(50rneU Hmneraitjj SItbrarg
Jtliara, Sftio ^nrh
BOUGHT W[TH THE INCOME OF THE
SAGE ENDOWMENT FUND
THE GIFT OF
HENRY W. SAGE
1891
fine Arts
Cornell University Library
ND 1262.L38
Greek and roman methods of paintingisome
3 1924 016 811 006
Cornell University
Library
The original of tliis book is in
tine Cornell University Library.
There are no known copyright restrictions in
the United States on the use of the text.
http://www.archive.org/details/cu31 92401 681 1 006
GREEK AND EOMAN
METHODS OF PAINTING
CAMBRIDGE UNIVERSITY PRESS
IJonlixm: FETTER LANE, E.G.
C. P. CLAY, Manager
lEllillllutiilj : 100 PRINCES STREET
Berlin : A. ASHER AND CO.
Eeipjifl: P. A. BROCKHAUS
Neta Sottt : G. P. PUTNAM'S SONS
Bomfiaganll (lalcuUa: MACMILLAN AND CO., Ltd.
[All HghtB reserved]
PICTURE IN MELTED WAX BY MR. BURNS
l''rontixp!ii
GREEK AND ROMA:Nr
METHODS OF PAmTING
Some Comments on the Statements made
by Pliny and Vitruvius about Wall and
Panel Painting
BY
A. P. LAUEIE, M.A., D.Sc.
FORMERLY FELLOW OF KING's COLLEGE, OAMBRIDGJE
Cambridge :
at the University Press
1910
vi PREFACE
of the Pompeian frescoes was correct, and conclusive
evidence that the ingenious but much criticised
theories of Herr Berger have no foundation in fact.
I have also, I believe, thrown fresh light on the
method of painting with wax on panel, the nature
of r^dvaxn'i, and the Greek method of painting on
marble surfaces.
In conclusion, I wish to bhank Miss R. F. Forbes
for her valuable assistance in translating difficult
passages, and Professor Baldwin Brown, Professor
Hardie, the Rev. G. C. Richards, Fellow of Oriel,
and Dr. Degering of Berlin (who is now engaged
on a German translation of Vitruvius), for help and
advice in many ways.
I am also much indebted to Mr. A. B. Thomson
and Mr. R. Burns, A.R.S.A,, for their experiments
in wax painting.
A. P. Laurie.
HerIot-Watt College,
Edinburgh, January 1910.
Note. — Since this book was completed Mr. Noel
Heaton has shown (Journ. Royal Soc. Arts, Jan. 7,
1910) that the frescoes at Knossos are examples of
Buon Fresco, that there are no signs of joins, and
that the pigments are bedded in.
EBEATA
Page 9. For "Valentine" read "Valentin" Rose; and for
"editions" read "edition."
Students of the subject are referred to the article by F. Gerlioh,
"Die Teohnik der Eomiseh-Pompejanisohen Wandmalerei" (Neue
Jahrbiicher fur das Klaasische Altertum, 21. 2. 1908), which, on the
whole, agrees with the conclusions come to in the text.
SOME COMMENTS ON THE STATEMENTS
MADE BY PLINY AND YITRUVIUS ABOUT
WALL AND PANEL PAINTING.
If we investigate the technical processes of past ages
we find that changes were very slowly introduced,
and that a method of attaining the desired result,
having been brought to perfection, remained unaltered
from generation to generation. This conservative
habit is peculiarly to be found associated with the
artistic crafts, which even to-day, in spite of the
discoveries of chemists and the introduction of
machinery, have undergone little modification.
The history of painting, with which we are here
especially concerned, illustrates this habit very
clearly. Although the properties of drying oils were
known in the sixth century, and the use of such
oils in painting was described by Theophilus in the
Schedula Biversarwm Artiv/m, in the twelfth century,
yet it is only in the fifteenth century that oil begins
to replace egg, as the almost universal medium.^
^ This statement probably requires modification as far as northern
countries are concerned, judging by the records of Ely Cathedral
and St. Stephen's Chapel.
A
2 GREEK AND ROMAN
The reasons for this conservative habit are easily
understood. In the first place, the mastery of an
artistic craft is only slowly and laboriously attained,
and the pupil, having by years of practice gained
facility of expression by means of the technique taught
him by his master, finds the new technique inexpres-
sive and nonplastic. In the second place, each tech-
nique leads to a particular form of artistic expression,
and the artist having at last attained perfection in
this expression desires no further change. Centuries
may therefore pass before a man arises of sufficient
originality to aim at new modes of expression, and
to try to obtain these by mastering a new
technique.
Even when the artist has at last accepted the new
medium as preferable to the old, he uses it at first
to imitate the old effects, and generations may pass
before the full possibilities of the new medium are
reahsed. The truth of this statement is so clearly
revealed by a study of the pictures painted in oil
from the fifteenth century onwards that it requires
no further demonstration.
We are therefore quite justified, in an inquiry of
this nature,in supplementing the information supplied
by Pliny and Vitruvius by information gathered from
older writers and older works of art.
Nor need we confine the source of such additional
information to the neighbourhood of Rome, as we
METHODS OF PAINTING 3
may safely assume that such technical knowledge
as was possessed by the civilisations which had grown
up round the Mediterranean basin would be con-
centrated in Rome at the time of the empire.
Our inquiry may also be legitimately assisted by
the examination of worksrof art proidjiced at a time
somewhat later than that'-of^PlmyiJ'n so far as we
find on examination that they assist us in elucidating
the descriptions of technical processes given by him.
For we cannot expect to find again in the history
of art a series of wall paintings buried in the volcanic
ashes of the eruption which killed the author who
is one of our principal sources of information, and
thus signed, sealed, and dated, forming the appro-
priate illustration to the author's own works.
- We may also find in later writers information
which will assist us in th« interpretation of passages
contained in the author under consideration, and, in
the modified practices of the art in later times,
indications of the methods of an earlier technique.
But the consideration of the extent to which we
are justified in using information from later sources
to enable us to understand completely the technique
used at the period we are inquiring into, leads us to
lay down the rule by which such an investigation
must be governed.
The rule is this. The information supplied by the
writers of the period under consideration . must be
4 GREEK AND ROMAN
regarded as in essence complete, until it is clearly
demonstrated by exhaustive experiments that some
essential information has not been supplied by them.
The necessity for thus limiting and defining our
inquiry has been so little recognised by many writers
that it requires some further explanation.
Let us take a particular example.
In order to understand the technical processes
used by painters in the time of Pliny, it is of the
utmost importance to know whether they were fami-
liar with the properties of the 'drying' oils which
are used by painters to-day.
On examining the writings of Pliny we find that
he describes the preparation of nut oil, which is
one of the 'drying' oils, but makes no mention of
its property of oxidising into a transparent elastic
solid, or of its use in painting.
In the earlier writings of others there is also no men-
tion of the properties of drying oils, and Dioscorides,
who is almost a contemporary, while adding the
description of the preparation of poppy oil, also
fails to note their ' drying ' properties.
The first mention of the ' drying ' properties of
nut oil, and its use as a varnish for pictures, occurs
in Aetius, a writer of the early sixth century.
If, therefore, we are to apply the rule which has
already been laid down, we are bound to commence
our inquiry with the assumption that the properties
METHODS OF PAINTING 5
of drying oils, and their consequent use in painting,
were not known in the first century of the Christian
era, but were discovered subsequently.
Against this conclusion many arguments can be
used. In the first place, comparatively few of the
books written in classical times have survived, and
many technical treatises must have perished.
In the second place, portions of the authors which
have survived have disappeared, and consequently
a missing chapter may contain the information which
is wanting.
In the third place, the authors themselves are
often compilers with no practical acquaintance with
the technical processes they describe, and probably
preferred to copy from older compilers, rather than
take the trouble of collecting information from the
craftsmen themselves.
In the fourth place, as the craftsmen were taught by
tradition, and jealously guarded the secrets of their
craft, they may well have made use of processes
which were only recorded centuries after their dis-
covery, or have disappeared without ever being
recorded at all.
In the fifth place, the author considered may often
omit facts with which he must have been perfectly
familiar, just as Pliny fails to tell us what medium
was used by one school of Greek painters, although
it is quite evident that some medium must have
6 GREEK AND ROMAN
been used, and Pliny must have been familiar with
its nature.
In reply to these arguments, only one thing
requires to be said. While we are justified in using
all the information supplied by writers existing
before the date which we have fixed as the limit
of our inquiry (in this case the end of the first
century of the Christian era), we are not justified
! in using new information contained in writers sub-
sequent to this date, because we cannot prove that
it is not the outcome of discoveries made after the
period we are considering.
We are no more justified in assuming a knowledge
of the properties of (irying oils in the first century
because they were described in the sixth century,
than we should be justified in assuming a knowledge
of aniline dyes in the first century because they
are described in works on chemistry Avritten in the
nineteenth century.
Therefore the only possible working hypothesis is,
that the descriptions given by our authors of techni-
cal processes are in substance correct, and that what
is not stated by them was not known at the time
when they wrote. A less rigid method of inquiry
leads us to endless loose speculations, and makes all
real progress impossible.
If this is to be our working hypothesis, let us next
consider what conditions are necessary to justify
METHODS OF PAINTING 7
us in abandoning it. These are, in the first place, the
discoYery of an earlier or contemporary writer con-
taining additional information. In the second place,
the demonstration by means of the chemical analysis
of a work of art of undoubted authenticity that our
information is incomplete or inaccurate. In the
third place, the demonstration, after exhaustive
experiments, that a technical process described by
the author is unworkable, or that a given contem-
porary work of art can only be reproduced by using
methods not mentioned by contemporary writers.
It is in this third case, however, that the utmost
caution is necessary.
The original craftsmen had the advantage of the
accumulated knowledge of centuries of tradition
and of years of training in the use of the lost
technical process. The modern experimenter has
nothing to guide him but a few brief words of
description.
His experiments must • therefore be ^of a very
thorough and exhaustive character before he ven-
tures to decide that the process is impossible, or that
the work of art must have been produced by another
method.
In conclusion, where there are obvious omissions
or obscurities in the information at our disposal, it
is better to state frankly that we do not know, than
to take advantage of the opportunity for ingenious
8 GREEK AND ROMAN
speculations. We must also resist the temptation of
inventing a technical process which the contemporary
writer does not describe, and then getting over
the difficulty of his omission by assuming that one
or two chapters of his book have been lost, which,
no doubt, contained the proof of the correctness of
our ingenious guesswork.
I may have seemed to have laboured too long
in laying down the conditions essential to a scientific
inquiry ; but those who are familiar with the many
writings on this subject, and the way in which
ingenious hypothesis has been piled on ingenious
hypothesis, will realise the necessity of bringing this
department of research into line with the many
other departments which have come in recent years
under the control of those methods of exact in-
vestigation which we owe to the students of natural
science.
Having now discussed at some length the prin-
ciples on which such an inquiry as this should be
conducted, I propose to describe briefly the pigments
used in classical times.
Those who wish an historical survey of the various
writers who have dealt with this subject can consult
Die Maltechnik des Alterthums, by Ernst Berger,
Mtlnchen, 1904.
As much that I have to say will involve a criti-
cism of some of Herr Berger's conclusions, it is
METHODS OF PAINTING 9
only fitting to state in this place, that all students
of this subjeot ^^^ immensely indebted to him for
his learned publications.
Before going further, however, it is necessary to
mention the actual texts of Vitruvius and Pliny
consulted and quoted from in the course of this
paper. These are : the edition of Pliny's Natural
History, edited by Carl Mayhoff, and issued by
Teubner in 1906 ; and the most recent edition of
Vitruvius, De Architectura, by Valentine Rose, issued
by Teubner in 1899, and containing some corrections
on the earlier editions by Rose and Mliller-Strtlbing.
To return to the pigments used in classical times,
they can be briefly dealt with, because, though there
are obscure descriptions in Pliny, and many pigments,
especially those prepared from organic colouring
matters, are difficult to identify, there is a sufficient
amount of exact information to enable us, partly by the
analysis of actual samples, and partly by the descrip-
tions given by Pliny, Vitruvius, and others, to decide
without doubt on the nature of a sufficient number
of pigments to supply the painters of classical times
with an extensive palette. As, moreover, there is no
dispute as to the identity of those I shall mention,
and I have little new to add to what is common
knowledge, it would be tedious to give all the proofs
and references.
It is, in the first place, evident, both from the
10 GREEK AND ROMAN
descriptions in Pliny, Vitruvius, and others, and
from the analysis of pigments, that the painters in
classical times were familiar with what are now
roughly called the earth colours, such as the yellow
and red ochres, terre verte, and probably the siennas
and umber. The references to red colours are often
somewhat obscure, but Pliny describes ochres from
various sources,^ and Vitruvius mentions both yellow
and red ochres,^ and the roasting of yellow ochres to
form purples.^ More than one white earth is men-
tioned by Pliny* as having been used in painting,
and chalk, gypsum, and plaster of Paris were known.
Other pigments described by Pliny have been identi-
fied with the blue and green copper ores.^
Blue carbonate of copper has been used as a
pigment up to comparatively recent times^ and was
apparently known to early Italian painters as azzurro
della magna. I have prepared a blue pigment from
azurite. It is a beautiful blue and very difficult
to distinguish in appearance from real ultramarine,
as is correctly stated by Mrs. Merrifield.^ It is of a
sandy nature, and therefore difficult to handle in oil,
in which it settles, letting the oil float to the top.
Hence the directions, after the introduction of oil
I Pliny, Bk. xxxiii. c. 56. = Vit., Bk. vii. o. 7.
» Vit., Bk. vii. u. 11.
* Pliny, Bk. xxxv. oo. 18, 19, 30 ; xxxvi. 59.
" xxxiii. 26, 57. " Ancient Practiee of Painting , cxcyii.
METHODS OF PAINTING 11
painting, to mix it with size or to dust it on dry, and
the belief that it turned green in oil.^
The native sulphide of mercury, cinnabar, was
also used as a pigment.^ I have prepared a red from
native cinnabar, which is very little inferior in
brilliancy to vermilion, that is, the artificially pre-
pared sulphide.
The native sulphide of arsenic, orpi ment. was also
used,* and has only recently disappeared from the
artist's palette. It is a yellow of peculiar beauty,
and unlike any other yellow in appearance. It is
fugitive in oil, but permanent in a pure spirit varnish
or pine balsam.
Coming to artificial pigments, they understood the
preparation of white lead (cerussa),* somewhat in-
accurately described by Pliny, but more clearly by
\Vitruvius. They were also familiar with several
vpreparations of the oxides of lead, such as litharge,
red lead, and other similar preparations.^
Sir Humphry Davy found an orange yellow oxide
of lead pigment on ' a piece of stucco in the ruins
near the monument of Gains Gestius.' "
The preparation of verdigris is also described by
1 Ancient Practice of Painting, cciv.
2 Pliny, Bk. xxxiii. c. 36; Vit., Bk. vii. c. 8.
' Pliny, Bk. xxxiii. c. 22 ; xxxiv. c. 56; Vit., Bk. vii. o. 7.
" Pliny, Bk. xxxiv. o. 54 ; Vit., vii. 12.
° Pliny, xxxiii. 35, 40; xxxiv. 51, 54; xxxv. 22; Vit., vii. 12.
8 Phil. Trans., vol. 105, p. 105.
12 GREEK AND ROMAN
Pliny and Vitruvius,^ and the preparation of lamp
black, charcoal black, and probably bone black.^
I One of the most interesting of the ancient pigments
was the copper frit which is frequently present both
in Egyptian and Roman remains, and which occurs in
Egypt as early as the twelfth dynasty.
There is no clear description of this pigment in
Pliny, though it is probably referred to in chapter 57
of the thiT.ty-third book, and again in chapter 38
of the thirty-seventh book, but Vitruvius gives a
perfectly clear account of its preparation.
Vitruvius tells us that it was formerly manu-
factured in Alexandria and afterwards manufactured
by Vestorius at Puteoli,^ and that it is prepared by
heating together copper, sand, and soda.
The analysis of the blue, however, reveals the
presence, in addition, of lime and alumina. These
may have been deliberately introduced, or have been
present in the impure sand and soda used.
Professor Russell has made a special investigation
of this blue, and has also prepared a green. A com-
plete account of his experiments will be found in the
Proceedings of the Royal Institution, London, 1893.
The blue is easily identified by its appearance under
the microscope, consisting of transparent glassy frag-
ments, and by its giving the usual reactions for copper.
' Pliny, xxxiv. 26 ; Vit., vii. 12. ' Pliny, xxxv. 25 ; Vit., vii. 10.
3 Vit., vii. 11.
METHODS OF PAINTING 13
In addition to these pigments they were accustomed
to prepare pigments from vegetable and animal dyes.
Both Pliny and Vitruvius mention the use of the
purple dye of the murex for this purpose.
Pliny states that the pigment is prepared by
adding to the dye a white earth which is used for
cleaning silver, and which absorbs the dye more
readily than wool.^ This earth is probably infusorial
earth, which has the property of absorbing and fixing
basic aniline dyes without the use of any mordant or
precipitating agent. Vitruvius mentions mixing the
dye with honey.^
In addition, Vitruvius mentions the use of other
vegetable colouring matters, including madder for
dyeing chalk,^ and Pliny describes a process for
dyeing 'chrysocoUa' (probably green carbonate of
copper) with ' lutum,' in the same manner as flax or
wool is dyed by the use of ' alumen.' *
An account of the nature and properties of 'alumen'
will be found in chapter 52 of the thirty-fifth book.
The question as to what this substance was and
how it was prepared or obtained, and how far it is
identical with the double sulphate of alumina and
potash known to-day as alum, is worthy of further
investigation. It is evident from the account given
in the thirty-fifth book that a white variety was
' Pliny, xxxY. 26. ^ Vit., vii. 13.
s Vit., vii. 14. ^ Pliny, xxxiii. 26.
14 GREEK AND ROMAN
known, and this white variety was used in the process
of dyeing wool scarlet, and was also used for preparing
vegetable lakes. This certainly seems to point to the
substance being either sulphate of alumina, or alum,
however obtained and prepared. We should, however,
be departing too far from our present inquiry if we
followed this question any further.
Such dyes as weld, woad, kermes, and madder were
well known, and doubtless used for the preparation of
pigments.
Sir Humphry Davy ^ examined a pot of pigment
found in the so-called bathsof Titus,of a pale red colour,
which proved to be a vegetable or animal dye on a
chalk base, and containing some silica and alumina.
But one of the most interesting of recent discoveries
has been the identification of a rose pigment obtained
by Mr. Flinders Petrie in Egypt, as a madder lake.
Professor Russell,^ who examined this pigment,
found the base to be gypsum, and identified the colour-
ing matter as alizarin by its absorption spectrum.
He then succeeded in successfully imitating the
pigment by heating madder root with lime water and
gypsum, an entirely novel method of preparing a lake.
Indigo, both from India and from woad, was also
known, and probably dragon's blood.^
The only other question of interest is whether the
1 Phil. Trans., vol. 105.
2 Rnsaell on Ancient Egyptian Pigments, Royal Institution, 1893.
* Pliny, xxxiii. 57 ; xxxiii. 38 ; viii. 12 ; xxxv. 27 ; Vit., vii. 14.
METHODS OF PAINTING 15
preparation of ultramarine from lapis lazuli was
known in classical times.
No pigment was more prized by the fifteenth and
sixteenth century painters of Italy, and its prepara-
tion is described in great detail in more than one
fifteenth century MS.^
Among his descriptions of precious stones Pliny
describes a stone called ' cyanos.' ^ From his descrip-
tion this stone is evidently lapis lazuli, and he goes
on to speak of an artificial variety discovered by an
Egyptian king. This account is probably taken from
Theophrastus, because he also speaks of the discovery
made by an Egyptian king of an artificial cyanos.
But the place where this statement is made in
Theophrastus is chapter xcviii., in that part of the
work where he describes earths, and more especially
those suitable for pigments.
An earlier mention of ' cyanos ' occurs in chapter
Ivi., where Theophrastus is describing the various
precious stones and gems. He again refers to it in
chapter Ixv., contrasting it with the ' sapphire,' ^ and
again in chapter Ixx., where, if this chapter be taken
as a continuation of the preceding chapter, he is
apparently referring to the golden particles of pyrites
the stone often contains. On the whole, therefore, the
' MS. Jehan Le Bfegue, Ancient PrtKtice of Painting, vol. i. p. 96.
MS. Cennino Cennini, page 47 of translation by Mrs. Herringham.
2 Pliny, xxxvii. 38, probably also xxxvii. 39, called 'sapphiros.'
3 A variety of lapis lazuli, not the modern sapphire.
16 GREEK AND ROMAN
evidence is in favour of Theophrastus having de-
scribed two, if not more, different blue stones under
the one name, the first being lapis lazuli, and the
last being of the nature of a blue earth used in
painting, and the artificial variety being probably the
Egyptian copper frit.
We may therefore safely say that there is no
evidence in favour of the view that the prepara-
tion of ultramarine from lapis lazuli was known in
classical times. The process, which is an elaborate
and difficult one, either came from elsewhere, or was
discovered at some time during the middle ages.
Some writers have gone so far as to assume a
knowledge of the preparation of arifi/?cmi^ ultramarine
from the statements made by Pliny.
It is evident from the foregoing account that the
painters of classical times had a very complete palette
at their command, a palette which remained the same,
with a few modifications, right through the history of
painting and up to the dawn of modern chemistry.
The principal modifications were, the introduction
of ultramarine, and the disappearance of the copper
frit, and of pigments prepared from murex, and the
introduction of artificial in place of native vermilion,
of lakes prepared from dye woods, and after the
discovery of Mexico of the cochineal lakes in place ol
kermes.
METHODS OF PAINTING 17
With the exception of the introduction of ultra-
marine none of these changes can be regarded as
improvements on the palette used in classical times.
While I have refrained from going into detail,
much more might be said on this subject, a com-
parison of the statements made by Pliny and
Vitruvius respectively about the same pigments being
alone of great interest. Pliny evidently copied some
of his facts from Vitruvius, and in many cases copied
inaccurately. The statements made by Vitruvius
about the nature and preparation of pigments are
evidently much more reliable, and we may suppose
are due to actual practical knowledge. The state-
ments in Pliny are in many cases quite unintelligible,
and his evident ignorance of the method of prepara-
tion of so well known a pigment as the blue copper
frit is very significant.
The suggestion that the white earth he describes
as being used to prepare a pigment from the murex
dye is infusorial earth is, I believe, new, and the
experiments by Professor Russell on the Egyptian
blue and Egyptian madder lake are not as well
known as they ought to be. His experiments on the
madder lake are especially interesting, as nearly all
traces of lakes from madder disappear in the mss.
of the middle ages, right through to the seventeenth
century, when a receipt is given in the Ars Vitraria.
We shall now proceed to consider what mediums
£
18 GREEK AND ROMAN
were used for painting on panel and canvas in and
before the time of Pliny.
A medium suitable for the artist must fulfil several
functions. It must, when mixed with the pigment,
form a convenient means of transferring the pigment
to the brush and from the brush to the picture,
and it must serve to attach the pigment to the
surface of the picture, and in addition some mediums
serve to protect the pigment more or less perfectly
from chemical and mechanical injury.
Among primitive mediums which serve to attach
the pigment may be mentioned gum, glue, and white
or yolk of egg. Such mediums can all be dissolved
in or mixed with water, and while serving to attach
the pigment do not really protect it from attack by
air and moisture or prevent its easy removal.
Of the three, probably egg protects and preserves
the pigment best.
But the discovery that certain vegetable oils have
the property of ' drying,' that is of being converted by
the oxygen of the air into elastic transparent solids,
has supplied the artist with a medium which protects
the pigment, and gives the picture a surface which
is practically waterproof Experiments ^ which I have
made have proved that this protection is not so
complete as has been supposed, and that dried films
of linseed oil are pervious to water vapour, but there
' Appendix I.
METHODS OF PAINTING 19
is obviously a great difference between such an oil
surface and a surface consisting of gum, glue, or egg.
The best known of these vegetable ' drying ' oils are
linseed oil, poppy oil, and walnut oil. Castor oil
gets sticky when exposed to the air, but does not
' dry ' with sufficient completeness to be of any
practical use to the painter. There are other drying
oils, but we need not consider them here.
If we wish to protect a surface from moisture we
can use other means besides the use of a ' drying oil.'
Many resins exude from the trees in a semi-liquid
form, and are known generally as balsams. This semi-
liquid condition is due to the presence of volatile oils
which subsequently evaporate and leave the hard
resin behind. The semi-liquid turpentines from the
various species of pine are excellent examples, such
as Canada balsam from the abiea halsamsa, Venice
turpentine from the larch, and others. On distilla-
tion the volatile spirits of turpentine, or ' turpentine '
as it is usually called, is obtained, and the resin is
left behind.
Such semi-liquid turpentines, on slightly warming,
can be used as varnishes, but do not lend themselves
readily to use as painting mediums without further
admixtures on account of their sticky nature.
At the present day varnishes are prepared in one
of two ways.
The dry hard resin is dissolved in a volatile
20 GREEK AND ROMAN
medium, such as spirits of turpentine, alcohol, or
rectified petroleum, and brushed over the surface.
The volatile medium then evaporates and leaves a
layer of the resin behind. Such varnishes correspond
to the natural varnishes or pine balsams, which may
be regarded as natural spirit varnishes, in which the
resin is dissolved in spirits of turpentine. (Distinc-
tions are drawn by chemists between balsams, tur-
pentines, and solutions of resin in spirits of tur-
pentine, but the above statements are sufficiently
accurate for our present purposes. ^
The other modern method of making varnishes is
to dissolve the resin in a drying oil, which solu-
tion, when exposed in a thin film, ' dries ' to form a
hard elastic coating, owing to the oxidation of the
oil, and contains the oil and resin intimately associated
together. Such varnishes are usually diluted with
turpentine.
It is evident then that if we wish to paint a picture
which will resist the attacks of moisture, we have
several different methods available.
We may use a drying oil as our medium, or we
may use a natural balsam, or a spirit varnish, or an
oil varnish.
In practice the last three mediums, if used pure,
will be found to be too sticky to lend themselves to
a successful use.
' Dieterioh on Analysis of Resins, Balsams, and Gum Resins.
METHODS OF PAINTING 21
Or we may paint our picture with gum, egg, or
glue, and then protect the surface afterwards by
varnishing with any of the above substances. The
sticky nature of the varnishes will not prevent their
use for this purpose, though making them unsuitable
as painting mediums (see Appendix I.).
There is still, however, another substance which
has not been mentioned, and which will serve to
protect the pigment, and that is beeswax.
Beeswax is easily melted and forms semi-fluid
semi-solutions, with turpentine and oil, and dissolves
in fused resins. It is not transparent, but in thin
coats is so nearly transparent as to make it quite
suitable for solid painting.
While a fused resin is quite unsuitable for a
varnish, as it is impossible to lay it on in smooth
coats, and it cracks all over on cooling, the dissolving
of a little beeswax in it only diminishes its trans-
parency a little, and makes it quite suitable for use
as a varnish. The melting point is considerably
lowered, and the varnished surface does not crack.
After these preliminary explanations we are in a
position _to consider intelligently the information
available as to the mediums and methods used by
painters in the time of Pliny.
In the first place we shall begin by stating what
knowledge they had of the various substances men-
tioned in the above statement.
22 GREEK AND ROMAN
In Pliny's account of remedies derived from eggs,
he states that a cement for glass can be made of white
of egg and lime.^ He says that white of egg (ovi
candidum) is suitable for laying gold leaf on marble.^
He also mentions egg as suitable for a painting
medium in a particulai; case in the following words : *
— ' Pingentes sandyce sublita, mox ex ovo inducentes
purptirissum fulgorem minii faciunt. Si purpurae
facere malunt, caeruleum sublinunt, mox purpurissum
ex ovo inducunt.' 'Painters put on sandyx as a'
ground colour; thereafter, laying on purpurissum
with egg they produce the brilliance of vermilion.
If they prefer to produce the brilliance of purple they
put on caeruleum as the ground colour, and then lay
on purpurissum with egg.' If we now proceed to
consider this statement of Pliny's, it is evident, in
the first place, that he does not state whether he
means the whole egg or the white or the yolk to be
used. Formerly, when describing how to lay on gold
leaf, he states definitely that ' candidum ovi ' is to be
used. It seems therefore probable that the medium
in this case is intended to be the whole of the egg.
In the second place, he only directs the purpurissum
to be mixed with the egg, and fails to mention what
medium the other pigments are mixed with.
If the medium for these other pigments was not
also egg, it certainly could not be wax, as the paint-
' xxix. 11, ^ xxxiii. 20. » xxxv. 26.
METHODS OF PAINTING 23
*■
ing over it with the egg medium could not be
successfully carried out. Probably, therefore, it was
a water medium, that is egg, gum, or glue.
There is also another possibility, and that is that
the passage refers to fresco painting. Purpurissum,
he tells us later, was one of the pigments that could
not stand the action of wet lime, and therefore it may
have been customary to paint it on the dry surface
with egg. Similar directions for touching up true
fresco with pigment mixed with egg are to be found
in Cennino Cennini.^
It is also not clear from the statement whether it
is the process of glazing with purpurissum to which
he wishes to direct attention, or the use of egg.
There are other passages with which we shall have
to deal later, in which a process is stated either to
be peculiar to a particular painter, or only to be
used in a special case. In such cases we are justi-
fied in stating that such processes were not universally
used.
In this passage it is impossible to say from the
context, whether or not we are to understand that
the use of egg is exceptional, and whether only for
this special purpose.
The passage at any rate proves conclusively that
the use of egg as a medium was a familiar idea to
Pliny for this special purpose, and there is certainly
1 Cennino Cennini, translation by Mrs. Herringham.
24 GREEK AND ROMAN
no evidence to be drawn from the text that he in-
tends egg to be used only in this case, and there-
fore there is no evidence that egg was not the
ordinary accepted medium for other purposes.
Pliny mentions the preparation of glue from hides
of oxen,^ and the preparation of fish glue? and the
use of glue by carpenters.^
He also mentions glue in connection with pig-
ments in his description of the preparation of
black pigments.* The actual words used are as
follows : —
' Omne autem atramentum sole perficitur, librarium
cumme, tectorium glutino admixto.' ' The prepara-
tion of every sort of atramentum is completed by
exposure to the sun, that used for writing having an
admixture of gum, that for use on plaster-work an
admixture of glue.'
With this statement the following statement made
by Vitruvius can be compared : —
'Inde coUecta partim componitur ex cummi sub-
acta ad usum atramenti librarii, reliqua tectores
glutinum adraiscentes in parietibus utuntur. Sin
autem hae copiae non fuerint paratae, ita necessi-
tatibus erit administrandum, ne expectatione morae
res retineatur. Sarmenta aut taedae schidiae com-
burantur, cum erunt carbones extinguantur, deinde in
' xi. 94. 2 xxxii. 24. " xvi. 82, 83.
' XXXV. 25.
METHODS OF PAINTING 25
inortario cum glutino terantur. Ita erit atramentum
tectoribus non invenustum.' ^ 'When it has been
collected thence ' (i.e. when the soOt arising from the
combustion of resin has been collected from the walls
of a specially constructed furnace), 'part of it is
blended with beaten up gum for use as writing ink,
the rest is used on walls by plasterers, who mix it
with glue. Should such material not be at hand, the
necessities of the case must be so dealt with as to
prevent undue delay ' in carrying out the work.
Twigs or pine- shavings are to be burnt, when con-
verted into charcoal they are to be extinguished, and
then powdered in a mortar with glue. This will
provide plasterers with a quite pleasing black.'
It has been assumed that these passages prove the
use of glue for painting on walls. They are, how-
ever, capable of another explanation. Lamp black
is of such a light powdery nature that it is very
unmanageable as a pigment, being both bulky to
handle and difficult to moisten with water. It has
long, therefore, been the custom of the trade to
grind it with water and a little glue, press it into
convenient shapes, and sell it in such a form as
' drop black.' The pigment in this form can readily
be mixed with oil or any other medium, the amount
of glue being just sufficient to bind it together.
Charcoal is also very unmanageable, and would
' Vitruvius, vii. 10.
26 GREEK AND ROMAN
require fine grinding and similar treatnaent. It is
therefore quite possible that this method of prepar-
ing black for use was equally familiar in the time of
Vitruvius, and that the passages quoted have no refer-
ence to the actual medium used for painting.
Glue was used for preparing gesso grounds on
coffins in Egypt, and possibly also as the painter's
medium.^
Pliny mentions various kinds of gum, including
gum Arabic, which he states 'to be the best, and he
also describes the use of gum for medicinal purposes.^
The references to the use of gum for making ink
have already been quoted. There is a passage in his
description of gums in the thirteenth book to which
more attention might well have been given. In
speaking of a gum which he calls sarcocoUa, he
says : — ' Fit et e sarcocolla — ita vocatur arbor et
cummis — utilissima pictoribus et medicis, similis
pollini turis, et ideo Candida quam rufa melior.' (A
gum) 'is also obtained from sarcocolla — both tree
and gum are so named — which is very useful to
painters and physicians. It resembles the dust of
frankincense, and for that reason the white is better
than the red.'
This is supposed to be the Penaea Sarcocolla of
Linnaeus, the gum of which is brought from Abys-
' See Appendix II.
•= xiii. 19 ; xiii. 20 ; xxiv. 64, 65, 67.
METHODS OF PAINTING 27
sinia. Pliny mentions its medicinal uses later on,^
and it is also described by Dioscorides.^ It would
be of great interest to follow up this reference, and
if the gum can really be identified, carefully examine
its properties.
In conclusion, the artistic possibilities of these
three media, glue, gum, and egg, require no demon-
stration. Glue is the medium of the scene-painter,
gum is the medium of the painter in water-colour, and
egg was the medium of the early Italian painters.
Nor can there be any doubt that all these sub-
stances, being familiarly known, would be tried as
media -by artists, though they might ultimately
select one in preference to another.^
We shall now proceed, to examine carefully the
statements made by Pliny about oils, resins, and
beeswax.
A very large number of resinous bodies are men-
tioned by Pliny, including those used for perfumes,
such as balm of Gilead and frankincense, and the
various sources from which they are obtained are
described.
It is unnecessary to include all these in our present
1 xxiv. 78. 2 B. iii. o. 99.
^ A medium of egg and gum mixed, with the addition of bile to
make the colour flow easily, is mentioned in a Papyrus found at
Thebes of the third or fourth century. Papyri Oraeci Musei
Antiquarii publici Lugduni Batavi, edited by C. Leemans : torn.
ii. papyrus x.
28 GREEK AND ROMAN
inquiry, but it is sufficient to give the names and
sources of some of them. In the 22nd chapter of
the twenty-fourth book, he mentions resins extracted
from the pine, the pitch-tree, and the larch, the
terebinth, the lentisk, and the cypress. He distin-
guishes in this chapter between the dry and the'
liquid resins, speaking, for instance, of the resin of
the pitch-tree as unctuous and of the larch as thin
like honey. It is a matter of some difficulty to
identify the various kinds of pine referred to by
Pliny. The terebinth, which he says comes from
Syria, is supposed to be the pistacia terebinthus, while
the lentisk yields the resin known as mastic.^
An exhaustive inquiry into all the resins men-
tioned by Pliny, and an attempt to conclusively
identify the various trees described by him would
involve a long and laborious research, and would be
of no value for our present purpose. It is sufficient
that such balsams or natural varnishes -were well
known, and were obtained from many sources.
A clear distinction is drawn by Pliny between resin
and pitch.2 The ' resin ' is the natural product of
the tree, 'pix' is the result of a special process,
namely, heating the broken up timber of the tree,
when a tar flows out and is collected. Such a tar
1 xxiv. 22.
^ Dioscorides mentions the trees pitch is obtained from, but not
the method of preparation. B. i. 94.
METHODS OF PAINTING 29
or pitch is still known and used to-day, and is of
a rich brown colour. It differs in its behaviour and
properties from the natural balsam or resin, and from
the hard resin obtained when the turpentine has been
distilled. The preparation of this pitch will be
found described in the 21st chapter of the sixteenth
book. This pitch, Pliny informs us, was used for
tarring ships. There is also another artificial
preparation described by him called zopissa, which
was apparently used in medicine, and was scraped
from the bottoms of ships. This he states to have
been a mixture of pitch and beeswax. The bees-
wax would doubtless be introduced to toughen the
pitch. A description of this substance is given in the
23rd chapter of book sixteen. The resins already
referred to were used, either to add to wine, or to
paint the inside of the wine vessels. They were
also, of course, used in medicine. There is no hint or
suggestion in Pliny of their being used to varnish
pictures or other objects. Below will be found the
references to the chapters where resins are men-
tioned, with the exclusion of those used as perfumes,
such as frankincense.^
He tells us that all resins are soluble in oil,^
describes the extraction of a resin by oil,* and de-
1 xii. 36 ; xiii. 11 ; xiii. 12 ; xiv.' 25 ; xvi. 16-23 ; xvi. 76 ; xxiv.
11 ; xxiv. 22, 23, 24, 25, 26, 28.
^ xiv. 25. ' xxiv. 1.
30 GREEK AND ROMAN
scribes the preparation of liniments by dissolving
resins in oil.^ When the nature of the oil is not
specially mentioned, no doubt olive oil is intended.
We are not justified, therefore, in supposing that
the preparation of an oil varnish by dissolving the
resin in a ' drying ' oil was understood, on the
evidence merely of these receipts for medical pre-
parations by the use of olive oil.
A large number of oils were also known in the
time of Pliny, though no distinction was made be-
tween fixed oils like olive oil and essential oils. A
list of these oils will be found ^ in the 7th chapter of
the fifteenth book, and some account of their method
of extraction. They are either extracted by pressing,
or by boiling the substance containing the oil with
water, or by mixing the substance first with olive
oil and then pressing, by which a solution of the
required oil in olive oil would be obtained.
It is in this chapter that the preparation of wal-
nut oil is mentioned. There are two other references
to walnut oil. It is mentioned as a remedy for
certain complaints,^ and a receipt for bird lime consists
of walnut oil mixed with the juice of the mistletoe
berry.* This receipt is very interesting, as no doubt
the drying properties of the oil would make the
mixture stickier on exposure to the air.
^ xxiv. 22. '' XV. 7.
'^ xxiii, 45. ^ xvi. 94.
METHODS OP PAINTING 31
The preparation of linseed oil and poppy oil are
not given, but the preparation of poppy oil is given
by Dioscorides.^
Several other interesting oils are raentioiaed, such
as castor oil.
It is evident from this account of the oils and
resins known in the time of Pliny that the wMe-
rials for oil painting and for making oil varnishes
were known. We shall next consider whether the
materials for making spirit varnishes were also avail-
able. These are spirits of turpentine, alcohol, and
petroleum.
There are several references to petroleum in Pliny.
It is mentioned in the 109th chapter of the second
book, in the 39th chapter of the thirty-first book, and,
under the name of liquid bitumen, in the 51st chapter
of the thirty- fifth book. He distinguishes between
different kinds, and states that it is used for burn-
ing in lamps. Some writers have therefore assumed,
on the strength of these references, that it was used
in classical times as a painting medium and in the
preparation of varnishes. Natural petroleum is quite
unsuitable for such a purpose, and requires to be
most carefully rectified by distillation before it can
be used. So that we may dismiss this hypothesis,
as the true art of distillation was apparently un-
known at the time of Pliny, the first descriptions
' Diosoorides, B. iv. 65.
32 GREEK AND ROMAN
occurring at a later date.^ For this same reason
alcohol was not available.
A description of a crude method of distillation is,
however, to be found in Pliny in his chapter on oils,^
where he states that, by boiling pitch and covering the
boiling pot with fleeces and then wringing them out,
a yellow oil is obtained which he calls pissinum. This
oil is again referred to as pisselaeon.^ It has been
assumed that this statement in Pliny justifies us in
stating that in classical times spirits of turpentine
was known and doubtless used in paiijtLng. It has
not been noticed apparently that this oil was not
distilled from the pine balsam or oleo resin, but from
the pitch, the method of preparing which has already
been described. Such an oil would be very different
from spirits of turpentine, and of very doubtful value
as a medium, or for the preparation of varnishes.
Finally, it may be pointed out that the assump-
tion that the art of distillation was not known in the
time of Pliny, because it is not mentioned, is streng-
thened by the description given by him of the process
for obtaining an oil from pitch. This crude process
no doubt represents the first attempt at distillation
before the idea of a closed-in vessel, and a descending
pipe leading from it, had occurred to any one.
It is evident then that spirit varnishes artificially
prepared were not known in the time of Pliny,
1 See Appendix III. ^ xv. 7. ^ xxiv. 24.
METHODS OF PAINTING 33
though natural spirit varnishes or balsams were
available. Such varnishes were used in Egypt in
the 19th dynasty for varnishing cofSns, and the
varnish is still in good condition,^ and therefore
there is no reason why they should not have been
used by classical painters. The evidence provided by
Pliny is, however, against their having been generally
used, for, speaking of Apelles Pliny says, ' Inventa ejus
et ceteris profuere in arte ; unum imitari nemo potuit,
quod absoluta opera atramento inlinebat ita tenui, ut
id ipsum, cum repercussum claritatis colorum omnium
excitaret, custodiretque a pulvere et sordibus, ad
manum intuenti demum appareret, sed et lurainuTn
ratione magna, ne claritas colorum aciem offenderet
veluti per lapidem specularem intuentibus et e lon-
ginquo eadem res nimis floridis coloribus austeritatem
occulte daret.' ^ ' His innovations in the art of paint-
ing have also been useful to others ; but one of them
nobody has been able to imitate. He used to cover
his pictures after their completion with a layer of
atramentum so thin, that while it created a reflexion
of the brightness of all the colours and protected
them from dust and dirt, it was itself visible only to
one examining very closely. But the chief purpose
was to prevent the brightness of the colours from
offending the eye (they were as if looked at
through talc), also that when seen from a distance
' Appendix II. ^ xxxv. 36.
C
34 GREEK AND ROMAN
the too florid colours might be imperceptibly
chastened.'
The general impression made by this description
is that of the result of varnishing a picture painted
in some medium like tempera, by which the dead
dull look of the colours is at once removed.
The use of the word atramentum, and the remarks
about the softening effect, suggest, however, that the
artist covered his picture with a thin glaze of bitu-
men. Some of the more fluid bitumens could pro-
bably be used for such a purpose. The beautiful
effects which bitumen is capable of giving were only
too well known to the early painters of the nineteenth
century as their pictures now testify.
Atramentum properly means black, and is con-
fined by Vitruvius^ to lamp black and charcoal
black. On the other hand Pliny, in speaking of
atramentum,^ mentions an atramentum which exudes
from the earth like the brine of salt pits. This
must surely be some form of bitumen, in which
case the use of the word atramentum, in speaking
of the process used by Apelles, would be justified.
It is evident that Pliny himself did not understand
the process, and that it was some method of varnish-
ing, though possibly with a bituminous or semi-
bituminous varnish. A varnish, for instance, in
which bitumen was dissolved in a pine balsam like
' vii. X., Vitruvius. ^ xxxv. 25, Pliny.
METHODS OF PAINTING 35
Venice turpentine, would fulfil all the conditions laid
down in Pliny's description.
It was also a process peculiar to Apelles, for Pliny
speaks of it as having been used only by him, stating
that in this no one could imitate him.
The CYidence, therefore, of this passage is in favour
of the conclusion that the varnishing of pictures was
not practised by classical painters, with this one excep-
tion. On the other hand it is difficult to understand,
since they had natural varnishes in abundance, and
since the use of them was so well understood in Egypt,
why they were not used by classical painters. This
is one of the cases where we have to decide between
the facts derived from the actual examination of
contemporaneous objects, and the authority of our
author. While it is difficult to imagine that Pliny
was not familiar with the artists' studios of his time,
or that the special significance of the passage lies in
the use of the word atramentum, the dark glazing
being the special invention of Apelles, yet it is not
easy to see why other artists should have had any
difficulty in repeating his process. On the whole I
am disposed to think that the classical painters
were probably quite familiar with the advantage of
varnishing their pictures with the natural balsams,
and that Pliny is speaking in ignorance of the tradi-
tions of the studios.
Whether the pictures were varnished or not, we
36 GEEEK AND ROMAN
have so far identified three possible media for paint-
ing, gum, glue, and egg. For pictures that did not
require to be waterproof these media would be
sufficient, but as soon as a demand came for a
medium which should be waterproof, and at the same
time pliable in handling and not too brittle when
dry, it is evident that we have a choice between two,
a drying oil, like nut or linseed oil, and beeswax.
We have found no evidence of a knowledge of
the properties of drying oilj and therefore we are
bound to assume, as a working hj'pothesis, that such
knowledge was absent. If we find definite proof of
the use of beeswax, it is a further confirmation of
the absence of, at any rate, a practical working
knowledge of drying oils, as no one would use bees-
wax if he knew of something more convenient, and
any one who has worked in both media will un-
hesitatingly prefer oil.
The problem of obtaining a waterproof medium
would probably arise, in the first instance, in connec-
tion with ships. The substance already described,
zopissa, the mixture of beeswax and pitch which
was used for coating the sides of ships, would be
a translucent brownish yellow body. If a desire
arose for decorating ships it could be easily done
by mixing pigments with this substance and laying
them on with the tar brush in bold designs, as the
colour of the mixture would not affect the colour of
METHODS OF PAINTING 37
the pigments more than a dark varnish would do.
Experience in this rough decorative treatment might
well lead to a more refined use of wax mixed with
pigments, as a medium, without the sticky dark
coloured tar, both for ship decoration, and for paint-
ing on panels.
At this point, however, the technique derived
from the practice with the tar brush would meet
another technique, namely that of modelling in
coloured waxes, and from the two combined a special
kind of pictorial art would develop. We shall see
how far this conjectural view agrees with the infor-
mation at our disposal. We shall begin by considering
the statements made by Pliny about the preparation
and properties of beeswax itself, by quoting in full
chapter 49 of the twenty-first book : —
' Cera fit expressis favis, sed ante purificatis aqua
ac triduo in tenebris siccatis, quarto die liquatis igni
in novo fictili, aqua favos tegente, tunc sporta colatis.
Rursus in eadem oUa coquitur cera cum eadem aqua
excipiturque alia frigida, vasis melle circumlitis.
Optima quse Punica vocatur, proxima quam
maxime fulva odorisque mellei, pura, natione autem
Pontica, quod constare equidem miror inter venenata
mella: dein Cretica, plurimum enim ex propoli
habet, de qua diximus in natura apium. Post has
Corsica, quoniam ex buxo fit, habere quandam vim
medicaminis putatur. Punica fit hoc modo : ven-
38 GREEK AND ROMAN
tilatur sub diu saepius cera fulva, dein fervet in
aqua marina ex alto petita, addito nitro. Inde
lingulis hauriunt florem, id est candidissima quaeque,
transfunduntque in vas, quod exiguum frigidae habeat,
et rursus marina decocunt separatim, dein vas ipsum
aut aquam refrigerant. Et cum hoc ter fecere, juncea
crate sub diu siccant sole lunaque. Haec enim
candorem facit, sol siccat, et, ne liquefaciat, pro-
tegunt tenui linteo. Candidissima Tero fit post
insolationem etiamnura recocta. Punica medicinis
utilissima. Nigrescit cera addito chartarum cinere,
sicut anchusa admixta rubet, variosque in eolores
pigmentis trahitur ad reddendas similitudines, et
innumeros mortalium usus parietumque etiam et
armorum tutelam. Cetera de melle apibusque in
natura earum dicta sunt.' ^ 'Wax is made from
honeycombs out of which the honey has been
pressed. Having been first cleaned with water and
dried for three days in the shade, the combs are on
the fourth day melted on the fire in a new earthen
vessel with water enough to cover them, and then
strained off in a wicker basket. The wax is again
boiled in the same pot with the same water, and is
poured into cold water contained in vessels the
interior of which has been smeared all over with
honey. The best wax is that called Punic, the next
that of a very yellow colour with the smell of honey,
1 xxl. 49.
METHODS OF PAINTING 39
which, though of Pontic origin, is unaffected, I am
surprised to find, by its poisonous honey. Next best
is the Cretan, for it has a large proportion of propolis
of which we spoke when treating of bees. Next to
these is the Corsican, which, as it comes from the
box-tree, is believed to have medicinal qualities.
Punic wax is prepared as follows : Yellow wax is
exposed to the outside air for some time, then boiled
in sea-water taken from the open sea, with nitrum
added. Then the flower, that is, the whitest part, is
skimmed off and poured into a vessel containing a
little cold water. Again it is boiled in sea- water by
itself, then the vessel, or at least the water, cooled.
When this has been done three times the wax is
dried in the open air on a mat of rushes in the light
of the sun and the moon. For the latter makes it
white, the sun dries it, and lest it should melt it is
covered with a thin linen cloth. It will become
exceedingly white if it is boiled again after the
exposure to the sun. Punic wax is the most useful
for medicines. Wax becomes black when papyrus
ash is added to it. It becomes red when mixed with
alkanet ; with pigments it is made to assume various
colours in order to represent true likenesses of objects.
It is useful to men in numberless ways, even serving
as a protection for walls and weapons. Other par-
ticulars concerning bees and honey we have stated
when speaking of the nature of these insects.'
40 GREEK AND ROMAN
It is evident from this passage that it was cus-
tomary to mix pigments with wax, and the coloured
mass was used for modelling portraits, and wax was
used as a protective coating for walls and armour. It
is necessary here, however, to make a somewhat long
digression in order to consider the statements made
by Pliny about Punic wax and the method of purifying
it with nitrum. A description of nitrum is given by
Pliny in xxxi. 46. The description is obscure, and he
may well be referring to more than one substance,
but we are on the whole justified in identifying it with
a natural efflorescence found in some of the desert
regions on the Mediterranean, and proving to be a
double salt of carbonate and bi-carbonate of soda.
For experimental purposes, therefore, we can replace
it by ordinary sodium carbonate, which will, how-
ever, be the more powerful alkali of the two.
Fee ^ and John ^ both rightly concluded that the
treatment with soda and salt water would have no
effect on the purification of the wax, and the matter
might be left here if Herr Berger had not revived
an old idea in a new form, namely, that a special
medium of wax and soda was used in classical times,
consisting of a wax-soda soap or emulsion, an idea
which he bases largely on this passage in Pliny.^
' F^e : Sur la MatUre MddicaU et la Botanique de Pline.
" John : Malerei der Alien, Berlin, 1836, p. 204.
' 'Der Zusatz von Nitrum bezweokt die Verseifung oder nur
Bmulgierung des Waohses,' etc. Die Maltechnik des AUerthums,
vol. i. p. 100.
METHODS OF PAINTING 41
It is impossible to give all the references in Berger
dealing with this subject, as the principal object of
the whole book is to prove that this medium alone,
or in combination, was used for panel and wall
painting.
In the first place, therefore, it is necessary to consider
whether the process described by Pliny would result
in the saponification of the wax. Beeswax consists
principally of myricyl palmitate with about ten per
cent, of cerotic acid. Myricylic alcohol is insoluble
in water, and the soda salts of cerotic acid are also in-
soluble. While, therefore, beeswax readily saponifies
in an alcoholic solution of soda, it is not probable that
it would be saponified by a solution of soda in water.
Moreover, in the British Pharmacopoeia the test for
the purity of beeswax is to heat it with a solution of
soda in water. If any soap is formed this shows the
presence of adulteration with fats. This test might
be regarded as conclusive, but the question as to
saponification is so important that I determined to
test the matter further. I therefore took some
beeswax of undoubted purity, of which I knew the
history, and boiled it under an inverted condenser with
a saturated solution of carbonate of soda for twelve
hours. At the end the whole mass was treated with
ether, and both the water solution and the solution
in ether examined. Neither contained the slightest
indication of a soap having been formed. The question
42 GREEK AND ROMAN
of saponification may be considered therefore as
finally settled.
We shall next consider the question of the forma-
tion of an emulsion. Herr Berger states quite cor-
rectly that when beeswax is melted in pure water
and then shaken up| no emulsion is formed, but that
the introduction of a small quantity of soda at once
makes it possible to form an emulsion in this
way. The question therefore is whether Punic wax,
prepared in the way described by Pliny, can be con-
sidered to be an emulsion of beeswax and soda
solution. If the beeswax is merely boiled with the
soda without shaking up, it collects on the top and
very little emulsion is formed. If the emulsion is
formed by shaking up, prolonged warming in a water
bath causes the wax to separate again and float on
the top. If, then, we suppose a considerable quantity
of wax to be boiled in this way with soda, we should
expect to find the bulk of it collecting on the top,
just as when it is boiled with water.
Following the description by Pliny, we find that he
states that the whitest part of the wax is skimmed off,
is poured into cold water, and is bleached iii the open
air upon a mat of rushes. It is quite evident that
this bleaching could only be applied to dry, solid
beeswax, which could not possibly therefore be an
emulsion as supposed by Berger. It might certainly
accidentally contain a little soda mechanically caught
METHODS OF PAINTING 43
in the wax. But the boiling with sea water alone
would serve to remove this accidental impurity, and
as I have shown no soda is chemically combined
with it. The resulting product will therefore simply
be bleached beeswax, and will not differ from the
wax prepared with water alone.
In order to strengthen the view that Punic wax was
something quite different from ordinary wax, and was
specially used for treating walls, Berger points out that
when talking of ordinary encaustic Pliny speaks of
wax, but, when discussing wall painting, he speaks of
Punic wax,^ and that similarly Vitruvius, in speaking
of the treatment of walls with wax, speaks of Punic
wax.^
In Mayhoff's edition of Pliny the word is given as
pvmica. But in the corresponding passage in Vitru-
vius (vii. 9) we find that Eose, in his later edition
alters the punica of his former edition to pontica.
From a note to the text we find that in the oldest
MSS. (see below, p. 78, note) the word appears, not as
punica, but as pumica. Rose refers us further to
his edition of the works of the physician Theodoras
Priscianus, where we find the following note : ' pontice
fere legitur ut pomice, unde ponice, punice, quod
delendum.' In another note on ' cera pontica ' : ' falsa
lectio punica' (in certain MSS. of Theod. Prise.) 'ut
etiam apud Plinium alibi.'
' Pliny, xxxiii. 40. " Vit. vii. 9.
44 GREEK AND ROMAN
It is evident that whether Rose is right in his
reading or not, no argument can be based on the use
of a word which a commentator of the authority of
Rose does not consider to be a correct reading.
It will also be noted that Pliny states that Punic
wax is best for medicinal uses, yet when discussing
the medicinal use of wax ^ he speaks simply of wax
without the word Punic. If he regarded Punic wax
as something quite distinct from ordinary wax, and
used the word Punic whenever he wishes it and
not ordinary wax to be used, it is difficult to under-
stand why he does not do so in the chapter on the
use of wax for medicinal purposes.
If we inquire further from what source this idea
of the use of an emulsion of wax and soda comes,
we find two sources.^
In the ffermenem, obtained by Didronfrom the Mon-
astery of Mount Athos, and supposed to be written in
the sixteenth century, claiming to be the teaching of
a painter of the eleventh or twelfth century, a receipt
is given for a medium consisting of equal quantities
of potash lye, wax, and size melted on the fire. While
this medium is mentioned, the evidence to be drawn
from the Hermeneia as a whole is in favour of Qgg
being the customary medium.^
' xxii. 55.
^ Berger's receipt for what he calls Punic wax will be found on
page 100 of the Maltechnik des Alterthums, vol. i.
' Didron : The Hermeneia, a MS. found by him at Mount Athos.
METHODS OF PAINTING 45
A similar receipt exists in the MS. of Jehan Le
Begue/ which begins thus: 'Prenez une livre de
chaux et douze de Flandres puis prenez eaue
boulant,' etc. This receipt refers to the presence of
potash, only if we accept Mrs. Merrifield's emenda-
tion of ' Cendres ' for ' Flandres.'
Such a medium would in reality be a glue medium
with wax intermixed. It is hardly necessary to
point out that a medium containing free soda or
potash is very objectionable, and would probably
soon result in the destruction of the picture.
But it is not necessary for us to discuss the merits
or demerits of this medium. It is sufficient for our
purpose that it appears for the first time in MSS.
which are not earlier than the end of the fourteenth
or beginning of the fifteenth century, and that there-
fore they have no bearing on the subject we are
discussing, namely, the medium used before and up
to the end of the first century of the Christian era.
Sufficient has been said to show that Pliny clearly
does not describe such a preparation, and that we
may dismiss it as one of the ingenious fictions
that have so long obscured the scientific investiga-
tion of the classical methods of painting. I shall
have to refer to it again, however, when discussing
Herr Berger's theories about metho'ds of wall
painting.
1 Mrs. Merrifield, Ancient Practice 0/ Painting, vol. i. p. 307.
46 GKEEK AND EOMAN
Having now cleared the way we can proceed to
consider the actual information given by Pliny about
pictures on panel, linen, or parchment, leaving the
question of wall painting for later consideration.
In the first place, it is necessary to consider the
evidence in favour of two mediums having been
used, one by painters ' with the brush,' and the other,
wax, ' with the cauterium and the brush.'
In the 34th chapter of the thirty-fifth book.
Pliny begins by stating that he is going to enumerate
as briefly as possible the more eminent among
painters, 'Nunc celebres in ea arte quam maxima
brevitate percurram,' etc. After discussing in this
chapter the early beginnings of painting, and in the
next chapter the institution of pictorial contests
at Corinth and Delphi, he begins to describe the
great painters in detail in the 36th chapter, and
speaking of Apollodorus of Athens, says: — 'Hie
primus species exprimere instituit primusque gloriam
penicillo jure contulit.' 'He was the first to depict
objects as they really appeared, and he first conferred
a just glory on the brush.' And again further on,
speaking of Zeuxis of Heraclea, he says : ' Ab hoc artis
fores apertas Zeuxis Heracleotis intravit olympiadis
Lxxxxv anno quarto, audentemque jam aliquid
penicillum — de hoc enim adhuc loquamur — ad mag-
nam gloriam perduxit,' etc. ' Through the gates of art
thrown open ' (by Apollodorus), ' Zeuxis of Heraclea
METHODS OF PAINTING 47
passed in the fourth year of the 95th Olympiad, and
brought the brush — for it is of the brush we are stiU
speaking — to the great glory at which it was already
aiming.'
He is evidently speaking here of artists who
painted with the brush. This, however, is not con-
clusive evidence that they were not painters in wax,
as we shall iind that he tells us that one method of
painting in wax was with the brush. But we shall
find, as we follow up the quotations, that he is
evidently referring to a school of painters who used
a different medium from the wax medium. These
pictures with this medium were evidently painted
on panels, as frequent references to panel, one of
which has already been quoted, are made in this
chapter.
The pigment mixed with this medium could be
easily removed from the picture with a sponge,
because he tells us that Protogenes, not satisfied with
his success in painting the foam on the mouth of a
dog, frequently removed the paint with a sponge,
and at last threw the sponge at the picture, when
the wet colour on the sponge was thus replaced on
the picture and gave the required effect.
This story certainly suggests the use of a medium
soluble in or mixed with water, though it might
equally happen in the case of oil painting. It could
not have happened with a wax medium, except in
48 GREEK AND ROMAN
a very hot climate like that of Egypt during the
summer months. We also know from this chapter
that the pictures of this school of painting were
destroyed by damp.
In chapter 37, Pliny proceeds to discourse on
painters of an inferior rank, who, however, were
' painters with the brush.' He says, ' Namque subtexi
par est minoris picturae celebres in penicillo : '
' It is well to add some mention of those who won
fame by the brush in a minor kind of painting.'
In this chapter he mentions the painting of stage
scenery and the painting of walls, so that there is
clearly no distinction in his mind between the
method used for these three purposes, namely panel
pictures, stage scenery, and wall decoration: they
are all ' painters with the brush.'
It is also in this chapter that the method of
varnifihing pictures invented by Apelles is mentioned.
In chapter 38 a method of frightening birds by
picture of a dragon painted on parchment is described.
In chapter 39 he discusses who was the inventor
of painting in wax, and then^ proceeds to give an
account of famous painters in this medium. Speak-
ing of Pausias, who had been taught encaustic
painting, he says that he retouched ' with the brush '
some walls at Thespiae, but was not successful, as
this kind of painting was evidently not in his line.
^ Pliny, XXXV. 40.
METHODS OF PAINTING 49
' Pinxit et ipse penicillo parietes Thespiis, cum
reficerentur quondam a Polygnoto picti, multumque
comparatione superatus existimabatur, quoniam non
suo genere certasset.' 'He also painted with the
brush some walls at Thespiae when they were being
repaired. They had been previously painted by
Polygnotus, and the work of Pausias was considered
to suffer much by comparison, for he was contending
in a style not his own,' thus apparently indicating that
wax painting was not the method usually used for
painting on walls.
The next sentence, however, throws some doubt on
this conclusion, as Pliny says : — ' Idem et lacunaria
primus pingere instituit, nee camaras ante eum
taliter adornari mos fuit; parvas pingebat tabellas,
maximeque pueros.' ' He first started painting coffer-
ceilings, nor had it been the custom before him to
decorate arched roofs in this way. He usually
painted small pictures, chiefly of children.' Unless
we are intended to draw a distinction between instituo
and pingo, and to understand that while he intro-
duced the decoration of ceilings, not necessarily exe-
cuted by his own hand, he painted, with his own
hand, on panel small pictures of children.
Further light is thrown on the process of painting
in wax by the following quotation: — 'Hoc aemuli
interpretabantur facere eum quoniam tarda picturae
ratio esset ilia. Quam ob rem daturus ei celeritatis
D
60 GREEK AND ROMAN
famam absolvit uno die tabellam, quae vocata est
hemeresios, puero picto.' ' This his rivals interpreted
as due to the slowness of that manner of painting.
Wherefore wishing to gain the reputation of
celerity he completed a picture in one day. It
was a portrait of a boy, and was called "a day's
work." '
So tedious a process, in which the execution of a
picture, it is to be presumed of small size, in a
single day was regarded as a remarkable feat, is,
apart from other reasons to be considered later,
further evidence that the wax technique was not
convenient for large surfaces of walls.
Further on in this chapter, after mentioning
various painters and their principal works, Pliny
says: — 'Hactenus indicatis proceribus in utroque
genere non silebuntur et primis proximi: Aristo-
clides, qui pinxit aedem Apollinis Delphis.' ' Though
I have so far mentioned only the leading painters
in each branch, I will not pass over in silence those
of the second rank: Aristoclides who painted the
temple of Apollo at Delphi.'
We have here a further proof that there were two
schools of painters, those who painted 'with the
brush,' and those who painted in wax.
It is not clear, however, whether in the remainder
of the chapter Pliny is speaking still only of the wax
painters or including both schools.
METHODS OF PAINTING 51
In confirmation, in the table of contents we find
as follows : —
' Qui penicillo pinxerint
De avium cantu compescendo
Qui encausto cauterio vel cestro vel penicillo
pinxerint.'
' Those who painted with the brush
On silencing the singing of birds
Those who painted in encaustic with the
cauterium, the cestrum, and the brush.' ^
These three methods of encaustic painting will next
have to be considered, but the distinction is clear
between those who painted with the brush, and those
who painted in encaustic with the brush. The
story about the birds divides for us the part of the
book dealing with the one kind from the part deal-
ing with the other kind of painting.
In the 41st chapter the different kinds of wax
painting are described, and in the 42nd chapter
a special method of painting or dyeing cloth used in
Egypt, and so the chapters on painting end.
It is, I think, clearly estabhshed by these quota-
' I have adopted the reading 'encausto cauterio' restored by
Mayhoflf on the authority of the Codex Bambergensis, the most
ancient extant. It was Sillig who altered ' cauterio ' into ' aut
ceris,' wishing to harmonise the Index with the Text xxxv. 41,
where it runs ' cera et in ebore cestro.' For MayhofPs proposed
emendation of xxxv. 41, see below, p. 60.
52 GREEK AND ROMAN
tions that the painters in classical times painted on
panel as well as on walls, and that there were two
quite different methods of painting panel pictures,
the one defined by Pliny as ' painting with the
brush,' and the other painting with wax.
While the evidence is not conclusive, it is on the
whole in favour of the view that Pliny regarded
'painting with the brush' as equally suitable for
panels, large canvas pictures, scenery, and wall paint-
ing, while the painting in wax was confined to panels,
probably of small size, and that the process was
recognised as a tedious one unsuitable for large sur-
faces.
The medium used for ' painting with the brush '
is nowhere described, but the story of the use of the
sponge certainly suggests that it was a medium
which could be mixed with water.^ The fact that
the medium is not mentioned is strongly in favour
^ If the following simile of Plutarch's may be pressed so far, I
would suggest that we find in it strong evidence of the use of a
water medium for panel pictures, and that it agrees with Pliny's
statements as to the destruction of paintings. It is unlikely that
Plutarch is alluding to buon fresco, which is extremely durable :
7} yap 6}pLS hiKe rets fih fiXXas (pavTa^ias itp' {lypdls ^wypafpelv, Taxi/
fxapaivofUva? Kal diroXenro^ffas t^v dtdvoiaVj at 5^ tCjv ^pwfj^vwv dK^ves
hir aiTTjS olov iv ^yKaijf/.a(n ypatpdfievat 6tct irvpbs ei'SwXa rah fivrifxaLS
ivaTToKeliroucrt KLVoi^ieva koX ^Covra Kal tjideyyb^eva Kai irapap/Jvovra rbv
&\\op xpl"""'- ' For, indeed, sight seems to paint all other fancies
on a wet ground, so swiftly do they fade and leave the mind, but
the images of those beloved, painted by it as it were in encaustic
by means of fire, leave behind in the memory shapes which move
and live and speak and remain for ever.' — Amatorius, 759 0.
METHODS OF PAINTING 53
of the conclusion that it was either egg, gum, or
gluej or any one of these according to the fancy of the
artist. These mediums are so obvious that it might
well never occur to a writer to mention them. The
same omissions are to be found in later MSS., where
the special methods and materials for laying on
gold, etc. etc., are mentioned, but it does not occur
to the writer to say with what he mixed his
colours.
Of the three . mediums mentioned, the early and
almost universal use of egg as a medium in Italy
and elsewhere, and the Byzantine traditions, point to
this being most probably the medium used in classical
times.
There is no need, however, to imagine that there
is any mystery in the matter. These substances
were known, just as they were known in the
fifteenth century, and were used probably one for
one purpose and one for another, just as they were
used in later times.
We find, for instance, that in order to prepare a
proper painting surface the Egyptian of the nineteenth
dynasty covered the wood of the coffin with a gesso
of chalk and glue, and we find Oennino Cennini, in the
fifteenth century, directing panels to be prepared in
exactly the same manner, while the directions given
by Pliny for using white of egg in gilding agree with
the directions given by Oennino Cennini. It is surely
54 GREEK AND ROMAN
reasonable, therefore, to suppose that before the
revolution produced in painting by the introduction
of the use of drying oils, artists made use of the
three obvious mediums at their command — egg, glue,
and gum, and that probably just as egg was found
to be the most suitable in the fifteenth century,
so it was found to be the most suitable in the first
century.
At any rate, until we have further information,
there can be no justification for devising mediums,
made of mixtures of wax and soda, for which there
is not a particle of historical evidence, and for exer-
cising our ingenuity in getting away as far as possible
from the obvious.^
We have next to consider what information can be
derived from Pliny about painting in wax.
Some writers "have suggested that the wax medium
was the wax soda emulsion, others have suggested
that the wax medium used by Greek painters was
an emulsion of wax in turpentine, or of wax in a
mixture of drying oil and turpentine, or of a drying
oil, varnish, and turpentine. Such mediums, on the
purely negative evidence which we have obtained
from a study of Pliny, are quite inadmissible.
We shall have to consider three difit'erent chapters
of the thirty- fifth book in order to understand clearly
^ Herr Berger believes this medium to have been a wax soda or
potash emulsion.
METHODS OF PAINTING 55
what Pliny has to say about encaustic painting,
In the 31st chapter, after discussing the various
pigments, he says : ' Ex omnibus coloribus cretulam
amant, udoque inlini recusant purpurissum, Indicum,
caeruleum, Mehnum, auripigmentum, Appianum,
cerussa. Cerae tinguntur isdem his coloribus ad eas
picturas, quae inuruntur, alieno parietibus genere,
sed classibus familiari, jam vero et onerariis navibus,
quoniam et pericula expingimus, ne quis miretur et
rogos pingi, juvatque pugnaturos ad mortem aut
certe caedem speciose vehi.' ' Of all colours those
which love a chalk ground and refuse to be laid on a
damp surface are purpurissum, indigo, caeruleum,
Melian white, orpiment, Appianum, and white lead.
Waxes are stained with these same colours for pic-
tures in encaustic, a kind of painting unsuitable for
walls, but commonly used for ships of war, and now
also for merchant ships. Since we paint even those
vehicles of danger, no one should be surprised if we
also paint our funeral piles, and like to have gladiators
conveyed in splendid carriages to death or at least to
carnage.'
The first part of this chapter dealing with the
pigments which cannot stand a wet surface we .shall
consider later.
We have it here clearly stated that all these pig-
ments were used for colouring wax for encaustic
painting, and that the process was not suitable for
56 GREEK AND ROMAN
walls, but was used to decorate ships. As we know
from the chapters already quoted, encaustic painting
was not confined to ship decoration. We have here
a confirmation of the former probable conclusions
that encaustic was seldom used, at any rate in
the time of Pliny, for wall decoration, but we know
from the 64th chapter of the thirty-sixth book that
Agrippa had the potter's work in the baths painted
in encaustic (figlinum opus encausto pinxit in calidis,
reliqua albario adornavit). This must have been an
ornamental terra-cotta work introduced as part of a
scheme of wall decoration and painted either before
or after it was put in position. Evidently Pliny
regarded this as a proof that he would have spared
no expense, and would therefore have used mosaic
if it had been invented in his time. This is a close
approach to treating the whole wall with encaustic,
if the terra-cotta was painted after erection. It will
be evident, as we proceed to consider in more detail
the nature of the process, that while not impossible
to apply to walls, the process would be difficult.^
' In the inaoription from Athens, which records the building
accounts of the Erechtheion, occurs the entry of a sum paid to the
encaustic painters for having painted the cymatium on the epi-
sbylium of the interior : iyKUXKnai^ rb Kv/juiriov ijKiavTL t4 ^Tri ti?
iinaTvKlif ti^ ivrbs (Cecil Smith, article 'Pictura' in Smith's Ant.).
That encaustic was used at a later date for ceiling decoration seems
clear from the statement in Procopius that Justinian, on restoring
the imperial palace, had the ceilings decorated, not with paintings
in melted wax, t(^ ki)/)iJ! iuTaKhri re /tai SiaxvOivr^, but with mosaic.
METHODS OF PAINTING 57
After discussing the artists who painted 'with
the brush/ he speaks as follows in chapter
39:—
'Oeris pingere ac picturam inurere quis primus
excogitaverit, non constat. Quidam Aristidis in-
ventum putant, postea consummatum a Praxitele;
sed aliquanto vetustiores encaustae picturae exsti-
tere, ut Polygnoti et Nicanoris, Mnesilai Pariorum.
Elasippus quoque Aeginae picturae suae inscripsit
iveKaev, quod profecto non fecisset, nisi encaustica
inventa.' 'It is not agreed who first thought of
painting with wax colours and making a picture by
heat. Some think the art was invented by Aristides
and afterwards brought to perfection by Praxiteles.
But there are in existence encaustic pictures of a
date somewhat earlier than theirs, such as those by
Polygnotus, and by the Parians Nicanor and Mnesi-
laus. Elasippus also wrote on his pictures at
Aegina iveKaev, which he certainly would not have
done unless encaustic painting had been invented.'
In the former chapter we have had the coloured
wax associated with the verb inurere. In this
chapter inv/rere is brought into association with
encaustic, and eyKaieiv, so that there can be no doubt
of their being used to describe the same process of
painting.
The opening sentence of this chapter, ' Ceris pin-
gere ac picturam inurere,' has given rise to serious
58 GREEK AND ROMAN
misconception as to the probable nature of tbe
process. It has been assumed by many writers on this
subject, of whom Eastlake and Conner are examples,^
that this means that the painting process was done
first, and the heating of the surface done afterwards.
They have therefore tried to devise mediums which
would on the one hand be fluid under the brush
when cold, and could be subsequently fused by
heating without injuring the picture. I do not like
to say that to produce a picture under such con-
ditions is impossible, but the experiments I have
made with wax dissolved in turpentine, and the
attempts I have afterwards made to fuse the sur-
face, do not encourage me to believe that anything
can be done ,in this way. Unless the wax is in great
excess the fusing process is quite useless as a method
of attaching more firmly the comparatively easily
removed wax turpentine painting, and if it is in
sufficient excess to fuse, the picture itself is inevitably
injured. Such wax turpentine painting, even after
subsequent heating, does not make nearly so durable
a job as starting with the fused wax right away.
The introduction of oil instead of turpentine increases
the difficulties of the operation.
There is no need, however, to translate this
sentence as if the ' inurere picturam ' followed the
^ Eastlake, Materials for History of Oil Painting, p. 152.
Conner, Mitteil. des archdol. Inst. Bom., Abtheil. xiv., 119.
METHODS OF PAINTING 69
process of 'ceris pingere/ as Mayhoff^ has correctly
pointed out. Without entering into the subtilties
of the various uses of ac, it is sufficient for our
purpose to state that ac denotes a closer connection
than that implied by et, and can be used therefore
to represent the same idea in a slightly different
form. This sentence may be therefore regarded
as an attempt to give a complete description of the
process of encaustic painting, by mentioning that it
was a process for painting, it involved the use of wax,
and it also involved the heating of the wax during
the painting process.
Moreover, Mayhoff points out that if Conner's
view be accepted the passage should run, 'Ceris
pingere pictaque inurere.'
The correct translation of ejKaio), and therefore
of inv/ro is difficult. Mayhoff, in the communication
to Herr Berger, which has been already referred to,
considers that the correct translation is not ' einbren-
nen,' but rather 'aufbrennen' — i.e. heiss auf den
Malgrund auftragen.^
At the end of the section dealing with encaustic
painters, the following chapter occurs^: — 'Encausto
1 Communication to Ernst Berger, printed in Die Maltechnik des
AUerthums, vol. i. p. 189.
• '■' This view is oonflrmed by the following allusion to the process
in Plutarch : eUbves iv iyKai/jiain ypa<fi6/jievaL Blo, irvpds — ' images
painted in encaustic by means of fire.' — Amatorius, 759 0.
^ Pliny, XXXV. 41.
60 GREEK AND ROMAN
pingendi duo fuere antiquitus genera, cera et in ebore
cestro, id est, vericulo, donee classes pingi coepere.
Hoc tertium accessit resolutis igni ceris penicillo
utendi, quae pictura navibus nee sole nee sale ven-
tisve corrumpitur.' 'In ancient times there were
[only] two methods of encaustic painting, with wax
and on iYory with the cestrum, that is with a sharp
pointed tool, until it became the custom to paint
ships of war. Then the third method was added,
that of melting the wax colours with fire and laying
them on with a brush. This kind of painting applied
to ships is not injured by sun, wind, or salt water.'
If we include along with this chapter the state-
ment made in the contents, 'qui encausto cauterio
vel cestro vel penicillo pinxerint,' we have evidently
here three methods of encaustic painting — with wax,
using the cauterium, on ivory, using the cestrum, and
again with wax, using the brush.
Professor Mayhoff has suggested an amended read-
ing of this chapter, replacing the word ' cera ' with
the word ' cauterio,' thus making the passage run,
that there were two ancient methods, 'one with
the cauterium, and the other with the cestrum on
ivory.'
Whether we accept this amended reading or not,
there is no dispute as to what one of the three
methods was. The cauterium is the name used
for the branding-iron for cattle, and for the instru-
METHODS OF PAINTING 61
iiient used in surgery for performing cauteries. It is
therefore a hot metal instrument, with which the
coloured sticks of wax could be moulded and
modelled on the panel into the required picture.
The only other instrument mentioned in this con-
nection is the pa^Siov, which Mayhoff, with great
probability, identifies with the cauterium.
The painting on ivory with the cestrum remains
up to this day an unexplained process. Many in-
genious suggestions have been made, but they are
merely suggestions unsupported by evidence. A
very complete discussion of possible explanations is
to be found in Herr Berger's Maltechnik, for which
the reference is given.^
Leaving then the painting on ivory with the
cestrum unexplained, we come to the third process,
namely, painting with the brush.
Pliny's description of this process is perfectly clear
and definite — 'hoc tertium accessit, resolutis igni
ceris penicello utendi '—and yet most of the writers
on this subject have refused to accept the process as
described by Pliny. Eastlake, for instance, at once dis-
misses it by saying that it is impossible to paint with
melted wax alone, as it cools too rapidly on the
panel.
The discovery, however, of actual wax portraits by
Flinders Petrie at Hawara, painted apparently in the
1 Vol. i. p. 223.
62 GREEK AND ROMAN
second century, and very little later than the time
we are considering, threw fresh light on the matter.^
Many similar pictures have since been discovered,
and they are remarkable for the fact that in many
both the modelling process with the cauteriuni and
work with the brush can be seen side by side.
Flinders Petrie was, I believe, the first to point out
that in Egypt, for many months, wax remains melted
in the sun, and that therefore far from presenting
difficulties from too rapid cooling on the canvas, the
difficulty would be the other way, and he mentions
a particular case where the eye of a portrait has got
smudged by a careless finger.^
The ease, therefore, with which melted wax can be
used as a medium for the brush is a question of
temperature ; as we move from Egypt to Greece, from
Greece to Rome, and from Rome to France, the
difficulty would increase, and special methods would
have to be adopted.
The experiments on wax-painting made by Herr
Berger are of great interest, and he has doubt-
less correctly identified the spoon-like instruments
found at St. Medard, with the cauterium, but on
this question of brush work with melted wax his
account of his experiments is not clear. He seems
^ Hawara, Biahmu, Arsinoe, Flinders Petrie. The Leadenhall
Press, 1889.
2 Hid., p. 19.
DIRECT PHOTOaRAPH PKOM OEIGINAL IN NATIONAL GALLEBT OF
POBTKAIT FKOM HAWABA (FLINDEEB PBTEIE).
To face page 62.
METHODS OF PAINTING 68
to have used a medium of wax, resin, and nut oil,
and to have varied the proportion, and he does not
state what these proportions were. The introduction
of a drying oil is, in the light of the information we
have derived from Pliny, quite illegitimate. The
introduction of a very small quantity of olive oil,
especially if counteracted by hardening the wax with
a resin, produces a mixture which is firm when cold,
but the addition of an appreciable quantity of olive
oil results in a soft mess which will not harden. If
the olive oil is replaced by a drying oil, then the
ultimate hardening of the surface depends on the
oxidation of the oil, and to all intents and purposes
we are merely painting an oil picture with the
superfluous introduction of wax and resin.
Pliny says the picture is to be painted with the
brush with wax melted over the fire, and speaks of
no other admixture. What we wish to know, is
whether this process is practicable in more temperate
climes. We have seen that in Egypt there would
be no difficulty. I determined therefore to experi-
ment upon this point with the assistance of Mr.
Thomson, a young art student.
The beeswax mixed with the pigments were kept
melted in little pots resting on a hot plate. The
hot plate also served for mixing tints. Ordinary
artists' brushes were used, and the painting done on
wooden panels. The secret of success proved to be
64 GREEK AND ROMAN
the warming of the panel. This can be done either
by holding near the surface a lump of hot' metal,
such as a soldering bolt, and then painting on the
surface thus warmed, or better, by. warming the panel
over a gas flame or in front of a fire. The painting
must be executed rapidly and with certainty. There is
no difficulty in over-painting or in keeping the panel
just warm enough to enable the colour to be freely
laid on without fusing what has already hardened.
Probably no such warming of the panel is necessary
during the summer either in Greece or Rome.
Just when about to commence these experiments
I found that Mr. Burns had many years before
devoted a year to experimenting on painting with
melted wax. The powdered pigments mixed with
the wax were kept in front of a fire in bottles, and the
picture was painted on a canvas which was kept
sufficiently warm by placing it so that the back was
exposed to the warmth from the fire.
The following photographs of one of the experiments
made by Mr. Thomson and myself, and the beautiful
work by Mr. Burns,^ prove conclusively that there
is no difficulty in painting with melted wax with the
brush even in northern countries. The process does
not lend itself to glazing, but if the finished work is
polished with a piece of linen, it has all the appear-
ance of an oil picture, while either thin washes or
^ See Frontiapieoe.
PAIBTINO IN MELTED. WAX BY MH. THOMSON.
To face page 64.
METHODS OF PAINTING 65
impasto painting are at the command of the
artist.
The remains found at St. Medard throw further
light on this question. For a detailed description oi
them the reader is referred to Berger's Maltechnik
vol. i. page 211.
Among the contents of the vessels was found bees-
wax, and a mixture of beeswax and resin.^ Some
long thin metal rods, with a shallow spoon at one
end and a flattened portion at the other, have been
supposed by Herr Berger, with every probability, to
be specimens of the cauteria used at that time. There
was also a bronze box, measuring 12 cm. by 20 cm.
and 10 cm. deep, and divided into compartments
which are covered with silver lids, stamped full of
square holes, and thus forming a silver grid over the
top of the box.
Herr Berger suggests, with every probability of
truth, that this box was filled with glowing charcoal
to melt the wax colours and warm the cauteria. It
would, however, serve another purpose still better,
with its large hot surface covered with the silver grid,
namely, the warming of the panel at intervals during
the process of painting.
' The fatty acids found in one vessel by Chevrenl are doubtless,
due to the oxidation of the beeswax. Herr Berger's Mcdtechnik,_
vol. i. p. 268, and again the experiments by Georg Buohner,
Munich, p. 273, which explain Chevreul's results.
E
66 GREEK AND ROMAN
In conclusion, the significance of the resin mixed
with the wax does not seem to have been completely
grasped. By dissolving a dry resin in the wax we
raise the melting point, and make it much more
difficult to manipulate.
But by mixing it with the semifluid pine
balsam existing in quantities in the surrounding
forests, the painter of St. M^dard would lower
the melting point of the wax, and make it easier
to manipulate. The resulting medium would also
be more translucent, and ultimately harder than
the wax alone, though taking longer to become
firm.
Experiments I have tried with beeswax and Canada
balsam have clearly demonstrated the statements
made above. In the course of time the volatile
turpentine of this balsam would evaporate, and the
final result would be the mixture of wax and resin,
free from the original turpentine found at St.
Medard.
It seems a very natural development of the tech-
nique in more northern lands, where pine balsams
were plentiful, and where the high melting point of
the wax seriously increased the difficulties, to use
this mixture. The wax removes the stickiness which
the balsam would have if an attempt were made to
use it alone.
To sum up, the statements made by Pliny, the
METHODS OF PAINTING 67
discoveries of wax pictures at Hawara, the materials
found in the grave at St. Medard, and Herr Berger's
experiments and my own combine to give us a com-
plete picture of this old technical process about
which so much speculation has been indulged in.
The origin of the technique from two sources,
modelling in wax and tarring ships, is clearly shown
in Pliny's account of the process; the two resulting
methods of painting with the cauterium and the
brush are also correctly and clearly described by
Pliny, and the two methods are found side by side
in the pictures from Hawara.
Finally, the difficulties that have been supposed to
exist are completely removed when we allow for the
climate of the region where wax painting originated,
and when we find how easy it is to overcome the
difficulties by warming the panel. The introduc-
tion of pine balsam and the bronze box, with the
painting and modelling tools found at St. M6dard,
complete the picture, and show us a modification
due to northern climes of the pure and simple wax
medium. Large panels could of course only be
painted in warm weather, or by heating them with
a hot bolt or brazier. Wall surfaces could probably
be directly painted in a warm chmate, but in a
colder climate would require to be heated with a
brazier. Vitruvius describes the use of such a brazier
for heating walls which were being varnished with
68 GREEK AND ROMAN
beeswax. Varnishing is a simpler matter than
painting, and I am disposed to agree with Pliny that
the process was not suitable for wall painting in
Rome, though possibly in Egypt and in Greece.
We shall now proceed to consider the methods
used for painting on walls.
Before looking at the information to be obtained
from Pliny and Vitruvius, we must first shortly con-
sider the possibilities in the use of mediums for wall
painting which the information already collected
brings before us.
In the first place we have the possibility that the
pigments mixed with water were painted on to the
wet plaster. This is the process known as Buon
Fresco, and is probably the simplest and most
obvious form of wall painting.
Or before being laid on the wet plaster the pig-
ments might be mixed with gum, glue, or egg ; or
again, the plaster might be allowed to get dry, and
then the medium might be used to attach the
pigments to the dry surface.
In addition to these water mediums there is another
possible one, namely, milk. The caseine in milk
forms a very strong and durable cement with lime,
and consequently milk ^ is a very suitable medium
for wall painting.
Besides these water mediums the wall might be
, ^ Pliny, xxxvi. 55 ; xxxv. 56.
METHODS OF PAINTING 69
painted in encaustic, though this would evidently be
a troublesome process, and finally the wall surface,
painted by any of these methods, might be var-
nished with wax to preserve and polish the
surface.
There is evidently no reason why only one of these
methods should be used, or why more than one
method should not be combined with another, just
as we find Cennino Cennini directing work to be
painted in buon fresco, but also directing it to be
finally finished, when dry, with egg tempera.
A wall, for instance, might well be begun in buon
fresco, painted with a size or egg medium, and finally
varnished with wax.
There is no reason to suppose that only one of
these methods was selected and always used, and
therefore it is not surprising that authorities should
differ or that the results of analysis should vary.
The question, however, that has recently been
reopened by Herr Berger is whether the process
known as buon fresco was ever used, and whether
the process which is called stucco lustro was not
the universal process.
I have not included stucco lustro among the
possible processes which I have suggested above, but
we shall have to consider it in due course.
The principal authority whom we shall have to
consult in this matter is Vitruvius, but before
«
70 GREEK AND ROMAN
doing so we shall consider first the statements made
by Pliny which throw light on the subject. We
have already noted one of these, in which he says
definitely that encaustic painting is not suitable for
walls. In the opening of the same chapter, which
I here requote, he mentions certain pigments as not
suitable for painting on a wet surface.
'Ex omnibus coloribus cretulam amant udoque
inlini recusant purpurissum, Indicum, caeruleum,
Melinum, auripigmentum, Appianum, cerussa.' This
sentence is quite meaningless if it is intended to
apply to water only, but if it means a wet surface
of lime it becomes intelligible, as many of these
pigments would be destroyed by wet lime. The
reference to wall painting in the next sentence makes
this meaning all the more probable.
But this chapter must be considered in conjunc-
tion with another chapter, xxxiii. 56, where he is
describing different varieties of ochre. One variety
he calls marmorosum sil, and then goes on to say,
' Hoc autem et Attico ad lumina utuntur, ad abacos
non nisi marmoroso, quoniam marmor in eo resistit
amaritudini calcis.' 'This and the Attic sort they
use for high lights ; for panelled spaces none but the
marmorean kind, because the marble in it resists the
acridity of the lime.'
If this mention of a special ochre which resists
lime be read along with the mention of the special
METHODS OF PAINTING 71
colours which do not resist a wet surface, I think the
combined evidence shows clearly that he is speaking
in both cases of the action of wet lime, and therefore
is familiar with the process of painting on wet
plaster. There is, however, another interesting point
to be noticed in the last quotation. One of the
arguments against the use of buon fresco in classical
times is the large area covered at one time by many
of their wall paintings, as it is held that buon fresco
involves the treatment of limited areas at a time,
and therefore joins should be visible. We shall have
to consider these matters presently at greater length;
but in this chapter Pliny distinctly suggests, by the
use of the word abacus, that limited spaces or panels
only are painted on the wet plaster, and that con-
sequently a pigment which could resist wet lime was
selected for the painting of such limited areas. It
may well have been that, while this wet painting was
used for important decorative pictures, the cheaper
decorative colouring round the margins was painted
on dry plaster in a less durable manner.^ We shall
find some remarks of Vitruvius bearing on this point.
In the meantime the evidence is clear and un-
mistakable from Pliny that he was familiar with the'
operation of painting on wet plaster. This does not,
however, exclude the possibility of some medium
1 Wiegmann holds this view : Die Malerei der Alten.
72 GREEK AND ROMAN
like size being mixed with the pigments laid on this
wet surface.
We shall next consider the information to be
obtained from Vitruvius.
In the seventh book of his work on Architecture,
after describing the making of concrete floors and the
preparation of lime, and the plastering of arches and
cornices, he proceeds as follows, in the middle of the
8rd chapter : —
' Coronis explicatis parietes quam asperrime trul^
lissentur, postea autem supra, trullissatione subares-
cente, deformentur directiones harenati, uti longi-
tudines ad regulam et ad lineam, altitudines ad
perpendiculum, anguli ad normam respondentes exi-
gantur. Namque sic emendata tectoriorum in picturis
erit species. Subarescente, iterum et tertio indu-
catur. Ita cum fundatior erit ex harenato directura,
eo firmior erit ad vetustatem soliditas tectorii. Cum
ab harena praeter trullissationem non minus tribus
Coriis fuerit deformatum, tunc e marmore graneo
directiones sunt subigendae, dum ita materies tem-
peretur uti cum subigatur non haereat ad rutrum,
sed purum ferrum e mortario liberetur. Graneo
inducto et inarescente, alterum corium mediocre
dirigatur. Id cum subactum fuerit et bene fricatum,
subtilius inducatur. I^ cum tribus coriis harenae
et item marmoris solidati parietes fuerint, neque
rimas neque aliud vitium in se recipere poterunt.
METHODS OE PAINTING 73
Sed et liaculorum subactionibus fundata soliditate
marmorisque candore firmo leYigata, coloribus cum
.politionibus inductis nitidos expriment splendores.
Colores autem, udo tectorio cum diligenter sunt
■inducti, ideo non remittunt sed sunt perpetuo per-
manentes, quod calx in fornacibus excocto liquore
facta raritatibus evanida, ieiunitate coacta corripit
in se quae res forte contigerunt, mixtionibusque ex
aliis potestatibus conlatis seminibus seu principiis
una solidescendo, in quibuscumque membris est
formata cum fit arida redigitur uti sui generis pro-
prias videatur habere qualitates. Itaque tectoria
quae recte sunt facta neque vetustatibus fiunt hor-
rida, neque cum extergentur remittunt colores, nisi
si parum diligenter et in arido fuerint inducti. Cum
ergo ita in parietibus tectoria facta fuerint uti supra
scriptum est, et firmitatem et splendorem et ad
vetustatem permanentem virtutem poterunt habere.
Cum vero unum corium harenae et unum minuti
marmoris erit inductum, tenuitas eius minus valendp
faciliter rumpitur nee splendorem politionibus propter
imbecillitatem crassitudinis proprium obtinebit.
Quemadmodum enim speculum argenteum tenui
lamella ductum incertas et sine viribus habet remis-
siones splendoris, quod autem e solida temperatura
fuerit factum, recipiens in se firmis viribus poiitionem
fulgentes in aspectu certasque considerantibus
imagines reddit, sic tectoria quae ex tenui sunt
74 GREEK AND ROMAN
ducta materia non modo sunt rimosa, sed etiam
celeriter evanescunt, quae autem fundata harena-
tionis et marmoris soliditate sunt crassitudine spissa,
cum sunt politionibus crebris subacta, non modo
sunt nitentia, sed etiam imagines expressas aspici-
entibus ex eo opere remittunt. Graecorum vero
tectores non solum his rationibus utendo faciunt
opera firma, sed etiam mortario conlocato, calce et
harena ibi confusa, decuria hominum indueta, ligneis
vectibus pisunt materiam, et ita ad certamen subacta
tunc utuntur. Itaque Yeteribus parietibus nonnulli
crustas excidentes pro abacis utuntur, ipsaque tectoria
abacorum et speculorum divisionibus circa se pro-
minentes tiabent expressiones.'
There are so many difficult and doubtful points
involved in the translation of this chapter that I now
give a rendering into English which has been done
with the greatest possible care, and after consulta-
tion with more than one classical scholar. This will
enable us to discuss more easily the difficult passages.
' When the cornices are finished, the walls are to
be trowelled as roughly as possible, and thereafter,
when the trowelling is somewhat dry, over it the
directions of the sand-mortar are to be so traced out,
that in length it must be true by the rule, in height
by the plumb-line, and the angles by the square.
For thus the surface of the plaster will be faultless
for pictures. When this (first coat) is slightly dry,
METHODS OF PAINTING 75
a second is to be laid on, and then a third. The
firmer and sounder the laying on of the sand-
mortar, the more solid and durable will the plaster-
work be. When besides the trowelling not less than
three coats of sand have been set out, applications
of marble-dust ^ are to be used. This stuff is to be so
tempered that in the spreading it does not stick to
the trowel, but the iron comes out of the mortar
clean. A coat of marble-dust^ having been laid on
and getting dry, another rather thin coat is to be
applied. When this has been beaten and well
rubbed, another still finer is to be put on. Thus
with three coats of sand and as many of marble,
the walls are so firm that they cannot crack or be-
come defective in any way. And moreover, solidity
being secured by rubbing with planes, and smooth-
ness from the hardness and sheen of the marble,
the walls will give out with great brilliance colours
applied with polishings. For colours, when they are
carefully laid on damp plaster, do not get loose, but
are for ever permanent, for this reason, that the lime,
losing all its moisture in the kiln, is so dry and
porous that it readily imbibes whatever chances to
touch it, and solidification taking place from the
mixtures of the various potentialities whose elements
or first principles are brought together, the result-
ing substance, of whatever it is composed, when it
' i.e. marble-dust mortar.
76 GREEK AND ROMAN
becomes dry, is such, that it seems to have special
qualities peculiar to itself. Thus plaster-work which
is well executed neither becomes rough from age nor
when it is washed does it give up the colours unless
they have been laid on carelessly and on a dry surface.
If, therefore, plaster- work on walls is carried out as
above described, it will be firm, lustrous, and very
durable. But when only one coat of sand and
one of marble-dust are used, its thinness renders
it liable to be easily broken, nor can it take on a
proper brilliance from the polishings owing to its
lack of substance. For just as a silver mirror when
jnade from a thin plate gives back a wavering and
uncertain image, but if made from a plate of solid
temper takes on a high polish and reflects to the
spectators bright and faultless images, so plastering,
when its substance is thin, is not only full of cracks
but also quickly decays, while thatwhich is firmly coni-
pacted of sand-mortar and marble, when it has been
rubbed with many polishings, is not only glistening
but also clearly reflects to the spectators the images
falling on it. Greek pla:sterers, indeed, use not only
the above methods to make their work firm, but also
putting the lime and sand together in a mortar, they
have it thoroughly pounded with wooden staves by a
number of men, and use it after it is so prepared.
Hence from their old walls people cut out slabs and
use them as panels, and those plaster slabs so cut out
METHODS OF PAINTING 77
for panels and mirrors have fillets in relief round
them.'
Before considering this chapter as a whole, there
are one or two points in the text and translation to
be considered.
In the earlier edition by Rose, in speaking of the
finishing processes of plastering, the word baculorum
is used, which in the later edition is replaced by
liaculorum, that is ' smoothing planes ' instead of
'beaters.' These smoothing planes would probably
not differ from the modern plasterer's trowel.
Though beating the surface might be part of the
earlier stages of preparation, smoothing planes
would obviously be necessary for the final process.
In line 3, p. 73, the words ' cum politionibus ' have
been rendered ' with polishings.' This translation is
justified by the repeated use of this word in this
chapter and elsewhere by Vitruvius in a sense which
can only be translated by the word polishings, and by
its agreeing with the whole description of the nature
of the process used. Herr Degering suggests that here
the word refers to the material used in the mortar.
The view adopted by Herr Reber ^ is that the correct
translation is the colour laid on at the same time
with the plaster. On the other hand Rode and
Conner interpret the passage as we do. It is difficult
' Reber, Des Vitruvius Zehn Biicher iiber Archilektur, Stuttgart,
1865.
78 GREEK AND ROMAN
to reconcile Reber's translation with the fact that in
the very next line Vitruvius directs 'the colour to be
laid on the plaster. It is true that Vitruvius some-
times uses the verb 'polire' in the sense of 'to
plaster.' But when we take into account the use of
the substantive in its plural form here, its undoubted
signification elsewhere in the chapter, and the state-
ment made in the next line that the pigments were
laid on the wet plaster, we are, I think, justified in
our translation.
Herr Berger suggests as an alternative translation
that in this passage ' coloribus cum politionibus in-
ductis ' should be translated the ' colours mixed with
polishing substances.'
In line 4, p. 73, we find the words ' udo tectorio.'
In the four MSS. mentioned in the note,^ this
word is not udo but nudo, and the emendation was
made by Jocundus in the sixteenth century. As
this sentence contains the most direct and important
statement to be found in either Pliny or Vitruvius
in favour of the use in their time of buon fresco,
it is curious that the opponents of buon fresco have
not made more use of this fact that udo is an
emendation on nudo. If, however, the whole passage
1 Harleian MS. Brit. Mus., 2767 (about ninth century). Sclet-
staiensis, 1153 (tenth century). Wolfenbuttelensis Oudianus, 69
(eleventh century). Epitomati Vitruvii, Wolfenb. Oudian., 132
(tenth century). These four are regarded by Rose as the most
reliable authority for the original text.
METHODS OF PAINTING 79
is read, and especially the remarks lower down, con-
trasting the dry with the wet method, this correc-
tion is justified. Possibly the copiers of the ninth
and eleventh centuries, unfamiliar with buon fresco,
put in the n to make sense of the passage, while
Jocundus, familiar in his time with buon fresco,
did not hesitate to strike it out again.
Herr Berger also proposes to translate ' colores
autem udo tectorio cum diligenter sunt inducti,' ' when
the colours are laid on with the wet plaster ' (' udo tec-
torio als Ablativ,nicht als Dativ gefasst ') ' wenn sie init
dem feuchtenTectorium aufgetragen sind,' p. 94. This
is quite inadmissible. Whatever doubt there may be
about the translation of the line above, the only
possible translation of this passage is the colour laid,
or smeared, on the wet plaster.
In line 10, p. 73, the phrase ' in quibuscumque mem-
bris est formata,' gives rise to considerable difficulty.
If it had been 'quibuscumque membris' the translation
we have given would be justified. The Rev. G. C.
Richards suggests, ' whatever be the shapings of the
divisions of the plaster,' and Herr Degering that it
refers to markings. Reber translates it as we have. I
confess I am unable to understand either of these
suggestions, or to make sense of them when intro-
duced into the passage. We are here dealing with
the chemical union of the lime, water, and pigments
into a homogeneous whole on the smooth plaster
80 GEEEK AND ROMAN
surface. Another translation which has been sug-
gested is, ' in whatever shape or form.' The phrase,
however, is not necessary to the meaning of the
passage as a whole.
We shall now consider the information to be de-
rived from this passage.
In the first place, the instructions for preparing
the plaster surface are perfectly clear and definite.
In the second place, whatever doubt there may be
about the translation of line 3, line 4 can only be
translated as meaning that the pigments are to be
laid on the wet lime.
In the passages after line 4 there is a most interest-
ing attempt to explain the way in which the lima
and pigments ultimately form a homogeneous whole.
If, instead of speaking of the lime losing its moisture,
Vitruvius had said losing its carbonic acid, the pas-
sage might with this emendation have been written
by a modern chemist describing the scientific basis
of buon fresco. I do not understand why this most
interesting passage has been condemned as obscure.
In the next place, it is to be noted that Vitruvius does
not speak of this as the only method of wall painting,
but as the most cki/rable method, and contrasts it with
the results obtained by painting on a dry surface.
We have here definite evidence that painting on
dry walls was also customary, in which case some
medium like glue would doubtless be used, and this
METHODS OF PAINTING 81
goes a long way to explain the conflicting conclusions
of investigators and chemists.
Vitruvius in effect tells us that he is familiar with
painting on dry walls, necessarily with some binding
medium, and with painting on wet lime, and he
regards the wet lime painting as the more permanent.
He would not have come to this conclusion without
a wide experience of both methods.
We have next to consider whether, in painting on
wet plaster, any medium such as glue was intro-
duced. No such medium is mentioned by Vitruvius,
and we are therefore bound to conclude that it was
absent, until chemical analysis or carefully conducted
experiments prove the contrary. There is no neces-
sity for its introduction in buon fresco. As I have
already pointed out, no conclusive evidence is to be
derived on this point from Pliny.
I have already discussed the references in Pliny and
Vitruvius to the mixing of glue with black, and have
shown that they do not prove the use of glue as a
medium, as they are capable of quite a different and
equally plausible explanation.
Herr Berger suggests that the whole of this
passage applies to the preparation of coloured grounds
by mixing pigment with the last layer of plaster,
and has nothing to do with the decoration by paint-
ing on the wall, which might be carried out on the
top of this coloured ground. He claims to have
82 GREEK AND ROMAN
found such coloured grounds in Pompeii, and states
that the thickness amounts to '01 mm. (page 96).
So thin a coat as this can hardly be described as a
final coat of plaster and pigment laid on with a
trowel. But Chevreul seems to have found cases of
pigment mixed with the last layer of plaster.' There
is no reason why, if they wished, the decorators should
not prepare such coloured grounds in certain cases,
and it is quite probable they did so. In reply,
however, to the statement that this description
applies only to the preparation of such coloured
grounds, I have to say that, in the first place, it is
impossible, without malting a distorted translation,
to translate ' udo tectorio cum diligenter sunt in-
ducti ' in any other way than laid on the wet plaster.
In the second place, Yitruvius is in this book dis-
coursing of wall painting, and gives no other method
than this for carrying it out.
It is true that there is an obvious hiatus in the MSS.
about the 6th chapter of this book, but this does not
justify the assumption that other methods of painting
on walls were here described and have been lost.
To sum up, then, the information which we have so
far derived from Pliny and Vitruvius, it is clearly
demonstrated that both painting on dry plaster, with
a suitable medium, and painting on wet plaster in
buon fresco was practised.
' Hittorf, I' Architecture Polychrome.
METHODS OF PAINTING 83
A further study of this chapter shows clearly that
this method of fresco painting was very different from
the method used in the time of the Renaissance, or
to-day, as there are frequent references to the polish-
ing of the surface during the process. Herr Berger
was, I believe, the first to point out the significance
of these polishing processes. In order to understand
the meaning of these passages, we shall have to
consider more clearly the nature of buon fresco itself.
When the pigment is flooded over the wet surface
of the plaster, the particles settle into the hollows of
the surface, bathed in a solution of lime. As this
solution of lime becomes carbonated and precipitated
by the carbonic acid of the air, the particles of
pigment are packed round with the precipitated
carbonate, so that the holding of the pigment to the
plaster is more of the nature of a mechanical than a
chemical process.
Moreover, lime is so slightly soluble in water, and
the carbonating of the lime is so slow a process, that
each time the surface is flooded with water, fresh
unaltered lime is dissolved and brought to the surface
of the plaster, for many days.
There is no need, therefore, for the immediate
painting of a surface as soon as the last layer of
plaster is put on, though, on the other hand, it is as
well that the plaster should be kept damp, in order
to keep a soft bed beneath the pigments, into which
84 GREEK AND ROMAN
they can become more or less incorporated. If this
is done there is no reason why a large surface should
not receive its final coat of plaster and then be
painted on in a leisurely manner, as long as, by
means of damp cloths or occasional sprinkling over
with water, it is prevented from getting too hard. The
particular technique, therefore, adopted by the
Renaissance painters, a small portion at a time re-
ceiving its final coat of plaster and then being painted
on, is not of the essence of the buon fresco process.
The Roman plaster was not only very thick, but
the numerous coats were to be put on before the last
coats were completely dry.
Such a mass would hold the contained water for
some time, and could easily be kept damp if necessary,
while the painting could be proceeded with in a
leisurely manner.
It is next necessary to consider carefully the state-
ments made in Vitruvius about polished surfaces.
As he does not direct the addition of any foreign
substances, we must first try whether such a polished
surface can be produced by the methods he describes.
In order to test this I had a series of shallow wooden
trays made, into which I introduced first a layer of lime
and sand, and then laid on this, when partially dry,
two layers of marble dust and lime, in the proportion
of two of marble dust to one of lime.
While the final coat was quite wet it was subjected
METHODS OF PAINTING 85
to the process familiar to plasterers of 'closing in.'
That is to say, it was worked repeatedly on the surface
with the long, straight, slightly rounded steel edge of
the plasterer's trowel, the trowel being held at an angle
to the surface of about forty-five degrees. This closed
in surface, although inside a building, took several
days to dry. As it got drier and firmer the working
of the surface with the rounded edge of the trowel
was repeated, by drawing it across with quick firm
strokes. Ultimately we obtained a dry, hard, compact
surface with the appearance of polished marble.
This satisfied me that, in so far as uniform plaster
surfaces are concerned, the polished surface described
by Vitruvius can be got without the introduction
of any material beyond lime, marble dust, and water,
although great technical skill is doubtless wanted,
and great expenditure of time and patience to
produce a satisfactory result. It is, however, evident
from his account that such polished plaster surfaces
were highly prized, and were not the work of the every-
day plasterer.
The next experiments were made with pigments.
I had had the opportunity of examining some
portions of Koman fresco obtained some years ago
by a friend from the Palatine. One of these was
coated with vermilion,^ another with red oxide of
' The plaster immediately below the vermilion was stained
yellow, as if some wax and oil had been used. This appearance
was absent in the other examples.
88 GEEEK AND ROMAN
iron, and a third with the blue Egyptian frit. In
the case of the vermilion, and to a great extent in
the ease of the oxide of iron, the coating of pigment
appeared homogeneous under the microscope. But
in the case of the coarse particles of the copper frit,
which had to some extent weathered off, it was evident
that the particles were imbedded in the plaster
and flush with the particles of marble dust. In order
to try to reproduce this appearance another panel
of plaster was prepared and closed in. It was then
allowed to dry for a day before being painted on, and
then it was painted with a thin coat of cobalt blue, in
one part, — laid on so as to show the brush-marks and
different depths of work, — with a thick uniform coat
of vermilion in another part, and with yellow ochre
in a third part. It was then left for another twenty-
four hours, and the whole surface then pressed firmly
with the flat of the plasterer's trowel. This could be
done without any disturbance to the painted surface,
but with an evident improvement in the vividness of
the colouring. On examining under the microscope
the whole surface appeared uniform, and the pigments
flush with the plaster, the edge of the vermilion and
the edge of the plaster being in focus at the same
time, and the particles of cobalt blue imbedded
among and flush with the particles of marble dust.
(This pigment was selected because it is of a com-
paratively coarse grain, and corresponds most nearly
MIOKOPHOTOSRAPH IN THREE COLOURS
OF POETION OF FRESCO, SHOWING
EGfTPTIAN FRIT IMBEDDED INTO
LIME AND FRAGMENTS OF
MARBLE DUST.
To face page S6.
METHODS OF PAINTING 87
therefore to the old Egyptian blue.) The surface was
allowed to dry further, and then the attempt was
made to polish it with the edge of the trowel. This
was only partially successful ; in some places a polish
was obtained, in others the pigment was disturbed,
owing to a want of perfect smoothness in the edge of
the trowel. Such a process of polishing does not
seem, however, to be impossible, even in the case of a
painted surface, if the right tools were devised and
sufficient practice attained. But even the first
stage of the process produces a smooth surface
with some degree of shine about it, and compacts
the whole mass together. When dry it can be
washed with water or rubbed up with beeswax and
turpentine.
If, then, we are to accept the account given by
Vitruvius, the Roman method of buon fresco differed
very widely from the one at present practised,
involving, in the first place, the ' closing in ' of the
plaster, and in the second place the partial or com-
plete smoothing and polishing of the surface after
the colour had been laid on. This is probably the
most durable method of painting ever devised.
While this polishing process might be facilitated by
the introduction of glue,^ or, judging by the results
1 Wiegmann, as a result of his experiments, came to the conclu-
sion that some size was present. Die. Malerei der Alien, Hanover,
1836.
88 GREEK AND ROMAN
of Herr Berger, still more by the introduction of
soap, it is not necessary to introduce any such sub-
stances in order to carry out successfully the process
as described by Vitruvius, and the durability would be
diminished. It is also evident, that even if it were
found preferable to plaster only a small portion at
a time, as is suggested by Pliny, such divisions would
disappear in the subsequent treatment, though they
might, as was found by Donner,i reappear in course
of time as cracks running along the lines of the
former joins.
We shall next consider the special treatment to
which the plaster surface was subjected in order to
protect vermilion, according to the statements of
Pliny and Vitruvius.
In the 40th chapter of the thirty-third book, speaking
of native vermilion, Pliny says ' Inlito solis atque lunae
contactus inimicus. Remedium, ut parieti siccato cera
Punica cum oleo liquefacta candens setis inducatur
iterumque admotis gallas carbonibus inuratur ad
sudorem usque, postea candelis subigatur ac deinde
linteis puris, sicut et marmora nitescunt.' ' When
laid on, the exposure to sun and moon is harmful.
The remedy is: when the wall is dry spread on it
with a brush melted Punic wax mixed with oil and
glowing hot, and again, heat it to sweating point by
^ Die erhcdtenen antihen WandmaUreien in techniscJter Beziehung,
1869.
METHODS OF PAINTING 89
placing charred gall-apples near it, afterwards rub
it with candles and then with clean linen cloths
as marble is made to shine.' And Vitruvius, also
speaking of vermilion, says in the 9 th chapter
of the seventh book, 'apertis vero id est peristylis
aut exhedris aut ceteris eiusdem modi locis, quo sol
et luna possit splendores et radios inmittere, cum ab
his locus tangitur, vitiatur et amissa virtute coloris
denigratur. itaque cum et alii multi turn etiam
Faberius scriba cum in Aventino voluisset habere
domum eleganter expolitam, peristylis parietes omnes
induxit minio, qui post dies xxx facti sunt invenusto
varioque colore, itaque pro minio locavit inducendos
alios colores. At si qui subtilior fuerit et voluerit
expolitionem miniaceam suum colorem retinere, cum
paries expolitus et aridus fuerit, ceram ponticam igni
liquefactum paulo oleo temperatam saeta inducat,
deinde postea carbonibus in ferreo vase compositis
earn ceram a proximo cum pariete calefaciundo
sudare cogat, atque ut peraequetur deinde tunc
candela linteisque puris subigat, uti signa marmorea
nuda curantur. haec autem <ydva)<Ti<i graece dicitur.
ita obstans cerae ponticae lorica non patitur nee
lunae splendorem nee solis radios lambendo eripere
ex his politionibus colorem.' 'But in open places,
that is, in peristyles and loggias and the like, into
which sun and moon can dart their bright rays, the
[painted] part when touched by these is marred, and
90 GREEK AND ROMAN
the quality of its colour being destroyed it turns
black. Thus it was that when Faberius the notary
wished, like many others, to have his house on the
Aventine hill richly decorated, he covered all the
walls in the peristyles with vermilion. After a month
they became ugly and uneven, and accordingly he
bargained with the contractors to lay on other colours
instead of vermilion. But a more discerning person,
who wishes his vermilion decoration to keep its
colour, should, when the wall is well polished and
dry, lay on with a stiff brush Pontic wax melted in
the fire and tempered with a little oil, then bringing
an iron pan of glowing coals near to the wall, he must
heat both it and the wall and make the wax sweat,
and thereafter, to make the surface even, he must
rub it with a candle and clean linen cloths, as nude
marble statues are treated. This process is called
ydvcoai^ by the Greeks. The coat of Pontic wax
being in front does not allow the play of the sun's
rays or the sheen of the moon to take away the
colour from such decorations.'
In the first place, it is evident that this is a process
for varnishing a surface already painted, and not for
painting a surface. There is no inconsistency there-
fore, as some have held, in Pliny's saying in xxxv.
81 ' alieno parietibus genere,' where he is discussing
the use of wax as a medium for painting, and on the
other hand recommending it for varnishing an
METHODS OF PAINTING 91
already painted surface. Mastic varnish, for instance,
is quite suitable for varnishing pictures, but would
make a very inconvenient and unsatisfactory medium
to paint with. In the second place, both Vitruvius
and Pliny confine the use of this process to a special
purpose, namely, the protection of vermilion, and
only when exposed to direct sunlight. Vitruvius
states definitely that it is not necessary where ver-
milion is used for interior decoration.
This is a conclusive proof that this process was not
a universal one. If the process had not been men-
tioned at all, it might have been omitted by accident
and still have been used. But to mention a process
and at the same time confine it to a particular
purpose, shows quite clearly that it was not the
general method of treating all wall surfaces. It will
also be noted that Vitruvius says the varnish is to be
applied ' cum paries expolitus et aridus,' clearly
indicating that the decorative treatment with ver-
milion has been executed on the wet surface, which,
after drying, is then varnished with wax.
In Pliny's statement the wax mentioned is Punic
wax.
As has been already explained (p. 43), in the older
edition of Vitruvius by Kose, punicam is given as
the word, but in the four MSS. already referred to,
the word is written puTnicam (sometimes written
pomicam), which Rose has come to the conclusion is
92 GREEK AND ROMAN
a corruption of the word ponticam (the short t and
the n being perhaps mistaken for m by the tran-
scribers).
It will be remembered that Pliny, in his chapter on
beeswax, mentions both these waxes, and I there
pointed out that Herr Berger lays great stress on
punicann being the word used both by Pliny and by
VitruYius in this connection.
After the correction by Rose, it is at any rate
open to doubt whether punicaifn was the word used
by Vitruvius, and therefore it is unsafe to base any
argument on the presence of this word.
The use of the word candela has given rise to some
difficulty. Some have suggested that it means that a
lighted candle was used to warm the surface, others
that a roller shaped like a candle is meant. Candles
seem to have been known, made both of wax and
tallow, and therefore others have said that the
meaning is that the surface was finally rubbed with
a wax candle and a linen cloth. Another possible
view is that a tallow candle was used for the final
polishing.
It will be noticed also, on examining the context,
that it is not at all clear whether the statement
about the polishing of marble applies to the whole
process, or merely to the rubbing with candelae and
linen cloths. Either view would satisfy the trans-
lation.
METHODS OF PAINTING 93
I tried the experiment of polishing marble
with tallow, with wax melted with olive oil, and
then strongly heated after being applied, and with
a lump of solid wax alone and a linen cloth.
I failed to obtain a polish either by rubbing up
with tallow or rubbing up after treatment with hot
melted wax and oil. But I found that if the marble
was very lightly rubbed over with a lump of solid
beeswax, and then rubbed hard with a hard rough
linen cloth, a beautiful polish was at once obtained.
The only precaution necessary is to avoid putting on
too much wax when rubbing with the lump of bees-
wax. The layer is so thin that the marble is not in
the slightest discoloured, but gets at once a glossy
surface, which gives it depth and transhicency. The
marble used had already, of course, been smoothed
and polished as far as was possible by merely treat-
ing the surface of the stone itself
A piece of wax, already shaped as a candle,
would be very convenient for this purpose, being
readily held in the hand while the end would be
rubbed over the marble.
The rubbing with wax candles is thus completely
explained, and it is evident, that while in the case of
a porous plaster surface it is necessary to fiU up the
pores with hot wax to begin with, before polishing
with wax candles and linen, in the case of marble
this process is not necessary, the process called
94 GREEK AND ROMAN
ydvaai'; being the rubbing with wax candles and
linen alone.^
In conclusion, vermilion, when exposed to direct
sunlight, does change colour in the way described,
and is to some extent protected by being covered
with a glossy surface either by varnishing or in the
way described by Yitruvius.^
We shall next proceed to consider the evidence
supplied by chemical analysis. But before doing so
it is necessary to state clearly two propositions.
In the first place, the presence of organic matter is
not a proof that an organic medium like glue or egg
has been used. Such organic matter may be present
accidentally in the original material, may form part
of one of the pigments, as for instance a pigment
prepared from murex or madder, or may have
soaked into the buried plaster surface.
If we can extract the organic substance and
identify it as glue, gum, beeswax, or what not, that
is another matter altogether.
In the second place, if organic matter is absent,
and at the same time the pigment is firmly adhering
' Pliny uses the word ' nitesount,' therefore the process must
give a shine. No doubt armour, to which he refers, would be
polished the same way.
^ Tj Si jAviiKri.s ToS Ayd\/iaTos AvajKala • raxli yci.p l^avBS rb hIXtivov (fi
Th TraXaio, tCiv dyaXfidrui' IxP'^t""- 'The " Ganosis " of the statue
is necessary, for the vermilion with which the ancient statues are
painted soon loses its colour.'— Plutarch, Quaest. Horn., 287 D.
METHODS OF PAINTING 95
to tlie plaster, this may be taken as proof positive
that the use of an organic medium was not of the
essence of the original painting process.
For it is inconceivable that if the binding material
has been removed by the action of fire, or of water,
the pigment would still remain firmly adhering to
the plaster, though of course it might be found lying
on the plaster in a loose powdery condition.
Therefore the presence of a firm adhering coat
with no organic matter, or merely traces of organic
matter, may be taken as a proof that buon fresco
was the process used. Herr Berger's experiments
with portions of fresco painted with wax colours and
then treated with hot ashes are quite inconclusive,
as after the baking he did not analyse the pigments
to see if the wax, all or in part, was not still there,
though driven in by the heat (p. 148).i Such hot
ashes might suck up some, but not all the wax, and
some would still be present.
Those who believe the Koman frescoes at Pompeii
to have been painted with wax are on the horns of a
dilemma.
Sir Humphry Davy, John, and Chevreul, all
agree that wax is absent. If this absence is due
to the action of hot ashes, then how is it that the
pigments adhere ?
How can any one explain the miraculous process
^ The teat with acid for effervescence is quite inconclusive.
96 GREEK AND ROMAN
by which the binding material has been removed
and the pigments left firmly adhering ?
A very interesting statement by Presuhn is noted
by Herr Berger in this connection (p. 149). He finds
that monochrome washes on walls in Pompeii dust
off, but in wall pictures the pigments are firmly
adhering. This clearly points to the use, for cheap
colour work, of an organic material, such as glue, on
the dry plaster, which has dissolved and therefore
left the pigment loosely adhering, while the more
important frescoes were executed in the more reliable
method of buon fresco as directed by Vitruvius.
The determination of the actual organic sub-
stances is a difl&cult matter; the amount of bind-
ing medium necessary to fix a pigment to a surface
is very small, and as a rule chemists can only get
small fragments to examine, as they cannot be
allowed to destroy large surfaces ; and such organic
substances, even if they remain in situ, change their
character in course of time from the chemical actions
of moisture and oxygen, and therefore it becomes
more and more difficult to decide definitely what
they were originally composed of.
The earliest experiments were made by Sir
Humphry Davy and Chaptal. Chaptal's experi-
ments were, however, confined to pots of pigment
found in Pompeii, and therefore throw no light on
the question of the employment of a medium, as it
METHODS 0^ PAlNT'ING 97
may well be that the medium was to be subsequently-
added. Sir Humphry Davy, however, examined
actual pieces of fresco, and as it has been stated by
Herr Berger that he only examined minute frag-
ments taken from the Aldobrandini wedding, I pro-
ceed to quote the whole of the sentence from his
paper dealing with this matter.
'I have examined several pieces of the painted
stucco found in the different ruins, and likewise
the Aldobrandini picture, with a view of ascertaining
if any application had been made to fix the colour,
but neither by the test of alcohol, nor by heat, nor
by the action of water, could I detect the presence of
any wax varnish, or animal or vegetable gluten.'
(Phil. Trans., vol. 105, p. 119.) He then suggests
that the lightness of vegetable blacks might require
the use of glue in their case, as stated by Pliny and
Vitruvius, though not for heavier pigments. This
is very much the explanation . I have already
suggested for these passages.
It is to be regretted that Sir Humphry Davy does
not enter into greater detail, but while the identifica-
tion of a particular organic medium and the deter-
mination of the amount of it present is difficult,
the absence of organic substances is not difficult
to settle, even though very small fragments are
available.
This statement of Sir Humphry Davy's may
98 GREEK AND HOMAN
be taken, therefore, as conclusive evidence that no
binding medium was present in the samples
examined by him. As, however, he does not
describe the condition of these samples, this is not
conclusive evidence that a binding medium had not
once been present.
J. F. John (Die Malerei der Alien, Berlin, 1836,
p. 155), in examining a piece of red tectorium from
Pompeii, found no binding material or organic matter
beyond a trace of fatty matter extracted by ether.
Ether, unless very carefully purified and rectified,
always leaves, on evaporation, a slight greasy residue,
and therefore, considering the early date of John's
researches, I think it quite possible that the fatty
matter was contained in the ether and not in the
pigment.
At any rate, such a trace of fatty matter is quite
rightly neglected by the chemist.
We now come to Chevreul's results.^
(1) In a red-coloured plaster, which was polished
hard and shining, distillation yielded ammonia, ' show-
ing a very appreciable quantity of an organic sub-
stance to be present,' evidently of a nitrogenous
nature.
A water solution residue also yielded an ammoniacal
distillation, and alcohol extracted a trace of fat.
1 Memoirs of the Academy of Science, xxii. 1880. Reprinted in
Hittorf'e I' Architecture Polychrome, p. 912.
METHODS OF PAINTING 99
The greater part of the organic substance was not
dissolved by the water or the alcohol.
It is evident that here an organic substance is
present, either forming a constituent of the pigment,
or forming a binding medium, as Chevreul comes
to the conclusion later on that there is more of it
present thaii can be accounted for by accidental
impurities.
The experiments would agree either with the
presence of egg or glue or milk, all nitrogenous
bodies yielding ammonia, and all containing oils or
fatty matter. Such fatty matter is present as an
impurity even in the glues of to-day, and would be
sure to be present in the glues made in ancient
times, when the methods for completely separating
fatty, matter were not fully understood.
It is also possible that an organic pigment or
lake had been mixed with the red oxide of iron,
which was found to be present.
(2) The next piece examined was a black colour,
and had been painted over a pink plaster, of 1 mm.
thickness.
It proved to be an organic black, and yielded
ammonia, and resinous bodies were extracted from
it by alcohol.
The presence of such resinous bodies was to be
expected in an impure vegetable black. The presence
of ammonia suggests that some animal black had
100 GREEK AND ROMAN
also been mixed with it, or that the ammonia was
due to the glue, which we know was introduced
into these blacks as a matter of course.
This analysis, therefore, cannot be held to throw
light on the question as to whether a medium wa$
present or not.
(3) was coated with yellow ochre. Boiling water
separated an organic substance, and boiling alcohol
a ' wax or resin like matter.' It seems improbable
that any organic pigment would be present in this
case.
(4) This was a fragment containing various colours
on white plaster.
The white plaster contained an organic material
of a nitrogenous character, partially soluble in water,
and a trace of fatty substance soluble in alcohol.
The mortar underneath contained very little or-
ganic matter.
(5) In this fragment the red ground was over-
painted with yellow, and the yellow with white and
brown lines.
The red ground contained organic matter, but the
yellow colour mere traces of organic matter.
This is a very remarkable case, as at first sight
the painting over of one colour by another seems
almost to necessitate a binding medium, and yet if
it had been present it had disappeared from the
over-painting. We have no information in this case
METHODS OF PAINTING lOl
of the condition of attachment of the surface pig-
ments. It is not impossible, I believe, to execute
over-painting in buon fresco if carried out in the
manner described by Vitruvius. The partially dried
and pressed under painting could quite well receive
an over-painting, which could then again be pressed
and polished. The final white over-painting softened
very much with water.
This completes the Pompeian fragments. The
next analysis was of a portion of Roman fresco
found in the Palais de Justice, Paris.
Only a trace of organic matter was found, but the
red plaster was firm, shining, and very compact. In
this case the essence of the process cannot have been
the introduction of an organic binding material.
Next we have the examination of two examples
from St. Medard des Pres. In this case there were
two or three over-paintings, yet alcohol only ex-
tracted a trace of fatty matter, and there were only
traces of ammonia set free, and in dissolving the
fragment in acid so as to decompose any soap present,
and then treating with alcohol, no fatty matter was
obtained. In this case, therefore, the evidence is in
favour of no medium having been used, and the
pigments were adhering firmly enough for one layer
to be removed above another, revealing those under-
neath. The probable explanation is that the painting
was done in the way suggested below.
102 GREEK AND ROMAN
The second fragment from St. Medard had over-
paintings, but was also free from organic binding
material! In the case, therefore, of the fragments from
Paris and St. Medard, Chevreul comes to the con-
clusion that no organic binding material was used,
and possibly milk of lime was used as the medium.
This method of painting on a dry wall with milk of
lime is described by Theophilus.
In the case of this second fragment from St.
Medard, the layers of pigment could be lifted off
one from another, and each layer contained carbonate
of lime, which may well have been originally milk
of lime.
In the Pompeian fragments the quantity of organic
matter Chevreul concludes is too great for us to
suppose its presence accidental.
To sum up, in all Chevreul examined nine samples.
Of these four contained only traces of organic
matter, three contained nitrogenous organic matter
and traces of fatty matter, and two contained nitro-
genous organic matter and resinous or waxy bodies.
One of these two would necessarily contain resinous
bodies, so that of these one only contains unexplained
resinous or waxy bodies. The suggestion that the
organic medium had disappeared through time will
not account for its absence in the samples from
Paris and St. Medard, certainly not in the case of
the Paris sample, where the firm hard nature of the
METHODS OF PAINTING 103
coating is specially mentioned, as the removal of
the medium would destroy the adherence of the
coating.
In the St. Medard samples, the layers of colour
are found one on the top of the other, as if laid on
a dry surface, and can be split off from each other.
This would hardly be the case if each coat had
been pressed home after being painted on the still
damp plaster ; and at the same time the description
of these layers of pigment, which can be split off
from each other, is not consistent with the presence
of a binding material which has perished. The
presence of carbonate of lime in each layer explains
the whole matter as Chevreul points out. In this
case the layer of pigment was mixed with milk of
lime as a medium.
From what I have said before of the probability
that organic binding mediums were sometimes used,
it is evident that there is no inconsistency between
the results obtained by Chevreul and Sir Humphry
Davy.
We can draw the following probable conclusions
from these results.
That many wall paintings were executed in
buon fresco without the use of organic binding
material, and that in some cases a modified buon
fresco was used in which the pigments laid on the
dry surface were mixed with milk of lime.
104 GREEK AND ROMAN
That in many cases an organic medium was used
which is present in the under plaster, and which was
of a nitrogenous character and contained traces of
fatty matter, and which might therefore be egg, milk,
or glue. If glue, it had become considerably modified,
or originally contained insoluble matter, as it was
only partially soluble in water.
That the fatty matter present had evidently not
become saponified by the enormous excess of lime
present.
That in one instance Chevreul searches for the
presence of soaps and finds none.
That there is no evidence to support the view that
wax, or any preparation of wax, was used except in
one doubtful instance, where a ' waxy or resinous '
substance is found.
That there is no evidence to show whether the
medium used in the Pompeian fragments mentioned
was utilised to paint on a dry surface, or was used to
paint on wet lime as a form of modified buon fresco.
The results obtained by the chemist Geiger
(Ohemisohe Untersuchungen, 1826) are unfortunately
open to doubt, as the fragments examined by him
had apparently been treated for preservation after
discovery, and therefore I do not propose to discuss
them.
Before concluding I have a few suggestions to
make about the painting of marble walls and statues in
METHODS OF PAINTING 105
Greece. The problem in this case is quite different
from that of painting on heavily plastered walls. The
colour is laid on a smooth marble surface, or a very
thin coat of plaster, and we have no information,
either in Pliny or Vitruvius, as to how this was carried
out. Faraday, Landerer, and Semper ^ have analysed
some of the Greek decorations and find in all cases
organic matter, and in most cases identify the
presence of beeswax. It is not, however, clear from
their results whether beeswax was the only bind-
ing material present. Having identified beeswax
they seem to have been satisfied and looked no
further.
I have experimented in the painting of smooth
marble surfaces with the following results.
In the first place, if the marble is warmed the
pigments can be applied with melted wax. This, as
has been already pointed out, would be a very incon-
venient process on large wall surfaces, unless the
wall was warmed to a suitable temperature by the
sun.
I have tried to paint surfaces with Herr Berger's
' Punic wax,' but have never succeeded in doing so,
the wax collecting in sticky particles, even with
the addition of soap as well as soda.
' Faraday's Analyses, Hittorf, p. 547. Landerer's Analyses,
Antiquitis ffelUniquea, by A. R. Ramsgate, i. p. 63. Semper';
Analyses, Hittorf, p. 489.
/tliziqmi/tis jicuen'iqu
Analyses, Hittorf, p,
sra
106 GREEK AND ROMAN
I have also tried the Mount Athos receipt. Size
alone emulsifies roughly with wax, but not in a very
satisfactory manner. The Mount Athos receipt gives
a perfectly workable medium, and the resulting surface
can, as stated in the receipt, be polished with a cloth.
There is another way, however, in which I have
obtained a better result.
As I have already pointed out, the statements as to
the polishing of marble Avith beeswax made by Pliny
and Vitruvius (Ganosis) may well refer simply to the
rubbing of the surface with wax candles and polishing
with a cloth, a process which is the best I have found
for polishing marble.
It occurred to me that this process was capable of
further application.
I therefore painted a piece of marble with a
pigment mixed with a little glue, allowed it to dry,
and then rubbed it lightly all over with a lump of
beeswax, taking care not to scratch off the pigment
until the whole surface had been passed over with
the beeswax, and then began rubbing with a linen
cloth, at first lightly, then with hard rubbing, and
finally lightly again. In this way a beautiful polish
was obtained, which was carried on to the bare
unpainted marble as well.
The result is to give the whole surface a uniform
shining coating, the pigment appearing like a dark
stone let into the marble, and the aesthetic effect being
METHODS OF PAINTING 107
therefore quite diiferent from any other method of
painting the marble which I have tried. Such a
pigment, if analysed, would show the presence of wax.
Further analyses are required of Greek paintings
on marble before the question, as to whether this was
the method, can be settled, but the process is very
simple, beautiful in its results, and quite in agreement
with the scanty information given by Pliny and
Vitruvius.
The information to be obtained about Egyptian
wall painting is very confusing and not satisfactory.
In samples I have examined the pigment is easily
sponged off, and therefore is probably put on with size
or gum. In other cases apparently the paintings
resist water, and probably wax has been used, possibly
in the way described above.
Herr Berger quotes John as saying that ' wax soap '
is present, but giving no demonstration of the evidence
for this impossible conclusion. Evidently further
research is required.
In conclusion, it is necessary to refer briefly to the
theory developed by Herr Berger in his Maltechnik
des AlterthuTns. There is a modern process known as
stucco lustro, used in Italy for making an artificial
marble, in which a plaster of marble dust and lime
is polished by treating it with an emulsion of lime
and olive oil soap, and polishing with hot iron. Herr
Berger holds that this was the method of wall paint-
108 GKEEK AND KOMAN
ing used in classical times, the medium being
an emulsion of wax in soda or potash, which he
calls 'Punic wax,' mixed with oil, which would be
saponified by the soda, and then applied to the
walls. He also believes other organic mediums were
used, along with this stucco lustro process.
He has developed this ingenious theory at great
length, and those who wish to study the evidence he
brings together must be referred to his book.
In the first place, the attempt to connect the
modern stucco lustro process with the ancient
technique through Pliny's description of the prepara-
tion of Punic wax breaks down at the outset, as it is
quite obvious that Pliny's description does not refer
to an emulsion of wax and soda. Yet the presence of
a watery mixture containing soda is necessary to Herr
Berger's theory, because he depends on the solution
of soda to saponify the subsequently added oil. In
fact the wax is superfluous and unnecessary.
In the second place, in the account given by Pliny
and Vitruvius of the application of wax mixed with
a little oil to walls, we are directed to melt the wax,
not to boil a watery emulsion.
In the third place, the process described by Pliny
and Vitruvius is meant for only limited application to
special cases.
In the fourth place, the results of chemical analysis
have proved the absence of wax from Roman frescoes.
METHODS OF PAINTING 109
Therefore the proof is complete that ' the wax soda
oil lime medium, suggested by Berger, is not men-
tioned by classical writers, and is disproved by
analysis. The most that can be said for his theory
is, that in those cases where an organic medium has
been found, the use as one constituent of the medium
of a lime olive oil soap without wax has not been
disproved, as it has not been looked for.
To sum up our conclusions, we have found that
when painting panel pictures the ancients either
painted with a medium which was probably egg, glue,
or gum (of these egg being the most likely), or with
wax, and when painting in encaustic, made use
either of the cauterium to model the wax surface, or
the brush using melted wax. That the difficulties
supposed to underlie this process have proved to be
imaginary, when climatic conditions are allowed for,
or the panel is artificially warmed.
That they were accustomed to use more than one
method of painting on walls. That the best method
of painting on plaster was considered to be buon
fresco, the pigments being laid on the wet lime,
though it is not absolutely clear that they did not
sometimes modify this by mixing the pigments laid
on the wet surface with glue, milk, or egg, this modi-
fication not affecting the essential nature of the
process.
That they in certain cases painted either on the
no GREEK AND ROMAN
wet or the dry surface with one of these three
mediums.
That they also sometimes painted on the dry sur-
face with milk of lime.
That an essential part of their best work was the
polishing during and after the preparation and
painting of the plaster.
That this polishing process is quite possible on
marble plaster without the addition of any organic
medium.
That such a treatment, even if only carried a little
way, increases the durability of the painting, and
gives a smooth surface capable of being washed or
varnished or polished with wax.
That in the case of vermilion painting exposed to
direct sunlight they varnished with wax.
That the process known as Ganosis among the
Greeks consisted, in all probability, in simply rubbing
the surface over with a beeswax candle and a linen
cloth.
That the easiest way to paint on marble is to lay
on the pigment mixed with glue, and then polish the
surface when dry with a lump of beeswax and a linen
cloth, and that this is the simplest possible explana-
tion of the presence of wax in Greek paintings on
marble walls and statues.
That in order to get further light on the extent to
which organic mediums were used in painting on
METHODS OF PAINTING 111
plaster, not only should a systematic analysis be made
of existing remains, but plaster painted in different
ways should then be subjected to analysis, in order
that Ave may have some indication of what sort of
results to expect.
And, finally, that the more faithfully we examine
the writings of these authors, without allowing our-
selves to be carried away by plausible hypotheses or
later inventions, the more we are convinced of the
completeness and accuracy of their statements.
Note. — While in making this inquiry I have as far
as possible avoided referring to any information of a
later date than Pliny, yet there is a passage in the
famous MS. in the library of the cathedral at Lucca,
of the eighth century, which bears so directly on later
practice that it is worth quoting : —
' Ita memoramus omnium operationes quae in parie-
tibus simplice in ligno cere commixtis coloribus in
peUibus ictiocollon commixtum.'
' Thus we mention operations with all of them, on
walls unmixed, on wood the colours being mixed with
wax, on skins fish-glue being mixed.'
This quotation I take from Mrs. Herringham
(Gennino Cennini, p. xxiv.), who copies it from a
photograph of the original, and which I find is
exactly the same in Muratori's work on Italian
Antiquities, vol. ii. p.. 377, except for an error in
112 GREEK AND ROMAN
spelling and a comma. It is the more necessary to
introduce this quotation as it is incorrectly quoted
by Herr Berger (Beitrdge, Part iii. p. 18), the 'in'
between ' simplice ' and ' ligno ' being omitted, and
the whole sense thus altered, ' cere ' being thus made
to apply to painting both on wood and walls, and
' simplice,' which means ' unmixed,' becoming mean-
ingless.
METHODS OF PAINTING 113
APPENDIX I
Professor Baldwin Brown supplied me some time ago
with a portion of an Egyptian mummy case, of about the
time of the nineteenth dynasty, which he had obtained from
Flinders Petrie.
The wood, which was much decayed, was covered with
a layer of gesso, on this was laid a black pigment, and on
the black ground a yellow pattern had been painted, the
whole being covered with a varnish of a warm colour.
In many places the varnish had powdered off, but in
other places it was intact. It was smooth, shining, with
straight cracks running through it, brittle, and broke up
under the needle point, with a conchoidal fracture. In fact
it had all the characteristics of a spirit varnish, either
artificial by dissolving a resin in a volatile medium, or
natural, that is, a liquid balsam or natural turpentine. I
have discussed this matter at some length in the text.
On treating with alcohol in the cold it readily dissolved,
showing it to be neither an oil varnish nor one of the
more insoluble resins, but rather behaving like a pine
balsam or some similar balsam. To identify it further
would require an examination of the balsams from Syria
mentioned by Pliny, such as terebinth and mastic, and
from other possible sources. The quantity present was
not sufficient to make anything but solubility tests. The
varnish had a distinct orange colour which was imparted
to the alcohol.
By treatment with alcohol the varnish could be removed
without affecting the painting underneath, which was
evidently executed with a medium insoluble in alcohol.
H
114 GREEK AND ROMAN
The black pigment proved to be coarsely powdered
charcoal, and the yellow pigment was a yellow ochre,
which had, however, not been floated according to the
modern practice, as it contained coarse white particles
apparently of quartz. The gesso was carbonate of lime.
On boiling with water the painting and gesso were
completely disintegrated, and the water solution, on filtra
tion and evaporation, left a residue of translucent brown
flakes, which on heating with soda lime gave oif quantities
of ammonia. The solution was precipitated by tannic
acid, but did not give the xanthoproteic reaction. It
was examined for me by Dr. Jerdan, of Cox's gelatine
and' glue works, and he came to the conclusion that it was
slightly altered glue, which, very probably owing to slow
oxidation, failed to give all the characteristic reactions..
This fact that Egyptian glue does not give all the char-
acteristic reactions of fresh-made glue has been noted by
other chemists.
The medium used to lay the pigments on the ground
could not be separately identified, as the painted surface
could not be removed in sufficient quantity from the glue
underneath. The question of most interest raisfed by this
fragment is the source of supply of the natural balsam
with which it has apparently been varnished.
The varnish itself was too glossy and transparent to
have been made by diluting a hard resin with wax.
APPENDIX II
The possibility of using natural pine balsams as painting
media is of consideriable interest.
The receipts for lakes given in the MSS. from the twelfth
to the sixteenth century are nearly all for their prepara-
tion from dye wood or from kermes, or cochineal, madder
being very rarely if ever referred to, .
METHODS' OF PAINTING 115
Sach lakes are very fugitive.
Moreover, the only really brilliant green in the receipts
is verdigris, which has the reputation of turning black
in oil.
It is therefore difficult to understand how the pigments
used for glazing by Van Eyck, for instance, have stood the
test of time so remarkably well. It is usually assumed that
oil, and still more oil varnishes, protect pigments from mois-
ture, and amber varnish has been treated with a special
superstitious reverence. Some experiments I made in
1891,1 and which are published in the Proceedings of the
Society of Arts, h'ave evidently escaped the attention of
recent continental writers. By using ignited sulphate of
copper with various mediums, and exposing the painted
surfaces when dry to moist air, I found that oil and oil
varnishes, including amber, are readily permeated by
moisture and the pigment attacked.
On the other hand spirit varnishes and natural pine
balsams protect the pigment from attack.
The question therefore arises whether the Van. Eyck
mediums were not more of the nature of a balsam than
an oil.
De Mayerne gives a receipt for making from verdigris
a green which will never change by dissolving the
verdigris in a pine balsam. This receipt gives a most
beautiful green, and it is the only green, either used as a
glazing colour or mixed with yellow, with which I have
been able to match the magnificent greens to be found in
Van Eyck's pictures, when confining myself to the palette
available for him.
The whole question has been carried further by the
Hon. Neville Lytton, who has been painting for some
time with a medium of Canada balsam and mastic
varnish, containing spike oil, and a little amber varnish
(copal oil varnish would do just as well) to give it tough-
ness and elasticity. He prefers to emulsify this medium
' Figments and Vehicles of the Old Masters. Cautor Lectures,
Society of Arts, 1891.
116 GREEK AND ROMAN
with water. Though it is not so easy to manipulate as an
oil medium, the surfaces painted with it are sufficiently
pliable, and tests which I have made with it show it to be
capable of protecting pigments from both moisture and
pernicious gases. White lead, for instance, in this medium
is not affected by sulphuretted hydrogen. Unfortunately
the possibility of using it with sufficient facility for fine paint-
ing has not been completely demonstrated, as Mr. Lytton
has been mixing it with pigments ground in oil, thus intro-
ducing an excessive amount of oil into the picture, instead
of grinding the dry pigment in the medium itself. A little
oil or oil varnish can be added with impunity, but the
amount must be small
In my own experiments I have found an emulsion of
Canada balsam, turpentine, and white of egg quite nice
to paint with, though of course drying rather rapidly.
Curiously enough this emulsion, if the white of egg is not
in excess, dries to form a hard transparent layer, the egg
merely serving to make it easier to manipulate under the
brush. A picture painted on paneL with this medium
would resist chemical change, but would be brittle and
easily injured mechanically.
The problem is, therefore, how much drying oil can be
safely introduced to give toughness without spoiling the
protecting value of the balsam.
The whole subject is deserving of further investigation,
but it is, I think, at any rate possible that the Van Eyck
medium was principally a natural pine balsam, like Venice
turpentine, toughened with a little oil, and perhaps
emulsified with white of egg.
The solid painting might be pure egg, while the upper
glazings would be nearly pure balsam with a little oil.
The green would be prepared by dissolving verdigris in
Venice turpentine by heat, and then probably adding a
little oil.
METHODS OF PAINTING 117
APPENDIX III
Distillation
Its history in ancient times and during the Middle Ages.
Extracted from the account by M. Berthelot in
La Grande Encyclopidie, by kind permission of the
publishers.
' The operation of distillation does not appear in the
history of the sciences before the Christian era, although
the industrial practices relative to the preparation of
mercury are certainly more ancient. But towards the
iirst centuries the concordant texts of Pliny and Dioscorides
prove that it was in use. They explain that cinnabar
being placed in an iron cup which is itself enclosed in an
earthen vessel with a head (afifii^) carefully luted all
round, and the whole heated on a coal fire, the mercury
sublimates and condenses by cooling on the head, whence
it is collected by scraping. It is a real sublimation which
is thus described. These authors describe also a prepara-
tion of essence of turpentine carried out by heating resin
in a pot and condensing the vapour in wool spread on the
orifice^ this wool being afterwards squeezed in order to
extract the volatile oil. To the same order of ideas might
also be referred the preparation of pompholyx, or oxide
of zinc, described by Pliny, the ore (calamine) being
thrown into a furnace and the vapour condensing into
smoke in the midst of a second chamber superposed.
' More exact indications as to the results of distillation
are to be met with in a writer who lived in the third
century, Alexander of Aphrodisias, a commentator on the
118
GREEK AND ROMAN
" Meteorologica," which is a writing of the Aristotelian
school. It is said there that sea-water can be rendered
drinkable by heating it in a boiler and collecting the
vapour in covers placed above. Wine and other liquids,
it is added, furnish water in the same way.
' Besides these documents drawn from classical sources
we find others, more detailed, and accompanied by draw-
ings of the distilling
apparatus in the writ-
ings of the Greek al-
chemists of the same
epoch. The most an-
cient are derived from
a treatise, now lost,
written Ity Cleopatra, a
learned Egyptian lady
who lived some time in
the early centuries of
the Christian era, and
who also wrote on
weights and measures.
Along with the mystic
figures of her "Chry-
sope6 " we find a distil-
ling apparatus con-
sisting of a mattrass
surmounted by a large
tube which debouches
into an upper receiver
provided with two inclined bent side tubes by which
the condensed liquid flows. The whole is placed on a
furnace.
' Similar figures more detailed are found in the works
of Zosimus, the oldest writer on alchemy whose authentic
writings we possess. He seems to belong to the epoch of
Clement of Alexandria and of TertuUian (about 200 A.D.).
Some of his figures have two or three lateral tubes, the
upper receiver being called /irJKO's.
W
Figure from a MS. in St. Mark's, Venice.
METHODS OF PAINTING
119
' We shall give only the accompanying figure (1) from
Zosimus. It is that of a true alembic.
'An apparatus (2)
still more analogous to
our alembics is de-
scribed and pictured in
Synesius, an author
who lived at the end of
the fourth and begin-
ning of the fifth cen-
tury. In Synesius we
have a boiler heated
over a bain-marie, or a
bath of cinders.
'The agreement be-
tween the MSS. con-
taining the figures here
given, and the indica-
tions in the " Meteoro-
V:?
The MSS. containing these figures were
copied in the 11th and 15th centuries, but they
reproduce more ancient MSS., and the figures
correspond exactly to descriptions in the text.
logica " by Alexander of Aphrodisias leaves no doubt as to
the existence of distilling apparatuses founded on the same
principles as our own, in the time of the Eoman empire,
particularly towards the third and fourth centuries of our
era. It is to be observed that the covers, heads, condens-
ing vessels then bore the names of S-ix^l^ and of /SijKos.
This is doubtless the origin of the word alembic, modified
only by the addition of al, the Arabic article.'
It is clear from this account that there is no evidence
of distillation with covered vessel and descending tube so
early as the time of Pliny, though apparently invented
shortly after. It is still more improbable that such inven-
tions were used at this early time for the preparation of
turpentine, alcohol, or rectified petroleum in considerable
quantities.
INDEX
Abacus, 71.
Abies bakamea, 19.
Aeid test, 95.
Aetius, 4.
Agrippa, baths, decoration of,
53.
Alcohol, 20, 31.
Aldobrandini Wedding, 97.
Alexandria, 12.
Alizarin, 14.
Alkanet, 39.
Alum, 13, 14.
Alumen, 13.
Amatorius (Plutarch's), quota-
tion from, 52, 59.
Analyses of frescoes, 94-104.
Apelles, 48.
- — varnish used by, 33, 34,
35.
ApoUodorus of Athens, 46.
Appianum, 55.
Arabic, gum. See Gum.
Architectura (Vitruvius), 9.
Aristides, 57.
Aristoclides, 50.
Arsenic, sulphide of, 11.
Are Vitraria, 17.
Atramentum, 24, 33, 34.
Azurite, 10.
Azzurro della magna, 10.
Baldwin Brown, 113.
Balm of Gilead, 27.
Balsam, 19, 20, 28, 29, 32, 33,
34, 35, 66, 67, 114-116.
Baths of Titus, 14.
Beeswax. See Wax.
120
Berger, 8, 40, 42, 43, 54, 59, 61,
62, 65, 67, 69, 78, 79, 81, 88,
92, 95, 96, 97, 107, 112.
Bird lime, 30.
Bitumen, 31, 34.
Black, 12.
charcoal, 25.
lamp, 12, 25.
pigments : preparation of
with glue, 24, 25, 26.
Blue, Egyptian, 12.
Bono black, 12.
Bronze box at St. M^dard, 65.
Buchner, 65.
Buon fresco, 23, 68, 69, 71, 78-
88, 95, 96, 101, 103, 109.
joinings in, 71, 88.
Burns, wax picture by, 64.
Caeruleum, 22, 55.
Gains Cestius, 11.
Canada balsam, 19, 66, 115, 116.
Gandela, 92.
Carbonate of copper, 10, 13.
Caseine, 68.
Castor oil, 19, 31.
Cauterium, 46, 60, 61, 62, 65, 67,
109.
Cement, 22.
Cennino Cennini, 15, 23, 53, 69,
111.
Cerotic acid, 41.
Cerussa, 11.
Cestrum, 60, 61.
Chalk, 10, 13, 53.
Chaptal,'96.
Charcoal black, 12, 25.
INDEX
121
Chevreul, 65, 82, 95, 98, 99,
102, 103, 104.
Chrysocolla, 13.
Cinnabar, 11.
Cochineal lakes, 16.
Codex Bambergensia, 51.
Coffins, Egyptian, 26, 33, 53.
Coloured grounds in Pompeii,
82.
Copper carbonate, 10, 13.
frit, 12, 16, 86.
ores, 10.
Cyanos, 15.
Cypress, 28.
Davy, Sir Humphey, 11, 14, 95,
96, 97, 103.
-De Architectura (Vitruvius), 9.
Decoration of ships, 36, 37, 55.
Degering, 77, 79.
Didron, 44.
Dieterich, 20.
Diosoorides, 4, 27, 28, 31.
Distillation, 31, 32, 117-119.
of pitch, 32.
Diversarum Artium Schedida, 1 .
Donner, 58, 59, 77, 88.
Dragon's blood, 14.
Drop black, 25.
Drying oils, 1, 4, 5, 6, 18, 19, 20,
30, 36, 54, 63.
Dye, purple, 13.
■woods, 16.
Dyes, 13, 14.
Earth colours, 10.
infusorial, 13, 17.
Earths, white, 10, 13.
Eastlake, 58, 61.
Egg, 18, 21, 22, 27, 36, 53, 68,
94, 99, 104.
tempera, 69.
iyKaloi, translation of, 57, 59.
Egypt, varnish used in, S3.
Egyptian blue, 12, 17, 86.
coffins, 26, 33, 53.
frit, 12, 16, 86.
gesso, 53, 113.
Egyptian blue, wall painting, 68,
107.
Elasippus, 57.
Emulsion from wax, 40, 42, 44,
108.
Encaustic, 48, 51, 54-68, 69, 70,
109.
Essential oil, 30.
Faraday, 105.
F6e, 40.
Fish glue, 24, HI.
Frankincense, 27, 30.
Fresco, analysis of, 85.
chemistry of, 83, 84.
microscopic examination of,
86.
experiments on, 86.
painting, 23.
polishing of, 86, 87.
touching up with egg
medium, 23.
Frit, Egyptian, 12, 16, 86.
y6.v<a<n^ 89, 90, 94, 106, 110.
Geiger, 104.
Gesso, 26, 53, 113, 114.
Egyptian, 53.
Glue, 18, 21, 24, 26, 27, 36, 53,
68, 87, 94, 96, 99, 104, 106,
110.
in black pigments, 24.
fish, 24, 111.
Gold leaf, receipts for laying on,
22.
Grounds, coloured in Pompeii,
82
Gum', 18, 21, 26, 27, 36, 53, 68,
94, 109.
Gypsum, 10, 14.
Hawara wax portraits, 61, 62,
67.
Hermeneia, wax potash medium,
44.
Herringham, 15, 23, 111.
Hittorf, 82, 98, 105.
Honey, 13.
122
INDEX
Index, Pliny's, 51.
Indigo, 14, 55.
Infusorial earth, 13, 17.
Ink, composition of, 25.
Inurert, 57.
picturam, translation of,
58.
Iron, oxide of, 86.
Ivory, 60.
painting on, 61.
Jehan le Bequk, 16, 45.
Jocnndus, 78, 79.
John, 40, 96, 98, 107.
Kbrmes, 14, 16.
Lake, madder, 14, 17.
Lakes, coehineal, 16.
• vegetable, 14.
Lamp black, 12, 25.
Landerer, 105.
Lapis lazuli, 15, 16.
Larch, 19, 28.
Lead, oxides of, 11.
Lentisk, 28.
lAaculorum, 77.
Lime, 22, 68, 79, 83, 84, 85, 104,
107.
Linseed oil, 18, 19, 31.
Litharge, 11.
Lucca, 111.
Lutum, 13.
Madder, 13, 14.
lake, 14, 17.
Maltechnik des Alterthums, 8.
Marble, polishing of, 93.
dust, 75, 86.
experiments on painting,
105-7.
Marmorosum sil, 70.
Mastic, 28, 91, 113.
Mayhoff, 9, 43, 51, 59, 60, 61.
Medium, 18, 19, 20, 21, 27, 36,
94, 105, 114.
probable nature of, 52-
64.
Medium of wax resin nut oil,
63.
of wax and potash in
Hermeneia, 44,
— — in Thebes Papyrus, 27.
Mediums, evidence for two, 46-
54.
for wall-painting, 68,
81.
Melian white, 55.
Mercury, sulphide of, 11.
Merrifield, 10, 45.
Method of inquiry, 1-8.
Milk, 68, 99, 104, 109.
Mistletoe berry, 30.
Modelling in wax, 37.
Mortar, 74, 75.
Mount AthoB, 44, 106.
Mnesilaus, 57.
Muratori, 111.
Murex, 13, 16, 17.
Myrioyl palmitate, 41.
Natural History (Pliny), 9.
Nicanor, 57.
Nitrum, 39, 40.
Nut oil, 4, 36, 63.
OOHKBS, 10, 70.
Oil, 21, 27, 29.
from pitch, 32.
essential, 30.
linseed. See Linseed Oil.
oUve, 30.
poppy, linseed, castor, 31.
varnish, 20, 31.
walnut, 30.
Oils, drying, 18, 19, 20, 36, 54.
vegetable, 18.
volatile, 19.
of walnut, linseed, poppy,
castor, 19.
Olive oil, 30, 63, 93.
Orpiment, 11, 55.
Oxide of iron, 86.
Oxides of lead, 11.
Painters ' with the brush,' 46.
INDEX
123
Panel, paintings on, 46, 47, 48,
' 49, 52, 53, 61, 67, 109.
Papyrus, Thebes, medium in, 27.
Papyrus ash, 39.
Pausias, 48.
Petrie, Flinders, 14, 61, 62.
Petroleum, 20, 31.
Pigments, 8-17, 24, 37, 40, 55,
80, 96,
Pigments injured by wet lime,
23, 55, 70.
Pensea saroocoUa, 26.
Pine tree, 19, 28.
Pistaoia terebinthus, 28.
Pisselaeon, 32.
Pissinum, 32.
Pitch-tree, 28.
28, 29, 36.
oil from, 32.
distillation of, 32.
Pix 28
Plaster', 75, 76, 83, 85, 107, HO,
111.
of Paris, 10.
Pliny's Index, 50.
Natural History, 9.
Plutarch, 52, 59, 94.
Polygnotus, 49, 57.
Pompeii, 82, 95, 96.
Pontic beeswax, 39.
Poppy oil, 4, 19, 31.
Portraits, wax at Hawara,
61, 62, 67.
Praxiteles, 57.
Presuhn, 96.
Prisoianus, 43.
Protogenes, 47.
Punic wax, 39-45, 91-92, 105.
Punica, correct reading, 43.
Parpurissum : laying on with
egg, 22, 23, 24.
55_
Puteoli,' 12.
papSlov, 61.
Keber, 77, 79.
Red lead, 11.
ochre, 10.
Resin, 19, 20, 21, 27, 28, 29, 31.
Richards, 79.
Rode, 77.
Rose, 9, 43, 77, 91.
Russell, 12, 14, 17.
St. Medabd, 62, 65, 66, 67, 101,
102, 103.
Sand, 75.
Sandyx, 22.
Saponification of wax, 41.
Sapphire, 15.
SarcoooUa, 26.
Semper, 105.
Ship decoration, 36, 37, 55.
Ships, tarring of, 29, 36.
Sienna, 10.
Sillig, 51.
Size, 44, 69, 87, 106, 107.
Soap, absence of, 104.
from wax, 40.
Sodium carbonate, 40, 41, 42, 44,
45, 54.
Soot, 25.
Spirit varnish, 20, 31, 33.
Spirits of turpentine, 19, 31, 54,
116.
Stucco lustro, 69, 107, 108.
Sulphide of arsenic, 11.
of mercury, 11.
Syria, 28.
Tar, 37.
Tarring of ships, 29.
Terebinth, 28, 113.
Terra-cotta, 56.
Terre verte, 10.
Thebes papyrus, medium in,
27.
TheophiluB, 1, 102.
Theophrastus, 15.
Titus, baths of, 14.
Turpentine, spirits of, 19, 20,
21, 31, .32, 54.
Venice, 19, 35, 116.
Ultramarine, 10, 15, 16.
Umber, 10.
124
INDEX
Varnish, 4, 19, 20, 21, 28, 31,
32, 34, 35, 37, 115.
used by Apelles, 33, 34,
35.
used in Egypt, 33.
oil, 20.
spirit, 20.
Varnishing of ships, 36.
Vegetable oils, 18.
Venice turpentine, 19, 35, 116.
Verdigris, 11, 115.
Vermilion, 11, 16, 86, 94.
protection of, 88, 91.
Vestorius, 12.
Vitruvius, 9.
MSS. of, 78.
Walls, painting on, 48, 49, 50,
52, 68-84, 80, 82, 111.
Walnut oil, 19, 30.
Wax, bees', 27, 29, 36, 39, 55,
60, 69, 93, 94 105, 106, 110.
and resin, 65.
candles, 92, 93, 94, 106,
110.
emulsion, 40, 42,44, 54, 108.
medicinal use of, 44.
mixed with pigments, 37,
39, 55.
Wax, bees', mixed with resin, 63.
mixed with balsam, 66.
mixed with oil, 63.
mixtures of, 54.
modelling in, 37.
• on marble, 105.
painting, 54-68.
portraits at Hawara, 61,
62, 67.
preparation of, 37-45.
Wax, Pontic, 39.
potash receipt, Jehan le
B^gue, 45.
soap, 40, 4] .
treatment with soda, 41.
turpentine painting, 58.
for walls and armour, 39,
67.
Punic, 39-44, 105.
Weld, 14.
Wiegmann, 71, 87.
White earth, 10, 13.
lead, 11, 55.
Woad, 14.
Yellow ochre, 10."
ZB0XIS of Heraclea, 46,
Zopissa, 29, 36.
Printed by T. and A. Constable, Printers to His Majesty
at the Bdinbnrgli University Press
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11