to make the expense of a more beautiful polish, after having horsetailed it, one polishes [it]
with rottenstone, reduced to a powder, passed through a silk sifter, ground up with a bit of olive oil of
the best quality, which one spreads over the work with an ordinary straw bundle polisher and which
one wipes later with a rubbing cloth. Then one takes a wood burnisher (wrapped with a bit of buffalo
Page 864
118
or beaver skin), and begins to polish the work while spreading a little bit of whiting, to remove stains,
and one wipes it again with a linen rag, smooth and clean.
The rottenstone is used differently, according to the different types of wood. One grinds it with
olive oil for ebony, red wood and knotty boxwood, with some tallow for red wood and with water for
Wood from China.
Ebony is also polished with coal oil [a heavy kerosene], whether with powder charcoal or with a
whole block of charcoal, which one shapes well at the end [of the block], that is to say, making the end
smooth and straight. These different types of polishing always end with the buffalo polisher and a bit
Plate 296
of whiting.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Finishing Marquetry
rom the perspective of a finisher with more than 40 years of experience, the finishing room of an
18th-century atelier is a vaguely familiar place with some peculiar tools, finishes and practices. As with
most of his volumes, Roubo tended to assume certain knowledge on the part of the reader and thus his treatise is often incomplete and most certainly can be confusing (especially when you attempt to read it as a strict
linear narrative) if you do not have broad experience with the varnishers trade.
His terseness is frustrating. While he recites an excellent account of the materials used in the finishing
process beginning with the preparation of the wood through the application and (almost) final detailing of the
surface, the reader would be well forgiven if the response is, Yes, but how do I actually use these materials?
Here are a few thoughts on the finishing process for the bnistes atelier, along with some contemporary
functional analogs. These comments might not always correlate exactly with the original text, but I have attempted to integrate Roubos descriptions with those of other historical observers and emerge with an understanding of the trade in the 1760s such that a contemporary artisan might replicate the process conceptually,
if not literally.
First, the surface must be properly prepared for the application of the polish/finish. This begins with re-
moving excess glue remaining on the surface from the assembly, followed by a toothing plane using a very
light touch to carefully smooth out the multiple marquetry elements (one cannot pay too much attention),
followed by a finer toothing plane, then a wire-edged scraper or even a fine file to make sure everything is
level.
instantaneous.
119
Most colors of wood, whether from the Indies or from France or if they are dyed wood, lose
their luster over time. Since it is very important to preserve these colors, one cannot do better after
having finished them with horsetail and pumice, or with whiting, than to varnish them with a white
varnish called commonly Venetian varnish. This varnish of which I am speaking is little different
from the last one other than it is white, or better said, without any color, making it appropriate for
Page 865
otherwise brittle]
Plate 296
All melted in double boiler without letting it boil. When this varnish is cooled, one filters it through
cotton so that no solid residue remains.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
At this point a number of abrasive materials can be employed in a manner to accomplish that for which a
21st-century finisher uses sandpaper. Included in this family of tools for lightly abrading the wood surface one
might include a block of fine pumice that is rubbed over the wood much like a modern sanding block. This
was especially useful for smoothing long stretches of architectural mouldings as the block could be shaped to
conform to the mouldings contours. A pounce of linen filled with pumice powder may also be used, with or
without any lubricant, with special care to avoid moisture that might delaminate the veneer. A solid block of
pumice was more widely used for smoothing solid wood. A piece of sharkskin or rayskin might also be employed, provided it is fine and supple enough. Because I do not have a steady source for unprocessed shagreen
I have used a piece of chamois onto which I have glued a powdered abrasive using glycerine-laced hide glue.
Finally, some split horsetail rush held by fingers or wrapped around sanding sticks, much like an emery board,
serves well for small spaces.
Interjecting my own practice, experience and common sense, I find that briskly scouring the surface with
a dry (unwaxed) bundled polissoir (see below) at this point infuses it with astonishing brilliance and serves as
an excellent foundation for the application of subsequent coating materials.
followed next.
mirrored surface.
120
One can put many layers of this varnish on works of cabinetry without fearing that it will obscure
the colors, paying attention not to put on the second application before the first is perfectly well dry.
When one has put on two, four or six layers of varnish, and it is all perfectly dry, one polishes the
whole with a pad made of the edges of rolled cloth, or with some buffalo, onto which one puts a little
rotten-stone dampened with water. Then one washes the whole with clear water, and one wipes with
white and clean sheeting.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Plate 296
Surprising to most, me included at the time of first learning, the bnistes application of finishes often
begins with a grain-filling primer of solid beeswax. The surface is rubbed with a chunk of beeswax until there
is a generous deposition, which is maneuvered around the surface and into the grain pores with a polissoir until
the pores are completely filled. Excess wax is gently scraped off with dull metal or wooden scrapers, usually
made from a hard wood such as boxwood, walnut or apple, followed by buffing with a soft, coarsely woven
fabric.
For dark or reddish woods with open grain, the process includes the addition of pulverized seedlac or even
On solid wood, Roubo asserts that rubbing the surface with a tallow/beeswax paste (common polish),
followed by melting the paste into the grain with hot coals or a large heated iron, completed by buffing with a
brush, yields a surface as lustrous as glass. Our use of commercial paste wax mixtures suffices for this, and
for other uses as detailed below.
Roubo acknowledges that his is a brief and incomplete presentation of the available finishing options. It
is possible that a fine surface could be accomplished by the simple buffing of wax then considered finished. A
more plausible recitation entirely in keeping with the narrative is that in order to accomplish the mirror-like
surfaces that were preferred for these magnificent pieces, one final step was necessary. I remain convinced
that for whatever reason this final step of the wax finishing process, for fancy marquetry at least, has been
omitted: The spit polishing of the surface.
Using a balled pad of soft damp fabric and a dab of the common polish (our commercial paste wax) the
surface could be rubbed repeatedly, with the cloth pad being recharged with moisture and paste wax as necessary, and polished until it was a mirror in short order. We might think it sounds like French polishing with
shellac. But at the conclusion of the later chapter on Boulle-work, Roubo refers to the formulation of a shellac
and amber spirit varnish that he calls English polish as one to be used only as a baked-on coating for bronze
hardware.
In short order and with a little effort, the true and traditional wax-based French polishing yields glorious
results if it is done well. It is eye-catchingly noteworthy, as anyone who has ever admired military dress footwear can attest. One aspect about Roubos wax-based polishing that I find especially pleasing, as compared
to the shellac-based spirit varnish polishing (which is captivating in its own brilliance), is that the wax system
seems to emphasize the character and texture of the underlying
surface. Shellac polishing tends to homogenize the surface. Neither result is better. They are just different. Both yield brilliant, appealing surfaces that are a feast for the senses.
revolves around the uses of pigmented oily polishes and oil/spirit varnish concoctions. The concern about the fading of brilliantly colored exotic woods was the motivating force behind these
efforts.
Fig. 7.7 Rather than obscuring the textured nature of the substrate, a wax polish surface enhances the character.
121
This method of finishing cabinetry is a little costly and more subjective than others. Also it has
the advantage of being the most perfect because the varnish, in blocking all the wood pores, holds
the color, which, not being able to oxidize, remains always in the same state. This is of a very big advantage given that the liveliness of these same colors in great part composes the beauty of works of
marquetry, of whatever type they may be. It is the only means that I know of for giving brilliance to
W
and an oil-of-lavender mixture suspended in alcohol. It was allowed to dry thoroughly between each appli-
Plate 296
cation. The method of application is not mentioned. After all is done and dry, the surfaces are abraded with
a rolled cloth or some buffalo hide bearing a rottenstone/water slurry. While this would no doubt impart a
brilliant surface, Roubos promise that this method would prevent the fading of the woods underneath was,
sadly, false.
Make and Use a Straw Polissoir. One influential tool that was unknown to me at the start of this project is a polissoir (polisher), which I call a corn straw burnisher. Roubo offers fewer than 100 words describing
the tool and its use, yet that tool has fundamentally changed parts of the
way I work. As the last tool to touch the surface prior to the application
of finish, or in some instances the tool that actually applies the finish, vigorously scouring the surface with it imparts radiance to the substrate that
cannot be adequately described. It must be experienced.
Fabricating your own burnisher is fairly easy, and the raw materials
You will need a hank of straws between 1" to 2" in diameter. Take your
bundle of straws and bind them together with several hose clamps of the
appropriate size side-by-side, leaving about 1/2" of straw sticking out at one end. Leave a little gap between
the hose clamps about halfway down the length of the bundle. At this opening, wick in a copious amount of
glue all around the circumference and let it sit overnight. Any glue is fine.
Then take a string and tie a loop at one end of the cord. I simply double up the end and tie it into a knot
leaving 3" or 4" inches of tail. Make a noose from the loop and the string leading to the ball. Place this about 1/8"
from the end of the straw bundle, cinch the noose and start wrapping it as tight as you can without breaking
the string. Remove the hose clamps as you work your way down the bundle. When you get to the other end of
the bundle (about 1/8" shy of the end), hold the string in place with a spring clamp. Soak the string wrapping
with dilute hide glue and let it set until dry. You can skip this gluing step if your burnisher will be cooked in
molten beeswax as the final step. Then reverse the direction of your wrapping (same rotation but now you
122
Different polishes are applicable not only to the types of veneer work of which I just spoke, but
even to all the other types which I am going to address later, except for those types of marquetry
where one uses tortoiseshell, mother-of-pearl and the metals, for which polish is made in a different
manner, as I will speak of this later.
W
Plate 296
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
are working back toward your starting point) to return to the starting point. When you get there, tie off the
wrapping string with the tail I mentioned earlier.
Soak the whole string surface with white glue or cross-linked hide glue and let it sit. Trim the straw bundle
ends as needed. If your goal is to make a dry burnisher, you are done. If you want a wax-impregnated burnisher, melt some wax in an appropriate vessel and allow the wrapped bundle to soak in it until it is fully saturated.
Remove the burnisher from the molten wax with appropriate caution and wipe the excess wax off. As soon
as it cools to hardness it is ready to be put to work.
Fortunately I have located an artisan broom maker who fabricates my polissoirs for me, so I do not have to
Work the tool as it is illustrated in Plate 296, Figure 13, pressing it down perpendicular to the surface and
scrubbing back and forth with the grain. The stiff bristles of the straw compress and smooth the wood fibers,
bringing the surface to a high glistening sheen without undue effort. However, as with almost all processes
related to the decorative surface, any one step can only be as good at moving toward the successful completion
as the step previous. There are no shortcuts.
The effect of the polissoir on the prepared surface is simply astounding. In Fig. 7.13, the surface is one-third
hand planed (bottom third), one-third dry burnished (center section) and one-third wax filled with a polissoir
(top).
he type of cabinetry that I am going to treat here is the most precious of all for
Chapter 12
both the delicacy of work and for the knowledge that is required on the part of the
worker. As I said at the beginning of this section, the cabinetmaker should well understand design of all kinds and possess knowledge of the tone and nuance of colors
in order to represent as perfectly as possible all sorts of subjects, like buildings in
perspective, ornaments, flowers, landscapes, figures, etc. These subjects are rendered entirely of
wood in keeping with natural color or tints, inlaid on a cabinetry base, whether of veneered wood
or solid wood, which are treated equally as far as the manner of working, or at least very close to,
as one will see next.
The big difficulty of this type of cabinetry consists not only in the arts of dcoupage and inlaying
wood for representing different subjects as well as possible, but also in the art of giving to each piece
the shading that is necessary. This is done by the means of fire or acids, and which requires much attention and experience to do well.
To give the description of this part of cabinetry [marquetry] the proper coverage, and at the same
time avoid repetition, I am going to content myself with giving the most general and the most necessary examples of different ways of operating, which could be applied to all cases. Besides being
very general in the descriptions being presented here, recognizing that the worker will know which
type of furniture or other type of woodworking is being created, these examples will be applicable
to whichever, since one can use them equally with all forms and types according to the taste of the
artist. It is better said that the general design of the piece to be covered with mosaic or marquetry, and
the expense that one wishes to make, is in fact considerably more expensive than simpler furniture
described earlier [Note: Roubo is referring to an earlier volume that is not in this translated work].
Because of this substantial increase in cost there is no allowance for any type of mediocrity in its execution, which makes it very costly when being very well made.*
* Nothing is so common as to see veneered pieces of cabinetry
made. This is due less to the fault of the workers than that
lation to the theory and practice of the art. The former [the-
of the collectors, who for the most part are without taste and
and the ornamentation with mosaics. For the most part these
and made for the space that they occupy. [All too often these
123
Page 8 66
124
Page 867
While I have treated summarily the different types of veneered cabinetry on page 765, I have
put the topic here in a particular designation as being absolutely different from the two other types
in order to give more order and clarity to my work. However, it is rare at present for mosaics [marquetry] to be used all alone, that is to say, on a wooden foundation alone. On the contrary, one uses
it almost always with cabinetry of the first type, that is to say, veneered marquetry, upon which one
W
inlays it, which seems to unify the two first types of cabinetry, and to make but one, whether they be
different one to the other, as much in the theory as in practice.
Section I
E l e m e n t a ry
pr i n c i pl e s o f
P e r s pe c t i v e ,
which
k now l e d g e i s a b s o lu t e ly n e c e ss a ry
fo r
Cabinetmakers
F all the knowledge necessary for cabinetmakers, that of the rules of perspective is one of the
most essential. However, it is the most often neglected discipline, nothing being more common than to see work that often lacks merit where the rules of perspective are absolutely violated,
whether in the design that the cabinetmaker compose themselves, or in those that they copy, which
they often render badly (supposing that they are well made in the first place), and which lack the
knowledge of these principles that have served to put these designs in perspective. To eliminate these
difficulties, and to facilitate the cabinetry of the makers for which this work is particularly written, I
have not been able to dispense with the notion of this science, all of which is demonstrated mechanically, without going into each detail of the optical rules and other parts of mathematics, which is
necessary for the perfect theory of perspective but which cabinetmakers can dispense with [when
working at the bench].
al] the different objects that strike our view, like buildings, landscapes and even men and animals,
Page 868
not like each of these different objects are exactly (which is the object of geometric designs), but as
to why and how they appear in our view by reason of the distance between us and them, and of the
means by which they are presented to us where they diminish more or less in length, and seem at the
Plate 297
Elements of
Perspective Necessary
for Cabinetmakers
same time to change in form. This difference is an optical effect, knowledge of the cause of which is
not necessary here.
We distinguish three types of perspective: that of lines, that of planes and that of bodies; which,
all three, are executed in the same manner, which is totally natural, since they all depend on the same
principles, as I will demonstrate here and now.
catalogs or brokers in marquetry appliqus. If the fit was not
Principles of Perspective
127
When we look at any object, the space that our view embraces forms a circle of which our eye
is the center, such that we cannot perceive anything but the objects visible within. The circle side
aa, Figs. 1 and 2, is called a visual circle and its center b, point of view, toward which point all
objects tend when they are presented perpendicularly to us, that is to say, at the base of which we are
positioned. Point b is called the horizon point because line cd,
W
Plate 297
When one wishes to put an object in perspective, one begins by tracing a geometric plane below
the land line, as for example line hi, Fig. 5, which one lengthens just to the land line that it meets at
point n. From this point one takes a line to point of view b, which gives first of all a visible slant to the
line that one wishes to put in perspective. Then from points hi, which are the extremities, one leads
two lines to the transposed distance line, which come to cut line nb, Fig. 1, to points op, giving the
length to the line in perspective. If one wishes to put line lm in perspective, parallel to that of hi,
Fig. 5, one would effect as for the latter, and one would have line qr in perspective, Fig. 1. If instead of
perpendicular lines hi and lm, Fig. 5, one would like to put in perspective the horizontal lines of the
same figure, one would always use the same method as for perpendicular lines, that is to say, that after
having raised up on the ends of the horizontal lines the perpendicular lines in and ls, points n and s,
one leads two lines to the viewpoint b; then from points i and l, one leads two others to the transposed
distance point, which cut the lines leading to the point of view b, and give points or, by which one
passes a line that is parallel to the land line, which is il, Fig. 5, viewed in perspective, which differs
only by its length.
Line hm, put in the same perspective, in leading from the ends hm, two lines to the transposed
distance point, which gives Fig. 1, horizontal line pq, which is shorter than that of or, the reason
being that line hm, Fig. 5 is more elongated than the land line il.
If it happens that lines il and hm, whether still parallel between them, are of an unequal length,
or whether of equal length, they are not placed below one another, as for example, that of ut, one
always uses the same method for putting them in perspective, noting to lead as many perpendicular
lines to the land line, and from there to the viewpoint, where the ends of lines will be, which is very
easy to imagine.
Once one puts perpendicular lines and horizontal lines in perspective, it is easy to quarter them,
since they are not formed except by combination of these two types of lines, and that by demonstrating the manner of putting them in perspective, I have right away constructed a square, as you can see
Page 869
128
Plate 297
in Fig. 1. That is why I will not talk anymore about this, given that it would be just a repetition of what
I already said.
[Note: end of text from page 869, jump to plate 299, page 878 Editor.]
S e c t i o n II
On
Page 878
Ornaments
o f wo o d
HATEVER kind of ornament the mosaic is, it first requires the design to be made on paper
as well as possible, as in Fig. 1, observing all shading necessary, according to how these parts
Plate 299
should be placed, whether on the right or the left of the work, or even according to how it will be
illuminated, assuming that it is going to remain in that place, as for example, the design elements in
The Way to
an apartment, or even furniture that is made expressly to occupy one place from which it will not
Prepare Marquetry
leave, which is essential to note. There is nothing more ridiculous than seeing an ornament that,
Ornaments
being made to appear in relief, is shaded from the side where it is illuminated, which often happens
when you do not take the precaution that I am recommending here.
When the ornaments are thus designed on paper, one takes the tracing and transfers it to wood,
which is done in many ways. The most common, and that which is used by cabinetmakers, is to lay
another piece of blank paper over the design [and] then to put them together with light coming from
below. This is done by putting them on a sheet of glass placed vertically to daylight, or even on a
square windowpane, provided that the lines of the design can be perceived through the paper placed
over the design. Then one takes a pencil and traces exactly on the blank paper all the lines of the design. This method has no other fault than being a bit inconvenient, given that it requires you to design
this work vertically, and even, if possible, a bit inclined toward you so that the light hits better from
above, which makes the worker who is obliged to make the design very tired. That is why I believe it
would be better to take the tracings in the manner of the engravers; that is, by applying on the design
some waxed or varnished paper, or even tissue paper, through which one can read all the lines of the
design, which one then traces by pen with a hardened China black pencil on the oiled paper; [or]
with a fine tip [that is] a bit rounded at the end for the glazed paper, and with Chinese ink quill or
pencil for the blank paper.
One can also take the tracing of a design by rubbing the back of the paper with the original design
with a supply of lead [graphite powder], or with pulverized red chalk, and by placing the design lightly on top of a piece of blank paper. Then one takes a fine point, with which one traces over all the lines
of the design. By this method, the design is transferred to the blank paper that is beneath the original
Page 879
design, and which one marks again, whether in ink or pencil, to ensure the lines that one has made.
This last type of tracing is applied and is stuck on the wood that one wishes to cut up, which is done
in the ordinary way, as I will describe later.
When one has many pieces of ornament of a similar form, as for example, the rose represented
in Fig. 1, one is therefore obliged to take many tracings to glue on the wood. When the item is not
of a very great consequence, one makes a tracing on paper a bit stronger, which one rubs with red
chalk on the back, and which one transfers to the wood with an ordinary point, or one even pricks
130
this same tracing with a pin, by following all the contours as closely as possible, then with a little
fabric sachet filled with fine lead [graphite powder] or pulverized red chalk, one rubs on the tracing,
which being placed on the wood, leaves there a tracing of dots, which indicate the shape of the piece
to be cut out.
This last method is only good for large pieces. For smaller ones, it is better to make as many tracW
ings as one has similar pieces to make, which makes the execution much more perfect.
In this case, one can still use a pattern of tin plate or very fine copper shaped in the same manner
as the pieces that one wished to cut out. This is much more accurate than the paper tracings, which
Plate 299
are very difficult to execute identically, especially when one has a great
number of similar shape. This should be the preferred method that I
gave here, which is not more expensive than that which one ordinarily
uses because if one loses on the side of the price of materials one gains
on the side of saving time which is a major consideration and the
great perfection that one makes in the work. Whats more, the dice or
patterns thus made of tin plate or of very thin copper can be used repeatedly without suffering any alteration in their shapes. This double
advantage makes them preferable to paper tracings at least in the case where one has a large number
of similar pieces to cut out, as with the roses represented in Fig. 1 or any other of this type.
One should, however, note that these models, whether in tin plate or in copper, cannot work ex-
cept for exterior contours of pieces, and that the interior contours are traced ordinarily, at least when
one does not wish to make as many models when each piece would be composed of different parts,
which cannot happen except in the pieces of a certain size.
Ornaments in general are regular [symmetrical or rectangular] like those in this plate, or are ir-
regular [asymmetrical or curvilinear]. In the first case, one must take great care that all the parts
which compose them be perfectly similar and equal to each other, at
least each one with its pattern, which one can do easily by measuring
from the design the dimensions and the distances between each part,
and by indicating them by straight and horizontal [layout or projection] lines. [When laying out elements that are symmetrical in relaPage 880
The mosaic ornaments being thus traced and transported to the piece of veneer destined for this
use, it is necessary to cut them out with the marquetry saw of which I gave the description above and
the manner of using it, on page 843 etc.
When one cuts out ornaments, it is necessary to take great care to follow the contours as per-
fectly as possible, the interior as well as the exterior, and one should also observe to cut out all the
protruding parts and to detach them one from the other, so as to be able to shade them as one judges
appropriate. See Fig. 3, which represents the rose design of Fig. 1, all cut out and the pieces separated
one from the other in order to prepare them for the fire and to give the shading, as is indicated in
the design of Fig. 1.
131
When the details are too small to be achieved by separating the elements by sawing, as with the
sides of the four leaves A, B, C, D, Fig. 3, one does not split them with a saw but instead you should
accentuate them with the engraving tool, as I will teach next. However,
it is much better to saw them, at least as long as they are not really too
small to be sawn, because it is always easier to shade them when they
W
are not so small. When the shading is done, whether by fire or by means
of acids, the shades are much softer and more natural than those done
with the engraving tool.
Plate 299
as for the circular pieces of which I already spoke, on page 842 etc., and
one should take great care in cutting out all the exterior contours a bit on a slant [bevel cutting], so
that they crowd a bit when you put them in place, and consequently they line up better. What is more,
this slant improves the exterior surface of the ornaments by compressing all the joints when you
put them in place, such that they are not visible except in the difference in color or of shading of the
pieces, of which they are composed. This is necessary to give the work all the perfection that can be
accomplished.
When one cuts out the ornaments or other pieces of mosaic, it is necessary to take great care to
handle the saw in the manner whereby it can cut out the greatest number of parts possible without
being obliged to remove the saw from the piece you are cutting. This is always easy to do, at least if you
just pay a little attention. One should set each cut-out piece aside, close by and in a clean location, always according to the place that it occupies, as one can see in Fig. 3. This being done, one reassembles
them one after the other, and one places them on a paper coated with a bit of clear glue to keep them
together and keep [retain] all the parts, which often being very small are likely to be misplaced.
This observation is essential, especially in the case of similar ornaments, such as those that are
represented in this plate, which have many such pieces. It is very important not to change their places, given that whatever precaution that one takes in cutting them out it is hardly possible to make
them identical and it is consequently necessary to take great care against their changing places.
What is more, as these sorts of ornaments are hardly cut out singly [it is rare to need only one of
any particular element], it is to the advantage of the worker to cut out right away the number that he
needs. It is therefore necessary to take these precautions so that these ornaments neither be exposed
to loss nor confused one with the other.
There are workers who, instead of sticking the cut pieces flat on some paper, are contented to
stick them down immediately with clear glue, placed one distance to the next, so that the pieces hold
together one to the other [in the completed composition]. This is less good than the first method that
I just explained. In order to stick the pieces of ornaments right away, the glue, however clear, should
be consistent and not leave a thickness that fills the joint, and is necessary to remove before inlaying
the ornaments. [Individual pieces are glued down so that they do not get lost, then have to be unglued
in order to shade them then reassemble and reglue them to the carcase.] This is subject to making
splinters when one wishes to unglue them for shading next, as I am going to explain.
The ornaments and other parts of the mosaic are shaded in two manners: that is, with fire (or
better-said, hot sand) or with acidic liquids. The first method is the most used, and that with which
I will begin, and though simple [it] requires much attention and experience on the part of the artist,
which cannot be acquired except by long practice given the differences that are encountered, in both
Page 881
132
the colors and in the softness or hardness of the woods used. I will present here only the general rules
of shading wood, which is perhaps easier to do well than to
describe given that it is an exercise purely of experience and
practice.
fine river sand which one puts in an iron pan, and which
one heats on a stove just until the heat of the sand is capable
of browning the wood without burning it. One knows this
Plate 299
Page 882
Since the pieces to shade are often too small to be able to hold them with your fingers without
burning them, you take them with a pick, Fig. 9, and you press the pick in the
middle to handle the piece, which is a little inconvenient because the pieces
can turn and even fall in the sand, which exposes them to burning, or at least
to be shaded where you do not wish or in an opposite direction. That is why
I believe that instead of a pointed pick, Fig. 9, it would be better to use a pincers or forceps of iron, represented in Fig. 8, with which one could hold the
pieces to be shaded, however small they be, without fear that they turn nor
that they will fall. This cannot happen with forceps, even if one opens the
hand with which one holds the forceps, as the tines of this last cannot move
themselves since they are held by a spring EF, which you must compress to
make them open.
Although the use of these types of pincers is more convenient, the wood-
Woods are not shaded on the first brief submersion in the hot sand, but rather by plunging them
in several repetitions in the sand, observing to thrust them more or less deeply, as one wishes to force
the shading closer to the edge of the piece, which otherwise would shade itself naturally when softening [the gradations toward the center of the piece].
As one continues to shade a piece it is necessary to reposition it in its place within the composi-
tion so that it not be separated from other elements of the composition in order to determine if it is
shaded to the degree that is appropriate. This determination is made by a comparison with the design
of the work and the other pieces already shaded, so that they are perfectly in accord between each
other and the design. It is necessary that they imitate the desired effect as much as possible, which is
not easy to do well without much patience and precaution, and above all of that, without the help of
an accomplished practice and a very great understanding of the design.
One also used acids to shade wood, as I said above. These acids are lime water [at first glance it
133
seems as though he is describing citric acid, but the larger context later on indicates the chemical is
calcium oxide, which is in fact not an acid but an alkali], in which one puts some corrosive sublimate
to strengthen the force. One uses also spirits of nitre [nitric acid] and oil of sulfur [sulfuric acid]. Of
these three ingredients, spirits of nitre is that which makes the best effect on the wood. It is necessary
to take care to shade before dyeing, because the spirits of nitre totally destroys dye colors. This acid
W
gives to white wood a rosy color and penetrates them in an instant. The action of oil of sulfur is less
violent: it gives to white woods a wine-brown tint, and augments certain colors instead of destroying
them. That is why in most cases one can use it after having tinted the wood.
It is necessary to take great care and not to use but a very small quantity in using one or the other
Plate 299
of these two acids because they spread out a lot [via capillary action], especially the spirits of nitre,
of which 3 lines diameter extends at least to 6, [and] which makes [can extend] close to three times
the area.
Lime water, although less violent than the two chemicals of which I spoke above, leaves only some
burned wood, whether white or colored, and I would prefer the last, especially for the larger pieces.
In general, the method of burning wood by means of acids is not used anymore at present, at least
by most cabinetmakers, and I do not know why. Their use is very convenient, especially in the parts
surrounded by lightness [for example, when you are shading the middle section of a piece of work].
These cannot be easily shaded with fire, at least when cutting out the spaces to be shaded, which is
not always possible.
As to the manner of using acids, it is very simple: it concerns only placing with a brush or the
end of pen, the quantity sufficient on the wood that one wishes to burn. To begin the operation
as much as one judges appropriate, which is an affair purely of experience, and for which one can
hardly give certain rules, the liquid is increased or decreased in the appropriate area according to
the required case. That is why I am content to indicate here the name and the usage of appropriate
chemicals for browning wood, leaving it to the artists to direct the proper use and different occasions
for doing it, which are so varied that one cannot describe each circumstance. Any attempt to do so
would become cumbersome without being particularly useful.
As to the chemicals for burning the wood, they sell them all ready at the druggists under the
names that I have indicated. It is only the lime water that one must make oneself, which is very easy
since it concerns only mixing quicklime in ordinary water and taking the liquid that surfaces when
the lime is diluted.
When all the pieces are thus shaded, one puts them in their place, gluing them all together on the
facing side (face down, that is to say, the side where they were traced and cut out) on a piece of paper,
and one lets them dry to be able to inlay them in place where they are designated, which is done in the
following manner.
Page 883
134
position this piece in the place where it should be, and you trace a line around its perimeter with a
very fine point, observing to follow closely the contours of the piece to be inlaid. Then you make the
excavation for the inlay with the trimming [shoulder] knife and the other tools appropriate for this
work, as I have taught it on page 832, etc., in speaking of inlay of pieces of marquetry, straight as well
as circular.
When the space that the ornaments should occupy is completely empty, as represented in Fig. 5,
one positions the piece to be inlaid, Fig. 4, to see if it fits easily, and one finishes by adjusting it, assuming that there is something to be done. That being done, one coats the inlay with glue on the underside,
Plate 299
as well as in the cavity that should receive it, and one puts it in place. Do not put on too much glue and
place the inlay such that there is a little daylight around the edges so that the excess glue can squeeze
Page 884
out easily. Afterward, one finishes seating it by tapping overall lightly above, with the head of a veneer
hammer. When the piece that one is gluing is of a large size and composed of many pieces, you must
take care to take a solid piece of wood that covers it entirely, on which one hits with the hammer so that
it not be exposed to rupture, but on the contrary it fills all the space, and that all the joints close up at
the same time. This happens automatically when it is well adjusted, and the contours were cut out on
an angle as I recommended above, page 880.
When the piece is being glued, one lays it in the cavity with the hammer to see if it fits well over
all, and one coats with a bit of clear glue overall including the face, so the moisture from below does
not cause it to curl [from unequal dampness]. Afterward, you cover the work with a slightly damp
rag, and you let it dry in a location without too much wind or humidity. Then you remove the clear
glue and any paper facing and you smooth the whole work, as I taught above, page 857, observing
however to polish only the parts that should be polished, which is done in different manners, as I am
going to teach in the following paragraph.
10, 11 and 12. It is a piece of steel about 34 thumbs in length, by 2 lines wide at
the most, and a little more than a line of thickness on the back side, that is to say
up above, tapering to nothing below. The shape or cross-section forms a very
elongated triangle, as represented in Fig. 13.
The middle of the thickness of this burin is empty along almost all its inner
length, parallel to its sides. Its upper extremity a, Fig. 10, is cut on a chamfer,
from ab, such that the cutting edge of the tool along side bc, although very
acute, cuts the wood very finely, and without making any splinters, given that
the position of the chamfer ab cuts as much at the sides as at the point, and
as little as the latter enters in the wood, makes the two sides of the tool make
135
the first cut. The shaving passes between the two sides, and becomes more or less thick, according
to how far one forces the tool down into the wood.
Page 885
dle], Plate 299 Fig. 10, of which the bottom is flattened, so that
one can hold the burin as close to the wood as possible, and to
W
Plate 299
at the thickness of the handle on this side, gives the worker the
ability to enter in the wood as little and as much as he judges
appropriate in order to make the engraved line fine or bold.
The burin is held in the right hand while the entire length of
The burin is held like this when one wishes to make straight lines, whether perpendicular or
horizontal. When one wishes to make contours of a very small circle, one slides the index finger
Plate 300
The Way to
Engrave Marquetry
Ornaments
One uses a burin to indicate the little details that one cannot do with a marquetry saw, and to
form the shading by means of lines or hatches. In the latter, the lines are made in two different
ways, namely that of a single direction, as in Fig. 13, or even with two directions, as in Fig. 14. The
engraving with a single direction is the most beautiful and is done with parallel lines of the burin,
which one makes deep or shallow, long or short, accordingly as one wishes the shading be more or
less bold along the length or along the width, as one can see in Fig. 13.
Engraving in two directions is done in the same way as the preceding one, with the exception
that one puts [makes] the lines with the burin in the form of cross-hatching, resulting in tiny diamond-shaped lozenges, which does not make a particularly nice effect in the type of engraving that
is concerned here, where it is good that all the lines follow the direction of the pieces that one is
engraving. That is why one would do well to engrave all the works with a single direction, as I have
Page 886
137
observed in different examples, which I gave in Plates 299, 300, 301, etc. This is the most normal,
which the engraving on marquetry does not serve to make the shading, but only to augment it, which
requires less black and can consequently dispense with requiring a second pass.
There are occasions where, instead of a burin, one can use a point [of] a bit like that used by wood
engravers. This tool, represented in Figs. 5, 6 and 8, is made of a [wooden] sleeve 45 thumbs long,
W
split in two along its thickness of about 2 thumbs in length, as that of EF, Fig. 6, to place blade GH
between the two [note that G is not shown on the plate], and to be able to push and pull as one judges
appropriate.
Blade H is nothing but a piece of very thin steel, which one guides from the end, and one holds
Plate 3 00
by the handle, by closing the latter with a string that wraps around it as one can see in Fig. 5. For this
blade not to move in the handle you make a little relief on both sides of the handle, Fig. 7, into which
one places the blade to provide solid support to the back edge of the blade, Fig. 8, which as I just said,
is very thin, often being made with springs from watches or from little pendulum clocks. Instead of
using springs, there are cabinetmakers who use old darts onto which they attach a handle made from
a small piece of wood, in order to use them to make extremely delicate inlay, or for engraving their
works.
When one engraves with a point, one holds the latter in the right hand, a bit like the way that one
holds a pen, and one makes the cut by pulling the point toward oneself, as one can see in Fig. 2 and
Fig. 4, side C. This cut is nothing but a simple line pressed into the wood, which one enlarges then by
passing the point in the same line many times, and by leaning the tool slightly to the right or the left to
cut the wood on the side of the line.
When engraving with a point, one places the work in front in the same way as for engraving with
an engraving pen, with the exception that the left hand is more extended on the work, as one can see
in Fig. 4, Side D.
Engraving with a point is very useful for pieces that are very delicate or when one fears that the
burin will make splinters, or that it will knock loose a small piece of the inlay. The rest depends mostly
on the taste and habits of the worker, who uses one or another of these tools as he judges appropriate.
I am not going to extend myself more on the engraving of wood marquetry, given that the theory
of this part of embellishing woodwork owes much to practice, which varies according to different
situations. Each artist adopts his own way of doing this, which fundamentally is immaterial so long
as the work be well made. I have hardly given the general rules addressing engraving, which will be,
I believe, sufficient to give a precise and fair idea of this part of marquetry cabinetry, in which it is
not possible to excel without having a great [deal of] practice in different types of design, which will
serve well to direct and determine the quantity and size of engraving lines. All that I could say on this
subject, on which one can hardly give but general rules, is what I have done.
When you have finished engraving whatever piece, you trim the edges, that is to say, you remove
all the burrs produced by the burin on the edges of the cuts. This is done with the scraper or with
the paring chisel. Then one fills the cuts with mastic, whether black, brown or another color, which
one keeps warm in a vase like that of I, Fig. 15, which one applies or, better said, one introduces
into the cuts with a wooden spatula. You let it dry and scrape it again afterward, then inspect each
piece to see if you missed any and if the engraving has made the desired effect. If you have missed
making some cuts, one completes them and fills them with pigmented mastic as done ordinarily.
This being done, one finishes by polishing the work, which is done with much care and precaution
Page 887
138
given the differences in wood grain, and the many parts of which it is composed. This observation
is essential, not only for the polishing of these sorts of works, but also when one begins to plane
them level, after they have been inlaid.
The ornaments, flowers, etc., are inlaid not only in wooden veneer but also on solid wood, chang-
ing nothing as to the manner of operating in general. However, while one makes the inlays into solid
W
wood, the base of these inlays is a bit more difficult to excavate than when it is into wood veneer.
That is why to the list of tools appropriate to making inlays of which I made the above description
on page 840, one should add a chisel with an elbow, Figs. 9 and 10,
Plate 300
chisels with a square tip, Fig. 11, and rounded gouges on the edge,
both straight and elbowed, Fig. 12. These tools should be of many
different sizes from 56 lines in width, diminishing down to 1 line, to
be able to use in the smallest places, so as not to score the wood when
excavating for inlays, and to do it with the most perfection possible
so that when the work is finished no joint is apparent, or at least the
joints are not apparent except by the difference in grain line or the
color of the wood. This requires much precision, such as in creating
the spaces destined to receive the elements, in cutting out these same
elements, of which the contours made with the marquetry saw can
still be trimmed with little shaping cuts to make them as perfect as
possible, presuming that the saw has made some unevenness. There are those cabinetmakers who
are very sure not to make any irregularities in their sawing, and to shape the ornaments with such
perfection that there is no trimming left to do.
Page 888
There you have it; all that can be said addressing the theory of veneer work/cabinetry of the sec-
ond type, that is to say, of mosaics or painted wood. There remains now to apply these general rules to
different subjects in order to exhaust the discussion of this part of veneered cabinetry, which as one
has been able to see is the most complicated of all, as much as for the knowledge that it requires as for
the difficulty of its execution.
S e c t i o n III
H ow
t o r e pr e s e n t
and
Plate 301
F l ow e r s , F ru i t s , P a s t u r e s
Figures
i n wo o d
HE object of mosaic being to imitate painting, or better said, [to imitate] the nature of different
subjects which the latter [painting] is but a copy; it is necessary, as much as is possible, that
The Design of a
Bouquet, Proper for
the works of mosaic resemble the subjects which one wishes to represent, like flowers, fruits, etc.
Being Executed in
Wood
In order to be able to give to works of mosaic this character of realism that they merit, one must
combine the knowledge of execution that I just spoke of above to that of speculative theory, which
has for its object order and agreement, which should rule in the assembly of a work. All the pieces
that compose it should be perfectly in place and in harmony, one with the other.
This second type of theory, which is nothing other than what we call taste, cannot be taught
Figures in Wood
139
in a book. It cannot be but the result of reflections that are born as a great habit from design, and
following examination and reflection of things that one represents by means of the latter. That is
why, supposing this theory is acquired for
the most part by those devoted to this type
of cabinetry (which, basically, should be
In general, one must pay attention that all the leaves or petals that make up a flower not be the
same nuance of color, and that when it is [the same color], for example in the rose, that this nuance be
very perceptible, as it also is in others that are all of one color, like the pomegranate and bluebottle,
etc. This should be preferred for using colored woods for making flowers of which the petals be all
the same color, and white woods for those where the petal color is nuanced [of a slightly different
shade, color or density], so as to be able to tint them after cutting them out. This gives the ability
to more or less deepen the color, as much as one wishes to make nuanced flowers of many bits of
wood of different colors, whether tinted or natural, which can be done, and makes very expensive
works, especially if they be of woods of a natural color.* This manner of making the flowers adds a
lot to the price of the work. That is why it is preferred to take whatever flowers from a sole piece of
wood, which one later tints after having cut them out as usual, that is to say with a marquetry saw.
The tinting of flowers or of a different piece of mosaic is done with the same ingredients and
the same way as for the wood in simple cabinetry, as I just spoke of on page 792. However, since it
is sometimes necessary that some of these pieces [such as, for example, the flower petals, and parts
of the sky, etc.] be nuanced on their surface, and even change sometimes their color, one must take
* That is how they make mosaics in Florence and in Rome,
others with pebbles sawn into leaves, the two of them being
Page 889
Figures in Wood
141
care in tinting them that they not take up too much color, or that they not take equally all over, which
is done in the following manner.
You begin by giving the wood a bath or dye of the color and the strength that should be given to
the lightest pieces of the work. Following which, when it is perfectly dry, you design the shape on
top that should be the deepest nuance. This being done, you cover in wax that which should remain
W
light, that is to say, [to resist] the first bath of dye. Then you put the piece in the latter again to deepen
the color just until it comes to the point that it should be. If it happens that a piece needs three or four
shades, you begin again, three or four times, the same operation.
You should observe, in covering pieces with wax, to put a bit more wax in the places where the
Plate 301
colors should differ, because when one resubmerges the wood in the dye, a bit seeps all around the
wax edges, more or less according to the quality of the dye, and according to the density of the wood
that you are using. I do not know how to give here certain rules, more or less extensive, that one
should give to a coat of wax, given that this varies a lot. There is only long experience that can serve
to direct this operation.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Floral Marquetry
nce you get the hang of what Roubo is saying, his description of creating floral marquetry is brilliant. But in the meantime theres the getting the hang of it part. As soon as you get past the learning
curve on this technique you will undoubtedly look on this photographic essay with, He didnt really need to
include this, did he? And at that triumphant point, you are a marqueteur. Until then you are on a really interesting journey.
Here is a fairly simple exercise taken directly from Plate 301. For the convenience of both my aging eyes
I started with a photocopy of one of the central elements and some hand-sawn tulip poplar veneer sheets
thicknessed to about 3/32". I cut them to fit into an assembled packet for the subsequent sawing. Above and
below the selected veneer I placed a piece of thin plywood as backing support during cutting. Traditionally,
the pattern transfer would have been accomplished by means of tracing the pattern on to sized paper, or by
pressing through a piece of graphite-coated carbon paper.
later on, all that is needed for the first step of saw-
menu of materials: at least one sheet of light-colored veneer, two facing sheets of thin plywood,
gummed paper tape, the pattern and glue to affix
the pattern to the outside of a plywood facing sheet.
142
Page 890
As to the means [methods] that I propose here, one cannot only augment the tone of the color of
a piece, but still change it as you judge appropriate; that is to say, one can tint a piece in two or three
different colors, observing all the while to coat with wax the under and upper sides of places where
the tint should be, and that with each change in dye.
When one changes the shade of a tinted piece, or when one tints in various colors, the place of
W
separation should not be very obvious. You do not pass right away from pale red to deep red or from
green to yellow, etc. The different dyes or colors soften at their meetings and mix together, which
works very well for the colors that one wants only to deepen. For those which are different, it hapPlate 301
pens sometimes that their meeting gives a shade composed of the two, which requires much attention and care to prevent this third shade, whether in making the first color preferable to another, or
whether in augmenting or diminishing the extent of the wax, which is something [learned] purely
from experience.
All the flowers that compose a bouquet like this one or another are cut apart and shaded each one
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
of the designs lines close to the center of the overall composition. This hole serves as an entry point for the blade, which
must be threaded through the opening and mounted in the saw
frame. By sizing the hole only slightly larger than the blades
cross-section, and locating it at an intersection, it is unnoticeable when everything is done.
Fig. 8.6 Sawing marquetry, like many motions, should be first and foremost attempted with
whatever posture you find most comfortable. My
preference is to cut vertically against a horizontal
workpiece, with the sawing surface slightly lower
than my chin. Directional lighting approximately
45 back and 45 elevated works well.
Plate 30 2 . The Development of the Principal Flowers Represented in the Previous Plate
Figures in Wood
145
according to the place that it occupies, which requires only a bit of care on the part of the artist. **
The way to cut out, shade and assemble various parts, which compose flowers, is the same as
The Development of
for the ornaments that I spoke of up above, page 878, etc. However, to facilitate the understanding of
what I just said concerning the placement of the flowers of the bouquet on Plate 301, I have given in
Represented in the
the following plate the detail of each part of these flowers, namely: Tulip D all cut out, Fig. 1; the same
Previous Plate
shaded in the place that they will occupy. In the most common
and shaded some from one direction and others from another,
ing whether they go well one with the other, and if they are
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
For this exercise it is critical to cut individual elements at the center of the composition first and then work
your way out to the outer edge, one piece at a time. This is because the only thing holding the packet together
is the tape, pins or glue around the outside edge.
Roubos marqueteur is shown in the traditional posture of sawing horizontally on a vertical workpiece.
Like him, I typically use an 18th-century-style saw frame, but my own working habit and experience is to saw
vertically on a horizontal work surface. For this project I used some 4/0 blades, considerably larger than my
usual 8/0 blade. Do not get hung up on this detail; breathtakingly glorious marquetry can be cut with even
fairly large blades if the skill and technique are correct.
In the end you are left with as many perfectly sawn compositions as you had layers in your packet. In this
case I had five completed patterns because I had three of the presentation veneers along with two backing
boards.
The virtuosity of excellent pictorial marquetry incorporates many elements of skill and technique, with
one of the main ingredients being the modeling of the individual elements of the composition through shad-
edge into the hot sand scorches the pieces all the
you desire.
146
flower reassembled and covered with its paper, Fig. 2, and all reassembled without paper, so that one
can see all the joints, which are made according to the shapes that form the outside and the reverse of
the petals of this flower, Fig. 3.
W
Plate 302
Page 891
When all parts that should compose a piece of mosaic are all prepared,
like the bouquet represented in Plate 301, Fig. 1, as I just taught up above, you
proceed with their embellishment, which is done in the following manner.
Plate 303
You first begin by tracing on the background, only with pencil, all the
assembled design, or at least the exterior shapes of all the parts that compose
the bouquet, as I have observed on this plate Fig. 1. Then you begin to inlay
and Ornaments in
the flowers by tracing the outline on the work with a very fine-point knife
Inlaid Wood
[more like a scalpel], in the place that each element should occupy.
While you inlay the flowers, you must begin with those that are found beneath, that is to say,
Figures in Wood
147
where the exterior form is not entirely visible, like those of M and H, Fig. 1. When you have inlaid
the flowers that are all whole, like that of G, you trace the outline on both the background and on the
flowers already inlaid, as I noted in this plate, where the position of Rose G is but a void, instead of
the flowers that border it, which are completely inlaid.
This observation is essential, because if you begin by inlaying the flowers that are visible in their
W
entirety as whole flowers, and if the shape of these flowers is very detailed, like that of the Sweet William A, it will be very difficult to adjust the other flowers or leaves that border it, and to join them perfectly at all the angles and shapes that produce the outer edges of this flower. I have not represented it
here completely inlaid, in order to better explain the reasons why you should not inlay these flowers
Plate 303
That is why (as a general rule) one should always begin by inlaying the pieces with the most out-
line against the background, then inlay those that are not inlaid against the background except by
one part, and finish with those where the contour will be totally visible, like those of A and G, and
in general all those that are in the same situation. [In other words, when inlaying several elements
that proceed into the background as compositional elements, it is best to begin with those elements
farthest back from the viewer and progress toward the front.]
As for those flowers that are isolated, like C, D or F, it is unimportant whether to inlay them first
or last. One must pay attention to whether it is the end of their stem that makes the contour, like in
Page 892
Fig. E, or if it is the base of a flower, like in Fig. D, because in the first case one must inlay the flower
first, and the stem later. In the second case, you must inlay the flower last, for the reasons that I just
detailed above.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
ing. Usually the shading was accomplished by the immersion of the elements in a pan of scorching-hot sand.
Fully complete, a proper marquetry panel really does resemble what would be called painting in wood.
The composition was mounted onto a paper backing with hot hide glue, ready for insertion into the back-
ground.
That insertion is a three-step process. First trace the outline onto the background.
Next cut the outline into the background with a shoulder knife and remove that background. In this case
the removal is straightforward because it is a veneer background glued to a substrate. Wet the veneers to be
removed, wait a while and remove them.
Once the void is cut into the background, the compositional element is glued in place with hot hide glue,
148
What I just said about the inlay of flowers should also be applied to the ornaments like in Fig.
2, so that the shapes of
the parts that are positioned on others are
W
Plate 303
These ornaments, like those represented in Fig. 2, side O, are called Moresques and are normally
made with wood of a single color applied on a different colored background. Sometimes these ornaments are of two colors, especially when they are composed in a way that they seem to be interlaced
with each other. In either case it is very rare that you shade them; you satisfy yourself with shaping
their outlines, and in indicating the interior contours with kerf lines of a cutting [marquetry] saw,
or by lines made with an engraving tool or pen. See Fig. 2, side O, where one part of the ornamental
running frieze is placed in this manner, and the opposing side P, is an excavated void prepared and
positioned to receive the rest of the inlay, as well as one part of the mass of the bouquet from Fig. 1.
I just did this in order to better explain the steps in execution of these types of works, which, as you
can see, cannot be treated with too much precision and care. It demands a great deal of experience in
practice, which is as necessary in this part of marquetry cabinetry as the most perfect theory.
Plate 304
Trophies, Ciphers
and Flowers in Inlaid
Wood
Figure 1 of this plate represents the design of a panel of marquetry cabinetry, when the middle is
ornamented with a medallion supported by branches of Rose and Laurel and crowned with garlands
of flowers, which is constructed in the manner that I described above.
The medallion can be filled in different ways, whether by a composition like that of the first fig-
ure, or only with a number or monogram. In either case, as in Figs. 1, 2 and 3, if the background of the
medallion is different from the background of the work (as it should always be), it is good that the
medallion composition be fully executed and all filled in with its subject before inlaying it on the base
of the panel.
When one makes a medallion or another work of this type, you begin by cutting out all the piec-
es that should be inlaid, so as to trace each one in its place. Next you cut out the middle of the oval
Page 893
with a fret saw, which you introduce by a hole pierced in one of the places that should be voided,
or even that one makes at the place where the figure approaches most closely the side of the oval,
like points A or B, Fig. 2. This should cause no difficulty, because the blade of the saw is so thin
that when the two parts come together, the joint is not obvious. When the middle of the oval is cut
out, you glue the piece of infill. When you insert the saw at the sides of the oval, like in Fig. 3, you
should, before positioning the parts of the infill, glue the fillet in place at the edge, like in Fig. 3, and
even the entire frame, Fig. 4, before joining up the joints. The pieces of infill should be locked into
place, as you can see in Fig. 3, which represents the medallion of Fig. 2, all filled in and viewed from
below [behind], that is to say, from the side that is applied directly to the panel.
When the medallion is dry, you glue it in place on the base as for ordinary marquetry/cabinetry,
I am not going to speak of the flowers or of other parts of the detail on this plate, given what I just
said above in explaining Plate 301 this should suffice. However, as there are many little pieces in the
flowers of Fig. 1, which cannot be inlaid, like the thorns, the ends of little stems, etc., you can be satisfied with engraving them on the background of the work from which they stand out, either in black
or in brown, according to the color of glue with which the details are filled.
Figures in Wood
151
Figure 1 of this plate represents a war trophy composed of different arms and instruments old as
well as modern, which are placed on a finial, which serves to support it.
This piece, although very complicated, can be executed in marquetry cabinetry. There is in the
Kings Offices, at Choissy a table on which there is a trophy of the same type as this one, and which is
executed with all the precision and delicacy possible. It gives great honor to its maker, whom I would
W
gladly name here if I knew him, given that these examples are very rare and you could not hire those
Plate 305
War Trophies
Appropriate
to be Executed
in Wood
The trophy that I represented in Fig. 1 is executed in the same manner as the other pieces of mo-
saic that I just described above; that is to say, one must always glue the big masses together before
inlaying them into the background of the work, as I have represented in Fig. 2. Always begin with the
pieces that are the most encumbered [or obscured], and finish with those that are totally or at least
partly isolated. One must also pay attention to place the grain of the wood in the direction where it
will be the least contrasting when one cuts it out, and [also pay attention to the grain] in the longest
pieces, as I have observed in the development of the principal pieces of this trophy, represented in
Fig. 2, sides B, C, D, F, H, I and R.
When the pieces are to be rounded, like those of A, M, E, N, O, G, P and Q, one must as much as
it is possible, put the grain of these pieces parallel to their roundness, because in shading them, the
action, whether from the fire or the acid, works better in this direction than the other, and softens
more naturally along the grain line than at the end grain. The inequality of a piece of wood shaded on
Page 894
the end grain takes more or less effect from the action of the fire or the acids, making the wood shaded
in this manner wavy at the ends of the shaded places. This does not happen when you place the wood
grain in the same direction as the shade, as I recommend here.
Figure 1 of this plate represents a panel on which I have assembled fruits of different species, birds
and other animals. I will not enter here into the detail of the construction of this piece of mosaic,
Plate 306
given that this would be only a useless repetition of what I just said. I have not made the design of this
plate, nor those of Plates 304, 305 and 307, except to give an example of each type of mosaic, and giving
each one the detail of the principal parts, and the differences that can be found in the manner of executing them.
In Fig. 2 of this plate, I
have represented the detail of the principal pieces
of Fig. 1, which are sides C,
D, E, F, G, I, L, M, N and
areas which are constructed ordinarily. There is only
the fur of the animals and
the feathers of the birds
that are a bit different, because they are done almost
all with an engraving burin,
in order to better indicate the fur and the lines of the feathers. That is why I represented in Fig. 2, sides
A and B, one wing of a bird all detailed and then a solid mass, so one can better judge [the work] than
in the other parts of this plate, which become a little small.
Figure 1 represents a landscape with a shepherd placed in the front playing the flute and a sheep
Plate 30 6 . Various Sorts of Fruits and Birds appropriate for Execution in Inlaid Wood
154
Plate 307
Landscape and Figure
next to him who seems to be attentive to the sound of the instrument, which is to him, they say, natural.
At the background of the plate, I have represented in Figs. 2, 3, 4, etc., the principal parts that compose
this piece of mosaic. The background should be a piece integrated into the composition, that is to say,
in Inlaid Wood
with Its
Explanation
can say about the theory and the practice of veneer-work cabinetry of the second type, that
is to say, of marquetry. I am only too happy if
Page 295
In the description of marquetry cabinetry, I have given no example of perspective, which I did
not believe necessary, given that it is only the first type of cabinetry where the pieces are cut relative
to the place they will occupy, to represent whatever object, according to the rules that I have given,
on page 868, etc. This changes nothing in the manner of operating, which is the same as for cabinetry
in veneer.
However, it is to be hoped that when one represents pieces of architecture in perspective or even
geometrically, that one imitates, in using the mixture of wood, the manner of using marble. That is
to say one could, for example, in the case of a frontispiece ornamented by the orders of architecture,
make the flutes of the columns of one type of wood, the bases and capitals of another, the architrave
and cornice of another, and finally the frieze and the other adjacent pieces of another type, which
would certainly work very well. Different grains of wood mix well and in opposition one to the
other, without however contrasting too much whether among themselves or with the background
of the work.
What I am saying here, relative to a portico decorated with the orders of architecture, can equally
be applied to any other object, without that one being represented in perspective. This would be a
type of marquetry cabinetry very nice to see, and which would be positioned between the first and
the second type of veneered cabinetry of which I just made the description. I am going to follow this
section with that of solid cabinetry and ornamented cabinetry, which will end this part of my work.
*** In the description of the two first types of veneered cab-
understands this part, and whose talents procure for him the
means of putting them into use, and merit his being more well
he third and last type of veneered cabinetry that remains to be presented, as I have
already said at the beginning of this part of my work, is the one where along with
wood, one uses metals, shell, ivory and other materials, and even precious stones.
This type of veneered cabinetry is named by its workers as marquetry, to distinguish
it from the type where one uses only wood veneer. This denomination, although the
most generally used, does not seem to be accurate because the cabinetry of which I speak is, above
all, a type of very rich mosaic. By means of these Mosaics one can represent all sorts of subjects, not
only colored as in works of marquetry or Painted wood, but also by the mixing and the contrast of
[positive and negative composition elements] different materials that one uses, which are separate
one from the other and create Compositions of the greatest beauty, whose [perfection of] execution
is often argued based on the richness of the materials.
This precious cabinetry technique is very old, and without having recourse to the testimonies of
ancient writers (often little verified or badly understood), we often see works of this type worthy of
admiration by connoisseurs in one of the rooms in a Gallery in Florence, and of the magnificence of
the Medicis, who commissioned them.
Following the examples of these beautiful works, we have made others in France, even by those
who did not share the times of Queens Catherine and Marie de Medici.* Since that time the taste for
this beautiful Art has been maintained in France up to the end of the last century, but presently one
hardly makes cabinetry of this type, except for some small thing or clock case. The makers who have
been named the Pendulistes [clock and watch-case makers] are those brethren who are occupied
especially with this work.
* There is in a salon in the Florence Gallery, called the Tri-
retired. (ibid.)
zul, and the bases and capitals are of a massive gold. It is the
tilly some cabinets of this type, which, although less rich than
are also found in the homes of some of the rich, who under-
stand their value, and which they keep with much care. You
155
Chapter 14
Page 982
156
Page 983
I will not extend myself much more on this last type of veneered cabinetry, not only because it
is not fashionable, but because I do not have enough practical experience on this matter, and it has
not been possible to find all the help that I would need to enter into all the details that are required for
the practice of this beautiful Art. I will content myself to describe them in their current state and to
indicate the means of working in this ancient art, hoping that someone more able than I can achieve
W
what I can only approach here. [It is indeed fascinating to note that by Roubos account, the practice
of Boullework marquetry (tarsia a encastro) had become almost unknown by the 1760s, to the point
where he could find no one to instruct him thoroughly in the practice, as he implores someone more
expert to write an authoritative treatise on the subject. This recitation is not strictly supported by the
historical record, as many talented Roubo contemporaries were well-known in the art.]
The appropriate tools for working this third type of veneered cabinetry are similar to those of
other veneered cabinetry, of which I made the description in the body of this part of my work. That
is why I am not speaking much more about it, and I will pass right away to the practice of this work,
after having said something about the materials that enter into the construction, which I am going to
do in the following Section.
Section I
Descr iption
o f t h e d i f f e r e n t m at e r i a l s
t h at o n e us e s i n t h e co n s t ru c t i o n o f t h e
t h i r d t y pe o f v e n e e r e d
Page 984
C a b i n e t ry
HE different materials that one uses in the construction of the third type of veneered cabinetry or marquetry, are (without considering the previous and aromatic woods that I described at
the beginning of this part, page 767, &c.) of two types the animal ones and the metal ones. Those of
the first type are tortoiseshell, ivory, horn, mother-of-pearl, burgau [mother-of-pearl with a greenish color] and whalebone. Finally those of the second type are copper, tin [pewter], silver and gold.
in their materials. Among others are that of the main altar of
so bad, since they are giving more than the value of the silver
that they cost. On the other hand, these sorts of works will not
ent works are at least of the last century, and we dont make
endure for as long as they please. But the really bad result is
last century, which we have no regard for any more, and for
elders who have worked or have seen this worked this way.
For the young, the greatest number of them dont know this.
the caprice of those who have invented it. It is true that the
works being made now have a good appearance and cost less
it is no less true, as much for the part of the art that I practice,
in comparison with those of the last century. But they are also
157
Since all these materials are of a different nature, and they require to be worked differently, it seems
to be necessary to give them a description here, which although abbreviated, can introduce the workers to a level of knowledge of the nature of these different materials.
1. Shell
W
We name shell the covering of an animal named turtle, which is amphibious and crustaceous, that
is to say, covered with scales. This covering is more or less large, according to the size of the animal,
and is of an oval and convex form, a little like an old shield. It [the shell] is not all a single piece but it
is composed of many pieces of different forms and sizes, which overlap each other, and have a compressed movement or dilation according to the wish of the animal, and that by means of the muscles,
which attach the scales to the shell or roof of the turtle. [The reference to scales, referred to as scutes
in the trade, that overlap in a shingled pattern is definitive in identifying the species of hawksbill
turtles. Greenback turtles plates are abutted and have a clean join line between them.] There are
turtles in the Asian and African oceans, but the most beautiful of them are taken from around the isle
of Quibo, in the South Sea, where there are four kinds, namely:
1. The Franche Turtle [also known as hawksbill turtle], which is of a medium size, and which is
only recommended for its flesh, which is very good to eat. It also has scales, but they cannot be used
for anything because of their thinness. [In fact, hawksbill is the thickest and most desirable shell.]
2. The Carette, or Caret [also known as loggerhead turtle], which is smaller than the first, of
which the flesh is not as good, but which gives the best shells, of which the turners and toymakers
make much use, and which one uses in beautiful works of cabinetry.
3. The Kaouanne, or Cahoane [also known as greenback turtle; in English, Loger-hu; in Spanish
Caivava]. This type of turtle is much bigger than the first two. Its flesh is not good to eat, but we use
its oil. Its shells are much larger than those of the Caret, but they are thinner and much less valued.
These are the shells the cabinetmakers use most commonly, as much because they are less expensive
than the others, and because being much thinner, there is less work to do in order to harvest and prepare them, and by consequence less waste.
4. The fourth type of turtle is the largest of all. It is absolutely good for nothing, if only to burn its
oil. There are also land turtles, but they do not have such shells on their backs, and the shells they do
have cannot be used for works of cabinetry. That is why I am not going to speak of them here.
The shell or top of the Caret turtle is composed of 13 sheets [scutes], namely, eight plates that
are placed equally with four on each side, and five that are curved/convex, and which are placed on
the center line of the back.
These last five are the thickest, and are almost equal in thickness on their entire surface. As to the
eight others, there are four that form an oblong and are a bit parallel, and the other four are rounded
on one side to follow the contour that forms the carapace on top of and at the end of which they are
placed.
These shells, as well as the four others, are not absolutely straight on the surface, but they are all a
bit curved/convex, especially these last ones at the extremities. The ridges of both taper away to nothing at their edges, which does create a lot of waste unless one welds many of these pieces to each other,
as the turners do for their different works. The longest of the sheets of the Carets is from 1215 thumbs,
by 78 thumbs in width. The flesh side of the sheets of shell, that is to say, that which is concave, is al-
Page 985
158
ways the less beautiful, and its surface is almost all covered with a type of white vermicule [filigree-like
markings], which often appears as a beautiful design. I do not know if it is the sign of their connection
with the membrane or tissues that attached them to the carapace or top of the animal.
Shell has three distinctive colors, namely blonde, brown and true black. Sometimes one or two
of these three colors dominate, but they are rarely alone. I have seen in a large store, a sheet of shell
W
totally blonde, which the merchant valued highly as something very rare. There is also, especially in
the Caret tortoise, some sheets that are mottled and mixed with a little brown of different nuances,
and with white, of which some areas are similar to mother-of-pearl.
In general, shell is transparent, hard and very fragile. Although similar to horn, it is much less
flexible than horn, which comes from having less fatty parts that bind the parts one to another.
Shell is, however, very malleable, and acquires much stretching/elongation [bending or flatten-
ing] by means of fire or boiling water. But when it is cooled, it remains in the form that one has given
it, and becomes as rigid as before.
Whenever shell is completely whole, it is subject to shrinking when heated. That is why one
should take great care that it be very dry when one uses it. When it is humid, with just a bit of heat, it
shrinks considerably.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
159
Shell has one very singular property, that is, one can weld it without having need for any particu-
Ivory is a bony substance that one finds in teeth, or better said, in elephant tusks. We call them morfils when they are one piece and we give them the name ivory when they are cut up.
The tusks of elephants come from Asia and Africa. The latter are the smallest, and are only around
four feet long. The first are up to 10 feet long. The smallest are from the African coast, especially from
Riofresca, of the Gambian river, from Senegal, and of the coast of Dents (Ivory Coast). Those
from Asia come from the island of Ceylon and the Kingdom of China, from Pegu, from Thailand
and from Oracan. It is said that the ivories of Ceylon never yellow. That is why they are priced more
expensively than the others.
Whether ivory be of the bone genre, it is much more compact and heavier than bone. Its pores
being very tight makes it capable of taking a very good polish, which lasts a long time.
We distinguish two types of ivory: namely the green and the white. One can distinguish them
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
shell scutes is nearly nonexistent. There is almost no research on scute formation in any turtles. Knowledgable speculation and forensic observation posits that the shell plates are formed via annual, or at least periodic, depositions of additional layers of pigmented keratin, enlarging and thickening the scutes as the turtle ages
and grows. The scales are deposited in thin sheets sequentially, resulting in a laminar structure, which, in concert with concentric growth lines along the scute boundary, can serve to support a careful estimation of age of
some species. This characteristic is especially notable on hawksbill scutes as they are flattened and scraped,
revealing growth tide lines similar in concept to grain in wood or ivory. On fully mature greenback turtles
the largest scutes can measure up to 20 cm x 40 cm, with a very even thickness generally of 12 mm. Hawksbill
scutes tend to be slightly smaller but more irregular in thickness, especially in the center where they can reach
6 mm in exceptional cases.
Shell Composition and Coloration. Tortoiseshell plates, like human hair, fingernails, horses hoofs
and animal horn are composed largely of keratin, a protein biopolymer. Keratin is a complex polymeric matrix
secreted by the outermost layer of skin. Since tortoiseshell grows as a complete layer at a time and not directionally as from a cuticle bed, it is more homogeneous (not fibrous) and isotropic than horn, with much tighter
laminae than the easily delaminated horn.
The remarkable mottling of the shell comes from inclusions of melanin, the nearly universal zoological
pigment responsible for all skin and hair color. In chemistry, appearance and feel, it is easy to envision a tortoiseshell scute as a giant fingernail with freckles.
If examined microscopically, the mottling of tortoiseshell is seen to be made up of spherical spots of color.
The closer together the spots occur, the deeper the color, rather like the effect seen in screen printing. In the
case of imitations of tortoiseshell, plastics such as celluloid, casein, Bakelite and some others give excellent
results. The color is shown to be in patches or swathes, and there is no dot-like or disc-like structure seen when
microscopically examined.
The plastron plates are uniform in pattern and comparatively colorless. These blonde translucent plates
are often called yellow-belly in the trade, and are prized for their own unusual aesthetic, especially as used
in ladies folding fans.
Page 986
160
both by the color of their outer shell. Green ivory has a brown and blackish shell and is a bit light;
white ivory has a white or a darkish yellow shell. These marks are not really certain, and it is good to
cut off the end of the tusk to judge the real color of its interior.
One prefers green ivory to the white because it has the finer grain, and as a consequence is less po-
rous than the white, which often has disagreeable visible grain, and has the fault of becoming yellow
W
over time. In green ivory, the grain is not very visible, and the small green tint passes over time, to be
replaced by a nice milky white, which has the advantage of not yellowing.
Green ivory has the fault of being more fragile than the white, and it shrinks more than the latter.
Following the analogy of hard wood or soft, it should not be. But that is not the only time that nature
deviates from the rules that one prescribes for her.
One cannot straighten ivory [or bend or mould it] over fire, such as with shell. It is necessary to
cut it up with a saw, then to finish it with a file, as if it were a piece of copper or other metal.
The morfils or tusks of elephants are a bit curved on their length, and end in a point. They are
not solid down their entire length, and in cutting them up, it is necessary to take great care to reach
the depth of their interior with the saw, in order to take full advantage.
3. Horn
Page 987
The horn that the cabinetmakers make use of is a type of white horn sold in Paris under the name
English horn, from where it is carried in little barrels. It is this horn that the tinsmiths use for closing up lanterns. They sell it more or less thick; those that are the whites, which have hardly any spots
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Shell Properties and Working. The acquisition of the carey as raw material for use in craft work involves separating the rigid plate from the supporting bone and flesh. This is accomplished by exposing the
outer shell to high temperatures, preferably in the form of boiling water, and then peeling the outermost layer
of mottled keratin from the underlying shell structure. In some cases this was actually done while the turtle
was still alive, in the belief that the living turtle would deposit new layers of epidermal plates to replace those
removed during the harvest. This is in fact the case, but the shock of immersion and relatively unprotective
nature of the damaged shell meant that the turtle was not likely to live very long once the harvesting and release were complete.
Once in hand, the carey plates were necessarily worked extensively to make them the material of art, as
they were not yet suitable for those purposes straight off the turtle. First they were flattened by placing them
in hot salt water, and then by pressing them between heated cauls for several days. They were then filed with
a tool called a graille, which is almost certainly no longer being manufactured, although they do appear periodically at antique tool markets. An excellent substitute for a graille can be found by using either a farriers
file, or a similarly configured tool called a Vixen file made for the plastics trade. Filing would be followed by
scraping, sanding and finally polishing with increasingly fine abrasives until the surface was mirror-like.
Many artistic processes for tortoiseshell exploit a very important characteristic of the scutes: they are par-
tially thermoplastic. The keratin plates respond to changes in relative humidity by taking up or losing water,
which in turn affects their mechanical properties and physical dimensions. As the moisture content rises,
the material will become weaker but more pliable. If it is steamed or boiled, some of the disulphide bonds are
broken and any deformation of the softened material will be retained upon cooling due to the formation of
new bonds. In short, under controlled heat and pressure, tortoiseshell can be made to conform to any shape
provided by a mold or caul, and on cooling will retain that shape.
A more extreme exploitation of thermoplasticity is in the artisans ability to weld tortoiseshell scutes to-
161
and which are entirely transparent, are the most cherished by the cabinetmakers, unless they want
to make fake shell. Then, they will use red horn, which imitates the light of the shell, of which they
will imitate its nuances with color, of which I will speak in its place. I will not extend myself further
on the subject of horn because you can find it by the sheet already prepared, and it only concerns the
purchase choice that one wishes to make. As to its usage, I will speak of this when treating the pracW
tice of marquetry.
4. Mother-of-Pearl
We call mother-of-pearl or simply pearl, the shell of one type of oyster in which pearls are formed.
These oysters are three or four times larger than ordinary oysters. Nacre is heavy and very hard.
Its exterior is of a rosy grevy and all wrinkled/rippled. The first exterior layers of this shell, once
removed (which can be done by means of aqua fortis [nitric acid] and a turners wheel, or even
simply by rubbing with a small millstone and water), will look as beautiful as the interior, where its
color is of a beautiful silvery white, very shiny, which is mixed with more beautiful colors of an iris or
a rainbow. When you look at it, you can see all at once the hints of yellow, red, violet, blue and green,
which change unceasingly according to which direction you look. This is caused by the various ways
the pieces [that] make up the Nacre receive the light, and reflect it to your eyes. This changing
of color is called orient [referring to the water origins of pearls]. Thus, we say that Nacre has a
beautiful orient, when these changes in color are greatly varied, and the difference in the latter is
obvious. Nacre has the singularity of appearing wavy on its surface, although it is perfectly uniform.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
gether to form either larger sheets or thicker blocks. For larger sheets, the margins of the scutes to be joined are
prepared by precisely thinning and conforming them to each other. Then the margins are soaked with water,
sometimes by applying dampened rags to the region. Finally the weld is accomplished with the overlapping
area being pressed together between hot soldering irons or similar tools. For making thicker sheets, cleaned,
prepared scutes are stacked together and pressed between rigid cauls and then immersed in hot water. By
employing these processes, large surfaces could be veneered with a welded single sheet of tortoiseshell, and
complex shapes such as frames for spectacles or objects dart could be fashioned from solid blocks. The chemistry of such welding is purported to be based on the hydrogen in the boiling water acting on the hydrogen and
hydroxyl in tortoiseshell, producing a hydrogen bond.
Most sources advise strongly against attempting to fuse tortoiseshell with dry heat, as the concentration
of such heat is generally too great and results in embrittling and deepening the color of the shell. Tortoiseshell
is not infinitely malleable, and it cannot be made soft enough to be molded or through-pressed (also known as
die-punching or die-cutting) without risk. Whenever possible, manufacturers chose to cut delicate ornamental patterns by means of drills and fine saws, rather than attempting to press open designs with dies.
As for gross analytical features, genuine tortoiseshell fluoresces with a yellow-brown color in the light of
an ultraviolet lamp. It rates about two on Mohss hardness scale, has a specific gravity of 1.29, and a refractive
index of 1.55. When it is being sawn, tortoiseshell gives off a characteristically acrid smell; when burned it
emits the unmistakable odor of burning hair.
Tortoiseshell does not show the corrugations or striations of horn. Antique tortoiseshell, however, often
develops a surface patterning visible at low magnification and in reflected light, that is reminiscent of watered
silk. These meandering lines probably represent the edges of the sheets laid down annually as the shell grows.
As the surface of the scutes are far from flat, the pressing processes distort the sheets; and the sawing, scraping and polishing cuts across the natural peaks and troughs, revealing the edges of the successive sheets.
Excerpted from Tortoiseshell and Imitation Tortoiseshell Donald C. Williams.
162
This appearance strongly approaches reality, when one sometimes touches it to assure oneself, of the
illusion that is made to your eyes.
These waves and these changes in color are only apparent when parallel to the surface of the
mother-of-pearl. When it is split, its thickness is of one equal and uniform color, of a matte white,
tending toward a pinkish grey.
Page 988
Mother-of-pearl is extremely hard. You cannot cut it except with a saw, after which you level it
out on a millstone, which I will speak of later. Mother-of-pearl has the defect of being very breakable
and sometimes damaged by worms, sometimes deeply.
It also splits sometimes in its thickness, which separates in sheets. It is these sheets or layers that
being more or less opaque, one to the other, or at least positioned one to the other in a wavy manner
and a bit parallel, give Nacre its waves and different colors that you note on the surface. The most
beautiful mother-of-pearl comes from the Far East Indies. You can fish them also in America and on
the Scottish coast.
There is a type of clam named Burgaut or Burgaux, and by the workers named Burgos,
which is found in all the American islands, where the shell makes a beautiful type of mother-of-pearl.
But as the largest of these shells are not as large as a fist one cannot gather them except in very small
pieces, given that they have Orient-like pearls only from the [inner] side of their surface, or at least
parallel to the latter. The Burgaut sometimes has more vivid colors than the mother-of-pearl, which
is preferred, especially when you only have need of small pieces; the rest are split and worked the
same as for mother-of-pearl.
5. Baleen: Whale
Baleen [the feeder/filter membrane in whales] is little used in cabinetry. However, one can use them
in good part for black filet/string, or another type of work. Baleen is gathered from the baleen plates
that act like teeth in the fish, which has the name of Baleen whale. These plates are composed of a
stringy substance a bit similar to that of goat horn. The baleen plate polishes very well. It comes in
grey and black. The latter is the most beautiful and should be preferred for works of cabinetry.
6. Of Copper
Copper is a metal, medium weight, of a hard quality, homogenous and ductile. Its color is of a shiny
red; although it is a bit less malleable than silver, you can make very delicate strings/wire/threads
by using a drawing plate. We distinguish two sorts of copper: namely, natural copper which is that
which I just spoke of, which is named rosette or red copper, and artificial copper, named yellow
copper or Laiton [brass], which is composed of about two-thirds rosette, and one-third zinc. This,
while changing the color of copper, does not get rid of its malleability, provided that you work it cold
and reheat it from time to time. If you wish to hammer it when hot, it breaks into dust, which is caused
Page 989
by the difference in materials in their response to the heating action of the fire, thus self-destructing.
The yellow copper [brass] is less dense than the red, and this difference is in a ratio of 548 to 648
pounds per cubic foot.
Cabinetmakers will use only yellow copper [brass] for the ornaments that they adapt for their
use. When they make pieces of marquetry, they use brass sheet, which they choose in stores according to the thickness that is convenient. It is not that they could not also use red copper, which would
163
work very well in certain occasions, like, for example, if one inlays on ebony, where it would have less
contrast than with brass. Lastly, it is a custom from which we should break away sometimes, as I will
speak of later.
The sheets of brass should be chosen of the most equal thickness that is possible, without any
inequality nor cavities nor cracks on the surface [grain], so that when used they are capable of
W
achieving a beautiful polish. This is essential for works of marquetry, where the least fault in the
brass would do great harm to the beauty of the work.
7. Pewter [in the earliest dictionaries]
Pewter [zinc or tin] is the lightest of all the metals. Its color is white, a bit like that of silver. It is
flexible and soft, and makes noise when you fold it. It is compatible with all the materials, but it
deprives them of their malleability and makes them break like glass. Pewter comes from various
countries, like China, Japan and the Orient, particularly from Malaga. It also comes from Bohemia,
but the highest regarded is that from England, known as pewter of Cornwall and Devonshire, provinces of England, [and] which is named in Paris [as] fine pewter. As pewter in general is a bit soft
and pliable, it is good to alloy it with a bit of red copper, which the pewterers call, to give an alloy,
which is nothing but two to three pounds of red copper melted apart, in which you mix a bit of tin,
and sometimes a pound of bismuth or shiny tin. You mix all of it in one hundred pounds of molten
tin, which will strengthen it, making it more sound and able to receive a good polish. Cabinetmakers
make little use of pewter nowadays, although it makes a very nice marquetry. You find pewter by
the sheet at the pewterers, who flatten and polish it to sell to the engravers of music. These sheets
could work very well for the cabinetmakers, which I will speak of later.
For lack of flattened sheets, one could use a rolling mill. However, those that are flattened are
Silver needs a bit of an alloy to be worked. The least quantity of this alloy is of a thirty-second,
and at the most a twelfth. The alloy of silver is made with the most pure red copper, which makes it
very elastic. One can have, using a rolling mill, some sheets of silver of such a thickness as you judge
appropriate, which is very suitable for works of cabinetry, where you use it with brass [copper], or
even with gold, or even all alone with tortoiseshell.
9. On Gold
Gold is the more precious and highly regarded of all the metals. It is also the most compact and the
heaviest, for a cubic foot of gold weighs 1,326 pounds, 4 ounces. It is a perfect metal, incorruptible,
of a yellow color that has a bit of brightness. It is neither elastic nor resonant [does not ring when
Page 990
164
struck], but it is very malleable, even more than silver, when it is alloyed appropriately. It acquires
much elasticity, whether one works it with a hammer hot or cold, and with a rolling mill.
The alloying of gold is made with silver or with red copper. The latter is preferred, however,
because the alloying of silver whitens the color of gold, while the red copper augments it. The finest
alloy of gold is ordinarily one/twenty-fourth, that is to say, twenty-three parts of gold to one part of
W
copper. However, to give it more hardness, we mix in one/sixth part copper. That is what is called
gold of 20 karats. The most ordinary alloy is a twelfth, which is called gold of 22 karats, or Kings Gold.
Gold works and polishes very well. It is of good use for works of marquetry, where one uses it
with silver, shell and mother-of-pearl. Sometimes we use it alone with these two last materials. To
give the most beauty to a work, you use gold of different colors, like white gold, yellow gold, red gold
and green gold, etc.
Page 991
After having summarily treated the different materials appropriate to be used in the last type of
veneered cabinetry, there remains the discussion of the methods to bond them together or to the base
of the work, as well as those that serve to hold together the joints of the metals, or better said, to solder
them together.
The first are the different sorts of pastes and mastics, and the second are the types of composition
of metals, which are called solders, which differ according to the metals you wish to solder. These
should always be of a much lower melting point than these last [the metal elements], so as to go more
quickly into a molten state, which I will speak of later.
I will not speak of strong glue, given that I treated this in the first part of my work, page 80. How-
ever, I do not know how to recommend too strongly to cabinetmakers to only make use of the good
English glue, for all parts of their works. That of Paris is of absolutely no value for gluing shell and
other materials of which I spoke above. There are instances where one must use fish glue, which I
will speak of.
10. Isinglass [Fish Glue]
Fish glue is the best that one can use for gluing hard woods and metals. It is made with the skin, nervous and mucilaginous parts of certain large fish [sturgeon], which are found in the Russian seas. It
is in the north where fish glue is made, from where the English and the Dutch bring it to us, especially from the Port of Archangel, where it is a good business. Good fish glue has hardly any odor, and
should be of a white color, clear and transparent. One must pay attention that is not contaminated,
that is, mixed of heterogeneous parts.
To make fish glue melt, you take it in the following manner: You begin by cutting the hard, dry
glue in little pieces, then you put it in a clay pot or a glass vessel with good brandy, noting that the latter covers the glue. Then you bottle up the vessel, which one must fill only half full, and you put it all
on hot cinders just until the glue dissolves perfectly. Or, you can cut the glue as above, and you soak
it in the brandy until it has softened, then you make it melt in a double boiler, as is normally done.
There are workers who, instead of brandy, put the fish glue in ordinary water to which they add
a garlic clove. This is rather good, but is not the same as brandy, to which one can add a bit of garlic,
which can only augment the strength of the glue [garlic serves as a surfactant and plasticizer].
One can do the same thing with good English glue; that is to say, put [it] in brandy and garlic. I
have done it many times, and that has always been successful for me. I even believe that in the case
165
where one only has to glue wood with shell or other materials in which the heat produces good action,
and consequently opens the pores, one could do without the latter glue, like one does every day.
Page 992
We name various compositions cement or mastic that have a glutinous and sticky nature appropriate for binding together many pieces, whether they are homogenous or heterogeneous, or speaking
in the common language, whether they are of the same or of different species.
We make different sorts of cements. That which is used most generally for holding metals is com-
posed of four parts of refined pitch [rosin or colophony], two parts of yellow wax, and one part of
black pitch [asphaltum]. One melts this all together in a glazed ceramic pot, on very low heat, and
even on hot cinders. When these are all perfectly melted, we mix in some pulverized brick powder
that has been passed through a silk sieve, and we put in a quantity sufficient to give the mastic the
consistence of a soft paste. We take care to mix this well, so that all the parts are blending well with
one another. We make another mastic that is almost similar to this last one, which is composed of four
parts of refined pulverized rosin, one part of yellow wax and one part of pulverized brick. These two
types of mastic are used while hot.
For glass and transparent stones, and even marble, we make a mastic composed of quicklime,
rye flour and egg whites, mixed together with salt water, or even one part pulverized quicklime, two
parts of brick passed through a silk sieve, and all of it soaked with walnut oil.
There are other mastics serving the same use, and for mosaics of glass and enamels, which is com-
posed of lime, or hard stone mixed with well-pulverized brick, with tragacanth [a natural gum from
tree sap] and egg whites. But that which is used the most commonly for these sorts of works is composed of slaked lime by air, powdered marble (or even with whiting [pulverized chalk], which works
just as well) boiled together with linseed oil, and reduced to the consistency of soft paste, to which
one gives more or less body, according to how one adds or diminishes the amount of marble powder
or whiting. It is this mastic that serves to construct the mosaics of Rome, with the exception that in
the place of marble powder or whiting, one uses there the soft stone of Tivoli, otherwise known as
Tibur at the time of the ancient Romans. If I am a bit long-winded on the different sorts of mastics,
it is because they are in widespread use for works of inlay, or where one uses metals or hard stones.
12. The Different Sorts of Solder
Solder is, as I said up above, a metal mixed in a way such that is more fusible than that with which it is
blended, and of which it fixes the different parts. Each metal should have a solder that is appropriate
for it, which is more or less strong, based on the strength of the pieces which one wishes to solder, and
in the manner by which they are soldered. We call strong solder that which approaches most closely
the nature of the metal that you wish to solder, and which, by consequence, is less fusible than the
weak solder, which one uses for small works whose capacity creates concern that the heat of the fire
does not melt them, which requires using for these sorts of works only solder of a very low melting
point, which consequently is very easy to melt.
You find solder ready-made, however it sometimes happens that it is not easily found, or
that those who sell it are secretive of their processes. I am going to give the way to make the dif-
Page 993
166
ferent solders, in the same order that I followed in making the description of different metals.
Solder of Copper
To make a strong solder for copper, one must put in a crucible 10-and-a-half pounds of scrap
brass. When it is well melted and very hot, you add 3-and-a-half pounds of zinc (which will melt very
promptly). You mix these materials, and as soon as you see that the zinc is perfectly melted, you add
five ounces of fine tin. You mix it all again for a little bit, and you throw the material to the ground in
W
a clean area, and the thinnest possible. Then you grind all of it in a mortar, and you pass it through a
sieve where the holes are of different sizes, in order to have grains of solder of an appropriate strength
for the work that you wish to solder.
One still makes solder for copper with red copper and some zinc only. It is more or less strong by
The strongest solder is of five parts red copper against one of zinc. The average is of three parts of
red copper against one of zinc, and the weakest is of two parts of red copper against one of zinc. The
last solder melts very easily, but it has the defect of being very brittle.
You make solder for soldering brass with the latter mixed with zinc. I believe it is better to use
average solder of red copper that I just spoke about above. In the case of small works, one would do
much better to use a solder composed of two parts of silver and one part of brass.
To make solder of copper that I just spoke about, you begin by melting the red copper in the cru-
cible. When it is molten, you add the zinc. When all is well blended, you pour it into a mold. When
Page 994
Solder for Pewter
the ingot is cold, you flatten it with a hammer into very thin sheets to make tinsel of solder.
Solder for pewter is made with fine tin and new lead, of which one puts about one part or one part
and a half, against five of tin, which you melt together. In order to be sure this solder has the correct
grade, you take a little bit that you pour onto a dry brick. When the solder is cold again, if it appears
on its surface like small eyes [crystal flecks] of average brilliance, that is a sign it is good for soldering
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
On one hand, the word mastic is a botanical exudate (harvested tree sap) from the Mediterranean. This
Fig.
10 .1
167
pewter. If on the contrary these eyes [crystal flecks] are very large and shiny, it would be better to add
a little lead. In general, this solder is flat and white when the tin dominates too much, and it is flat and
grey when the lead dominates.
Solder for silver is made by putting in the same crucible three parts of good silver the like of 11
deniers [deniers were small French silver coins of about 1.19 g] with one part of brass. You let it melt,
W
then pour it into a mold. After the ingot is cold, you hammer it cold to reduce it to very thin sheets
from which to make tinsel. If this solder splits in hammering, one must put it back into the fire and
let it cool again before beginning to hammer it. This solder is called solder in thirds [a 3:1 silver:brass
alloy]; you make it also in quarters and sixths, this last being the strongest of all.
Solder for gold is composed of one part gold, two parts silver and one part red copper. When you
wish that it be stronger in color, you increase the quantity of gold. This solder is made in the same way
as that of silver that is why I am not going to speak more about it.
It would be very difficult to make use of these solders that I just spoke of for the union of different
metals if one did not add some borax, a type of salt or white and transparent substance, a bit like that
of alum. Borax has the property of facilitating the melting of metals and uniting all parts, and guaranteeing the effects of air and fire by enveloping them with a type of thin glass. [In short, borax makes
an excellent flux.] That is why one uses it in soldering, because not only does it accelerate the melting
of solder and prevent it from burning before melting, but because it makes it more fluid and blends in
all parts of the piece where you have spread it.
One must not use borax without previously having calcified it, or more properly said, having
melted it apart, which is done by putting it in a crucible, around which one must have a moderate fire,
because it will glaze under too large a fire. That is why one must take care to remove it from the heat
source when it stops boiling. One must also have a large crucible, because when the borax is boiling,
it bubbles excessively. Most workers do not do enough to melt the borax, or, as it is commonly said,
to calcinate it. They are content to heat it on an iron shovel, and to pull it out when it ceases boiling.
I am not going to extend myself more on the description of the materials that enter into the con-
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
rapid cutting of brass filigree patterns, which were tacked to a substrate. The interstices, or voids between the
brass filigree, were filled with hot animal glue within which were blotches of pigment to replicate the patterns
of true tortoiseshell. In short, the tortoiseshell was nothing more than a painting with glue as the medium.
Once the partly pigmented mastic tortoiseshell had dried completely it was abraded level, so as to be
even with the thickness of the brass, and the entire surface was polished with light abrasives, probably pumice and Tripoli in wax/turpentine or mineral spirits. Because the tortoiseshell fill was both moisture- and
heat-sensitive, great care was needed to smooth and polish this surface during fabrication.
Without stable integral plasticizers and hardening agents, this decorative technique was doomed from the
start, as the fill was susceptible to too much moisture, too little moisture, fungi and insects. An intact artifact
of the era is virtually unknown.
My own recent explorations of this general technology has resulted in much greater success, because the
chemical technology at my disposal is dramatically greater than that of more than a dozen decades ago. By
adding glycerin or polyvinyl alcohol as water-soluble plasticizers for the glue base, and protein cross-linking
agents (sold as photographic gelatin hardeners), a much more fracture- and heat/moisture-resistant material can be derived; and with the recent additions of lightfast organo-metallic water-soluble colorants for
creating the tortoiseshell pattern, a very convincing replication of true tortoiseshell can be achieved.
168
struction of the third type of cabinetry. It suffices to have given an idea, so the workers who read
Page 995
this work will to be able to make a difference and to put into use with knowledge of the reason. As
to other materials, like colored glass, enamels, marble and precious stones such as lapis, agate, tourmaline, chalcedony [a large variety of silica-based semiprecious stones, including agate], emeralds,
turquoise and even rubies and others, I am not going to speak of them at all, because strictly speaking,
W
the work of these different materials is not the jurisdiction of cabinetmakers. This knowledge will
not be totally useless if the boundaries that I have described and still more the little practical knowledge that I have of the different talents where one makes use of these materials does not oblige me to
describe work that is above my ability, whatever desire I have to be useful to my colleagues.
S e c t i o n II
Works
fo r w h i c h o n e us e s
t h e t h i r d t y pe o f
Cabinetmak ing
said above that one hardly ever uses anymore the third type of cabinetry other than for orna-
Plate 333
menting some clock cases. I have not given any examples of these sorts of furnishings in the sec-
ond section of this third part of my work, even though the construction of clock cases is absolutely
[within] the competence of veneered cabinetmakers. I cannot avoid giving the details of these sorts
of works, since the manner of handling the assembly, whether to receive marquetry or veneer work,
or simply for painting, is highly fashionable presently.
We distinguish two sorts of case clocks, namely the tall, which are 56 feet in height, and the
small, which have hardly 2 feet in height, and which are ordinarily supported by feet or pedestals. We
make smaller boxes than these latter, which take the name porte-montres [carrying watches]. These
sorts of boxes do not have feet or console supports, and have hardly a foot or 15 thumbs in height.
Ordinary cabinetmakers, in other words the vast majority, do not make their clock carcases
themselves. They are made by workers who are occupied only with this type of work, and who use
only the least materials and workmanship and design that is possible. There are others who are comPage 996
mitted enough to their work to make them themselves, or have them made on their premises, in order
to watch over their execution, and who use only very good, dry, oak wood from the Vosges. As I just
said, it is not the greatest number of cabinetmakers who take this precaution. It follows that there are
many works badly made and not very solid.
The large boxes are those where we normally place the secondary pendulums, of which the work
is set by a regulator or pendulum of 3 feet 8-and-a-half lines of length, which makes it such that these
boxes have about 5-and-a-half to 6 feet in height, and contain the lantern or upper part, in which is
placed the clockworks, and the pedestal on which the case is positioned. The normal width of these
cases is from 1518 thumbs in their largest part, which cannot have less than 10 thumbs to one foot in
their interior, in order not to hinder the vibrations [movements] of the pendulum, which are from
2 to 2.5 thumbs from each side of the bob or weights, which are placed at the base of the shaft of the
pendulum. As to their thickness or depth [of the case], it should be from 56 thumbs from interior to
interior, that is to say, from the front of the backboard to the rear of the door of the case.
The shape of these sorts of cases is rather arbitrary, as long as they do not hinder the work of
170
the mechanism that they enclose. That which is designed in Fig. 5 is the form the most generally followed. This case is composed of three parts, namely the light A [bonnet], the body of the case B, and
the pedestal C, Fig. 5. There are many of these
cases to which one makes the body and the pedW
Plate 333
From the rear, the circumference of the bonnet envelops the backboard of the case, which enters
the whole thing within. However, I believe it would be better to make the diameter of this backboard
a bit wider, about 56 lines, and to make a rebate all around the interior of the lantern [bonnet], into
which this backboard would set, which would better guard the works from dust, which almost always
enters at the joint of the bonnet with the backboard of the case.
Page 997
As these sorts of carcases are made to be recovered, whether with wood of the Indies or with
marquetry, their structures are but a plain wood mass glued as solidly as possible, always observing
that these assemblies do not present any wood on end grain, at least as much as that could not be done
otherwise. It is for this reason that one glues the circumference of the bonnet with bands or staves.
As to its face, it is made of solid wood, about a thumb in thickness, positioned perpendicularly, and
you glue wood blocks to increase the thickness in the places where it is necessary. As it is to be hoped
that this mass is not limited along its width, it would be good to fit it in along the grain line from the
bottom [such as breadboard ends]. This would be better than being content to apply a little straightgrain batten crosswise, as is done ordinarily.
The body of the case is composed of two sides cut to shape [in profile], and consequently com-
posed of many pieces glued together, with a rear [back piece] that comes up as high as the bonnet,
but which does not come down except to the height of the pedestal (because the space that remains is
filled by a door that opens at the rear when one judges it appropriate), and of a front composed of two
pieces, of which one forms the front of the pedestal [that] is adhered to the sides. The other, which is
movable, forms what we call the door to the box. This door opens above the pedestal just below the
bonnet. It is good to encase it along the grain line [with breadboard battens] at both ends, to give it
more solidity.
171
The entire face of the clock case such as that I described is ordinarily arched/waisted in the fore-
ground, as indicated in Fig. 10, which ordinarily requires gluing [stock of pine], much as on the door as
on the front of the pedestal, because they say this wood takes the glue well. The
real reason for which it is preferred is that it costs less than the tender and dry
Vosges oak, which would be of a better use. This wood deteriorates less quickly,
W
The curvature of the door of the case is not equal along all its length. It is good that it be a bit more
so [curved] toward the base, the place of the oval opening to allow the pendulum to be viewed, as well
as its movements and vibrations. One must pay attention when making this curvature [bulge, convex
Plate 333
shape] that all the circumference of the oval opening be in the same plane, that is to say that all the
ridges of this opening level out, whether to be parallel to the back of the door or to be inclined. This
observation is essential since this opening, being closed by a flat pane of glass, is necessary that the
place it should occupy be so as well. What I am saying for the opening at the rear of the door should
apply to that of the face of the bonnet in which is placed the clock dial, or to better clarify, from where
it will be seen.
One should have [pay] the same attention to the construction of the front of the pedestal, as much
for the assemblies as for [also] making them more solid, in fitting it from the two ends at wood grain
[such as mitering].
When the pedestal holds [is attached] to the body of the box, as in Fig. 6, one joins the front of the
pedestal with the sides [of the box] that come down to the base, and the latter are secured on the back
of the box, where it could be good that they hold [are secured] by tongue and groove, as well as in the
front, which is much more solid than to glue them flat-joined [butt joined], as you do ordinarily. Since
the latter descends only to the height of the pedestal, it is necessary that it be fitted from the bottom
and that this fitting is assembled by inserting the back panel into grooves inside the sides of the case,
which must be wider in order to accommodate this assembly [rather than the dimension if the back is
simply butt joined along the rear edge of the side panel]. This would work better if you did not make it
descend to the base, and that you carved out the place for the door, leaving on each side a casement of
about 3 thumbs wide. This will not prevent putting one fitting from the top of the opening assembled
in the ordinary, and by tongue and groove at the ends. Whatever method you take, one must always
put a crosspiece at the base of this opening, as much to serve the closing of the door, as to receive the
base of the pedestal, which should be assembled by tongue and groove.
It is also good to place one or two crossbars from the rear side and of the thickness of the back of
the case, so that it remains stable. The sides are glued along the grain, and it is necessary to make a
tongue or tail that enters into the sides of the bonnet. I believe it to be better that these tails be set by
end grain, and since the wood is thick in this area, you return the fitting that carries the protruding
cornice by a floating tenon after these tails, as one can see in Fig. 6, side E.
When you put a stand under the overall body of a clock [the pendulum box], as I did here, it is con-
structed ordinarily, that is to say, that one assembles it on grain, at least on the front, and that its upper
side [the top] is glued by tongue and groove all around. See Figs. 6 and 7, which represent an overall
body for a clock as much on the face as cut sideways, constructed just as I explained, and according to
the design represented in Fig. 5.
The body of clocks of which I just spoke are capable of much richness both in the style as in the
material. As for their shape, they can vary according to the wishes of those who make them, noting
Page 998
172
always that the clockworks has enough space [allowing the pendulum to swing freely], and that the
distance from the middle or center of the clock face to the middle of the oval hole of the door of the
box be at 34 thumbs of distance. [Because] the length of the pendulum being of 3 feet and eight-anda-half lines from its point of suspension to the center of its swinging movement, which is a bit higher
than the center of the pendulum bob, the distance of 34 thumbs is rather close to where it must be
W
for the center of the bob to be found with regards to the oval hole in the door of the box. This comes
back rather well in following the measure that I gave here, because the point of suspension of the
pendulum is found, at least in ordinary clocks, at around 3 thumbs above the center of the clock face.
Plate 333
One should never make these sorts of works, especially when they are of consequence, without previously consulting a clockmaker who should make the clockworks, and who should give the general
Page 999
measurements of the box in regard to its basic works, and the way in which he wants it to be installed;
thus in working in agreement together, the body should be made for the clockworks and not haphazardly, as it sometimes happens.
After the clockmaker has given the measurements of his pendulum, or at least has marked the
distance which should be found from the center of the clockface to that of the bob, one must be careful of the distance from the floor where the oval hole is found, or better said, the center of the bob,
because the lower it is, the more one will need to raise the first one [the oval hole], for the reason of the
viewpoint being more elevated than the hole. The center of the latter [the hole] must necessarily be
found on a line taken from the center of the bob and bending to the point of view, that one supposes
to be elevated at 5 feet 3 thumbs, and move away from the front of the box from a similar distance.
These sorts of cases are ordinarily ornamented with bronze or better said melted copper that is
later chased and gilded, whether only in gold plating or even in moulded gold foil [it is unclear here
as Roubo describes a foil rather than melted gold that is used in conjunction with the pattern mold
noted below], which is very rare, given that this gold costs a great deal. The copper ornaments that
one puts on the pendulums are the principal cause for the shape of a pendulum, which once deemed
very fashionable is replicated for many subsequent copies, given that a new design requires the making of a new pattern and mold at great cost, since one must first make the designs, then sculpt the
models in wood, on which one makes, with wax, the models for the ornaments, such that they could
be executed in copper. Instead, one can melt ornaments in a mold based on a pattern already made.**
However one makes these clock cases, it is here simply a question of different shapes; they are
made of a form to follow very closely the shape of the work and pendulum, [they are] also [made] in
tapered form; currently they are made where the body resembles a pedestal on which is positioned
the part that encloses the clockwork. Of these different forms, those that are tapered, represented
in Figs. 8, 9, and 12, appear to be the best. That is why I have given here a drawing from the front face
and from its side view. This box, or at least one a bit similar, has been executed in Indian wood by Mr.
Lancelin the younger, who communicated to me the drawing, to which I made some changes that
appeared to me to be necessary.
**All that I am saying here about the ornamentation of the
of this genre. They seem to have been all made in the same
places and the cabinets where you use them. If you pay good
with others that are made by the dozen, without being em-
173
This box is 6 feet in height from the top of the vase, which crowns the pyramid, and [has] the same
height as the clock face of that represented in Fig. 5. It differs not only in the general shape, but also
because its door opens for the full height, and the bonnet is also open on the sides, as one can see in
Page 1000
Fig. 9, [so] all the movement of the clockworks is visible. The bonnet
is removable as in the other box, and [it] has no base, like this last one.
W
The pyramid [tapered waist] and the pedestal of the box in Figs. 8
and 9, hold together so they are more solid. All the borders or mouldings that carry glass, on the door as well as the bonnet, are in copper.
As to its construction, it can be covered in veneer. That which I have
seen executed is made in solid wood, [and was] assembled with all
the skill and firmness possible. The details of its construction would
merit a very long dissertation, which I could not make here without
adding considerably to this part of my work.
almost the only ones where one uses the embellishments of shell and copper [brass]. They are composed of the clock box strictly defined, its crown
and its pendant or reversed finial, which serves to support it.
The body of the box, Figs. 3 and 4, is composed of two frames, of which one forms the front and
Plate 333
174
the other the back. These frames leave a void of around 6 thumbs in the middle of its width, and
Page 1001
are shaped on the exterior according to the design of the work represented in Fig. 2. The void in
the frame in front should be ended 3 thumbs higher than the center of the clockface, and descend
from 34 thumbs downward from this same center. There are clocks, like that in Fig. 1, side F, for
example, where the void is lengthened to the top of the
W
The frames from the front and the rear are joined together by four crosspieces that follow the ex-
terior contour of the case. Those up high are normally arched fully in the center, and those of the base
only curved on the top, as one can see in Fig. 3. The length of these crosspieces should be from three
to three-and-a-half thumbs, taken at the extremity of their exterior edges, which makes about 4 lines
less for these lengths than the interior depth of the box. One must make a groove on the thickest part
of the frame battens and some rebates at the crosspieces to receive the doors on the side of the case,
with copper [bronze or brass] mouldings to retain the glass, as well as the door that closes the opening of the front of the pendulum [clock]. The pendulum [clock] case is enclosed at the top and at the
base by ordinary bottoms. The lower one should be placed in a way that it leaves a place for a double
bottom or floor, which one embellishes with marquetry, which should be movable and placed a bit
lower than the crosspiece of the frame in front, as one can see in Fig. 1. The other bottom is placed at
the center of the squared opening up above, so long as the watchmaker does not wish otherwise. The
crown of the case is composed of three pieces with mitered joints, and shaped to the exterior according to what is required by the design in Fig. 2. These three pieces are glued onto a fourth that forms the
top of the crown. In order for the mitered joints to hold most solidly, one places in the interior some
Page 1002
wooden cleat glue blocks that join the two sides at an angle, and [they] are glued there. The crown is
held to the box by four iron pins placed at the four angles, and which enter into the latter. See Fig. 1,
side G, and Fig. 4, where this crown is viewed on the cutaway.
The corbel is constructed the same way as the crown, as one can see in the figures. When there
is a lot of shape or an arch you glue many pieces together in places where it is necessary, and observing always that they be along the grain line, so the veneer work holds better. The assembly of corbel
and crown of which I just spoke, however simple in appearance, requires you to pay attention when
tracing the cuts, because of the diagonal cuts of the pieces that compose it. Without having recourse
to the theory of the art of tracing, each worker has his method that works successfully enough, especially for the works that are never obvious. The simplest method is to begin by tracing the cuts of the
corbel like that in Fig. 1, side G, and Fig. 4, so that each piece be at a slight angle along their interior
length. When the pieces are placed according to their angle, you trace the joints with the mitered-tri-
175
angle, supposing that the side projections are equal to that of the front. If they are not, like in the figures above, one must trace a plan of these different projections, Fig. 11, so that
at their meeting you have the required cut and the true length of the joints.
The watch-carrying boxes differ from the pendulum cases of which I just
spoke in that they do not have a corbel, they are smaller than half the size of
W
the latter, and their crown opens from above to facilitate the removal of the watch or alarm that is
placed within.
The height of these sorts of boxes is of one foot at most by 67 thumbs in width, and 2 to 2.5
Plate 333
They are composed of two boards of 45 lines thick, between which you glue other pieces of
wood to give it the appropriate thickness, and which are shaped the same as the boards above and
below, which are both laid out cross-grain, to give more strength to the work. Toward the upper extremity of the front piece, you make a round hole of about 2 thumbs of diameter at the most, and you
reduce the interior ridges by chamfering, so that the watch draws close to the copper bezel, which
is applied outside. The entire body of the box should be solid up to about 1 thumb downward of the
opening of the front, and the rest should form a type of box of about 3 thumbs wide, which is closed
by the crown of the case, which is held on the left side with a hinge, and stopped on the right side by a
little lock placed in the thickness of the case.
As these boxes are not made expressly for a specific watch placed within, you fill up what is too
much space in the empty interior by use of wooden spacers that you adjust by reason of the width of
the watch that you wish to place there. It is good to embellish the whole interior later with fabric, in
order not to damage the surface of the watch box.
The watch boxes are ornamented with bronze like the pendulum boxes [clock cases], which I
spoke of above, whether they are ornamented with marquetry or just simply painted and varnished,
which is being done a lot at present.
In general, when the clock cases are prepared for being painted and varnished, they are construct-
ed in the same way as when they are to be covered with marquetry, adjusting the stock thickness as
appropriate to take the thickness of the veneer into consideration after the contours of the design are
given to the work. Instead, the thickness of the paint should be counted as nothing, which requires
finishing the frames as if they were visible, at least as to their exterior forms. When the clock cases
are to be painted and varnished, you build them with white woods, like linden and even chestnut.
That is because these woods take paint better, they say, which I can barely believe given that you see
every day really beautiful works of this type, applied on hard woods. It must be believed that it is for
reasons of economy that we use these sorts of wood, or perhaps they require less preparation and care
on the part of painters. Because they make these sorts of work at such a modest sales price, that they
are obliged to spare on materials and care in making them in order to make a living.
As to the clock cases covered by marquetry, I cannot give here all the details of different forms
that one can give to this, which depend totally on the taste and savoir faire of the artist, and even more
on the expense you wish to spend. Furthermore, this detail becomes very extensive without being
entirely useful, at least at the present, this description of the art of cabinetry being more a recounting
of technique rather than to give examples of the different sorts of decoration we use for various types
of cabinetry.
Before moving on to the description of working with shell and other materials, which enter into
Page 1003
176
the construction of the third type of veneered cabinetry, I am going to give the description of a machine named the movable cross or square, which serves to cut out ovals and even to shape mouldings,
supposing that one cannot make them with a guiding tool, as happens sometimes. This machine has
the advantage of shaping ovals in the same form as wood lathe, which is exceedingly helpful when it
concerns adjusting or embellishing some copper [brass] ovals or other materials made with a wood
W
Page 1004
lathe that are more flattened at their ends than on the main axis [rather] than those which are traced
with a compass.
As to the quarter oval ABC, Fig. 1, (the part in front being taken for the whole) of which the
Plate 334
length of the two axes has been given and traced with the compass, as usual, as indicated by the punc-
Machines
tuated lines of the figure, one must take the excess of the semi-major axis AC, which one carries to
the little B to D. After having divided the distance CD into as many parts as you judge appropriate,
like points a, b, c, d, e and f, you take the distance CD, which you carry to each of these points on the
major axis AC, which touch at points g, h, i, l, m and n. The points of division on the little axis, and
those of the latter, which are their correspondents, are as long as point C is from point D. You draw as
many lines leading to the circumference of the oval, and you give each of these lines, thus elongated,
a length equal to that of the semi-major axis, that is to say, that you make the distances ao, bp, cq,
dr, es and ft, equal to those of AC, or BD, which is the same thing. The distances go, hp, iq,
l-r, ms and nt equal that of BC. Which must be since the distances ag, bh and c, are equal to CD.
Then with the points A, o, p, q, r, s, t and B, you trace a curve more flattened than that which is traced
with a compass. This curve is that which describes the oval curve, of which the mechanism is based
on this demonstration, which serves equally for the construction of the cross or the movable square,
and which produces consequently the same effects, as you will see later.
The movable cross or square EF, GH, Fig. 2, is composed of two pieces of wood assembled
squarely and notched one into the other. In the middle of each piece is carved a channel dug [also]
into the sides. To facilitate the passage of a mantonet [T-nut], ab, [of] Fig. 3, and ef [of] Fig. 6, which
is retained by two little shoulders, but has the ability to move freely, as one can see in Fig. 3, which
represents the side view of one of the branches of the cross, and of this mantonet, of which the middle
is pierced to allow passage of a pivot pin c, which has a head in its lower end, to prevent the mantonet
Page 1005
from escaping, in which it must turn freely. The upper extremity of this pivot pin is held in place with
a cover of iron or brass, (Figs. 3 and 4), into which passes the bar or rod IL, Fig. 2, and side M, Fig. 3,
and which one fixes by means of a pressure screw that does not rest directly on the wooden rod, but on
the sheet of iron or copper attached to the latter, or which holds it only at the width of the cover, both
sides of which are raised to form a T-channel void into which the sliding T-nut fits, so that it cannot
escape.
One could move this cover, making, as in Fig. 5, a pivot that passes across the mantonet and the
wooden bar, which one holds with the pivot by means of a wing nut, and by a base carrying the pivot
pin, above the mantonet. This way of holding the bar with the pivot pin is more simple than the first
one, but it requires that you make a [through] groove in the crossbar to let the mantonet pass while one
advances or pulls back on the bar, according to the different diameters [axes] of the ovals.
The mantonet ef, Fig. 6, should be a bit elongated [oblong], so that when it crosses the meeting of
the grooves of the two branches, it does not waver but engages in the upcoming one before exiting the
trailing one.
One should also take care to round the ends of the mantonets, so they slide more easily from one
Plate 334 . Machines Appropriate for the Tracing of Ovals; the Way to Prepare Shell
178
groove into the other, and that they do not bind at the intersection of the cross. One must always have
two similar mantonets to stop the bar at the cross, because to do this operation, you must have two
centers of movement, as one will see later on.
When you wish to make use of this machine, it is very easy. After having determined the diame-
ters of the oval, like those of NO and PQ, you adjust the cross at the middle of the opening and of the
W
work, where you stop it with wedges, like in this figure, noting that the middle of the grooves correspond perfectly with
the axis of the oval.
Plate 334
At the end of bar IL, one can place there a cutting tool as in the compass with a shaft, of which I
spoke on page 848. One can even adjust the tools for moulding in the same way as the masons adopt
their calibrations [probably referring to the forming of mortar or plaster mouldings with a template].
The cross can be more or less large, according to different needs, and consequently highly proportioned in all its parts. When there is a certain largeness in size, it would be good to place some squares
Page 1006
[braces] in the angles [corners], as I did here, so that there cannot be any movement.
For the construction of mantonets and the parts on which they depend, one could adjust [them] as
one judges appropriate, provided they meet the same requirements, namely that the pivot turn freely
in the mantonet, and that the bar be stopped in the same manner with the fixed pin, and always with
the ability to change place when needed. You must, however, pay attention that the mantonet be of
two pieces on its width, so as to be able to position the pin, and that one could not do otherwise, without making its construction very complicated. So that the rubbing of the pin not use up the wood too
fast, it would be good to position a brass cylinder there, as I indicated in Fig. 3. This could likewise be
179
of two pieces, and held in place with each side of the mantonet, which one joins later with screws, the
heads of which should be counter-sunk into the wood, so that they do not interfere with the passage
of the mantonets in the grooves of the arms of the cross.
When the piece will not be hollowed out, like that of Fig. 2, one attaches the movable [cross] on
top [of the work]. That does not change anything in the manner of operating, if it is not necessary to
W
use bar IL, of which the part that holds the apparatus hangs downward to reach the thickest part of
the cross placed there, that is to say, attached on top of the work.
Plate 334
S e c t i o n III
H ow
t o wo r k t h e d i f f e r e n t m at e r i a l s t h at a r e
M a rqu e t ry ,
S h e l l , I vo ry , H o r n , e t c .
us e d i n t h e co n s t ru c t i o n o f
like
hell is the material that is used most commonly in the construction of this marquetry, whether
one mixes it with copper [brass], pewter, silver or gold, against which it serves ordinarily as the
background. The sheets of shell are normally curved on their surface, as I said above. That is why the
first thing to prepare them for use is to flatten them, which is done in the following manner.
After having made the choice of scutes or sheets that you want to flatten, you heat some water in
a cauldron, Fig. 15, or some other uncovered vessel, which is capable of containing the sheets of shell without their touching the sides of the cauldron, for
fear that the heat will burn them. When the water is boiling, you dip the sheets
inside and let them stay just until they are sufficiently softened, which you will
know when pulling up a sheet with forceps [tongs or pliers] or even with the
fingers (since shell loses its heat promptly). If when out of the hot water the
shell folds easily from its own weight, that is a certain sign that it has softened
to the necessary degree. There is a little press, Figs. 7 and 8, of the width necessary to contain the
largest sheet of shell, and into which you put the sheets thus softened, noting [that] when there are
enough, to put between each sheet some plates of iron or of
copper [brass] about 2 lines thick, [that are] completely flat
and smooth on their surfaces. One should take care to heat
these plates previously, so as to conserve longer the malleability of the sheets of shell that they just acquired by means
of the boiling water.
When the sheets of shell are very curved, and you fear
that they are not ready for the action of the press, whether
because of their curvature or because they cooled too fast when you put two or three sheets in the
press, one must tighten [close] the press slightly and plunge the whole thing in boiling water, which
will resoften the shell. After this you tighten the press fully and you take it out of the water to let the
shell cool down slowly, which is better than immersing it in cold water, which chills the shell too fast
and makes it more breakable and more subject to warpage.
Page 1007
180
When the shell is totally cooled, you remove it from the press and it will be found perfectly straight,
and will always keep this new form, provided that you do not dip it anymore in the boiling water, then
it will again become curved like previously. One must also pay attention that the shell expands in hot
water, but it shrinks when cooling. That is why when you are shaping the shell by pressing in molds,
it is necessary that the molds be a little bigger than necessary, so as to give the shell the means to freely
W
emony. Most are content to place it, upon leaving the hot water,
Plate 334
between planks about one thumb thickness, and close them with
one or two clamps [holdfasts], as in Fig. 14, which is not good, because if the shell is not hot enough, you expose it to possible breakage, supposing that the shell has reached only medium heat, and
it will not remain straight when you take away the weights, which
requires starting the operation again. That is why it is better to make use of a press and of hot irons,
as I just said above. As to the size of the press, it is rather arbitrary, but it would be appropriate that
it be constructed like Figs. 7 and 8, that is to say, that it be composed of two uprights or posts, with a
pressure screw in the middle so that the movement of the press can be made more quickly.
Page 1008
The screw, to be good, should be of iron, and be held by the upper platen of the press (in such a
manner that it can turn freely) so that it be tightened or released as you judge appropriate.
The nut of this screw is placed at the summit above, and you make the screw move by means of a
handle, which passes through the head of the screw, as one can see in Figs. 7 and 8. As to the size of this
press, a foot of length between the two uprights, or 15 thumbs at most, is sufficient to place the large
sheets of shell along their length.
Shell straightens also with fire, which is made by holding it in front of the flame of a bright fire,
making sure to move it in all directions with much haste. If you leave it an instant unmoved over the
flame, it will burn, and it will form on its skin a type of whiteness caused by too much action of the
fire, which separates the layers of the shell and even burns them.
This manner of straightening shell [with fire] is barely used, having as much to do with accidents
that I just spoke of as because many workers state that the fire ruins the color. Knife makers, however,
work with nothing else for straightening the knife handles made of shell, which truthfully, is nothing
but small pieces that are well heated. That is why they are content to straighten them with the flame
of a lit candle.
When the shell is straightened, you work its thickness, whether with a grattoir [a grattoir is a
scraper or scratching knife very much similar to a die-makers scraper; conversely the tool graille
also frequently used to work tortoiseshell and horn, is a specialized file very similar to modern plastics floats such as Vixen files], Fig. 13, (which is a type of blade with two edges, which each one has a
bevel, as one can see in this figure) or simply with the toothed plane, which bites rather well on top,
which cannot be done with the ordinary plane iron, because the shell is composed of a gritty substance that is destroyed rather quickly in cutting it [the language structure leaves unclear whether it
is the shell or the tool that is quickly destroyed; experience suggests both are true], which makes us
prefer the toothing irons, the rasps and the files that only attack the surface evenly overall.
When you want to work the sheets of shell, you begin by straightening out the underside (that
is to say, the side that adheres to the body of the animal) with the toothing plane. You then turn
181
them over to the other side to reduce the thickness appropriately. You make use of rasps, scrapers
[or grailles] and toothing planes. This operation requires great caution and skill on the part of the
workers. Because when the shell is too thick, and there is consequently much material to remove, it
sometimes happens that one breaks the sheet of shell, which, as I said above, is very fragile. You can
only keep it in one piece by treating it with great care, observing especially that the shell is supported
W
equally in all its parts on the bench on which you are working it.
You remove the excess thickness out of the sheets of shell from the side that was to the outside of
the turtle, because the flesh side is normally the most beautiful, and where the variations are the most
obvious, at least that is what the workers say I never saw a big difference.
The thickness of finished shell varies according to the different works where you use them. This
thickness hardly exceeds one line-and-a-half, and cannot be less than three-quarters of a line, since
when it is thinner, it is worked with difficulty its lack of consistency exposing it to being broken
with the least effort.
Whatever thickness you give to shell, one must take great care that it be equal everywhere, so that
it be equally flexible. Before putting it to use, one must keep it for some time in a place dry and even a
bit warm, so that it shrinks before being used. One barely curves sheets of shell before cutting them
out and gluing them with copper [brass], as I will speak of later. If you wish to curve it in a certain
form while in the sheet, you do it in the following manner.
You begin first by tracing the curve that the shell should have when it will be moulded, as for example in Fig. 9. You then trace the full length of the curve on a straight line [establish the linear length
of the contour outline], Fig. 10, to define the correct length of
the sheet of shell, which is good to cut a bit longer than is necessary. You next make the interior and exterior molds, Fig. 11,
of a form similar to that of Fig. 9, creating there, in each part
a groove a-b, into which the shell can expand when the heat
of the boiling water and the pressure of the mold forces it to
spread out.
When you want to mould the shell, you do not put it straight in the mold.
After having flattened and worked the thickness, as I said above, you soak it in
boiling water. When it is softened, you begin to form it by hand so that it approaches the form that it should have. Then, you put it in the mold, and then into the press, Fig. 7.
Close the mold only as much as necessary to prevent the shell from slipping out, after which you soak
the whole thing in boiling water, and you close the screw of the press to the measure such that you can
perceive that the shell is making no resistance. When the mold has closed as much as necessary, and
consequently the shell has taken the form of the mold, you take the press from the water, and you let
it cool as is normally done, which should be done without any difficulty.
Plate 334
The molds of which I just spoke are normally made of hard wood, which is sufficient
when you do not have many similar pieces to mould, or when they are very large. Other-
Page 1009
182
wise, you would do well to make the molds in copper because they will resist better the action of boiling water than those of wood, and they keep better their shapes and they last longer.
Page 1010
It is rather rare that you mould shell before it is cut out. There are occasions where that is neces-
sary, so I believed I was not able to avoid entering into some details on this subject.
I said above that one welds shell without the help of any agent, which is done in a very simple
W
16. You do enough to each piece, observing that the two slants or chamfers have the
same incline. This being done, you place them against each other to see if they join
up well together. The joint being good, you put the pieces together and you surround them with strong paper [usually moistened], which you put in three or four
double layers, and you hold it all with some string. You then heat up some tongs/
pliers large enough so they hold [grasp] the entire length of the joint, which you close with them, just
until you can see that the shell is becoming soft, folded by its own weight, or at least responding easily
to the finger. You then take out the forceps [holding the shell] and let it cool down and it is perfectly
welded.
You must take care that the tongs/pliers are not too hot; they will burn the shell without welding
it. That is why, before closing the joint with the tongs/pliers, you must try them out on paper. When
they only make a bit of paper scorched without burning it; that is a sign that they have reached the
appropriate temperature.
In order to weld shell perfectly, it would be good to have tongs/pliers specially made, of which the
jaws are a bit thick, and be positioned in a way that when they seize the shell and the paper surrounding it, they be exactly parallel, so that their pressure is perfectly equal across the entire joint. It is a bit
like the manner in which the Chinese weld the horns of their lamps, as you can see in the Economic
Journal of the month of September 1756, page 92.
They still weld shell by means of boiling water, which is made in the following manner.
You position the joint in a long roll, as I just said above. Then you put the two pieces in the press,
Figs. 7 and 8, between two pieces of copper [brass], observing that the joints overlap a bit. Then you
close the screw to better fasten the pieces in place. When you are certain that it is going well, you put
all of it in boiling water. You tighten the screw of the press to help the joint bond together while the
Page 1011
shell is softening.
Whatever manner you weld shell, you must take great care that the joints be very sharp and very
clean, because the least particle of grease or dirt will prevent the action of the welding. One must
even avoid passing your breath on these joints, and even touching with the fingers.
Cabinetmakers weld shell only rarely, in part because they do not know how, and in part because
they do not care to; besides it is not the custom. However, this resource is helpful in many occasions
where it is necessary that large areas of shell be made to appear as though they were of one piece.
When you wish to weld two pieces of shell together, one must choose a way such that at the place
of the joint, their colors or their pattern be rather similar, so that the joint not be obvious, at least as
much as is possible.
Cabinetmakers do not use shell completely alone; that is to say, they do not simply apply it directly
on the wood. After having flattened and thinned it, they double [it] with a backing so that the glue and
183
the nuances of the frames [grain of the carcase] do not appear through it. This doubling is nothing but
a layer of black or red pigmented glue spread on the shell on the skin side and [then] covered with some
paper that they apply at the same time as the color, which serves as a mordant for holding the paper.
These two colors are made one with the black from smoke, and the other with some vermilion
both being diluted and ground with fish glue, which is preferred to the English glue. Fish glue holds
W
better and is more clear and fluid, and [it] does not ruin the color of the vermilion, which one should
use as pure as possible so that it gives a more beautiful color to the shell. These two colors are the only
ones to give to shell, at least normally, and the red is the one that is the most used nowadays.
You could, however, enhance the shell with other colors, which would be good in different occa-
Plate 334
There are little pieces of marquetry, like the watch-doors [watch cases] and others, where instead
of shell, you use horn painted from behind, first in vermilion red applied in spots and then coated
completely with some black. This usually makes a bad imitation of shell, and which can hardly be
mistaken with even a little knowledge. However, you could, with a bit of skill, not only imitate shell
by painting the horn from the rear, but also painting various subjects by this method which would
last a long time. For as few as these paintings are well made, it is still a method to enhance the magnificence of these works of marquetry.
When the sheets, whether of shell or horn, are doubled, you let them dry for cutting them out
later, as I am going to explain in the following section, when I will have said something of the work of
ivory and mother-of-pearl.
Ivory being, as I said above, a bony substance, you cannot melt it nor straighten it, such that it
cannot be made ready to be used except by cutting it with a saw along its length and on its thickness,
and always at the expense of the latter, which does not produce much waste however. With a bit of
experience and skill, you can split it in very thin layers without breaking much, an advantage that
you do not have with shell, since it is necessary for working the thickness, to lose some part, which is
sometimes quite considerable.
You saw ivory with a carving saw, of which I gave the figure on page 801. The blade of the saw with
which one cuts ivory should be tempered a bit thinly, and have teeth of medium size and perfectly
equal throughout, so they are not consumed, and they are not bound up in the ivory, which will make
it split.
You saw ivory in two ways, namely, dry and wet. The latter way is the most utilized and the most
advantageous, especially for green ivory, which heats up easily by the rubbing of the saw, which not
only splits it but also ruins the color, which becomes scorched by the effect of the heat. One remedies
this by pouring a bit of water in the kerf without taking out the saw (a bit like the stonecutters do),
which one should do as soon as one senses the least resistance, so as not to give the saw the time to
engage nor heat up the ivory.
Page 1012
184
When one saws dry ivory, as soon as you sense that the saw is beginning to bite in, rub it with a bit
of leather or wool on which you have spread some grease, which will facilitate the movement of the
saw, but which is less good than water, which in filling the same purpose has the advantage of keeping
the ivory fresh.
When you cut up ivory, as well as shell and horn, you place it in a vise. When there are small piecW
es that you wish to cut up, you secure them in a little vise held inside the big one, so that the pieces are
exposed to less stress. An intelligent worker always anticipates this, whether by restraining the hand
that drives the saw, or by supporting with the other hand the bits that are being cut up.
Plate 334
Ivory is worked rather well with ordinary tools, but it would be better to use large and small files
and rasps, of which I made the description on page 937. You also use files for straightening and finishPage 1013
ing off the edges after they were cut out by a saw, supposing that was necessary.
Of all the materials of which cabinetmakers make use during the construction of cabinetry that
I am discussing now, mother-of-pearl is the hardest, the most unproductive and the most difficult to
work. There is nothing but tempered saws, hard stone and files that bite into it. Furthermore, as it is
very breakable it requires much effort and care to work it.
When you cut up mother-of-pearl, you must take great care to saw the pieces perpendicularly to
the surface. Then, you split the thickness, not exactly as much as what you need, but more than is
necessary for smoothing later on the grindstone, because it is not easy to split it straight.
There are large pearl shells that split themselves at their thickest points. They are less beautiful,
and however large it often happens that you can only pull off very small pieces because of the waviness of the layers, or better said, the different layers of which they are composed.
When the shell is cut up, you straighten it and put the thickest part on the grindstone, which is
Mother-of-pearl, as well as ivory, can be made to a thickness of ordinary veneer, that is to say, that
185
you give it all more than one line thickness, so that when inlaying them, they are found to be all of the
same thickness.
I will not speak anymore here of the work of other materials that enter into the construction of
such marquetry, like copper [brass], tin, etc., because cabinetmakers buy them ready made at the
necessary thickness, such that they have nothing to do other than cut them up, as I will explain in the
W
following section.
S e c t i o n IV
The
m a n n e r o f co n s t ru c t i n g
I n l ay
and finishing it
he construction of this marquetry, or better said, of mosaic in metal, although a bit similar
to that of mosaic in painted wood, of which I made the description, page 866, requires however
Page 1014
to be addressed separately, due both to certain ways of working that are different and the quality of
Plate 335
materials that one uses. These materials, copper [brass] and tortoiseshell, are used as in ordinary
Design of a Piece
marquetry techniques, the one forming the designs and the other the background of the work.
Since the materials that one uses for marquetry, even the most common, are of a certain value and
of Marquetry
Appropriate to be
are rather difficult to put to work, we look for ways to spare both handwork and materials at the same
time. That is what we accomplish in cutting out two sheets of different materials (like copper [brass]
Copper
and shell) one on top of the other [simultaneously], in a way that you have created at the same time
two pieces of identical marquetry. The flowers or ornaments of the one can be placed in the void of
the other, of which consequently, you can remove the flowers and ornaments to place them in the
void of the first one. This manner of cutting out marquetry is called working in contra-partie.
Marquetry where the tortoiseshell makes the background of the work, and brass the ornaments
is the most beautiful, and is called la partie, that is to say, the principal piece of work. When the
copper [brass] makes the background and tortoiseshell the ornament, it is called contra-partie, [the
second part] and is less valuable, although one can achieve making it very costly, which I will speak
about in its turn.
When you wish to make a piece of marquetry, you begin by making the general design of the
work to be executed, as, for example Fig. 1, where you take one or many tracings, according to what is necessary the initial design is to be conserved for
using to finish the piece.
Then you take a sheet of shell of a convenient size and a sheet of brass
[in the original French text, here the author is specifically referring to brass
while elsewhere Roubo may be referring to brass or copper and we have often
chosen to use the term brass or copper when it seems more sensible], which
you glue one on top of the other with ordinary glue, not all over but only in a
few spots, which is sufficient for fastening them together. Before gluing the
sheet of brass, it is good to scrape the underside with a coarse file used in different directions, or even with a toothing plane iron, so that when it is cut out,
it bonds better with the glue.
Plate 335 . Design of a Piece of Marquetry Appropriate to be Executed in Shell and Copper
187
When the two sheets are thus joined together, you glue the tracing on that of the brass, and you
let it all dry, after which you cut it out with the marquetry saw in the same way as for mosaic in wood.
Since the ornaments of marquetry or mosaic in metal are ordinarily very delicate and of a form
Page 1015
often very complicated, it is necessary to use extremely fine saw blades. You often start by piercing little holes in places that
When the design is entirely cut out, you separate the two sheets one from the other, whether by
plunging them in hot water, or simply by passing between them a very thin blade of a knife, which is
sufficient, when you take care to place nothing but a few spots of thin glue, and simply that which is
necessary to hold the two sheets together, as I said above. After the sheets are separated, you disassemble them, that is to say, you take out the ornaments for repositioning one piece inside the other,
so each piece remains empty, like that represented in Fig. 2.
In the case of a large piece of marquetry, like that represented in Figs. 1 and 2, it is hardly possible
to make the whole thing of a single piece, both for the difficulty of finding the sheets of shell large
enough and the difficulty caused by the great number of shapes of which this piece is composed.
That is why you first make separately the flat bands of the perimeter, like those of AB and CD, Fig.
2. In the body of the piece you make the joints where the ornaments intersect one another, and where
they separate the background by large pieces, as for example, in the places E, F, G, H, I, etc., which is
Plate 335
188
a thing of experience and of taste. This is especially true for shell where it is necessary, as much as is
possible, that the different sheets not be too obvious.
After the background sheets, whether brass or shell, have been hollowed out, you fill them with
their ornaments, which are not made except in bulk, like the one represented in Figs. 1, 3, and 5, and
Figs. 2, 4, 6.
The first ones are in partie and the second ones are in contra-partie, that is to say, that the base of
the first is in shell, and the other is in brass, as one can see in these figures, and as I already said. Each
partie and contra-partie being thus completed and made even with the outside, or better said, from
Plate 336
the side that should be applied to the cabinet, you glue the whole together on a piece of heavy paper
Manner of Executing
with a good English glue, or even isinglass as with mosaic in wood. You leave it all to dry before apply-
a Piece of Marquetry
ing, which is done a bit the same as for other veneer work, with the exception that one cannot press
Represented in the
with enough force with a hammer to make the glue stick to the veneer. This metal mosaic requires
Previous Plate
being placed in a press with cushions or ticking sacks filled with heated sand or very fine stone, in
Page 1016
order to keep the glue warm for a longer time. It is even good to put some pieces of fabric soaked in
boiling water then wrung out between these cushions on top of the veneer. These hot and humid linens serve to keep the shell moist, which makes it flexible under the weights that you put on top, and to
make it take the glue. You leave the veneer thus in the press just until it has cooled and the glue is dry.
You leave it for some time in a dry place, where there is hardly any humidity to worry about, nor too
much heat, which could make the veneer lift.
When the surface of the cabinet is curved, the marquetry is more difficult to veneer because it is
necessary to curve it in advance of the application to the substrate, that is to say, after the ornaments
and the base have been assembled and glued together on a sheet of heavy paper.
When you want to curve a piece of this marquetry, after it has been cut out and assembled ac-
cording to its design, beginning with the center of the mosaic you hold the piece near the fire to warm
and soften the shell, without however melting the glue, which holds all the different pieces together.
You then position it on the cabinet in the place it should occupy and you cover it promptly with some
linens dipped in boiling water. You put the cushions on top of it, and you clamp the whole thing together as if you wanted to glue it, noting however not to force it entirely closed, whether a press or
bed boards are used. Proceed gradually and gently, so as not to force the shell; it is better to repeat the
process many times to achieve the form required.
Instead of bending the brass and shell marquetry as I just described, it would be much better to
make molds of a similar form to the cabinet contour so that instead of using a cushion you make a caul
of curved wood in the opposite contour to the cabinet. To curve a piece of marquetry, after having
heated and even bent it by hand, one would have only to put it between two molds. Always take the
precaution of putting between these and the piece of marquetry some linens dipped in boiling water,
or even some wool pads of a medium thickness.
This method of bending marquetry is excellent, and was formerly much in use. I do not know
why it is not being used anymore. It is true that it takes a bit longer than the first, but also it has the
advantage of being much more certain and convenient, being able to do all this work in a press and in
Page 1017
front of a fire, which makes the pieces of marquetry keep their heat longer, and consequently lends
better to the effort you are making to bend them.
The upper side of the moulding that I just spoke of can also serve as a fitted caul to glue the curved
Plate 336 . Manner of Executing a Piece of Marquetry Represented in the Previous Plate
190
When the marquetry is glued to the cabinet you let it dry, after which you finish it by first re-
moving the paper and the glue. You then flatten and smooth it with files of different sizes, and then
you scrape it before engraving it to achieve the detail and shading of the ornaments that are still in
bulk form. You cannot have the effect you
W
Although in general contra-partie of marquetry is less esteemed, one could still produce a very
good piece of work. You could even, by sacrificing one part of the shell that forms the ornaments,
make a very brilliant and rich mosaic by substituting horn of various colors and even paint. You could
also cut and position other pieces of shell of which the shades reflect the hue of certain flowers or
other objects that you would like to represent.
This type of marquetry, executed by an able artist, would be very beautiful and would perhaps
have the default that is reproached to the ancient mosaics of Saint Sophia in Constantinople and
Saint Marcs in Venice, where the gilded backgrounds undermine some part of their original intents.
Cabinetmakers do not engrave their works themselves, or at least there are workers who occupy
themselves only with this sort of work. The tool appropriate for engraving marquetry (at least for
metals) are the same as those of engravers of copperplate, namely burins, scoopers, scrapers, a grater
Plate 337
The burin, Figs. 1 and 2, is a piece of steel about 45 thumbs long, and about one line and a half
Different Sorts
squared, like that on side A, Fig. 1, of which the upper end is sharpened on a bevel from one angle
of Tools Appropriate
to the other. There are some burins B, Fig. 2, lozenge-shaped in their cut. They are sharpened in a
for Finishing
Marquetry
Page 1018
diamond-shaped cross-section and serve to make deeper tracks or lines that are less wide than with
squared burins.
The sides opposite the bevel of a burin are called le ventre [belly]. They should be sharpened very
sharp and perfectly straight. The burins are held in little handles of white wood, which are flattened
on the bottom, so that when they engrave, the belly of the burin is held almost parallel to the surface
of the work and only engages the surface as much as you deem appropriate.
Scoopers differ from burins, in that the sharp edge of the ventre [belly] is a little broken down in
192
a way such that they do not have a point at the end, but they present a little face [a secondary bevel],
which is more or less in size, according to the edge of the die-makers file. The use of this tool is to
quickly make large and square cuts in the background. The grater [scraper], Fig. 3, is nothing but a
file with a pointed tip, sharpened at the edge on its three faces. Its use is to remove the burrs on the
brass for the burin or scooper to pass over when subsequently making engraved lines that intersect
W
previous lines.
The deburring tool differs from the grater [scraper], in that it has four faces all sharpened on
edge. Its use is a bit the same as that of the grater [scraper], but we prefer sometimes the scraper for
Plate 337
fine work, because its edges are less sharp. [Note: in contemporary engraving, both tools are called
scrapers, although the four-sided scraper is rarely used.]
about 34 thumbs long, tapered and curved toward its tip. The crosssection of this tool is in the form of a very elongated olive. Its use is to
erase the scratches made on the brass, in pressing together the parts on
either side of the scratch, or better said, in compressing the material on
itself to fill the scratch.
must design them to be made on the inlaid brass, or whatever other materials that might be used, but it would be incorrect to do this with the
stylus directly on the mosaic. It is better to use the following technique to trace them from the same
design, or on a well-made copy, provided that all the shapes of the design are well rendered.
For copying thus the ornaments, one must coat the brass with a light layer of white wax, which
one applies after having heated up the brass a bit. When the wax has cooled, you place the tracing
over the element, after having previously rubbed the underside of the paper tracing with some pulverized blood-stone or with pencil lead. With a sharp point, you trace all the contour of the design,
which is then applied onto the wax.
Page 1019
You only outline ornaments like this when there is a rather large amount to do. For the little pieces, you content yourself to design them in place.
As to the way to engrave on brass, it is a bit the same as for ornaments in mosaic that I spoke about
before, page 884. That is why I am not going to speak about it further here.
Also, the detailed instruction of engraving with a burin more properly belongs
to the art of engraving, and would require a very long description to be done
with the precision and clarity that this technique is capable.
Brass [copper], tin, silver, gold and mother-of-pearl are engraved with a
solid burin, represented in Figs. 1 and 2 of this plate. Shell and horn can hardly
be engraved unless using a hollowed burin [similar to a V-groove chisel or veiner], as one can see in the figure in Plate 299, Figs. 10, 11, 12, and 13. Since these
materials are composed of many layers positioned one on top of the other, the
solid burin would split them, no matter what precaution is taken. This does
not happen when one makes use of the hollowed burin, as I demonstrated in
speaking of the use of this tool on page 884.
193
When marquetry is engraved, you fill in all the details with melted pitch, or some other stuff that I
spoke about on pages 863 and 887. You then finish with a polish, which is done in the following manner.
You take a very smooth pumice stone [that has been] dipped in olive oil. You rub the work in all
directions, noting not to apply too much oil, which, in soaking into the veneer, could lift it. You then
take some of this same pumice stone reduced to powder, which you sift through a very fine strainer.
W
You mix in a bit of oil and you rub it on the work with a polishing wood, pressing strongly from above,
and making the latter move in different directions. In place of a polishing wood, you could use a polishing rush, especially in the flat parts, or even a type of solid wood of medium hardness, like pear or
another wood of this type, which spreads this polishing compound well.
Plate 337
After the pumice stone, if the marquetry is composed of ebony and brass, you would use charcoal
After the pumice stone and charcoal, you make use of rottenstone, pulverized and diluted with
a bit of olive oil, and applied on the work with polishing wood or a pad stuffed with buckskin. Then
you make use of dry rottenstone and Spanish whiting next, for finishing the polish on the piece, and
for leaving no spots.
Whatever the materials you use in the construction of marquetry or metal mosaic, you always
follow closely the same procedures of which I just described. There is no difference other than some
precautions that one must take by reason of the nature of these different materials, in the detail of
which I will not go into because it is a business that would make too long a discussion, without being
absolutely necessary.
When you add silver or gold in the construction of marquetry, one must take great care in cutting
out these metals to surround the base of the vise with a bit of sheepskin onto which may fall the sawn
bits and filed bits, which are very valuable given the costliness of these metals, especially that of gold,
which is valued at close to l00 pounds/ounce.
Since in general, metals do not take glue very well, one must, when there are not large pieces,
hold them in the wood with little nails of the same metal, which you drive as accurately as possible, of
which you bury the head a bit so they join up better. You will, as much as possible, place these nails in
the bronze ornaments, or even in the scroll of the engravings, in such a way that they are not visible.
Instead of a wooden mosaic or brass one, of which I just made the description, you sometimes
directly cover the frame works with lacquer or varnish from China or Japan, of which normally the
background is black and enhanced with raised gilded ornaments.
The lacquer that one uses normally for cabinetry is taken from the panels of a cabinet or screen
that has come from China or Japan. For the most part these are varnished and painted on both sides,
and which you split in the middle of its thickness to be thinned later with a plane and adjust them
enough to be placed on the carcases of ordinary woodworking. One must take many precautions in
splitting and thinning these panels for fear of melting or cracking the varnish. That is why, when you
split them, you must put them in the press between cushions or wool coverings. You must take the
same precaution while planing from behind, that is to say, you must put on the bench a cover folded
double, so the unevenness on the surface of the varnish at the locations of flowers and other ornaments go into the thickness of the padding.
In thinning the wood panels that have lacquer or varnish from China, you must take care to leave
enough thickness so they do not break. That is why one must leave them with at least one line of
thickness. After that you place them on the surface as is normal, while taking the precaution to let
them heat up as well as the frames that are to receive them, and to place on top of the lacquer some
Page 1020
194
coverings, on top of which you put cushions or wooden cauls with bed boards or glue presses, as
you judge necessary. You must never make use of holdfasts for making these assemblies, for fear
that in pressing on top of them, you [will] disturb the varnish and make it split.
Page 1021
As much as possible, you surround the edge joint of the panels of lacquer with ornaments or
frames of brass, because whatever precautions you take in cutting the sheets of lacquer, it is difficult
W
not to make some splits, which appear at the edges, and which gives a disfiguring effect. Furthermore, even when you cut lacquered panels with all the tidiness possible, the edges of the openings
that could not be trimmed will slowly be ruined by lifting up, which will always look bad.
Plate 336
In France we imitate the varnish from China (at least as well as is possible at present [with Vernis
Martin, an imitation lacquer]), which makes the work appear more solid than those on which we
apply panels with the varnish from China [recycling disassembled and retrofitted Chinese pieces].
In the latter case, that is to say, when you varnish the pieces, you must take great care that their
frames be constructed of very dry good wood, with all firmness possible as I directed in the course
of this work.
What I just said concludes, just a bit, all that is essential to know touching on the practical the-
ory of marquetry or mosaic in metal. With that I have now finished with the description of this last
type of veneered cabinetry along with that of the two other types, having presented general rules
applicable to all cases without giving all the particular examples of the application of these same
rules.
I will not speak any further of the mixing of different materials that enter into the construction
of marquetry because this subject is inexhaustible and this mix depends entirely on the taste of
the artist, the nature and the opulence of the different cabinetry. While such works make for very
great magnificence, they depend for their construction on the labors of different workers like the
cabinetmakers, the game table makers, the marble workers, the painters, the gold workers, the
stone setters, the chiselers, etc., such that cabinetmakers do not possess all the knowledge relative
to these different talents, which is not entirely impossible to find, however it is rare.
As this level of detail is beyond my abilities, I am going therefore to finish this section by giving
a general idea of the different types of marquetry where you make use of the mosaic of which I just
spoke. I will enter into some detail on bronzes [hardware and mounts] with which you ornament
the works of general cabinetry, which will end this third part of my work. I will then proceed to
the description of the art of treillage [trellis-making], or woodworking for the garden, which will
be the subject of the fourth and last part of the art of woodworking, as I already announced at the
end of the description of the woodworking for carriages, page 596.
Page 1022
The art of mosaic is from the most ancient antiquity. It has taken its origin from Asia, from where it
was transported to Greece, and from there to the Romans and all the rest of Europe.
This Art, as well as all the others, was very simple in its origins; it had as its object the decoration
of the floors of apartments, and consisted of an arrangement of some variously shaped bricks, or
stones, or colored marbles, placed symmetrically on the place of these same floorboards. Bit by bit,
this art was perfected and evolved into representing various subjects by employing stones of a very
195
small size, of which the various colors render the latter with much accuracy.
At the fall of the Roman Empire, this Art was annihilated along with many others. It was not until
after a very long expanse of time that many masterpieces of mosaic were found among the rubble
of buildings of the ancient Romans. Thus, it was that we admire the famous mosaic of the Temple
of Fortune at Praeneste or Palestrine, made under Sulla. The mosaics from the house of Cicero at
W
Tusculum, today called Frascati; those of the town of Adrianne, close to Tivoli, formerly Tibur, all
merit the attention of connoisseurs and are seen as having such perfection, which the ancients have
imparted to the art of mosaics.
They have also found some mosaics in the ruins of Herculaneum, [a] Greek colony, and swal-
Plate 336
lowed up by the lava of Vesuvius at the beginning of the Christian era. One of these is composed of
very small stones and is but 18 thumbs long by 12 in width, in which represent four masked figures of
which one dances, another plays the tambourine, another plays two flutes and the last the castanets.
They also have found a wardrobe in which were many manuscripts. This wardrobe, of which
only fragments remain (the rest having been destroyed by the passage of time and by the heat of the
materials that covered Herculaneum) was made of marquetry in various shapes. It is the only monument that remains to us of the marquetry of the ancients, who surely made as much progress [in
wood] as mosaic in stone and in marble, but of which we have only found one in existence, as well
as this last one, because the woods do not last as long as stone or marble. They have also found some
mosaics in France, near Nimes, at Reims and at Cepoy, near Montargis, which, without being as
beautiful as those found in Rome or nearby, make us understand that this Art was spread over all the
Roman Empire.
The first type of mosaic about which I am going to speak is comprised of tinted glass [and used]
at the time of the low empire. This mosaic was composed of blocks of glass, doubles of one colored
sheet and glued on top, of a nature that it cannot be separated except by means of boiling water (which
returns us to the horn used in the contra-parties of marquetry). Most mosaics of the Church of Saint
Marc in Venice are of this genre. They are composed of little cubes or pairs of colored glass, and were
executed by the Greeks from Constantinople, around the years 1071 to 1092.
This mosaic had its birth with that commonly called Roman Mosaic. This is the most beautiful
of all, [and is valued] for the precise use that was made for all sorts of subjects, the color and for the
shape. It is composed of little parallelograms of inlaid enamel, or better said, placed on a base of mastic, as I will explain afterward. This type of mosaic had its birth in Constantinople, where they made
it since the 9th century. It was perfected in Rome, and one can still see the most beautiful works of
this genre, including the portrait of Pope Nicolas IV made in 1239. Works of this type are still being
made in our own time in the church of the Vatican, otherwise called Saint Peters of Rome. The
masterpieces of painting with which this magnificent temple was ornamented are for the most part
copied in mosaic here, with an astonishing realism.
Not only have the artists who made the modern mosaics rendered with precision some large
paintings, such as the large tableaus in Saint Peters in Rome, which are up to 16 feet high by 89
feet wide, but also some little trestle tables representing flowers, fruits and even portraits, like, for
example, that of Queen Christine of Sweden, which is at Saint Peters in Rome, above the tomb of
this princess. The mosaic of Pope Paul V, of which the face is, they say, composed of more than one
million, seven hundred thousand pieces, which must surely be very tiny.
The enamels from which the Roman mosaics are made began with some glass that is melted, and
Page 1023
196
in which they mix some colored materials, whether minerals or metals, which in blending with the
glass [then] stain them various colors and render them opaque because they do not vitrify entirely.
They make enamels in little flat pans of 56 thumbs in diameter, and 67 lines of thickness, like the
Page 1024
enamels of Holland and Venice, which pass as being the best, and of which the enamellers of Paris
make use. For large mosaics they melt the enamel onto panels and cut them up, while they are hot,
W
into little parallelograms of 34 lines square, by one-and-a-half to two thumbs long. Others cut the
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Plate 337
Metal Casting
he most concise description of metal casting might go something like this: Make a void in some
inert solid material, and fill that void with molten metal. In this exercise of replicating a drawer handle,
the inert solid material is a foundry sand known as Petrobond, a fine sand with an oily binder holding it together, and the molten metal is pewter for ease of use and safety while we were trying to photograph the pour
in real time. The following description provides a simple, perhaps even simplistic portrait of the process. For
those serious about pursuing casting, you can find some excellent resources that will further describe these
techniques with little searching.
In keeping with many of the annotations in this volume, this exercise is not a slavish copying of Roubo or
the techniques of his time, but it is rather adapted to make the technique appropriate for our modern age. In
Roubos time, for example, the sand used would have been green sand, a careful mix of fine sand with a small
Fig. 11.3 Press the gently sifted sand into the flask,
throughout.
Fig. 11. 4 Using the butt end of the ram, pound the
Fig.
complete.
11.6
197
enamel while cold by placing them on a steel peen wedge [probably part of an anvil], and hitting the
enamel directly over the steel peen wedge with a hammer of which one side is sharp and of tempered
Page 1024
steel, the same as the peen wedge. Or while someone else is holding the enamel, you hold the sharp
peen hammer with your left hand and you hammer the enamel with another hammer you are holding
in your right hand.
When splitting enamel, you must pay attention that the parallelograms be of a conical shape,
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
Plate 337
portion of clay and just enough water to make the whole mass stick together. Another difference includes the
use of electric burners rather than coke or charcoal hand-pumped blast furnaces. For this exercise I used a
high-temperature glass cooking pot crucible to hold the molten metal (pewter). (In higher-temperature metal
casting, a ceramic crucible is still utilized). One other component of metal casting, the two-part topless and
bottomless box used to hold the inert solid material remains essentially unchanged in concept from the
ancient days. This box is called a flask, easily fabricated by anyone with even basic skills. You can still use
the traditional methods, and in fact I do, on occasion.
Here is the one and only safety note in this entire volume: Working with molten metal is an inherently
dangerous thing. If you do not know what you are doing, dont do it! Seriously. Even with pewter melted on
a kitchen hot plate, maiming and disfiguring injuries can occur. With hotter materials, such as bronze or even
aluminum, a simple mistake can prove fatal. When I am casting bronze I simply will not allow any bystanders
to be present, even if it is my writing partner just trying to get good photographs.
The first of many ingredients necessary for metal casting is a three-dimensional pattern that will estab-
Fig.
11.7
Fig.
11.9
moved.
Fig.
to remove them.
on the bench.
11.12
198
so that they join up better at their upper ends, which is the face of the tableau, and so the glue
makes a connection between them. You must however observe that they not be too conical, because in polishing the work, you may uncover the joints, which are essential to be as fine as possible.
The tableaus in mosaic are made on stone tables excavated out over their entire surface to a depth
of 23 thumbs, according to the size of the work. The bottom of the area is then cut with channels
W
excavated in the shape of dovetails and carved in various directions to retain the glue on which you
place the enamels. This is done in the following manner:
You begin by placing the stone vertically next to the picture that you wish to copy, and of which
Plate 337
you trace the contours on the glue. You take some enamels from a box where they are arranged according to the gradation of colors, and you place them on the glue, where you press them down with
a wooden hammer, observing that the surface of the mosaic be as uniformly flat as possible. When
the work is finished, you let it dry, then you polish it by rubbing it with some pieces [of] or a full sandstone grinding wheel held on a plank that is called a polishing wood.
This plank ends at its extremities in the form of a round handle held by two workers moving it
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
lish the void to be filled later with molten metal. In some instances, the creator of the pattern must consider
shrinkage inherent in metal as it cools from liquid to solid. That is why patternmakers such as myself use
shrinkage scales when engaged in the fabrication of new components that require precise tolerances.
For this exercise however, we made a simple replication of a standard drawer pull.
First place the pattern, in this case an original pull, on a board and dust all the surfaces, board included,
with an even application of parting compound dust. This makes certain that the sticky sand will separate
from the pattern easily when the time comes.
Once the surfaces are dusted, place the bottom half of the flask on the board, upside down. In the vernac-
ular of the craft, this unit is called the drag. Gently sprinkle some of the casting sand into the flask, and pack
it tightly around the pattern.
Then fill up the flask to the top and pound the sand firmly with a ram until it holds together firmly.
Using a straight wooden stick, strike off the excess sand such that the surface is exactly level with the
Place another support board on top of the filled flask, and gently flip it over. Remove the board and set it
aside.
With a stiff brush clean away any sand that is now on top of the pattern to reveal a complete underside of
the pattern.
With your fingers gently press the almost-always undulating surface so that it is firm and smooth.
gates.
199
over the mosaic, and who sprinkle from time to time some sandstone powder to hasten the work. You
check the mosaic to see if it has any faults, which you repair, whether by adding other enamels or just
some glue composed of wax and enamel powder chosen of a convenient color. You end by polishing
with an emery board and some linseed oil, a bit like how you polish glass.
Small subjects of mosaic are made on a table of sheet steel, of which the sides are lifted up about
W
two thumbs. The inside of this table is strewn with little burrs, which adhere to it, and which serve
to hold onto the glue, which must always be kept fresh and in a state of softness appropriate to receive
Page 1025
the enamels. That is why for large mosaics you only add glue as the work advances. Each time you
leave the work, you cover it with moist linens, which prevent the glue from drying out too fast.
Other than the mosaics of which I just spoke, there is still another type that is named Florentine
mosaic, which is constructed with pebbles, marbles, hard stones and even precious stones, that one
uses in little pieces of 34 lines of diameter more or less, like with other mosaics, but rather in their
entire size according to the variation of their colors and the forms of the things they are to represent
can permit.
m,,,,,,,,,,,,,,,,,.,,,,.,,l;',,.,.,,,,,,,,,,,,,,,,,,,,/
When this is completed, dust the surface completely with more parting compound.
Place the second half of the flask, known as the cope, in place on top of the now-right-side-up drag.
The cope and the drag by necessity always have alignment pins or some similar arrangement to make sure
they always go together in the exact same manner. Again, gently sprinkle the casting sand over the pattern,
and pack it tightly over the pattern with your fingers.
At this point, place two large dowels vertically and fairly close to the pattern by pressing the dowels down
into the drag half perhaps 1/2" or so and fill, ram and strike the flask. These dowels will form the channels
or gates through which the molten metal will flow into and out of the mold. When the flask is filled and
struck, carefully withdraw the large dowels, leaving behind cylindrical voids through the cope all the way to
the face of the drag.
Place a support board on the now-completed two-part flask and turn it up on edge. Carefully separate the
two halves of the flask. Leave the cope on edge and lay the drag back down.
Using a tool as simple as a piece of looped wire or bent sheet metal, cut a channel or sprue from the im-
pression left by each dowel to the pattern. Loosen the pattern with a few light taps and remove it carefully
from the sand in which it is embedded. You want this to go perfectly, as the quality of the void left by removing
the pattern will determine the quality of the mold and thus the casting itself.
Check to make sure there are no loose grains of sand and that the packed sand is firm, especially at the top
of the gates.
the drag and the cope again, then lay them both down together,
right side up. You now have a completed mold ready to be filled
with molten metal.
Fill it with molten metal, pouring the metal into one gate and
Once the metal has cooled and solidified, you shake out the
sives and some careful polishing, the rough casting can be turned
Fig. 11.16 The completed casting, ready for finishing with saws, files, polishes and patination chemicals as necessary.
Plate 337
200
This mosaic technique was invented in Florence. The beautiful octagonal table that you see in
one of the rooms of the gallery named La Tribune carried the coat of arms of the Grand Duke Ferdinand II, and was made during his reign of 1621 to 1670 when he died.
This mosaic is the most precious of all for the richness of the materials that are used and the time
necessary [because of] the difficulty of the work. The hard stones that are used could only be cut with
W
saws of which the blade is of copper and has no teeth, the same used by the stone sawyers. To assist
the action of the saw, you can use emery granules with water, instead of sandstone, which is used by
the latter [stone sawyers].
Plate 337
One of the great difficulties of this type of mosaic is to find the stones of which the nuances match
well with the shade of the design that you are copying. This required the artists working on these
beautiful works to have an infinite number of cut-up stone of all colors and shades possible, to be able
to choose those which appear to be the more appropriate. In Florence you can see 125 different marble samples and 125 different cut stones of various types.
It is only in Florence where they make this type of mosaic. All the workers who work there,
numbering approximately 40, are on the account of the Grand Duke, and do not do much work. This
is not surprising given the great difficulty of this art, which makes these works extremely costly. A
square foot of mediocre execution is valued, they say, at 600700 pounds.
They have done this work of mosaic in Paris, in the Royal Manufactory of Gobelins in the time
when the famous Mr. Boule [sic] made these beautiful works of cabinetry of all types. But after the
death of Louis XIV, or more importantly, of Mr. Colbert [who was the minister of finances under
Louis XIV, and who improved French manufacturing and regulated the guilds], these beautiful establishments were abandoned, and the work of mosaic in hard stone returned only to the Florentines, who have conserved the art with honor to this day.
I do not believe I could address this topic without giving a little detail touching on mosaics, both
ancient as well as modern, given the great rapport that the latter has with marquetry or mosaic in
metals, which are sometimes used. As to the rest, those who wish to have a knowledge more detailed
of these works of art, could consult The Dissertation on the Making of Mosaic, by Mr. Fougerous de
Bondaroy, The Voyage of a Frenchman in Italy, the Encyclopedia [this one might be Diderot], and the
Dictionary of Commerce, under the entry mosaic.
Of whatever nature the works of cabinetry are, they are almost always ornamented in bronze, whether inlaid from above, as with foliage, the clasps, garlands, etc., or even when they are part of the same
work and they are sometimes embedded or form frames, compartmented borders [borders to sections], etc.
The work of the first types of bronzes depends absolutely on the modelmaker, the caster and
the other types of workers who finish it. The first, that is to say, the modelmaker and caster, should
not work except in concert with the cabinetmaker, who should direct their operations regarding
the shape of the bronzes, especially on their interior surfaces where they are applied to the cabinet.
They should follow these contours very exactly, and should also decide the relief or projections of
these same bronzes, and the places where they should be cut to facilitate the various openings in the
Ornaments in Bronze
201
These bronzes, Fig. 6, are normally attached with little nails of brass, or even more solidly
with some wood screws with gold-plated rounded heads, which you
place in locations that are less obvious, like the bases, the reverse of
leaves, etc. In the case of large ornaments, you do very well, if space
W
Plate 337
pass through the work, to be fixed on the inside with some nuts, as
represented in Fig. 8.
Brass mouldings are made in two different ways, namely with a wire-drawing die-plate and by
casting with molds. Those that are made with a wire-drawing die-plate are the quickest made and
the most perfect. But they have the defect of being ruined every so often by the pressure from nippers or pincers, which bite into them. Furthermore, when you need a moulding that is a bit shorter,
and where the profile is not found in the wire-drawing die-plate, which often happens, it is better
to make a pattern for the moulding that you need and to cast it, or to be more exact, to make it of
melted brass, which costs much less than to make a custom die plate.
Cast mouldings have their difficulties, however, because they do not usually come out perfectly
clean, and they often have on their surfaces some unevenness or cavities that you cannot repair except by inserting [brazing] tiny pieces of brass and adjusting [them] in the cavities as best as possible.
For the cast mouldings to work well, you must use their patterns so they do not leave more
material than is necessary, especially on the side of the mold that will be visible. [Essentially they
are casting only a thin metal veneer of the moulding contour, not a solid moulding.] These patterns
should also be made with all their returning angles relieved or
tapered, and in general all their corners and ridges should be a bit
rounded so that when pulling them out of the sand mold, you do
not lift up any parts. It is also good that these patterns be made of solid and smooth wood, and when
they are finished, they are varnished to fill all their pores, and prevent them from being stuck to the
sand mold (base of the mold). See Figs. 9 and 10, which represent one of the profiles of a finished
mold, and the other of the pattern of the same mold.
When the mouldings are cast, you adjust them with files and rasps of all sorts, which I spoke
about above, page 935. For them to be most perfect, one should use bent riffler files [or die-makers
Page 1027
202
sinker files] like we use sometimes, but also fasten these same files to stocks similar to those of moldmaking tools, which would make the work much more valuable, and at the same time require less
time to finish.
As with mouldings of brass made by a wire-drawing die-plate [a steel plate with a hole shaped as
an exact negative of the cross-section of the moulding; the moulding is formed by pulling a piece of
W
tempered brass wire through the hole, edges of which form the moulding profile], and those that are
cast, each has their difficulty, and you search for a more expedient means to make them quicker and
cleaner. We have not found anything other than to turn on a lathe a circular brass plate with an edge
Plate 337
profile similar to the moulding that you wish to make. When you have turned these circular profiles,
you cut them at whatever location on their circumference and then straighten them out, such that
you make a straight line, which makes very beautiful real-looking mouldings. This does not always
succeed well because you cannot make mouldings other than those with a very low profile given the
Page 1028
difficulty of straightening them, especially when the diameter of these circles is small. In such cases
it sometimes happens that they break on the inner line of the ring, or at least the moulding does not
become perfectly straight in its length. That is why I believe it is better to hold onto the first two ways
of making mouldings of brass. I spoke of this third way of making them, but it was only to indicate a
resource that would sometime be useful in certain occasions.
I will not extend myself further on the construction of bronzes, because their work is the special-
ty of other workers than cabinetmakers, at least normally, and that will be treated in other arts with
all the extent appropriate, which I cannot do here without diverting myself from my subject.
Since it is the cabinetmakers who make the casting molds, I cannot overemphasize the impor-
tance of taking all the care and attention possible in both removing the pattern from the sand mold
and cutting the sprues to allow the molten materials to flow into the proper spaces, as I just said up
above, which can be applied to all sorts of works of the same nature.
After having summarily treated the different sort of bronzes, there only remains now to treat the
different ways to solder the different metals that are used in cabinetry, which will be the object of the
following paragraph.
III. The way to solder the Metals which one uses for different
works of Cabinetry
By the term to solder we understand the action of subjecting or keeping together two pieces of the
same or different metals, and that with the help of an agent or composed metal, named solder, of
which I made the description above. ***
There are two ways to solder, the one called cold solder, or more properly, soldering with an
iron, and the other called hot solder, which is made by means of a coal fire or a burning lamp [a
torch].
*** In general, all metals are soldered as I just described.
means of brass, you do not call it a solder, but a braze [or hard
which when you unite two pieces of iron without help of any
203
You rarely cold solder except for pewter [or tin], whether with brass or with itself, because since
pewter [or tin] is a soft metal it would melt quickly if you put it on a fire. You only use hot irons, which
Page 1029
melt the solder, of which the heat is sufficient to heat up the pewter to the point that it incorporates
easily with itself.
Soldering irons are of two sorts: the ones, like Fig. 12,
W
Plate 337
When you solder with an iron, you must have next to you a brick or a type of palette plated with
tin, named an etamoir, on which you rub the iron from time to time, after having made it touch a
piece of refined rosin [as a flux, as with rosin core solder available today], by means of which the
solder attaches more easily.
When the solder joint on a piece of tin should be readily visible [on the finished work], you must
make use of the same tin as the piece for soldering, so that when it is polished, the solder joint does
not appear in any way.
When you solder a piece of brass to a piece of tin, you always do it as I just described above, ob-
serving only that it is good to heat the piece of brass a bit, and to rub the contact surfaces with salt of
ammonia [ammonium phosphate or sal ammoniac], so that the solder of the tin attaches better.
The second way to solder, the hot solder, is done by means of a coal fire or with a torch, as I said
above. In either case you always prepare the pieces to solder the same way, which is done in the following manner.
You begin by positioning the pieces as they should be placed,
whether edge to edge or overlapping (which is the best way, whenever possible). You tie the pieces together with some soft small
wire, in such a way that they cannot move, like in Fig. 11. You moisten the spot that should be soldered
in order to attach there the solder tinsel, and especially the borax flux, and you cover the entire joint
area. You place the piece thus assembled on hot coals in an iron pot, and you cover it up with other
coals in such a way that you leave visible the place that should be soldered. When it is all assembled
like this you blow on the fire in directing its action such that the part that should be soldered heats
equally and thoroughly. This operation requires much attention and practice, because as soon as
Page 1030
204
you see the solder melt (which is done quickly) you must stop blowing immediately, otherwise you
would run a great risk of melting the piece, or vaporizing the solder.
When the pieces that you wish to solder are of such a nature that they cannot be tied together, but that the solder joints are at their ends [or edges], or these
same pieces should be put end to end at an angle, like Fig. 15, you
W
wish to solder. See Fig. 16, which represents the string of wire
onto which is placed the part of the frame represented in Fig. 15.
When the pieces are too small to be soldered on a large fire, you
Since the pieces to be soldered with a torch are for the most part very small, and only a very small
amount of borax is needed, you put the latter in a box of tin or of brass, Fig. 17, called rochoir, at the
base of which is a little pipe with a very small opening for dispensing a very small quantity of borax at
one time.
If it happens that the pieces that one wishes to solder together have some means by which you can
hold onto the joints, you could solder with pewter, which I am going to teach.
You first begin by heating up the joint on which you rub salt of ammonia [sal ammoniac], and
which you tin-plate with the iron, Fig. 12, observing to well extend the tin in all parts of the assembly. You do the same thing, if possible, to the two pieces that should go together. When they are well
tin-plated, you heat up both of them just until the tin is almost melted. You assemble them precisely
and then let them cool. They are then perfectly soldered on the interior of the joint without being
visible at all on the outside.
Page 1031
This way of soldering pieces together is very advantageous because if you wish to separate them,
you would only have to heat them up to soften the tin, which one cannot do with other alloys of solder.
There you have it, a summarized version of all that pertains to the manner of soldering the dif-
ferent metals of which we make use in cabinetry. I am going to end this chapter and this part of my
work with a description of varnish appropriate to put on works of copper, known in France under the
name English Varnish.
205
than here, since I just finished all regarding bronzes and brass mouldings with which you ornament
different types of cabinetry.
As you cannot find this varnish at any merchants, and that those who know the composition make
it a secret, I believe that I ought to give here a detail of the ingredients of which is composed, and the
manner of thinning them in liquor, made in the following manner.
W
You take a half ounce of yellow amber that you reduce to a very fine powder and sift through fine
silk; a half ounce of gum of seedlac, put in a powder like yellow amber, nine grains [whole grains
of saffron] of powdered saffron; 10 grams of dragons blood tears [a dark red resin] crushed, and 10
ounces of alcohol perfectly distilled, which is made in the following way:
Plate 337
You put a pinch of gunpowder in an ordinary spoon, and you fill it with the alcohol that you want
to proof. You light the alcohol with a piece of paper. If it is good it must be entirely consumed at first,
the gunpowder burning as well as if it had not been soaked [another way of saying that the gunpowder should burn like it was dry]. If on the contrary the powder does not burn suddenly, this would be
a mark that the alcohol is not perfect enough to make this varnish.
All the ingredients being well chosen, you take a pint bottle, dry and clean, in which you pour in
the alcohol and yellow amber, observing to stir the bottle to mix the two ingredients. You then plug
up the bottle with a moistened parchment draped over the opening, which you bind around the neck
with a well-tied thread, and you make a hole in the parchment with a pin that you leave there, to give
some air to the bottle as necessary. You put on the fire a cauldron of a size convenient for containing
the bottle, and you put some hay at the bottom of the cauldron so the bottle does not touch in any way
and remains isolated in the middle of the water, which you put into the cauldron in sufficient quantity
so that the bottle is well immersed. To ensure that the bottle cannot turn over, you fasten the neck to a
bar placed across the top of the cauldron, below which you make a fire sufficient so the water is always
very hot, without boiling.
As the water heats, you remove from time to time the pin, so that the alcohol that is gassing out
does not break the bottle, which is returned to the cauldron every half hour, and always kept warm
close to the fire, for fear that the cold air [might] break the bottle. Each time that you take it from the
cauldron, you must agitate it for a moment, observing always to remove the pin to give it some air. You
continue the same maneuver for 45 hours, after which there is no more fire, then let the water in the
cauldron cool down, and consequently the bottle.
When the bottle is completely cool, you take it out of the cauldron, and you open it to add the
other ingredients the lac, the saffron and the dragons blood and you close it up again like before.
You stir it up and put it back in the water, which you heated a bit previously and you heat again like
before for 45 hours. You let the bottle cool down without stirring it any more. At the end of 45 days,
you empty very slowly the varnish into another bottle as it will clear. What remains, you can filter
through fine linen, after which you close well the bottle.
If you wish to make a larger quantity of varnish, you must take a larger bottle or other glass con-
tainer. Note always that the varnish only occupies a quarter of the capacity of the vase, without which
capacity it would break when the ingredients are heated up.
If you wish to gild with silver or tin with this varnish, you must double or even triple the dose of
saffron or dragons blood. As to the manner of using the varnish, you do the following:
When the piece of [metal mosaic marquetry] furniture is to be well polished, you must degrease
it by rubbing with talcum, diluted in a bit of alcohol. Instead of talcum, you can use Spanish whiting,
ground to an indiscernible powder. You heat the workpiece on a tin plate placed on an oven, ob-
Page 1032
206
serving that the workpiece is well heated all over and to a degree of heat such that you cannot easily
touch the surface with the back of your hand. You pour some varnish into a little vase, in which
you dip a brush of very soft badger hair. After having wiped off a bit on the edge of the vase, you
brush it lightly over the piece without much pressure. You must do it rather carefully, so that the
Page 1033
renewed and overlapping layers do not become obvious, and that there be no unevenness or spots
W
If, however, it happens that there is some waviness, you can make them disappear, at least partly,
by approaching the varnished piece with the heated tin plate from the side where the waves are,
Plate 337
If it happens that you wish to varnish some pieces which, by their shapes or for another reason,
cannot be heated up in this manner, you varnish them cold and immediately afterward place them
near the fire so that they are heated uniformly in order to bake the varnish and give a luster to the piece.
When the polish becomes soiled, you clean it with some tepid water and white fine linen. You
must never use any stones or polishing powders, like Spanish whiting, tripoli, etc.
This varnish is hard and very beautiful, it masks the smell of copper [brass], giving it a beautiful
brilliant color approaching that of gold. What is even more advantageous, it prevents it from vert
de gris oxidation by preventing the effect of humidity, such that you can touch the pieces of copper
[brass] thus varnished, without fear of making spots, neither of spoiling the polish.
The English have long made a big mystery of the composition of this varnish, and of the way to
use it, however it was communicated to some French Artists in 1720 and 1738. In 1761, it was detailed
in the Memoirs of the Royal Academy of Sciences, where it would still be unknown to Artists, as well
as many other discoveries that would be very useful to [an] infinity of Workers, if not for the research
made by the Reverend Father Dom Bedos. He was not content to copy the recipe of this varnish as
it was given in the memoirs of the academy, but he joined the practice and theory in making himself
some varnish, which does not cede anything to the English in any fashion. It is because of him that I
have the knowledge to make this varnish and to put it to use.
he art that I just described is one of the most respected and sophisticated of the mechanical arts, and is even the most useful and the most widespread, after the arts of
Page 1255
We can presume that the earliest applications of woodworking were the making of doors, win-
dows and some furniture, such as beds, chairs, tables or something like that, whatever the name or
the shape. After these first works of necessity (at least for certain people), we made furniture pieces
for security, such as boxes, armoires, etc. When buildings were constructed using stone and other
materials, we outfitted the interior apartments and even of Temples with woodworking to make the
habitation of them more healthful or more magnificent. To the ornaments that would be proper and
natural for works of this art, such as mouldings and walls, we added those of sculpture and goldsmithing, and embellished them with different metals and precious materials, such as gold, silver, copper,
fine stones, ivory, etc.
It would be impossible and futile to try to fix here the era of beginning and progress of this art, of
which the origin is lost in antiquity, or the various changes to which it was subject, whether regarding
the techniques of working or the names and forms of the tools of this work, or the different names
under which this art has been known to various peoples who have used them. Similar research could
hardly be but conjecture, and would serve more for the erudition of the author than for the perfection of the art and the instruction of those for whom I have particularly written. I will content myself
therefore to say that the art of woodworking is now, especially in France, an art of necessity and of
taste, which has become considerable, that the workers who profit from it are obliged to self-separate
into five branches that are distinguished from each other as much for the manner of working as by the
diversity of these works, as one could note in reading the description of these different arts.
I will not extend myself any more to prove the utility of this beautiful art, necessary to all, rich
as well as poor, and that by reason of their different needs. In effect, we are indebted to it for one
207
Page 1256
208
part of the conveniences of life, both in the city as in the countryside, and even when traveling
and in healthfulness, comfort and safety of our homes. It is with its assistance that we have come
to decorate with so much richness and magnificence the Kings and Princes Palaces, the hotels of
the Seigneurs, and even our temples, which sometimes owe to this art a part of their decoration,
without speaking of works of the same arts that are privately consecrated.
W
The comfort and elegance of our furnishings, and especially our carriages, were carried forward
from the last century, when all the arts seemed to be united to supplement woodworking and enhance its magnificence.
This became more obvious throughout our landscape, where in its decoration the unification was
a great success after the art of Trellising joined with the art of woodworking to create the fifth and last
branch.
Page 1257
Our theaters, finally, draw from woodworking one part of their current splendor, as much for
their construction as for that of the machinery that carries and moves the set decorations.* How
many of the other woodworking arts are not as useful, for the construction of the machinery and
instruments necessary for the preparation or execution of works of these same arts?
Woodworking has the advantage of furnishing the citizen of well-being and with hard-working
habits, an object of occupation altogether most useful, and which exercises equally the body and
the soul. As it is one of the sciences, like that of experimental physics, mechanics, etc., where this
art is normally necessary, it is hardly possible to dispense with the elementary knowledge of this
art for those who occupy themselves with these sciences. This gives many men, illustrious by their
rank and their knowledge, a right and a pleasure to work in a profession little known [to them], and
of which the most able workers could go on a par and even be preferred to most of these artists,
truthfully skilled and ingenious, but of which all the merit consists just in knowing how to make
precious, unnecessary things, which often has no price except that which opulence or caprice give
it, by those who use it.
If woodworking is, by itself, an important art should not the knowledge of it be acquired (or at
least attempted) as much in theory as in practice? Sadly, this is not very common at the present. That
is why it is hoped that woodworkers (and especially the young ones) are persuaded of these truths,
so that they are accustomed to seeing their art as an honest and distinguished profession. They spend
all their efforts to acquire the knowledge necessary on the part of woodworking to which they are totally dedicated, to a specialization so that the analogous arts are relative to theirs. This is what I never
cease to recommend to them in the course in my work that they pursue dedication to study and to
work, which will not really be useful to them other than it would add good will and love for their position, having only this means for acquiring the knowledge and practices of their profession. Books
can add to their talent, but never do so, however well-written and illustrated they can be.
As to this, it is to be hoped that perfection results from the care and zeal with which I created these
the woodworkers, who, for the most part, dont know this
here. Instead I will reserve this discussion for the part which
209
volumes. I am persuaded that if it were perfect, the Public, and especially my young companions,
would have nothing to desire in this regard. I can well assure you that I have not spared pain nor ex-
Page 1258
pense in creating this work,** so that nothing could compete with its perfection either in the order
and arrangement of materials which are presented here, nor regarding the research and experiences
that I have been obliged to recount.
I have also not taught anything by combining example and precept without making remarks
of the advantages and disadvantages, and the different occasions where certain forms of works, or
certain ways of operating, should be preferred over others, and of which I have shown the advantage
on other occasions. This has sometimes placed me in the position of making critiques, perhaps a
bit lively, but when speaking badly of a piece I have always respected the worker, at least that was
my intention.
As to the arrangement and division of this work [set of books], I have paid attention to do it so
that the woodworkers could access the part necessary to them while remaining independent of others, although all the portions fit together and make a complete body of work they should acquire for
the reasons which I gave above. Finally, after seven years of continuous and assiduous work spent in
composing and making the designs of the many plates of my work, I enjoyed the sweet satisfaction of
seeing it completed, I have had enough happiness to have attained the goal that I had set, not in beginning this work, since I did not understand the full scope of this project nor the serious commitment
I made to the public and myself when I finished the first part of my work, and well advanced the second, I perceived the immensity of work that remained to do. I remain committed to neglect nothing
of what could be contributed to give order and clarity to my work, so that if it not be enjoyable, at least
it would be useful.
I have had to face many obstacles because with the exception of the art of building furniture
and the art of drafting, I was not familiar with many other areas of woodworking, and consequently
I had to study particular areas to combine theory and practice as much for knowledge, selection,
the use of materials and tools as for obtaining the perfect knowledge of objects to which one or the
other is being employed.
I have not even been privileged with the help of strangers, my being the first person who has
written on the art of woodworking without the benefit that a careful education could furnish but
that I did not have, and which is found rarely in the class of citizens among which Providence has
placed me.
Born of honest parents, but poor like those of all the workers who do not have other resources
than the work of their hands, they were, consequently in no position to give me a good education or
access to tutoring, and I received no other education than that of the public schools of charity where
I nearly learned to read a dead language that I did not understand (and I still do not understand), and
that which was my own naturally. I did not learn to speak one nor the others of these two languages
since they didnt teach it, at least in the schools of which I am speaking.
I did not enjoy this education for long, totally imperfect that it was, because at the age of 11, (it was
and that which has been the most onerous, is the loss of my
exempt!
Page 1259
210
in 1750) my father, who was and who is still of the Guild of Furniture Makers, made me begin to work
with him in order to transmit his knowledge and his position to me, the only good that he had to give
me, and that he had likewise received from his father, also of the Guild of Furniture Makers.***
From this time up until 1768 [at the age of 29], when I began to make my only object the Descrip-
tion of the Art of Woodworking, I was more occupied with the practical knowledge of my position, than
W
those which, however appropriate to expanding the soul and mind, were strangers to my principal
Page 1260
purpose. I have had the happiness to make progress in my profession during the 18 years that I practiced manual work, and still more to know the late Mr. Blondel, with whom I worked for five years,
during which time he gave me all assistance possible in giving to me freely all the help and counsel
which could be useful to me both in public as in private, to help me acquire the knowledge necessary
on architecture in general and relative to my position.
I responded to his kindness with all the zeal and application of which I was capable, and given the
bit of time that I had, given that the only free time I had was Holidays and Sundays, and a few hours
taken from my rest. I have recognized, by my own experience, that there are no obstacles that good
will and hard work cannot overcome.
I have acquired much knowledge in theory and the practice of my art. How much did I still miss
that I am now able to transmit to others by means of this work, which would have been absolutely impossible given my primitive education if I had not found in reading the good French authors, (such as
Mr. Rollin, Montesquieu, Mr. de Buffon, J. J. Rousseau and others) wisdom, which could not have
been furnished to me even by a more careful education. Fortunately it is wisdom often given and received in an age where a man is still capable of profiting from it.
Instead these readings, although absolutely unconnected to my goal [of learning my craft and
subsequently writing this book] but obtained at a time where my reason began to develop (to which
the study of geometry has served me greatly), put me in the position to render my thoughts without
Page 1261
a florid and elegant style, but at least with precision and a sort of clarity. They have also allowed me
to know and to feel the order that is necessary in a didactic work such as mine, where all the content
should tie together and often in sequence. Repetitions are sometimes inevitable and perhaps even
necessary to better know the truth and importance that one is advancing, and the consequences that
one should or one can derive.
There is also the order and arrangement of materials to which I am the most attached in the course
of my work, without attaching too much worth to the beauty of the style, which should be simple and
reachable to the workers, my confreres, for whom I have particularly written.
It is not, similarly to the Consul Romain (Marius), that I despise the arts I do not know, especially
of beautiful letters and the art of eloquence. On the contrary I know how much the style, selection
and placements of words enhance a book and make its reading more enjoyable. However, as I mentioned this before, these skills are beyond me and I am certain that if I had wanted to write in a more
correct and perfect style, I would have given to exaggerations and unusable information, thus my
simplicity is certainly preferable.
One of the biggest obstacles that I have had to overcome is the cry of the public against big books,
which they will not buy because they are too expensive, or they buy but do not read because they are
*** Claude Roubo, my grandfather, died the month of July
211
too voluminous. But how could I do otherwise? Should I fool the Public in pandering to their taste
but against their interests by giving them an abridged and consequently less-expensive edition, but
where they will learn nothing, or at the most only words or names of the arts? ****
I must however affirm that I have been the first surprised by the extent of my work being known,
which I never believed would be so considerable, when the Royal Academy of Sciences gave me the
W
honor of accepting my work and in permitting the imprint under its name and with its endorsement.
I have been carried away in spite of myself by the abundance of materials which I dealt with, as I said
Page 1261
above, and which I made the remark in the advertisement at the heart of the Second Part.
Whats more, it is not possible, for as little as this work has been read, provided that one be of
good faith and without prejudice it is not possible, I say, to not confirm that the details of the different types of woodworking are immense. However concise it be in detail, it must still be considerable.
It is not the work that is in question here, like history or fantasy, where one is content to expose facts
to the eyes of the reader or to amuse him, but where one leaves him the liberty to make application of
what he has read, in not preventing him from his own judgment, which would become boring to the
reasonable reader.
Here, on the contrary, and in the description of all the arts in general, where it is a question of
teaching, one must not only tell everything, but tell how it is done, and why it is done. Showing the
different ways of operating every day, and in making visible the advantages and the disadvantages,
and the situations where one method is preferable to another, requires describing the minutiae of
works of art, whether whole or in part.
I know that the Arts consist not only of certain practices and fabrications, which, once
wellknown, are applicable in all cases, at least in nearly all. But in other instances, especially those
of which I just made the description, where this fabrication changes with each work of a different
material, and where, with the knowledge necessary and relative to the situation, one must combine
Page 1262
with other processes that are related, and of which one must give at least elementary ideas.
The detail of this elementary knowledge is found, they say, in books, where they are treated with
the most exposure that one could in this art (which is partly true). These same books are very expensive, and one could almost drown in the infinity of knowledge needed by the worker, and with which
it is not always easy to distinguish those things that are really useful. Whats more, the language of
scholarly books is not accessible to everyone. When necessary, why require a worker, who paid the
expense for the description of his art, to buy yet other books to find there what he needs, when one
could get it for him in the description of this same art? [Why should he have to buy more than one
book, when he has already bought the principal one why cannot he find all that he needs in one single volume?]
That is why I have believed myself unable to dispense with giving in my books at least elementary
notions of the sciences or parts of other arts that are relative or accessories to the different types of
woodworking, so as to render my work more useful, and to improve the understanding of the relationship and rapport between all of the arts. In general with the sciences of reason and of style, some
**** What I say here is the incontestable truth: Nothing has
done more wrong to the sciences and the Arts, than the con-
Page 1263
212
parts of mathematics, architecture, the design of different genres and other subjects without it, it is
not possible to make real progress in the art of woodworking and where it is absolutely necessary to
add the light of theory to the habit of the practice.
These are the different reasons that have induced me to omit nothing that could serve the per-
fection and usefulness of my work while responding, as much as it was within me, to the views of
W
the Royal Academy of Sciences that always added to the progress and the perfection of the arts. To
testify to this knowledgeable and illustrious group, I am moved by the honor that they have given me
in associating me with their work, for cooperating with me under their auspices, in the perfection of
the most beautiful monument that one could raise in the human mind.
Appendix
Andr-Jacob Roubo
W
ndr-jacob roubo , of a family originally from Soissons, was born in Paris in 1739. His father, a simple journeyman joiner, was one of those coarse workers who, incapable of managing
their own lives, are even less able to raise their children wisely. That was the case with Roubo, who
was entirely neglected as a child. There can be no doubt that the impulsion, the drive, that marked
his entire life, came from within himself and helped the young apprentice overcome the enormous
obstacles that arose at the start of his career. His fathers example was the first serious danger to him,
but the force of his character luckily preserved him. Put in an apprenticeship by his father, in the
same profession, Andr-Jacob soon understood, despite his youth, that if he didnt want to be buried
in the lowest ranks of workers, learning just the common skills of his trade would not be enough. His
talent for, and love of, woodworking energized him and made him feel that he would need more extensive instruction at a level higher than commonly given to apprentices. So he launched himself into
his studies with all the force of energy and ardent effort that would mark his work for the rest of his
life. The innumerable difficulties he encountered could not stop him. The young apprentice could
expect no help from his father; he was in fact very fortunate that his love of learning did not attract
the kind of punishment that other parents might have inflicted upon their children for being lazy.1
To obtain his first lessons and a few indispensable books, for a long time he needed to set aside
the necessary money from the pittance his father allowed him to feed himself. To Roubo the toughest
hardships seemed light, as long as he had his books, models and the time to read and draw. Even after
he finished his apprenticeship and started working as a journeyman, he was still so poor that to study
during the long winter nights, to stay up a few hours later, and rise a few hours earlier, he was obliged
to take recourse to the cheapest kind of lighting, and was often reduced to scavenging or begging a bit
of grease or tallow to fuel a simple oil lamp. The story of this period in Roubos life, on which we have
few details, should not be thought to have been less honorable or un-useful to write about. So much
courage had its rewards!
His strong character, the ardor and sustained zeal for life that distinguished this young worker
could not long remain unremarked. Jean-Franois Blondel was the nephew of the celebrated architect who created Pariss St. Louis Gate, and himself a member of the French Academy of Fine Arts.
He was a distinguished and enthusiastic artist, but first and foremost an excellent professor, and
as generous as he was zealous. Understanding the nature and social position of the young Roubo,
Blondel admitted him, for free, to his school of architecture and directed his studies for five years.
The memory of Blondel deserves to be here honored as much as that of his student.
Roubo, for the first time in his life, by an unexpected stroke of good fortune, found himself sup-
ported, guided and encouraged. He profited from the lessons of such a distinguished teacher. During
the day, in his work as a joiner, he was proud to be known as a student in a respected school of archi215
216
tecture, and for the talent and finesse he showed in his work. Evenings, and in the spare moments
of free time his trade left him, he devoted himself to the study of mathematics, classical mechanics,
architecture, perspective, the different types of drawing, and soon he had proved himself superior
in the theory as well as the practice of the art of joinery. The network of contacts and acquaintances
he acquired and the ease and style of writing he had developed in the course of his assiduous studies
W
soon inspired him to consider writing about the art that he himself practiced.2
At that time, there were no books on the art of joinery at all worthy of the subject. There were
only two authors, Marin Legerest, in 1619 in his General Methods for Drawing Rampant Curves,
etc., and after him, Edme Blanchard in 1739 published Treatise on Working Wood for Arches,
Rear Aches, etc. who had written seriously on joinery, but these works described only a few isolated topics, and were themselves far from being well and clearly written. In Roubos time the French
Academy of Sciences was in the process of compiling the extensive and useful series of books that
would become the Description of the Arts and Trades, but even this organization, made up of the
most highly educated men in the country, could not and did not not hide the reality that for many of
the arts and trades upon which they needed learned and complete books, there was no one in the
Academy who had any practical knowledge at all. There were already a large number of treatises that
had been written on a variety of professions, but one glaring hole in the collection was the lack of a
book on a trade of such enormous everyday importance as joinery. In the end the work was entrusted
to an engineer, Jeaurat, the author of a treatise on perspective, and who became a respected astronomer, but who had never been a joiner. So it was that Roubo, who had found a second patron in the
person of the Duke of Chaulnes, dared to present the first part of his Treatise on the Art of Joinery.
This section was consecrated to the art of joinery in buildings. As soon as Jeaurat learned who his
competition was, with a generosity that was not without a certain prudence, he abandoned the work
he had in progress and declared that it would be presumptuous to believe himself as knowledgeable
on the subject as a trained professional would be.
[Henri-Louis] Duhamel du Monceau, one of the most learned men of the 18th century, and au-
thor of a large number of treatises in the collection and the one that the members of the Academy
judged to be the best qualified to judge the quality of Roubos work, was charged with examining the
submission. The report that he made on the treatise, which was entered into the Academys official
register on 17 December, 1768, contains the most complete praise of the first effort of the joiner-theoretician.
This first part, Duhamel wrote after a methodical analysis of the work, consists of 50 drawings
which were all drawn by Mr. Roubo. I can assure that we find much order and clarity in this work,
that it is written in a style well-adapted to the subject, and I am persuaded that those that read this
treatise will be surprised to see, in the title, that it was written by a journeyman joiner. How satisfied
the Academy would be, if, in all the trades, we could find workers as capable of recording the knowledge they have acquired during their long training and experiences, etc., etc.
We see by this report that the Academy had decided that Roubos work would become part of the
collection of descriptions of the Arts and Trades. This first part, the Art of Joinery in Buildings, dealt
principally with windows and doors, called mobile joinery,3 and was published in 1769. Roubo
was 30 years old.
To this first honor, the Academy added a second: It asked for Roubo, by the intervention of the
Lieutenant-General of Police, De Sartine, letters conferring on Roubo the title of master joiner,
Life of Roubo
217
something that would not be at all easy to obtain. Moreover, the Council of State, in a very unusual
act, also declared Roubo a master joiner, and gave him the right to use the title professionally as recognition of his talents. So in the title page of the next installation of the The Art of Joinery, published in 1770, Roubo could proudly add the title master joiner to his name. The second book dealt
with sleeping joinery.3
The first section of the third volume dealt with building carts and wagons and wheels; the second
with building everyday furniture; the third with building fine furniture; and the fourth dealt with
trellises and garden structures. These parts were published in 1771, 1772, 1774 and 1775.
This large body of work, taken together, was complete and dealt with all the major branches of
joinery. This is remarkable, given that the subject had essentially never been systematically described
at the time the books were published. Not only had Roubo drawn all the illustrations, of which there
are a large number and are remarkable for the clarity with which the subjects are presented, but he
had, amazingly, also engraved many of the plates used to print the drawings. In this he demonstrated
that rarest of combinations, a worker who could combine those four talents: worker, writer, illustrator and engraver.
Roof Geometry, lArt du Trait4 the discipline that is central to all types of construction, was
described by Roubo at the end of the second part, in a way that was superior to all previous attempts.
The Jesuit priest Derends, Larue and Fraiseir in stone masonry, Mathurin Jousse in roof carpentry,
and the two authors mentioned previously in joinery, preceded Roubo in describing the process, but
his work left all of them far, far behind. Roubo studied and gathered together all the different types of
this work, and was the first to have arrived at a systematic description of the general principles of the
art. This part of his work used geometric principles with great clarity, and is perhaps the most admirable of his contributions in that he wrote the section at a time before the genius Gaspard Monge had
systematized the concept of descriptive geometry!
Roubo, in addition to his other qualities, had the humility to remember the help he had been
given, and had not forgotten to thank his benefactors in print. In this way we learn that the Duke of
Chaulnes had, to allow Roubo to set up his own business, hired him to do all the woodworking that
the Duke needed done on his mansion in Paris. Unfortunately, the premature death of the Duke,
in 1769, deprived him of this important source of support. Ardent and tireless, and despite putting
enormous effort into making a success of his new workshop on the Rue St. Jacques, across the street
from the church of St. Jaques du Haut Pas, he published, in 1777, the first volume of his treatise on
the Construction of Theaters and Theatrical Machinery, with interesting research into Greek and
Roman theaters. He did not publish the second part, but did publish as a last installment of his Art
of Joinery, in 1782, on making boxes and luggage5, and was also invited to become an associate of the
prestigious Fine Arts Society of Geneva.
A second occasion arose to offer Roubo the chance to join the theory and practice of joinery.
Pariss market for grain and flour had long complained that there was not enough space in the circular building of the Grain Market. The vast courtyard in the center of the building, encircled by
arched galleries, had been conceived to be open to the sky, but had over time been partly filled with
roughly built sheds. These were dismantled for the celebrations at the birth of the crown prince in
1781. A huge cloth canopy was stretched over the courtyard, and the clarity of the light revealed the
beauty and magnificence of the newly open courtyard. Two young architects, Legrand and Molinos,
recently returned from Rome, were particularly struck by the possibilities of putting a roof on the
218
court. The directors of the market decided to replace the cloth canopy for the party with a wooden
roof, but it was important that the roof not be too heavy for the ancient foundations and the existing
building. But where might one find a carpenter capable of designing and building such a structure?
At a meeting with the architects to discuss the project, an assistant said there was only one person in
Paris that could get the job done: the renowned joiner Andr Roubo. The architects went to speak to
W
him. He asked for an evening to consider the task. The next day he declared with assurance that he
would take on the job, but only on condition that he would be free to build it as he intended. The
condition was accepted and Roubo got to work. It fell to Roubo to reintroduce the work of Philibert
Delorme, the architect of the French King Henri II, whose innovations had been forgotten in Paris
for 200 years. He adopted the method the great architect Delorme had employed during the construction of Muette Castle, which consisted of substituting for the usual huge beams, ordinary small
planks in spruce with their ends embedded in the tops of the walls to create arched roofs of any size.
The method had been described by Delorme in his book Inventions to Build Well at Low Cost
(Inventions pour Bien Batir Petits Frais).7
At last, after battling all kinds of difficulties, inspecting and adjusting himself each of the innu-
merable pieces of wood; aided by the carpenter Albouy and the blacksmith Raguin who made the
cupola in iron to crown the dome; after five months of work directed with such minute attention to
detail that nobody was killed during the construction, the dome was finished 31 January, 1783. It had
a diameter of 39.5 meters, which was only about 4 meters less than the Pantheon in Rome. As they
removed the scaffolding used during construction to support the immense vault, Roubo, very confident of his calculations of the strength of the dome, chose to remain up in the structure to watch and
feel for himself if it moved as it was gradually left to support its own weight. Nobody wanted to stay
with Roubo, considering it too dangerous. The supports were taken away and as the last one was
taken down, with the shouts of admiration from the bystanders ringing in his ears, the intrepid joiner
saw with his own eyes the perfection of his work. The officials of the market, amazed and enchanted
by the sight of the new dome, ran up into the building and grabbed Roubo, hoisting him in triumph
onto their shoulders. They paraded him in front of the crowd that had gathered to see the man who
had constructed such a novel and useful structure, and then through the streets back to his home.
Roubo, on this occasion as in all others, showed himself to be as unselfish as he was talented,
refused the profits that he might have demanded as contractor for the work, and only wanted a fixed
sum for managing the job. His enthusiasm for his art did not prevent him from understanding that,
later, this kind of structure would soon leave the art of joinery. The metalworker, Raguin, was singing one day the praises of his iron cupola as a masterpiece. Shut up, said Roubo. If I were a metalworker, I would have wanted to build the entire dome in iron. His idea was realized 28 years later.
His magnificent dome in wood was destroyed by fire in 1802 and rebuilt in 1811 by Brunet, but entirely
in iron and copper, and it is this structure that we see today.
The construction of the dome for the Grain Market made Roubo famous throughout Europe.
Proposals and requests for domes and vaulted roofs -la Philibert Delorme poured in from everywhere, even from other countries. Large projects, such as the roof of Pariss Cloth Market, a stairway in solid mahogany for Marbeuf Mansion, etc., should have enriched him. But the Revolution
came, and many who owed him money decided to leave the country. Ruined, Roubo nevertheless
remained a staunch and loyal patriot. A Lieutenant in the National Guard, he wanted, even though
he was sick, to attend with his guard company, the grand ceremony of the Federation on 14 July, 1790,
Life of Roubo
219
at the Champ-de-Mars. The fatigues of the day aggravated the illness that had been troubling him
for many years8. He died on 10 January, 1791; great honors were bestowed by the national guard on
the patriotic joiner. His two sons were given full tuition to the School of Children of the Fatherland.
One of them became a distinguished engineer-geographer, and one of his grandsons was a notary
in Paris. Roubos widow received from the Convention9, a pension of 3,000 pounds, in a decree of
W
18 fructidor, 03 (4 September, 1795), to reward her for the services her husband had rendered to the
Fatherland in perfecting such a useful art!
Louis-Auguste Boileau, Joiner in Paris
From The History and Portraits of Useful Men (1836)
This simple account of the life of a joiner, written by a joiner,
with the idea, first and foremost, of being read to an audience
of young workers or apprentices in the same trade, was welcomed by the founder of the collection Useful Men with such
enthusiasm that I must here thank him, in the name of all the
families of our profession!
and then to derive the dimensions and joint angles of the in-
ly with his father, even after setting up his own shop. Some
ment and was proud that his readings extended into litera-
working.
and the like anchored into the walls and foundations. Much
mation.
like chairmaking is often a specialty within furniture making because of the extreme stresses chairs must withstand.
dor, 03.
Jameel Abraham
Steve Buchanan
Bryan Day
Marc Adams
Jason L. Buchta
Patrick J. Delaney
Kent L. Adkins
Brian P. Buckley
Russell Agle
Amos R. Bullington
Steve E. Dietrich
Graham T. Burbank
Leslie R. Dockeray
Sean Allbee
Michael J. Burda
Dan H. Dolan
Daniel J. Appel
J. Burks
Edward H. Donato
Ronald Aronica
Larry Bush
Joshua P. Doran
Steve Avery
Mike Cairns
Gerhard Dreo
Jamie Bacon
Vincent D. Cameron
Tim Dresser
Jeremy D. Ballenger
Samuel J. Cappo
William Duffield
Robert J. Carriou
Jon M. Dugan
Ron Barter
John Cashman
Christopher Dunn
Mark W. Bauer
David Cassidy
Todd Eckelberg
Rick Bean
J. Tredway Childress
Al Eichorn
Steven W. Bell
Yuri Chumak
William M. Elliott
Matthew Belovarich
Douglas Clark
Peter Ellison
Jerome F. Bias
Suzanne M. Ellison
Genevieve Bieniosek
Robert A. Engl
Thomas Bier
John P. Clifford
Ronnie W. Evans
Wallace Birgenheier
William E. Clift
Mark A. Fairchild
Ty Black
Jason G. Coen
Bob Farris
Michael Blake
Edward T. Coleman
Matthew M. Bobek
Mark Firley
Frank L. Bodner
Joseph J. Corigliano
Megan Fitzpatrick
Louis Bois
Christopher Cosentino
Craig Flaming
Niels Cosman
Peter Follansbee
Fedor Bondarev
Charles H. Courtney
Jack Folse
Douglas J. Boor
Michael P. Fontane
Nicholas Bourke
Jim Crammond
Todd A. Forman
Jared A. Fortney
Fred Brasch
Gary Cruce
John Braunreuther
Randy W. Cummings
Mark C. Frye
Bredesen
Joseph Curran
Rose Ftaya
Adam Brown
Eric S. Brown
Tim Davis
Mo Gambill
Steve Brown
Betsy Davison
Robert Gammage
223
224
Joseph E. Gansert
Graham Hughes
Derek L. MacInnis
James Gee
Ian Mackay
Brian M. Geyer
Wade A. Hutchison
James R. MacNaughton
Ian C. Gifford
Benjamin W. Hutton
Lee Marshall
Sergey Gololobov
Timothy Janssen
Michael S. Mavodones
Jacob Goodson
Donald JeffriesW
Adam Mayatt
Richard J. Gotz
Michael E. McCartan
David H. Jeske
Jedediah M. M. McCartan
Josh Graser
Scott D. Johnson
Ken McColly
Roger O. Green
Susan Johnson
James A. McGee
Paul M. Green
Gary Jones
Joseph E. McMahon
C. Martin Greever
Mark Juliana
Nicholas McNeely
Fred Grosse
John J. Kacerosky
Steven D. Meade
Christopher Grothe
Michel G. Kaplan
Woody Medina
Christian N. Groves
Stefan Karfakis
Gordon Meffert
Gene L. Gustafson
Tom Keane
Kelly Mehler
Randall Gwinn
David E. Keller
Olga Meliakina
Charles A. Haas
Steven Kirincich
Jeff Mestack
Edward T. Hagerty
David W. Metzger
Kim-Erik Hggblom
Robert Miller
Dan Haggerty
Hudson P. Kisser
Gerald K. Millheim
Per Hansen
Travis R. Knapp
Russell D. Moore
Rebecca M. Harkins
Michael L. Morris
Paul Kramer
William M. Morris
Joshua D. Harris
Michael L. Krogh
Philip B. Haynie
Kevin Kuehl
Antonio Murado
Colin Hayward
T.W. Murphy
Paul Hee
Alexandr Kutuzov
Paul Murphy
Stephen Heffernan
George La Chapelle
Thor-Aage K. Heiberg
Gary P. Laroff
Michelle Myers
Murray A. Heidt
Roger E. Myers
David S. Heller
Steven M. Lash
Narayan Nayar
Douglas N. Henderer
Brian LaShomb
Ken Neiswender
Christopher J. Henderson
Travis Lavallee
Raney Nelson
Glenn F. Hendrix
John G. Nelson
William T. Henrick
Justin Lawyer
Chuck Nickerson
Bruce S. R. Lee
Wolfram Herzog
Danny Nieforth
Donna R. Hill
Rod Nieforth
Nathan Hinkle
Jeremy O. Lewis
Bengt Nilsson
Phillip Hirz
Thomas Lie-Nielsen
Chris Norman
Jonathan W. Histed
Mike Livesay
Martin OBrien
Chris Hobbs
Larry OLoane
Elliot Hoefs
Frederick R. Lorthioir
Clifford Ober
John A. Hoffman
Kyle Lovell
Ralf Homberg
Jerome A. Olson
Robert Howard
Patrick R. Lund
John L. Overmire
Eric D. Hresko
Johan Lyrfalk
Kenneth J. Pacheco
List of Subscribers
225
Wendell Pahls
Carolyn J. Sanford
Anthony P. Tiberia
Adam Palmer
Paul G. Sanow
Steven M. Timmis
Wilbur Pan
Jos A. Santiago
Jesse R. Toland
Bernard S. Paniccia
Daniel W. Santos
Jamieson F. Turner
Peter Papasarantopoulos
Ryan D. Sather
Greg Turner
Joshua L. Parker
Tobias R SauerW
Khalid Tyabji
Rick Parker
Steven A. Schafer
James Uhrich
Rex S. Parkison
Jonathan Schroeder
Peri N. Urvek
Troy E. Passe
Rafael V. Vanni
Aaron Pederson
David Schwarzkopf
Alan Vondran
Thomas Pier
Chris Scimone
Matthew L. Vredenburg
Joel Shappell
Joseph L. Walker
Richard Portis
Alf Sharp
George R. Walker
John Pratt
Gregory L. Shomo
William L. Walmsley Jr
Devon W. Prescott
Jared Simms
Beau W. Ward
Thomas A. Priest
Brad Sitzman
Deneb Puchalski
Jared Wayne
James Purl
Tony Slama
Jason A. Weaver
Jon Quinn
Jared Slayton
Christopher Quinn
Jeremiah Smedra
Kenneth W. Wesenberg
David E. Raeside
Darrell Snodgrass
Bill M. Rainford
Linden E. Snyder
D. Eric Westberg
Paula J. Rairdan
Nathaniel L. Wiggers
Matthew A. Rausch
Rick Stillmaker
Randall Wilkins
Dan Reahard
Eric Stirrup
Regis J. Will
Michael Redmond
Thomas J. Stork
J. Norman Reid
Matthew Stott
WM T. Reilly
Timothy J. Talma
Donald C. Williams
Trevor Remster
Wesley B. Tanner
James Wilson
Michael Renfro
Robert Tatman
Glynn R. Wilson
Michael Rogen
Jasper Taylor
Freddy Roman
Mitchell Wilson
Glen A. Rundell
Michael Teniaev
Tadd Wilson
Shawn Russell
Richard Thames
Christopher M. Wilson
T. Alan Russell
David Thew
Bill Rypka
Sean M. Thomas
Robert L. Winter
Jeanpierre Sainfeld
Clayton R. Thomas
Michael Zilis
Nick Sandmann
Scott C. Thornton
Brian K. Zirkle
Matthew Sanfilippo
Nelson V. Thorpe
I N DEX
W
Terms that are found in tables or illustrations have the page number indicated in italics.
Acaja (Anacardium occidentale), 4. See also Cashew
Biblos, 8
Africa, 6, 7, 159
Agalochum Sauvage, 5
Boxwood (Buxus spp.), 11, 16, 17, 19, 118, 120; color, 21
21
Brasil of Fernambouc, 5, 27
Brazil, 4, 5, 6, 9
also Mahogany
Burgaut, 162
Amourette, 5, 6; color, 21
Calambour(e), 5, 6
Cambaye, 5, 6
Asia Minor, 8
Candie, 8, 26
Candlewood, 6, 8
Bald Cypress, 66
Baleen Whale, 162
21
Cedar
Benoit Fin, 5
227
228
Cement, 165
Central America, 14
2930; yellow, 28
Ceresier (cherry), 15
Ceylon, 4, 5, 6, 159
Cherry (Prunus cerasus), 16, 17, 19. See also wild cherry
Eagle Wood, 6
W
Ebony, 3, 6, 9, 19, 118, 163, 193. See also Black Ebony; Cytise;
Citronnier, 9
Citrous Sandalwood, 14
Clove Tree, 8
Cochinchine, 4, 5, 6
Cocoa Wood, 9
Fernambouc, 5, 6, 9, 27
Copper, 27, 38, 42, 76, 94, 100, 130, 156, 160, 16263, 164,
166, 167, 172, 173, 174, 175, 176, 179, 181, 182, 185, 186, 192,
200, 203, 204, 206, 207, 218
Coromandel, 7, 14
Cypress (Cupressus spp.), 6, 9, 35; color, 21. See also Bald Cypress
Cypresswood, 6
Gold, 25, 156, 16364, 167, 172, 179, 192, 193, 194, 206, 207
Cyprus, 8
Cytise, 10
Grooving Plane, 86
Guyana, 5, 6, 8, 9
Dalbergisterie, 1213
Design of Flowers, Fruits, Pastures, and Figures in Wood,
13854
de Veronne, Jean, 25
Holly (Ilex aquifolium), 16, 17, 18, 19, 59, 75; color, 21
Horn, 158, 159, 16061, 162, 180, 183, 184, 190, 192, 195
Index
229
Martinique, 11
Mastic, 165, 16667, 195
Meridional Americas, 5, 8
cabinetmaking, 450
Indigo (Haematoxylum campechianum), 11, 25, 26, 27, 114,
183. See also Campeachy
Middle East, 6
Miter Plane, 79, 80, 82, 83
Moluccan Isles, 6
Inlay, 18594
Mosaic, 123, 124, 128, 130, 131, 138, 139, 146, 151, 154, 155, 165,
185, 187, 188, 190, 192; metal, 193, 194; types of, 194200;
Italy, 9, 19
wood, 193
Mother-of-Pearl, 33, 42, 122, 156, 158, 16162, 164, 183, 184,
190, 192
Jamaica, 4, 6, 7, 11, 14
Jamaican Pear, 11
Jasmine, 9
Joining Plane, 45
Jointing Plane, 50
Juglans spp. See Walnut
Oak, 3, 9, 14, 15, 27, 34, 42, 46, 47; Vosges, 168, 171
Oeben, Jean-Franois, 12
Kaouanne/Caboane, 157
Orangewood, 7, 13, 21
Lamon, 5, 6
Logwood, 11
Lorraine, 15, 19
Madagascar, 6, 9, 10
Madder Root, 17, 25. See also Alizier/Alizie Sorbus Ariar
Mahogany, 4, 6, 35, 71, 117, 218; color, 21. See also Apple
Cashew; Cedar of San Domingo
Pear (Pyrus, spp.), 12, 16, 19, 193; color, 21. See also Jamaican
Pear
Peltogyne spp. See Amaranth(e)
Perspective, Elements of, 12428
Malabar, 4, 5, 6
Pewter, 39, 156, 163, 166, 167, 179, 196, 197, 202, 203, 204
Maple (Acer spp.), 5, 13, 16, 1718, 154; color, 21, 22; hard, 34,
Philippines, 6
35
Maragnan, 6, 9
Plane (tool), 43, 44, 4546, 74, 88, 112, 124, 127, 193; grooving, 86; joining, 45; jointing, 50; kit, 78, 80; low-angle,
230
45, 49; miter, 79, 80, 82, 83; rebating, 52, 71, 73; radial, 10,
22, 66; router, 73, 86, 98; toothing, 50, 56, 79, 86, 110, 113,
Sassafras, 5, 6, 8
Satinwood (Chloroxylon swietenia), 15; color, 21. See also Red
Satinwood; Yellow Satinwood
Pollio, Asinius, 9
Portugal, 6, 9, 10
veneer, 3436
Scraper, 50, 110, 11213, 11415, 116, 118, 120, 137, 180, 181, 190,
192, 201
Serviceberry/Service Berry (Sorbus domestica), 16, 17, 19;
color, 21
Red Cedar, 8
Siam, 5, 7, 14
Silver, 5, 25, 26, 156, 163, 164, 166, 167, 179, 192, 193, 205
Red Santal, 9
ters), 6, 8, 21
Red Satinwood, 7
Solor, 5, 6, 7, 14
Rhodes, 6, 14
South America, 5
Spain, 9, 17
Rose of China, 10
St. Martin, 6
Rottenstone, 112, 114, 117, 118, 121, 193. See also Tripoli
St. Maurice, 6, 9, 10
Sumatra, 5, 6
Syria, 6, 7, 8, 13
Tanasserin, 14
Timor, 5, 6, 7, 14
San Lucia, 15
Tobago, 6, 7, 10, 12
Tools, 3646
San Marthe, 5
Toothing, 50, 56, 79, 86, 110, 113, 118, 18081, 185, 201
San Maurice, 10
Tortoiseshell, 122, 156, 163, 166, 167, 180, 185; nature of,
Santal Citron, 8
Santalum spp. See Sandalwood
15861
Tripoli, 8
Index
231
Walnut of Louisiana, 19
Walnut of Virginia, 19
West Indies, 6, 7, 15
W
White Cedar, 8, 9, 21
White Ebony, 6, 9, 21
Varnish, 166, 193, 194, 204; for gild copper and other
metals, 2046
Veneer, 3, 10, 16, 17, 22, 24, 29, 43, 5153, 6263, 66, 71;
arrangement, 5460; background, 70; banding, 51,
54, 60, 65, 67, 6974, 80, 82, 103, 104, 106, 108, 201;
Windy Isles, 6
ic, 13; how to cut out, 54; how to split, 3032; plan-
making, 450
Walnut (Juglans spp.), 4, 14, 16, 19, 26, 27, 28, 96, 97,
The text type used in this book is Dutch Type Library Fleischmann, a revival
of late 18th-century designs cut by Johann Michael Fleischmann (17071768). Other display fonts and ornaments used were drawn by the books
designer from the original specimens of Fleischmann's contemporaries,
Jacques-Franois & Matthias Rosart, and the brothers Ploos van Amstel.
The photography of the plates from the original 1768 edition of LArt du
menuisier were made at 42-line by E. M. Ginger and Martha Blegen in
Oakland, California. The book was printed by Capital Offset in Concord,
New Hampshire, using Mohawk Superfine paper, and the edition of 600
copies was bound in Cialux cloth by Roswell Bindery in Phoenix, Arizona.
The design and typography are by Wesley B. Tanner at
Passim Editions, Ann Arbor, Michigan.