THE SCIENCE OF PAINTING. 
 
THE HORSE. 
 
 BY 
 
 E. DUHOUSSET, 
 Lieut.-Coloiiel. 
 
 A TRANSLA TION 
 
 LONDON : 
 
 PERCY YOUNG, 137, GowER Street, W.C. 
 
THE 
 
 SCIENCE OF PAINTING. 
 
 BY 
 
 J. G. vibert: 
 
 A TRANSLATION 
 FROM THE EIGHTH EDITION, REVISED BY THE AUTHOR. 
 
 LONDON : 
 
 PERCY YOUNG, 137, GOWER STREET, W.C. 
 
 1892. 
 
Books are dedicated to esteemed colleagues, to admired celebrities, or 
 to illustrious individuals, whose patronage the author thus hopes to 
 obtain. 
 
 This book would be dedicated to the Institute of France, were such a 
 dedication permitted to any hook. 
 
TO THE READER. 
 
 The author, at the conclusion of the course of public 
 lectures which he gave at the School of Fine Arts," has 
 willingly consented, at the request of many of his hearers, 
 to publish his lectures in this book, which he has now the 
 honour of placing before you. 
 
 Having no pretension to give to his colleagues lessons, 
 or even advice, except for the preservation of their works, 
 the author will pass over all questions of beauty of drawing, 
 of anatomy, of perspective and. of composition, which form 
 part of the art of painting, but which have no influence on 
 its durability; he will only take up the practical part, or 
 what might be vulgarly called the wire-pulling department. 
 At the most he will only make some scientific digressions, 
 necessary because certain effects cannot be properly under- 
 stood unless their causes are known. 
 
 This book, the fruit of thirty years of study and ex- 
 perience, is the most serious and honest work which has 
 been written on the subject. At least the author thinks 
 so ; and he must say so, for should he through excess of 
 modesty say anything else, he would not be believed. 
 
 Notwithstanding the qualities which its author believes 
 it to possess, will this book have the results which he 
 expects from it ? 
 
vi 
 
 TO THE READER. 
 
 Will it cause painters to adopt a more rational method ? 
 May we hope, thanks to it, to see the masterpieces of the 
 future preserve their brilliancy and freshness ? 
 
 Of that the author is not quite sure. Not daring to feel 
 certain of great success, he will feel himself rewarded for 
 his efforts if his book finds one reader from beginning to 
 end ; and you as that possible reader he now salutes. 
 
CONTENTS. 
 
 CHAPTER I. . PAGE 
 
 The Process of Painting at Different Periods . . . 9 • 
 
 CHAPTER II. 
 
 Light and Colours 17 
 
 CHAPTER III. 
 The Laws of Colouring. 
 
 Mixing of Opaque Colours. — Mixing of Two Colours. — Mixing of 
 Three Colours. — Superposition of Colours. — Superposition of 
 Transparent Colours on White and Black. — Superposition 
 of Transparent Colours over each other. — Superposition of 
 Translucid Colours. — Contrast of Colours by Juxtaposition. 
 — Contrast of Black and White. — Contrast of Juxtaposed 
 
 Colours . . . • . . . . . . . «31 
 
 CHAPTER IV. 
 
 With the Scientists . . . .48 
 
 CHAPTER V. 
 
 Colouring Substances 56 
 
 CHAPTER VI. 
 
 Oils, Grinding of Colours, Essences, Petroleums, and Sicca- 
 tives . 71 
 
 CHAPTER VII. 
 Resins and Varnishes. 
 Re-touching Varnish. — Painting Varnish. — -Picture Varnish . .84 
 
 CHAPTER Vin. 
 
 Canvas, Panels and their Sizings, Pastes . . . .96 
 
viii 
 
 CONTENTS. 
 
 CHAPTER IX. 
 Outline and Execution of a Picture in Oils. page 
 Outlines in Oil. — Execution. — Repainting. — Glazing. — Embus . no 
 
 CHAPTER X. 
 The Preservation and Restoration of Pictures. 
 Cleaning 122 
 
 CHAPTER XL 
 
 Wall Painting 130- 
 
 CHAPTER XII. 
 
 Crayons, Distempering, Egg Painting, Water Body-Colours . 135 
 
 CHAPTER XIII. 
 Water Colours. 
 Paper. — Gums. — Fixative. — Water Colours Fixed by Fire. — Sar- 
 
 cocol Water Colours ........ 146 
 
 CHAPTER XIV. 
 
 Conclusion . . . . 160 
 
 APPENDIX. 
 
 Recipes and Manipulations. 
 Good and Bad Colours.—Lead Whites. — Zinc Whites. — Chalk 
 Whites. — Earths, Ochres and Marls,' Iron Lake, Yellow and 
 Orange Colours. — Red Colours. — Blue Colours. — Green 
 Colours. — Violet Colours. — Brown Colours. — Verification of 
 Colours. — Cleaning of Pictures. — Varnishing of Pictures. — 
 Fixative for Water Colours and Dissolving Liquid. — Varnished 
 Water Colours. — Repair of little Daily Accidents. — Canvas 
 Blistered, Cracked or Broken. — Panels of Unpolished Wood. 
 Caséine or Cheese Paste. — Sizings of Caséine Paste and 
 Zinc White. — Sarcocol and Sarcocolline. — Gelatine. — Starch 
 or Farina Paste. — Albumen. — Dextrine. — Gum Lac. — Water, 
 Egg, Gum and Sarcocolline Varnish. — Inks. — To render Wax 
 Miscible with Water and Glycerine. — Prevent Vibration 
 of Canvas or Paper stretched for Crayons. — Instantaneous 
 Sizing. — Mastic for Plastering Up. — Method of Rendering 
 Paper Transparent. — Paper for Taking Tracings from Oil 
 Paintings. — Commercial Guarantee . . . . .162 
 
SCIENCE OF PAINTING. 
 
 ♦ 
 
 CHAPTER I. 
 
 THE PROCESS OF PAINTING AT DIFFERENT PERIODS. 
 
 As soon as men collected together anywhere to live as 
 a society, they had a religion and an art, whose first 
 manifestation has always been painting. 
 
 The primitive process employed was everywhere the 
 same. It still exists in all its simplicity with some savage 
 tribes who use it to paint their idols, their utensils, their 
 firearms, and even the bodies of their warriors. This 
 process, limited to the use of argillaceous earths, naturally 
 coloured by the presence of metallic oxides, or to dyes 
 obtained from the decoction and the sap of certain plants, 
 consists in mixing earth with water to form a coloured 
 paste, and then by means of a little stick, or merely with 
 the finger, spreading it over the object to be decorated. 
 This is a very simple method ! 
 
 And yet, although this first process is so very elementary, 
 it contains the germ of all the others ; and we are about to 
 see that the- numerous colouring-matters placed to-day by 
 nature and science at the disposal of painters are nearly 
 all combinations of three elementary substances — clay, 
 metallic salts, and vegetable dyes. 
 
 v.p. ' I 
 
lO 
 
 SCIENCE OF PAINTING. 
 
 This was afterwards improved by mixing gum. with these 
 elementary colours, to obtain more substance and more 
 brilliancy. 
 
 The Egyptians have bequeathed to us on their mummies 
 specimens of these first gum paintings, which they further 
 covered sometimes by melted wax, and which, being thus 
 preserved from contact with the air, have retained an extra- 
 ordinary freshness. 
 
 In the most ancient times the use of coloured clays, 
 incorporated with lime and cement, was customary, and 
 also the manufacture of terra cotta covered with vitrified 
 substances. Such are the coarse enamels and cements of 
 various colours, which, broken into little cubes and after- 
 wards encrusted in fresh mortar with other pieces of marble 
 or stone, constitute the earliest specimens of mosaic work. 
 
 In the heroic age of Greece, the painter introduced those 
 same colouring matters, reduced to fine powder, into melted 
 wax ; often he added resin to this, and spread those pastes 
 softened by heat on the surfaces which he wished to deco- 
 rate, by means of a bronze spatula slightly warmed. After- 
 wards, with the other end of the spatula, which was pointed, 
 he drew, on the yet soft surface of colour, the object which 
 he wished to represent ; and after having broken off all that 
 was outside the outlines he filled up the place left vacant 
 with paste of another colour. 
 
 In this manner were painted the prows of the ships which 
 carried the Greek kings during the Trojan war.* 
 
 In short, until then, the art of painting had much in 
 common with that of sculpture, and only consisted of flat 
 tints. Much later, with the use of the paint-brush, the first 
 attempts at shading appeared. 
 
 * This kind of painting has lately been resumed, and a picture painted 
 by this process was on view at the last Salon. 
 
PROCESS OF PAINTING AT DIFFERENT PERIODS. 11 
 
 The artists of those days could, by spreading ground 
 colours mixed with water on fresh mortar made of sand and 
 lime, produce graduated tints ; and the first pictures worthy 
 of the name were painted after this invention of fresco- 
 painting. They painted afterwards with resinous gums into 
 which was introduced wax, rendered miscible with water 
 by the help of lime. 
 
 This process finished, the painting was moved backwards 
 and forwards before a bronze chafing-dish, in the form of a 
 grating filled with hot coals and called a caitterium. The 
 wax and the resin, melting under the influence of the heat, 
 then formed, with the colours, a homogeneous whole whose 
 durability was almost unhmited. 
 
 Paintings done by this process have been seen in a 
 perfect state of preservation nine centuries after they had 
 been painted, although they were on outside walls exposed 
 to all the injuries of the weather. 
 
 In the same way the Greek sculptors coated over their 
 marble statues, to preserve them and to give them a verdi- 
 gris of which unmistakable traces are still found on some 
 fragments. 
 
 At last, when by distillation they were able to extract 
 essential oils from certain plants, painters made varnishes 
 by dissolving resins in these essential oils, and one may say 
 that then the painting of the ancients attained its apogee. 
 In fact, a picture of this period painted on lava was ex- 
 hibited at the Universal Exhibition of 1889 in the Industrial 
 Court, and all who saw it were struck with its extraordinary 
 state of preservation. 
 
 These different processes of antiquity were continued 
 almost until the Middle Ages, when the use of wax was 
 almost entirely discontinued, being superseded by egg 
 painting which rapidly became general. The miniatures 
 
12 
 
 SCIENCE OF PAINTING. 
 
 on the parchment of missals, the panels of costly shrines, 
 the pictures painted on three-folding panels which orna- 
 mented the altars and the walls of churches, were mostly 
 painted by this process which is very substantial, as it 
 admits of the introduction of resins — the yolk of egg con- 
 taining an oil capable of dissolving the resins. 
 
 Oil painting began in the fifteenth century. We are not 
 here going to enter into that great discussion, which is not 
 yet finished, as to whether oil painting was really discovered 
 by Van Eyck, but perhaps we may venture to say that its 
 discovery was really caused by a mistake. 
 
 In the first place, neither John Van Eyck nor his brother 
 ever acquainted anybody with their processes, and their 
 secrets died with them ; and if report is to be credited, they 
 were even so jealous that they stabbed one prying individual 
 who ventured into their studio. 
 
 In vain distinguished artists, attracted by the fame of the 
 two brothers, came from Italy to Bruges : they could learn 
 nothing, except that the belief spread that Van Eyck owed 
 the exceeding brilliancy of his colours to the use of oil. 
 
 Now, at that time the word oil was used indiscriminately 
 for all the slimy liquids which could be extracted from 
 plants without the use of heat, and some of which certainly 
 are essential oils. People did not know that such could be 
 obtained also from the hard yolk of egg, and that these 
 essential oils form with resins thick varnishes which have 
 all the greasiness of true oils, and give to colours a trans- 
 parency and a richness surpassing that given by oil alone. 
 
 What is certain is that, from the processes of Van Eyck, 
 only one word transpired, and that was — oil ! We now 
 know that this word could be used with a double meaning, 
 but for the painters of that date it was otherwise : they 
 hastened to grind their colours with any sort of oil in order 
 
PROCESS OF PAINTING AT DIFFERENT PERIODS. I3 
 
 to imitate the great inventor, and yet they never succeeded 
 in equaUing the powerful colouring of his pictures. 
 
 Now, among the adepts at this new style of painting there 
 were some great masters who ought to have been able to 
 make as good use of the method as Van Eyck himself : we 
 may therefore permit ourselves to say that the process, such 
 as it has been employed since his time, was perhaps not his, 
 or that it has been badly understood, for if this great painter 
 really did use the oils which are still used to-day, he never 
 used them except when much imbued with resin on panels 
 prepared with size and outlined in egg : so that the oil 
 colours applied semi-liquid or in glaze only merely exist as 
 pellicles in his pictures. 
 
 Not being able to penetrate the mystery which envelops 
 the invention of oil painting, and convinced that, if it was 
 Van Eyck who discovered it, he did not practise it as it 
 has been practised since his day, we shall content ourselves 
 with saying that the use of oil painting spread rapidly after 
 Van Eyck's death, which occurred in 1450, and that, with 
 regard to results obtained, the Flemish and Dutch schools 
 were those which most nearly approached the great painter 
 of Bruges — which allows us to suppose that they also 
 approached him most nearly in technical methods. 
 
 The custom of painting on surfaces prepared with size and 
 of adding resin to oil-colours was continued for more than a 
 century afterwards in Flanders. Otto Venius, the master of 
 Rubens, retained, the method, and Rubens himself painted 
 all his first pictures in the same manner. When, later on, 
 he used thickenings, it was only in the light colours, and he 
 never abandoned the use of coatings of size. 
 
 After this incomparable master, the method fell into 
 disuse. Resins were no longer employed, which necessi- 
 tated increasing the proportion of oil ; canvas and panels 
 
14 
 
 SCIENCE OF PAINTING. 
 
 were prepared with coatings of oil, and thick layers of oil 
 colours were put over them. To this abuse of oil was 
 added that of turpentine ; and the painting, gradually losing 
 its transparency, took a dull and floury appearance, and this 
 went on increasing until the end of the eighteenth century. 
 
 At the beginning of the nineteenth century all these 
 traditionary methods disappeared, neglected by those who 
 ought to have taught them, forgotten by those who ought to 
 have practised them, and at last entirely unknown to the new 
 generation. 
 
 This generation, that of 1830, hastening to reverse every- 
 thing, yet understood that a method was necessary, and 
 they created some new ones : they created many, even too 
 many, for not one of them was good. Never was so much 
 written about painting as at that time, and never was paint- 
 ing so bad in point of substance ; each group, each artist 
 even, had formulas which he kept secret. 
 
 One of them, facetiously inclined, wrote on the little 
 bottles which lay about in his paint-box, " Oil of Perlimpin- 
 pin," " Balm of Apollo," " Aerial fluid," etc. What looks of 
 envy the pupils in his studio cast on those treasures ! How 
 much they would have liked to have used them ! And 
 later on, when they saw how brown the pictures of this old 
 magician had become, how they congratulated themselves 
 that they had not been able to do so ! 
 
 To-day, artists have no longer any of that paltry jealousy, 
 and, disdaining even to occupy themselves with the material 
 part of their art, they leave the care of preparing their 
 canvas and their colours to grinders who come down in 
 hordes from the mountains, — we shall not say what mount- 
 ains, so as to wound no susceptibihties. 
 
 Those clever but ignorant colour-dealers invent oils, 
 pomades, siccatives, etc., and under pompous names they 
 
PROCESS OF PAINTING AT DIFFERENT PERIODS. 1 5 
 
 manufacture disastrous mixtures of colour. They furnish 
 everything : the box, the easel, the paint brushes, the subjects 
 of the pictures, even their. own advice and the favourable 
 opinion of the critics. 
 
 The modern artist, furnished with all this amateur stock 
 of wares and having no need to think of his requirements 
 is free to follow the dictates of his fancy. He paints 
 casually, without thought of the morrow, being merely 
 desirous to follow the fashion, for that is important ! There 
 are exquisite tints which it is necessary to have : veined flesh, 
 marble transparencies, dull softenings, dusky shading, etc. ; 
 there is not enough of that variety, — there must be more, 
 the journalists insist ! 
 
 Warm colours ? enamel ? 
 
 No more of that ! 
 
 And then there are the treasures of the virtuoso, the all 
 powerful touch lovingly exhibited ! 
 
 As for the vagaries of the paint brush invented by Paul 
 de Saint Victor, they are old. 
 
 And yet the masterpieces of yesterday, scarcely in the 
 museums, get dark, crack, and weep tears of bitumen ! In 
 the monuments, the new frescoes get mouldy, peel off, fall 
 from their places ; and the contemporaries who have seen 
 when they were new the works of Henry Regnault and of 
 so many others are obliged to tell to the young people who 
 are unwilling to believe them, that the colours in those 
 works were extremely brilliant and fresh when new. 
 
 It is only necessary to take a walk in the Louvre to see 
 that the preservation of the picture is in direct ratio to its 
 antiquity : that is to say that in the paintings of the fifteenth 
 century, to go no farther back than that, the colouring 
 remains more brilliant and the materials more substantial 
 than in those of the sixteenth century, and that in approach- 
 
i6 
 
 SCIENCE OF PAINTING. 
 
 ing our own time the painting deteriorates more and more ; 
 those paintings which are in the worst state of preservation 
 are not many years old. 
 
 Whose fault is this ? Alas ! it is that of artists themselves, 
 for their indifference surpasses all bounds. 
 
 But, are they seeking to excuse themselves ? " What are we 
 to do to make our painting imperishable ? Nobody ever spoke 
 to us of it. Our professors themselves knew nothing about 
 it." That is partly true. Therefore do not let us recrimi- 
 nate, and let us forgive those who have gone before ; but, 
 so that those who come after us may be wiser, let us find 
 again those lost traditions which we shall do our best to 
 perfect with the help of the discoveries of modern science. 
 
 With this object we shall study all the materials employed 
 by our predecessors, beginning always at the times of 
 greatest antiquity and coming up to our own days ; thus 
 learning, by the use which the ancients have made of their 
 methods and the results which have reached us, what ought 
 to be retained and what ought to be abandoned. We shall 
 thus be able, on a sure basis made by the experience of 
 many centuries, finally to establish the essentials of the 
 material processes of painting, and, shedding abroad the 
 light into the darkness of ignorance and empiricism, bring 
 forth from it a new science. 
 
CHAPTER IL 
 
 LIGHT AND COLOURS, 
 
 It is not enough to give to artists good material to ensure 
 the durability of their works, if those materials are badly 
 used ; and it is not possible to indicate in a general manner 
 how they ought to be used, as each object painted requires 
 a different style of execution. 
 
 The painter should obtain from his colours, by the 
 different manners in which he puts them on by means of 
 oils and varnish, all the effects of transparency, of opacity, 
 of brilliancy, of dulness, etc., which light is capable of pro- 
 ducing. Each of these particular manners of using the same 
 materials so as to give a durable result necessitates special 
 precautions which we shall indicate ; but in order properly 
 to understand how one can reproduce artificially all the 
 phenomena of light and of colour which present themselves 
 to our eyes, it is indispensable first to know how they are 
 produced in nature. 
 
 According to the most modern theory, the luminous 
 bodies by themselves, the sun, the stars, the flames, are 
 composed of molecules in a state of perpetual vibration. 
 This vibratory movement, communicated to the surrounding 
 molecules, produces circular waves which, rolling off into 
 space, cause to us when reaching the retina the sensation 
 of light, in the same way that the sonorous waves when 
 reaching the tympanum cause to us the sensation of sound. 
 
i8 
 
 SCIENCE OF PAINTING. 
 
 Some of these luminous waves are absorbed by bodies 
 which they encounter, and thereby undergo chemical 
 changes ; others make their way through transparent media, 
 such as air, water, glass : this is the phenomenon of trans- 
 missio7i. If they meet opaque bodies which stop them, 
 they rebound and are thrown back into space : this is the 
 phenomenon of reflection. 
 
 But, according to the density of the transparent mediums, 
 and according to the matter, the form or the nature of the 
 surface of the opaque bodies, they are transmitted or 
 reflected in different conditions. 
 
 In mediums of homogeneous transparency they are 
 transmitted in a straight line, but the direction changes as 
 soon as they pass into a medium of another density : for 
 instance, a stick plunged into water appears to be broken, 
 that is called refraction. When the waves are bent back by 
 an opaque body to a polished surface, the reflection is direct 
 or specular,, as by a mirror, for instance. In this case, if the 
 light strikes the object perpendicularly, it comes back upon 
 itself ; if it strikes obliquely, it is sent back in the same 
 oblique manner from the opposite side to the one whence 
 it comes. When the body has a rugged surface, the 
 luminous waves are sent back in all directions : this is 
 diffuse reflection. But a body may be more or less trans- 
 parent or opaque in its entirety, or composed of some 
 parts very transparent and others very opaque. It may 
 have a surface more or less polished or rough, or it may be 
 composed of some parts very polished and others very 
 rough ; so that, combinations of transparency, of opacity, 
 and of rough and polished surfaces being unlimited, the 
 phenomena of transmission and reflection of light are also 
 unlimited. 
 
 On those general laws are still to be grafted the effects 
 
LIGHT AND COLOURS. 
 
 19 
 
 of concave or convex surfaces — the convergence or diver- 
 gence of rays ; the effects of multiple refraction from surfaces 
 cut in facets ; the theory of prisms, etc., which constitute 
 the science of optics, upon which we cannot here enter, 
 but which painters would do well to study. 
 
 Now that we know how His Majesty the Sun gives us 
 this beautiful ray, we should also like to know of what it 
 consists. 
 
 Beautiful sunbeams for our information enter this dark 
 room, go through this little hole, and afterwards cross the 
 prism. There ! Now the beautiful ray is analysed, spread 
 out on a screen. It is now a luminous band divided into 
 seven distinct colours : red^ orange^ yellow^ green ^ blue, ultra- 
 marine, violet', and the sharp and learned analyst, getting 
 ready to dissect it, calls it at this point the Macabrean name 
 of solar spectrum. 
 
 Under the scalpel of science this spectrum reveals 
 strange secrets. It shows the process of spectral analysis, 
 thanks to which we are able to learn something of the 
 chemical composition, not only of the stars in our planetary 
 system, whose light comes to us in a few minutes notwith- 
 standing the millions of leagues which separate them from 
 us ; not only of the most distant stars, whose light takes 
 years to reach us, but even of the nebulae, which, yet 
 farther off, are lost in the infinitudes of space, where all 
 thought or calculation would mean madness, were it not 
 that there is found — God ! 
 
 By the help of various prisms and decomposed rays, the 
 analyst in his dark room places in juxtaposition and in 
 superposition colours which are very interesting, but which 
 it will be more profitable for us to consider later. 
 
 He establishes the fact that the coloured rays cannot be 
 decomposed, that their refrangibility increases when they are 
 
20 
 
 SCIENCE OF PAINTING. 
 
 taken from left to right, and that they have each different 
 undulations of length, which go on diminishing from red, 
 where they are 620 millionths of a millimetre, to violet, 
 where they are only 423 millionths of a millimetre.* On 
 account of this the number of their vibrations per second 
 increases from violet, which has 752 trillions of vibrations, 
 to red, which has no more than 514 trillions. Finally, the 
 analyst, reuniting the seven coloured elements of the poor 
 sunbeam by means of two prisms placed opposite to each 
 other, is thus able to reconstitute the white light. 
 
 Beautiful sunbeam ! you have taught us our lesson : you 
 are now free. 
 
 If the beautiful sunbeam could speak, it might tell us 
 that, since the beginning of the world, the secret of the 
 decomposition of light has been under every eye ; that sun- 
 rise and sunset, the rainbow, the iridescent fringe of the 
 glaciers, the thousand fires of the diamond, might have taught 
 us ; and that there was no need to drag a poor sunbeam 
 into a dark room in order to teach us. Whilst hearing this, 
 man, so proud of his learning, might well feel ashamed of 
 the time it had taken him to learn. 
 
 Before studying colours, we are confronted with one 
 primary question. 
 
 Why is a body red or yellow or blue, etc. ? 
 
 We do not know. But if we do not know the cause, at 
 least we may learn the effect of this phenomenon. 
 
 All bodies do not act aHke under the action of the light 
 which they receive. They have the mysterious property of 
 absorbing either the whole or part of that light, or some- 
 times after having decomposed the light they only absorb 
 certain coloured rays of it. Some bodies have only the 
 faculty of absorbing, others have only that of decomposing. 
 * A millimetre is 'ooi of a metre. 
 
LIGHT AND COLOURS. 
 
 21 
 
 Others, again, reunite in different proportions the faculties 
 of absorbing and of decomposing, and those are the 
 differences which determine the colour peculiar to each of 
 them ; for the colour of a body is the sensation which is 
 produced on the retina by the portions of light which it has 
 not absorbed, and which it sends back. 
 
 All absorbed light produces darkness. A body which 
 would absorb all the light which it receives would be quite 
 dark ; it would be impossible to distinguish either light or 
 shade. A body which would absorb nothing from the light 
 which it receives would be just as vivid as that light ; but 
 these two extremes, altogether unmodified, are not met with 
 in nature. 
 
 A body which absorbs part of the light and sends back 
 the rest is grey. The whitest objects, therefore, are only 
 very light grey, and the blackest very dark grey. However, 
 the light which a grey body sends back is the same as 
 that which is sent back by a white body : the difference is 
 merely in quantity. Why do those two lights of the same 
 quality give different sensations ? It is because, the light 
 being shed equally over all bodies, our eye, in comparing the 
 luminous intensity of several objects lighted by the same 
 light, can perfectly distinguish the quantity which each of 
 them absorbs, and our eye supplies the missing quantity 
 with an equal amount of black. Consequently, if a body 
 only sends back the half of the light which it has received, 
 our eye has the sensation of a grey composed of half black 
 and half white. 
 
 All decomposed light produces the seven coloured rays 
 of the spectrum. The bodies which, after having effected 
 this decomposition, absorb one or more of these rays and 
 send back the others are of the colour of the ray which they 
 send back, if they send back only one, or of the colour 
 
22 
 
 SCIENCE OF PAINTING. 
 
 composed by the mixture of the rays which they send back, 
 if they send back more than one. 
 
 Example : — A body which absorbs six rays and sends 
 back only the red is red, and so forth. 
 
 If a body, having absorbed a part of the light which it 
 has received, instead of sending back the rest decomposes 
 it, the colour of the coloured ray, or of the mixture of 
 coloured rays which it will send back will be more or less 
 darkened, according to whether it shall have absorbed much 
 or little of the light. 
 
 Example : — A body which absorbs the half of the light 
 received, decomposing the remainder and only sending back 
 the red ray, gives the impression of half-black and a seventh 
 of half-red, i.e.^ a dark red. 
 
 If, on the contrary, a body sends back a part of the 
 light received, and decomposes the rest, the colour of the 
 coloured ray or of the mixture of coloured rays which it will 
 send back will be of a colour more or less light, according 
 to the quantity of light sent back. 
 
 Example : — A body which sends back the half of the 
 light received, decomposing the remainder, and sending 
 back only the red ray, gives the impression of half-w^hite 
 and a seventh of half-red, ie. pink. 
 
 It is, then, in fact, the mixtures of colours between them- 
 selves, and the large or small quantity of light or darkness 
 which is added to them, that gives the millions of shades 
 discernible in nature and which in painting is called 
 colouring. 
 
 But in the mixtures of colours much difference exists 
 between mixing coloured rays and mixing coloured matter. 
 Two rays of the same light passing one through a pane of 
 blue glass, the other through a pane of yellow glass, give, 
 at the place where they meet, a white light. If, on ithe 
 
LIGHT AND COLOURS. 
 
 23 
 
 contrary, the two glasses are placed one over the other, the 
 light transmitted will be green; and if, reducing the two 
 pieces of glass to powder, the powders are mixed, the hght 
 reflected will still be green. 
 
 In order to explain the reasons of this phenomenon and 
 of many others, it would be necessary to enter into a number 
 of scientific causes sufficient to drive an artist mad. Besides, 
 as we never do have at our disposal on our palette the rays 
 of the spectrum, and consequently cannot utilise them, any 
 more than we can the effects of polarisation^ of phosphor- 
 escence^ of efflorescence^ etc., we shall leave all that for the 
 consideration of those material colours which are all that 
 remain to us when we pass from the study of physics to 
 that of painting. 
 
 Those colours which consist of matter ground with 
 different substances may be had in conditions more or less 
 transparent or opaque. They act on the eye by transinission 
 and reflection^ and consequently give different results 
 according to the conditions in which they are used. 
 
 In fact, if you add a colour to opaque white, you will not 
 obtain the same shade as if you were to rub a thin coating 
 of that colour on the white. In the same way a colour 
 added to opaque black does not give the same result as if 
 it were transparently put on that black. 
 
 By this rule, two or more colours have different effects 
 according to whether they are mixed in an opaque condition 
 or laid one over the other in clear coatings. 
 
 The use of colours of different degrees of opacity consti- 
 tutes what is called, in studio terms, solid and semi-liquid ; 
 and the use of different degrees of transparency, scumbling 
 and glazi?ig. 
 
 The appearance of a colour changes also according to 
 the nature of its surface and the manner in which it is 
 
24 
 
 SCIENCE OF PAINTING. 
 
 painted : the surface can be rendered dull by flaking^ 
 smooth by using the badger brushy or rough by dabbing \ 
 effects of vibration are to be obtained by strokes called 
 hatches. 
 
 In fact, a painter who knows his trade finds means of 
 making his colours take all the different aspects which they 
 present in nature; thus bringing into action unawares, as 
 M. Jourdain made prose, all the phenomena of refraction^ 
 of specular reflectio7t^ and of diffuse reflection^ by means of 
 which the coloured rays manifest themselves to our organs 
 of sight. 
 
 The artist obeys, therefore, immutable laws which he 
 does not know. How can he do it ? The reason is that 
 the laws of colour, like those of perspective, painters had 
 felt before science had discovered them, and that artists 
 have them all, more or less, by instinct. 
 
 How much quicker, however, they would arrive at this 
 facility by knowing, from the outset of their career, those 
 laws which are unchangeable and which they only arrive at, 
 using unconsciously later, after many attempts and jumbling 
 efforts ! 
 
 Yes ; but where are they to be learned ? 
 
 Not in treatises on physics, where they are mixed up 
 with all the questions of optics and of chemical and caloric 
 luminous rays. Neither is it in works specially written for 
 artists by professors, grave but inexperienced, who have no 
 proper knowledge of what painting really is. 
 
 Oh, no ! Without having seen them, one would not 
 dare to believe it possible to write the contents of some of 
 those volumes. 
 
 In one which has very recently appeared, the author tells 
 us : "In order to paint the head of a young man of white 
 race, twenty-one years of age " (not twenty-two or twenty- 
 
LIGHT AND COLOURS. 
 
 25 
 
 three — exactly twenty-one), mix such and such colours for 
 the brow, for the eyes, the lips, the cheeks," etc., and the 
 author draws out a plan of cipher for his directions. It is 
 true that this laughable author informs us that he did not 
 invent the notation (which is to be regretted !). 
 
 The predecessor had expressed the cheeks of the young 
 man by this formula : — • 
 
 but the new writer changes the formula and thus simplifies 
 it ! The young man's cheeks are written : — 
 
 Ah ! what genius ! 
 
 It is works of this kind that give the proportions of colour 
 to be mixed in order to make the shades of the leaves of 
 all the trees, the lakes and the rivers of Europe, the mount- 
 ains of Switzerland, etc. ; to imitate the brightness of the 
 rose, the velvet of the peach, or the flames of a fire ; to 
 reproduce the tints of the human skin in all races of the 
 universe, from the ebony skin of an African negro to the 
 pearly colour of a patrician ; and one limits precisely all 
 the shades according to whether one is painting an infant, a 
 child, a young girl, or a grandmother : they do not give us 
 the recipe for painting a woman who conceals her age ; 
 perhaps all the colours should be mixed up together. 
 Finally, they even furnish us with the colour of the passions 
 — deadly paleness, the blue of anger, the red of shame, and 
 the carnation of bashfulness ! All that is a good deal. Too 
 much to be discovered by a single artist ! 
 
 Let us suppose now that a pupil of this professor paints 
 a picture, following out from point to point his programme : 
 
 V. p. 2 
 
 4 red f^, 4 — 5 shade ; 
 
 R I a 
 R I i 
 
 Oi No 
 Oa Nu 
 
26 
 
 SCIENCE OF PAINTING. 
 
 it is soon done. It is of a noble simplicity, a triumph of 
 local colouring, a masterpiece of to-morrow : it is sold, or 
 offered to the Louvre, and here we leave it. 
 
 But if a thoughtful pupil were tempted to try in his turn, 
 he would first think in this manner : We take a ball and we 
 smear it all over with any colour, say vermilion. We shall 
 next make a copy of this ball in painting. Do you think 
 that for this purpose we can use the same vermilion ? Yes, 
 but only for one stroke of the brush — that which corresponds 
 to the exact middle of the ball ; for, at all the other points, 
 the colour is modified by the bordering of light, the reflec- 
 tions of neighbouring objects, the shadow and the half-tints 
 which constitute the shading, and we should require almost 
 all the colours of the palette in order to reproduce these 
 modifications. 
 
 Now, here is a picture representing an orange tree laden 
 with oranges : we shall leave the fruit and proceed to darken 
 the foliage, and also the sky and the earth. In fact, we now 
 produce the effect of night, and the oranges, which have 
 not changed their colour, are become luminous balls like 
 those which adorn the trees on fête nights. 
 
 Now, let us take a white canvas on which we paint one 
 touch of a brilliant green. Well ! according to what we 
 paint round it, you may make of that touch whatever you 
 like : an omnibus lantern, with a background of Boulevard 
 at nightfall ; a leaf illumined by the sun, in a green bower ; 
 the sacred emerald which shines on the brow of the god 
 Buddha in a Hindoo temple. Without changing anything, 
 this touch may lose its brilliancy and play the rôle of a 
 lizard in half-tints amidst the effulgent odalisks of Diaz, 
 or it may be lost in the shadow of a stake outlined on the 
 Grand Canal at Venice, touched by the light of an Italian 
 sunset. 
 
LIGHT AND COLOURS. 
 
 27 
 
 Thus, a material colour, placed in a picture which remains 
 lighted by the same light, changes its tone according to its 
 surroundings. 
 
 Delacroix said : " Give me some mud, I will make of it 
 the skin of Venus, if you leave to me the choice of the 
 surroundings." 
 
 It is the same thing in nature, where the colour of a 
 body can be modified to any extent by the degree and the 
 colouring of the light which it receives from the reflection 
 or the mere contrast of the colours near it, etc. 
 
 Let us follow a cardinal, dressed in red, whilst he walks 
 in his gardens. At every instant the colour seems different, 
 according to whether he receives the blinding rays of the 
 sun, or the white reflection from a cloud, or shelters under 
 the verdant shade of a leafy grove. AVhether we see him 
 on the intense green of the sunny lawns, under the dark 
 green of the cypress, on the silvery surface of a lake, or 
 under the azure of the sky, he still changes. He changes 
 always, becoming pale before a bank of geraniums, and red 
 before the marble of the statues ; he gets dark in proportion 
 as the daylight fades, until he becomes of a dark purple, 
 and is dressed in black, like a simple priest, as he returns 
 to his palace by the dusky shades of twilight. 
 
 There is still another reason why the colour of a body 
 cannot be defined by a formula : it is that the colour does 
 not appear alike to every one. Although it is said that 
 *' Tastes and colours should not be discussed," and although 
 popular sayings have always a foundation of truth, that 
 would not be a sufficient proof ; we require better arguments. 
 
 We said that light had other properties besides that of 
 enabling us to see. It develops heat, and it acts as a 
 chemical agent. If we move a thermometer under the 
 different coloured rays of the spectrum, we shall see that 
 
28 
 
 SCIENCE OF PAINTING. 
 
 heat increases from violet to red, and passing red it con- 
 tinues to increase, although in the shade, to a culminating 
 point which is to be found left of red, at a distance equal to 
 that existing between red and yellow. Starting from this 
 point heat diminishes, but it is still appreciable at a distance 
 from red equal to the extent of the entire spectrum. 
 
 As for the chemical effects of light, it is impossible to 
 name them all. Let us merely think of those forming the 
 basis of all photographic processes. These may be of 
 great violence, since chlorine and hydrogen, which are 
 without effect upon each other w^hilst in darkness, combine 
 explosively when they are exposed to the sun. 
 
 The chemical power of the coloured rays of the spectrum, 
 ■contrary to those of heat, increases when going from red to 
 violet. Farther than the last violet rays, we finish by per- 
 ceiving another shade, to which the name of lavender grey 
 has been given ; in the same way that beyond the red is 
 now distinguished a beginning of a darker shade named 
 crimson. The truth therefore is that the solar spectrum 
 stretches from the extreme point, where heat begins to be 
 felt, to the other extreme point, where all chemical action 
 ceases ; but to us it is only visible in its central part. We 
 say now visible, because our organs of sight becoming un- 
 ceasingly, not stronger, but more delicate, we already see 
 better than formerly \ and perhaps a day may come when 
 we shall be able to distinguish yet more than we now can. 
 
 With this reasoning we must admit that the ancients had 
 not the power of seeing all the colours which we see. 
 
 Well ! but why not ? 
 
 Certain scholars have already tried to prove that the 
 ancient painters did not use either blue or violet, and that 
 the language of Homer has no words expressive of those 
 colours. Are we to infer from this that there was once an 
 
LIGHT AND COLOURS. 
 
 29 
 
 epoch when these extreme shades of the spectrum were not 
 yet seen ? 
 
 At all events, amongst organised beings there exist some 
 who, gifted with an organ of vision enabling them to 
 perceive light, nevertheless do not see colours, or only see 
 certain colours. Little crustaceae of a particular kind, in a 
 globe full of water, frisk about, and hasten towards a jet of 
 light, if that light is presented to them of a certain colour, 
 and do not move if the light is not of that particular colour. 
 
 The human race presents similar anomalies. Certain 
 people, affected with an eye malady called Daltonism, are 
 unable to distinguish all colours. It has been remarked, 
 since examination has been compulsory to the employés of 
 the railway companies, how great is the number of Dal- 
 tonians amongst those men who present themselves for 
 examination. The proportion is quite as great amongst 
 those who do not present themselves; only these latter 
 generally have no suspicion of their infirmity. Many are 
 artists, and some are even very talented. 
 
 One of them, whom we knew as a student, could not 
 distinguish red from green. To him, vermilion and Verona 
 green were identical. He was guided by the labels on 
 his tubes, and, knowing by hearsay the use of these two 
 colours, he painted moderately well. There certainly were, 
 here and there, some wandering touches which " howled 
 a little, in studio phrase ; that passed for originality. But 
 having one day, by inadvertence, taken the palette of a 
 neighbour who did not arrange his colours in the same 
 order as himself, the secret was discovered. All those who 
 have seen it must still be able to remember the academical 
 figure of the ancient wrestler, conscientiously painted in all 
 the greenest tints of spinach and leeks. One can imagine 
 the burst of laughter w^hich this caused amongst his com- 
 
30 
 
 SCIENCE OF PAINTING. 
 
 rades; it was spoken of for a long time. The poor boy, 
 in desperation, dating from that memorable day, cleared 
 from his palette all reds and brilliant greens, and contented 
 himself with painting subjects not requiring much effect 
 from colours. As he drew very well, had the idea of 
 relative values well developed, and keenly felt the poetry 
 of nature, he became, in spite of all, a great artist ; only 
 no picture ever left his studio without a friend sincere (but 
 indiscreet as you may see) coming to make sure that no 
 mistake had slipped in, and that the few indispensable reds 
 were well in their places. 
 
 We may therefore conclude from the preceding examples 
 that the colour of a body is perpetually modified by all sorts 
 of causes, and that it is not the same to every eye. 
 
 If, therefore, professors, imposing but ignorant, having no 
 knowledge of painting (we must repeat it), dare any longer 
 to wish to formulate the precise colours of objects, may they 
 be the laughing-stock of the lowest of the colour-grinders ! 
 
CHAPTER III. 
 
 THE LAWS OF COLOURING, 
 
 The seven colours given by the decomposition of light 
 seem to melt insensibly into each other ; but a practised eye 
 distinguishes almost a thousand shades in the extent of the 
 solar spectrum. 
 
 The presence has also been ascertained of about two 
 thousand dark rays, which separate those shades into 
 unequal parts. 
 
 All scientists, however, are not agreed about choosing 
 exactly amongst those shades the seven normal colours. 
 
 One considers red as being a little nearer to orange, for 
 instance, and the other considers it to be a little farther 
 off, etc. 
 
 That alone would render impossible the establishment 
 of a normal universal spectrum. It has been proposed to 
 decide with common consent, that the true red, the true 
 yellow, etc., should be between such and such rays, these 
 latter remaining always in the same place. Unfortunately, 
 if the rays are immovable the shades are not ; they change 
 according to the intensity of the light which produces them : 
 when the light increases, all the colours of the spectrum go 
 nearer to the centre, which is greenish-yellow : then the 
 yellow becomes a greener yellow, the orange yellower, the red 
 more orange, the crimson redder ; and on the other side of 
 the centre the green becomes a yellower green, the blue 
 
32 
 
 SCIENCE OF PAINTING. 
 
 greener, the ultramarine bluer, the violet more ultramarine. 
 On the contrary, in proportion as the light diminishes, the 
 colours get farther from the centre : the yellow becomes 
 more orange, the orange redder, the red more crimson, and 
 this last, going on towards darkness, becomes darker. On 
 the other side of the centre, the green becomes bluer, the 
 blue more ultramarine, the ultramarine more violet, and this 
 last Hke crimson, going on towards darkness, gets darker. 
 
 So that, if you decree, for example, that the true red is 
 between such and such rays, as the red moves to the left 
 or to the right according to the intensity of the light, the 
 red will not always remain between those rays, for the rays 
 do not change place. 
 
 It would therefore be necessary to establish some practi- 
 cal means of measuring this intensity of light, so that we 
 might say : it is at such a degree of luminosity that the 
 spectrum should be consulted. 
 
 Whilst admitting that that might be possible, there still 
 remains one objection : it is that light is differently coloured 
 according to the season, the climate or the hour of the day. 
 
 After these explanations it is easy to understand why 
 those who have tried to fix a universal standard have all 
 failed. 
 
 But even if we could overcome the difficulties which 
 oppose the realisation of this standard, what practical results 
 should we obtain from it ? 
 
 You ask a manufacturer for a stuff of a certain shade by 
 this standard ; but according to the material which will be 
 dyed to prepare the stuff ordered so will the shade differ. 
 Yarns of silk, wool or cotton, having surfaces more or less 
 shiny, will reflect back the white light in a small or great 
 degree, and, being twisted or woven in different ways, there 
 will be effects of 7nultiple reflection^ which lighten or darken 
 
THE LAWS OF COLOURING. 
 
 33 
 
 the colour. And if an artist wishes to make use of this 
 standard, if he put colour of the normal tone on his palette, 
 what will be the use of it to him ? As we have seen, the 
 material colours which he uses change their tone according 
 to the manner in which they are used, and according to 
 their surroundings ; without counting the chemical reasons 
 which cause them to vary still more. 
 
 If we are stopping in order to protest against this non- 
 existent colour standard, it is because it is being constantly 
 proposed in special works all very unpractical, as, for ex- 
 ample, in a recent chromatic summary^ where the author 
 gives chromo-typography tables in which all the shades 
 correspond to written formulae. Only, he artlessly tells us 
 that his shades have not come out very well, and, further, he 
 begs us to keep the book shut as much as possible, to prevent 
 his plates fading from exposure to the light. And in this 
 manner he professes definitely to settle colours. 
 
 That^would be very amusing were it not that such volumes 
 written for artists do great harm to those who read them. 
 In fact, their perusal is time wasted, and by always reading 
 without understanding one loses all desire for learning. 
 
 The solar spectritm^ as we have just seen, has only one 
 fixed point. It is a yellow-green. From it we can go to 
 violet, which is the last perceptible colour on the right, the 
 left extreme being a deep crimson. 
 
 As soon as the human mind seeks to penetrate the secrets 
 of nature, it finds always infinitude — that great hope ! 
 
 Infinitude is represented by the figure of a serpent biting 
 his own tail : this is a bad illustration, for the idea of 
 a circle, a closed and imperfectible figure, more properly 
 represents a proud despair. This idea fits in with those 
 which declare that when human sensation stops all is 
 finished ; and, as regards our present subject, somebody 
 
34 
 
 SCIENCE OF PAINTING. 
 
 should think of considering the solar spectrum as a definite 
 whole and uniting the two extremities. This suggestion 
 has not been lacking. It is fortunate that to prove the 
 folly of this conception we have other arguments than 
 philosophical fancies. 
 
 In the first place, the theory of luminous waves does 
 not allow us to suppose that the shortest waves can be 
 immediately joined to the longest ; further, the laws which 
 regulate the mixtures of the colours of the spectrum would 
 not apply at all if it were desired to mix the extremities- 
 Thus violet, returning to the side of red and mixed with it, 
 does not give the hue in which crimson takes its origin. 
 
 It is a very convenient theory that colours are arranged 
 in a circle, but as it is not true it must be renounced. It is 
 like the three primary colours which founded all the others. 
 But that is another mistake. It is true that green can be 
 made with yellow and blue, orange with yellow and red, 
 and violet with red and blue ; but this green, this orange 
 and this violet are darkened, by a phenomenon of interference 
 which it would take too long to explain here. The primary 
 colours would have to be soiled to make them harmonise 
 with the compound colours, and we should have to be 
 satisfied with a circumscribed palette which could have 
 no crimson. Yet crimson exists in nature, and painters 
 cannot dispense with it if they wish to reproduce all they 
 see. 
 
 Having explained why we must renounce, with reference 
 to painting, the study of the mixtures of the coloured rays 
 themselves, we shall therefore begin by treating them as 
 palpable substances, and copy the solar spectrum with the 
 material colours which are at our disposal. Leaving all 
 scientific experiments, we shall collect the phenomena 
 which produce the mixtures of these colours, their super- 
 
THE LAWS OF COLOURING. 35 
 
 positions and their contrasts, under the general name of 
 laws of colouring. 
 
 Mixtures of Opaque Colours. 
 
 Colours prepared to an equal degree of opacity and of 
 luminous intensity will be arranged on a black palette in the 
 order of the spectrum. Certain mathematical coincidences 
 which we find in the mixtures of colours being of use to 
 us, they are numbered so as to distinguish, immediately, the 
 odd numbers from the even, and the fixed point at yellow- 
 green is marked by a cross \ moreover, each space between 
 two colours representing the intermediate shades, is 
 separated into five equal intervals, which divide the spec- 
 trum into thirty-seven degrees. The shades are further 
 subdivisible, as the eye can perceive a thousand of them ; 
 but that would be useless for our demonstration. On left 
 and right, crimson and violet occupy the degree No. 3 of 
 the intermediate shades between the two extreme colours 
 and those unknown. 
 
 . Mixtures of Two Colours, 
 
 1. When two colours come together, their mixtures, in 
 all proportions, produce the intermediate shades with all 
 their luminous and colouring intensity. 
 
 Example : — Red and orange produce all the shades of 
 red-orange and of orange-red; blue and green, all the 
 shades of green-blue and of blue-green, etc. 
 
 2. The mixture in equal parts of two odd colours separated 
 by a single other, produces the colour which separates them, 
 but with less colouring intensity. 
 
 Example: — Red and yellow produce an orange greyer 
 
36 SCIENCE OF PAINTING. 
 
 Crimson . . ——— 
 
 Red . . 
 
 Red-orange 
 
 Orange . . 
 
 Yellow-orange . 
 
 Yellow 
 
 Yellow-green . 
 
 Green . 
 
 Blue-green . . 
 
 Blue . . . 
 
 Blue-ultramarine 
 
 Ultramarine . 
 
 Violet 
 
 I odd. 
 
 2 even. 
 
 3 odd. 
 
 fixed point. 
 
 4 even. 
 
 5 odd. 
 
 6 even. 
 
 Whilst reading what refers to it, you would do well to have a copy of 
 this list. 
 
THE LAWS OF COLOURING. 
 
 37 
 
 than the normal orange ; yellow and blue a green greyer 
 than the normal green. 
 
 3. The mixture, in equal parts, of two even colours 
 separated by a single other, produces a grey somewhat 
 approaching the colour which separates them. 
 
 Exmnple : — Orange and green produce a yellowish grey, 
 green and ultramarine a bluish grey, etc. 
 
 4. The mixture in equal parts of two colours separated 
 by two others, which consequently is always the mixture of 
 an odd and an even colour, produces a grey approaching 
 the odd colour. 
 
 Example : — Red and green produce a reddish grey ; 
 orange and blue a bluish grey ; yellow and ultramarine a 
 yellowish grey, etc. 
 
 But if one doubles the quantity of the even colour, 
 in the proportion of one-third odd and two-thirds even 
 colour, the perfect grey, corresponding to the mixture of 
 white and black, is obtained. 
 
 Example : — 
 
 The complement of red . 
 
 orange 
 yellow 
 green 
 blue . 
 
 No. I is 1 + 3 = 4, green. 
 „ 2 „ 2 + 3 = 5, blue. 
 ., 3 n 3 + 3 = 6, ultramarine. 
 5» 4 5J 4 - 3 = I? red. 
 5 j» 5 - 3 2, orange. 
 
 ultramarine 
 
 6 6 - 3 = 3, yellow. 
 
 To find the complement of an intermediate, we calculate 
 in the same way as for the colours nearest to it, deducting 
 from the found complement the same number of degrees. 
 
 Example: — To find the complement of crimson, we 
 shall seek that of the nearest colour: this is red, of which 
 the complement is green. As crimson is three degrees left 
 of red, its complement will be three degrees left of green, 
 /.^., yellow-green; violet being three degrees right of 
 
38 
 
 SCIENCE OF PAINTING. 
 
 ultramarine, its complement will be three degrees right of 
 the complement of ultramarine, which is yellow : it will 
 therefore be yellow-green. The two extremes have thus 
 for complement the fixed point — which is the centre. If 
 we seek the complement of the red-orange shade placed 
 at the first degree after red, we find it at the first degree 
 after green, etc. 
 
 The MixtiLves of Three Coloicrs. 
 
 1. The mixture in equal parts of the consecutive colours, 
 two odd and one even, produces the even one with less 
 intensity of colour. 
 
 Example: — Red, orange, and yellow, produce a greyer 
 orange than the normal orange, but also less grey than that 
 produced by the mixture of red and yellow ; yellow, green, 
 and blue, produce a green greyer than normal green, but 
 less so than that produced by the mixture of yellow and 
 blue, etc. 
 
 2. The mixture, in equal parts, of three consecutive 
 colours, two even and one odd, produces a grey approaching 
 the odd colour. 
 
 Example : — Orange, yellow and green, produce a yellowish 
 grey ; green, blue and ultramarine, a bluish grey, etc. 
 
 3. The mixture, in equal parts, of three odd colours, 
 produces perfect grey. 
 
 4. The mixture of three even colours produces also 
 perfect grey. 
 
 As a matter of course, results obtained by mixtures of 
 equal portions would be influenced by any colour the pro- 
 portion of which was increased : thus, red and yellow, which 
 in equal proportions produce orange, would produce a red- 
 orange if the proportion of red were increased, and an 
 orange-yellow if the proportion of yellow were increased. 
 
THE LAWS OF COLOURING. 
 
 39 
 
 As for all other combinations beyond those above specified, 
 they give greys more or less coloured. 
 
 The Superposition of Colours. 
 
 If we paint a transparent colour over so thin a white that 
 this foundation white has influence, it is as if we were to put 
 light behind that colour, which is then transmitted to us by 
 transparency instead of by diffuse reflection^ as when it is 
 opaque. The effects obtained by the laying on of colours 
 are more perfect in proportion as the colours used are more 
 transparent. Leaving those which we have just used, as 
 well as the black palette, we shall again copy the spectrum ; 
 but, this time, on a white palette and with colours which 
 we shall render all equally transparent with more or less 
 varnish, and leaving to them equally their colouring 
 intensity. As for the luminous intensity — the light coming 
 from below, produced by the white of the palette — it is 
 perfectly regular for layers of similar thickness. 
 
 The first observation which strikes us is that by this new 
 means we obtain a much greater colouring intensity for the 
 same luminous intensity : i.e.^ when, with opaque colours, 
 we wish to have, for instance, a very light red, we must add 
 to it so much white that we arrive at a pale pink where the 
 sensation of red is almost lost, whilst, by transparency, we 
 obtain at the same degree of light a pink much more vivid, 
 in which the sensation of red is still palpable. 
 
 Another example still will help to show this difference. 
 
 Being given red the most intense possible to make with 
 opaque colour, it would be useless to put more on the top, 
 — it would not render it more intense ; but if you painted it 
 over with transparent red in glazing, you would increase 
 considerably the colouring intensity without diminishing its 
 luminous intensity. 
 
40 
 
 SCIENCE OF PAINTING. 
 
 Superposition of Transparent Colours on White aitd on 
 
 Black. 
 
 A colour painted over white increases in luminous in- 
 tensity in proportion to its being in coatings more or less 
 thin ; but it gets near to the fixed point of the spectrum (i.e, 
 the yellow-green), according to the law that we have already 
 mentioned. 
 
 Example : — In coatings thinner and thinner : 
 
 Left of 
 the fixed 
 point. 
 
 Right of 
 the fixed 
 point. 
 
 c yellow becomes more and more yellow-green. 
 I orange yellow 
 
 \ crimson 
 
 green 
 blue 
 
 ultramarine 
 , violet ,5 
 
 orange, 
 red. 
 
 yellow-green. 
 
 green. 
 
 blue. 
 
 ultramarine. 
 
 On the contrary, if you increase the thickness of the 
 coatings, starting from the point where the colours are at their 
 greatest degree of colouring intensity, they go farther away 
 from the fixed point (j.e, of yellow-green), the luminous 
 intensity diminishing. 
 
 Example : — In thicker and thicker coatings : 
 
 Left of r yellow becomes more and more orange, 
 the fixed -! orange red. 
 point. \ red crimson. 
 
 And as after crimson is darkness it gets darker : 
 
 Right of r green becomes more and more blue, 
 the fixed \ blue ultramarine, 
 point. \ ultramarine ,, violet. 
 
 And violet, going also towards darkness, Hke crimson 
 becomes darker. On black, if it were absolute, a trans- 
 parent colour would not go. On account of its being 
 transparent, it does not send back the light which strikes it ; 
 
THE LAWS OF COLOURING. 
 
 41 
 
 it can only transmit what comes behind it : so, if there be 
 only darkness it can transmit nothing. But, as we mentioned 
 before, there is in nature no body of an absolute black. 
 
 If we divide into a hundred parts the quantity of light which 
 strikes a black screen, the screen reflects about five parts ; 
 yet this small quantity is sufficient to render perceptible the 
 transparent colour painted on the screen : the effects there- 
 fore will be the same as upon white, but considerably 
 darkened. In painting transparent colours upon lighter and 
 lighter greys, you may see the results of this law accentu- 
 ated in proportion as the gradations get lighter. 
 
 Superposition of Transparent Colours upon each other. 
 
 If you paint one transparent colour over another, there 
 are mixtures by transparency and those mixtures follow the 
 same laws as do the opaque colours. If the bottom colours 
 contain white or black, you add to them the results which 
 are given by the superpositions on white and black that we 
 have just been studying. 
 
 Superposition of Translucent Colours. 
 
 Bodies composed of very small minute particles exist in 
 suspension in a transparent gaseous medium either liquid 
 or solid, such as clouds of dust and smoke, the gummy 
 liquid of paste, mother-of-pearl, opals, tortoiseshell, etc. ; or 
 composed of little imperceptible drops of liquid in suspen- 
 sion in a medium, also transparent liquid but of different 
 density, such as all emulsions ; or in a gaseous medium, like 
 all the vapours in the air — mist, dew, fog, etc. 
 
 These bodies, which are called translucid, uniting the 
 properties of transparency and opacity, transmit a part of 
 the light which they receive and reflect a part ; but the part 
 which they receive takes a more orange tint, and the part 
 
 V. p. 'I 
 
42 
 
 SCIENCE OF PAINTING. 
 
 which they reflect a bluer tint. Those two colours, orange 
 and blue, of which tints the light, transmitted and reflected by 
 the translucent bodies, is, are placed in the spectrum, just at 
 the centre of each of the halves separated on the fixed point. 
 
 Example : — Smoke when escaping from the chimney of a 
 hut appears bluish on the dark green of the trees, because, 
 having no light behind to transmit, we only see the reflected 
 light ; but the same smoke appears reddish when it is on 
 the luminous sky. We see it then by transparency. 
 
 A glass of milk seen in transparency is orange ; but if 
 you spill some of it on a dark table, it becomes bluish, 
 because then we only see it by reflection. 
 
 The air itself, holding in suspension solid corpuscles, 
 obeys this law of translucent bodies. The sky is blue, 
 because, the terrestrial atmosphere, being unable to transmit 
 the infinite darkness stretching beyond it, reflects the light 
 of the sun. But turning aside, the atmosphere, transmitting 
 to us its light by transparency, is no longer blue, and in 
 proportion as the radiant orb descends towards the horizon 
 its rays, coming through the translucent coatings of the 
 mists which rise from the earth, are transmitted to us more 
 and more orange as the mists are thicker and thicker ; 
 and quite at the approach of night, when the luminous 
 intensity diminishes, orange, following the laws which we 
 have explained, proceeds towards crimson, which is the last 
 tint of the setting sun before darkness. 
 
 These are also the causes which make the sun appear like 
 an enormous incandescent ball on foggy days. 
 
 In the same way can be explained the blue colour which 
 the skin takes where the veins pass, the bluish tint of the 
 white of the eye, etc. 
 
 Artists who, often without knowing the causes of Nature's 
 phenomena, have yet observed them all, have for a long 
 
THE LAWS OF COLOURING. 
 
 43 
 
 time expressed instinctively this law of translucent bodies 
 by separating their colours into two classes, warm tones and 
 cold tones. Warm tones correspond to all that part of the 
 spectrum left of the fixed point, whose centre is orange. 
 Cold tones correspond to all the part right of the fixed point, 
 whose centre is blue. 
 
 Painters speak the truth when they speak of the green 
 leaves of plants lighted by transparency as of a warm green, 
 and of those lighted from above as cold green. 
 
 Poets are not equally correct. Their famous phrases 
 should be suppressed which speak of the golden sun of the 
 South, the empurpled skies of the East, etc. On the contrary, 
 the nearer the sun is to zenith the whiter its light is, and 
 the farther you go to the regions of the South, the atmosphere 
 there being less charged with vapours, the bluer is the sky 
 and the paler and more intense the light. We must go to 
 the North, to the countries of fogs, to find the warm colouring, 
 the golden reflections and the purple rays. 
 
 Besides, it is in those northern places — in Holland, in 
 Flanders, in England, at Paris — that the greatest colourists 
 have lived. In Italy they are much fewer, except at Venice, 
 which is a misty place. If we go thence south-eastward 
 we do not even find great artists. 
 
 If you add to the material opaque colours a transparent 
 liquid, such as varnish, or if you use them in thin coatings, 
 you render them translucid and they produce the results we 
 have just been describing. Rubbed on dark tones, they 
 approach blue and become colder ; rubbed on light tones, 
 they approach orange and become warmer. 
 
 A colour altogether opaque should have no transparency, 
 even in the thinnest coatings ; and a transparent colour 
 should always remain so, even in a very thick coating. 
 Now, as neither absolute transparency nor absolute opacity 
 
44 
 
 SCIENCE OF PAINTING. 
 
 exist in nature, and as oil and similar necessary mediums 
 render paints more or less translucid, the opacity which we 
 obtain from them is only relative, and it is by the degree 
 of thickness of the coating that we render this quality 
 perceptible. As for transparency, we can, by the help of 
 varnish, arrive more completely at it; but to light our 
 opaque colours in front and our transparent colours behind, 
 we have not the command of light at the greatest degree of 
 intensity which it displays in nature. 
 
 To light a white canvas by the sun we must content our- 
 selves with the diffuse light of a studio or museum reflected 
 by our oil-white, or our paper if it be water-colours that we 
 are using ; and this white absorbs already much light, for it 
 only reflects 40 per cent, of what it receives. 
 
 To represent a pane of stained glass transparently lighted 
 by the sun, as our back-light we are obliged always to use 
 those same feeble whites. In the matter of darkness we 
 are also limited, having only black, which reflects 5 per 
 cent, of the light which it receives, and which consequently 
 is already grey in comparison to an absolutely dark hole. 
 Yet if painters of talent arrive at giving us truthful repre- 
 sentations with those limited means, it is only by using 
 subterfuges — by exaggerating opposites, by sacrificing certain 
 parts and making much of others — that they succeed. In 
 fact, that is the art of painting, and the art is not to be 
 learnt in a book ; but what is to be learnt there is the 
 law of the contrast of colours, which may be made very 
 profitable. 
 
 The Contrast of Colours by Juxtaposition. 
 
 Contrast of Black and White, 
 
 White, placed beside black, makes the latter appear 
 blacker and itself whiter ; but this efl'ect is produced more 
 
THE LAWS OF COLOURING. 
 
 45 
 
 and more strongly the nearer the two are brought together. 
 Consequently, when a white band and a black one touch 
 each other, the part of the white band which limits the 
 black, seeming whiter than the rest of the band, the white 
 stripe appears shaded like a white stick ; in the same way 
 the part of the black band which limits the white, seeming 
 blacker than the rest, the black stripe has the effect of a 
 black stick. A succession of grey and white bands 
 alternately side by side would give the impression of a 
 fluted pilaster. This, applicable to all colours which are of 
 different luminous intensities, will produce each time only 
 two tones, containing each more or less white or black than 
 the other. 
 
 Cofitrast of Colours in Juxtaposition. 
 
 1. When two colours which follow each other in the 
 order of the spectrum are placed side by side, they take 
 more and more, in proportion as they approach each 
 other, the aspect of the colour which precedes or follows 
 them. 
 
 Example : — Red and orange : the red which verges on 
 orange approaches the crimson colour which precedes it ; 
 the orange which verges on red approaches the yellow 
 colour which follows it. Yellow beside green becomes 
 more orange ; green beside yellow more blue ; blue beside 
 ultramarine becomes greener ; ultramarine beside blue more 
 violet ; violet beside crimson more ultramarine ; crimson 
 beside violet redder, etc. 
 
 2. When two colours separated by another in the order 
 of the spectrum are in juxtaposition, the same results are 
 obtained. 
 
 Example :■ — Red beside yellow becomes more crimson ; 
 yellow beside red greener, etc. 
 
46 
 
 SCIENCE OF PAINTING. 
 
 3. When two colours are separated by two others in the 
 order of the spectrum they are, as has been said, comple- 
 mentary ; then they do not change when in juxtaposition, but 
 they intensify. 
 
 Example: — Red beside green increases in intensity — it 
 becomes a brighter red ; green beside red appears a brighter 
 green, as do all the colours with their complements. 
 
 4. When two colours separated by more than two others 
 in the order of the spectrum are in juxtaposition, each 
 approaches the complement of the other. 
 
 Example: — Red a7id blue. Red beside blue verges on 
 orange, complementary to blue it becomes more orange ; 
 blue beside red verges on green, complementary to red 
 it becomes greener. Red and ultramarine. — Red beside 
 ultramarine verges on yellow, complementary to ultramarine 
 it becomes more orange ; ultramarine beside blue verges on 
 green, complementary to red it becomes bluer. — Orange 
 and ultramarine. Orange beside ultramarine verges on 
 yellow, complementary to ultramarine it becomes yellower ; 
 ultramarine beside orange verges on blue, complementary 
 to orange it becomes bluer, etc. 
 
 After these explanations, it is easy to understand how the 
 aspect of a colour may be modified without changing it. 
 
 When you have exhausted all the resources of the palette 
 to make a colour intense, you can still increase its brilliancy 
 by cleverly surrounding it with objects of its complementary 
 colour; if, on the contrary, you find that a colour is too 
 pronounced, you can soften it by surrounding it with objects 
 of the same colour, more intense. 
 
 Example : — An orange drapery can be rendered yet more 
 orange surrounded by blue tones ; a red surrounded by green 
 tones will seem redder, etc. A portrait may have carnations 
 that one thinks rather too red, but does not want to change. 
 
THE LAWS OF COLOURING. 
 
 47 
 
 The normal aspect may be given by a red background : if it 
 is too pale it will become pinker with a green background, 
 etc. 
 
 Nature has at its disposal luminous and colouring intensi- 
 ties which we have not on our palette. We are therefore 
 obliged, in order to imitate the phenomena produced by 
 nature, to exaggerate them. It is therefore necessary to 
 know them. 
 
 Let us resume our former example, where we saw that the 
 smoke of a hut appeared bluish on the dark background 
 of trees, and reddish on the sky. If, after having painted 
 the trees and the sky, we take the uniform tone to represent 
 smoke, we shall not obtain the same result as in nature. Our 
 sky has not the luminous intensity of the real one, and to 
 give the appearance of it we have exaggerated the dark- 
 ness of the trees; we must therefore also exaggerate the 
 blue tone of the smoke when passing over the trees, and 
 its orange tone on the sky. 
 
 If we paint this smoke according to nature, we shall make 
 this exaggeration instinctively ; but if the hut does not 
 smoke that particular day, and we do not know these laws, 
 it is more than probable that we shall carefully paint a smoke 
 false in tone. 
 
 Finally, what we wish to have understood is, that a know- 
 ledge of the laws of colouring facilitates the work of the 
 artist, by enabling him to give reasons for the causes which 
 produce the different effects that are made by colours in 
 nature, in the same way that drawing becomes easier when 
 perspective and anatomy are known. 
 
CHAPTER IV. 
 
 WITH THE SCIENTISTS. 
 
 The laws of colouring, as they have been formulated in the 
 preceding chapter, unite almost all the results made by dis- 
 coveries up to the present, but presented in a form especially 
 suitable for artists ; for men of science have been more 
 inclined to the study of the coloured rays than to that of 
 concrete colours. 
 
 This theory, therefore, has nothing new but the form in 
 which it is presented. Rule without exception, absolutely 
 demonstrated, universally controlled and recognised, it is 
 at last the truth. 
 
 To explain the reason of it is quite another matter. 
 Learned men are not agreed upon this subject, and when 
 "doctors differ" the majority of people find it difficult to 
 form an opinion. 
 
 The best thing would be to wait, but artists are impatient 
 to know what is hidden at the bottom of this mystery, — all 
 the more so that we are enticed by treatises on harmony 
 of colours by following which every one can (at least so 
 they are promised in the prefaces) instantly become a 
 great colourist. This is very tempting. If therefore any 
 one is on the point of plunging into the furnace, let him 
 first listen to this. 
 
 A young artist, anxious to know the reason of everything, 
 prowled unceasingly round the domain of science, that sacred 
 
WITH THE SCIENTISTS. 
 
 49 
 
 grove where the pontiff scientists hide from profane eyes 
 the secrets they discover from nature. One day the young 
 artist finished by penetrating into the mysterious sanctuary, 
 where a tall old man with a keen and cunning look im- 
 mediately came to him. This grave personage carried over 
 one arm skeins of wool of all shades, and in the other 
 hand he held a disc, divided like a cake into parts of all 
 the colours of the rainbow. He spoke in a loud voice : 
 " Young man, the senior student, to-day a centenarian, 
 welcomes you, and knowing the object of your visit, he 
 congratulates you on having found immediately the best 
 guide whom you could follow. My system is marvellous : it 
 explains everything ; it replaces everything. No more of the 
 antiquated names by which colours have been named : we 
 shall hear no more turtledove grey, nymph pink, dead- 
 leaf, or Dauphin brown ; but for example, in such a case, 
 the fifteenth tone of the reduced yellow scale, seventh, 
 eighth or ninth shade, according to whether the Dauphin 
 brown were more or less green." 
 Yes," thought the young man. 
 
 ''And see how simple it is," continued the old man, 
 showing his disc. *' Do you seek the complement of a 
 colour? Exactly opposite, following the diameter, — no 
 error possible ; obligatory harmony for everything. 
 
 " I regulate the toilet of ladies, military uniforms, the 
 arrangement of gardens and salads ; I regenerate the art 
 of the upholsterer, of the glazier, of the artist, and of the 
 decorator. 
 
 " Ah ! this chromatic circle is a work of genius. And 
 mark well, young man, that in writing on these subjects a 
 volume of 730 pages, I have done nothing lightly. I tell 
 you so in my preface. I have submitted each experiment 
 to my pupils, to my friends ; and it is not the judgment of 
 
50 
 
 SCIENCE OF PAINTING. 
 
 my own eyes that I have written, but the average judgment: 
 of many." 
 
 " Pardon," insinuated the young artist ; " you appear to 
 confess by that, that your arrangements of colours may 
 present different effects to different people, in which case 
 your universal system would cease." 
 
 " One moment, young man. Do not try to make me say 
 what I think ; at my age we don't do that. It must be 
 admitted that my system is applicable to every one; with- 
 out that it would be useful to no one. The hypothesis must 
 also be admitted on which it is constructed, of three simple 
 colours, red^ yellow^ and blue ; and of three colours composed 
 by the mixtures of simple colours, orange^ green^ and violet ; 
 otherwise, I could not divide my circle into corresponding 
 parts : nothing would harmonise ; there would be confusion, a 
 mess — if you prefer a mess." 
 
 " No ! no ! only you are suppressing one of the colours of 
 the spectrum." 
 
 " Which, please ? " 
 
 " Well, ultramarine, it seems to me." 
 You say ultramarine ? Yes, what in my time was 
 called indigo : you see it has even changed its name. But 
 it is not really a colour. It is a spurious mixture of blue 
 and violet, the favourite of laundresses, in any case worse 
 than useless, an intruder, a hindrance. No, but do you see 
 it in my circle ? It prevents it from being perfectly round. 
 As well put up my disc at once, and restore the Gobelins 
 to Louis XIV." The noble old man, going away visibly 
 ruffled, exclaimed these last words : " Presumptuous young 
 man, from the moment that you defend indigo, you may 
 choose another guide." 
 
 In fact, the young artist, turning round, found himself 
 before two other personages. 
 
WITH THE SCIENTISTS. 
 
 51 
 
 " I present to you the celebrated Thomas Young, the 
 pioneer whom I am proud to have followed," said one of 
 the two, introducing: his colleague ; " the first half of our 
 common glory." 
 
 And I," said the other, " present to you in my turn the 
 celebrated Helmholtz, my worthy successor, the second 
 half — Well, young man, then you do not adopt the 
 hypothesis of that obstinate old man ? " 
 
 " Our system," continued the celebrated Thomas, " ex- 
 plains everything, replaces everything ! Our three funda- 
 mental colours " 
 
 " Ah ! you have also three colours ? " 
 
 " Yes, but they are not the same : ours are red, green, and 
 violet. We suppose the retina to be composed of nervous 
 rootlets united in threes, and we suppose that each of the 
 three is sensible of one of our three colours." 
 
 " Ah ! you suppose also ! " sighed the young artist. 
 
 But the " worthy successor," understanding the thought, 
 added diplomatically: ''Remark that Thomas had made 
 those suppositions long ago ; I had, therefore, nothing to 
 suppose when I presented that theory whose glory we 
 share. After reading our works you will be doubly con- 
 vinced." And the two men of learning took leave of the 
 young artist, each cordially pressing one of his hands. 
 
 The apprentice of learning, somewhat discouraged at this 
 beginning, sought no more guides, and decided to penetrate 
 the mysterious labyrinth alone, listening here and there, 
 sometimes to the pleadings of David Brewster, the intrepid 
 defender of the three primary colours, and of J. J. Muller 
 and Maxwell, Thomas Young's disciples ; sometimes new 
 theories about the retina containing three visual substances 
 with six fundamental sensations, or even the theory of the 
 waves producing on the retina different compounds according 
 
52 
 
 SCIENCE OF PAINTING. 
 
 to their length; sometimes also, the fantastic lucubrations 
 of demented brains, rules of music applied to painting, 
 chromatic pitch, chromometer, etc. Eve,n the lamentations 
 of the forgotten and ignored, such as Wiirnst, who professes 
 to be the discoverer of Young's theory, and poor Newton 
 who cannot forgive the theft of his disc : " I invented it 
 first," he often said with bitterness; only I made no 
 flourishes to attract loungers ! " 
 
 In short, when our young artist came out of this domain 
 of science, his hair was grey, and he was more than 
 fifty years old. Common Sense was awaiting him at the 
 door. 
 
 You are here at last ! cried he (Common Sense is a 
 companion of childhood that we often lose sight of in life, 
 but who always considers us children). " Have you, at last, 
 adopted a system ? " 
 ^^No." 
 So much the better ! " 
 Yes, but I have made one ! " 
 " So much the worse 1 " 
 
 " Oh ! don't laugh : my system is admirable ; it explains 
 everything, it replaces everything. Here is the principle : 
 bodies, whatever they may be, absorb the light which they 
 receive, up to a point variable in each body, when they 
 become saturated with light ; then only they begin to de- 
 compose it in a certain proportion still variable, and at last, 
 when they can neither absorb nor decompose any more 
 light, they send it back with the coloured rays which they 
 have not absorbed." 
 
 Very good ! that is no worse than anything else ; but 
 are you quite sure about it ? " 
 
 " Well 1 I suppose it : if my hypothesis be not admitted, 
 my system no longer exists." 
 
WITH THE SCIENTISTS. 
 
 53 
 
 Then Common Sense began to laugh — a laugh terrible to 
 hear. 
 
 " How ? " he resumed : " you find all systems bad because 
 they rest upon suppositions, and in order to replace them 
 you invent one more, also a supposition. It is time to 
 beware of that ! " 
 
 " Then, if my system is not good, what am I to say to my 
 colleagues, who are simple enough to expect my revelations 
 
 " The truth." 
 
 " The truth ! Am I then to tell them that, in harmony of 
 colours, there is no absolute rule ? " 
 " Exactly so." 
 
 " That beyond the elementary laws of colours, there is 
 nothing we can apply in all the theories known up till now ? " 
 " Yes." 
 
 That artists should only take advice from nature, and, the 
 better to understand it, should study the manner in which 
 the great masters have interpreted it ? " 
 "Wonderful!" 
 
 " Then, they must be told also that there were magicians 
 called Veronese, Rubens, Delacroix, and many others, who 
 knew more about colouring than any scientist in the world ; 
 for with their colours they have created a language which 
 speaks to the soul, which communicates feeling and life, 
 long before science has even suspected that coloured rays 
 influenced the brain." 
 
 " Bravo ! That really is the truth ! " 
 If this be the case, as well as immediately confessing 
 to my colleagues that I thought to astonish them by my 
 learning, I must tell them that all I bring back from my 
 journey to the scientists is limited to this advice : Don't do 
 as I have done, don't lose your time in seeking that which 
 does not exist ! " 
 
54 
 
 SCIENCE OF PAINTING. 
 
 " Well ! but acknowledge it without blushing — at your age 
 one may still acknowledge — this advice will be valuable, 
 should it serve to spare others the worry you have given 
 yourself, and that will restore a little the inflation of pedantry 
 of which I have just seen you deprived. The pedant, do 
 you see, is detestable because he lacks charity. He thinks 
 in his vain egotism : ' I have grown pale over volumes, I 
 have searched dictionaries, I have scrawled notes, I have 
 dissected, analysed, and that for years, to discover some- 
 thing ; and this something, I would go to tell in two words 
 to others, who w^ould thus know it without trouble. Ah ! 
 no, let them learn first to understand the noble phraseology 
 of science, for I do not speak a language comprehensible to 
 the vulgar crowd : I call a colour a pigment ; a paper horn 
 a spiral cone ; and the good God, Lord Archetype or 
 Demogorgon ! Afterwards the converts will swallow my 
 prefaces, the recital of my unsuccessful theories, my refuta- 
 tions to my adversaries. 
 
 " ' I will present problems without giving the solutions of 
 them, I will develop hypotheses which I shall afterwards 
 destroy, I will scatter Latin and Greek quotations without 
 translation, archaeological notes supported by documents, 
 etc., etc., and the useful substance I shall only give out 
 little by little, bits at a time. If I ever see a colleague 
 engaged in a search which I have followed, and which leads 
 to nothing, I will let him have his turn at it, let him get 
 embarrassed as I did, let him wander as I wandered, even 
 if he should perish where I was able to pass. Ah ! but it 
 would be too convenient then ; they would only have to 
 fold their arms, and to assimilate science from others with- 
 out seeking it themselves ! ' Dull pedant ! you can never be 
 covered with enough contempt ; but with how much respect, 
 honour, and love, may be hallowed the memory of those 
 
WITH THE SCIENTISTS. 
 
 55 
 
 true scientists who work throughout their Hfetime in order 
 to leave to the world a single phrase which may be written 
 in the book of human science ! 
 
 " But you make me forget my duties," cried Common 
 Sense, interrupting himself ; to-day I must preside at an 
 assembly of artists ; I have been absent from those assem- 
 blies several times, — they will begin to think that I don't 
 wish to appear there any more." And he went away. Let 
 us hope that we shall meet him again ! 
 
CHAPTER V. 
 
 COLOURING SUBSTANCES. 
 
 In taking, to determine the laws of colouring, the compounds 
 of colouring-matter which may best represent the free 
 colours of the solar spectrum^ we have made a theoretic 
 palette ; but the exclusive use of that palette would not be 
 practicable for painting. 
 
 In the first place, the substances which compose them 
 are not all solid, and some of their mixtures would produce 
 chemical reaction capable of changing them. Further, those 
 colours are not found ready made, and their manufacture, 
 which could scarcely be counted upon with absolute cer- 
 tainty, would be expensive. 
 
 Then why should we deprive ourselves of earths, ochres 
 and marls which are quite sufficient for broken tones 
 frequently wanted, and which join to a perfect fixedness the 
 precious qualities of covering well and of drying easily ? 
 
 Do not let us banish, from prejudice, any colouring-matter 
 from our palette ; for, as M. de la Palisse might have said, 
 "All that can be used are useful," always provided that they 
 are substantial, and that is what we have now to consider. 
 
 Ancient artists, in the time of Apelles, had only four 
 colours : chalk white, yellow ochre, red ochre, black. 
 
 Are we quite sure of that ? Pliny, who came some cen- 
 turies later, affirms it, and he tells us further that in his time 
 painting was already enriched by a large number of new 
 
COLOURING SUBSTANCES. 
 
 57 
 
 colouring-matters, whose nomenclature is here re-estab- 
 lished in modern language : — 
 
 Chalk white, variously produced. White lead, and its derivatives. 
 Massicot and minium orpiment (sulphur of arsenic, red, and 
 yellow). 
 
 Red lac — lakes of purple colours, prepared from shell-marl. 
 Natural and burnt ochres. ■ 
 Cinnabar (native vermilion). 
 Indigo. 
 
 Blue pulverised enamels. 
 
 Verdigris. 
 
 Erown earths. 
 
 Various blacks prepared from the combustion of different bodies, 
 
 e.g.^ ivory, grapestone, etc. 
 Sepia. 
 
 And Pliny laments this profusion of colours, regretting 
 the time when there were only four of them. He cries : 
 " To-day, when purple' even covers our walls, when India 
 procures to us the coloured sand from its rivers and colours 
 drawn from the blood of its dragons and its elephants, we 
 have no longer noble paintings. Then, when we were poorer 
 in materials, we really were richer in art. It is no longer 
 soul that one paints, it is the luxury of personages. It is 
 the material that is now appreciated in art." 
 
 Were it not that we know that Pliny wrote that more than 
 eighteen hundred years ago, we might believe that we were 
 reading an article of yesterday. Yes ! if art be always young, 
 we may say that criticism has always been old. 
 
 In spite of Pliny, the domain of painting continued to 
 increase. 
 
 Means of preparing yellow lakes from buckthorn, and red 
 lakes from cochineal and certain woods, and finally madder, 
 were discovered after his time. 
 
 For blue lapis was used, and the palette was enriched by 
 
 new terre-vertes and browns. 
 
 v,p, 4 
 
58 
 
 SCIENCE OF PAINTING. 
 
 However, chemistry did not yet exist. Painters, confined 
 to the productions of nature, cared only to procure them as 
 pure as possible. 
 
 In a barn of Antwerp Museum there exists a trunk which 
 belonged to Rubens, carefully preserved, and in which he 
 had stored away collections of pigments gathered during 
 his travels. Besides the inestimable value attached to this 
 souvenir of the artist, this trunk is for us a valuable lesson. 
 One learns by it the personal trouble which the great artists 
 took in selecting their pigments, and the authentic specimens 
 this trunk contains permit us to learn exactly the materials 
 with which the pictures we so much admire to-day were 
 painted, and to choose amongst them those which are best 
 preserved. 
 
 White lead, cinnabar (native vermilion), lapis, charcoals, 
 madder lakes, earths and ochres have resisted very well ; but 
 buckthorn, like all yellows, reds and vegetable greens, has 
 more or less disappeared. Those results should be noted. 
 
 Since the time of Rubens, many new colours have been 
 invented, and science constantly furnishes new ones ; unfor- 
 tunately more attention is bestowed upon their brilliancy 
 than upon their solidity. 
 
 The discovery of aniline amongst others has been quite a 
 catastrophe for art. 
 
 It is true that as soon as a colour is reputed bad, con- 
 scientious artists dismiss it from their palette ; but it quickly 
 reappears disguised, like swindlers who change their name 
 to make new dupes, and it becomes very difficult to recognise 
 those bad materials in the clever mixtures in which they are 
 concealed. 
 
 So, from fear of being deceived, certain artists fall into 
 an excess of distrust which prevents them from using new 
 productions that they might safely employ. 
 
COLOURING SUBSTANCES. 
 
 59 
 
 Without knowing enough of chemistry, they know too 
 much of it and take fright at certain names. Thus, know- 
 ing the misdeeds ^of chrome yellow, they hesitate to use 
 oxide of chrome (emerald-green), one of the best colours 
 existing, because they do not know that the most terrible 
 substances may become inoffensive when they are mixed. 
 For example, pour into one glass some muriatic acid 
 and into another glass a concentrated solution of caustic 
 soda ; then, into each of those glasses throw little pieces 
 of meat : they will be instantly dissolved under your eyes. 
 Then, mix the contents of the two glasses by pouring the 
 one into the other and — swallow the whole ! It will taste 
 just like a glass of sea-water, nothing more. Those two 
 rightful corrosives are transformed into chloride of sodium, 
 which is merely common kitchen salt. Chlorine is another 
 substance that one is still desirous to see amongst pigments, 
 for everybody knows it destroys almost all other substances. 
 When it is alone, yes ! — but with others it is often perfect, — 
 like a policeman, implacable in uniform, but tender and 
 gentle in the bosom of his family. 
 
 It will therefore be preferable, instead of studying pig- 
 ments separately under names which are not real, and 
 allowing ourselves to be guided by prejudices accompanying 
 these names, to group the pigments according to their deri- 
 vation, and to consider their general qualities and faults. 
 
 We can thus establish five categories of colours : 
 
 First Category : Vegetable colours. 
 Second Animal 
 Third Mineral ,, 
 
 Fourth Colours obtained by vegetable and 
 
 mineral combinations. 
 Fifth Charcoals. 
 
 The ^rst category includes coloured substances procured 
 
6o 
 
 SCIENCE OF PAINTING. 
 
 directly from vegetables, fresh, dried or torrefied, such as 
 iris green, chicory brown, coffee brown, indigo, gamboge, 
 saffron yellow, etc. All the colours of this category are 
 bad : they fade from exposure to light, some of them 
 change tone on contact with fat bodies, and others are 
 frequently destroyed by mineral combinations with which 
 they are mixed. 
 
 The second category includes coloured matter procured 
 from animal substances, such as purple, Indian yellow, 
 carmine of cochineal, sepia, etc. 
 
 Purple, which was produced from shell-marl, is now only 
 known by name, and the recipe for its manufacture is lost \ 
 but now a similar product is obtained from the purpurate of 
 ammonia obtained from uric acid. 
 
 It is said that Indian yellow is obtained from the excre- 
 ment of the chamois or of cows fed on certain plants ; in 
 any case it is a very alkaline substance, which changes oil 
 into soap, and may become, consequently, soluble in water, 
 if the "grinding is not carefully done. 
 
 Carmine fades in light. 
 
 Sepia is solid in water-colours, but does not grind well 
 with oil. 
 
 The third category includes all combinations of metallic 
 bases, whether natural or obtained by chemical processes ; 
 some are to be found in a free state, like carbonate of 
 lead (white lead) and sulphide of mercury (cinnabar), whilst 
 others are more or less firmly fixed in coatings of clay to 
 form the coloured earths to which we give the name of 
 ochres, or in silicious matters to form stones like malachite 
 and lapis. 
 
 All these mineral colours are generally solid, but they are 
 not of a good homogeneous tone ; and although after being 
 powdered, they may be washed with great care, one can 
 
COLOURING SUBSTANCES. 
 
 6i 
 
 never quite succeed in getting rid of the impurities which 
 they contain. This is detrimental to their brilliancy. 
 
 It will be understood, therefore, that in re-making by the 
 aid of chemistry the same combinations met with in nature, 
 and by establishing them on pure and uncoloured substances 
 like aluminum and silica, which are the bases of natural 
 clays, we obtain those same colours quite as solid and much 
 more beautiful. It is thus that marls reproduce all the red 
 and yellow earths coloured by oxide of iron, and that Guimet 
 ultramarine is the reconstitution of lapis-lazuli as vermilion 
 is that of cinnabar. 
 
 Chemistry does not merely confine itself to reproducing 
 combinations which may be found in a natural state ; 
 chemistry has already given us many others, and it is to be 
 hoped that it will continue to enrich this category of mineral 
 colours to which it has already added so much precious 
 matter. It is true that on the other hand it has given us 
 some useless and even bad, which we must be careful not 
 to use : we shall therefore devote a chapter to the practical 
 means of recognising them under whatever name they may 
 be called, and in whatever mixture they may be introduced.* 
 
 The fourth category includes all vegetable and mineral 
 combinations whose principle consists of fixing on aluminum 
 or silica vegetable dyes dissolved in water by means of 
 precipitation. 
 
 In this way are obtained red lakes, with dyes of madder, 
 Brazilian wood, Pernambuco wood, Campeachy wood, etc.,- 
 and the yellow lakes with dyes of woad, Avignon berries, 
 buckthorn, etc. These last often bear the primitive name 
 of stil de grain. 
 
 All the colours of this category are bad ; they fade in the 
 light, and are changed by contact with certain mineral 
 * See Appendix — Verification of Colours." 
 
62 SCIENCE OF PAINTING. 
 
 colours ; they should not be used. We must here, however, 
 except the madder lakes, which are comparatively solid, and 
 which are indispensable, because we have no others. 
 Besides, by taking certain precautions, we can minimise the 
 inconveniences. As for yellow lake, it can be advantageously 
 replaced by iron lake, which, in spite of the diffidence natural 
 to an inventor, we recommend as being of perfect solidity. 
 
 The fifth category includes charcoals produced by the 
 calcination of certain vegetable or animal matter, such as 
 peach kernels, wood, cork, vine shoots, chestnuts, ivory, 
 bones, etc. 
 
 This category furnishes only blacks, all of which are good. 
 
 We have only to reject bistre and lampblack, which are 
 bad on account of the tarry matters contained in them.* 
 
 Bitumen, which we have not classed with mineral colours, 
 will not find place with charcoal either, for it must be 
 formally excluded from painting. It is the plague, the death 
 of pictures. It never dries thoroughly, or at least it softens 
 and runs as soon as the temperature rises above 30° to 35° 
 per cent. By mixing it with a very siccative oil, you may 
 give it an appearance of solidity, but, sooner or later, it is 
 certain to run and exude from the canvas. It is all the 
 more destructive for being used to prepare foundations ; 
 and pictures thus painted upon foundations which are always 
 soft, must, of necessity, crack. 
 
 To give examples caused by ravages owing to the use of 
 bitumen, it is sufficient to refer to the canvases in our 
 museums which it has already been necessary to restore 
 several times, such as the " Shipwreck of the Medusa," by 
 Gericault, the Portrait of Cherubini," by Ingres, and so 
 many others, alas ! 
 
 * It is possible by careful preparation to get a lampblack free from 
 tar. 
 
COLOURING SUBSTANCES. 
 
 63 
 
 But the fanatic lovers of bitumen answer you : " That is 
 not the fault of bitumen, it is the manner in which it is 
 used." Here, then, is another example. A well-known 
 picture dealer, who is also an intelligent collector, has suc- 
 ceeded in collecting a large number of palettes which had 
 belonged to artists of renown. These palettes were taken 
 after the day's work of the artist was over, and not only 
 were they covered with the pigments used, arranged in the 
 order habitual to the artist, but they still showed all the tones 
 improvised whilst actually painting, and on many of them the 
 artist had painted a little bit, according to his own method. 
 We have thus many specimens of almost all the methods 
 in which bitumen can be used. Those palettes as well as 
 pictures are hung on the wall. They are taken good care of, 
 as you may suppose, and even if they are subject to sudden 
 changes of temperature, those changes are common to all. 
 Well ! all the palettes having no bitumen are in perfect 
 condition, exactly as they were left by the artists ; but in all 
 cases where bitumen has been used, the bitumen has run ; 
 not merely the small quantity arranged in its place with the 
 other colours, but also the tones mixed with it. Is this con- 
 clusive ? At least the bitumen enthusiasts cannot pretend 
 that no artist knows how to use bitumen. This unique 
 collection will be bequeathed by its proprietor to the Louvre 
 Museum, and it will be a great lesson for the future. 
 
 Some artists pretend to have substantial bitumen ; they 
 either deceive themselves, or are deceived. Colourmen 
 give them a false bitumen made with yellow lakes and 
 aniline black. Those false bitumens fade in the light. 
 Real bitumen also fades. 
 
 Hov/ever, if bitumen is essential to certain artists, it is 
 quite possible for them to make a bitumen with substantial 
 pigments united with normal resin soluble in oil, and 
 
64 
 
 SCIENCE OF PAINTING. 
 
 not melting at a low temperature. This imitation bitumen 
 will have all the qualities of the real bitumen in the way of 
 convenience, and will not have its disadvantages.* It will 
 have the additional advantage of not fading in light. As for 
 mummy-umber, that is still bitumen, and the few particles 
 of Pharaoh which it may contain must not induce us to 
 accept it. 
 
 We give (Appendix — " Good and Bad Colours ") detailed 
 explanations on good and bad colours ; but for those whom 
 these details may tire, here is the list of the colours which 
 may be used with perfect certainty : — 
 
 White lead (carbonate of lead) . 
 
 Zinc white (oxide of zinc). 
 
 Cadmium yellows (sulphide of cadmium). 
 
 Strontia yellow (chromate of strontia). 
 
 Zinc yellow (chromate of zinc). 
 
 Iron lake (oxide of iron on aluminium basis). 
 
 Vermilion. (sulphide of mercury). 
 
 Madder lakes (madder dye on aluminium basis). 
 
 Cobalt blue (oxide of cobalt on aluminium basis). 
 
 Emerald green (oxide of chrome). 
 
 Mineral violet (phosphate of manganese). 
 
 Cobalt violet (phosphate of cobalt). 
 
 Further, all the natural and burnt ochres, all natural and 
 burnt earths having oxide of iron for basis, are equally good, 
 as well as all the marl colours, which are oxide of iron at 
 different degrees of calcination on aluminium basis. As 
 for blacks, they are all good except lampblack, containing 
 some tarry matters. 
 
 From this list we exclude all colours on lead bases, such as 
 chrome yellows (chromate of lead), Naples and antimony 
 yellows, massicot and minium, which are white lead more or 
 less calcined, because those metallic combinations are liable 
 
 * See Appendix — " Brown Colours.'' 
 
COLOURING SUBSTANCES. 
 
 65 
 
 to get black when brought in contact with the air, and they 
 may be replaced by others ; still they may be used if certain 
 precautions are taken, and also Veronese green and certain 
 other colours. (See Appendix — "Yellow and Orange 
 Colours.") 
 
 If we have not rejected white lead itself, it is because 
 there is no equivalent for it known up till now : its covering 
 and drying qualities, and the hardness which oil imparts to 
 it, make it indispensable. Still it is very dangerous to use it 
 with sulphides such as vermilion and cadmiums, with which 
 we consider nothing but zinc white should be mixed. For 
 cadmiums this precaution is useless if they are well manu- 
 factured. 
 
 This question of the chemical solidity of colouring-matters 
 is excellent for the preservation of painting j unfortunately 
 artists will never give it their personal attention ; many of 
 them will not even read the chapter in which we shall give 
 precautions to be taken for the manufacture of pigments 
 and their use, as well as the practical means of analysing 
 them. (See Appendix — " Good and Bad Colours.") 
 
 Colourmen should therefore be forced to furnish good 
 materials, and here is a means which we consider excellent 
 for obtaining this result. 
 
 Artists should insist, when they buy a tube of colour, 
 that the ticket bears, in addition to the usual name of the 
 colour, its chemical formula according to the above list. 
 
 In this way, if the colourman does not furnish what he 
 professes, he can be summoned for fraud ; there is deception 
 about the quality of the merchandise. Whereas now we 
 cannot even complain, because the names Capucin lake. 
 Geranium lake, China rose, given to all sorts of concoctions, 
 profess nothing, any more than golden yellow, malachite 
 green, Venetian red, etc. 
 
66 
 
 SCIENCE OF PAINTING. 
 
 For instance, you buy a tube of celestial blue : this blue 
 is made of Prussian blue, or of anything else ; it is of no 
 consequence to anybody ; the colourman may call any- 
 thing he pleases celestial blue. You evidently consider it 
 sufficiently celestial since you have taken it. Yet if artists, 
 seeing that this celestial blue is not substantial, should cease 
 to ask for any more of it, it would discard its name and 
 reappear, with more or less white, under the names of azure 
 blue, peacock blue, turquoise blue, sapphire blue, Smyrna 
 blue, etc. We are still far from having exhausted all the 
 names of precious stones, flowers, towns, birds, of which the 
 colourmen's vocabulary is composed; and if the reform 
 which w^e propose should have no further result than to 
 render useless all these fantastic appellations, it would even 
 then do good service, for the number goes on increasing in 
 ridiculous proportions. If we do not stop soon, there will be 
 thousands of those new shades, which are, as a rule, merely 
 common pigments of bad quality dyed and done up with 
 aniline products, all fading in the light. 
 
 From a technical point of view, it is also puzzling if 
 different people give to the same thing different names, or 
 the same name to different substances. Thus, an author in 
 whom you have every confidence asserts to you that ver- 
 milion is solid ; you therefore make use of it, but instead of 
 the sulphide of mercury recommended to you, it is iodide of 
 mercury or dyed minium or anything given to you under 
 this name of vermilion : not merely do they sell to you at the 
 rate of 15 francs what is worth 40 sous ; but they compromise 
 your work. It is a swindle and an abuse of confidence, 
 which could not occur if you bought your colours under their 
 true names. Ask a druggist for an anti-fever powder, and 
 he may give you a little lentil flour sweetened ; but if you ask 
 for sulphate of quinine, he is obliged to give you some, as 
 
COLOURING SUBSTANCES. 
 
 67 
 
 the colourman will be forced to give you whatever may be 
 stated by the ticket on his tubes. But, some one will say, 
 this colourman is himself deceived, for often he does not 
 manufacture the colours he sells. In that case, when buy- 
 ing his primary materials, let him exact from his furnishers 
 the guarantees which his customer requires from him, and 
 you may be sure, should the subject ever be one for litiga- 
 tion, he will know how to make the true perpetrator of the 
 fraud responsible. 
 
 The only serious objection which could be brought 
 against it is this : you cannot force colourmen to make their 
 tickets as you ask. That is true, but neither can you force 
 artists to buy there rather than elsewhere ; and it will be 
 enough for one colourman to begin, for all the others to 
 be obliged to do as much under penalty of losing their 
 customers, for to refuse to give the guarantees given by a 
 colleague would be to confess intentions of inveterate fraud. 
 
 Further, this guarantee of chemical formulas, which 
 might be immediately obtained, is not the only one which 
 we might ask : in a future perhaps nearer than we think, 
 we shall have many more guarantees, and if the project 
 which we have submitted to the " Society of French Artists " 
 is accepted, the members will no longer have anything to 
 desire with regard to the purity and good fabrication of the 
 productions they require. 
 
 This is the style of project : — 
 
 I. That a permanent commission on the material 
 processes of the arts should be named by the committee. 
 It would be composed of members of the society — artists, 
 sculptors, engravers and architects ; whilst all branches of 
 art having, from a practical point of view, problems to solve 
 and advantages to draw from the work of this commission, 
 are interested in being there represented. 
 
68 
 
 SCIENCE OF PAINTING. 
 
 Members foreign to the Society of French Artists " — 
 chemists, manufacturers, etc., whom their capabilities would 
 render useful — could, be added to the commission by a 
 temporary or definite title. 
 
 The duty of this commission would be to study all inven- 
 tions and all processes, both ancient and modern, and to 
 indicate, in special reports, those which might seem to them 
 preferable ; to receive and to class communications made 
 to them ; to reply to questions which might be addressed 
 to them ; in short, to centralise all that might be said or 
 written on the matter. 
 
 The commission would have the power of preparing all 
 treaties and conventions with manufacturers and dealers 
 in products destined for the arts, which treaties and agree- 
 ments would not bind the society until after having been 
 regularly accepted by the committee. 
 
 2. There should be added to the monthly bulletin of the 
 society a rubric the direction of which should belong ex- 
 clusively to the said commission, who might introduce into 
 it all communications, correspondence, articles scientific or 
 otherwise (even when emanating from collaborators foreign 
 to the society), should their publication be found useful. 
 
 3. There should be founded a laboratory where the 
 society would maintain a chemist charged with studying the 
 questions'^submitted by the said commission, and with making 
 all analyses required by the artists, manufacturers or dealers. 
 The analyses might be charged at prices which would enable 
 the society to refund its expenses. 
 
 4. The treaties between the society and the dealers or 
 manufacturers should be made upon the largest possible 
 bases — the society would grant no monopoly to any one, 
 and would exclude no one. The society would be satisfied, 
 abandoning all idea of commerce on its own account, to 
 
COLOURING SUBSTANCES. 
 
 69 
 
 authorise absolutely without fee the said dealers and 
 manufacturers to put a mark determined by the society and 
 belonging to it, upon those of their products which the 
 society had recognised as good, after the decisions of the 
 said commission on the material processes. 
 
 Each dealer or manufacturer, on depositing the sample of 
 the product which he desired to have stamped, would bind 
 himself in writing, under penalty of a heavy fine in case of 
 non-fulfilment, to manufacture always a product identical with 
 the sample, which would be guarded under double seal to 
 bring forward as a proof should litigation ever become 
 necessary. 
 
 Artists, being afterwards informed that the products 
 stamped by the society are guaranteed of pure matter, of 
 good manufacture, and in no way injurious to the preserva- 
 tion of their works, will have the right, it will even become 
 their duty, to insist that their furnishers submit their goods 
 to this stamp, as it will be refused to no deserving person. 
 Let no one say that this project of treaty with manufacturers 
 is a chimera ; for we can name already the house of Lefranc 
 & Co., which is quite disposed to be the first to sign. 
 
 As for the necessity of assuring the greatest possible 
 durability to our works, some critics sneer at this as at 
 a ridiculous pretension, and some artists, through false 
 modesty, consider it may be neglected ; but we are of 
 opinion that it is only honest, and that honesty in everything 
 is the best policy. 
 
 In fact, as we sell our pictures, we make of them mer- 
 chandise, and all merchandise which deteriorates and loses 
 its value in the hands of the acquirer is soon discredited, — 
 all the more so that to-day the prices at which amateurs 
 acquire works of art give them the right to be exacting. 
 
 In letting it be supposed that modern painting can be 
 
70 
 
 SCIENCE OF PAINTING. 
 
 unsubstantial, it is not only to ourselves that we should be 
 doing wrong, it would be also to our successors. Yet it 
 must be so. But as we cannot know what the future alone 
 can demonstrate, it is well to let the fastidious care which 
 we take be known, and also the guarantees with which we 
 surround ourselves in the choice of the materials that we 
 use. 
 
CHAPTER VL 
 
 OILS, GRINDING OF COLOURS, ESSENCES AND 
 SICCATIVES, 
 
 The colouring-matters being chosen, and offering the best 
 guarantees of solidity, we have now only to grind them with 
 an agglutinative of some kind, so that, reduced to the smallest 
 possible particles, they may be in a condition suitable for 
 whatever style of painting they are destined. 
 
 We shall speak of gums, wax and other agglutinative 
 substances used for this purpose when we shall study the 
 different processes of wall-painting, water colours, distemper 
 painting, pastel, etc. 
 
 For the present, we shall only consider oil painting, which, 
 being the furthest developed and the most generally used, 
 is entitled to the first place. 
 
 Oils. 
 
 The name of oil has been given to many liquids which 
 are neither of the same composition nor possessing the same 
 properties. 
 
 I. The essential oils are extracted by distillation, from 
 certain plants, such as rosemary, lavender, fennel, etc., or 
 from certain resinous balms, such as turpentine. Some are 
 also obtained merely by pressure, like the essential oil of 
 lemon. Now the name of essential oil is fallen into disuse; 
 those products in general are called essences. Essences 
 dry by evaporation, but they leave a residue more or less 
 
72 
 
 SCIENCE OF PAINTING. 
 
 viscous ; some are used in painting, and we shall notice them 
 later. 
 
 2. Empyreumatic oils are produced by the analysis of 
 certain bodies by heat, such as oil of wax, oil of camphor, 
 etc. Nearly all the resins burnt emit vapours which con- 
 dense into empyreumatic oils. Bistre and certain tars are 
 empyreumatic oil of wood mixed with charcoal. These oils 
 also dry by evaporation, but very slowly, and they leave yet 
 more viscous residue than the essences, so they are not 
 used, except oil of wax, an attempt to utilise which was 
 made about 1830, but in a kind of encaustic painting with 
 a basis of wax and resin : the author who speaks of it merely 
 advises timidly the trial of it with oil paints. 
 
 3. Animal oils are obtained chiefly by boiling the feet 
 and hocks of animals, such as neat's-foot oil, sheep's-foot 
 oil, etc., and they are also found in a natural condition under 
 the skin of certain fish, such as whale oil. They are not 
 siccative, and are of no use for painting. An oil is also 
 prepared from eggs, either by pressing the yolks hardened 
 in boiling water, or by putting the yolks into sulphuric 
 ether, which dissolves the oil, that is afterwards collected 
 whçn the ether evaporates. This egg oil is not yet of any 
 use ; but it is well to know of it, because it is certainly one of 
 the principles of egg-painting, so much used formerly, and 
 which has given such substantial results. This oil does not 
 corrupt, like other animal oils, and perfectly dissolves cold 
 resins, with which it forms varnishes comparatively hard at 
 the end of some time, although used alone it is not siccative. 
 
 It is not in short used ; but perhaps some day it will be. 
 
 4. Fixed oils are procured by means of pressure from 
 different vegetable matters which contain them. They are 
 not all suitable for painting, and some are not siccative — such 
 as olive oil, for instance. 
 
OILS. 
 
 73 
 
 Those which dry present a peculiarity distinguishing 
 them from essential and empyreumatic oils; they do not 
 evaporate, on the contrary they increase in weight by 
 drying, and this increase is due to the quantity of oxygen 
 they absorb ; but, whilst increasing in weight, they diminish 
 in volume. In fact, if you put a certain thickness of oil into 
 a plate, at the end of some time a pellicle is formed whose 
 surface remains perfectly soft. Yet when the oil gets 
 thoroughly dried the surface is lined, because the fresh oil, 
 shut under the pellicle as under a skin, coming to dry in 
 its turn and diminishing in volume, the same phenomenon 
 it produced as when a balloon gets disinflated or a person 
 gets thin, the skin becomes too large and forms folds. 
 
 The term " to dry," applied to oil which passes from a 
 liquid to a solid state is not appropriate, as nothing evapo- 
 rates — quite the contrary. In reality, there is condensation 
 and absorption of oxygen ; but what happens with this 
 oxygen ? Is there a combination ? Is a new body formed ? 
 In any case the solidified oil no longer dissolves in any of 
 the solvents of fresh oil, except benzine. And, on the 
 contrary, the alcohols which do not dissolve fresh oil dis- 
 solve dry oil. 
 
 There has therefore been at least a serious modification 
 in the nature of the oil to produce such great differences. 
 
 Amongst the oils now used for painting there are two 
 which are generally preferred : linseed oil, and poppy oil. 
 
 Linseed oil is the more siccative of the two ; it acquires 
 greater hardness and remains more transparent whilst drying 
 than poppy oil ; but it is more viscous and gets more easily 
 sour. In this last case, chemical combinations operate in 
 the tubes with certain colours, especially those having an 
 aluminium basis — such as lakes, for instance, which become 
 like india-rubber. . It is said of pigments arrived at this 
 
74 
 
 SCIENCE OF PAINTING. 
 
 condition that they are greasy, and can only be used when 
 diluted with oil or essence ; but they should not be used. 
 They should be entirely discarded, because they will never 
 dry thoroughly. The use of colours ground with acidified 
 oils is one of the principal reasons why pictures crack. 
 
 We should always be careful, before grinding the colours, 
 to assure ourselves by means of litmus paper that the oil 
 is not sour; which often happens after submitting it to 
 operations said to be for purifying it and rendering it 
 siccative. 
 
 The Grinding of Colours. 
 
 In the process of oil painting, colours are used at different 
 degrees of opacity and transparency ; we must therefore give 
 to each of them one of those qualities, in the proportion 
 which suits it best, and it is by an intelligent and sensible 
 grinding that this result can be obtained. 
 
 Unfortunately, most of the people who grind colours for 
 artists are merely concerned with the commercial side of 
 this industry. To prepare colours which shall keep fresh in 
 tubes the longest possible time and in all climates, such is 
 their principal object ; but this object, good for the foreign 
 market, is not so for us. If, unlike the ancient artists, we 
 do not ourselves grind the colours, we should at least be 
 capable of doing so, in order to know if those to whom we 
 entrust the important care of preparing our colours do it 
 properly. 
 
 People are not agreed as to whether ancient artists 
 preferred to paint with linseed oil or poppy oil, and it is 
 now scarcely possible to clear up this mystery : now only 
 poppy oil is used, except for some dark colours, where the 
 use of Hnseed oil is extolled. We shall not trouble the 
 grinders with this subject, as it is really of little importance ; 
 
THE GRINDING OF COLOURS. 
 
 75 
 
 and since poppy oil is preferred as rendering the colours 
 less viscous, let us have poppy oil ! 
 
 But what we cannot admit is the exaggerated quantity 
 used — generally one-third more than really necessary. Now, 
 as oil is pernicious for painting, because it darkens it and 
 changes fatally with time, no more should be introduced 
 than just sufficient to obtain solidity. 
 
 Many artists, convinced of this excess, have adopted the 
 custom of leaving their colours some minutes on blotting 
 paper before using them. It would surely be more practical 
 not to put so much in than to take it out afterwards. But 
 the grinders allege that the colours would then be too dry 
 to grind, that a great many of their customers never find 
 them sufficiently liquid, and at last the eternal bad reason 
 that they have always done so. The truth of the matter is 
 that colours ground liquid keep longer, and then — oil costs 
 less than pigment, and the more there is of the one the 
 less there is of the other. Only the colours are so thickly 
 charged with oil that they would run on the palette if wax 
 were not added to restore some body to them ; the more 
 wax the more oil may be put : one might even do without 
 any pigment at all. With a few grains of aniline dye, you 
 may dye more than a pound of a gluten made of wax and 
 oil, and thus have colours superb in tone and of a very 
 consistent thickness. 
 
 In fact, we are brought back to the oil and wax method 
 formerly invented by Taubenheim ; and the great incon- 
 venience which results from it is that the oil, sinking into 
 the canvas or panel, leaves the wax on the surface, and that 
 the following coating does not adhere to the preceding one 
 because oil does not hold on wax. All artists have noticed 
 that very often, when wishing to scrape away roughnesses, 
 they raise a pellicle, and that by continuing they can take 
 
76 
 
 SCIENCE OF PAINTING. 
 
 away a large piece which has not assimilated with the sketch 
 over which it has been repainted. 
 
 Besides this great fault, compromising the solidity of the 
 painting, the introduction of wax into the colours reduces 
 their opacity and intensity : not only do they no longer cover 
 the canvas, but they deteriorate, for they derive their richness 
 neither from oil nor wax, but from colouring-matter. 
 
 Grinders allege still that wax gives body to the paste, 
 which yields better to the palette knife, and that it gives to 
 the colours a smoothness which all artists now seek. 
 
 To this we answer that they can perfectly well grind closer 
 than they now do (it is only a little more fatiguing). By 
 putting less oil the colours remain dull without the need of 
 wax ; and if the paste seems too firm in certain cases, it is 
 easier to add oil than to take it out. 
 
 Besides, there is a very simple way of satisfying everybody : 
 it is to grind pigment according to our wishes for those 
 artists who wish to return to sound tradition, and to continue 
 grinding in the modern way for those who paint trowel- 
 fashion, or undertake long voyages. 
 
 By grinding colours with only the quantity of oil strictly 
 necessary, we are able to introduce other useful substances 
 to perfect the different qualities that we require from each 
 of them. With more or less resin soluble in oil^ cold, we 
 shall increase at will the transparency of some, leaving to 
 others all their opacity ; and if we fear that in this condition 
 the paste will be too stiff to be convenient, we can increase 
 its fluidity by means of a petroleum but slightly volatile, 
 which, taking as long to evaporate as oil does to dry, will 
 give us, whilst working, the feeling of a pigment containing 
 much oil. If certain colours ground in these conditions 
 showed a tendency to run, like vermilion for example, one 
 might add to them, instead of wax, a salt of aluminum with 
 
ESSENCES. 
 
 77 
 
 which the consistency of pomade can be given to oil without 
 any inconvenience-. The colours, being thus conveniently 
 prepared, might be used as they are, especially for painting 
 solid work. But it is necessary, in certain cases, to be able 
 to increase fluidity, siccativeness, and transparency. It is 
 to answer the call of these three indispensable qualities that 
 the innumerable quantity of ingredients has been invented 
 which are mixed with colours on the palette, and which, 
 under the names of unguents, pomades, balms, mediums, 
 lotions, gluten, varnish, essences and siccatives, litter the 
 studios, and are, without exception, poison for painting — in 
 different degrees, however. We are about to try and put a 
 little order into all this collection. 
 
 Essences. 
 
 If we wish to render a colour liquid, to sketch clearly, 
 glaze or paint transparent and vaporous objects, we use an 
 essence which distempers the colour. Many essences are 
 used for this purpose : essence of lavender or aspic (a variety 
 of lavender), essence of turpentine. 
 
 All the essences procured by distillation from plants or 
 resinous balms have the fault of becoming resinous by 
 contact with the air. They get thick and yellow : it is 
 then said that they are greasy. Arrived at a condition of 
 bird-lime, they are no longer volatile, and consequently, not 
 drying, they become a cause of cracks in the painting. 
 
 One must be careful to rectify them, i.e.^ to distil them 
 afresh before using them. 
 
 In this rectification, the parts already resinified remain 
 in the alembic, and the essence once more becomes limpid 
 and colourless ; but in a short time the phenomenon occurs 
 again, and none but recently rectified essences should be 
 
78 
 
 SCIENCE OF PAINTING. 
 
 used, SO as to diminish as much as possible this incon- 
 venience ; as for suppressing it entirely, that is impossible, 
 if merely from the fact that it evaporates on contact with 
 the air ; the purest essence becomes resinified, and there 
 always remains a viscous residue which never dries, but gets 
 yellow and attracts and retains all dust. A true manufacture 
 of black. 
 
 In fact, all essences, without exception, cause paintings to 
 blacken, and become yellow. 
 
 If they have not hitherto been discarded, it is because 
 there has been nothing to replace them with ; but now we 
 find in petroleums liquids offering many more qualities 
 without inconvenience. 
 
 Petroleums. 
 
 Petroleum, by taking in distillation only volatile particles, 
 evaporates without forming any residue. It serves as a 
 means, and leaves nothing behind it. Spread on white 
 paper it is impossible when evaporated to find even any 
 trace of it. 
 
 Further, the facility of having it at various degrees of 
 volatility allows the work to be regulated according to the 
 desire of the artist. We have obtained some essences of 
 petroleum which evaporate in five minutes, others in an 
 hour, others in several hours, and even in several days (see 
 next page). At this point it will be useful to make a 
 little observation. It may seem that very volatile petroleum, 
 evaporating in five minutes for instance, and petroleum but 
 little volatile, only evaporating in several days, mixed in 
 different proportions, would be sufficient to obtain all the 
 degrees of volatility. That is an error. The petroleum 
 which evaporates in five minutes is distilled between loo 
 and 150 degrees of heat; that which only evaporates in 
 
PETROLEUMS. 
 
 79 
 
 several days is distilled between 300 and 350 degrees of 
 heat. Now, if you put half of the one and half of the other, 
 say twenty grains in all, into an open vase, at the end ,of 
 five minutes only ten grains will remain ; that which evapo- 
 rates in that length of time will be gone, but the remaining 
 ten grains will require some days to evaporate. 
 
 These essences of petroleum should therefore be care- 
 fully kept apart, and not mixed together except' in cases 
 where one might reasonably benefit from those differences of 
 volatility. For instance, having to paint a sky rapidly, you 
 wish your colour to be very liquid, to spread it on easily, and 
 afterwards suddenly less liquid, in order to finish. You will 
 take then, as a liquid to mix your colours, a mixture com- 
 posed of a petroleum evaporating very quickly, which will 
 be gone when you have finished the sketch, and another 
 much slower, which will remain as long as you require to 
 finish it. 
 
 Products derived from petroleum submitted to a special 
 treatment, become absolutely colourless, and can be ren- 
 dered almost inodorous, which gives them a great advantage 
 over all other essences, whose penetrating odours are in- 
 supportable even to professional artists. This advantage, 
 and that of leaving no residue, are not the only advantages 
 which petroleum has over essences. It is further superior 
 by its power of penetrating through bodies, and especially 
 through dry oil. A drop of petroleum on a canvas long 
 after painting, goes through the painting without dissolving 
 anything, and comes out at the other side there, where 
 essence of turpentine would not even penetrate. It follows 
 that resins, oil and transparent colours diluted with 
 petroleum are carried by it much farther into the depths of 
 the old coatings over which they are laid, than they could 
 be by any other medium. 
 
8o 
 
 SCIENCE OF PAINTING. 
 
 To understand the importance of this quality of petroleum, 
 it is necessary to know how one coat of paint operates 
 on another. It is not like gum, where the new coat, 
 distempering the preceding one, unites with it. Here the 
 fresh oil has no action over the dry oil, and the adhesion 
 is entirely mechanical. 
 
 The oil of the new coat filters into the old by a multitude 
 of little wells remaining open, which are like the pores of 
 the skin, and these filterings hardening become solid fila- 
 ments, remaining rooted in the old coat like the roots of 
 a tree in the earth. Now, according to whether the . oil- 
 painting be more or less dry, its pores are more or less open ; 
 a time arrives when they are sufficiently shut up for no new 
 oil to be able to enter; but petroleum penetrates these 
 perfectly, and if you have been careful to rub an old 
 painting over with it before repainting the upper surface, the 
 new oil dissolving in the petroleum will be carried by it to 
 where it could not have filtered alone. 
 
 If the colours contain wax, adhesion does not take place, 
 as has been said above, and it is now easy to understand 
 that in fact this wax, corking up the pores, does not allow 
 the oil to cast its roots ; in this case, petroleum, instantly 
 dissolving the wax, uncorks the pores, and adhesion can 
 take place. For all those reasons you will do well .to use 
 petroleum. If, however, your fancy inclines to turpentine, 
 only use it when rectified and renew it often, like the boy 
 who remains faithful to his friend but who changes the 
 friend every month. 
 
 Siccatives. 
 
 There is only one means known of rendering oil very 
 siccative, to add to it oxides of manganese and lead ; not 
 either, but both at the same time, and in equal proportions. 
 
SICCATIVES. 
 
 8i 
 
 Why so ? 
 
 We are about to have the boldness to try to explain. 
 
 We have said that in order to pass from the liquid state 
 to that of solidity, oil is required to make with oxygen a 
 kind of combination. This oxygen the oil takes from the 
 air, but it can absorb an appreciable quantity of it without 
 combining with it, and consequently without drying*. 
 
 Oxide of manganese has the property of taking the oxygen 
 where it finds it, and by the help of the oxygen transforming 
 itself into double oxide. 
 
 Oxide of lead has the property not of taking oxygen 
 wherever it may be, but from those bodies which had 
 already combined with it. 
 
 As for oil, it only combines well with oxygen when at 
 transition point — i.e.^ when leaving one combination to enter 
 into another. 
 
 In fact, we have three thieves who constantly rob each 
 other ; let us watch their operations when united. 
 
 Oxide of manganese, taking oxygen where it finds it, 
 takes even the oxygen brought by oil when not yet com- 
 bined with the oil ; by this means it gets speedily rich and 
 becomes bi-oxide, which is the millionaire of oxides. 
 Then oxide of lead steals its money, i.e.^ its oxygen, to 
 appropriate it to itself. 
 
 Oil appears afterwards as the third thief, and takes the 
 oxygen to itself at the precise moment when it is passing 
 from the manganese to the lead, for it is in those conditions 
 that the oil combines best with oxygen. Here is an 
 illustration. 
 
 A man has in his meadow grass which contains azote, 
 phosphorus, and other matters required for his nourishment, 
 but which, in this form of grass, would be difficult for him 
 to digest. He gives this grass to his cow, which eats it, and 
 
82 
 
 SCIENCE OF PAINTING. 
 
 thus absorbs the azote, phosphorus, etc. Then the dairy- 
 maid milks the cow, and the man, drinking the milk, re- 
 finds the azote, phosphorus, etc., which under this new 
 form are to him perfectly digestible, and he may then profit 
 largely. 
 
 If this explanation does not enable you fully to under- 
 stand what happens when oil is brought into contact with a 
 siccative on a basis of manganese and lead, it may at least 
 give you some vague idea of it. 
 
 The most common type of this kind of siccative is that 
 called Courtrai siccative. Its manufacture is at the same 
 time barbarous and mysterious. If you ask information on 
 the subject from t'he colourmen, they are as dumb as if 
 their back-shop were the seraglio of the Sultan ; or rather 
 they will tell you that they have their siccative direct from 
 Courtrai. Some one was simple enough to go and see. At 
 Courtrai they did not even know what it was ! Thé truth is 
 that the colourmen do not make the siccative themselves, 
 and the majority of them do not know how it is made. They 
 content themselves by insisting that it shall be very black 
 (this is absurd but quite true), so that the manufacturer, 
 who might with care make a better production, is quite at 
 liberty to let all impurities pass, since it is never sufficiently 
 dirty, and he is obliged to add lamp-black to it. 
 
 Colourmen receive this siccative in large cans, and they 
 content themselves with putting it into bottles, the smallest 
 possible. Those who are more conscientious than the 
 others (there are some) wait for some time until it clears, 
 before performing this operation; but then there remains 
 at the bottom of the can a large quantity of muddy deposit, 
 and one must be very conscientious to consent to such a loss 
 as that. 
 
 This siccative should be made chemically pure \ it would 
 
SICCATIVES. 
 
 83 
 
 then be much less coloured, and it might be introduced into 
 colours in the form of paste. One would thus have the 
 advantage of being able better to regulate the necessary 
 quantity ; for too much is always used, and this should not 
 be with colours that do not dry easily. If, however, you 
 choose to continue the liquid form to which we are 
 accustomed, it should first be made with petroleum instead 
 of turpentine for reasons already stated, and it should not 
 be kept within reach in a cup fixed to the palette, as is 
 generally done, because every instant the paint brush is 
 dipped into it, it is used as essence and in the end a 
 ridiculous quantity is used. This is all the worse, because 
 when used in more than a certain proportion it does not 
 aid drying, and it is destructive to introduce oxides in too 
 large proportion ; for, when the oil is saturated with them, 
 they continue their effects to the detriment of certain 
 colours with which they remain mixed. 
 
CHAPTER VIL 
 
 RESINS AND VARNISHES. 
 
 We cannot, for this study, follow the method which we 
 have hitherto used ; for, if it be indubitable that ancient 
 artists mixed resins with their oil-colours, it is impossible 
 for us to know, for several reasons, what those resins were. 
 
 1. Chemical analysis is still powerless in this respect. 
 Resins not having been seriously studied by any one, their 
 composition is not known; we know that they contain 
 hydrogen, carbon and oxygen, but the formulas are not 
 settled. The proportionate quantities of those three sub- 
 stances vary, not only in each resin, but they change 
 perpetually by contact with the air, whether in solution or 
 in a dry state ; so much so that in analysing a resin taken 
 from an old picture, we have the proportions of hydrogen, 
 carbon and oxygen at which it has arrived in this musty 
 condition, but it would be impossible to infer from this 
 statement what the proportions in its primitive state were. 
 
 2. The researches which can be made in old works treat- 
 ing of painting only furnish very obscure information about 
 the resins formerly used, whilst the names under which they 
 are designated are, for the most part, unknown to-day, and 
 those names which have traversed the centuries have always 
 been applied indiscriminately to all kinds. 
 
 The names of resins, even in our time, are not yet fixed ; 
 what is called copal gum in France is called quick resin 
 
RESINS AND VARNISHES. 
 
 85 
 
 in England, and vice versa \ this copal gum is alternately 
 named gum of Calcutta, of Zanzibar, etc. — not according 
 to the names of the places of its ' production, but merely 
 those of the countries where stores for it have been es- 
 tablished. Now, as the route of ships has been modified 
 through steam, the boring of isthmuses and other circum- 
 stances, the places of storage have been changed and the 
 names have been altered. Besides, arbitrary classifications 
 have been made. Thus several kinds of resin have been 
 arranged into three categories : hard copals, semi-hard and 
 soft copals, sometimes also known as male and female 
 copals. Hard copals were the male, and soft copals the 
 female. This classification permits the sale without fraud, 
 under the general name of copal varnish^ of the commonest 
 and least solid solutions of resins. 
 
 To increase yet more the confusion, it happens that the 
 custom-house has imposed a heavy tax on all resins imported, 
 leaving gums scarcely taxed. Immediately commerce 
 hastened to call gum all that had hitherto been called resin; 
 and now there are only coarse resins, refuse from the manu- 
 facture of essences manufactured in France, which have 
 preserved their names. The result of all this is that when 
 one finds anywhere a recipe for varnish, one never exactly 
 knows of what resin the author is speaking. It is thus that 
 we find continually processes detailed for dissolving the true 
 hard copal, whether in an essence, in alcohol, or in any 
 other solvent. Now, the true hard copal does not dissolve 
 in anything ! absolutely nothing ! Or at least if any 
 solvent exist, it is not yet discovered. It therefore follows, 
 that not to treat as deceivers those practitioners who 
 profess to have dissolved it, we must conclude that the resin 
 which they have taken for hard copal was not the same as 
 that which to-day bears this name. We have, therefore, 
 
86 
 
 SCIENCE OF PAINTING. 
 
 nothing to gather from the examination of old pictures on 
 this subject, and we are obliged to take the resins in the 
 condition in which modern commerce presents them to us. 
 
 We class them into four groups, not basing them either on 
 their names or their places of production — all that may still 
 be changed — but on their solutions, which do not change. 
 
 Group I. — All the resins which dissolve entirely in cold 
 alcohol, and not in essence, like gim lac. 
 
 Group 2. — All those, which are dissolved in their own 
 essential oil, and can be put over another essence, such as 
 Canada balm^ balm of Copaiba^ turpentines of Scio, " of 
 Switzerland, of larch, of Bordeaux, American resin (elemi), 
 etc. 
 
 Group 3. — All those which dissolve equally well in 
 alcohol and cold essences, such as mastic^ sandarac^ etc. 
 
 Group 4. — All those which do not dissolve in anything 
 without the help of fire, such as Manilla^ Madagascar., and 
 copal gums, etc. 
 
 What is to be done in order to find, in all that, the 
 elements of a varnish suitable to be used with colours ? 
 
 Works of reference are only filled with empirical recipes 
 eternally copied, twisted, and never actually tried by him 
 who describes them ! If some author brings occasionally 
 a novelty of his own, we find fantastic mixtures, as for 
 example to have oil cooked with lime to use as a pomade 
 for touching up. 
 
 The scholar who gives this recipe to artists had already 
 given them (he gave many) bitumen and greasy oil of sad 
 memory : one might say leprosy and plague. 
 
 He had been speaking of a journey to Italy, and he relates 
 himself how he made this fine discovery. One day, he 
 stopped in the street to contemplate a noble old man who, 
 on the threshold of his door, was painting a picture yet 
 
RESINS AND VARNISHES. 
 
 87 
 
 older than himself. This circumstance was not surprising 
 in a town where every one is more or less of a broker and 
 restorer of pictures ; but one peculiarity had forcibly struck 
 the traveller. This old man was not alone ; he was accom- 
 panied by a little pot ! Yes, a little earthenware pot in 
 which was some strange substance. Into this little pot, the 
 old man automatically dipped his paint brush at regular 
 intervals; and each time that, after this immersion, the 
 paint brush came back to the picture, the picture became 
 darker and older ! — Oh ! the secret of the little pot or death, 
 is it not ? (This was in 1830.) This secret the old man had 
 from his grandfather, who had it from his, who had in- 
 herited it from a friend one of whose ancestors had formerly 
 bought it from a robber ! — who had stolen it from a cele- 
 brated monk — in the neighbourhood of Florence. 
 
 Fra Bartolommeo ! — Hush, old man. — Be quiet. — Here 
 is gold ! — And the little benefactor bore the treasure over the 
 hill, without explaining that this mixture of baked oil and 
 lime was only a kind of black soap ! 
 
 We cannot enter into details of all the recipes which have 
 been published ; they may ultimately be reduced to this : — 
 
 1. Use balms in their natural state : 
 
 2. Dissolve one or more resins in alcohol and essences, 
 and use them as varnish : 
 
 3. Bake resin m oil to make oily varnish ; and add wax 
 to it to make pomade. 
 
 The use of balms is to be discouraged, because the 
 essential oil which they contain, requiring a long time to 
 evaporate, prevents the colours from drying regularly. It 
 forms a crust on the surface, whilst inside parts remain soft ; 
 and as essence of balm always evaporates at last,- even if it 
 should be through the pores of the dry soil, the lower 
 coatings then shrink, and the surface is liable to crack. 
 
88 
 
 SCIENCE OF PAINTING. 
 
 Alcohol varnishes are not used in oil painting ; but there 
 is one of them which is too frequently made use of on 
 account of its being convenient ; viz. : — Sochné retouching 
 varnish. It does not mix with colours, and it is only used 
 to get out an einbu'^ and to repaint over the top. This 
 varnish is absolutely pernicious, because it has a basis of 
 gum lac and being altogether insoluble in oil the coatings 
 of colour are thus separated into isolated flakes, making of 
 the picture a mass of leaves without cohesion. Besides, this 
 varnish is easily affected by damp, which makes it dull and 
 bluish. Essence varnishes have all one great drawback : 
 they do not evaporate completely, but leave a viscous and 
 coloured residue besides the resins with which those var- 
 nishes are made, are either not soluble in oil, or form with 
 it a mixture disagreeable to use and even decomposing in 
 the presence of siccative, such as mastic for example. 
 
 As for oily varnishes, only one is known — oil copal. It is ^ 
 the same as that used in the thirteenth century by a 
 learned monk to preserve the paintings in his convent. 
 
 He prepared it himself during the calm of the night, on 
 account of the dangers of the operation ; and when the 
 peasants returning from the fields passed near the old 
 chapel where he had installed his laboratory, they crossed 
 themselves when they saw the pestilent vapours, all red 
 with reflection from the brazier, escaping from the windows. 
 Since then nothing new has been discovered, except that it 
 is no longer in old chapels, but outside the barriers of towns, 
 that this dangerous and disgusting manufacture is carried on. 
 
 It has remained absolutely at the same point. Some 
 makers, not even yet knowing that mercury thermometers 
 
 * Emhii is equivalent to the English word sinking. It means a part 
 of the picture where the painting is dull owing to the foundation having 
 absorbed in the oil. 
 
RESINS AND VARNISHES. 
 
 89 
 
 exist registering to 400 degrees, have no other means of 
 ascertaining the temperature of oil than by frying in it a 
 clove of garlic and a crust of bread. 
 
 Let us therefore take this production such as it is and 
 see what we can learn from it in the cause of painting. 
 From all the resins which do not dissolve in oil without 
 the help of fire we shall choose the most beautiful and the 
 hardest, true copal, leaving all the others as only fit for 
 varnishing carriages. 
 
 To recognise resins at first sight it is not enough to have 
 read the description of them : one must be in the habit of 
 manipulating them. If young artists had at their disposal 
 specimens of all the resins and also of all materials used in 
 painting, they might, without much trouble, familiarise 
 themselves with all the products of nature or science the 
 very names of which are unknown to them at present : and 
 this would be of great use to them later. It seems that at 
 the School of Fine Arts a collection of this kind ought to 
 have its place in a laboratory, where a practical chemist 
 might show to the pupils the different elementary prepara- 
 tions which constitute art materials ! Let us hope ! True 
 copal dissolves at 370 degrees of heat. Subjected to this 
 temperature, it sends forth vapours which condense into an 
 empyreumatic oil, as do all resins soluble under the action 
 of fire. 
 
 When copal has thus lost in vapour 10 per cent, of its 
 weight it can be dissolved in the essence of aspic : this is 
 what it called copal essence forming the basis of Haarlem 
 siccative (but do not go there to seek it). In this condition, 
 copal neither dissolves in oil, turpentine, nor petroleum, and 
 it yields a bad result if mixed with colours. 
 
 It appears to cause them to dry because as soon as the 
 essence of aspic evaporates, not remaining dissolved in the 
 
90 
 
 SCIENCE OF PAINTING. 
 
 oil, the copal dries by itself ; but it also brings into the 
 painting a foreign matter which does not incorporate with 
 it. (We shall speak of its effects later.) In all cases, those 
 preparations are only false siccatives which in no way dry the 
 oil, since oil only solidifies when mixed with oxygen. When, 
 continuing the operation of dissolving copal by heat, the 
 copal has lost 25 per cent, of its weight, it can be dissolved 
 in oil and essences, even cold. This is the production 
 known as copal oil It can then be painted over essence 
 or petroleum, and it is no longer destructive to the painting 
 except that it is very yellow and becomes even more so 
 with time. It dries badly and leaves a dull surface. 
 
 In reality, this resin, dissolved in oil, is no longer hard 
 copal : if we separate it from the oil, it is dry, yellow and 
 friable between the fingers like colophony. 
 
 It is, therefore, not worth while to take the rarest and 
 most solid resin, since it can only be utilised by making it 
 like the commonest and most fragile. 
 
 Before the experiments which have permitted us to know 
 this were made, copal was put into oil over the fire, and, seeing 
 it melt there, it was imagined that it was thereby incor- 
 porated with all its qualities. To-day we are forced to 
 acknowledge that the same disintegration takes place in 
 warm oil. As soon as the oil attains the temperature of 
 370 degrees, the same vapours escape, and it is only when 
 the copal has lost 25 per cent, of its weight that it begins to 
 dissolve. Only it is almost impossible to prevent the oil 
 from rising to a higher temperature, and then the copal 
 burning becomes partly charcoal, which, with its empy- 
 reumatic oil, makes actual tar. It is the same with all the 
 resins of the fourth group requiring the help of fire. They 
 themselves are not dissolved; it is only the product of 
 their dissociation that is dissolved. Amber, succinum, or 
 
RESINS AND VARNISHES. 
 
 91 
 
 yellow amber about which so much has been said lately, 
 acts like copal. Yet persons, whose veracity may not be 
 doubted, have affirmed that they have found the means of 
 dissolving amber or copal without the help of fire. There 
 is only one remark to be made about this ; it is that copal 
 may be dissociated by some chemical means other than 
 heat, thereby obtaining the same result. 
 
 What we should like to see would be a varnish made with 
 copal, which, once dry, should be as brilliant, as transparent, 
 and as hard as copal itself. That has not y^t been seen ! 
 
 As far as we have arrived, the result is that no resin 
 known is suitable to be used in painting. 
 
 Must we, then, wait until others are discovered ? No ! 
 One must be made, or rather one must be purified. After 
 many trials and much patient study, we can affirm that 
 nearly all resins, as they are collected, are in a partial state 
 of disorganisation ; or, if you prefer it, just about to be 
 organised, which is almost the same thing. All their parts 
 are not soluble in the same solvents, are not coloured alike, 
 do not melt at the same degree of temperature, and do not 
 contain the same proportions of hydrogen, carbon, and 
 oxygen. 
 
 One single resin we have been able thus to separate into 
 twenty-three different parts. Those parts are in varied 
 proportions not only in each resin, but even in the same 
 resin the proportions are perceptibly variable from month 
 to month. Would they continue to vary indefinitely, or 
 does there exist a definite condition towards which all those 
 parts are travelling and at which they will stop ? We do not 
 know. We have never found any arrived at this condition. 
 On the contrary, we have remarked in several a more or 
 less considerable part no further divisible, uncoloured, hard, 
 crystallisable, perfectly transparent, soluble in cold oil and 
 
92 
 
 SCIENCE OF PAINTING. 
 
 petroleum, and which would be, were such a definition 
 admissible, the normal resin^ having as yet undergone no 
 transformation. This is the normal resin that you ought to 
 seek, if you want a substance invariably the same, what- 
 ever may be the climate of the country, the nature of the 
 soil where the tree grows which produces it, the manner in 
 which it is gathered, and the changes to which it has been 
 subjected ; for all those conditions which one cannot know 
 influence the composition of the resins, which consequently, 
 although of the same kind, are never absolutely alike. With 
 the normal resin we shall compose, by the help of petroleum 
 at different degrees of evaporation, the three kinds of 
 varnish which are necessary to facilitate the process of oil- 
 painting : 
 
 1. Varnish for re-touching ; 
 
 2. Varnish for painting (or oil varnish) ; 
 
 3. Picture varnish (or final varnish). 
 
 Varnish for Re-touching. 
 
 Dull patches or sinkings, of which we shall speak later with 
 regard to the painting of a picture, should not trouble the 
 artist, since, with a light scumbling of re-touching varnish he 
 can remove them whenever they appear. 
 
 As this varnish dries in a few minutes, it can be im- 
 mediately painted over ; and as it forms a soHd link between 
 the different coats of paint, it is even very necessary never 
 to repaint any piece without first painting it over with this 
 varnish as a preliminary. 
 
 It can be mixed with colours to make rapid glazings, but 
 it is too volatile to be used for solid painting.* 
 
 For this purpose painting varnish should be used. 
 
 * If, however, in certain cases, this quick drying is no inconvenience, 
 there is no other objection to painting with re-touching varnish. 
 
PAINTING VARNISH. 
 
 93 
 
 Painting Varnish. 
 
 This varnish, which should be kept in the trough of the 
 palette, is used for painting by mixing it, by means of 
 the paint brush, with colours whose fluidity, brilliancy, 
 and solidity it increases. It is of excellent service for 
 glazing, and, not being too siccative, it allows a long time 
 for shading in the fresh paste. It replaces re-touching 
 varnish in all cases where it is desired to paint on wet instead 
 of dry canvas ; and mixed with re-touching varnish in all 
 proportions it can be used for all requirements. All these 
 varnishes are composed of purified petroleum, more or less 
 volatile, leaving no residue after their evaporation, and 
 composed of solid matter perfectly soluble in cold oil. So 
 that, whether they are painted over, or directly mixed with 
 the colours, those substances are always penetrated by the 
 oil, dissolve easily in it, and, completely incorporating with it, 
 they give to it much more resistance, without withdrawing 
 from it its suppleness and without hindering its work of nar- 
 rowing. Further, those varnishes render the drying of the 
 painting quicker and more equal by a phenomenon which a 
 short explanation will make clear. As we have already said, 
 oil must be oxygenised in order to dry. Left to itself, it 
 borrows oxygen from the air. If siccative be added to it, 
 the siccative collects the oxygen and transmits it to the oil, 
 which is a quicker process ; but in any case, contact with 
 the air is always indispensable. 
 
 When the first film forms on the surface, this contact 
 having no further hold than through the pores of the dry oil, 
 the drying of the deep coatings is naturally kept back until 
 this film thickens. Now, amongst the materials which these 
 varnishes contain, is to be found a rather large quantity of 
 oxygen in a chemical condition which renders it very easily 
 
94 
 
 SCIENCE OF PAINTING. 
 
 assimilable with oil : the latter can, therefore, oxygenising 
 without the help of air, dry much quicker, and in an equal 
 degree at all depths. 
 
 It is again useful to explain that all re-touching varnishes, 
 and all glutens, pomades or siccatives that have been. until 
 now introduced into oil-painting, besides their characteristic 
 of becoming yellow and black, have the serious fault of 
 being largely composed of substances insoluble in cold oil, 
 so that the successive coats of paint are separated by 
 isolating pieces, and the oil with which these colours are 
 ground is divided by foreign substances, unequally dis- 
 tributed, which solidify quicker than the oil and become 
 harder and more brittle. 
 
 It follows that the oil, which contracts whilst drying, can 
 no longer do its work regularly, in the midst of all these 
 obstacles ; it then happens that successive solutions cause 
 the outlines of cracks that go on alway? increasing . as the 
 work of the oil proceeds. And this process of oil-shrinking 
 continues for years ! 
 
 These accidents are not prevented by the final varnish, as 
 their causes lie below. Often the accidents are rendered 
 worse by the final varnish, for it brings a new resistance, if 
 it is not perfectly supple, and it cracks in its turn when too 
 dry. The greatest care should therefore be exercised in the 
 choice of picture varnish. 
 
 Picture Varnish. 
 
 The final varnish which is painted over the finished 
 picture has for its object, whilst giving brilliancy and trans- 
 parency to the colours, to preserve the picture from direct 
 contact with the impurities liable to be deposited on its 
 surface, as well as from the gases in the atmosphere ; and 
 
PICTURE VARNISH. 
 
 95 
 
 that, without trammelling the, incessant work of shrinking 
 and dilatation of the oil, the wood and the canvas. 
 
 In order, therefore, that this object may be attained, the 
 varnish must remain uncoloured and transparent, at once 
 resisting and supple ; it must be possible to wash it when 
 dirty, and should it be desirable to take it off, it must be 
 possible to do so without spoiling the picture. 
 
 Now, all varnishes hitherto employed become more or less 
 yellow as they get old. They get blue with dampness, sticky 
 with heat, crack, and in time always finish by getting white 
 and opaque. They cannot be washed when dirty without 
 spoiling them, as they do not withstand damp. They have 
 to be renewed, and can only be removed by rubbing them 
 with the finger, or by damping them with alcohol and essence 
 of turpentine. All those processes, however careful one may 
 be, always attack the painting a little, — so much so that after 
 several removals of varnish the picture is half destroyed. 
 Many amateurs know this danger, and they often prefer to 
 keep their smoky pictures, almost invisible under deteriorated 
 varnishes, rather than risk experiments. 
 
 The varnish which we propose joins to all the requisite 
 qualities the great advantage of being easily cleaned as 
 often as may be desired,, without changing in any way, 
 which enables the pictures to be kept in a constant state of 
 cleanliness, absolutely necessary for their good preservation. 
 The petroleum with which this varnish is made, evaporating 
 less rapidly than the essence, renders the varnishing much 
 easier, by enabling the operator to apply the coating more 
 leisurely, without the drying being thereby retarded : quite 
 otherwise, for this varnish is completely dry at the end of 
 about an hour. Further, as it does not contain any viscous 
 matter, there is no stickiness to attract the dust in the air. 
 
 For directions for varnishing see Appendix — "Varnishing. ' 
 
CHAPTER VIII. 
 
 CANVAS, PANELS, AND THEIR PASTES.— SIZTNGS, 
 
 Having chosen with the greatest care the materials re- 
 quired for painting, we have still to know on what we are 
 going to paint ; the question of canvas, panels, etc., is one 
 of great importance. 
 
 A picture is composed of three altogether distinct ele- 
 ments : — 
 
 I The support, or the material substance painted on, as 
 wood, canvas, stone, paper, etc. 
 
 2. The coating or sizing used to cover the material support. 
 
 3. The final painting, formed by successive layers of 
 colour applied over the sizing. 
 
 If the painting be done under bad conditions, and con- 
 tain in itself the germs of its destruction (spoken of above), 
 the picture is irretrievably lost, and that in spite of the good 
 quality of the support, and of the sizing. 
 
 But, admitting the hypothesis that the painting is as per- 
 fect as it ought to be, it is also necessary that the two other 
 constituent parts of the picture {ix, the support and the 
 sizing) should be perfect. 
 
 The support will first be chosen as substantial as possible, 
 but it can still be preserved from many causes of destruction 
 by covering it behind and on the sides by means of mastic 
 and suitable protective varnish. (6*^^ Appendix — "Panels 
 of Unpolished Wood.") 
 
CANVAS, PANELS, AND THEIR PASTES. — SIZINGS. 97 
 
 If, however, it happens to spoil even when all precaution 
 has been used, a clever picture restorer can still save the 
 painting by transferring it to another canvas or panel, always 
 provided that the sizing sustaining it remains in good 
 condition. 
 
 A proof of this preservation of painting on destroyed 
 supports is to be found in an examination of the old triquet- 
 rous altar screens, whose three panels of the same wood, 
 painted by the same artist over similar sizings, are in the 
 best conditions to be usefully compared. Nearly always 
 the middle panel, remaining attached to the wall, is rotten, 
 owing to damp, while the two side leaves, being surrounded 
 by air, are yet perfectly sound. In those conditions, when 
 the sizings are good, the painting is equally preserved on 
 the middle panel and on the side leaves, whilst on the con- 
 trary on bad sizings, the painting is spoilt on the rotten 
 panel and preserved on the leaves in good condition. 
 
 In fact, therefore, the preservation of the painting depends 
 on the quality of the sizing. It is not enough that the sizing 
 should be solid: were it like steel, indestructible as the 
 diamond, we should still require to have it less brittle ; for 
 should it scale off, melt or powder, it would involve the 
 painting with it. It must even be sufficiently supple to lend 
 itself to the peculiarities of the painting, as it contracts 
 when drying, and yet it must be of sufficient resistance to 
 preserve the painting from too sudden movements of the 
 support. It should absorb the excess of oil and varnish 
 which comes to it from the painting, and it should reject all 
 which might penetrate through the support. In short, com- 
 pelled to obey the slightest caprices of the one and to resist 
 all the attacks of the other, it is required from it, as from all 
 servants, to be equal to the occasion. 
 
 Notwithstanding the importance of the rôle which the 
 
98 
 
 SCIENCE OF PAINTING. 
 
 sizing plays, and perhaps rather on account of this import- 
 ance, it is often suppressed— first, for painting on paper in 
 water-colours and pastel where it is not necessary, and 
 sometimes even for oil painting when done directly on the 
 support. 
 
 In this case supports should 'be chosen having something 
 of the qualities the sizing would have had — i.e.^ dilating little 
 and shrinking little, of a fine texture, uniformly porous, and 
 containing neither acids, resin, nor any other matter capable 
 of influencing the colour; if found too absorbent it will be 
 well before painting to brush the support over with a little 
 re-touching varnish, or painting varnish, or a mixture of 
 both, according to whether the painting is to be in a dry 
 style, or the colour is to glide smoothly over the surface. 
 {See Chapter IX.—'' Oil Sketching.'') 
 
 The ancients painted on wood, on tanned skins, on lava, 
 on marble, on slate, on metals, on walls covered with lime 
 mortar, and on stone itself, after having .saturated it with a 
 sizing of resin made to penetrate by the heat of a chafing- 
 dish. 
 
 Their portable panels were very thick and very solid. It 
 is related that at the siege of Rhodes the soldiers made use 
 of pictures by Apelles for a table. In the middle ages the 
 same substances were used to paint on, but especially wood, 
 chiefly oak and poplar. 
 
 The joins of the boards which formed the panel, and its 
 coating, were made of animal glue (Taurocoll), or of flour- 
 paste mixed with plaster or chalk, or even with a paste made 
 of cheese and lime, the recipe for which Theophilus gives 
 in a celebrated manuscript. This last paste is by far the 
 most solid. It has been proved that on old panels altogether 
 rotten the coating remained intact, and that even at the 
 place where the joins were solidified by bands of canvas 
 
CANVAS, PANELS, AND THEIR PASTES. SIZINGS. 99 
 
 prepared with this paste, the wood which it covered had 
 been preserved. 
 
 Sometimes, for valuable pictures, they gummed all over 
 the surface of the panel a leather coated with resin or covered 
 over with plates of gold. But as they could not make those 
 plates as thin as they are now made, they were valuable 
 enough to tempt the cupidity of the iconoclasts, and nearly 
 all the pictures painted on those foundations considered 
 indestructible were just the ones that were first destroyed. 
 
 The exclusive use of wood panels was preserved in Italy 
 till the time of Raphaël, and in the Low Countries much 
 later, till the time of Rubens. 
 
 At that time so strong was the conviction that panels for 
 painting should 'be scrupulously chosen, that government 
 had monopolised their manufacture. 
 
 Only perfectly dry and faultless wood was used, the 
 workmanship was of the best, and it was forbidden, under 
 penalty of fine, to paint on other panels than those of the 
 government, for the following reasons : that the genius of 
 an artist is the patrimony of his country ; that it is the duty 
 of the country to guarantee the longest possible duration to 
 its masterpieces, and that, to make this sure, equal pre- 
 cautions should be taken with regard to all pictures ; as a 
 painter, however celebrated he may become, always begins 
 by being unknown, he may by chance be modest, and con- 
 sequently ignorant of the ultimate value of the work which 
 he undertakes, and therefore the possibility of his compro- 
 mising the work by negligence or economy should not be 
 permitted." 
 
 This law, which to-day would make many people smile, 
 was rigorously enforced during the zenith of the Dutch 
 school, and it is perhaps to it that we owe the preservation 
 of many of the finest gems in European museums. 
 
lOO 
 
 SCIENCE OF PAINTING. 
 
 When canvas became more general, the sizings which had 
 been used for wood panels were continued, and it is on 
 the same concoctions of paste and chalk that the first canvas 
 pictures were painted. These concoctions were afterwards 
 abandoned by degrees — the sizings of paste, getting 
 thinner and thinner, were re-covered by others made of oil 
 and white lead, and finally no paste was used at all. But it 
 was soon observed that canvas, in direct contact with oil, 
 burned and became like touchwood. This sad fact estab- 
 lished, artists returned to the use of a first coating of paste, 
 merely to separate the canvas from the oil coating ; but this 
 first coating was of a gelatinous substance susceptible to 
 damp, and the canvas became rotten. 
 
 Another method was then tried — viz., putting behind the 
 canvas another coating of separating paste re-covered with a 
 second coating of oil and passing them between two cylinders 
 to press them. But, thus imprisoned, the canvas lost all 
 suppleness, and could no longer be stretched on the frames ; 
 then they stretched the canvas first, and the coatings put on 
 the stretched canvas were merely numerous coatings of oil- 
 white, and by therewith entirely covering the grain of the 
 texture, manufacturers tried to get a surface as smooth as 
 that of a panel. 
 
 This method had the inconvenience of encumbering the 
 studios with frames, each canvas requiring two or three 
 years of constant exposure to the air for its preparation ; 
 trying to shorten this time, siccative means were used — such 
 as replacing a portion of the oil with essence of turpentine, 
 using oil prepared with litharge, umber, or red earth mixed 
 with white and sometimes even using no oil at all. 
 
 These practices were not calculated to improve the 
 qualities of a coating already doomed to scale off fatally — 
 quite the contrary ; but they facilitated the manufacture of a 
 
CANVAS, PANELS, AND THEIR PASTES. — SIZINGS. 1 01 
 
 kind of canvas which assured to artists a rapid and valuable 
 execution in great favour for a long time especially amongst 
 portrait painters whose supreme ambition was to imitate 
 porcelain. 
 
 Did they suspect that they would imitate even its cracks ? 
 In any case, time has amply revenged the fiimsiness of those 
 guilty pretensions : to-day a portrait of an ancestor that is 
 cracked loses much of its value ! 
 
 After this period of brittle canvas, the ordinary laws of 
 reaction brought on a taste for supple canvas, and much 
 ingenuity was exercised in procuring this quality. A sizing 
 was added of mucilage paste, such as linseed, snail-slime, 
 honey, fig-juice, etc. An oil was also used, rendered 
 viscous by rancidity — ix.^ acidified and never drying. For 
 reasons of economy, manufacturers substituted for white 
 lead, chalk, whites of Troyes, whiting, white of Meudon, 
 white of Bougival, pipeclay, etc., which gave them results 
 just as good in the matter of suppleness, none of the car- 
 bonates of lime ever drying thoroughly when they are 
 ground with oil. 
 
 Every one knows that glazier's putty, which is made of oil 
 and whiting, will remain soft for many years under a hard 
 and wrinkled coating. 
 
 It is easy to understand, therefore, that painting must 
 crack when done over coatings that continue to shrink long 
 after the painting is dry. The invention of railroads, which 
 has changed so many things, is also not without influence on 
 painting-canvas, for like all other things pictures also now 
 travel, and the constant rolling and unrolling necessi- 
 tates the use of more and more supple canvas, until 
 suppleness is now the chief property considered in its 
 manufacture. 
 
 This is the present state of the manufacture. 
 
I02 
 
 SCIENCE OF PAINTING. 
 
 By an inexplicable anomaly, unbleached canvas costs 
 more at the French custom-house than prepared canvas 
 does. Now, painting-canvas being made chiefly in Holland, 
 and hand labour costing less there than at Paris, it has 
 become the custom to apply the coating at the place of 
 manufacture and to send the canvas ready prepared in 
 large rolls : this has the double advantage of costing less and 
 of not encumbering the shops. Further, the canvas being 
 rolled as quickly as possible, the oil deprived of the light 
 before becoming dry, gets rancid and diminishes the supple- 
 ness of the coating, as has been said above. 
 
 This canvas is therefore perfect ? For the tradesmen, yes ! 
 But it is yellow, it has a disagreeable odour, and pictures 
 painted on it remain to posterity blackened and cracked. 
 Of course manufacturers do not work for posterity. 
 
 By comparing the works which remain to us and by 
 examining the materials on which they are painted, it is 
 easy to convince ourselves that the best preserved pictures 
 have been painted on paste sizings. 
 
 Is that to say that all those which are on paste sizings are 
 well preserved ? Assuredly, no. But that does not affect 
 the principle, because, very often, the cause of destruction 
 is independent of the sizing, and sometimes the paste used 
 was of bad quality. 
 
 Paste should be incorruptible, supple, perfectly neutral, 
 and not susceptible to damp. In this condition it isolates 
 the colours from all chemical reaction liable to result 
 from the support ; neither contracting nor expanding, it 
 occasions no cracks, and its suppleness enables it to follow 
 the movements of wood and canvas ; whilst oil sizings, on 
 the contrary, oxidise, get yellow, produce chemical reaction 
 on certain colours, shrink, and in the end invariably become 
 brittle. 
 
PASTES. 
 
 103 
 
 The conclusion of all this is, therefore, that oil sizings 
 should be rejected and paste used. But it must be carefully 
 chosen. 
 
 Pastes. 
 
 How should we proceed to study pastes ? 
 As in all other studies, by consulting books treating of 
 the subject. 
 
 Yes ! but clever practitioners, well informed, have not 
 left books, and one gets mystified with a deluge of manuals 
 and collections scarcely intelligible. The recipes therein 
 found are often impracticable. They are frequently dis- 
 torted by successive errors of print, and in that condition 
 they are carefully collected, reprinted and perhaps translated 
 into several languages by one of the sham philosophers who 
 write about everything, satisfied with re-copying what has 
 been done before their day, without understanding it and 
 without ever making personal experiments themselves. 
 
 It is thus that the greatest errors and follies have come 
 to us through all ages, from book to book, as if they were 
 gospel truths. 
 
 If, however, you still desire to consult all those old 
 authorities, here is a kind of key to facilitate their study. 
 
 When you hear of oyster shells, crabs' eyes, stags' horns, 
 mother-of-pearl, pearls, you will know that by them is 
 meant carbonate of lime, of which the type is chalk. When 
 you hear of the blood of a pig or of any other victim, of 
 milk, of fresh cheese, of crust of Gruyère, of yolk of egg and 
 of powdered insects, thé principle of- all that is caséine, 
 fibrine or vitelline, which are almost the same thing. 
 
 White of egg is albumen ; if you hear of fishes' bladders, 
 cows' tails, old gloves, boot-tags, skins of rabbits, still-born 
 kids, or sheep's feet, all that is gelatine. 
 
SCIENCE OF PAINTING. 
 
 All the farinaceous féculents, wheat, barley, sago, tapioca, 
 rice, potato, etc., furnish starch and gluten. Many trees 
 and exotic plants having Latin names, not the same in all 
 books, yield gums and resins ; and if in some recipes you 
 read that a candle should be burned over the boiler, it is 
 only the wax which falls from it that can be useful. 
 
 As for quicklime, the urine of camels and dung, they 
 act like alkali, and can be replaced in almost all cases by 
 volatile alkali (ammonia). 
 
 You will also hear of other unimportant ingredients, which 
 must not be considered as of consequence. Amongst 
 them are figtree sap, syrup of snails, linseed, honey, all 
 used as mucilage to give suppleness, and not one of which 
 is equal to glycerine for this purpose. 
 
 All the pastes imagined and described up till now (there 
 are some thousands of them) are really made with only eight 
 substances, four of which are animal and four vegetable. 
 
 Animal. Vegetable, 
 
 Gelatine Starch 
 
 Albumen Gluten 
 
 Caséine (or fibrine, or vitelline) Gums 
 
 Wax Resins 
 
 Glycerine as mucilage, and ammonia as alkali, yield an 
 assistance often indispensable. 
 
 By taking, therefore, those substances chemically pure, it 
 is possible by mixing them to reproduce all the most com- 
 plicated ancient pastes. For this purpose it is enough to 
 consider those of the substances which contain the materials 
 composing the paste that we wish to reproduce. For 
 instance, here, in an old manuscript, is the simple recipe of 
 a paste used by the Benedictines to prepare the parch- 
 ment of missals on which they painted miniatures : " Gather 
 bees during the summer months after vespers ; pound them 
 
PASTES. 
 
 in a mortar with lime-water, and afterwards filter the mixture 
 through linen." 
 
 By analysing this strange mixture, we find that bees contain, 
 like all other animals, a little gelatine and fibrine, and bees 
 further contain wax and honey. 
 
 The presence of lime transformed the fibrine into paste, 
 and rendered the wax miscible with water ; filtering separated 
 the useless particles, the legs, wings and other impurities. 
 As for the advice to choose the summer months and the 
 evening hours, it is doubtless at that time of the year that 
 bees make most wax and honey, and in the evening that 
 they have most of it. 
 
 Let us now consider the subject synthetically. Instead of 
 fibrine we shall take lascine, which is analogous ; instead of 
 honey, glycerine ; instead of lime, ammonia ; we shall add 
 gelatine, wax and water, and, without even requiring to 
 filter it, we shall have made the Benedictine paste, 
 excellent for the purpose for which they used it. It is 
 not at all susceptible to damp, incorruptible, supple, and 
 perfectly good to preserve the leaves of a book destined 
 to be often handled. 
 
 And that is how science, which sometimes immolates so 
 many victims, can to-day save the bees from the terrible 
 pestle and mortar of the monks, and leave them to buzz in 
 peace amid the flowers of the meadows ! 
 
 With the eight substances mentioned, we are not limited 
 to re-making ancient pastes : we can compose new ones. 
 
 Only, when we consider that each of these substances can 
 be separately treated in different ways, and that certain of 
 them, such as gum and resin, offer different and numerous 
 types whose qualities should be separately studied, when we 
 consider the number of combinations that can be produced 
 with eight figures, that in each of these combinations the 
 
 I 
 
io6 
 
 SCIENCE OF PAINTING. 
 
 proportion may be changed many times, and that it is after- 
 wards necessary to try all these pastes, first on supports of 
 all sorts, to see how they are affected by wood, canvas, card- 
 board, paper, etc. ; and that afterwards paintings in all styles 
 should be subjected to dampness, dryness, sunshine, darkness, 
 and that it is prudent to wait several years to judge fairly 
 the results of these trials, it can be understood that it is only 
 toward the end of his career that an honest experimentist 
 can allow himself to say that he thinks he has finished. 
 
 Let us now see how each of these eight substances can 
 be transformed into paste or can be assimilated with it. 
 
 Gelatine^ dissolved in water, gives the paste generally 
 known as animal gluten and glue. Strength may be given 
 to those gelatinous pastes by the addition of quicklime, oily 
 varnish, or gum lac ; or they can be preserved liquid whilst 
 cold by acetic acid or other means ; glycerine, chloride of 
 calcium and indiarubber will give suppleness ; and they can 
 ultimately be rendered insoluble by chromic acid, acetate 
 of aluminum, and bichromate of potash. {See Appendix — 
 " Gelatine.") 
 
 Albumen is obtained by beating up the whites of eggs 
 in snow, and at the end of an hour collecting the liquid 
 which flows from it. It may also be bought in a dry state. 
 
 This paste can also be made supple with glycerine, and . 
 becomes insoluble when it is heated to loo degrees in a 
 damp state. 
 
 Caséine may be bought in a dry state from chemical dealers, 
 or it may be extracted from cheese. It becomes supple with 
 glycerine. {See Appendix — " Caséine or Cheese Paste.'') 
 
 Wax is a natural production. It should be chosen free 
 from the suet with which dealers mix it. It is rendered misci- 
 ble with water by volatile alkali (ammonia). {See Appendix 
 — " To render Wax miscible with Water and Glycerine.") 
 
PASTES. 
 
 Starch and Gluten furnish paste, dextrine, etc. (^See 
 Appendix — " Starch or Farinaceous Paste.'') 
 
 Gum dissolves simply in water. — By adding boric acid to 
 gum water it will keep a long time without corrupting. It 
 becomes supple with glycerine. 
 
 Resins dissolved in oil (oil varnish) or in essences mix 
 with paste in the state of emulsions, or directly by means of 
 alkali — as gum lac, for instance, which, dissolved in borax 
 or ammonia, can be thus introduced into sizings. {See 
 Appendix — "Gum Lac") 
 
 With the above any one can make for himself combinations 
 suitable for all purposes ; but we are here limited to the 
 search of better sizings for surfaces destined for oil paintings. 
 
 The great fault of those who make researches on a subject 
 and write about it is that they wish to make their readers 
 feel all the extent of their work, and they begin, before giving 
 the result at which they have arrived, by relating all the 
 unsuccessful attempts that they have made — like conjurors 
 who fail in their tricks several times to excite the public all 
 the more when they succeed. 
 
 Dismissing this pretentious practice, we shall say at once 
 that the best of all sizings that can be used to prepare canvas, 
 panels or cardboard destined to receive oil painting, is cheese 
 paste (or caséine paste) ; but there is a way of making it 
 which should be followed to the letter, if good results are 
 wished. (See Appendix — " Sizing of Caséine or Cheese.") 
 
 When the sizing has been executed on the canvas in good 
 condition, the next thing to think of is the wrong side of the 
 canvas, which also requires a sizing impervious to water 
 and oil, so that the excess of oil coming from the colours 
 shall not go through the tissue, as that might burn it, and so 
 that dampness shall not be able to penetrate it, as that might 
 cause it to rot. 
 
io8 
 
 SCIENCE OF PAINTING. 
 
 This sizing must also be supple, so that the canvas may 
 roll up easily. Many substances may fulfil those require- 
 ments — indiarubber dissolved in petroleum, wax and resin, 
 gum lac. But what is yet better, especially for canvas of 
 small dimensions, undesirable to thicken too much, is to put 
 two coatings of water-colour fixative. 
 
 It has been advised through excessive precaution to put 
 also a fireproof coating of some-kind as a preservative against 
 fire ; but this cannot be put on the front of the picture and 
 would be consequently of little use in case of fire. A better 
 plan would be to nail a metallic canvas on a light frame, 
 attached by hinges behind the picture (so as to be easily 
 opened at will). This metallic canvas, being thus held about 
 half an inch off from the painting canvas, would allow the 
 air to circulate and would preserve the picture from shocks, 
 such as contact with flame, of which there is much more 
 danger from behind than in front ; for it is noticeable that in 
 museums, at picture dealers' and in the houses of amateurs, 
 servants who have the greatest respect for pictures when 
 hung pay no attention to them when standing with their 
 faces to the wall : they drag ladders about, and move round 
 with lighted candles through piles of heaped up frames, 
 without taking the smallest precaution, although, on the 
 contrary, they ought to take the greatest care. 
 
 We might still, besides the sizings properly so called, 
 whose liquid basis is water, find in the substances which 
 chemistry has already discovered, and will discover, the 
 elements of a sizing more perfect than cheese paste ; but we 
 are unable to find one that offers the substantial guarantee 
 which this does of having been used for centuries, and of 
 having resisted all causes of destruction. 
 
 If, however, you have enough of confidence in the 
 science of an old practitioner who has consecrated his life 
 
PASTES. 109 
 
 to the study of painting, and if you wish to try something 
 new, here is a sizing which he proposes, and which he con- 
 siders absolutely perfect : — 
 
 Grind zinc-white with alcohol and mix it with water- 
 colour fixative. 
 
 Lay on one coating ; and when it is dry, which will be in a 
 few minutes, rub with glass paper. As a matter of course, 
 the more zinc-white you add, the more absorbent the sizing 
 will be, but it will be less solid in proportion. {See Appen- 
 dix — " Instantaneous Sizing.'') 
 
 Over all these sizings we can paint equally well in oil or 
 water colours. Only one question regarding the sizing 
 remains to be considered : and that is, what tone we should 
 prefer to give it. We do not hesitate to answer that the 
 sizing should be perfectly white, and that for several reasons. 
 First, to obtain effects of transparency without the necessity 
 of painting in oil-whites and afterwards glazing, which 
 involves waiting a long time until the under coatings are 
 dry, and sends us back into all the inconveniences of oil 
 sizing. Secondly, because the under coatings always work 
 up a little, and a picture, painted over a coating of any tone 
 whatever, will ultimately always become itself of that tone, 
 especially in those parts least covered. Certain pictures of 
 Poussin and of the French school of the eighteenth century, 
 painted over red coatings, are striking examples of this. 
 Some artists have preferred umber, and even pure black, 
 but their works have not lasted a century. Their pictures 
 have become quite invisible, in half tints, at the end of a 
 very short time. 
 
 Apart from the preservation of the picture, there is another 
 important consideration — viz., white foundations naturally 
 cause greater luminosity and that is a valuable quality ; 
 darkness always comes on soon enough. 
 
CHAPTER IX. 
 
 OUTLINE AND EXECUTION OF A PICTURE IN OILS 
 SINKINGS, 
 
 We have substantial colours, well ground ; petroleum 
 volatile in different degrees, siccative, liquid or in paste; 
 re-touching varnish, painting varnish and canvas or panels 
 •covered with excellent size, in short, for painting under 
 favourable circumstances, we have everything in good 
 condition. 
 
 We shall not trouble about the tools, brushes, paint 
 brushes, knives, pans, etc. ; each artist can make his own 
 choice ; they do not affect the solidity of the painting. We 
 shall only mention the palette. We advise it to be white, 
 and quite impervious to oil, because that allows a correct 
 judgment of the transparency of the colours, and also tends 
 to keep things light, and the sizing of the canvas or panels 
 being also white, tones will have the same effect on it as on 
 the palette. The latter being impenetrable, preserves to the 
 colours exactly the quantity of oil or varnish judged well to 
 add to them, and a white palette is easy to clean. The 
 utmost cleanliness in palette and all other requisites is quite 
 indispensable. 
 
 The Outline. 
 
 Whether we paint on a panel, or on canvas sized with 
 cheese paste, or on a panel of unpolished wood, or on 
 water-colour paper, the outline may be done either in oil 
 
THE OUTLINE. 
 
 Ill 
 
 or water colours. If in water colours, it is best to use 
 ordinary water colours, not paste — being careful, if on un- 
 polished wood, not to put on too much wash, so as to have 
 no misadventure with the wood. 
 
 1. If the sketch is to be done in water colours on canvas, 
 or panel sized with caséine paste, one very even coat of 
 painting varnish should be laid over it : this varnish will 
 partly soak in with the sizing, and, even before it is dry, the 
 oil painting may be continued. This is the process of Paul 
 Veronese. 
 
 2. If the sketch is to be done in water colours on water- 
 colour paper, one very even coating of water-colour fixative 
 should be laid over the paper : this fixative will soak 
 completely into the paper, and will be dry in a few minutes. 
 It is possible that this coating may not be sufficient, because 
 some papers absorb more and some less fixative. The fact 
 may be ascertained by putting a drop of oil on one of the 
 corners : if this oil does not penetrate into the paper, there 
 is enough of fixative ; if not enough, then put on a second 
 coating. 
 
 In any case those operations cannot take more than from 
 fifteen to thirty minutes, after which the oil painting may be 
 immediately continued. This method allows us to do in 
 water colours the part that they are capable of doing well — 
 viz., the sky, water, etc. — in fact, all the transparencies. The 
 solid parts which require much shading may be done in 
 oils. 
 
 The whole once varnished is of perfect solidity and of great 
 beauty of execution. 
 
 For this process, paper of fine grain should be chosen, 
 coarse grains always having a bad effect when varnished. 
 
 3. When the outline is done in water colours on un- 
 pohshed wood, it is well to give one good even coating of 
 
112 
 
 SCIENCE OF PAINTING. 
 
 water-colour fixative, and the oil painting can be continued 
 as soon as the fixative is dry, which will be in a few minutes. 
 
 Outlines in Oil. 
 
 If required to outline directly in oil : 
 
 1. On canvas or panel sized with caséine paste, paint 
 over the sizing one very even coat of painting varnish which 
 will partly soak in, without which this sizing would be too 
 absorbent. If, however, it is desired to make use of this 
 quality, a rather large quantity of painting varnish should be 
 mixed with the colours to make the work less tedious. 
 
 2. On water-colour paper, before beginning the oil 
 painting, one or two coats of water-colour fixing should be 
 applied, according to whether it is desired to have the 
 paper absorbent or not. 
 
 3. On a panel of unpolished wood, the outline may be 
 done directly in oil. It is also well to apply first a coating 
 of water-colour fixative, or of re-touching varnish, according 
 to taste, and according to whether the panel is to be absorb- 
 ent or not. Oil painting adheres admirably to the fixative. 
 When we said that for outlining in water colours impasta- 
 tion was not to be used, it was by way of caution, for it is not 
 a suitable process for this. But if the outlining be done 
 directly in oils, we must also dispense with it, and that 
 requires much care. 
 
 Oil colours for sketching should be used fresh ground, 
 being careful to put everywhere the same thickness (a light 
 semi-liquid), so that all the parts, however different in tone, 
 may have a generally equal coating, without blisters or harsh 
 ouches. 
 
 The merest trace of siccative should be put into 
 black, and the lakes, so that the oil of this first coat, not 
 
OUTLINES IN OIL. 
 
 drying too quickly, may penetrate deeply into the pores of 
 the object ; and the badger brush is not to be used, so that 
 this oil may not be brought back to the surface. 
 
 Each part of the picture should be outlined in flat 
 tints, of the lightest and most intense tone possible, of 
 the principal colour of the object which they represent, 
 because the lower layers of colour having always a tendency 
 to work upward, it is best that they should be very brilliant, 
 as this counteracts the tendency to darkness of the upper- 
 most layers. 
 
 Great care must be taken to blend the outline, and to leave 
 no careless stroke of the brush to be a trouble afterwards. 
 
 Before re-painting this sketch, any roughness which may 
 exist must be taken off with a very sharp scraper. 
 
 In short, this outline directly in oil is more a preparation 
 given to a support to make it suitable for painting, than an 
 inteUigent sketch. It may be answered : But this is dull 
 and monotonous ; it is a manœuvring method ; nothing is 
 left to inspiration or to chance. Firstly, let us always leave 
 as little as possible to chance ; and as for inspiration, it is just 
 in order to give it more liberty at the moment of finishing, 
 that we take so much pains with the preparatory part. If, in 
 the warmth of execution, you leave little empty spaces be- 
 tween your touches, you will find them very convenient : the 
 tone of the outline being almost definite no longer needs to 
 be blocked up everywhere ; whilst if you introduce another 
 tone into the sketch, all these little intervals have to be filled 
 up, which will soften the execution and give more liberty 
 to the painter. If just at the spot of a delicate detail you 
 fall upon an impastement or an empty place, as will be 
 found in an inspired sketch, you will be greatly troubled, 
 and it will be quite another bungling work to fill up the 
 holes and scratch off the lumps ; and frankly, if there i to 
 
114 
 
 SCIENCE OF PAINTING* 
 
 be any bungling, it is better it should be below than above. 
 Besides, if the sketch has any faulty features, or any particu- 
 larly strong points, as. in finishing the primary drawing is 
 not exactly preserved, these features are much less trouble- 
 some if the outlines and accents are but lightly marked. 
 
 Inspiration (here-it is !) might induce you, whilst executing 
 the painting, to be satisfied at certain places with a simple 
 scumbhng, which would be all right on a solidly and 
 evenly covered foundation ; but if you happened just at 
 that place to be on a part of the inspired sketch where 
 there was already almost nothing, those two almost nothings 
 one over the other would not suffice, and you would be 
 forced to fill up with colour this forgotten place, to the detri- 
 ment of that lightness of execution which is just what you 
 wanted to keep. 
 
 Besides, you can always in finishing make the tone as 
 soft, as broken as you wish, on a brilliant sketch, even 
 a little crude, whilst you will never obtain a colour at its 
 maximum of intensity on a dull sketch. On the contrary, 
 by putting an intense colour over itself, you increase its 
 power in a considerable proportion. 
 
 In short, it is to leave to the artist liberty to execute to his 
 own taste solid painting, semi-liquid, scumbling, and even 
 glazing, that we advise to him a sketch which never con- 
 strains, and which ensures to him the physical possibility, 
 by realising all his fancies, of working out his inspiration as 
 far as he has any. 
 
 And then this sensible and regular sketch is the guaran- 
 teed health of the painting ; it is its under garment. 
 
 That we may wear with impunity, at all seasons, light 
 fanciful clothing, we wear underneath a good flannel ; well ! 
 the sketch which we are recommending is the flannel of the 
 picture. 
 
THE EXECUTION. 
 
 The sketch being finished must of course be left to dry 
 before re-painting. 
 
 As its colours are not thick, and as the oil is partly 
 absorbed in the wood, or in the sizing, it is not necessary 
 to wait as long as if you had sketched on a non-absorbing 
 surface with impastements. A month in a very dry studio, 
 or a fortnight in fine summer sunshine, is quite enough. 
 
 When the sketch is very dry, a general coat of re- 
 touching varnish should be painted over the whole picture 
 before continuing. This varnish, as we said above, dries 
 in a few minutes, and may be immediately painted over. 
 The object of this varnish penetrating into the sketch is to 
 go and fill up all the empty spaces in the coating of colour 
 left by the oil absorbed into the support painted. Filling 
 up the spaces in the coating of colour, it will deposit on its 
 surface a skin of normal resin, which will serve as a link 
 between this coating and the following one, and it also 
 removes dull spots, and gives more brilliancy and trans- 
 parency. 
 
 The Execution. 
 
 We have said that, in the execution, the artist, released 
 from all trammels, might paint to please his own fancy ; we 
 shall, however, remark that excessive impastements, besides 
 not being always seen underneath, are rather destructive to 
 the solidity of the painting, and it is not necessary to make 
 much use of them, since finding always in places which are 
 in the light a luminous sketch, and in dark places a sketch 
 already dark, in order to cover them we shall not require 
 efforts which always show by a thickening of colour, when 
 we want to produce light on a dark ground, or darkness 
 on a light ground. 
 
 Here an objection naturally presents itself, if, in spite 
 
ii6 
 
 SCIENCE OF PAINTING. 
 
 of one's knowledge, one makes a mistake in the sketch, or 
 if one changes his mind — for instance, where he had 
 sketched somebody in blue wishes him to be in red, or 
 if, after having intended to put a coloured tablecloth on 
 a table, it is wished to replace it by a white tablecloth, 
 etc. 
 
 To this we reply that there should be no hesitation in 
 promptly and completely removing the sketch with ben- 
 zine at the place where the change is desired, and then 
 putting in a new sketch as wished. 
 
 Of course it is always better, when possible, to put every- 
 thing in at once without any changing, but it is certain that 
 it is not always possible to do this successfully. Certain 
 portions are sometimes unsuccessful, and glazings are neces- 
 sary. 
 
 When forced to go over a certain piece several times, it 
 should never be done without first rubbing a little re-touch- 
 ing varnish over it, and the same piece should never be 
 re-painted two consecutive times with solid colour : 
 avoid re-painting. 
 
 Re-Painting. 
 
 If a piece does not satisfy you completely — let us suppose 
 it is a head, whose shading you wish to soften, to colour, or 
 to brighten — with semi-liquids and glazing you can obtain 
 what you desire without being obliged to re-paint. But if, on 
 the contrary, the drawing of it was defective, features have 
 to be altered, the brow made higher, the mouth lower, etc., 
 that would necessitate new impastements in light colours on 
 the dark parts and in dark colours on the light parts. This is 
 what is called re-painting. Sooner or later the re-covered 
 parts will show again, and will be a cause of regret. 
 
 Here is a striking example of this kind of accident : There 
 
GLAZING. 
 
 117 
 
 exists at the museum of Madrid an equestrian portrait of 
 Philip IV., by Velasquez, in which the horse has eight legs, 
 not all equally apparent, but still they are discernible. It 
 is evident that four of these legs were not intended by 
 Velasquez. Not liking them as they were at first, he 
 had re-covered them with an earth-coloured tone, without 
 removing them from the canvas, and then he painted new 
 ones in other places. It is absolutely certain that when 
 Velasquez finished that picture, the effaced legs did not 
 appear, as he assuredly would not have left them there ; and 
 it is not less certain that in time they reappeared. 
 
 Re-painted pictures have still another inconvenience, 
 which is, that new colours thickly laid on other thicknesses 
 imperfectly dry at the bottom cause cracks. Here, again, 
 there should be no hesitation in scraping down to the out- 
 line the parts to be re-painted in full paste. 
 
 Glazing. 
 
 As for glazing, it should never be put on dry if it is desired 
 to have it even. 
 
 The following are precautions which it will be well to take. 
 If the place on which you wish to put a glazing has been 
 already covered with re-touching varnish, rub it over with 
 some petroleum oil and wait a few minutes until the varnish 
 gets thoroughly distempered ; then put on the glazing, mixing 
 with the colour a little painting varnish, and a little siccative 
 if the colour be a lake. When the place to be glazed has 
 not yet been covered with varnish, rub a little painting varnish 
 over it and apply the glazing as above. 
 
 Besides these preparations one may sometimes be, in 
 certain cases, desirous of rendering the colour used for 
 glazing more or less liquid, sometimes requiring it to be 
 merely of the consistency of a wash. For this purpose use 
 
ii8 
 
 SCIENCE OF PAINTING. 
 
 essence or petroleum oil, according to whether you wish it 
 to dry quickly or not. There is only one observation to be 
 made — viz., that the more petroleum you put in the glazing 
 the more necessary it is to see that the foundation is 
 thoroughly dry. It is a question of taste whether the colour 
 used for the work shall be more or less liquid or ropy. Each 
 artist, with his two varnishes that he may mix in any pro- 
 portions he pleases and his petroleums more or less volatile, 
 may act as he thinks best : nothing is forbidden, there is no 
 destructive excess to be feared. 
 
 The siccative which solidifies oil, on the contrary, should 
 be sparingly used, whether in paste or liquid. Very little 
 should be put into light tones, none into white lead ; and 
 it should be well mixed into the colour with a knife on the 
 palette, so as to be equally distributed. 
 
 Embus, or Sinkings. 
 
 Where, in the execution of a picture on account of repeated 
 re-touching, embus or sinkings appear, it is possible, as said 
 above, to make them disappear with a light scumbling of 
 re-touching varnish. This operation should even be repeated 
 as often as the stains appear ; for when an embu is obstinate, 
 it means that the interior vacuums are not filled up, and for 
 the preservation of the picture it is requisite that they should 
 be filled up. 
 
 Those embus are the disease of oil painting, and, in order 
 to cure it, it must first be well known. Until now artists 
 have been content with getting rid of its immediate effects, 
 but without studying their causes, and consequently without 
 bringing any remedy to the effects which this malady may 
 have in the future. We are therefore obliged at this point 
 to give some explanations. 
 
 Ground colours contain the oil necessary for keeping 
 
EMBUS, OR SINKINGS. 
 
 119 
 
 together the small parcels of colouring-matter, and for giving 
 to them, according to the laws of refraction in centres of 
 different density, an aspect more coloured and more trans- 
 parent than if these parcels of matter were in a state of free 
 powder; but that on one condition — viz., that the oil shall 
 remain equally distributed amongst all these parcels. And 
 that does not always occur. 
 
 Let us say firstly that, in the case which we are considering, 
 oil does not follow the laws of weight, but it obeys the 
 capillary law, which is that in channels of small dimensions 
 liquids disperse in all directions, as much from bottom to 
 top as laterally or from top to bottom. 
 
 Now, the interstices which the imperceptible parcels of 
 colour leave between themselves, form a number of little 
 canals very propitious for this phenomenon. One of the 
 best proofs of this is that the oil does not descend to the 
 bottom of the picture, although the latter is nearly always 
 vertical : it remains in suspension in the layer of colours 
 which contains it, but, in the thickness of that layer, it is 
 differently distributed according to circumstances. That is 
 what we are going to study. 
 
 When a first layer of paint is applied to a panel, a canvas 
 or any support — in fact, on the plane surface of any material 
 whatever, if that material is porous, such as wood, card- 
 board, a wall of stone or plaster — the oil gets partly absorbed 
 into it and the colour remains more or less dull, according 
 to the quantity of oil it has lost. If the materials on which 
 the painting is done have been prepared by means of sizing, 
 still the sizing will always be a little absorbent ; and this is 
 requisite, so that the oil by penetrating a little may throw 
 out the roots which enable it to adhere. In this case a 
 little oil is used in this manner, and the colour is still slightly 
 dull, but less so than if there were no sizing. 
 
120 
 
 SCIENCE OF PAINTING. 
 
 It would seem that, by continuing to put on one layer 
 after another, the oil from each would filter into the pre- 
 ceding one to fill up the vacuums there formed, so that we 
 should have a succession of layers all dull ; yet this does not 
 always occur, for the following reason : — 
 
 If you tread on sand which has been wet, water will 
 spring up without the soil sinking. If you even make a 
 little hillock of this wet sand and renew the experiment, 
 water again comes to the top. A similar phenomenon may 
 be noticed in the clay used by potters and in a damp wall 
 when knocked. The more you triturate the colour, the 
 more the oil comes up and collects on the top. Under 
 these conditions a layer of colour applied over another, 
 even if the under layer be sunken, remains bright. 
 
 It would still seem that in re-painting over this bright 
 surface the layer should be normal. That does not matter, 
 because the oil, in rising to the surface, has left, in the 
 inside of the layer, some colour deprived of oil : all the little 
 canals spoken of above, produced by the interstices between 
 the parcels of colours, are empty under this exterior sheet 
 of oil, and this is what happens when it is painted upon. 
 The oil, leaving the new colour, penetrates through the pores 
 of the dried oil on the surface of the preceding layer. It is 
 sucked up by the empty canals as if by a sponge, and the 
 result is still an embu. Now, the embu will come whether 
 you re-paint a dull or a bright surface, if that surface, 
 although dry, has not been long dry. But if the bright 
 surface is tolerably dry, and its pores are sufficiently shut to 
 prevent the new oil from penetrating, the painting will be 
 normal ; only the canals at the bottom will remain empty, 
 and as often as there are vacuums in the construction of the 
 painting it will be to the detriment of its durability. 
 
 It is like a wall made of dry stones merely covered outside 
 
EMBUS, OR SINKINGS. 
 
 121 
 
 with a coating of mortar. Should the sHghtest earthquake 
 occur, the mortar cracks and the wall falls. Now, for the 
 construction of a picture the movements of the canvas or 
 the panel are terrible earthquakes. The pomades used for 
 getting out the embus do not penetrate deep enough to 
 fill up the vacuums formed. A new mortar must therefore 
 be filtered into those little catacombs, and for this work 
 petroleum is useful. 
 
 Shall this mortar consist of oil ? 
 
 We said above that there was quite enough of oil in the 
 colours. Its being badly distributed is no reason for using 
 more of it ; it is better to take normal resin which will give 
 solidity, preserve and even increase transparency, without 
 the yellowness and darkness which time always produces 
 with oil. 
 
 For this purpose it is well to use resin dissolved in 
 petroleum ; if this resin dissolves well cold in oil, it is 
 because all the pores will remain shut up by this resin if 
 no re-painting be done, and should re-painting be done, the 
 oil dissolving this resin which shuts up the pores will be 
 able to become rooted. 
 
 This ought to explain how everything intended to get 
 rid of the embu, which is not soluble in cold oil, is alto- 
 gether disastrous. In fact, the want of adhesion makes 
 the picture scale off, and is most disastrous to it. 
 
 By observing carefully all the precepts given above, an 
 oil painting may be executed under the best possible con- 
 ditions for durability, and the maximum of brilliancy and 
 freshness can be obtained from the colours. 
 
 V. p. 
 
 8 
 
CHAPTER X. 
 
 THE PRESERVATION AND RESTORATION OF 
 PICTURES. 
 
 Once upon a time there was a little dealer in Swiss articles : 
 villas, cuckoos, milkmaids and mountaineers of white wood. 
 But, this style of thing being no longer the fashion, the 
 little dealer, in spite of his intelligence and courage, saw his 
 business on the verge of ruin. One day, when no longer able 
 to pay his landlord, he was obliged to forfeit all his goods 
 to him. 
 
 He took refuge, sad and desolate, in a lodging which he 
 took in a hurry. Fortune at last, compassionating his 
 distress, sent to his lodging a messenger in the form of a 
 servant from a large house. The servant thus addressed 
 him : " Sir, the Count, my master requires your services. I 
 have spoken to him of you, because in the neighbourhood 
 I was told that you were the nearest. I might have gone 
 to look for another, so I hope you will not forget my little 
 commission — all the more so as the Count is generous, and 
 if you deal with him properly, his business may amount to 
 some thousands of francs. We understand each other, do 
 we not ? I shall say that you are coming directly." 
 
 The servant ran joyously downstairs, leaving the unhappy 
 little dealer perfectly stupefied, unable to understand what 
 this unknown Count could expect from him. 
 
 Still, hypnotised by the words " several thousands of 
 francs," he mechanically took up his hat and stick to go out. 
 
THE PRESERVATION AND RESTORATION OF PICTURES. 1 23 
 
 As he shut the door, the mystery was revealed to him. 
 There it was ! Expert^ engraved in beautiful black letters 
 on a brass plate. This plate and the title of expert belonged 
 no doubt to the former occupant. 
 
 Might he appropriate them — profit by this mistake ? Oh 
 yes ! A drowning man does not trouble about where the 
 plank which saves him comes from. So, putting remorse 
 behind him, that is if he had any, and resolving to profit by 
 this opening, he had recovered his composure when ringing 
 the Count's door-bell. The servant, who expectei him, 
 ushered him into the drawing-room, announcing with dignity: 
 " The Expert ! " His fate was thrust upon him — he was an 
 Expert ! But expert at what ? 
 
 It was the Count who informed him. First the Count 
 wanted to have his opinion about the authenticity of several 
 old pictures, then to have some of them re-varnished, and 
 afterwards to ferret out the works of promising young artists. 
 For the Count had decided partly to renew his picture 
 gallery. He considered the old masters full of merit, but he 
 thought that an amateur should also interest himself in con- 
 temporary artists, and that it is more interesting to discover 
 them when rising than to follow with the applauding crowd 
 when past their primer. (We see that the Count possessed 
 intelligence far above thé average.) Finally, he counted 
 upon the enlightened assistance of the learned Expert to 
 help in realising his plans. The Expert approved in every 
 respect of the Count's ideas ; but he asked, before giving a 
 formal opinion, if he might study the pictures a little nearer 
 ... It was a serious matter ! . . . etc. In short, he 
 played his rôle admirably well, and when they separated, 
 the Count and the Expert were mutually pleased with each 
 other. 
 
 Next day, the Expert's brass plate, brightly polished, shone 
 
124 
 
 SCIENCE OF PAINTING. 
 
 like the sun on the door of the former dealer in Swiss 
 articles. 
 
 He knew already that he was doing no harm by keeping 
 the title, as it is one which anybody may take. He took 
 into partnership a picture restorer, as he was ignorant of 
 the first step in his new business. He attended sales, visited 
 artists, and in a short time he became a most important 
 personage in the commerce of pictures. 
 
 This anecdote, perfectly authentic, is to show how 
 indifferent certain amateurs are as to the competence of 
 the people to whom they confide works of whose great value 
 they are perfectly aware. 
 
 Were it a question of a valuable dog or horse, those same 
 amateurs would not only call in to attend them a properly 
 qualified veterinary surgeon, but they would require a 
 substantial guarantee as to the skill of this Esculapius; 
 whereas for the care of a picture the first comer will do. 
 
 We do not wish to depreciate experts in general. Some 
 of them are profoundly learned in their art, but there are 
 others quite ignorant. What we blame is the thoughtless 
 selection that is made from them; all the more so that 
 the pictures abandoned to their care have often damages 
 much greater than varnish worn off. They might be repaired, 
 for restorers exist who are masters of their art and do 
 wonders ; but if they fall into unskilful hands, they are 
 irretrievably lost. And our indignation is roused to think 
 that the support of a minister, the recommendation of friends 
 in good positions, is enough to cause the treasures of our 
 museums to be scoured by such hands. 
 
 The end of this will be, that no one shall be allowed 
 to take the title of expert in pictures without having passed 
 regular examinations, such as are passed by lawyers and 
 doctors. 
 
THE PRESERVATION AND RESTORATION OF PICTURES. 1 25 
 
 This would not give to all experts great ^kill, but it 
 would at least prevent any from being altogether ignorant. 
 
 We have left the picture carefully done and substantially 
 constituted. But on leaving our hands, its own life begins, 
 and that is sometimes a hard one. 
 
 The pictures which are not appreciated are the least to 
 be pitied, for, as we have just seen, it is especially from the 
 attentions bestowed upon them that pictures suffer most. 
 
 The first operation to which they are submitted is 
 varnishing. 
 
 Varnish is the dress of the picture ; it preserves it from 
 destructive emanations, it protects its surface, it enriches its 
 colours, and when it is faded, soiled by contact with impurity, 
 frayed, yellow, blackened by time and smoke, altogether 
 worn out, it is taken off and a new one is put on. Clearly, 
 it is a dress ! But a dress fitting so closely that we might 
 almost call it a skin, and a skin cannot be changed like a 
 dress. Each time that a picture has the varnish taken off a 
 little flesh comes with the skin. Unless very great pains are 
 taken the process is one of flaying. Wishing to have the 
 pictures ready too soon, they are always varnished too fresh, 
 in which case the varnish penetrates yet more into the 
 painting. We are not going to speak as did a celebrated 
 practitioner of " the varnisher's art " ; yet to varnish well 
 requires great care and long experience. {See Appendix — 
 " Varnishing.") 
 
 When the picture is varnished, it may be left alone for 
 some years ; but it is seldom so lucky. 
 
 Amateurs have a habit, when looking at a painting, of 
 making a circle with their finger to indicate the part which 
 they admire. 
 
 The little circle is imaginary ; but not so the knocks from 
 the admirer's nails. It is even fortunate if the excited 
 
126 
 
 SCIENCE OF PAINTING. 
 
 patron does not hold a lighted cigar or a metal eye-glass too 
 near the picture. 
 
 Sometimes the admirer will even rub the remarkable place 
 with a little saliva. The remarkable place consequently 
 becomes of a dark blue colour, and the darker it gets the 
 more the washing with this nicotine saliva is repeated. So 
 that, if the picture contains many remarkable places, it soon 
 becomes quite disgusting. 
 
 When the amateur leaves home, you might expect that 
 the picture would have nothing more to suffer. . Ah, but yes ! 
 Servants do not take the trouble to draw down the blinds 
 when the sun is at its zenith, and, during the summer 
 months the unfortunate picture receives every day at the 
 same hour a burning ray of sunshine, under which the 
 ^ varnish swells, to crack afterwards as it cools. And the 
 great cleaning days ! the knocks from feather brushes, whose 
 broken feathers scratch it in all directions. And the flies ! 
 it is really quite remarkable how they always collect on the 
 lightest places, not exactly to eat the picture ! 
 
 Then, winter comes back, with its cigar-smoke and coal 
 fumes and general admiration ; until some day the amateur 
 gets tired of the picture (which is very likely ! ) or perhaps 
 a tempting price is offered him for it, or he has losses in 
 "business, or ... he dies ! At any rate, sooner or later, the 
 picture . changes owners ; then the expert appears, which 
 always means cleaning and re-varnishing. Poor picture ! 
 There are unscrupulous people who, to get on faster, instead 
 of taking off the varnish with the finger, use alcohol, which 
 takes off the painting, or they even re-varnish it without 
 cleaning i't, for we cannot call what they do to it cleaning. 
 We have seen a very celebrated expert wash all the pictures 
 in a gallery, estimated at some millions of francs, with the 
 same wash-leather and bucket of water ; the dirt from the 
 
THE PRESERVATION AND RESTORATION OF PICTURES. 1 27 
 
 pictures of Rubens passed over to the pictures of Teniers, 
 and so on, but all were as dirty at the end as at the 
 beginning. 
 
 The precautions to be taken for the preservation of 
 pictures are, therefore, to varnish them carefully with good 
 varnish, to protect them from all abrupt changes of tempera- 
 ture, and from any chance of being touched or knocked 
 about, and to keep them always in a state of the greatest 
 cleanliness, that the taking off of the varnish may be seldom 
 necessary. 
 
 But a picture may contract other maladies yet more 
 serious than being merely dirty. These maladies come 
 sometimes from causes in the painting itself when that has 
 been carelessly done. 
 
 The maladies inherent to painting are of two kinds : 
 
 1. Change in the colours. 
 
 2. Defective solidity in the colours on the support. 
 The change of colours is due to two kinds of causes : — 
 
 1. Chemical causes; 
 
 2. Physical causes. 
 
 The chemical causes of change in colours are the influence 
 of light, which makes them fade or change tone, and the 
 reaction which produces and destroys their combinations. 
 
 This reaction is mostly the result of contact amongst 
 themselves, or with oil, varnish, etc., which fix and cover 
 them, or with the gases contained in the atmosphere. 
 
 All the alterations proceeding from these chemical causes 
 are irremediable, and often by trying to cure them the 
 existence of the picture is compromised. 
 
 As for the changes which have physical causes, they are 
 occasioned by the presence of gluten and disorganised 
 varnish, which, placed over each other, darken the colours 
 without changing their nature, or sometimes by defective 
 
128 SCIENCE OF PAINTING. 
 
 drying causing cracks, etc., or any other accident which 
 may interfere with the continuity of the painting. 
 
 The physical changes are reparable with intelligent care. 
 The want of solidity in the colours on their support may 
 be owing to the destruction of the support under the paint- 
 ing, or to the lack of adhesion between the support and its 
 sizing, or between the sizing and the painting. 
 
 The destruction of the support is produced : if it be a 
 panel, by dampness which rots it, or by worms which devour 
 it ; if it be canvas, by dampness which rots it, or by oil 
 which burns it. 
 
 The want of adhesion of sizing on a sound support arises 
 always from the sizing being of bad quality ; either it has 
 been affected by damp, or it has become spoiled in itself ; 
 but in either case there is no adhesion amongst its particles. 
 
 The want of adhesion of colours on the sizing is the 
 result of the same causes. The roots thrown out by oil will 
 not hold in a sizing without consistency ; the repeated 
 stretching of the canvas, as well as the play of the wood, 
 hasten the catastrophe. The repair of all these accidents 
 necessitates two operations : re-canvassing, done by gumming 
 a new canvas behind the old one to consolidate it; and 
 transferring, which consists in taking the painting completely 
 off its panel, canvas, or even off its wall to transfer it to a 
 new panel, canvas, or wall. 
 
 Let us say at once that when such operations are deemed 
 necessary they should be done by professional repairers ; 
 we have some who are true artists — extraordinarily clever, 
 and who have done the work of restoring in a manner more 
 wonderful than any painter could have done. We venture 
 to advise to those restorers the use of caséine paste, to the 
 exclusion of gelatine or rye pastes, which are much more 
 susceptible to damp. 
 
CLEANING. 
 
 129 
 
 Trivial accidents, such as split or bruised canvas, blistering 
 of paint, caused by excess of heat coming either from the 
 sun or a stove, little creases, slits occurring during the exe- 
 cution of the picture, through any neglect of requisite pre- 
 cautions, in short, all the accidents of the craft, daily occur- 
 ring, the artist himself should be able himself to repair; 
 and to do this he will find the simplest and most practical 
 means described in the Appendix ("Repairing of Small 
 Everyday Accidents 
 
 Cleaning. 
 
 Pictures sometimes only require to be cleaned. 
 
 In all cases, they should be cleaned before re-varnishing, 
 and the complications to be met with are not always easy 
 to foresee. 
 
 Besides dust, smoke, and all the other impurities with 
 which paintings become covered in course of time, it is 
 difficult to imagine all that the caprice of different owners 
 may have put on it : egg varnish, oil scumbling, bacon rind, 
 vaseline or other grease, burnt wax, carriage varnish, etc., 
 even collodion. 
 
 Each of these substances requires a different solvent, and 
 we require a drug store at our disposal to clean perfectly 
 certain pictures which perhaps may be only a few years old. 
 {See Appendix — " Cleaning.") 
 
CHAPTER XI. 
 
 WALL FAINTING, 
 
 For the style of painting called fresco painting, one of the 
 earliest in use, the sizing is of lime or mortar coloured with 
 paints ground and distempered with water, and applied 
 whilst this mortar is still fresh. 
 
 .The colours penetrate into the- mortar, and the solidity of 
 the painting depends upon that of the sizing itself. 
 
 The colour does not leave the sizing, but often the sizing 
 leaves the wall. 
 
 The wall to be painted is first entirely covered with mortar 
 made of lime and sand, put on about an inch thick ; this 
 first bottom layer making a rough surface is called rough- 
 casting ; over this rough-casting the sizing proper is applied 
 composed also of lime, but less coarse, and of finer sand. 
 Only as much as can be painted that same day is prepared 
 at once, and the painting is only done when the sizing 
 is sufficiently set to bear the pressure of a finger without 
 sinking. Long-haired brushes should not be used, as they 
 are apt to stick in the lime. 
 
 Re-touching is not possible. The sized part that has not 
 been overtaken during the day, and any parts that may have 
 been spoiled, must be demolished down to the rough-cast- 
 ing, and recommenced next day. 
 
 In truth, it is a barbarous and inadvisable process, only 
 spoken of here because it holds a high place in the history 
 of art, and we owe to it several masterpieces. But the 
 
WALL PAINTING. 
 
 bad condition in which 'they are, "to speak only of the 
 frescoes of Raphael and Michael Angelo, when compared 
 with pictures by the same masters executed in egg or oil, 
 does not do much to recommend a style of painting so 
 difficult to do and whose results are so limited. It is true that 
 means have been proposed and even tried for making fresco 
 painting easier to manipulate and to re-touch, by mixing with 
 the colours paste or mucilage enabling it to be resumed the 
 following day ; but that is really like distempering in egg, 
 fig or cheese paste, etc., according to the material employed. 
 The paint thickened and kept back by the suj)stance 
 added to it can no longer penetrate into the mortar, thus 
 diminishing its durabihty, and it is then quite useless to 
 paint over the lime, as it may destroy certain colours. Only 
 those whose use is applicable to the process should be 
 employed, i.e. carbonate of lime, white earths, enamel and 
 ultramarine blues, smoke black. Cinnabar, or vermilion 
 may also be used, but they miist be prepared with lime- 
 water, and that diminishes their brilliancy. 
 
 Wax has also often been .used for painting walls ; but on 
 walls wax is not sheltered from damp, and that causes it 
 to powder. 
 
 The oil and wax painting invented by Taubenheim 
 seemed for some time to be an improvement ; but the same 
 causes spoiled it, and simple oil painting is now used, but 
 with no better results. 
 
 The question has not been properly gone into. All 
 painting, good on canvas or panel, would be so on a wall ; 
 when the painting is bad, the wall is blamed. 
 
 It has been thought that this difficulty might be got over 
 by putting complicated sizing on the wall. In fact, the 
 colours do hold better on these sizings than on stone, but 
 ^ the sizings themselves disintegrate and fall. 
 
132 
 
 SCIENCE OF PAINTING. 
 
 There is only one way of preparing a wall to make it 
 suitable for preserving painting, ix.^ to size it with resin on 
 condition that the resin penetrates deeply into it. Besides, 
 stones get very easily permeated with resin, and then become 
 so hard that iron tools can scarcely break them. Here is 
 the process : First, instead of carefully filling up all the 
 joints and holes with cement, which sooner or later always 
 gets loose and falls off, you should, on the contrary, 
 thoroughly clean all the interstices, and put bare stones 
 everywhere. Afterwards dry it as much as possible by 
 means of a chafing-dish passed up and down before all 
 parts of the surface. 
 
 When the wall is thoroughly dry and still warm, give it a 
 first coating of resin dissolved in volatile essences, and when 
 this varnish is set, by means of heat make it penetrate into 
 the interior of the wall, repeating the coatings of varnish 
 in the same way, until the stone refuses to absorb any more : 
 and only then fill up any holes with a mastic made of the 
 same varnish, fine sand and wax. The last is applied hot, 
 with a glazier's knife. The wall thus prepared is now ready 
 to receive the sizing you select a§ suitable for the kind of 
 painting that you intend doing. 
 
 Should oil painting be chosen, first apply a thick coating of 
 re-touching varnish which, adhering by affinity with the resin 
 already imbibed by the wall, will serve as a link between 
 it and the sizing that you will next apply, composed of white 
 lead ground in oil and distempered with painting varnish. 
 
 If wax painting be the style that you choose, the sizing 
 must be composed of zinc-white ground with essential oil of 
 petroleum and wax dissolved in re-touching varnish ; this 
 sizing being dry, should be heated so that the materials 
 of which it is composed may incorporate properly with the 
 resin in the wall. 
 
WALL PAINTING. 
 
 Finally, should it be water colours that you choose, — 
 yes, water colours, and that would be the best process, — the 
 sizing must be composed of resin and zinc white, ground 
 together with water : four parts of resin to five parts of zinc 
 white by weight. After being applied and having dried, it 
 must be heated so that the resin in melting may assimilate 
 with the zinc white {see chapter on Water Colours, — " Water 
 Colours fixed by means of fire 
 
 This sizing may be rubbed when quite cold with 
 pumice-stone powder and cold water, and other coatings can 
 be applied above, until the surface becomes as smooth as 
 semi-vitrified porcelain of dazzling whiteness. 
 
 Over this hard sizing you should paint with colours 
 specially ground with a solvent and glycerine. Those 
 colours keep fresh for months if desired, and if you wish to 
 dry them you have only to heat them : the solvent melts like 
 the enamel of earthenware, and when all is cold, after a few 
 minutes' interval, wash with fresh water to take off the 
 superfluous glycerine. Re-touches are always possible, and 
 the painting is indestructible, almost as flat as pastels. It 
 may be washed with black soap or potash, and resists the 
 most violent acids. This process may be applied to all 
 sorts of materials. We had invented it to do water colours 
 on wood and even on paper. We shall have occasion to 
 speak of it again. {See chapter on Water Colours — " Water 
 Colours fixed by means of Fire.") 
 
 It is very easy to add wax to these colours, whether in 
 painting or afterwards, rendering the wax invisible with water 
 by means of ammonia {see Appendix — "To render wax 
 miscible with Water and Glycerine") ; in this case, it is melted 
 with resin after using the hot chafing-dish, and the painting 
 obtained by this means resembles the wall paintings of 
 Pompeii which are always, although wrongly, called frescoes. 
 
134 
 
 SCIENCE OF PAINTING. 
 
 There is still another means of decorating walls besides 
 painting them : that is, to cover them with painted canvas. 
 In this case the canvas is pasted on the bare wall, or on the 
 oil sizing with a paste made of oil and white lead, or with 
 rye paste. This operation is called priming. 
 
 It is better to prepare the stone with resin, as we said 
 above, and to mix picture-varnish with the oil of white lead. 
 The use of paste for this kind of work is absolutely bad. 
 
 On walls thus imbued with resin, especially in halls 
 or theatres where there is much emanation, from gas, 
 you may, by means of water-colour fixing, paste paintings 
 executed in water-colours on paper, which, afterwards covered 
 with several layers of the same fixative, would be as solid 
 as earthenware, and as easily cleaned, for neither potash, 
 acids, essences, nor even benzine, can attack these pictures. 
 
 With these processes, which are not crude dreams, as for 
 eight consecutive years specimens of them have been shown 
 at each exhibition of French water-colour artists, we no 
 longer ought to see what we have hitherto seen : master- 
 pieces covered with smoke in the lobby of the Opera 
 House, which if cleaned at all must be done by rubbing 
 with bread-crumbs for feâr of spoiling them, a process at 
 once expensive and insufficient. What is wanted is a 
 commission of competent men thoroughly to study these 
 methods, and if approving of them, recommend their use 
 to artists, who should upon certain conditions bind them- 
 selves to use them. Is what we desire impossible ? Im- 
 possible is a word that we are told is not in our language. 
 Alas ! we always find it in a question of administration. 
 
CHAPTER XII. 
 
 CRAYONS, DISTEMPERING, EGG-PAINTING, WATER 
 BODY-COLOURS. 
 
 When colours are ground merely with water, it is the 
 medium of cohesion added to them to make them stick on 
 the support, paste, gum, etc., that seems to determine the 
 numerous different styles of painting to which this principle 
 has given rise. But it is rather the quantity of mediums 
 than any difference amongst them which causes the variety 
 of processes. In fact, as the proportion of medium increases, 
 the transparency of the colour increases also, and the aspect 
 of the painting is no longer the same. 
 
 In classing, then, these processes by degrees of trans- 
 parency, we begin with the crayon, which is not transparent, 
 and in which the colours contain so little medium, that they 
 remain in a state of powder. 
 
 Next comes distemper, which is a half pastel. There is 
 only a very little medium in it, but yet sufficient to obtain 
 rather more solidity, and rather more transparency. Then 
 come water body-colours, and egg-painting, in which the 
 quantity of medium continues to increase. We arrive at 
 complete solidity, but not at the utmost possible trans- 
 parency. Ultimately, with transparent water-colour, the 
 quantity of medium always increasing, we can arrive at a 
 maximum of transparency almost equal to that obtained by 
 varnish. 
 
136 
 
 SCIENCE OF PAINTING. 
 
 The mediums used in these different processes, having all 
 the same solvent (/.^., water), and generally having much 
 affinity amongst themselves, can be easily mixed. Con- 
 sequently, there is nothing to prevent us from using several 
 of these processes for the execution of the same picture. 
 We can thus arrive at uniting all the qualities of opacity and 
 transparency desired ; but the solidity, for instance, will not 
 be equal in all the parts, and it is impossible to establish it 
 uniformly, for reasons which we are about to give, in con- 
 nection with the never-ending question of fixing crayons. 
 
 Crayons. 
 
 These colours are ground with pure water. The clay 
 added to them (generally pipeclay) is sufficient to preserve 
 them in the form of a stick when they are dry. Those 
 which do not contain clay are ground with water very 
 lightly gummed. 
 
 It is then in a state of powder that they are put on the 
 support — paper, cardboard, canvas, etc. — and they are only 
 kept on it by mechanical means, so it is necessary that 
 this support should have a rough natural surface, or 
 that it should be made so with a layer of pumice stone or 
 powdered sand mixed into some kind of paste. 
 
 There are some useful points to indicate in the preparation 
 of supports destined to receive crayons. Amongst the 
 numerous causes of destruction in this kind of painting, the 
 two principal are that the powdery colours easily leave 
 the surfaces on which they are stuck, and that not being 
 protected by any gluten, they, more than any others, are 
 susceptible to damp and to emanations from gas. 
 
 Now, the supports generally chosen for painting in pastels 
 are those most conducive to its destruction. Canvas or 
 
CRAYONS. 
 
 paper stretched on a frame is quite a sponge for dampness, 
 and at the same time a drum which every noise and the 
 rolHng of carriages keep in a state of perpetual vibration, 
 the result of which is to make the powdery colour come off 
 quicker still. On the contrary, such materials should be 
 chosen as are little vibratory, and rather thick, e.g., card- 
 board, and secured from damp behind, and in front, 
 either with water-colour fixative, or with painting varnish. To 
 this some people will object that to paint in crayons on 
 stiff objects is very difficult, that crayons break easily, readily 
 crumble, and that the flexibility of the stretched paper or 
 canvas is more agreeable. It is a question of lightness of 
 hand ; but as some people are unwilling to let a requirement 
 of solidity disturb their habitual customs, there is {see 
 Appendix — " To prevent the Vibration of Stretched Canvas 
 or Paper for Crayons") a means of stiffening canvas and 
 paper stretched on frames, after the work is finished. 
 Pumice-stone powder, sand or powdered glass, which give 
 roughness to supports, should also be applied by means of 
 impervious materials such as fixative or varnish, and not 
 with a susceptible paste liable to putrefy. 
 
 If the cardboard used be thin, it should first be. floored 
 to increase its stiffness. 
 
 We may give to the crayon a little more solidity than it 
 has naturally, but to fix it completely is not to be hoped for : 
 that is quite impossible without destroying the velvety look, 
 which is its principal quality. 
 
 To explain this it is useful to refer to the physical laws 
 of light. 
 
 The parcels of colouring matter which constitute crayon 
 colours are all in a free state, kept together merely by pres- 
 sure of air, and held on the support by natural or accidental 
 grapples, which catch on its rough or velvety surface, for the 
 
SCIENCE OF PAINTING. 
 
 power of the slight gluten which keeps them in sticks is 
 broken directly the sticks are rubbed. 
 
 These parcels reflect, by diffuse reflection^ the quantity of 
 white Hght, and the uncoloured rays of the decomposed light 
 absorbed, which determine their proper colour, increase 
 in intensity by the fact that they reflect each other. Further, 
 these parcels often have smooth surfaces, which reflect the 
 white light by specular reflection. Those of the surfaces that 
 are placed at a propitious angle of incidence, send direct to 
 our eye the light which they thus reflect ; the others send 
 that light on the neighbouring parcels, which are thus 
 further lighted, and their luminous intensity is consequently 
 increased. All the reflections, diffuse and specular, from 
 the parcels reach our eyes through a transparent homo- 
 geneous medium — viz., the air in which they are themselves 
 bathed. 
 
 All these parcels cannot be fixed except by uniting them 
 by some glutinous and transparent material. This material, 
 in order to penetrate easily through the powdery colours, 
 ought to be in a solution, but slightly concentrated in a 
 volatile liquid without influence, and, whether it be applied 
 behind a spongy object, or sprinkled in front by means of a 
 spray, it will penetrate, attracted by the vacuum, through 
 the interstices which separate the parcels, according to the 
 laws of capillary attraction, and will come to accumulate first 
 round those which are nearest to the support. 
 
 If this material is in small quantity so as only to seize 
 the parcels in the lower part of the layer of colours, it will 
 solidify them without the appearance of the crayon being 
 thereby changed, because those imprisoned parcels are 
 just what we do not see ; but the upper part of the layer 
 will not be fixed. 
 
 If we increase the quantity of glutinous material so as 
 
CRAYONS. 
 
 to ' reach the surface of the layer without covering it, the 
 solidity will naturally increase, since all the parcels will be 
 laterally bound by the agglutinative — that . is to say, they 
 will be bathed in a transparent medium of greater density 
 than air, and there will be produced, for the upper parcels 
 that we see, new effects of refraction which will somewhat 
 modify the sensation that they produce \ but the external 
 surfaces of the parcels which are the most visible not being 
 yet covered by the agglutinative, and consequently continu- 
 ing to give the first sensation, the general aspect of the 
 crayon will be little changed. 
 
 At this stage a distemper is being made of it : it no 
 longer falls in powder when shaken, but still it comes off 
 when rubbed ; if required to fix it entirely, the quantity of 
 agglutinant must be increased until it covers all the upper 
 parcels to a sufficient thickness. 
 
 Then these parcels are quite drowned in a transparent 
 medium thicker than air, and they change entirely. 
 
 This brings us to the aspect of a water colour more or less 
 impasted. 
 
 The crayon is quite fixed, but it has no longer either its 
 velvetiness, bloom, or dull softness ; it is no longer crayon. 
 
 So, in fixing crayon, it is better to stop short at half-fixing, 
 corresponding to the consistency of distemper. 
 
 This process is especially good for outlines and founda- 
 tions, which thus acquire a comparative solidity, and can 
 spare a little of their bloom as they will afterwards be 
 re-painted. A mistake is made, in using for this purpose 
 fixatives with a basis of gelatine, which they all have, from 
 the first invented by Latour more than one hundred years 
 ago, to the most recent — which, privately, is exactly the 
 same. 
 
 A substance is thus introduced into the colour that 
 
140 
 
 SCIENCE OF PAINTING. 
 
 attracts damp and causes fermentations of which the result 
 is often the loss of the crayon whose guarantee is desired. 
 
 Only incorruptible and impervious substances should be 
 used as agglutinatives. 
 
 Now, be quite convinced of this principle — that, whatever 
 the process of painting used, the results arrived at will be 
 the same. 
 
 By grinding any pigment with petroleum as well as with 
 another volatile liquid, either alcohol or water, the petroleum 
 once evaporated, that pigment would be crayon. By adding 
 an agglutinative soluble in petroleum, oil or resin for example, 
 in a very small quantity at first, the pigment would take, 
 when once dry, the consistency and appearance of distemper ; 
 and in proportion as you increased the quantities of oil or 
 resin, you would see solidity and transparency increase. 
 
 In short, whether the pigment be ground with more or less 
 of any agglutinative, or whether more or less agglutinative be 
 made to penetrate into a layer of colour already laid on, it 
 comes to the same thing, and there is no getting out of this 
 dilemma : — either the colour remains in powder and is not 
 solid on the support, or an agglutinative is added which 
 solidifies it, but then it is no longer in powder nor has it 
 any longer the qualities of crayons. 
 
 The colouring-matter used for crayons should be the same 
 as those considered substantial for other kinds of painting. 
 Carbonate of lime may even be used. It gives a desirable 
 white, in the sense that it is not affected by gas emanations 
 and cannot be used with oil because it does not adhere. 
 
 Unfortunately, manufacturers of crayons are induced more 
 than any others to give unsubstantial productions, because 
 with these same carbonate of lime whites dyed with aniline 
 dye, they obtain crayons of a good consistency and of 
 magnificent tones, but which fade on exposure to light. It 
 
DISTEMPER PAINTING. 
 
 141 
 
 would ,be well, therefore, for these colours, to exact the 
 same guarantees as for oil colours, and to be particular 
 about knowing whether, in the mixtures made by the manu- 
 facturer to prepare the gradations of broken tones which he 
 presents to us, there be no substances liable to produce 
 chemical reaction on each other. 
 
 Distemper Painting. 
 
 In this process the pigment is ground with water and 
 distempered in painting with animal gluten, rendered 
 liquid in a balneum. We shall only just mention this style 
 of painting, which is scarcely ever used except for theatre 
 decorations. 
 
 It is not the object of this book to teach all the different 
 styles of painting which it describes, but merely to give to 
 artists, whom we suppose to be already experienced, the 
 means of rendering their work more durable. Therefore, 
 why stop to consider a process which is only used by some 
 of them for outline work, especially as we consider this use 
 of it bad and only desire to dissuade them from using it ? 
 In fact, as distemper is much too absorbent to be re-painted 
 in oil, a preliminary layer of oil or varnish should be put 
 on, and this darkens the distemper considerably, because 
 it always resumes the tone that it had when wet ; all the 
 benefit of its characteristic quality of freshness is thus lost. 
 A good plan is to increase the quantity of paste, thus making 
 of it a water colour in paste, instead of in gum. 
 
 In this condition it is less absorbent and does not change 
 tone under the layer of varnish or fixative, with which it 
 should be covered before painting in oil. Only, gelatine in 
 so large a quantity has the fault of scaling off ; and as, for 
 other reasons already explained, we exclude it from sizings, 
 why bring it into outline work ? 
 
142 
 
 SCIENCE OF PAINTING. 
 
 Cheese paste, which has not the same inconveniences might 
 be used, but of it also a large quantity would have to be put. 
 
 Now, as we have just seen, it is not the kind of paste or 
 gum used which makes of a painting distemper, or water 
 colour: it is the quantity put in. In distemper, the 
 quantity of paste is so small that the parcels of colouring- 
 matter are held together without being covered ; in water 
 colours, on the contrary, the quantity of gum is large enough 
 for the parcels to be completely drowned in it. 
 
 Then with gelatine or cheese paste or gum, you can 
 graduate from distemper to water colour. 
 
 Well, it is in water colour, and not in distemper, that 
 outlining should be done, i.e,^ with much paste or gum in 
 the colour ; and as gum-arabic is more transparent than 
 paste, we prefer it for this purpose. 
 
 As for those authors who have written that Veronese 
 outlined in distemper, they have not stated definitely the 
 precise use of the word, and from this mistake have come all 
 the unfruitful attempts which have been made to recover 
 this process of the Venetian masters. Yes, they have 
 outlined in paste, but with sufficient paste for the sketch to 
 be transparent and impervious to oil in the necessary 
 measure, as water colours are, and not floury and absorbent 
 like distemper. 
 
 Water Body-colours. 
 
 Opaque water colour is merely distemper \ only its colours 
 are distempered with gum-arabic melted in water, instead 
 of animal gluten. We have, therefore, nothing more to say 
 on this subject. 
 
 Egg Painting. 
 This process, one of the most ancient, lends itself to a great 
 many combinations. It may extend from distemper to water 
 
EGG PAINTING. 
 
 colours, from the greatest opacity to the greatest trans- 
 parency ; and as it allows the addition of resins and of wax, 
 it may acquire a solidity which oil painting will never have. 
 
 Formerly the colours were ground with water, and dis- 
 tempered with yolks of egg fresh emulsioned with cold water 
 (this is egg distemper) ; to this was sometimes added natural 
 white of egg or albumen to increase transparency. To 
 introduce resins, they were first dissolved in essence and 
 as varnish mixed with the egg, either directly or emulsioned 
 with water. 
 
 To introduce wax, it had first to be rendered miscible 
 with water, through the medium of an alkali, and at that 
 time only one was known — viz., lime. 
 
 It is by cleverly-used mixtures that those paintings of 
 the middle ages, were obtained, which have stood out for 
 many centuries on damp walls, and which have often been 
 mistaken for frescoes. Similarly, many pictures on wood, 
 copper, slate, or other materials painted in the same way, 
 have been looked upon as oil paintings. The reason of this 
 mistake is that, in these paintings the flesh shadings have 
 been so harmoniously finished that they could only be done 
 with a viscous paste long malleable, like oil colour^ but the 
 painters of the middle ages obtained these same results by 
 directly triturating the yolk of egg without water, but with 
 resin; the oil contained in the egg dissolved the resin, 
 and colours dissolved with this paste had absolutely the 
 consistency of oil painting. 
 
 To-day, if we seek to imitate this style of painting, it 
 is not necessary to take eggs ] it is easier to analyse the 
 materials contained in those mixtures and, by taking only 
 those which are useful, thus reproduce this ancient process. 
 
 What is there in an egg ? 
 
 The yolk and the white. 
 
144 
 
 SCIENCE OF PAINTING. 
 
 What is there in the yolk ? 
 
 There is some vitelHne, analogous to caséine, an oil, 
 besides some sulphur and other materials that we need not 
 consider. 
 
 What is there in the white ? 
 
 Albumen and other matter not required for our purpose. 
 
 To reconstitute the useful parts of the egg, we therefore 
 only require caséine, oil of egg, and albumen. We can 
 introduce into it resins by dissolving them directly in oil o^ 
 egg, introduce into it wax with volatile alkali (ammonia), 
 which transforms at the same time caséine into paste ; and 
 we have then the egg painting, re-made chemically and purer 
 than before, since we have not the useless parts of the egg — 
 amongst others, sulphur, which would even be destructive, 
 if used with white lead. A painting executed by this process 
 is extremely solid, especially if covered over with a final 
 varnish of good quality. For this purpose we might try a 
 varnish composed of normal resin dissolved in egg oil in 
 the proportion of two-thirds resin to one-third oil. As this 
 varnish would be much too thick to be used, it might be 
 applied with volatile petroleum, which, once evaporated, 
 would only leave the oil and resin. But this varnish dries 
 very slowly, it remains sticky for weeks, and it would be 
 necessary to keep the picture sheltered from dust during 
 all the time it was drying ; therefore we do not give this 
 formula as a very practical one. 
 
 Oil of egg can also dissolve copal, and decomposes it 
 much less than any other oil, because the boiling point of egg 
 oil is just the same as the disintegrating point of copal — 370° 
 Centigrade — and the copal being well immersed in the boiling 
 egg oil, cannot thus be brought to any higher temperature. 
 
 The solutions of copal in the egg oil are uncoloured, and 
 there is no loss of weight. 
 
EGG PAINTING. 
 
 Varnishes thus obtained are therefore much stronger than 
 those of linseed oil, and do not get yellow. 
 
 All this, as we have said in connection with egg oil, is 
 perhaps prospective, but, for the present, no one has yet 
 studied this question, for the few experiments that we have 
 been able to make cannot be instanced as a sufficient study. 
 We have, alas ! been unable to continue our experiments, 
 for the never-ending reason that life is short, and that to 
 undertake everything is a sure way to accomplish nothing. 
 
CHAPTER XIII. 
 
 WATER COLOURS, 
 
 We have at length arrived at water colours. 
 
 Under this name is meant, to-day, a very complex style 
 of painting which includes impastements, opacities, and 
 transparencies. It is, in different proportions, a mixture of 
 the ancient paintings on missals with opaque water body- 
 colours and washes. All these styles of painting, which have 
 for liquid, water ; and for agglutinative, gum-arabic, are of the 
 greatest antiquity ; but water colour properly so called, which 
 is only a wash in colours, and where the paper is reserved for 
 the lights, scarcely dates back beyond the beginning of this 
 century ; for the artists who, before that time, used washes 
 for their sketches, confined themselves to a few neutral tints 
 which never went beyond a camaieu. Of all kinds of 
 paintings that can be done with colours ground with gum, 
 having the same causes of destruction and requiring the 
 same care for preservation, we shall not make special 
 articles ; we shall confine ourselves to giving the necessary 
 instructions for doing water colours in general, in a sub- 
 stantial manner. 
 
 Paper. 
 
 Water-colour paper should be made only from linen rags, 
 and should be bleached only by the action of pure water, 
 air and sunshine ; but it is very seldom that such is to be 
 found. 
 
PAPER. 
 
 Cotton is introduced into it and, as it is sold by weight, 
 kaolin or other heavy substances. To bleach it, chlorine or 
 other chemical agents are used, which burn the linen and 
 are often destructive to the colours. 
 
 The sizing, which prevents it from being too absorbent, 
 should be equally distributed in the paste and not put on 
 the surface just at the last ; otherwise, as soon as the outside 
 gets worn out by prolonged work, the inside is found to be 
 irregularly spongy, and this causes stains. Not only the 
 paper should be dry when used, but it should never have 
 suffered from damp from the time of its manufacture, 
 because then the sizing ferments, decomposes and loses its 
 properties. 
 
 English manufacturers, to whom we are indebted for all 
 water-colour requisites (one scarcely knows why !), send 
 enormous quantities of their paper into France, and, for 
 reasons of economy, the goods are conveyed to Paris by 
 water ; so that sometimes the paper remains for weeks at the 
 bottom of the hold in the ships, when arrived at the quay. 
 
 Besides, the business being carried on by wholesale 
 dealers, the latter, when they receive the reams of paper, heap 
 it up on the ground floor of warehouses, sometimes even in 
 cellars, until the paper is delivered over to the customers. 
 After that the paper dries : but it is spoilt. 
 
 We should either make our paper ourselves, which is not 
 practicable, or a society of important artists should arrange 
 with a manufactory (everything tends to this) to obtain paper 
 in good condition. 
 
 Whilst awaiting this golden age for artists, there is only 
 one means of getting good paper ; and that is by choosing 
 it. A paint box should be taken to the colourman's, the 
 trial of a sheet of paper there made and, when found suitable, 
 a stock of it should be bought. 
 
148 SCIENCE OF PAINTING. 
 
 Paper well preserved, in a dry place, at a moderate 
 temperature, does not spoil : on the contrary, it improves. 
 
 As for any observations that you might make to your 
 colourman for him to repeat to his dealer, thence to pass by 
 the traveller to the manufacturer, they would be only trouble 
 wasted. In this respect paper is not alone : it is the same with 
 every other article ; and the objections of artists, even when 
 of high reputation, are of no more importance to these 
 gentlemen than are the cries of the sheep in the slaughter- 
 house. 
 
 Gums. 
 
 The pigments for water colours are ground with gummed 
 water ; but the choice of the gum is of some consequence. 
 Gum-arabic was excellent for this purpose, unfortunately it 
 can no longer be used; for the very good reason that it is 
 not to be had. 
 
 How ? — it is not to be had ! All the druggists sell it. 
 . Yes, they sell a gum which they call arabic. 
 
 Is it then false ? 
 
 It is false in one sense, i.e. not being the genuine, it is 
 false ; but the genuine, no longer existing, there cannot be 
 any false. And then, such as it is, it is imitated. It be- 
 comes then genuine, since it is imitated. 
 
 If this somewhat strained reasoning suffices to quiet the 
 consciences of the druggists, their gum-arabic is none the 
 less very different from that of former times. It is obtained 
 from the same kinds of acacias, that have been planted in 
 Senegal and elsewhere, but which are no longer in the same 
 climatic conditions. 
 
 It is supposed that in Arabia the gum of these same 
 acacias owed its particular qualities to the heat of the sun 
 to which it was exposed for a long time on the trunks of 
 
GUMS. 
 
 149 
 
 trees ; and that the trees could only support the high 
 temperature of the country by the help of subterranean 
 rivers in which the trees plunged their roots. The explana- 
 tion is that these rivers are now dried up and become so 
 finally ; in the countries where the gum acacias grew, they 
 have now totally disappeared. It is certain that heat is a 
 necessary agent ; for, by subjecting the gums of Senegal to 
 high and progressive temperatures in stoves ; chemists have 
 now succeeded in restoring some of the qualities of the old 
 gum-arabic. 
 
 It is then understood that none but the gums of Senegal 
 and other provinces are found in the commerce of to-day, 
 and that those are the gums which are made suitable for 
 water-colour painting by subjecting them to a preliminary 
 treatment. 
 
 In these gums there is naturally much variety, and also 
 counterfeit imitation. Some of them contain no gum at all. 
 
 Here is a specimen : From dextrine, gelatine or other 
 cheap materials melted in water, a thick paste is prepared 
 which is passed through a strainer with holes of unequal 
 diameter, the edges of which are roughened with a pointed 
 file ; the paste comes through these holes by its own weight, 
 and the drops have their surface marked with stripes like the 
 true gum. When the pieces are dry they are broken, and 
 there is the gum ! 
 
 Is it sufficiently Arabic ? 
 
 It is necessary that the water colours should be easily 
 spread on the palette ; but at the same time they should be 
 sufficiently insoluble for it to be practicable to put a second 
 tint over a first without taking the first off. 
 
 This correct degree cannot be obtained for all colours 
 with the same gum. For almost all the gum that has been 
 through the stove is necessary, because it distempers easily ; 
 
SCIENCE OF PAINTING. 
 
 but with others, the natural gum of Senegal is preferable, and 
 for others such as emerald green, for instance, dextrine must 
 be had recourse to. 
 
 Cnromates render the gum insoluble ; and emerald green, 
 when not well manufactured, is affected in this way. 
 
 The colouring materials that may he used for water 
 colours are the same as for oil painting, and equal care 
 should be taken to make sure of their purity. {See 
 Appendix — " Verification of Colours.") 
 
 In the water colours of commerce we find one great 
 fault, especially if it be desired to use them for outline work 
 to be afterwards filled up in oil ; they are not all equally 
 gummed. 
 
 To obtain this equal gurnming, the colours should be 
 tried whilst grinding and gum added to them until the 
 colour, when dry, has not the same tone as when wet. 
 Once arrived at this equal gumming, colours imbibe no 
 more at one place than at another. 
 
 Artists, according to their tastes, like colours more or less 
 damp ; and to satisfy them large or small quantities of honey 
 or glycerine are added : sometimes even sugar is had 
 recourse to ; but all these have drawbacks : first, the sugary 
 matters attract dampness ; if there is any excess, the colours 
 get so soft that, often, some water colours in a box stick to 
 each other. 
 
 But it is not only dampness which sugar attracts : it also 
 attracts flies, and the harm which they do is irremediable. 
 They leave everywhere, in the colour of the fresh tints, 
 marks of their visits like so many little touches from a 
 microscopic sponge. We have once seen a woman's head, 
 very beautifully shaded, attacked by flies, which made her 
 appear as if she had had smallpox. 
 
 Often, to facilitate execution, especially for large pieces 
 
GUMS. 
 
 like skies for instance, very rapid drying is not desired, and 
 it is necessary to keep the paper damp. For this a Httle 
 glycerine may be put into the water, but very little ; or a 
 solution of gum tragocanth or of chloride of calcium may be 
 used. Some ancient miniaturists used for this purpose the 
 slime of snails or the juice of the figtree ; but all these means 
 are more or less efficacious, and are not without danger. 
 
 Nearly all substances which take a long time to dry, only 
 possess that quality, thanks to the affinity which they have 
 for water : this affinity with water gives them the power of 
 retaining it for a long time, but also that of resuming it 
 very quickly, and might transform water colours into 
 barometers. 
 
 A liquid should be sought which should have no affinity 
 for water, but which should naturally evaporate much 
 slower than it. We have found this liquid ; it has even the 
 advantage of taking an hour to dissolve gum, so that the 
 top can be re-painted without dissolving the bottom, and one 
 may shade in full paste, as in oils, for several hours. 
 
 If, after having worked thus for some time, you desire the 
 work to dry up quickly, by putting the water colour in the 
 sun or before the fire this famous liquid will evaporate in a 
 few minutes. 
 
 Only (there is an only) this liquid is not an article of 
 commerce ; at present it is but a curiosity of the laboratory, 
 and actually costs ;^i2 for a quarter of a gallon. 
 
 This is not very available. 
 
 Let us hope, however, that its production at a reasonable 
 price may soon be arrived at, and that water-colour painters 
 may be able to utilise this marvel — for it is one. 
 
 In the process of water-colour painting, the uppermost 
 colours adhere perfectly, as the second layer distempers the 
 first and mixes with it. The only precaution to be taken 
 
152 
 
 SCIENCE OF PAINTING. 
 
 for solidity during execution is to avoid impastements. 
 Should white have been used and a light tone be wanted 
 instead of a dark one, instead of thickening the touch until 
 the dark tone has no more effect by transparency, it will be 
 preferable to wash it off first ; and on the unpainted surface, 
 even if soiled, much less impastement will be required to 
 obtain the same tone. 
 
 Water colour does not contain in itself anything which 
 can change or destroy it ; but it is affected by the action of 
 the air, water, and many other accidents, and if complete 
 solidity be required it should be fixed. 
 
 Fixative. 
 
 Why, some people will ask, should water colours be fixed ? 
 They are generally protected by glass, and it is not usual to 
 wet them with water. To this we might answer that an 
 excess of precaution can never destroy, and that it is well 
 to preserve them from any possible accident. But the 
 fixative has another object. Water colours, even under 
 glass, are often spoilt by many causes. The dampness 
 which penetrates into the paper, facilitating the fermentation 
 of the gum with which the colours are ground, causes the 
 formation of microscopic fungi. The painting is then said 
 to be getting rotten or sticky. 
 
 Excess of heat, produced by the proximity of a fireplace 
 or the rays of the sun through glass, brings the gum to such a 
 state of dryness that it scales off in imperceptible quantities, 
 and leaves the paper, taking the colour with it. The water 
 colour is then said to fade : as an art critic remarked, it 
 forms a breakfast for the sun. 
 
 Now, the object of the fixative is to prevent the sun 
 from breakfasting upon water-colours, and dampness from 
 cultivating fungi on them. 
 
FIXATIVE. 
 
 T53 
 
 Penetrating deeply into the paper, it surrounds the gum 
 and the colouring-matters with a transparent cement, always 
 supple, as hard as glass, incorruptible and quite impervious. 
 It cannot be attacked by any acid, so that colours which 
 cannot be used habitually, because of the influence of the 
 gases in the atmosphere spoiling them, remain, when thus 
 protected, perfectly solid. Besides these two destructive 
 agents, dampness and dryness, which the fixative paralyses, 
 water colours may be ruined in other ways. If left exposed 
 to air, the paper gets yellow in time. Even in a box they 
 catch dust, and even under glass coal smoke reaches them 
 and spoils them. The fixative remedies all this. The 
 paper does not get yellow when the water colour is fixed, 
 and, however dirty and greasy it may become, washing with 
 a clean soft sponge, water and black soap restores its 
 brightness and freshness. 
 
 These are the services which fixative renders to the 
 amateur anxious to preserve the water colours which he 
 buys and which he loves. 
 
 But to the artist, whom modesty forbids to care about 
 the duration of his works, fixative offers other advantages in 
 the help which it gives to his work. 
 
 The whole or part of a water colour may be fixed in the 
 course of execution, to be retouched afterwards above, and 
 that several successive times. 
 
 With a dissolving liquid the whole or part of the fixative 
 may be withdrawn from a water colour which can be restored 
 to what it was before the fixing, so that any amount of 
 re-touching is possible. {See Appendix — " Manner of with- 
 drawing the Fixative.") 
 
 In a case where the colour did not take easily on certain 
 parts of a water colour already fixed, it is only necessary to 
 put with a paint brush a little dissolving liquid on the parts, 
 
 V. P, 10 
 
154 
 
 SCIENCE OF PAINTING. 
 
 and to let it evaporate. The colour will afterwards take 
 readily. 
 
 The fixative is, besides, an infallible detective- of the bad 
 quality of any colours which should be banished from the 
 palette. 
 
 In order to be convinced of this, it will suffice to put on 
 paper specimen touches of all the colours and to fix them, 
 or merely pass above them some dissolving liquid ; all those 
 which contain aniline or gamboge, and consequently are 
 detestable because unable to bear light, will dissolve in the 
 fixative and should never be used. 
 
 Water Colours Fixed by Fire. 
 
 We have spoken, in connection with wall painting, of 
 process which might be used for water colours. 
 
 This process, to which we have given the name of fixed 
 water colours {see Appendix — " Commercial guarantee"), can 
 be used on paper, wood, or stuff, and all materials that can 
 resist the degree of heat necessary for fixing them — from 
 1 20 to 150 degrees. 
 
 The colours enclosed in tubes can be preserved indefinitely, 
 and on the palette remain fresh for months; when used 
 they are dissolved with water, or with half water and half 
 glycerine, if required to be kept a long time fresh whilst 
 working. 
 
 If left to dry on paper they take the appearance of crayons, 
 but they can be again damped by means of a vaporiser, and 
 when fixed by means of heat they resume the appearance 
 which they have when wet. 
 
 The fixing is done by means of a spirit lamp furnished 
 with a bent tube, the orifice of which is moved about in 
 front of the picture; when the colour takes a shining 
 
WATER COLOURS FIXED BY FIRE. T55 
 
 appearance under the influence of heat it is left to cool and 
 washed in plenty of water to take oif the glycerine. This 
 washing can be done under the tap of a fountain, or with a 
 sponge pressed over the water colour, so that the water flows 
 vertically, or, if the object is not very large, by plunging it 
 into a bucket of water several times. Then it is allowed to 
 dry naturally, or the drying is hastened by a lamp held 
 pretty far off, and when all trace of water has disappeared, 
 the lamp is brought nearer ; at the time the water colour is 
 somewhat pale^ and it resumes its colour perceptibly in 
 proportion as the heat increases. 
 
 When all the colours are restored to their primitive tone, 
 the water colour is fixed. It may then be painted over 
 without any fear of melting anything ; it may even be washed 
 with a sponge like an oil painting ; only it must not be 
 rubbed too hard because it is not merely the colour that 
 would come off but the paper which would tear like any 
 wet paper much rubbed. 
 
 Wax may be introduced into this painting. The wax is 
 prepared by dissolving it in glycerine with ammonia (see 
 Appendix — "To render Wax miscible with Water and 
 Glycerine ") : the solution keeps quite well in bottles. 
 
 The wax thus prepared mixes with the colours whilst 
 painting, but the best plan is to apply a layer of it before 
 painting. 
 
 When this wax has been used once in a water colour it 
 will be necessary to apply a layer of wax every time that it 
 is wished to repaint it, because the heat of the fixing makes 
 it melt and the new colours will not adhere to it, but with 
 this layer of prepared wax they adhere perfectly. 
 
 The use of this process requires some practice, as may 
 be understood, and in heating there is a knack of hand to 
 be acquired ; but when one is accustomed to it it is very 
 
156 SCIENCE OF PAINTING. 
 
 easy to do, and the results — of a perfect solidity — hold a 
 middle place between crayons and water colours. It is well 
 to reserve some paper for lights, so as not to be obliged to 
 put impastements of white, and the white should be oxide 
 of zinc, for the carbonates of lead or lime do not bear the 
 necessary heat of fixing. 
 
 When the water colour thus made contains no wax it 
 may be varnished with a light egg varnish, and over that 
 with re-touching varnish, fixative, or picture varnish. 
 
 Water colours without wax may also be varnished with 
 gum-lac, dissolved in water with borax {see Appendix — 
 Gum-lac"). 
 
 When the water colour contains wax, it is preferable to 
 confine oneself to passing over it a layer of wax prepared 
 with glycerine, and heating it ; the wax melts, and the 
 glycerine springs to the surface in tiny drops like lye ; a light 
 washing takes it off at once, and a final heating unites the 
 whole by spreading the wax equally over all the surface. 
 
 When this process is used for coloured substances, such 
 as wood, cardboard, etc., it is well to cover them with a sizing 
 which produces a white surface. In this case to avoid 
 washing and double heating, the white of zinc should be 
 prepared in water without glycerine, and put on very evenly 
 with a swallow-tail ; when this sizing dries, naturally it will 
 be powdery like crayon, and one heating will be enough to 
 fix it."^' If in the course of the work it be wished to change 
 some parts of the water colour already fixed, to get a white 
 surface so as not to impaste, the piece to be taken off should 
 be done with a linen rag saturated with the dissolving liquid 
 (water-colour fixative), and afterwards with water. Over 
 this place a layer of water-colour zinc-white should be put, to 
 
 * The zinc- white ground in water contains the flux which is to be fixed 
 by the heat. 
 
WATER COLOURS IN SARCOCOL. 
 
 reconstitute the sizing. If it is on paper that is useless ; 
 for after one washing in the dissolving liquid, the paper 
 will be found quite white. 
 
 These colours may be mixed with gum, in which case less 
 glycerine should be used in the grinding, and they should 
 be moistened with pure water. They are used just as in 
 water colours, and after being heated there is no need to 
 wash them, as there is no glycerine to take off. 
 
 Colours fixed by heat are more waterproof than those in 
 pure water colours, but are not altogether impervious, as they 
 contain a little gum. 
 
 It is true that recourse can always be had to fixing them 
 with fixative, but then it must be done rapidly, and the same 
 place must not be done over several times, because the 
 fixative dissolves the other materials which unite the colours 
 with the gum. 
 
 Water Colours in Sarcocol. 
 
 We have sometimes used sarcocol for uniting the pigments 
 in water colours. Sarcocol is a gum resin soluble in water 
 and in alcohol. 
 
 It was known to the ancients, and was much used in early 
 times, especially in Italy. Pliny speaks of it in his day as a 
 substance very useful to artists and to surgeons (we may 
 infer this from its name) ; bandages were saturated with it, 
 and used to bind up wounds. Being replaced in pharmacy 
 by cerecloth, and artists not having used it for several cen- 
 turies, it is to-day difficult to obtain ; still it can be got, and 
 nature furnishes an abundance of it ; it only requires to be 
 harvested. 
 
 After being gathered it must undergo some preparation 
 to discolour it and extract the sarcocolline, which is the 
 only part to be used. {See Appendix — " Sarcocolline.") 
 
SCIENCE OF PAINTING. 
 
 Colours ground with sarcocolline, which acts like gum, are 
 of an extraordinary richness and colouring intensity. They 
 have the advantage of softening much less quickly than those 
 ground with gum, consequently allowing glazing over water 
 body-colours, and rendering the superposition of tints much 
 easier. On the other hand, they have the inconvenience of 
 being scarcely soluble when once dry on the palette, and 
 the tenacity of this gum resin is so great that in drying it 
 takes off the enamel from paint-boxes and leaves the tin 
 bare. 
 
 They can therefore only be used easily by keeping them 
 damp in little bottles with large mouths, whence, they are 
 taken with the point of a knife. In this way they can be 
 easily kept, by being careful to put occasionally, half water 
 and half alcohol into the bottles, the presence of the alcohol 
 preventing fermentation. 
 
 These colours may be moistened with water mixed with 
 alcohol in various proportions, and even with pure alcohol. 
 We have thus a water colour drying very rapidly, which is 
 sometimes very convenient, especially with objects more or 
 less waterproof, on which the tints in pure water have a 
 tendency to melt, and to run into each other. Indeed, it is 
 all the more useful that, on materials which absorb little or 
 none of the superfluous liquid used to lay on the colour, this 
 liquid evaporates all the quicker for being but slightly 
 absorbed. 
 
 Colours in sarcocolline, moistened with alcohol, have 
 also the advantage of adhering very readily to greasy surfaces. 
 By this means, painting in water colour may be done over 
 foundations painted in oil. This process is very speedy, 
 and produces wonderfully powerful effects. 
 
 It is very convenient for the decoration of inhabited rooms 
 where the smell of oil paint is feared ; and it gives a very 
 
WATER COLOURS IN SARCOCOL. 
 
 substantial painting, especially if covered with wax or varnish, 
 according to whether a dull or shining effect be desired. 
 
 This kind of water colour cannot be fixed with the fixative, 
 because the alcohol which the fixative contains would 
 dissolve the sarcocolline ; but it is easy to prepare a fixative 
 which would not have this inconvenience. 
 
 One word more, for those water-colour painters who have 
 the bad habit of sucking their paint brushes : sarcocolline 
 has a bitter taste, something like aloes and similar medicinal 
 qualities. 
 
 Finally, all these processes may lend each other a mutual 
 aid, according to the style of work ; we have often mingled 
 several, and sometimes all, together in the same water 
 colour ; but strict attention must be paid to their respective 
 qualities and to the different methods in which each should 
 be fixed. 
 
CHAPTER XIV. 
 
 CONCLUSION, 
 
 In finishing this book, we shall repeat what we said in its 
 preface. It was not our object to write a treatise in order 
 to teach artists that which they know as well as ourselves, 
 nor to teach amateurs that which they can only learn by the 
 help of a professor. We have limited ourselves to trying to 
 destroy the prejudices of ignorance, the destructive practices 
 of routine, and to furnishing to those artists who are willing, 
 the necessary instructions for making the best use of their 
 tools, for procuring substantial materials, and for using them 
 in such a way as shall let their works remain for the longest 
 possible time in the condition in which they leave them, 
 whatever may have been the processes used to produce 
 them. 
 
 The hope that this book will be useful to those who will 
 read it is very natural, but we have still the ambition to 
 believe that it will have good results, even for those who will 
 not read it. 
 
 The kind reception which the pupils of the School of 
 Fine Arts have given to the lessons on the material pro- 
 cesses of painting, and the diligence with which they have 
 attended to them, prove that they begin to be no longer 
 indifferent to those questions ; and our pupils — in them 
 lies the future ! 
 
 The taste for science coming always in proportion to the 
 
CONCLUSION. 
 
 l6l 
 
 Study given to it, we shall soon see the young generation 
 occupied in seeking for lost traditions, and in applying 
 themselves to profit from the lessons of the past. Already 
 the Society of French Artists has appointed a committee 
 which will render, we hope, immense service. 
 
 The bad fabrication of the materials used by artists is an 
 acknowledged fact, and the idea of remedy is beginning to 
 become general ; to-day this question is considered, and 
 yesterday it was thought good form to ignore it. On all 
 sides symptoms manifest themselves of an approaching re- 
 action against ignorance; we see coming a psychological 
 evolution which shall lead painters no longer to blush at 
 knowing their business; and the moment is not far off 
 when a gentleman who shall say in a drawing-room I do 
 not care what becomes of my pictures after they are sold,'' 
 will be thought as pretentious and as ridiculous as would be 
 an architect if he made light of the solidity of the monu- 
 ments entrusted to him to construct. 
 
 This roving school on the fields of fancy has lasted long 
 enough. Therefore let artists resume the ancient route 
 traced by the old masters : they will find there noble 
 examples to follow. Rubens, one of the most learned men 
 of his time ; Vandyck, a distinguished chemist ; Leonardo 
 da Vinci, engineer and mathematician ; Michael Angelo, 
 painter, architect, sculptor and poet, and so many others, 
 have well proved that knowledge does not destroy genius. 
 
 I>et us hope, then, that science and painting, after having 
 been so long separated,! will remain closely united in the 
 future, as they have been in the past, and we shall then 
 consider as a great reward of our work, that we have been 
 somewhat instrumental towards that result. 
 
APPENDIX. 
 
 THE COMPOSITION AND MANIPULATION OF VARIOUS 
 
 PAINTS, 
 
 We have collected into this Appendix all the practical 
 directions necessary to those artists whose love of the 
 science is great enough to induce them personally to 
 make trial of the precepts which this book contains. 
 
 But we do not venture to hope that this part of the book 
 will find many readers ; and by thus calling attention to it, 
 we almost acknowledge beforehand the dulness which is 
 inseparable from it. 
 
 To all those readers who find that statements of figures 
 put them to sleep, to those ultra-refined beings to whom 
 manual work is distasteful, to those lazy brains whom the 
 slightest attention fatigues, to the superficial minds who take 
 no interest in serious things ; finally to all those who only 
 read in order to amuse themselves, we are conscientious 
 enough to say before they turn this leaf : " Beware ! 
 
 Abandon all hope of amusement ! or go no farther ! 
 We are about to teach !" 
 
 Good and Bad Paints. 
 
 Chinese White or White Lead (Carbonate of 
 Lead). 
 
 Formerly this was obtained by a process (called the 
 Dutch process), which was as follows : — 
 
 Pieces of lead soaking in vinegar were put into earthen 
 
APPENDIX. 
 
 pots, which, ranged in shelves, were buried in ditches the 
 bottoms of which were filled with refuse straw and stable 
 manure. The acetic acid of the vinegar formed with the 
 lead acetate of lead, which was changed into carbonate of 
 lead by the carbonic acid escaping from the dung. By 
 chemical agency the same result may be obtained in a 
 laboratory. 
 
 Since that time white lead has been prepared by solution. 
 This is called the Clichy process. The white thus procured 
 is not entirely free from the water used in its preparation, 
 and it does not cover a surface so well. In reaHty the white 
 which is used, is made by processes almost similar to those 
 which are the secret of the manufacturers ; but as the chief 
 object of the manufacturers is to produce an article at the 
 lowest possible price, so as to prevent competition, they 
 have thus arrived at monopoly of production in some articles, 
 so that being unable to procure any other, the artist must 
 content himself with such white as he can get and that is 
 frequently not of a proper consistency. 
 
 It is the same thing with other colours : manufactures on 
 a large scale have extinguished private industry ; and now, 
 in order to obtain paints suitable for artistic painting, it 
 would be necessary either for artists to make their paints 
 themselves (which would be impossible), or for them to 
 arrange to have manufactured for them paints especially for 
 their use ; and this, we hope, will soon be realisable. 
 
 Carbonate of lead, by means of vapours of sulphuretted 
 hydrogen which are often sent abroad into the atmosphere, 
 forms sulphur of lead which is black. It is thus that certain 
 water body-colours and certain white pencils, made of car- 
 bonate of lead, become of a more or less dark grey shade. 
 One may by moistening them with oxygenised water cause 
 them to return to their primitive whiteness, because then the 
 
164 
 
 SCIENCE OF PAINTING. 
 
 black sulphur of lead becomes a sulphate which is white. 
 But if such an accident occurs in a painting, the remedy is 
 of no use because the presence of the oil and varnish is an 
 obstacle to the said transformation. 
 
 Nevertheless, white lead or Chinese white remains, for 
 the time, necessary in oil painting ; because it is the only 
 kind which adheres. 
 
 It must be banished from all encaustic or other processes 
 requiring the help of fire, because it decomposes in any 
 heat exceeding too to 120 degrees. Neither must it enter 
 into any combinations which might contain sulphur, as that 
 would blacken it, nor acetic acid which decomposes it, — as 
 for instance in a distemper, where one would use the cold 
 paste very liquid, to avoid the trouble of the water bath. 
 
 Zinc White (Oxide of Zinc). 
 
 The white of zinc is procured by vapours of zinc oxidised 
 by a current of air. 
 
 It is not alterable like white lead, but it does not acquire 
 with oil the same solidity'; it remains more brittle, is spent 
 more easily and does not cover so well. Yet one can give 
 it more body by compressing firmly powder of oxide of zinc 
 by the hydraulic press before grinding it, or also by grinding 
 first this powder, and with water making it into cakes and 
 drying them in a stove. Zinc white may be used in all kinds 
 of painting : it is indispensable for mixing with vermilion, 
 to the exclusion of white lead, and also for mixing with 
 cadmium, should there be any fear about the manufacture 
 of the latter. 
 
 Finally, it is an excellent substance and very agreeable to 
 use on all those occasions when, employing much white on 
 the support by way of light, much impasting is not required 
 
APPENDIX. 
 
 Spanish Chalk White, Troyes Chalk White, Bougival 
 Chalk White, Meudon Chalk White, Carbonate 
 OF Lime. 
 
 All these whites are solid when used with water, but 
 when used with oil they become grey and transparent and 
 never dry thoroughly ; therefore they should not be used 
 even for water colours, when painting in distemper, water 
 body-colour or any other kind of painting that is to be 
 varnished or touched-up with oil. 
 
 Like carbonate of lead, they do not bear strong heat, and 
 like it they decompose with acetic acid. 
 
 Earths, Ochres, Marls, Iron Lake. 
 
 All these colours, consisting of oxide of iron more or less 
 calcined, naturally fixed on clay or artificially on alumina, 
 are perfectly solid ; but the catalogue of them is difficult to 
 limit, for different kinds of coloured earth are constantly found. 
 
 We mention here some of the colours most extensively 
 met with. 
 
 When we say that those colours are good it is understood 
 that their genuineness is first to be proved ; for it is not 
 enough that they are called by the names which we are 
 about to give, it must still be proved that they are what 
 they ought to be — viz., oxide of iron. 
 
 Italian earth, 
 
 Natural and burnt sienna, 
 Red earth, 
 Rose earth, 
 Yellow earth, 
 Foundation ochre. 
 Red ochre, 
 Prussian red, 
 Venetian red, 
 Antwerp red, 
 Nuremberg red, 
 
 Indian red, 
 
 Colcothar or crocus martis, 
 
 Vandyck brown, 
 
 Red brown. 
 
 Mars yellow, 
 
 Mars orange, 
 
 Mars red, 
 
 Mars violet. 
 
 Mars brown, 
 
 Iron lake. 
 
i66 
 
 SCIENCE OF PAINTING. 
 
 It is quite understood that this list cannot be complete. 
 Oxide of iron is presented and will continue to be presented 
 under many other names until artists shall insist upon . its 
 bearing merely its true name. 
 
 Orange and Yellow Colours. 
 
 We have classed separately those earths, ochres, and 
 marls which are yellow, red, and brown, because those 
 substances form a distinct category ; but all the other kinds 
 of colouring matter, whose chemical formulas are different, 
 we shall class according to their colour, so that they may be 
 more easily recognised, desiring it also here to be noticed 
 that it is impossible for us to give a complete Hst of the 
 numerous fancy names with which the prospectus of an 
 artist's colourman is covered. In the first place we do not 
 know them all, and before this book is printed many others 
 will have appeared. We limit ourselves to the best known. 
 
 Lead, in combination with other bodies, furnishes many 
 yellow colours more or less bad. Such are : — 
 
 Certain yellows of a lemon shade (iodide of lead). 
 Chrome yellows (chromate of lead). 
 
 Orpiment, king's yellow (arsenite of lead), obtained also by 
 
 sulphur of arsenic. 
 Mineral yellow (on a basis of oxy-chlorure of lead). 
 Minium . ^ 
 
 Mineral orange V protoxide and bioxide of lead. 
 Saturn red J 
 
 Straw yellow | s„bsulphate of lead. 
 * Mmeral turpeth J 
 
 All these colours get black or decompose the metallic 
 combinations with which they are associated. 
 
 * The subsulphate of mercury, also called mineral turpeth, could not 
 be used in painting. 
 
APPENDIX. 
 
 167 
 
 The following are also bad : — 
 Ultramarine yellow (chr ornate of baryta). 
 
 Certain yellows of the chrome kind made with chromate of lime and 
 all the mixtures bearing the pompous names of 
 Green Naples yellow, 
 Brilliant yellow, 
 Buttercup yellow, etc. 
 
 The following vegetable yellows are also bad : — 
 
 Saffron yellow, 
 Gamboge. 
 
 And all the yellow lakes which fade on exposure to light 
 and are changed by certain metallic combinations : — 
 
 Wold lake, 
 Yellow lake, 
 Yellow alumina, 
 Golden yellow lake. 
 
 We have said what we had to say about Indian yellow : 
 it is a fairly substantial colour, but it requires to be very 
 carefully ground so as not to be soluble in water, whilst, 
 owing to its alkaline properties, with oil it forms a kind of 
 soap. As we can very well dispense with it, being able 
 to replace it admirably with iron lake and cadmiums, we 
 always think it wiser not to use it. Naples yellow (lead 
 antimony) and antimony yellow (lead antimony and lime 
 silicate) are two substantial colours, but very seldom pure 
 and very often badly made ; and as they are also not very 
 useful, we have excluded them from the palette. 
 
 The following are solid yellow colours : — 
 
 Orange cadmium yellow 
 Cadmium yellow 
 Light cadmium yellow 
 Lemon cadmium yellow 
 
 Before mixing the light shades of cadmium with white 
 
 sulphur of cadmium. 
 
i68 
 
 SCIENCE OF PAINTING. 
 
 lead, it will be well to ascertain that they do not blacken, 
 which would happen if, being badly made, they contained 
 an excess of sulphur. 
 
 The following are good to use as yellow colours : — 
 
 Ochres, earths, marls, iron lake, chromate of strontian and cadmium 
 yellow. 
 
 The firm of Lefranc & Co. now make an orange of a 
 superb tone which may be very useful ; this colour, a com- 
 bination of coal acid and naphthaline, is perfectly solid, but 
 it must be used alone and, above all, not mixed with white. 
 
 Red Colours. 
 The following are bad : — 
 
 Realgar (Sulphur of arsenic) blackens and decomposes metallic 
 colours.' 
 
 Scarlet (Iodide of mercury) fades in the light. 
 All the lakes obtained by cochineal fade in the light : — . 
 Carmine, 
 Lake Carmine ; 
 
 or those procured by dyes from woods of Pernambuco, Brazil, 
 Campeachy, etc., under whatever names they may be pre- 
 sented, as well as those which are due to aniline dyes : — 
 
 Geranium, 
 Nasturtium, etc. 
 
 The madder lakes are comparatively substantial. 
 The following are also good : — 
 Cinnabar ^ 
 
 French Vermilion \- Sulphur of mercury. 
 Chinese Vermilion J 
 
 but care must be taken never to mix them with lead whites, 
 but with zinc white only. 
 
 Red colours good to use are therefore all the earths, ochres, 
 and marls, vermilion and madder lakes. 
 
APPENDIX. 
 
 169 
 
 Blue Colours. 
 
 The following are bad : — 
 
 Mountain blue (carbonate of copper). 
 Berlin blue 
 
 Prussian blue 
 Turnbull blue 
 Antwerp blue 
 Celestial blue 
 Turquoise blue 
 Mineral blue 
 
 different combinations — formula, Ferrid 
 cyanide of iron. 
 
 All these blues of the same origin are decomposed by 
 the majority of metallic oxides and disappear when exposed 
 to light ; some, however, reappear in the shade (excessive 
 modesty). 
 
 Still bad :— ' 
 
 Indigo (a vegetable substance which gets black and green from 
 contact with greasy bodies). 
 
 The following are not bad, but useless, as they can be 
 replaced, and grind badly : — 
 
 Pompeii blue (double silicate of copper and lime). 
 Azure blue 
 Smalt blue 
 
 Saxony blue ^ double silicate of potash and cobalt. 
 ' Starch blue 
 Blue of Zaffre 
 
 Enamel blue (known to the ancients as Alexandria frit). 
 The following are good : — 
 
 Cobalt blue r Combination of aluminum and oxide of cobalt 
 Dresden blue \ obtained at a high temperature. 
 
 r Combination of silicate of aluminum, silicate 
 Ultramarme | ^^^^^ ^^^^ sulphur of sodium. 
 
 Ultramarine badly made may sometimes blacken Chinese white ; a 
 preliminary trial of it should therefore be made unless the artist is very 
 sure of its quality. 
 
 V. P. 
 
 1 1 
 
SCIENCE OF PAINTING. 
 
 Blue colours good to use may therefore be reduced to 
 two, which, besides, are quite sufficient : Cobalt and ultra- 
 marine. 
 
 Green Colours. 
 
 The bad are : — 
 
 Prussian green (double salt of cyanure of iron and cobalt ; turns to 
 
 a grey-red in the light). 
 
 Verdigris (acetate of copper). 
 
 Scheele gfreenl r 
 
 , ^ Y arsenite of copper. 
 Mineral green J 
 
 English green \ j^i^tures of chrome yellow and Prussian blue. 
 Green cinnabar J 
 
 Green lake (mixture of wold lake and Prussian blue). 
 Schweinfurt green (arsenite of copper and acetate of copper). 
 Vienna green (arseniate of copper). 
 
 Bladder green, a lake made with buckthorn dye, and iris green are 
 also bad owing to their want of substance. 
 
 Green earth is also not to be recommended ; it does not 
 adhere. As for malachite green, or mountain green, or 
 Bremen green, or Hungarian green, it is very scarce and 
 difficult to grind. It is well imitated under the name of 
 Brunswick green. The true malachite green is substantial 
 but none of its imitations are. Veronese green is an arseniate 
 of copper of a particular make ; it may be used on condition 
 that it is left by itself and carefully covered with varnish to 
 avoid any contact with other colours. 
 
 The green colours which are really good are : — 
 
 Cobalt green (oxide of cobalt and oxide of zinc). 
 Emerald green (oxide of chrome). 
 
 Violet Colours. 
 
 All the violet colours obtained by aniline are bad, such 
 
 • Magenta violet, 
 
 Solferino violet, etc. 
 
 Vegetable violet obtained by dye of Campeachy wood 
 
APPENDIX. 
 
 171 
 
 mixed with salts of lead, and violets derived from cochineal 
 are also bad. 
 
 An ultramarine violet is now made which is as solid as 
 ultramarine, but does not adhere well and to which for that 
 reason we prefer cobalt violet (phosphate of cobalt and 
 silicate of cobalt), and especially mineral violet (phosphate 
 of manganese). This last colour, very solid and adhering 
 well, is especially useful ; it may be had in two different 
 shades, both equally good. . 
 
 Therefore, for violet colours, preference should be given 
 to cobalt violet, mineral violet and Mars violet. 
 
 Brown Colours. 
 
 Amongst brown colours the worst of all is bitumen. It 
 fades in light, melts in heat, causes the painting to crack, 
 runs and compromises the existence of all the pictures in 
 which it is used. 
 
 Next comes Cologne earth, which is only wood in a state 
 of decomposition. 
 
 Cassel earth, chicory brown, and natural or burnt umber 
 (bioxide of manganese), are also bad. Umber is penetrating 
 and blackens all the tones with which it is mixed or over 
 which it is laid. 
 
 Prussian brown (calcined Prussian blue) is completely 
 useless, as it may be advantageously replaced by Mars 
 brown, which is excellent as a covering brown. 
 
 Imitation bitumen, of which we have spoken and which we 
 have the honour to present to artists under the name of 
 Vibertbrown^ will give an excellent transparent brown, having 
 all the good qualities of bitumen without its defects. This 
 colour, perfectly substantial, is a mixture of lithanthrax and 
 oxide of iron on an aluminum basis. Brown colours, there- 
 fore, should be limited to Mars brown, and Vibert brown. 
 
172 
 
 SCIENCE OF PAINTING. 
 
 As for blacks, as we have said, they are all good, except 
 those which contain tar. 
 
 Verification of Colours. 
 
 Unless one be a chemist and have at his disposal the 
 various requisites of a laboratory, it is impossible to make a 
 complete analysis of colours. But with the- simple materials 
 that are always at hand, either in kitchen or studio, the 
 colours when pure may be verified and their principal 
 adulterations may be discovered. 
 
 The following is a list of the necessary materials and 
 objects. They are neither numerous, costly, nor difficult 
 to piocure. 
 
 Strong vinegar. 
 
 Kitchen salt (in wet solution). 
 
 Copper water, (This can easily be prepared by purchasing 
 from the grocer some oxalic acid in powder and making a 
 solution by dissolving it in water.) 
 
 Aquafortis, Composed of one part nitric acid to five parts 
 water. 
 
 Soda crystals in a wet solution. 
 A spirit lamp. 
 
 Some little glass vases and funnels, some paper filters, and 
 an iron spoon not tinned. 
 
 The colour to be analysed should first be reduced to 
 powder ; if an oil colour, it should first be dissolved in a 
 glass of benzine and allowed to settle, the liquid then gently 
 poured off, and the process repeated with new benzine. 
 This should be done several times, until all the oil be 
 eliminated and the powder which remains quite free from 
 grease : it should then be put to dry on blotting-paper.* 
 
 * We insist on the colour being completely free from grease, because 
 in the reaction obtained by fire any trace of oil would give rise to 
 smoke, which might cause an error. 
 
APPENDIX. 
 
 If the colour to be analysed be a water colour, the same 
 process should be gone through with water to remove the 
 gums, honey, glycerine, etc. In any case the liquid which 
 comes off must not remain coloured, as that would indicate 
 the presence of cerjain aniline colours.* 
 
 It is therefore quite understood that, in all the following 
 operations we treat the colours as in a state of dry powder 
 and perfectly free from grease. 
 
 White Lead or Chinese White (Carbonate of 
 Lead.) 
 
 Put into a glass a small quantity of the white, and pour 
 over it about six times that amount of aquafortis. If the 
 white be pure, it should dissolve entirely with effervescence : 
 an insoluble residue would indicate the presence of sulphate 
 of baryta. To this solution add a large quantity of kitchen 
 salt in solution, it will form a precipitate white which is lead. 
 Filter, and to the liquid which comes off clear, add a solution 
 of soda crystals until the bubbling stops. If a precipitate 
 white be formed, it shows that the white contained oxide of 
 zinc or carbonate of lime. 
 
 Zinc White (Oxide of Zinc). 
 
 Oxide of zinc should dissolve in vinegar : if a very heavy 
 white residue remain which falls at once to the bottom of 
 the glass, it shows that the white contains sulphate of baryta. 
 When the oxide of zinc is dissolved in the vinegar, add the 
 solution of salt (twice the volume of zinc) the liquid should 
 remain clear : if it form a precipitate, there is carbonate of 
 
 * Sometimes the colours are gromid so fine, such as ultramarine for 
 example, that the water remains coloured for a long time, and patience 
 is required until the deposit is properly formed. 
 
174 
 
 SCIENCE OF PAINTING. 
 
 lead. Put back some of the clear solution of oxide of zinc 
 into the vinegar, dilute it with fifteen times its volume of 
 water and add some copper water, about the same volume 
 as there was of the zinc : the liquid should still remain 
 clear; if there be a precipitate, it shows that the white 
 contained carbonate of lime. 
 
 Strontian Yellow (Chromate of Strontian). 
 
 Put the chromate of strontian into a glass with ten times 
 its volume of aquafortis : it ought to dissolve entirely. The 
 solution becomes orange-coloured if a solution of salt be 
 added to it ; should it precipitate a yellow residue, that 
 would be chrome yellow. 
 
 Cadmium Yellow (Sulphur of Cadmium). 
 
 Put the sulphur of cadmium into a glass with four times 
 its volume of nitric acid : it ought to dissolve ; the solution 
 is milky, and a certain quantity of sulphur floats on the top. 
 If the solution be more or less grey, and after the deposit 
 the purified liquid be coloured a greenish blue, it shows that 
 the sulphur of cadmium contained chromate of lead ; if 
 the residue be of an intense red, it shows that it contained 
 vermilion. 
 
 Vermilion (Sulphur of Mercury). 
 
 Heated in an iron spoon, the sulphur of mercury should 
 be completely volatilised. Should a residue remain it shows 
 that it is not pure. Put to boil in water and filter when 
 boiling ; the clear liquid should not, on cooling, precipitate 
 any red powder : that powder would be iodide of mercury. 
 
 Madder Lakes. 
 
 Madder lakes boiled with a weak solution of crystallised 
 soda give a red colouring ; cochineal lake would give a violet 
 
APPENDIX. 
 
 colouring. The solution of crystallised soda should be 
 made with thirty times as much water as lake and an equal 
 quantity of crystal and lake. 
 
 Lakes which colour alcohol contain aniline. Whether or 
 not they will bear light should be ascertained. 
 
 Ultramarine. 
 
 Ultramarine should dissolve cold in ten times its volume 
 of copper water cleared with sulphuretted hydrogen (smelling 
 of rotten eggs). The solution is of a milky grey, but with- 
 out the slightest tint of blue ; should there be any blue tint 
 it contains Prussian blue. 
 
 Heated in the iron spoon, the ultramarine whilst burning 
 should not emit any vapour of a beautiful purple colour, 
 that would show that it contained indigo : these vapours 
 have a strong and characteristic smell. Ultramarine is 
 rarely adulterated with indigo. 
 
 Cobalt (Oxide of Cobalt). 
 
 Cobalt put in a glass with ten times its volume of copper 
 water should preserve the same tone that it would have in 
 pure water; boiled, and left to settle for an instant, the 
 floating liquid should be uncoloured. If it were blue, that 
 would indicate the presence of Prussian blue. 
 
 If it emitted an odour of rotten eggs, that would indicate 
 ultramarine. 
 
 Cobalt should get red in the spoon without vapour, like 
 ultramarine. 
 
 Emerald Green (Chrome Oxide). 
 
 Boiled with pure nitric acid emerald green should not 
 change ; if water be added and the liquid strained it should 
 
176 
 
 SCIENCE OF PAINTING. 
 
 be uncoloured. If it be yellow, it is from the presence 
 of chromate of strontian, lead or zinc. Boil the emerald 
 green with copper water ; after it has settled for a short time, 
 if the liquid on the top be coloured blue, that shows it 
 contains Prussian blue; heated in the spoon, it should 
 emit no vapour; should it emit a white smoke with an 
 odour of garlic when put on a red-hot spoon, that would 
 show that it contained arseniate of copper. 
 
 Cobalt Green (Oxide of Cobalt). 
 This is like emerald green, only it dissolves entirely in 
 aquafortis mixed with water and the solution when pure is 
 of a light pink. The solution will be orange if it contain 
 chromâtes or does not entirely dissolve; Prussian blue 
 would remain at the bottom if there were any. 
 
 Blacks. 
 
 All blacks which are the product of carbonised animal or 
 vegetable substances, when kept red-hot for an hour in the 
 spoon, will leave an amount of ash equal to the amount of 
 black paint. 
 
 Lamp blacks leave little or no ash. 
 
 Earths (Ochres). 
 These colours are not generally adulterated ; impurities 
 they contain vary in degree ; the only deception practised 
 is to enrich their tones with aniline dyes : . by putting them 
 into alcohol we can at once see if the alcohol colours. 
 
 Cleaning. 
 
 To clean a picture, the first thing to do is to wash it with 
 lukewarm water and a soft sponge to take off the dust and 
 coal slag from its surface. 
 
APPENDIX, 
 
 177 
 
 Afterwards, if it is varnished, the varnish may be taken 
 off with the finger, washing it from time to time and leaving 
 it to dry before resuming the de-varnishing until there is no 
 longer any- dust to be felt under the finger. Then put on 
 a little essence of petroleum, applied with a linen rag the 
 surface of which must not be fluffy : the picture is then 
 ready to re-varnish. But if under this varnish first taken off 
 there is another which does not yield to the finger, or if this 
 first varnish itself does not yield to the rubbing, there would 
 be every reason to suppose that we had come to an oil 
 varnish or an ordinary varnish over which some kind of 
 greasy matter had been passed. The picture should then be 
 covered with essential oil of petroleum, and at the end of 
 twenty-four hours a layer of finely powdered chalk passed 
 through a sieve should be sprinkled over the oil on the 
 picture laid horizontally. The object of this chalk is to 
 suck up the petroleum charged with fatty matter. Twenty- 
 four hours after this operation, the picture should be washed 
 with lukewarm water to take off the chalk, and afterwards 
 with water slightly soaped to finish taking off the grease. 
 
 If these means are not sufficient, it may be washed with 
 acetum and water, but it should be cautiously done, and 
 first tried on a small corner to see the effect. 
 
 After each washing, the finger should always try to 
 de-varnish ; if, however, the varnish persists in not coming 
 off, we may be sure that there is an oil varnish. In this case 
 it would be dangerous to try and take it completely off — it 
 would be necessary to try alcohol or benzine, which would 
 attack the painting. We inust be satisfied with taking off 
 merely the surface with acetum and water and a silk rag, and 
 doing only a very small part at once ; or it may be done with 
 egg oil mixed with the dissolving liquid of water-colour 
 fixative. But if, under this varnish that we do not wish to 
 
178 
 
 SCIENCE OF PAINTING. 
 
 take entirely off, the picture seems very dirty, the reason is 
 that this varnish had been appHed over coatings of old grease, 
 which happens very often. In such a case what is to be 
 done ? It must be taken off ; but there is a risk of destroying 
 the picture, and the best thing to do is to entrust it to the 
 care of a restorer who will only do what you would do 
 yourself, but who will do it more cleverly. In any case it 
 is always a hazardous operation. 
 
 To take off the colour from dried oil on a picture, either 
 in part or entirely, a rag soaked in benzine, or even a scraper, 
 should be used. On an unpolished panel this last means 
 is good, but on sizings or paper, benzine is preferable ; 
 with it there is no risk of making either holes or scratches. 
 
 Varnishing, 
 
 Before varnishing a picture it should always be washed 
 in lukewarm water, as said above, and left to dry thoroughly. 
 
 Next the picture should be placed horizontally, and a 
 first layer of essence of petroleum applied (it is indis- 
 pensable that the picture to be varnished should be 
 sufficiently dry) and immediately afterwards the layer of 
 picture varnish, applied by passing the swallow-tail brush 
 evenly from right to left and left to right, until the entire 
 picture is covered with varnish, after which repeat the 
 operation with- the swallow-tail from top to bottom and 
 bottom to top, so that the passage of the brush may have 
 been everywhere crossed. At the end of an hour the 
 picture may be placed upright, and two hours afterwards 
 the varnish will be sufficiently dry to prevent any dust 
 adhering to it. This would not be the case with essence 
 varnishes, as they remain sticky sometimes for days. 
 
 For varnishing dry weather should be preferred, and the 
 
APPENDIX. 
 
 179 
 
 work always done in a room warmed at least to twenty 
 degrees. As a matter of course all precaution should be 
 used to ward off dust during the two hours required for 
 drying. 
 
 At all times, when from any cause a picture has been 
 washed with water, it should never be left wet, but gently 
 and thoroughly dried with a silk rag, to prevent the dampness 
 from penetrating. 
 
 Water-colour Fixative and the Dissolving 
 Liquid. 
 
 The object of the dissolving liquid sold with the fixative 
 is to liquefy the latter when too thick. It is also useful to 
 take oif the fixative when desired, and to wash the paint 
 brush after the fixing. 
 
 Method of Fixing. 
 
 When required to fix a water colour it should be laid flat 
 and one or two coatings of dissolving liquid applied with 
 a flat or soft brush. The paper being thus saturated, the 
 fixative will be better distributed over all the surface, and 
 will penetrate better into the interior. Before the dissolving 
 liquid evaporates a layer of fixative should be immediately 
 applied with the same brush, going over it first one way 
 and afterwards crossing it, as in varnishing. 
 
 It should be left to dry in the air, and at the end of some 
 minutes or a quarter of an hour, according to the temperature, 
 the liquid having evaporated, the water-colour will be 
 sufficiently fixed to allow new tints to be painted without 
 melting those below; if a more complete fixing be desired 
 allowing the water-colour to be washed freely in water, a 
 second layer of fixative should be applied, but this time 
 without the preliminary layer of dissolving liquid. Besides, 
 
t8o science of painting. 
 
 the need of this second layer will be greater or less 
 according to whether the fixative has been applied thick or 
 thin, or whether the paper has been much or little sized. 
 
 Only one bit at a time should be fixed : e.g,^ apart already 
 shaded that is to be glazed at leisure, without fear of 
 melting what is below it, or, on the contrary, a flat tint over 
 which the shading can be done without destroying the local 
 tone beneath. 
 
 Method of Taking Off the Fixing. 
 
 To withdraw the fixing, a layer of dissolving liquid should 
 be applied, and it should be afterwards sponged with linen 
 or blotting-paper but not rubbed, so as not to wear off the 
 colour nor grains of paper. This operation should be 
 renewed until all the fixative comes off, which may be 
 ascertained by trying on a corner if the colour melts readily 
 in water. This de-fixing is useful in certain cases. First, 
 although it is quite easy to paint over the fixative, some- 
 times it is desirable to retouch by washing in order to 
 accentuate the lights, and it is necessary that the colour 
 should melt. Besides, it will often happen that the fixative 
 has only been used in a provisionary state as a means 
 of work. 
 
 Thus, being satisfied with a head, for instance, the back- 
 ground behind it may be objected to ; therefore the head 
 alone should be fixed with a small paint brush and the back- 
 ground washed off all round ; but to repaint another 
 background and blend the contours, the head must be 
 defixed. It would be the same if on the contrary the head 
 were to be washed out and the background preserved. The 
 latter should then be fixed and defixed afterwards when the 
 head has been washed off. 
 
APPENDIX. 
 
 i8i 
 
 When the fixative is used in this way there is no harm 
 in putting a good deal of it, as it will afterwards be taken 
 off. 
 
 Sometimes, also, it is very tiresome all the time that 
 a water-colour painting is being done, to reserve little white 
 places like the brilliant buttons of a doublet, the ornaments 
 of a sword handle, the pattern of a stuff, etc., and they are 
 consequently done in Chinese white afterwards ; but this 
 method is despised by genuine water-colour painters. If, 
 therefore, they want to do away with these disgracing touches, 
 they have only to border them off with fixative, and after- 
 wards, washing it off they will find the paper white and intact 
 and the touches will remain quite clear. 
 
 In this way falling snow may be done, and the method is 
 useful on many other occasions. 
 
 Even when a water-colour is not definitely fixed after 
 being finished, the fixative can still render good service 
 during its execution. 
 
 Varnished Water-colours. 
 
 To varnish a water colour, it should first be fixed ; and 
 to know if one layer is enough a little varnish should be 
 applied in a corner over the first layer. If the varnish 
 does not penetrate into the paper, that is enough ; if the 
 varnish penetrates, a second layer of fixative should be 
 applied, and so on until the varnish no longer penetrates ; 
 then the picture is to be varnished as if done in oils. 
 
 Repair of Little Daily Accidents. 
 
 Indented Cmtvas. — Wet the canvas behind the place of 
 the indentation with a sponge soaked in lukewarm water and 
 let it dry in a medium temperature. 
 
l82 
 
 SCIENCE OF PAINTING. 
 
 Blisters. — Soften the blister on the top with 'an equal 
 mixture of essential oil of petroleum and essence of petro- 
 leum ; next, pierce it in the middle with the point of a 
 needle, and drop into this hole with a tiny syringe (like 
 those used for dropping medicines in single drops) a little 
 picture varnish. Press with a rag, to flatten the blister and 
 at the same time to take off any superfluous varnish ; and 
 afterwards have the canvas pressed under a piece of zinc 
 or glass, so that the borders may be smooth and show no 
 mark. A small piece of paper sized with paste should be 
 put between the glass and the blister, because any varnish 
 which might come out would stick to the glass, whence 
 it would be awkward to get it off, whereas the paper can be 
 taken ofl" with a little lukewarm water. 
 
 Splits^ Flaws ^ Cracks. — When the colour is split — which 
 often happens when the repainting is done too soon, 
 especially in a dark tone over a light one — it should be 
 rubbed with a Httle re-touching varnish and the splits filled 
 up with a knife and a special mastic that we have invented 
 for the purpose. (See " Mastic for Stopping Up.") ^ - 
 
 Burst Canvas. — When canvas bursts it is customary to 
 glue behind it a little piece of fine linen saturated with hot 
 melted wax. This is a very bad method. In the first place 
 the wax alone is not impervious to damp, and these mendings 
 often rot. Next, the wax passes to the other side of the 
 canvas through the crack, and as the canvas must always be 
 re-painted to hide the repair, the re-paintings are not sub- 
 stantial, oil paintings (as said above) not adhering to w^ax. 
 It would be better, before glueing the piece, to apply with a 
 knife a little caséine and zinc-white sizing (see Sizing of 
 Caséine Paste and Zinc White "). This sizing would prevent 
 the wax from passing through, and instead of pure melted 
 wax, a mixture of half wax and half dammarine resin melted 
 
APPENDIX. 183 
 
 together in a water bath should be used. This would 
 diminish the chance of dampness. 
 
 But the following is another means which we consider 
 much superior. 
 
 The picture should be placed horizontally, face down- 
 wards, on a slab of marble or metal separated from the 
 painting by a sheet of paper sized with essential oil of 
 petroleum. The edges of the burst canvas should be 
 brought well together by the help even of a flat iron if 
 necessary, being careful to heat this iron only over boiling 
 water. 
 
 All the destroyed part should now be coated with caséine 
 paste according to the formula. When the dampness has 
 evaporated a little and the paste feels sticky under the 
 finger, a sheet of paper, thin, but sized, {ix, not blotting- 
 paper, so that the paste may not come through. It should . 
 then be well flattened with the moderately hot iron. Not 
 the slightest flaw must . remain. All this being dry, a new 
 piece of sized paper should be applied, and a piece of fine 
 canvas rather smaller than the paper, because the canvas 
 would not adhere well to the other canvas. It should be 
 flattened with a warm iron as before ; and everything being 
 quite dry, one or more successive layers of re-touching 
 varnish should be applied, letting each layer dry before 
 putting on the next. The varnish should pass the borders 
 of the paper and touch the old canvas, so that the edges of 
 the pieces of canvas and paper are well secured and have no 
 tendency to rise if scratched by anything. 
 
 When the last layer of varnish is dry, the picture should 
 be turned and placed upright. If, by chance, the paper ' 
 sized with essential oil of petroleum which was put between 
 the table and the picture should have stuck, it can be taken 
 off with a little lukewarni water. All that now remains to 
 
SCIENCE OF PAINTING. 
 
 be done is to fill up the traces of the crack with filling-up 
 mastic (see " Mastic for Filling Up or with a little zinc 
 white and caséine paste. A mastic can even be made with 
 caséine paste and a coloured powder of the tone desired. 
 
 If caséine mastic be used, it should be painted over with 
 painting varnish and left to dry before being re-painted. 
 
 Panels of Unpolished Wood. 
 
 Choose the grey or Dutch poplar, because its pores are 
 open and its grain regular. The oak and mahogany have 
 veins some of which are hard and some soft, consequently 
 the oil from the pigments gets more deeply absorbed in the 
 soft veins than in the hard ones, and there results a want of 
 smoothness in the painting which increases layer by layer 
 to the end. 
 
 The panel should be rubbed with sand-paper, and for 
 the last rubbing the paper should be soaked in essential oil 
 of petroleum. By this means rays are avoided, the petro- 
 leum evaporates and the panel remains smooth without 
 having lost either its grain or its porosity. 
 
 The panel should be floored and kept in a dry place 
 until it has shrunk — which will be about one inch in a 
 hundred. Only then should the back of the panel be sized 
 with a flooring composed of equal parts of thick linseed oil 
 and re-touching varnish without siccative, and, when this 
 layer is dry, with another layer of picture varnish. The 
 painting should be done over several layers of white lead, 
 and between each layer the panel should be rubbed with 
 glass paper, and painting varnish with a little siccative 
 should be mixed with the colours. The object of these 
 precautions is to preserve the panel from the attacks of 
 dampness and worms. 
 
APPENDIX. 
 
 185 
 
 Paste of Caséine or Cheese. 
 
 Take of caséine of commerce, about 20 grammes* 
 according to quantity of paste required. Put it to soak in 
 TOO grammes of cold water for some time. Occasionally it 
 requires to soak for some hours. The length of time 
 depends on the condition of the caséine, which may be 
 more or less hard through drying on the stove. 
 
 Whilst stirring, add 4 grammes of ammonia by degrees, 
 stirring with a spatula, or a stick of wood, glass, or horn 
 but not metal. When the ammonia is incorporated with 
 the mixture, leave it alone for some minutes, then stir it 
 anew every ten minutes for a few seconds each time. The 
 mixture thickens gradually, and the paste is ready when it 
 streams from the end of the spoon in a thick syrup without 
 lumps. Still stirring, add 20 grammes of glycerine. Mix 
 well in stirring. The paste is then fit for use. Next day it 
 is still good, but after that it loses its strength and becomes 
 spoilt. 
 
 If you have no caséine, and wish to use cheese, the 
 method is as follows : — White cheese (called new cheese), 
 ICQ grammes, to which, whilst Stirring, add 4 grammes of 
 ammonia, and stir until the cheese becomes paste. Next 
 add to it 10 grammes of glycerine. It is always preferable 
 to use caséine, because, the cheese being more or less 
 drained, it is impossible to judge exactly the quantity of 
 caséine contained in it. The following is the method of 
 obtaining the caséine from the cheese. 
 
 Put about ICQ grammes, according to quantity required, 
 of white cheese into muslin, and rub it whilst thus confined 
 in warm water, to extract all the, watery particles, spread it 
 
 * A gramme is about ^\th of an ounce avoirdupois. It is 15 '432 
 grains. 
 
i86 
 
 SCIENCE OF PAINTING. 
 
 out on blotting paper and let it dry naturally on the stove, 
 but at a temperature below 80 degrees. The caséine thus 
 obtained should be in the proportion of about 12 grammes of 
 caséine for 100 grammes of cheese used. 
 
 SiziNGs OF Caséine and Zinc-white Paste. 
 
 Grind the zinc white in water in the proportion of 100 
 grammes of water to 100 grammes of zinc, and mix this 
 zinc white with 200 grammes of caséine paste already pre- 
 pared. This paste should be kept moist with a wet linen cloth 
 over the jar, so that it does not dry. The wooden panel to 
 be sized, having been well smoothed with the glass paper, 
 should be washed with a very clean rag and benzine, to take 
 off any trace of greasy or resinous matters which would 
 prevent the paste from adhering. 
 
 When the benzine has quite evaporated, give, with a 
 swallow-tail brush, a first layer of paste and white. 
 
 Let this dry at a mild temperature. Under the influence 
 of the water contained by the paste, the pores of the wood 
 will open, and, rising up, will become so many little links to 
 facilitate the adhesion of the following layers. 
 
 Be careful, when this first layer is dry, not to polish or rub 
 the panel, so as to leave to the wood any roughness which it 
 may have acquired. 
 
 Give a second layer. Let it dry naturally, or if wanted 
 quickly, put it in the sun covered by gauze in a room not 
 warmer than 25 degrees, so as not to make the panel 
 crack. 
 
 It may then be polished with glass paper or powdered 
 pumice stone by means of a cork or dry rag.* 
 
 * According to the manner in which they have been prepared, the 
 zinc whites are more or less dense, and consequently more or less 
 heavy. The proportions given are for the heaviest, which weigh about 
 
APPENDIX. 
 
 187 
 
 If a brilliant white be desired, the third coat may be 
 doubled, or even trebled ; but between each coat it must be 
 left to dry, and polished. 
 
 When the sizing is satisfactorily white, a last coat should 
 be given. 
 
 This last C(;at should be put on very evenly, and so as 
 to require no rubbing, as that would give a disagreeable 
 polish, on which the painting will not look so well. 
 
 All the coats should be given thickly, i.e., the brush 
 well charged with matter. As, according to the temperature, 
 the w'ater of the paste evaporates more or less quickly, if 
 the paste get too thick, a little water should be added. 
 The liquid sizing should always come uninterruptedly to 
 the end of the brush when it is lifted out of the jar. 
 
 The canvas should be stretched on a frame, and freed 
 from grease with benzine. The canvas should be rubbed 
 with glass paper after the first coating, because the grain 
 of the canvas will always give sufficient adhesiveness, and it 
 is necessary to get quickly rid of the straws, threads, and 
 knots which would interfere with the following coats : 
 
 After this first coating it is a good plan to lay on a very 
 light coat of re-touching varnish, which must be allowed to 
 dry thoroughly. 
 
 800 or 9CX) grammes to the litre (a litre =1*76 pint). Those are the 
 best to use, but there are some like snow-white which only weigh 450 
 to 500 grammes per litre. Should the latter be used, it is evident that 
 the weight given in the formula should change in proportion. Besides, 
 these formulas can only be approximate. The nature of the wood, 
 cardboard or canvas, desired to be covered, as well as the degree of 
 absorption that the artist desires to obtain, will compel each one to 
 decide for himself the proportion most suitable. The only caution to 
 be given is that each time the coating cracks and lacks suppleness, 
 it contains too much white. Other whites than zinc whites may be 
 used, but in that case the proportions of paste would not be the same. 
 
i88 
 
 SCIENCE OF PAINTING. 
 
 Second coat, No. 2. 
 
 Third coat, No. 3, to be put as desired according to taste, 
 to the thickness of the canvas, and to whether it be desired 
 to preserve much or little grain. 
 
 Sarcocol. 
 
 Sarcocol, about 50 grammes, according to quantity 
 desired. Treated by alcohol at 96 degrees, boiling until all 
 boiled away. The distilled alcohol leaves a red residue, 
 soft, looking like turpentine. This residue still contains 
 much alcohol, and is dried in a stove at from 70 to 80 
 degrees with much difficulty ; this is the sarcocolline, of 
 which there will be 35 grammes. 
 
 Sarcocolline is soluble in water, except a very small por- 
 tion which dissolves in sulphuric ether. This watery solution, 
 treated by a solution of acetate of lead, forms a scanty 
 precipitate, which is separated by filtration, and the liquid is 
 treated with sulphur. To eliminate the lead, this filtered 
 liquid is evaporated until dry, and thus a very slightly 
 coloured production is obtained, preserving all the properties 
 of sarcocolline. 
 
 It is a glucoxide in which a resin is combined with the 
 elem'ents of a sugar. 
 
 The alum and acetate of aluminum considerably diminish 
 its solubility without rendering it completely insoluble. 
 Practically, one may be satisfied to put the sarcocol in solu- 
 tion in boiling alcohol. The liquor allowed to cool is filtered, 
 and may be whitened by leaving it for some time in contact 
 with bone black exposed to great light, filtered afresh, and 
 the solution preserved in a well-corked bottle exposed to 
 light. 
 
APPENDIX. 
 
 189 
 
 Gelatine. 
 
 To render gelatine liquid {ix. to prevent it from becoming 
 a jelly when cooling), dissolve the white gelatine in its weight 
 of the following liquid : — 
 
 Half acetic acid and half water . . . . f 
 Alcohol at 95 degrees \ 
 
 To render the gelatine supple, a little glycerine is added, 
 in different quantity according to the degree of suppleness 
 desired. 
 
 To add strength to the gelatine, some gum lac may be 
 added to it, either in alcoholic solution, or in solution with 
 water and borax. {See " Gum lac") 
 
 Gelatine mixed with linseed oil and our picture varnish 
 makes an emulsion which, ground with zinc white or white 
 lead, makes sizings drying slowly but of wonderful solidity ; 
 when they are to be painted over we do not advise their 
 use, preferring the sizing of caséine paste to them, but in 
 certain cases they may be useful. To render the gelatine 
 msoluble, it should be added to a mixture composed of one 
 part chromic acid to five parts of gelatine or of bichromate 
 of potash and left to dry in plenty of light ; but in this 
 condition the gelatine is of a yellow colour. 
 
 If it be desired to preserve the gelatine uncoloured and 
 liquid whilst rendering it insoluble, we prefer the following 
 means : — 
 
 Gelatine . . . . \ \ 
 
 Acetate of aluminum 3 ) 
 Add a little alcohol when all is melted. On all occasions 
 when gelatine contains acetic acid, care must be given not 
 to mix it with colours which are carbonates. 
 
 Add a little alum. 
 
 melted together in a water bath. 
 
190 
 
 SCIENCE OF PAINTING. 
 
 Starch or Farina Paste. 
 
 The starch or farina should be put into cold water, and 
 thoroughly mixed ; when the starch or farina no longer 
 forms lumps, put it on the fire, keep stirring, and take it off 
 after it reaches boiling point. Starch paste with gelatine and 
 turpentine (resin) makes a very good paste, that dries quickly. 
 
 Albumen dissolves in alcohol, in water and in glycerine ; 
 the paste obtained from it may be mixed with gum lac 
 like gelatine and like it, becomes insoluble by acetate of 
 aluminum. 
 
 It also becomes insoluble if heated above 100 degrees, 
 when still damp. 
 
 Dextrine. 
 
 Dextrine may be extracted from itself by the following 
 means : — 
 
 Starch needles 100 gr. 
 
 Water 40 
 
 Azotic acid . . . . . . 4 
 
 Make a paste, let it dry in the open air, heat it afterwards 
 in a stove for an hour at a temperature of 120 degrees. 
 
 Gum Lac 
 
 Brown gum lac dissolves in alcohol, as does also white 
 gum lac, cold or in the water bath. 
 
 Brown gum lac dissolves in ammonia cold after soaking 
 
 Boil in water 
 
 Starch 
 
 Gelatine 
 
 Turpentine 
 
 100 gr. 
 SO „ 
 
 Albumen. 
 
APPENDIX. 
 
 191 
 
 for five weeks ; white gum lac takes two or three months. 
 Gum lac dissolves also in water by the help of borax. 
 
 Gum lac . , . . . . . 60 gr. 
 Borax . . . . . . . . lO 
 
 Distilled water . . .... 500 
 
 With very long boiling all the gum lac will dissolve. 
 
 Afterwards filtered and evaporated to a syrupy consistency, 
 this liquid is a good deal coloured. 
 
 It may be whitened by means of oxygenised water in large 
 quantity. 
 
 Water Varnish. 
 
 Varnishes may be made with water, and have the 
 advantage of being easily taken off with a sponge and 
 some warm water. 
 
 Gum Varnish. 
 
 Water saturated with borax. Senegal gum in sufficient 
 quantity according to the thickness desired for the varnish, 
 sugar to one-eighth the amount of the gum, all melted cold 
 and passed through muslin. 
 
 Egg Varnish. 
 
 Beat the whites of eggs to a froth with half a piece of sugar 
 reduced to powder for the white of each egg. Leave it to 
 settle for an hour, then draw off the transparent liquid fallen 
 to the bottom of the jar. This varnish does not keep, and 
 must be used fresh. 
 
 Sarcocolline Varnish. 
 
 Make a solution of sarcocolline in alcohol at 60 degrees, 
 filtered and purified, with the addition of some drops of 
 glycerine. 
 
192 
 
 SCIENCE OF PAINTING. 
 
 This varnish adheres very well over oil paintings, even 
 when rather greasy. 
 
 We may also use as varnish a solution of gum lac in water 
 and borax ; but this once dry would not come off with water, 
 — it would require alcohol. 
 
 IndeHble ink which may be used with steel pens : 
 Aniline Black . . . . . . 4 gr. 
 
 Add concentrated hydrochloric acid 60 drops. 
 
 Drawing ink which may be washed. — Good Chinese ink 
 has this property ; but where it cannot be procured, it may 
 be made with a solution of gum lac in borax added to 
 aniline black dissolved in alcohol. There is no deposit 
 from this ink, but before shutting it up in a bottle, the 
 alcohol should be allowed to evaporate for a day or two. 
 
 To RENDER Wax Miscible with Water and 
 
 Take 10 g^-ammes of wax broken into little pieces, and 
 put it on the fire in 100 grammes of water, or in 100 
 grammes of glycerine. 
 
 When the wax is melted, drop in gradually whilst stirring 
 rapidly some drops of ammonia : the wax will soon become 
 a liquid emulsion, and may be kept in that condition for a 
 long time ; the bottle into which it is poured should be left 
 uncorked for several days, so that the excess of ammonia 
 may evaporate. 
 
 Inks. 
 
 Alcohol 
 Gum Arabic 
 Water 
 
 15 
 
 6 „ 
 90 » 
 
 Glycerine. 
 
APPENDIX. 
 
 To Prevent Vibrations of Canvas or Paper 
 Stretched for Crayons. 
 
 A wadded mattress should be prepared and placed behind 
 the canvas or paper, and between it and a sheet of cardboard 
 fixed on the frame by screws. Mineral cotton in spun glass 
 or a stuffing of asbestos may be used, as they are both in- 
 destructible by fire or dampness. 
 
 The cardboard or shelf which is to give resistance behind 
 should be carefully rendered waterproof. 
 
 Instead of the cotton, cork or any similar body might 
 be used, provided the body were impervious and very light, 
 and had no tendency to sink to the bottom of the frame, 
 and that its pressure remained equally distributed over all 
 the surface of the paper or the canvas. 
 
 Instantaneous Sizing. 
 
 If a sizing be desired which shall dry immediately, we 
 advise it to be made as follows : — 
 
 Water-colour fixative . . . . 20 gr. 
 
 Dammarine resin 3 
 
 Oxide of zinc . . . . . . 3 ,, 
 
 The whole well mixed together and applied with a knife. 
 
 If it be desired to apply this sizing with a brush, it may 
 be liquefied with water-colour fixative dissolving liquid. 
 
 Either water-colour or oil painting may be done over this 
 sizing. 
 
 Mastic for Filling up. 
 
 Take powdered colour, according to the tone desired for 
 the mastic, and mix it with dammarine resin, equally pul- 
 verised in the following proportions : 
 
194 
 
 SCIENCE OF PAINTING. 
 
 For almost all tones, two grammes of colour and one 
 of resin. For vermilion three grammes of colour and 
 one of resin. For cobalt a little more resin is required. 
 For lakes, there should be one-half more of resin than for 
 the other colours, but they should not be used in filling up 
 mastics for which opaque and adhesive colours are greatly 
 preferable. 
 
 The colour being mixed with its proportion of resin, the 
 whole should be ground with water-colour fixative to a 
 liquid paste, and some minutes should elapse before using 
 it, to let it thicken to the consistency of mastic. Paste of 
 different colours may be prepared before being wanted, and 
 kept in bottles. 
 
 Should they dry through being badly corked, they can 
 easily be moistened with the water-colour fixative dissolving 
 liquid. 
 
 An equally good mastic can also be made with caséine 
 paste and powdered colours. For this purpose the paste 
 should be prepared a Httle thicker than for sizings, or more 
 patience will be required until it thickens by evaporation. 
 Over this paste mastic retouching or painting varnish should 
 always be applied before painting. 
 
 Manner of Rendering Paper Transparent. 
 
 If in want of tracing paper, it can easily be prepared, by 
 applying a good layer of essential oil of petroleum to any 
 kind of white paper, and when that oil shall have thoroughly 
 penetrated through the paper, wipe off any excess, from 
 both sides, with a clean rag. 
 
 Ink and pencil take very well on paper thus rendered 
 transparent. The transparency lasts from ten to twelve 
 hours according to the temperature, and the petroleum is 
 completely evaporated at the end of two days. 
 
 I 
 
APPENDIX. 
 
 This method is very useful in certain cases, as it saves 
 the trouble of tracing and retracing several times. For 
 instance, if you wish to make a pen-and-ink drawing of 
 a water colour or other picture with the ordinary tracing 
 paper, you have first to trace, then take off the tracing on 
 the white paper, and finally outline with ink ; whereas by 
 preparing with petroleum the paper itself on which you are 
 going to draw, you may directly at one tracing make the 
 first outline with the pen. Next day your paper is no longer 
 transparent, and you may continue as if the paper had never 
 had any petroleum. 
 
 Paper for taking Tracings from Oil Paintings. 
 
 To take a tracing from an oil painting, instead of using 
 paper rubbed with black lead or red chalk, or white lead or 
 lamp black, the tracings on which have the disadvantage of 
 disappearing as the painting proceeds, and sometimes of 
 soiling the tones, it is preferable to prepare a tracing paper 
 of the shade desired by rubbing oil colour mixed with 
 siccative on an ordinary piece of tracing paper. In order 
 that very little colour may remain, any excess should be 
 well wiped off with a rag pad. 
 
 The layer of colour being very thin and very siccative, 
 this kind of paper is only serviceable for a few hours. 
 
 Its use has great advantages : the tracings obtained from 
 it when dry look as if painted with a very fine paint brush ; 
 they do not run when painted with new colour — they may 
 even be done over with semi-liquids or glazings ; and they 
 bring no foreign body into the painting. For designs on 
 stuffs, tapestries or other ornaments which have to be 
 traced on backgrounds, and which require going over, this 
 kind of tracing is very convenient, because by doing it with 
 
196 
 
 SCIENCE OF PAINTING. 
 
 the exact tone, there is no need to repaint the tracing. 
 No others should be used for tracings of perspective or 
 architecture often done in ink, seeing that the ink does 
 not adhere well on the painting ; and if it happen to come 
 "off, it brings off the top paint, and the tracings reappear, the 
 tones of the under part being thus laid bare. 
 
 Commercial Guarantee. 
 
 All the new productions spoken of in this book may be 
 found in the house of Lefranc & Co., 64 & 66, Rue de 
 Turenne, Paris. 
 
 Not only are these productions fabricated according to 
 our directions and formulas, but we have only authorised 
 their manufacture on condition that they should be at all 
 times under our supervision. 
 
 One can understand that an author may from self-esteem 
 desire his inventions to yield results such as he professes, 
 and that - he should surround himself with all possible 
 guarantees, but at the same time one can understand that 
 he cannot be responsible for all that may be fabricated 
 and has been already fabricated with his methods more or 
 less badly carried out. 
 
 That, therefore, is the reason why the author of this book 
 declares, that amongst the productions which commerce 
 may present as made according to his methods, he only 
 guarantees those which bear the trade mark of Lefranc 
 & Co. 
 
 With regard to colours ground specially for Water Coloicrs 
 to be fixed by Fire^ colours on this principle are prepared by 
 the house of Mary & Son, 26, Rue Chaptal. This house 
 has the monopoly of them.