REPORT TO THE SCIENCE AND ART DEPARTMENT OF THE COMMITTEE OF COUNCIL ON EDUCATION ON THE ACTIOir OP LIGHT OH WATER COLOURS. to iotti fioujfeii of ^arUamont ComtnanQ of Her ^ajeiftfi. LONDON: PRINTED FOR HER MAJESTY’S STATIONERY OFFICE, BY ETBE AND SPOTTISWOODE, PEINTEBS to the qheeh’s most excellent majestt. And to be purchased, either directly or through any BookseUer, from EYRE AND SPOTTISWOODE, East Harding Street, Fleet Street, E.C., and 32, Abingdon Street, Westminster, S.W. ; or ADAM AND CHARLES BLACK, 6, North Bridge, Edinburgh ; or HODGES, FIGGIS, & Co., 104, Grafton Street, Dublin. [C.— 6463.] Frice 2a. dd. 1888. REPORT TO THE SCIENCE AND AET DEPAETMENT OP THE COMMITTEE OF COUNCIL ON EDUCATION ACTION OF LIGHT ON WATER to iinti) of parltamcnt fig CnmmaitH of Her iWajctftg. FEINTED EOK HER MAJESTY’S STATIONERY OFFICE, BY EYRE AND SPOTTISWOODE, PEINTERS TO THE QUEEN’S MOST EXCELLENT MAJESTY. And to be purchased, either directly or through any Bookseller, from EYRE AND SPOTTISWOODE, East Haeding Steeet, Fleet Street, E.C., and 32, Abingdon Street, Westminster, S.W. ; or ADAM AND CHARLES BLACK, 6, North Bridge, Edinburgh; or HODGES, FIGGIS, & Co., 104, Graeton Street, Dublin. ON THE LONDON: 1888. [C.— 5453,] Price 2s. 9d. PREFACE. In April 188G tne Lords of the Committee of Council on Education requested Dr. Russell, F.R.S., and Captain Abney, R.E., E.R.S., to carry out an exhaustive series of experiments on the action of light on Water Colour- Drawings. These gentlemen having expressed their willing- ness to undertake the investigation, the sanction of the Treasury was obtained to the expenditure of a small sum of money for the provision of materials and in the payment of a student in the Art Training School for the preparation of the necessary tints. Shortly afterwards a resolution of the Royal Society of Painters in Water Colours was received urging “"tire “ desirability in the interests of Water Colour Painters, “ of the appointment of a Water Colour Painter in “ association with Dr. Russell and Captain Abney in the “ work of investigating the effects of light of various kinds “ upon Water Colour pigments.'’ Thereupon their Lordships of the Committee of Council passed the following minute, dated 12th June 1886 : — “ Seeing the interest that the question of the action of light on Paintings in Water Colours has excited, and the great importance of that question to Artists in this country ; it appears to my Lords desirable that there should be a representative Committee of Artists appointed to consider the matter from the Art point of view. Request the Royal Society of Painters in Water Colours and the Royal Institute of Water Colour Painters each to name two representatives to serve on a Committee, which Committee my Lords will themselves invite some other distinguished painters to join. “ Request Dr. Russell and Captain Abney, who have already undertaken to investigate as a scientific question the action of light on the various pigments used in painting, to inform the Committee of the method and nature of their inquiry. “ When the Committee have this information before them they will be in a position to judge whether there are any further points they would desire to suggest for investigation to Dr. Russell and Captain Abney or whether there is any A 54947. Wt. 25151. A 2 investigation wliich the Committee would themselves wish to carry out.” The Committee thus appointed consisted of : — Sir F. Leighton, Bart., P.R.A., Chairman. Mr. L. Alma Tadema, R.A. Mr. T. Armstrong. Mr. Sidney Colvin. Mr. Frank Dillon. Mr. Carl Haag. Sir James D. Linton. Mr. E. J. Poynter, R.A. Mr. Henry Wallis. Mr. Arthur Torrens, Secretary. This Committee have held four meetings, and Dr. Russell and Captain Abney have had numerous communications with individual members, from whom they received valuable information on the subject of their inquiry. Their first report, now presented, and dealing with the physical effects of light on water colours — the investigation into the nature of the chemical changes involved being deferred to a second report — was carefully considered by the Committee of Artists, who unanimously adopted a resolution, which was communicated to my Lords, that the Committee accepted the first report, and desired to record their sense of its very great value and of the thoroughness and ability with which so laborious an inquiry had been conducted. And further that they were of opinion that it would be of great advantage if the experimental research, which Dr. Russell and Captain Abney have been conducting into the action of light on water colours, were extended to its action on colours when used with oil and other media. By Order, J. F. D. DONNELLY. Science and Art Department, .30th June 1888. FIRST REPORT BY DR. W. J. RUSSELL, F.R.S., AND CART. W. DE W. ABNEY, C.B., R.E., E.R.S. Part I. — Introduction. I. An investigation of the cause of the fading of colours naturally divides itself into at least three parts. Plrst, the nature of the optical changes ; second, the nature of the chemi- cal changes ; and third, the causes which initiate and accelerate these changes. To carry out such an investigation necessarily requires much time, many of the changes, more especially the chemical ones, can only be brought about by brilliant sun- light, acting for a considerable length, of time. We have therefore divided our report into two sections, the first of which we now present. In it we have confined ourselves to the first division of our subject, the nature of the optical changes, and in some measure to the last division, reserving for our supplementary report a description of the results obtained in the second division. From the experiments described in this report certain obvious conclusions can be drawn, and these we shall indicate in due course, II. Before giving our results we have thought it advisable to make a few preliminary remarks on the optical properties of pigments and the different characters of light to which they may be exposed. Colours are popularly talked of as greens, blues, reds, &c., and to these distinctive appellations are added by artists. Thus they talk of emerald green, cobalt blue, Venetian red, &c. Though to some extent to the trained eye an idea is thus given of the hue and luminosity of the colour, yet to the scientific experimentalist, the definitions of colours require supplementing, and in esti- mating any change which may take place in them an exact quantitative value of each is a desideratum. As to why a pigment is coloured science has hitherto not furnished a satisfactory answer, nor for the inquiry which we have undertaken is it at all necessary that an answer should be given. The question, however, as to how a pigment pro- duces the impression of colour is one which can be answered. Colour is due to the selective action of the pigment, or stain, on light. That is to say, the rays of certain wave lengths (or colours) are transmitted and reflected by each pigment, or stain, in a more or less perfect manner and the others A 2 Division of the investi- gation. Nature oj colour. 4 - !ire absorbed. The transmitted and reflected rays will be shown to be identical — except in certain cases which it, is unnecessary to consider here — and it is to these that the colour of the pigment is due. Colour of III- If we decompose a thin slice of white light into its ■pigments illu- component colours by means of a prism we get what ^pectrmi'^ known as the spectrum, and if we allow this variegated band to fail on a white surface, such as white card or a surface of zinc oxide or barium sulphate, that surface becomes luminous where the colours fall, and they are presented to the eye with the greatest brilliancy possible. If, however, for the white surface we substitute a surface covered with some coloured pigment we at once jiercoive a difference. That which is generally supposed to be the colour of the pigment is lost and we have a stripe of the surface illuminated with the different rays of the spectrum, but their different colours are presented to the eye with their brilliancy unequally reduced. Probably no one coloured ray is reflected with the same brilliancy as it was from the white surface, and a large part of the spectrum is very much dimmed. In other words the pig- ment does not reflect any component of white light with the same intensity as does the white surface. If by proper means we collect the different coloured rays of the spectrum reflected from the coloured surface, and recombine them, we get back again the colour of pigment. If we measure the brightness of each colour reflected from the pigment in terms of the brightness of the same colours reflected from white paper, we have a quantitative measure of the light reflected from the pigment, such light being that by which the pig- ment produces the impression of colour. Thus, if we measure the light reflected from cobalt blue in the various parts of the spectrum, we may produce the colour of cobalt on white paper by reducing in proper proportions the different rays of the spectrum formed by white light, and then recombining them. The colour of a pigment, we may say, is dependent on the amount of the components of white light which it reflects back to the eye. Further it will be seen that no matter what the source of light may be, — whether the yellowish- white light of gas, the purer Avhite of the electric arc light, or of the sun, or the bluer white light from the sky, — the comparative measures of the brightnesses of the different colours reflected from the two surfaces will always be the same. Thus, if the stJectrum formed by gaslight be used, and the bidghtnesses of the different parts of the spectrum reflected from a white surface and from a coloured surface be compared, the com- parative values of the different rays thus obtained will be the same as if the source of light had been the sun. To see what 5 will be the colour of a pigment by gaslight, the rays of the spectrum of gaslight must be reduced in the proper propor- tion found for the particular pigment, and be again re- combined, and this will give the colour of the pigment seen by gaslight. In the same way, if the spectrum be formed by sunlight, the different parts of that spectrum must be reduced in exactly the same proportion, and the recombined spectrum will give the colour of the pigment as it would be seen in sunlight. Careful measurements of the various simple and mixed water colours which are usually employed by artists have been made and are given in Part III. The source of light employed was the electric arc light, and the relative intensities of the difterent coloured rays were measured by a method which one of us Avith Major-General Festing, F.R.S., has described in a paper recently read before the Royal Society. IV We will now trace the sequence of phenomena which happen when white light falls on a coloured surface. Colours may be divided into two classes, those which are insoluble and those which are soluble in water. Many of the soluble colouring matters are precipitated as lakes, and the same arguments hold good for these as for the solid pigments. In the case of an insoluble pigment a ray of white light falls on one of the particles, but only some components of the light can pass through it, and it emerges after its passage as coloured. This strikes the next particle and part is reflected back, I’eaching the eye as coloured light, but part penetrates through the next particle. This in its turn is partly reflected back and partly transmitted, and so on. At the same time, however, a certain amount of white light is reflected from the surface of the particles and mixes with the coloured light which has passed througli one or more particles. Hence the colour of a pigment always contains a certain per-centage of white light, together with the coloured light. Thus the colour of Prussian blue is principally due to the transmission through the particles of a large proportion of the violet blue and blue-green rays, together with a small quantity of white light reflected from the surface of the particles to which the original light falling on it had direct access. It may be remarked that when thin washes of these colours are Avashed over white paper the number of coloured particles are comparatively few, and that the Avhite paper reflects relatively more Avhite light to the eye. A microsco[uc examination of such a surface reveals this fact in a very interesting manner. The colour is therefore less intense in hue, and whiter. The nearest approach to the real colour of a water-colour pigment is seen Avhen a mass of the moist colour is on the pallette, the white light reflected being Production of colon r. 6 Identity of the light reflected from, and tra7ismitted by, pigments. then at its minimum. The same argument applies to those colours which are soluble, and which, consequently, we may take to be continuous. The light which penetrates such a colour will be of the same hue as that reflected to the eye from the paper. The white light passes through the stain, loses certain portions of the original spectrum colours, is reflected from the white paper beneath, and is once more transmitted through the colour, reacliing the eye as a coloured light. Thus the colour of the carmine (cochi- neal) is due to the transmission by the dje of most of the red and a good deal of the blue of the spectrum of white light. The effect of mixing a white with a coloured pigment is to cause more white light to be reflected, and thus to give a pale and less transparent appearance to the colour. V. It may be of interest to compare the light reflected from colours with that transmitted by them. The following diagram will give the results of two such colours, Prussian blue and Carmine. Fig. I. VIOLET BLUE GREEN YELLOW ORANGE RED Note. — The letters in this and in all other diagrams refer to the principal Fraunhofer lines of the solar spectrum. Curve I. is the light reflected from Prussian blue after deducting white light. Curve II. is the light transmitted through Prussian blue. „ III. „ t eflected from Carmine. „ I\'^ „ transmitted through Carmine. 7 In the Prussian blue the reflected and transmitted light are very nearly identical, but in the Carmine the transmitted light was less than the reflected. The small difference in the first case and the larger difference in the second is due to the fact that the depth of colour in the reflected and transmitted lights was not equal. VI. In the analysis of white light by the prism we have only so far mentioned those radiations which are visible, and which are called “ light.’^ These have been considered first, as it is these alone from which a pigment derives its colour. But there are other radiations which coexist with the visible radiations, and though invisible to the eye may have to be taken into account. Of these invisible radiations some lie beyond, or, as it is generally termed, below the red in the spectrum, and some beyond, or above the violet. Those which are below the red, and have a longer wave length, experiment has shown to possess more energy or capacity of doing work than all the radiation light”) above the red. What that work may be we shall touch upon briefly later on. It would be beyond the scope of this introduction to show how the comparative energies of the radiations, visible and invisible, forming the spectrum can be measured. It will sufiice to say that by allowing the different parts of the spectrum (light and dark) to fall on an undecomposable substance (lamp black) which can absorb them all (or very nearly all), the measurement by thermo- electric means of the rise in temperature of the lamp black produced by the different parts of the spectrum gives a comparative measure of the energy of radiation of these parts. It must be remembered that the energy of the slice of white light decomposed by the prism, and which includes the invisible radiations, is the sum of all the energies of the different radiations of the spectrum. Figure II. shows the comparative energies at different parts of the spectrum of sunlight, the electric (arc) light, and an incandescent light which is the same as that of gaslight. The heights of the curves denote the energies at different points of the spectrum produced by means of a prism. VII. Figure III. gives the luminosity of the different rays of the spectrum of sunlight on a day in July, of the electric light, and of gaslight. The ordinates or heights of the curves show the comparative brightness to the eye of the light at the different parts of the spectrum. The dark rays and the energy of the spec- trum. Luminosity and energy of the spectrum. 8 Fig. II. ■« ~ ^ ^ j\. VIOLET BLUE GREEN YELLOW ORANGE RED 9 In Figure IV. we have shown the portion of the energy curve of the electric (arc) light of the visible part of the spectrum together with Its luminosity curve, and It needs but a glance to show that the brightness of the visible spectrum bears no sort of relation to Its energy. The brightest part of the visible spectrum is in the yellow ; whilst, as before said, the point of maximum energy is just below the red. Had we used the luminosity and energy curves of any other source of radiation we should have arrived at exactly the same result. Fig. IV. darh ray a. VIII. The dark rays when falling on a pigment behave in H<:/lectiun of exactly the same manner as those which are visible ; some disappear, and others are transmitted, and though they give no colour to the pigment (it would appear of the same colour if they were excluded) are just as much in the reflected light as those rays which ai*e visible. We have shown that the light which is transmitted through a pigment is of the same general character as that reflected, and the same holds good for the dark rays. Fig. V. will give an idea of the total rays which are transmitted and reflected by Prussian blue and carmine as measured, not hy their briylitness, but by their cncryies, using the method indicated in Fig. VI. 7'he toj) curve shows the energy of the oi-iginal light, and the difference in heights between this curve and the two inner curves will give the energies of the rays reflected from the two pigments. 10 Fig. V. VIOLET BLUE CREEN ORANGE RED DARK It will be seen on comparing the two innermost with the outer curve that in the Prussian blue there are very few rays in the red, yellow, and green, v/hich are transmitted and consequently reflected, but that in the dark rays we again have a transmission of radiation. In the case of carmine more of the red is transmitted and less of the dark The energy existing in the invisible region beyond the violet is very small, but since all radiation consists of a wave motion in what physicists have called the ether, it may happen, and does indeed happen, that the wave period of this portion of the spectrum is of such a nature as to cause a destructive vibratory motion on the atoms of the molecules of the pigment on which the light falls. It is necessary, therefore, to take this action into consideration. We are now in a position to see that all radiation, visible and invisible, is not reflected or transmitted through the various pigments, and the question arises as to what has become of the rays which are apparently lost. The radiation 11 which is not reflected from the colouring matter has dis- appeared in passing through it ; in other words the pigment has absorbed certain of the radiations visible atid invisible, and with very close approximation to the truth it may be said that the rays absorbed are complementary to those reflected. Hence in Prussian blue (see Fig. I. & V.) the rays absorbed are principally in the dark part of the spectrum, and in the red and yellow, and partly, but in a minor degree, in the green, blue, and violet. Similarly in carmine, the rays absorbed are principally in the dark part, the yellow, and the green, of the spectrum, and less in the blue and violet of the spectrum. IX. Now it is a received axiom in physical science that in Absorption any body which absorbs radiation the energy so disappearing performs work of some kind in that body. The work so done is a raising the temperature of the body, a chemical decomposition of the body, or a re-arrangement of its molecular condition, as for instance in the iodide of mercury. Each kind of work may be done at the same time, but the energy which is expended on one form, cannot be again expended on another form, of work. On elements such as carbon, gold, &c., the work done only raises the temperature of the body above that of the surrounding objects, but in chemical compounds, such as nearly all the pigments in use are, there may be besides a chemical decomposition. The chemical decomposition of a colour means a fading or an alteration in its colour, but the raising of its temperature to the small degree which the visible (light), or invisible, radiation to which it is ordinarily exposed can effect, does not alter its composition or colour. That the temperature above that of the surrounding objects to which a colour can be raised, even by sunlight, when freely exposed, is small is not only shown by theoretical reasoning, but by direct experiment. In the case of a wash of water-colour, the particles on which the radiation falls are very small, and they consequently have a large surface compared with them volume, j^s the rapidity of loss of temperature in equal volumes under the same conditions is in the ratio of the radiating area, it follows that the loss of temperature by radiation in the small mass is very nearly equal to its gain. In other words there is an equilibrium of tempera- ture established which is but very little higher than the temperature of surrounding objects. When chemical decom- position takes place by light, however, the results are different. The decomposition once effected remains, and the quantity of the matter decomposed increases with the length of exposure. The outside of the particles is first acted upon, and then gradually (as light continues to act) the inner portions are decomposed, until finally the whole particle is 12 changed. Evidently those colours, the effect of light on which is to bleach them, are the most rapidly acted upon. Again, too, the large area of the surface of particles, as in a water-colour, compared with tlieir small volume, is favour- able for the rapid effect of the action of light in altering their composition. As work depends upon absorption, it is important to remember that when the radiation (light) is decomposed into its prismatic components it is only those rays of the spectrum which are absorbed that can do this work. Thus, if a pigment only absorbs in the red, it is only ihe red rays which can do work and no others, and so on. Estimation of X. In estimating the chemical action effected on a body by chemical action radiation there are thus two factors which have to be taken into account, viz., the intensity of the radiation acting, and the time during which it acts. To obtain the same amount of action in two cases the product of these two amounts must be the same. Thus if a certain tint be exposed to an intensity of radiation which we will call 100 and bleaches it in, say, one hour, then if a simitar tint be exposed to an intensity 1, it will require 100 hours’ exposure to it to effect the same bleaching. This has been fully proved by experi- ment. Theie is an idea abroad that if the light be very feeble a bleachable colour, no matter what length of exposure be given, will not facie. This, however, is not the case. The same proportion of the total energy absorbed by the body which, with an intense radiation, effects chemical decom- position, is expended with a feeble radiation in doing the same kind of work. To appreciate this we may very briefly allude to what the deductions from scientific experiment lead us to believe to be the manner in which light acts on the molecules of which a body is composed. Action of XI. In a compound body, the molecules must at the very “ lic/ht ” waves ]east consist of two ultimate atoms, and these oscillate to and another, each atom having its own constant time of completing an oscillation, the molecule itself oscillating in a period of its own. It need scarcely be said that the time of these oscillations is not to be measured even by millionths of a second, nor the extent of the oscillation by the millionths of an inch, but by standards very far smaller. A ray of light of any pure colour is due to a continuous series of oscillations or waves of a known and measurable length in the physicists’ “ ether ” which we have already mentioned. If it happens that the time of oscillation of some “ light” wave agrees with the time of oscillation of one of the atoms, the length of swing of the oscillation of this last is increased with each beat of the ether, till, if the number of beats of the ether be sutficiently numerous — that is if the light be allowed to play upon the molecule long enough — the length of its swing 13 is increased till finally the atom will swing off from the molecule, thus changing its composition. This liberated atom may join itself to the molecule of some other matter which may be present, such as oxygen or water. The amount of increased swing the waves of light can give the atom depends on the amplitude of the waves (the amplitude in a wave in the sea is the height fi'om trough to crest), the square of which is a measure of their energy as it is of the intensity of the light. To take a very familiar example, suppose we have a heavy church bell hung without any friction on its supports, and without any resistance to its motion, and that when vibrating freely, it would make a complete swing once in a second. Suppose to the end of the bell rope was attached a small horizontal plate, and that at intervals of a second 1,000 grains weight of water fell from a fixed height on to the plate, the bell would gradually oscillate, and finally the oscillations would become so great that it would ring. If instead of 1,000 grains falling from the same height we had one grain of water falling every second it would take 1,000 times longer before the bell rung, and if it was yoVo ^ grain of water that fell every second it would take 1,000,000 times as long before it rang. The work done by the dropping water may be looked upon as the work done by the amplitude of the wave, and the church bell as the atom, moving without friction and without resistance. XII. It will also be noticed that it is only those rays whose Chemical waves beat in unison with the oscillation of the atom which <}ct[on mat/ increase the swing of the atom. The wave motion is then theTye! ^ destroyed, and that particular ray disappears, i.e. is absorbed. It must be recollected that the visibility of any change effected on a body merely means the number of molecules altered. In feeble light the numbers altered in a given time are much fewer than when light is intense. We have a good instance of this in a photographic plate, where the effect of the exposure for of a second to sunlight on a salt of silver is invisible to the eye, whilst a second’s exposure is rendered visible. We know, however, that Trru.TCo ^ second’s exposure to sunlight has chemically altered some minute portion of the silver salt on which it fell, as what is termed development proves it. Again, we have a further definite proof in the case of certain colours that the smallest intensity of radiation if sufficiently prolonged effects a chemical change in them. In photographic processes the chloride of silver is only sensitive, roughly speaking, to the extreme violet of the visible spectrum. When any one of certain colours which are fugitive are applied to stain the silver chloride, and the part of the spectrum which the colour absorbs is below the violet, then after exposure in the spectrum on applying a developer, as it 14 Light to which water colours are exposed. Choice of light for experimen- tal work. is called in photography, the action of the spectrum in decom- posing the colour of the dye is shown by a deposit of silver taking place in that part which the colour absorbs. Thus, if carmine (cochineal) be applied to tbe chloride of silver an action will be shown to take place in the green where the colour, as will be seen, absorbs. It must be remembered that without that colour no such deposit would be possible. The spectrum of sunlight will cause this phenomenon to appear in a few seconds, and the spectrum of skylight, or of candlelight will equally cause it if the exposure is pro- longed. That is to say a feeble radiation (light), if sufficiently prolonged, will give the same effect as a radiation (light) which is several thousand times as intense. And it further demonstrates that chemical decomposition takes place in a colour long before such change is visible to the eye. The heating effect on a body may be taken to be an increase in the amplitude in the oscillations of the molecules rather than of the atoms (though the two are closely con- nected), pointing to the fact that the shorter wave-lengths which have a greater rapidity of oscillation are those which would be most likely to increase the amplitude of the oscillations of the atom, and thus to produce a chemical change in the body. We shall see further on that this the case. XIII. As to the light to which pigments in water-colour drawings are ordinarily exposed in a room, a few remarks must be made. There is no doubt that pictures are, as a rule, care- fully protected from direct sunlight, but it is nevertheless true that the greater portion of the light they receive is reflected sunlight. On a bright day clouds reflect sunlight, and on a dull day the principal part of the diffused light is also sunliofht which is reflected according: to the laws of geometrical optics from particle to particle, a certain pei’- centage eventually reaching the earth through the clouds. There is of course also a fair proportion of the light due to the sky, and this light is bluer than reflected or diffused and weakened sunlight. In cases where the windows of a gallery are in the vertical walls and have an uninterrupted view of the horizon, the blue light reflected is compara- tively small, the light near the horizon being distinctly more like sunlight than is that nearer the zenith. In galleries lighted like those at South Kensington by skylights the light to which pictures are subjected is on the whole bluer. The artificial lights to which water colours are exposed are gaslight, the arc, and Incandescence electric lights, and as we shall see presently the first and last are very deficient in blue rays. XIV. In conducting our experiments it became necessary to choose the light which would most readily adapt itself to giving a clue as to which colours were affected by exposure in 15 a time which would be measured by months instead of by years. A careful consideration led us deliberately to avail ourselves of as much sunlight as we could secure in this rather sunless climate of ours, together with the diffused and sky light, when sunshine was absent, which would act less ener- getically. We are aware that writers have expressed themselves as disinclined to accept deductions as to the fading of pigments when exposed to this bright light of the sun, but they have, as far as we are aware, never given any serious reasons for their disinclination. Their arguments have usually been based upon their own convictions rather than oil experimental proof of any kind, of if experimental proof has been quoted from other writers, half the truth or more is most frequently and probably unwittingly concealed. Probably, however, they express what is in the minds of many, so we shall enter somewhat fully into the arguments which decided us to adopt the step we did. XV. To the eye the hue of the lights mentioned above Cause of the undoubtedly differ considerably, and unless the cause of the difference in difference had been tracked out experimentally, and with ‘!^arious^Ughts, scientific exactness, it would have been unwise to have chosen out any one of them with which to conduct experiments, since the results obtained with it might not be applicable to any other. Happily, however, for such work, the spectro- scopic analysis of light furnishes irrefutable evidence that from the results obtained from exposure to one light, correct deductions may be made as to what would happen were the exposure made to another. If, by a prism, we analyse all the different kinds of light mentioned above, we find that in the visible spectra so obtained no colour is absent,* but if we compare the intensity of the same colours in the different spectra we find that there is a variation. For example, if we compare the spectrum of sunlight at mid-days in May with gaslight we find that there is considerable less violet, blue, green, and yellow light in the latter than in the former, and in light from a blue sky considerably less red and yellow. The following diagram shows the proportions in each colour. * We are not here taking into account the Fraunhofer lines, whieh in sun light and sky light are present. Even in these there is diminished radiation present. 16 Fig. VI. The intensities of the spectrum colours of smrlight near mid-day in May are not very different from those of the electric (arc) light, whilst the intensities of the colours of the incandescence electric light when rendered normally incandescent are a very close approximation to those of gas- light. The above diagram indicates in a striking manner tliat the light of the gas and therefore of the incandescence electric lamp, is yellower than that of snnlight, and therefore of the electric (arc) light, owing to the increasing diminution of comparative intensity of the colours from red to violet, whilst the light from the sky is considerably bluer than that from the sun. One of us has shown recently how by cutting off from the electric arc light (or from sunlight of known comj)osition) the proper proportions of the different spectrum colours, the exact liue of gasliglit or skylight can be produced. The difference in hue of the light from any of the sources we have been considering we may repeat is due to an excess or defect, but not to a total absence of Intensity of the different parts of the spectrum. As regards the dark rays of the spectrum the same argument holds good. Fig. II. will enable a judgment to be formed of the different proportions of dark rays in sunlight, the arc light and gasliglit. In light from the blue sky the proportion VIOLET GREEN YELLOW ORANGE RED BLUE 17 of dark rays is much smaller, and no very accurate measure- ment of their intensity has been made, but as it will shortly be shown that the rays below the red do not cause chemical change in any of the pigments we have tried, this want of accurate knowledge is of no great moment in the present instance. XVI. Since, then, all sources of light emit the same rays, Results from but of different intensities, which can be measured, it follows of that if we know which rays are chemically active, and the a%lilTto amount of work which, when of a certain intensity, they another source perform, we can, from the work done by the light from one of light. source, deduce the work that would be done by another. The most perfect manner of noting the action of light would be to expose for a given time the pigments to the action of the spectrum formed by an unvarying source of light, and to measure the amount of chemical action (fading of the eolour in most cases) which had taken place in every part of the spectrum. When the relative intensities of the different parts of the spectra from other sources of light compared with this standard spectrum were known, then the length of time during which it would be necessary to expose the colour to any one of them to produce that same total effect could be calculated. Unfortunately for our experiments, even in full sunlight, which is the most powerful light we can work with, months are often required to effect a visible chemical action on some of the pigments ; it was therefore useless to take a narrow slice of lisht, say ytg- inch in width, and ^ inch in height, and form a spectrum with it on a surface of coloured paper four inches in width and one-half inch in height, and wait to see when the bleaching took place. XVII. To avoid this impracticable method resort was had Use of coloured to the use of coloured glasses to ascertain the part of the spec- d^ouses m the . , . 1 i j.- • j ■ J.1 T • experiments. trum which was most active in producing the fading action. The glasses used were red, green, and blue. It may be well here to say something regarding what is meant by the terms red, green, blue, &c. as applied to glasses. It is a very popular idea that light coming through coloured glass is really white light, which is in some way transmuted into red, green, blue, &e., as the case may be. There is no transmutation. The effect of colour is merely produced by the abstraction of certain rays, or of a proportion of them, from the white light by the material of which the glass itself is composed. The following diagram shows the different parts of the spectrum, and the intensities of the rays which the above glasses transmitted, taking the spec- trum colours of white light as unity throughout. X 54947. B 18 Fig. VII. YELLOW ORANGE VIOLET The difference between the lieiglits of the three curves and the line which would occupy a horizontal position at 100 on the vertical scale had the diagram been continued up and which represents the intensity of the different colours forming the original white light before passing through tl'e glasses will give the amount of the different rays cut off. A table of these curves will be found in Part III. Practically the visible spectrum was divided into three parts, and a reference to the table of colours will show that in every case where any fading took place it was always found beneath the blue glass, very much less often and to a far less degree under the green, and only twice under the red glass, and was then barely perceptible. The blue glass also allows most of the dark rays beyond the violet to be transmitted. Experiment has shown that these rays are chemically active, but not to the same degree as those which are visibly transmitted through the blue glass. This might be expected as their energy or capacity of doing work is far less rroportion XVI IT. The following diagram will show the proportion of dark rays j.rj^yg ’^vldch pass through the different glasses. transmitted .id o through the coloured glasses. 19 VIOLET BLUE GREEN ORANGE RED DARK Fig. VIII. They are nearly entirely transmitted through the red glass, very slightly through the blue and green glasses. Had the fading of the colours we have examined been due to the dark rays, it ought to have been shown beneath the red glass far more than under the green or the blue glass. This was not the case, as a reference to Table VIII. will show. We may therefore say that the blue, violet, and ultraviolet rays are tliose which are by far the most active in producing a change in the pigments with which we have experimented. As the intensity of tlie different rays of the spectrum coming from the different sources of light have been measured, it follows that the total intensity of the rays from each source of light transmitted through the blue glass can be calculated. We may take these total Intensities as a very close approximation to an inverse measure of the time to which the colours would have to be exposed to produce an equal result as regards fading. B 2 20 Joint effect of heat and light. Deductions made from the cut that chemical changes take place in the blue rays. XIX. It might very properly be objected that although it has been shown that the dark rays do not affect chemical de- composition, it has not been proved that the heating effect they have on a pigment might not aid the rapidity with which the decomposition takes place. Direct experiments were undertaken with this object in view. The backs of papers coloured with pigments which we have proved to be fugitive were placed in contact with a tin containing boiling water and exposed to light, together with similar papers merely resting; ag;ainst wood. Some few of those colours which are affected by heat without light in an atmosphere saturated with moisture, see page 28, did fade with very slightly greater rapidity where exposed as above, but with the majority the rate of change was, if anything, slower. Further experiment has also shown that if the dark rays be cut off from sunlight by proper means the rate of fading in colours freely exposed is not diminished. In our experiments in the open tubes which will be described presently, the temperature was only a very few degrees higher than the temperature of outside atmosphere, and therefore the experiment made by heating the pigmented paper by contact with a vessel at the tempera- ture of boiling water was an extreme example of the effect of heat. A reference to the results of experiments shows that damp is often a factor in the I’apldity of fading, and as heat tends to lessen the moisture present in the paper and pigment, it might be expected that in the majority of cases fading would result more slowly when the pigment was heated whether by radiation or by heat applied as above. XX. That fading should principally take place in the blue rays was to be expected, from experiments that have been conducted with other objects, and is of great practical importance. We have already stated that it is only those parts of the spectrum which are absorbed by a colour that can do work on it. Of all colours, the reds, yellows, and greens absorb principally in the blue part of the spectrum {see diagrams of colours), and the blues much less. Hence we may expect that the former pigments would fade more rapidly than the latter if they are fugitive. And what is moi’e important, it locates the action of the spectrum to the I’egion which is least luminous, and which varies enormously in the different kinds of light to which pigments are exposed. Of the different kinds of daylight, viz., sunlight and sky- light, to which water-colours are exposed, sunlight is the safest wher; reduced to equal intensity, since it contains a far less propoi-tion of blue light than does skylight. 21 Fig. IX. 120 VIOLET BLUE CREEN YELLOW ORANGE RED Fig. IX. shows the luminosity of each part of the spectrum for skylight, sunlight, and gaslight, derived from Fig. VI., when the total illuminating effects of each to the eye are equal. The relative energies (capacities for doing work) of the light which is transmitted by blue glass in three cases are : blue skylight, 112'0; sunlight, 57’5; gas- light, 5’7, or very nearly in the proportions of 20, 10, and 1. XXI. We may now consider the amount of exposure given Duration of to our pigments, and thence deduce, within limits, the time ttie exposure that would be taken to produce a similar elFect in the light to which they would ordinarily be exposed. Between the middle of May to the middle of August, that is, from the time when the exposures were first made to the time when the first series of readings were made, there was registered at Kew Observatory 705 hours of bright sunshine, and at Greenwich 652. We may therefore take it that there were about 675 hours of the same bright sunshine at the place in which our colours were exposed. Not only, however, did they receive this sunshine, but also the light from the sky. The total number of hours of effective daj light which they received in the same time was about 1,700 hours. We can 22 make a very fair comparison between sunlight and skylight from the results obtained by experiment. When the sun is bright, we may take it that the sky is not overcast, but is fairly clear, and if, by photometry, we may measure the brightness of the total light from the sky and of sunlight, and of the skylight separately when illuminating a surface placed vertically and facing the direction towards which the papers were exposed, we can calculate the ratio of the brightness of sunlight and skylight. It would be manifestly of little use to take measures of the whole of the components of the light, for we have shown that if Ave take a unit of light of blue sky and a unit of sunlight, we have much more blue light in the former than in the latter. And, as it is the blue rays that we have shown to be effective in acting on those pigments which do fade, the photometry had to be confined to these rays alone. The average intensity of the blue rays in direct sunlight, (i.e., with the receiving sur- face held normally to the direction of the sun’s rays, is attained A^eiy nearly at 3'30, thus in the middle of August it is about •63 of tliat of the maximum. In the case of our tubes, however, the pigments were not exposed so that the surface was normal to the direction of the sun’s rays ; but always v/ith the surface vertical. As they were cylindrical, it might, at first sight, have appeared to be a matter of some difficulty to say whether the photometric measurements should be made with a vertical surface facing east, south, or west. A reference to the tubes themselves, however, solved the ques- tion, as it was found that the greatest fading took place in that part of the paper which was parallel to the building against which the tubes were hung, and it was this direction in which the surface of the photometer to be illuminated was placed. The conditions of exposure required that the effect should also take into account the reflection from the glass, and this was duly attended to. As a result, it Avas found that though the illumination by the sun near mid-day of a A'ertical surface facing 20'^ east of south was on an average nearly 4‘5 times more intense in blue light than Avas the sky, yet there was a steady diminution in the ratio after about three hours on each side of Avhen the maximum Avas attained ; owing to the greater inclination the paper had to the solar rays and also to the increased reflection from the glass from the same cause. A table of the Aertical intensity of sunlight at different times of the year Avill be found in Part III. It may be fairly taken that the average intensity of the blue of sunlight throughout the day is close upon 2'75 that of the average light from the sky. Although the sun, soon after its rising, shone upon the pigments, yet at about 3.30 in the afternoon they were shaded from the sunlight, and this reduces the number of hours’ sun Avhich they received to 23 about 500 hours. The above average value of sunlight to skylight was taken with a knowledge of this fact. XXII, We thus arrive at the conclusion that when the sun was shining for 500 hours, the pigments received blue light equal to l,875hours of that of a blue sky fully illuminated when the sun shone on them. Besides this, the pigments received 200 hours of blue sky towards sunset wlien the colours were in the shade, which may be taken as about equal to 50 hours of average skylight illumination. The light from a sky v/hich is cloudy has very much the same composition as sun- light itself, as we have repeatedly proved. Supposing, however, we take it that the light was half due to blue sky and half to that of sunlight, this is equivalent to the pig- ments being exposed for 600 hours to liglit of the same composition as that of a blue light from the sky and 600 hours to light of the same average composition as sunlight. Now, for equal units of illumination skylight, as already stated, is almost exactly twice as rich in blue rays as is sunlight. Therefore, the 600 hours to which the pigments were exposed to degraded sunlight is equivalent to 300 hours of light of the quality of skylight. Hence we may take it that the pigments were exposed when the sun was not shining to 900 hours of light of the same quality as that coming from the blue sky, but inferior in illumination. We have next to take some measure of Avhat this inferiority may be. Measui’es taken show that the light coming from the sky at the time of year when the exposures were made varies from |ths to -j^th, and sometimes less in brightness of that coming from an unclouded sky, the measurements being taken at the same time of day. If we assume that the average illumination of an overcast sky is ^rd of that of a blue sky, we shall not be far wrong. Applying this factor, we find that the pigments were exposed for an equivalent of 300 hours of average bright blue sky beyond that to which they were exposed when the sun was shining. It may therefore be said that the pigments received a total illumina- tion equivalent to 2,225 hours of average blue sky, which is made up of the 1,875 hours, the 50 hours, and the 300 hours. This of course is only an approximate estimate, owing to the very variable quantitie.s dealt with, but still it will give an idea of the illumination by the blue rays which were effective in causing the fading. XXIII. We may now go a step further, and calculate approximately the amount of illumination which a picture hung in a gallery, such as those at South Kensington, would receive during the same period. No direct sunlight would be admitted, and therefore the illumination due to the direct light from the sun would be eliminated. Photometric measurements show that the blue light illuminatinof o O Duration of ex- posure betivee?i May and August I88G, in ttms > f mean skylight. Calculation of exposure in a South Ken- sington gallery to produce fading. 24 Calculation of exposwe neces- sary to be given to gaslight to produce fading. a picture in these galleries varies between 4^!^^ and -^th of that to which it (when no blinds are used for subduing the light) would be subjected if it were placed where our pigments were exposed and illuminated by the sky alone. Now when there is a blue sky the ratio is least, and we shall be safe in taking it as yV^h. For the 700 hours when the sun was shining we should therefore have an equivalent inside the gallery to about 9'3 hours of average blue sky. For the 1,200 hours of light from an overcast sky we may take the factor of ^th both for the 600 hours of light which was of the quality of light from a blue sky and also of the light for the 600 hours which we supposed to be of the same quality as of sunlight, which, both together, we took to be equivalent to 300 hours of light from an unclouded blue sky. This would give an exposure equivalent to 7'5 hours of the average light from a blue sky, such as that to which the pigments were exposed, or in all 16 8 hours. This would make the exposure of a picture inside the gallery about 1^'^^ given to the pigments during the same time. If the whole year was of the same daily average brightness as that between 15th May and the 15th August, the same effect would have been pro- duced on the pigments located against the walls of one of these galleries in about 32 years. Seeing that the daily intensity and continuance of light is so enormously diminished in the autumn and winter we shall not be overstating facts when we say that it would have taken 100 years in the gallery in question to have arrived at the same degree of fading as to that to which the pigments had arrived by our sunlight experiments up to August 1886. XXIV. We will now endeavour to make an approximate estimate of the time which would have been required had our experiments been conducted in light falling on the walls of the same gallery when lighted by gaslight or the electric glow lamps. General Festing has furnished us with the measurements taken by him with the Preece photometer of the illumination of the walls of some of the galleries so lighted. The N.E. water colour room (gas) 1‘81 candles at 1 foot off. The S.E. water colour room (gas) 2'32 candles at 1 foot off. The Jones Bequest Gallery (electric glow lamp) 1'72 candles at 1 foot off. Eaphael Gallery electric (arc) light 2'26 candles at 1 foot off Sheepshanks Gallery electric (arc) light 3-12 candles at 1 foot. off. The glow lamp light and the gaslight have very closely the same composition, and we may, therefore, take it that 25 the mean illumination on the walls lighted by gashght and electric incandescence lamps is equal to two candles at one foot off. The illumination of the same galleries by daylight has been measured, and the mean light for the whole year may be taken as about six candles at 1 foot off. In Part III. details of the candle value of the illumination for the brightest months in the year has been given. The illumina- tion in the winter months is so small that this average may be taken. That is, the mean illumination by day is three times better than by night, but the blue rays in one unit of the illuminating value of light of gaslight are only Y^th on cloudy days to on days when the sky is clear of those contained in a unit day light. We may take x^th as a probable proportion, and on this assumption the blue light illumination by the latter is about of that of the former. That is to say, that one hour’s exposure to mean daylight is about equal to 45 hours of gaslight. We have already estimated that to produce the fading which took place in the colours between May and August in direct sunshine at least 100 years would have been required had the exposure been made in the gallery. Allowing for the duration of darkness, it would have taken at least 2,000 years continuous illumination to have produced the same result in gaslight or in the light from the electric glow lamps. With the arc electric light giving an illumination of 2^ candles at one foot off we calculate, on similar data, that the same result would have been obtained in not less than 200 years. XXV. Our final results were obtained after an exposure between May 1886 and March 1888, during which time there had been about 3,000 hours of sunshine in all, 2,100 of which we may take it fell on the colours. Making the same estimates as before, we find that this was equivalent to 8,800 hours of mean blue sky light. There would be left about 8,000 hours of light during which no sun was shining. Taking it as before, we should find that this was equal to 6,000 hom’s of subdued blue sky light, and when reduced to Jrd, ^would be equivalent to 2,000 hours of mean bright blue sky light. This, with the sun light, would give a total of 10,800 hours of the blue sky light to which the pigments were exposed, or about 4'8 times the amount which they received between May and August 1866. To produce our final result there- fore we should have had to expose them in the gallery for at least 480 years, and to the gas light continuously for 9,600 years. Had we exposed to the whole of the light coming from the southern sky alone, shielding the colours from direct sunlight, we should have had to extend our observations for four years, and if to a northern sky probably for nearer 10 years, since the mean brightness of the latter is considerably less than the former. Exposure necessary to be given in a South Ken- sington gallery to produce final results obtained in simligkt. 26 Final remarks on exposure necessary to produce fading. XXVI. With these estimates before us, it is not surprising that vve should liave preferred to use an illumination which would give us results which we ourselves should be able to discuss. We may remark that a certain amount of impatience has been exhibited in some quarters at what to them appears the prolonged time which has elapsed since our experiments were commenced. We trust that the statement we have made regarding the approximate lengths of exposure which are required will show that experiments of this nature are not capable of being hurried, or, when partially completed, of being discussed except after weighing all causes which are operating. We must here enter a protest against the loose way in which comparative exposures in sun light and sky and subdued light are spoken of. For instance, the fact that colours have been exposed for so many hours sunshine as compared with so many hours of light from the sky has but little meaning or value (a) unless the total effective light during tlie whole period of both be measured, or (/>) unless the exposure has been so prolonged that it is safe to resort to averages. We have carried out experiments in both directions and the one confirms the other. We may again emphatically repeat that experiment has shown that knowing the composition of the light used and intensity of its different components, and the effect whicii they produce on a pigment in a given time, it is only a matter of calculation to arrive at the time necessary to produce the same result with any other light whose compo- sition is known. The details of the experiments on which the foregoing calculations are based have in part already been communicated to the lioyal Society by one of us and Major-General Testing, F. 11. S., and some of the remainder will be also communicated to that body, and others will be embodied in our second report. 27 PART II. Description and Result oe Experiments with Various Colours. I. In the following experiments the moist colours of one Colours em- firm were employed. We have tried, as will be seen, the action of light on the single colours, and on mixtures two or more colours. Sir J. D. Linton P.R.I., Mr. E. J. Poynter, R.A., and oUiers kindly supplied us with an account and samples of the different mixtures they employed, and these were copied with as much accui’acy as possible. Those mixtures were avoided in which a change would of necessity take place without the action of light owing to the known chemical composition of the colours. We have not confined ourselves solely to the above makers’ colours, but have experimented with colours from other makers both dry, in pans, and in tubes. An account of these experiments we reserve for our next report, merely mentioning that tiie results so far do not greatly differ from those recorded below. II. The paper which we have used is Whatman’s, and in em- order that no variation of quality should occur in different ployed. experiments, we obtained at once sufficient for the whole investigation. This paper was examined and found to con- tain only a trace of thiosulphate and in every square foot nearly 1 grain of sizing matter. The amount of moisture present, as we shall show, is a matter of con- siderable importance, and obviously will vary with the condition of the surrounding atmosphere. We have found that the “ hot pressed ” paper is capable of absorbing from a moist atmosphere as much as 12 46 per cent, of its weight of water, the “ not pressed” 12'20, and the rough 12’07 per cent. This absorption of moisture goes on slowly, even when the paper is fully exposed to a saturated .air. Twenty-four hours elapse before the paper is perfectly saturated.. III. The published experiments on the action of light on ofpre- colours are few, and have been made under undefined con- paring the ditions. We therefore commenced our investigation by taking <^oloured a very large number of the colours which are ordinarily used and exposing them to conditions rigorously determined. These conditions were selected so as to give us definite information as to the natui'e of the changes, if any, which occurred. The colours to be tested were applied, by a practised hand, to the paper in a series of washes, the first wash extending over the whole sheet, the second one leaving a strip 1 in. wide and the length of the paper untouched. The following figure represents one of these strips. In most cases as many as eight washes were applied, giving thus a complete series of 8 tints. In the following experiments strips 2 ins. wide and 8 ins. long, having all the tints upon them, were used. Fig. X. represents one of these strips. 28 Experiments 171 dap and sun light. Fig. XI IV. In the first series of experiments the colours were subjected to conditions similar to those to which pictures are subjected, but to an exaggerated extent. The experiments were carried out as follows : two strips of the coloured paper cut from the same sheet were carefully introduced into a glass tube f in. in diameter and 2 ft. long, open at both ends, the upper end being bent over to prevent the entrance of wet and dirt. The tubes were hung vertically out of doors against a wall facing nearly south, where all the sunshine until after 3.30 could fall upon them. A piece of American cloth was carefully bound round one half of the tube, thus effectually protecting one strip of the paper from light. The two pieces of identically tinted coloured paper were therefore under exactly the same conditions in all respects, except that one was exposed to light, and the other was in the dark. In these experiments the colours were exposed to the action of light, air, and moisture, as are pictures, only to a greater extent. They had to bear the action of direct sunlight, the air circulated through the tube, and the paper was in contact with the outer air. This free circulation of the air also prevented, as already indicated, any very appreciable rise in temperature in the tube. These papers were exposed fi’om May 1886 till March 1888. They were observed for the first time on the 14th of August 1886, again in December 1886, and in July and in November 1887, and finally in March 1888. The results of the first and last examination are given in Table I. 29 Table I. Open Tube. Name of Colour. August 14th, 1886. Mai’ch, 1888. Remarks.* Carmine - - 1 Gone - Gone - - Crimson Lake - Gone to 7 Gone - - Scarlet Lake - Pink gone.Ver- Gone - — milion left. Vermilion - No change - Gone black - — Rose Madder No change - Faded to 4, ~ and bluer. Madder Lake No change - SI. faded — Indian Red - No change - No change - - Venetian Red No change - No change - - Brown Madder Changed to 4 Faded to 8 - - Burnt Sienna No change No change - - Gamboge Faded to 2 - Faded to 7 - - Aureolin - No change - Faded to 4 ■ - Chrome Yellow No change - No change - - Cadmium Yellow - No change - Gone - - Yellow Ochre No change - V. si. faded - - Lemon Yellow No change - No change - Naples Yellow Faded to 6 - Gone - - Indian Yellow Faded to 2 - Faded to 6 - — Raw Sienna - No change - No change - - Emerald Green No change SI. brown — Terra Verte - No change - No change - — Chrom. Oxide No change - No change - - Olive Green - Blue gone Gone brownish Antwerp Blue pink. Paler Gone green - Blue revived. Prussian Blue A little lighter. No change - Indigo Blue no green. Gone to 4, all Faded to 8 - Cobalt Blue lighter. No change - No change - — French Blue No change - No change - — Ultramarine Ash No change - No change - — Leitches (cyanin) Blue V- si. faded - No experiment _ Permanent Blue No change - Faded to 4 - — Paynes Grey Gone red to 7 Gone - — Violet Carmine Red nearly Bleached to 6 — Purple Carmine gone. Red nearly Bleached to 6 Purple Madder gone. Faded to 7 - Faded to 8 - Sepia ... Faded to 1, all Faded to 8 - Vandyke Brown lighter. Faded to 4 - Gone - Burnt Umber - No change ■ V. si. faded - - mrathin the ™ column took place after the pigments had been about one Note— SI. means slightly ; V. si. means very slightly; No. 1 is the faintest tint. Effect of light on colours exposed in open tubes. 30 A’'ame of Colour. August 14th, 1886. March, 1888. Remarks. Brown Pink Indian Yellow and Rose Madder Rose Madder and Raw Sienna Raw Sienna and Venetian Rod Venetian Red, Madder Rod, Indian Yellow. Vermilion and Chrome Yellow Indian Red and Rose Madder Indian Yellow and Rose Madder Burnt Sienna and Naples Yellow Indigo, Indian Yellow, Raw and Burnt Sienna. Indigo and Gamboge Prussian Blue and Gamboge Burnt Sienna and Antwerp Blue Raw Sienna and Antwerp Blue Prussian Blue, Raw and Burnt Sienna. Indigo and Vandyke Brown Prussian Blue and Burnt Sienna Prussian Blue and Raw Sienna Indigo and Raw Sienna Indigo and Burnt Sienna - Leitches Blue and Burnt Sienna Leitches Blue and Raw Sienna Indigo, Raw and Burnt Sienna Prussian Blue and Vandyke Brown - Indigo and Venetian Red - Prussian Blue and Indian Red Coh.alt and Indian Red Prussian Blue and Venetian Red Antwerp Blue and Rose Madder Prussian Blue and Crimson Lake Antwerp Blue and Crimson Lake Indigo, Venetian Red, Yellow Ochre Prussian Blue, Yellow Ochre, Vene- tian Red. Indigo, Ra w and Burnt Sienna, and Indian Yellow. Intligo and Indian Red Gone to 3 Gone pink to 0, yellow go- ing in all. No change No change - Rodder in all No change - No change - Part faded, yellow gone to 3. Browner Gone reddish in all. Gamboge go, ing in all. Yellow goin, SI. redder SI. redder Blue gone Vandyke brown going, all gone as far as 2. Blue gone Browner Reddish SI. Browner - Browner Browner BroTOer Blue and fading to 3, Van- dyke brown gone in all. Redder Gone red in all. No change - Gone red in all. Gone pinkish Red all gone, blue lighter. Red all gone - Redder Redder Browner SI. redder Faded to 7 Gone pink V. si. yellowed to N 0 change - SI. faded, gone pink. Darkened No experiment Gone pink V. si. faded - Gone rod Faded to 6 - Gone blue to 7 Gone s', brown N 0 change - Brow'n Faded to 5 - Gone brown - Gone brown •' Gone yellow to 5. Gone brown - Gone brown - Gone brown - Gone red Blue gone and a little brown left. Gone red Gone red No change Gone pink Gone blue Gone blue Gone reddish Gone red Gone brown Red to 7 . Blue revived. Blue I'ovived 1 to 3. Blue revived. Blue revived I to 5. Blue revived. Blue revived. Note.— W here Nos. are given in this and in following tables the higher the number the deeper the tint referred to. Thus, in these experiments. No. 1 is the faintest tint and No. 8 the deepest tint. Note.— SI. means slightly; V. si. means very slightly; No. 1 is the faintest tint. 31 In the fourth column we have noted the curious change wliich often occurs with Prussian blue, namely, the return of the blue colour if the faded paper be placed for some time in comparative darkness. This very interesting change will be discussed in detail in the second part of our report. In sotne cases the colour entirely disappeared, in carmine for instance. In the majority of cases only a part of the colour disappeared, the thinner washes fading out; but the following pigments were able to withstand this most trying ordeal, and remained unchanged. They are : Indian red, Venetian red, burnt sienna, chrome yellow, lemon yellow, raw sienna, terra verte, chromium oxide, Prussian blue, cobalt blue, French blue, and ultramarine ash. The following mixtures also underwent no change : raw sienna and Venetian red, raw sienna and Antwerp blue, cobalt and Indian red. The table will show that other colours and mix- tures under these extreme conditions are only very slightly acted on, and that no actually sharp line can be drawn in these cases. V. Table II. shows approximately the order of instability of the single colours in open tubes which we have tried, of colours be^innlno; with the most fugitive. TABLE II. exposed in an ordinary atmosphere. Carmine. Crimson Lake. Purple Madder. Scarlet Lake. Paynes Grey. Naples Yellow. Olive Green. Indigo. Brown Madder. Gamboge. Vandyke Brown. Brown Pink. Indian Yellow. Cadmium Yellow. Leltches Blue. Violet Carmine. Purple Carmine. Sepia. Aureolln. Rose Madder. Permanent Blue. Antwerp Blue. Madder Lake. Vermilion. Emerald Green. Burnt Umber. Yellow Ochre. Indian Red. Venetian Red. Burnt Sienna. Chrome Yellow. Lemon Yellow. Raw Sienna. Terra Verte. Chromium Oxide. Prussian Blue. Cobalt. French Blue. Ultramarine Ash. Of these 39 single colours 12 were not acted upon at all by light, and two others were only after this long exposure to direct sunlight very slightly faded. 32 All of these, except Prussian blue, are purely mineral colours. Of the 34 mixtures tried only three remained from first till last unchanged, but six mixtures containing Prussian blue, although at first altered, on placing in the dark for six weeks more or less returned to their original colour. It is of considerable interest to note that in the cases in which any change occurred it had commenced before our record made in December 1886, though not in all cases before August 14th. Exposure of YI. In another series of experiments, carried out at the \n °di^yair!^^ ^ same time with mostly the same pigments, the atmosphere to which they were exposed Avas free from all moisture. The glass tube was first heated, allowed sufficiently to cool, the dried tinted paper carefully introduced, and the glass tube hermetically sealed. As before, two similarly coloured strips of paper were introduced into each tube, one was protected from light, the other expc>sed fully to it. Table III. gives the results obtained. Thirty-eight experi- ments were made with single colours ; but under this altered condition, 22 instead of 12 were found to be perma- nent, principally those colours which in the former experiments were only very slightly faded. In two cases the colour in the open tube was not acted on, while that in the dry tube was ; these cases are brown madder and Prussian blue. The colours Avhich were unchanged in dry air, but were acted on in ordinary air are madder lake, cadmium yellow, Naples yellow, emerald green, olive green, Paynes grey, sepia, and burnt umber. Again, Avith the single exception of madder lake, all the above which were not acted on in dry air are mineral colours. TABLE III. Name of Colour. Dry Air. Carmine ... _ Faded to 7. Crimson Lake - - Gone to 5. Scarlet Lake - - - - Faded and darkened. Vermilion - - Gone black. Rose Madder - - No change. Madder Lake - - - No change. Indian Red - - No change. Venetian Red - - - No change. Brown Madder - - - Faded to 4. Burnt Sienna - - - No change. Gamboge - - Faded to 3. 33 Name of Colour. Dry Air. Aureolin . . ^ - No change. Chrome Yellow - - - - No change. Cadmium Yellow - No change. Yellow Ochre - - - - No change. Naples Yellow - - - - No change. Indian Yellow - - . Faded to 4. Raw Sienna - - - No change. Emerald Green ... No change. Terra Verte - - - - No change. Chrom. Oxide - - - - No change. Olive Green - - - - No change. Antwerp Blue - - - Faded to 3. Prussian Blue - - - - Faded to 5. Indigo Blue - - - - Faded to 7. Cobalt Blue - - - - No change. French Blue - . - . Nochange. Ultramarine Ash - - - - No change. Leitches Blue - - - - Faded to 5. Permanent Blue - - - - No change. Paynes Grey - - - - No change. Violet Carmine - - - - Faded and brown. Purple Carmine ... Faded. Purple Madder ... Faded to 4. Sepia - - - - - No change. Vandyke Brown ... V. si. faded. Burnt Umber - - - - No change. Brown Pink - Faded to 4. Note. — SI. means slightly ; V. si. means very slightly ; No. 1 is the faintest tint. VII. In the next series of experiments the colours were exposed to air fully saturated witli moisture. The paper was saturated with moisture, and was sealed up in tubes containing moist air. Thirty-seven experiments with single colours were made ; the results are given in Table IV. Only 10 colours withstood the action of light under this condition : these were Indian red, Venetian red, burnt sienna, yellow ochre, raw sienna, emerald green, terra verte, chromium oxide, cobalt, and ultramarine a.sh. No vegetable colour is in this list, and both Prussian blue and Antwerp blue were entirely destroyed. Twenty-nine mixtures were also tested in a similar way, and only two remained unchanged ; these were raw sienna and Venetian red, and cobalt and Indian red. A 54947. Exposure of colours to light in moist air. 34 TABLE IV. Name of Colour. Moist Air. Carmine Gone. Crimson Lake - - Gone. Scarlet Lake - - - Faded and blackened. Vermilion • - Gone black. Rose Madder - - - Faded to 6. Madder Lake - - Faded to 5. Indian Red - - No change. Venetian Red - - No change. Brown Madder - - Faded to 4. ' Burnt Sienna - - No change. Gamboge - - Faded to 3. Aureolin - - F'aded to 6. Cadmium Yellow - - Faded to 3. Yellow Ochre - - - No change. Lemon Yellow - - - Faded to 4. Naples Yellow - - Faded to 8. Indian Yellow - - _ Faded to 6. Raw Sienna - - - No change. Emerald Green - - - No change. Terra Verte - - - No change. Chrom. Oxide - - No change. Olive Green - - - Gone brown. Antwerp Blue - - Gone. Prussian Blue i - - Gone. Indigo Blue - . Faded to 8. Cobalt Blue - _ No change. French Blue - - Bleached to 2. Ultramarine - - No change. Permanent Blue - Faded to 5. Paynes Grey - - Faded to 6. Violet Carmine - - Gone brown. Purple Carmine - - - Gone brown. Purple Madder - - Gone. Sepia - - - - Faded to 7. Vandyke Brown - - Faded to 5. Burnt Umber - - No change. Brown Pink - - - Faded to 7. Indian Yellow and Rose Madder - - Pink. Rose Madder and Raw Sienna - - V. si. faded. Raw Sienna and Venetian Red - - No change. Venetian Red, Madder Red, Indian Yellow - Gone red. Vermilion and Chrome Yellow - Gone black. Indian Red, Rose Madder - - Gone pink. Burnt Sienna and Naples Yellow - - V. si. faded. Indigo, Indian Yellow, Raw and Burnt Sienna Gone red. Indigo and Gamboge - - Faded to 5. 35 Name of Colour. Moist Air. Burnt Sienna and Antwerp Blue - Eaw Sienna and Antwerp Blue Prussian Blue, Raw and Burnt Sienna, and Indian Yellow. Prussian Blue and Burnt Sienna Indigo and Vandyke Brown - - - Prussian Blue and Burnt Sienna Prussian Blue and Raw Sienna Indigo and Raw Sienna - - - Indigo and Burnt Sienna - - - Leitches Blue and Burnt Sienna Leitches Blue and Raw Sienna Indigo, Raw and Burnt Sienna Indigo and Venetian Red Cobalt and Indian Red - - - Indigo and Indian Red . . . Prussian Blue and Venetian Red Antwerp Blue and Rose Madder Prussian Blue and Crimson Lake - Antwerp Blue and Crimson Lake Indigo, Venetian Red, Yellow Ochre Gone brown. Gone brown. Gone brown. Gone red. Gone blue. Gone red. Gone brown. Gone red. Gone bl own. Gone brown. Gone brown. Gone red. Gone red. No change. Gone red. Gone red. Gone pink. Gone. Gone blue. Gone red. Note. — SI. means sliglitly ; V. si. means very slightly ; No. 1 is the faintest tint. VII. To eliminate the effect of oxygen, a further series of experiments were made in which the colour was exposed in an atmosphere of moist hydrogen gas. Thirty-six experiments were made with single colours, and of these no less than 22 remained unchanged ; even carmine and crimson lake did not alter, neither did madder lake, Indian red, Venetian red, brown madder, burnt sienna, chrome, yellow, yellow ochre, raw sienna, terra verte, chromium oxide, olive green, indigo, cobalt, French blue, ultramarine ash, permanent blue, Paynes grey, sepia, Vandyke brown, and burnt umber. The results are given in Table V. TABLE V. Name of Colour. Moist Hydrogen. Carmine No change. Crimson Lake - - - No change. Scarlet Lake . - - No change. Vermilion - - Gone black. Madder Lake - - - - No change. Indian Red - - No change. c 2 Exposure oj colours to tiplil hi the presence of hyih-ojen. 36 Name of Colour. Moist Hydrogen. Venetian Red _ No change. Brown Madder - - No change. Burnt Sienna - - No change. Gamboge - - SI. faded. Aureolin - - Faded. Chrome Yellow - - No change. Cadmium Yellow - - SI. faded. Yellow Ochre - - - No change. Naples Yellow - - Faded. Indian Yellow - - Faded. Raw Sienna - - No change. Emerald Green - - - Black and pink. Terra Verte - - - No change. Chrom. Oxide - - No change. Olive Green - - - Gone brown. Antwerp Blue - - Gone. Prussian Blue - - - Gone. Indigo Blue - - No change. Cobalt Blue - - No change. French Blue - - No change. Ultramarine - - - No change. Leitches Blue - - Faded to 6. Permanent Blue - - No change. Paynes Grey - - No change. Violet Carmine - - Black. Purple Madder - - Faded to 3. Sepia - - - - - No change. Vandyke Brown - - No change. Burnt Umber - - No change. Brown Pink - - Gone yellow. Note. — SI. means slightly ; V. si. means very slightly ; No. 1 is the faintest tint. Exposure of colours to light in vacuo. VIII. Another series of experiments were made, and in these the air and moisture were, as far as possible, removed from the tubes containing the coloured papers ; the papers were carefully dried, and the air pumped by a Sprengel pump out of the tube, which was then hermetically sealed. Thirty-nine experiments with single colours were made, and it will be seen by Table VI. that hardly any colour under this condition is acted on by light. Violet carmine and purple carmine slightly darkened ; Prussian blue and purple madder and sepia slightly bleached ; but in all cases the action was very feeble. Twenty- four experiments were made with mixed colours, and the results are of much interest and importance. The mixtures containing I russlan blue changed, the other colour becoming dominant. Vermilion also blackened. With other mixtures hardly any change occurred. TABLE VJ. Name of Colour. Vacuum. Carmine . No change. Crimson Lake - - No change. Scarlet Lake - - - No change. Vermilion - - Gone black. Rose Madder - - No change. Madder Lake - - - No change. Indian Red - - No change. Venetian Red - - - No change. Brown Madder - - No change. Burnt Sienna - - No change. Gamboge - - No change. Aureolin - - No change. Chrome Yellow - - No change. Cadmium Yellow - - No change. Yellow Ochre - - No change. Lemon Yellow - - - No change. Naples Yellow - - No change. Indian Yellow - - - No change. Raw Sienna - - SI. darkened. Emerald Green - - No change. Terra Verte - - No change. Chrom. Oxide - - - No change. Olive Green - - No change. Antwerp Blue - - No change. Prussian Blue - . V. si. faded. Indigo Blue - - No change. Cobalt Blue - - No change. French Blue - - No change. Ultramarine Ash - - No change. Leitches Blue. - - No change. Permanent Blue. - - No change. Paynes Grey. - - No change. Violet Carmine - - - - - SI. darkened. Purple Carmine - - SI. darkened. Purple Madder - - V. si. gone. Sepia ... - - SI. faded to 6. Vandyke Brown - - - No change. Burnt Umber - - No change. Brown Piuk - - No change. Indian Yellow and Rose Madder - - No change. Rose Madder and Raw Sienna - - No change. Raw Sienna and Venetian Red - - No change. Vermilion and Cbrome Yellow - - More yellow. Burnt Sienna and Naples Yellow - - V. si. faded. indigo, Indian Y^ellow, Raw and Burnt Sienna No change. Indigo and Gamboge - - Gone blue. Prussian Blue and Gamboge - - Gone green. Burnt Sienna and Antwerp Blue - - Gone red. 38 Summation of results. Name of Colour. Vacuum. Raw Sienna and Antwerp Blue - - - Prussian Blue, Raw and Burnt Sienna, and Indian Yellow. Prussian Blue and Burnt Sienna Indigo and Vandyke Brown - - - Prussian Blue and Burnt Sienna Prussian Blue and Raw Sienna Indigo and Raw Sienna - - - Gone brown. Gone brown. Gone brown. Faded. Gone brown. Gone red. No change. Indigo and Burnt Sienna Indigo, Raw and Burnt Sienna Prussian Blue and Vandyke Brown - Indigo and Venetian Red Prussian Blue and Indian Red Indigo and Indian Red Prussian Blue and Crimson Lake Antwerp Blue and Crimson Lake Indigo, Venetian Red, Yellow Ochre Prussian Blue, Yellow Ochre, Venetian Red No change. No change. Gone brown, No change. Gone red. No change. Gone pink. Gone pink. No change. Gone red. Note.— SI. means slightly; V. si. means very slightly ; No. 1 is the faintest tint. IX. The action of the surrounding medium on colours ex- posed to light is strikingly shown by the above experiments with moist air, dry air, and a vacuum, and in order to show at a glance the different effects produced, we append Table VII., in which we have grouped the colours thus acted on under the three heads of No change. Altered, and Destroyed. Under the heading of Altered, are included changes ranging from a very small destruction of colour to a very con- siderable one. Of course such a classification can only be approximate. TABLE VII. Single Colours. Moist Air. Dry Air. Vacuum. No Change. No Change. No Change. Indian red. Indian red. Indian red. Venetian red. Venetian red. Venetian red. Burnt sienna. Rose madder. Carmine. Yellow ochre. Madder lake. Crimson lake. Raw sienna. Aureolin. Scarlet lake. Emerald green. Chrome yellow. Rose madder. Terra verte. Cadmium yellow. Madder lake. Chromium oxide. Yellow ochre. Brown madder. Cohalt. Naples yellow. Burnt sienna. Ultramarine ash. Raw sienna. Gamboge. 39 Moist Air. i Dry Air. 1 I Vacuum. 1 No change. No change. Emerald green. Aureolin. Terra verte. Chrome yellow. Chromium oxide. Cadmium yellow. Olive green. Yellow ochre. Cobalt. Lemon yellow. French blue. Naples yellow. Ultramarine ash. Indian yellow. Permanent blue. Emerald green. Paynes grey. Terra verte. Sepia. Chromium oxide. Burnt umber. Olive green. Burnt sienna. Antwerp blue. Indigo. Cobalt. French blue. Ultramarine ash. Vandyke brown. Burnt umber. Brown pink. Altered. Altered. Altered. Scarlet lake. Carmine. Raw sienna (slightly Vermilion (black). Scarlet lake. darkened). Rose madder. Vermilion (black). Prussian blue (slightly Madder lake. Brown madder. darkened). Brown madder. Gamboge. Violet carmine Gamboge (very slightly). Indian yellow. (darker). Aureolin. Antwerp blue. Purple carmine Cadmium yellow. Prussian blue. (darker). Lemon yellow. Indigo. Purple madder (very Naples yellow. Leitches blue. slightly faded). Indian yellow. Violet carmine. Sepia. Olive green (brown). Purple carmine. Vermilion (blackened). Indigo. Purple madder. French blue. Vandyke brown (very Permanent blue. slight). Paynes grey. Violet carmine (brown). Purple carmine (brown). Purple madder. Sepia. Burnt umber (very slight). Brown pink. Vandjke brown. Brown pink. Destroyed. Destroyed. Destroyed. ■ ■ ■ - Carmine. Crimson lake. None. None. Antwerp blue. Prussian blue. ' 40 Effect of Ox galL Experiments with the arc electric light. X. To determine whether a mixture of ox gall woxild influence the results, the following colours were mixed with that substance and tested in the same way and under the same conditions as the colours above had been tested. In no case did we find that the addition of the ox gall altered the result : — Scarlet lake. Rose madder. Indian red. Rose madder and raw sienna. Indian yellow. Indian yellow and rose madder. Indian yellow, raw and burnt sienna. Indigo and gamboge. Indigo and raw sienna. Indigo and burnt sienna. Prussian blue. Indigo. Indigo and Venetian red. Indigo, Venetian red, and yellow ochre. XI. In selecting the colours to be exposed to the electric light we have chiefly taken those most easily acted on. These were exposed, some in a frame with a glass in front, others under the most favourable conditions for fading, viz., in sealed tube with moist air to the full action of the electric light for about 84 hours. The light falling on these colours is estimated as having an illuminating value of 2,000 candles at a foot off. Under these conditions the only single colours in the frame which underwent change were crimson lake (bleached and more red), and carmine (bleached and more red). Antwerp blue, Prussian blue, Naples yellow, brown pink, purple madder, rose madder, Vandyke brown, sepia, vermilion, Venetian red, Indian yellow, gamboge, indigo, Leitches blue underwent no change. With mixtures, Prussian blue yellow ochre and Vandyke brown, Prussian blue and raw and burnt sienna, and Prussian blue and burnt sienna all became slightly browner. Indigo and burnt sienna, indigo and gamboge, indigo and Indian red, Prussian blue and gamboge, Prussian blue and purple carmine, Prussian blue and Vandyke brown, Prussian blue and Indian red, Antwerp blue and raw sienna, rose madder and raw sienna, chrome yellow and vermilion, Venetian red and raw sienna underwent no change. In the tubes with moist air, carmine and Naples yellow very slightly altered, and crimson lake altered more. The following colours underwent no change : rose madder, vermilion, madder lake, gamboge, Indian yellow, Prussian blue, Antwerp 41 blue, indigo, Leitches blue, Vandyke brown, brown pink sepia ; neither did the mixtures of Prussian blue and Indian red, Prussian blue and gamboge, indigo and burnt sienna, indigo and Indian red. From tiic nature of these experiments tliey could not be carried on fur an indefinite length of time. XII. To determine whether heat, without light, would have Experiment any effect, the strips of the following colours, cut from the Mh heat but same sheets as those used in the previous experiments, were sealed up with moist air in glass tubes and heated for seven hours a day for three weeks in boiling water, all light being excluded. Indian yellow changed very decidedly, a mixture of Prus- sian blue and gamboge went brown, also a mixture of Prussian blue and burnt sienna and Prussian blue and raw sienna, Antwerp blue, Leitches blue and permanent blue all bleached. In a second experiment the permanent blue did not change. Indigo and Venetian red also bleached very slightly and became more red in colour, whereas car- mine, crimson lake, vermilion, Venetian red, Prussian blue, indigo, sepia, and brown pink underwent no change. Two tubes with rose madder were used, one of them bleached very slightly, the other not at all. XIII. This part of our Investigation has, at present, only Action of the reached its first stage, and will be more fully considered in products of the the second part of our report. The results already obuiined ^J'as^on*pig-^^ are, however, of interest, and may be briefly stated. An ments. ordinary gas jet, burning two cubic feet per hour, was kept burning day and night for three weeks in a cupboard 6 ft. 6 ins. long, 2 ft. 6 ins. wide, 5 ft. 6 ins. high. At the top of the cupboard was fastened a board on which strips of the following colours were pinned : Indian red, madder red, rose madder, carmine, crimson lake, Venetian red, scarlet lake, madder lake, vermilion, burnt sienna, Indian yellow, burnt umber, gamboge, Naples yellow, chrome yellow, cadmium yellow, aureolin, chromium oxide, emerald green, Prussian blue, cobalt, indigo, Paynes grey, Leitches blue, Antwerp blue, French blue, purple carmine, sepia, indigo and gamboge, indigo and Venetian red and yellow ochre ; Prussian blue and raw sienna, Prussian blue and Venetian red, Prussian blue and crimson lake. The temperature to which these colours were exposed was 82° Fahr. Of course no moisture deposited upon them, but the window to the cupboard was bedewed ; under these circumstances hardly any change occurred. Crimson lake was slightly bleached, madder lake became a little redder, and Antw^erp blue and Prussian blue a shade greener. The changes which pro- bably occur under slightly different conditions we shall treat of hereafter. 42 Experiments with body colours. Experiments with coloured glasses. XIV. We have already pointed out the action which must occur on diluting any colours with a solid white medium such as Chinese white. But few comparative experiments have yet been made. Mixtures of rose madder and Indian yellow, and of Prussian blue and gamboge, when mixed with Chinese white faded more rapidly than without it, and a mixture of Prussian blue and crimson lake are strikingly acted on, for without Chinese -white the mixture becomes blue, but if mixed with Chinese white, it becomes of a bright pink colour. The fading of Prussian blue appears to be brought about by the addition of Chinese white. The following mixtures exposed in a frame at a window receiving about half the day’s sun from May 19th, 1887, till January 18th, 1888, did not show any marked difference although mixed with Chinese white. Indigo and burnt sienna, vermilion and chrome yello-w, Antwerp blue and burnt sienna, purple madder and burnt sienna, rose madder and Indian yellow. The following single pigments were also tried but the addition of Chinese white did not appear to hasten their fading. The vermilion mixed with the Chinese white darkened a shade more than without it, and the fading of Prussian blue was increased by it. Carmine. Brown pink. Rose madder. Purple madder. Crimson lake. Violet carmine. V ermilion. Paynes grey. Gamboge. Antwerp blue. Indian yellow. Prussian blue. Vandyke brown. Indigo. Sepia. Burnt Sienna. Straps of the following colours with and without Chinese white were sealed up in tubes with moist air and exposed to all the sunlight there was from May 20th till July 29th 1887. Rose madder, Paynes grey, sepia, vermilion, violet, carmine, crimson lake and Prussian blue, indigo and burnt sienna, and Prussian blue and burnt sienna. The only case in which the addition of the Chinese white made a perceptible difference was in the mixture of Prussian blue and crimson lake. The change was the same as above described. XV. The following single and mixed colours chosen on ac- count of their instability (see Table II.) were exposed from May 1887 till January 1888, under such conditions as to receive about one half the skylight and half the possible amount of sunlight. The composition of the light trans- mitted through the red, green, and blue glasses used, has been already stated. Fig. VII. Taking first the action of the red light, only four single colours were acted upon but slightly by it, these vvere indigo, Leitches blue, crimson lake, and Prussian blue. With the mixed colours!, the only cases in which action occurred were those in which indigo was present they faded, for the indigo was destroyed. Under the green glass five single colours and two mixtures showed alteration, viz., Ueitches blue, Paynes grey, crimson lake, Vandyke brown, carmine, Prussian blue, and crimson lake, also Prussian blue with raw sienna. All these changes occurring under the red and green glasses were very slight. A reference to the diagrams which give the optical composition of the different colours will show that these particular colours absorb in the red and in the green, hence their fading. We come now to the blue glass, and it will be seen that the amount of fading in this case is nearly, though not quite so much, as under the white glass, the dilference being due to the opacity of blue glass even for blue rays. (See Fig. VII.) All the colours under the blue and white glass (burnt sienna excepted) were acted on. This clearly shows that it is the blue end of the spectrum which is active in producing the fading of colours. — White. Blue. Green. Red. Purple Madder Faded to 2 - Faded to 1 - — - Antwerp Blue - No experiment Faded - - - Leitches Blue - SI. faded SI. faded Darkened Darkened. Violet Carmine Faded to 1 - Faded to 1 - - Paynes Grey - Faded to 1 Bluer - Blue - Indigo - No experiment Faded to 1 — SI. faded. Prussian Blue No experiment SI. faded - V. si. faded. Rose Madder - (2 experiments.) SI. bleached - SI. faded — — Brown Pink - No experiment Faded to 3 - - Crimson Lake - No experiment Faded - SI. faded SI. faded. Vandyke Brown No experiment Faded to 1 SI. faded - Vermilion Darkened V. SI. darkened - ~ Carmine - No experiment Faded to 3 SI. faded - Gamboge No experiment Faded to 1 - _ Indian Yellow No experiment No change - - Sepia - - - Become lighter Become lighter - - Burnt Sienna No change - No change - — - Mixtures. Indigo and Burnt Sienna, 2 experi- ments. Indigo and Venetian Red. Indigo and Raw Sienna. Indigo and Vandyke Brown. Bleached to 2 Bleached to 1 — Bleached Red to 1 — No experiment Gone red - Lighter Lighter — Si. red. lading. Browner. 44 _ — ■White. Blue. Green. Red. Indigo and Indian Very si. faded _ _ Red. Indigo and fJamboge Gone blue Gone blue . - — — Prussian Blue and Crimson Lake. Gone bluish j^reen. Gone to a neu- tral tint. Pinkish blue - - Prussian Blue and Gamboge. Much gone, little blue in No 1. Same as under white. Prussian Blue and Burnt Sienna. Brown, Prus- sian blue quite gone in 1 and 2. Same as under white. Prussian Blue and No experiment faded to 1 - SI. faded — Raw Sienna. Antwerp Blue, Rose Become bluer Become bluer — — Madder, and Indian Yellow. Prussian Blue and Quite green - Green — — Purple Carmine. Antwerp Blue and Bxirnt Sienna. Burnt sienna only left. Burnt sienna left. — — Indian Yellow and Rose Madder. Bose madder left. Became pink — — Chrome Yellow and Blackened • SI. faded Vermilion. Colours mixed with Chinese White. Prussian Blue and No experiment faded to 2 - Raw Sienna. Rose Madder and Indian Yellow. No experiment Rose madder left. — — Antwerp Blue and Brown to 3 - Brown to 3 - — — Burnt Sienna. Indigo and Raw No experiment Bleached — — Siemia. Indigo and Burnt Sienna. Burnt Sienna left. SI. red — — Indigo No change - No. change - — — Antwerp Blue No experiment Bleached - - Prussian Blue No experiment Bleached - - Purple Madder Bleached Bleached - - Burnt Sienna No change - No change - - - Gamboge No experiment SI. bleached - - - Indian Yellow No experiment SI. bleached - - - Vandyke Brown No experiment Bleached - - Brown Pink - No experiment Bleached to 3 - - Crimson Lake No experiment Bleached to 3 - SI. faded. Carmine No experiment Bleached to 3 - - Vermilion Blackened Blackened under 1 and 2 - - Rose Madder SI. bleached - V. si. bleached — — Violet Carmine Bleached to No. 1 and darkened to 2 and 3. Same as under white glass. Paynes Grey Bleached to 1 Become bluer Become bluer — Sepia - Lighter Lighter - - I’russian Blue and Burnt Sienna. Prussian blue gone in 1 and 2. Same as under white glass. Prussian Blue and Crimson Lake. Became of bluish green. Became of a neutral tint. Pinkish blue - Note.— SI. means slightly ; V. si. means very slightly ; No. 1 is the faintest tint. 46 XVI. In addition to the severe tests, both with regard to Ej^ect on surrounding atmosphere and amount of light, to which we subjected the different colours, it was clearly of interest and light of aroom. importance to have similar specimens of colours subjected to milder treatment, and under conditions approximating to those to which pictures are usually subjected. We have therefore taken strips of our tinted papers and exposed them in a picture frame under a glass, so that one half of each paper was exposed to light, and the other half, bent back, was entirely shielded from the light. The back and glass were carefully pasted into the frame so as to exclude dust, as would be done with a picture, and the frame was then exposed in a room to very bright light, but not to direct sunlight. During a part of the time the frame was hung up against a window, during the remaining time it was at a little distance from the window. The frame was first ex- posed to the light on August 4th, 1886, and was op ned, and the colours examined on May 6th, 1888. The following colours were in the frame ; — Antwerp blue. Prussian blue. Leitches blue. Indigo. Gamboge. Brown pink. Indian bellow. Naples Yellow. Lemon yellow. Vandyke brown. Venetian red. Crimson lake. Vermilion. Rose madder. Carmine. Prussian blue and burnt sienna. Prussian blue and Indian red. Prussian blue and Vandyke brown. Prussian blue and g-amboo-e. Indigo and Vandyke brown. Indigo and burnt sienna. Indigo and gamboge. Indigo and Indian red. Rose madder and raw sienna. Antwerp blue and raw sienna. V ermilion and chrome, yellow. Of the single colours we found that the gamboge, indigo, and Naples yellow had slightly faded. Browm pink had faded perceptibly to 6. Carmine had bleached to 3 ; Vandyke brown had faded to 1, and was fainter to 4, and crimson lake had faded to 5, and all the darker shades had become paler. With mixtures, Prussian blue and burnt sienna had changed, the blue had faded ; with Prussian blue and Vandyke brown, and with indigo and Vandyke brown in both cases, the Vandyke brown had faded ; with Prussian blue and gamboge, the gamboge 46 Conchisions. had slightly gone in 1 ; with indigo and burnt sienna the indigo had gone in 1, and all the shades had become browner, indigo and gamboge had become paler throughout, both colours apparently fading ; with indigo and Indian red, the indigo had gone completely in 1, and in part in all the tints. The other colours, both single and mixed, had not changed. XVII. In a subsequent report we hope to be able to make further deductions as to the causes which operate in producing the fading of pigments, I:;ut we can summarise the conclusions which are clearly to be drawn from the results of the experiments which we describe in this first report. Mineral colours are far more stable than vegetable colours, and amongst those colours which have remained unaltered, or have only very slightly changed after an exposure to light of extreme severity, a good gamut is available to the water colour artist. The presence of moisture and oyxgen are in most cases essential for a change to be effected, even in the vegetable colours. The exclusion of moisture and of oxygen, particu- larly when the latter is in its active condition, as experiments to be described in our next report show, would give a much longer life, even to these, than they enjoy when freely exposed to the atmosphere of a room. It may be said that every pigment is pei’uianent Avhen exposed to light “ in vacuo,” and this indicates the direction in which experiments should be made for the preservation of water colour drawings. The effect of light on a mixture of colours which have no direct chemical action on one another is that the unstable colour disappeai’S and leaves the stable colour unaltered appreciably. Our experiments also show that the rays which produce by far the greatest change in a pigment are the blue and violet components of white light, and that these, for equal illumination, predominate in light from the sky, whilst they are less in sunlight and in diffused cloud light, and are present in comparatively small proportion in the artificial lights usually employed in lighting a room or gallery. The experiments have also shown that about a century of exposure would have to given to water colour drawings in galleries lighted as are those at South Kensington before any very marked deterioration would be visible in them, if painted with any but the more fugitive colours ; and that when the illumination is of the same quality as that of gaslight, or of the electric glow light rendered normally incandescent, and of the same intensity as that employed in those galleries, an exposure to be reckoned by thousands of years would have to be given to produce the same results. We have here not taken 47 into account the action, if any, which might arise from the products of combustion where gaslight is the illuminant, and which our experiments so far have shown to have but a trifling effect, nor of any modification of hue which might be due to change in the whiteness of the paper on which the paintings were made, but simply to the change in the colours themselves. When It is determined what is the minimum illumination In which a water colour drawing can be well seen, the length of time during which the drawing will retain almost its pristine freshness can be easily calculated. Since it Is the blue light which causes the fading, it might be thought that the glazing of skylights with a glass of a slightly yellow tint should be adopted. It must be recollected, however’, that in ordinary diffused sunlight this would entail an alteration in the brilliancy of the blues of a picture, and a change in their tone. It is well known to artists that a picture painted or illuminated by blue skylight looks colder than one Illuminated by diffused sunlight, whether the diffusion be caused by white blinds to the gallery or by cloud. The cause of this will be apparent on looking at Fig. VI. ; the red and yellow light are very deficient, whilst the blue light predominates. When the illumination is from the blue sky a yellow glaze or blind might be useful in imparting the warmer tone of diffused sunlight, and at the same time it would reduce in intensity the destructive component of the light. We h ave to thank the Committee of Artists, appointed by the Science and Art Department for suggestions which they individually and collectively have made to us during our investigations, and for the criticisms which they have made on various points In our report. To us, who could only follow the subject from a purely scientific point of view, the ripe experience of those who are masters of its technique has been of the greatest value. W. J. Eussell. W. DE W. Abney. 48 PAET III. Appendix I. ^'^thTinfemUy already referred to the measures made of the of light colours with which we experimented, and we now give the reflected from results of the measurements ma-le. For convenience of pigments. reference the curves of several ])igments have been put in one diagram, and as far as possible those of the same colour appear together. These diagrams we believe will be found of great utility, since they show the exact optical properties of the pigments we used, and will be a guide in future investigations for ascertaining whether the same material is being dealt with. They also indicate the amount of fading which has taken place in those pigments which have not absolutely bleached, or in which no appreciable change has taken place. For instance, a yellow pigment to the eye might be of approximately the same hue and luminosity as gamboge, but it is perfectly possible that the spectrum value of the two might be totally different. We have the same thing occurring in the mixture of spectrum colours. Thus, by mixing pure spectrum red with pure spectrum green, a yellow may be formed which to the eye may appear identical with the yellow found in the spectrum itself. Such a pigment might behave in the light quite differently to gamboge, either by being more permanent or more fugitive. Again, too, even the preparation of the same colour may vary. Thus cadmium yellow may be of a variety of tones, and of this variety some might behave somewhat differently in the light to that variety with which we have experimented. In a paper recently communicated to the Royal Society, General Festing and one of us have shown how from these curves the exact colour of the pigment, or the faded pigment can be reproduced from the spectrum, and thus will enable any one who possesses the necessary apparatus to follow the results which w^e have obtained. We have adopted in these curves the nomenclature of colours of the spectrum suggested by Professor Rood, as given in his “ Modern Chromatics.” Beneath each part of the curve the colour measured has been shown, and this may be of use for those who are not familiar with the Fraunhofer lines of the solar spectrum, the principal of which are indicated by B, C, D, &c. In diagrams 13 and 14 we have given the readings of a series of tints of two colours with a view of showing the reflection-capacity ” of varying depths of tint. In 49 diagram 13, No. 5 curve refers to No. 8 tint (see Fig. X.), and No. 1 curve to No. 4 tint. There were only three tints on the paper from which Diagram XIV. was made. In the following tables * before the colour signifies that the colour was unchanged after exposure to light ; t that the colour bleached after exposure to light ; ft that the alteration action was so small that it was not considered necessary to measure it. Indigo (Plate 1). TJnfaded. Faded. Wave Length. Intensity of Refleeted Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensitv of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 25-0 5,200 19-0 4,300 40-2 5,500 47-4 4,500 26-0 5,500 16-0 4,500 45-0 5,700 46-5 4,600 25-0 .5,700 14-5 4,600 46-7 6,000 45-5 4,700 24-0 6,000 14-0 4,800 48-2 6,200 45-8 4,800 22-7 6,200 13-6 4,900 48-5 6,500 48-0 5,000 20-7 6,900 13-5 5,100 48-2 6,700 50-0 5,300 47-8 6,900 52-0 Antwerp Blue (Plate 1). Unfaded. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,200 76-5 6,000 32-0 4,200 47-5 5,500 57-0 4,300 75-5 6,100 30-8 4,300 53-0 5,600 53-0 4,400 74-8 6,200 30-2 4,400 57-8 5,700 49-5 4,500 75 -0 6,400 28-7 4,500 61-5 5,800 46-4 4,600 76-5 6,600 27-5 4,600 64-5 5,900 43-4 4,700 78-0 6,900 26-2 4,700 66-3 6,000 40-0 4,800 77-0 4,800 67-8 6,100 37-5 4,900 73-2 4,825 68-0 6,200 35-5 5,000 69-5 4,900 66-0 6,300 34-0 5,200 62-0 5,000 63-8 6,400 33-0 5,400 53-5 5,100 63-0 6,500 32-3 5,600 45-5 5,200 64-0 6,600 32-0 5,800 38-0 5,300 63-2 6,900 32-0 5,900 34-6 5,400 61-0 A 54947. D 50 * Cobalt Blue (Plate 1). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,350 44-0 5,000 41-5 5,600 6-5 6,400 23-5 4,500 51-7 5,100 29-0 5,750 6-0 6,500 28-5 4,600 53-3 5,200 20-0 6,000 7-5 6,600 34-5 4,700 54-0 5,300 15-0 6,100 10-5 6,700 40-0 4,800 53-6 5,400 11-5 6,200 14-0 4,900 50-5 5,500 8-5 6,300 18'5 — French Ultramarine (Plate 1). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,350 38-0 4,900 27-0 5,500 8-0 6,500 7-5 4,600 38-0 5,000 21-5 5,750 7-0 6,620 8-5 4,750 35-0 5,200 12-5 6,300 6-8 7,000 6-5 Prussian Blue (Plate 2). Unfaded. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,200 60 • 5 5,800 34-5 4,200 64-0 5,500 45-5 4,500 67-0 6,000 30-0 4,500 71-0 5,800 37-0 4,600 68-0 6,300 25-5 4,600 74-5 6,000 33-0 4,800 64-0 6,500 23-5 4,650 74-75 6,300 29-0 5,000 59-0 6,800 22 0 4,800 71-0 6,500 27-5 5,200 49 • 5 7,000 21-0 5,000 65-5 6,800 26-0 5,500 41-75 5,250 54-5 7,000 25-0 51 t Pekmanent Blue (Plate 2). Wave Lengtii. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 57-0 4,900 44-0 .5,700 16-6 6,300 20-8 4,400 60-0 5,000 38-5 5,800 16-0 6,400 22*2 4,500 Gl-8 5,200 30-8 5,900 16-0 6,500 24-0 4,600 60-8 5,400 23-5 6,000 17-0 6,600 25-5 4,700 58-0 5,500 20-5 6,100 18-0 4,800 51 0 5,600 18-0 6,200 19-3 t French Blue (Plate 2). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,250 40-0 4,750 58-5 5,750 24-3 6,500 34-5 4,400 47-0 4,900 48-0 5,850 24-0 6,600 36-5 4,500 54-0 5,000 43-7 6,000 25-2 6,700 37-5 4,600 61-0 5,100 40-2 6,200 28-2 4,625 61-25 5,200 36-5 6,300 30-2 4,700 60-5 5,400 31-0 6,400 32-7 * Ctanin Blue (Plate 2). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 45-5 4,900 50-5 5,500 26-8 6,100 10-5 4,400 49-2 5,000 46-5 5,600 23-0 6,200 9-5 4,500 52-6 5,100 42-8 5,700 19-5 6,300 9-0 4,600 55-5 5,200 38-5 5,800 16-5 6,500 8-2 4,700 57-0 5,300 34-5 5,900 14-0 6,800 7-0 4,800 54-8 5,400 30-5 6,000 12-0 D 2 52 tt Emerald Green (Plate 3). Wava Lenat-h. Intensity 1 nf 1 Reflected 1 Light. 1 Wave Length. Intensity of Reflected Light, j "Wavp Length. Intensity of 1 Reflected j lAght. ! Wave ^ Length. Intensity of Reflected Livht. 4,200 25-0 5,000 72-5 5,700 39-7 6,400 9-0 4,300 25-75 5,125 75-0 5,800 32-0 6,500 8-0 4,500 29-0 5,200 73-5 5,900 20-0 6,600 7-6 4,600 34-5 5,300 69-5 6,000 20-5 6,750 7-0 4,700 49-0 5,400 63-0 6,100 16-5 7,000 6-5 4,800 60-0 5,.500 55-5 6,200 13-4 4,900 68-0 5,600 47-5 6,300 10-0 * Green Oxide of Chromium (Plate 3). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,250 16-0 5,100 35-5 5,600 35-0 6,300 26-0 4,400 17-6 5,200 41-5 5,700 22-0 6,400 28-0 4,600 20-2 5,300 44-6 5,750 20-5 6,500 29-3 4,800 22-6 5,310 45-0 5,850 19-8 6,600 30-8 4,900 26-0 5,400 43-5 6,000 20-5 6,700 31-4 5,000 30-5 5,500 40-5 6,200 24-0 6,850 33-0 * Terra Verte (Plate 3). Wave Length. Intensity of Reflected Liglit. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 19-6 4,800 24-0 5,400 20-5 6,000 14-0 4,400 20-0 5,000 25-0 5,500 18-7 6,200 13-4 4,500 21-0 5,100 24-7 5,600 17-0 6,300 13-0 4,600 22-4 5,200 23-8 5,700 15-8 6,500 13-0 4,700 23-4 5,300 22-2 5,800 14-8 53 Prussian Blue, Raw Sienna, and Burnt Sienna (P late 4). UlfFADED. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave length. Intensity of Reflected Light. Wave Length. Intensi ty of Reflected Light. 4,250 13-0 5,900 21-0 4,250 15-0 5,450 29-5 4,400 14'5 6,000 20-5 4,400 17-0 5,600 32-0 4,500 15-6 6,200 18-5 4,500 19-5 5,700 33-0 4,600 16-6 6,400 16-8 1 4,600 22-0 5,800 32-7 4,700 17-0 6,500 16-0 4,700 23-3 5,900 32-0 4,900 18-2 6,700 14-5 j 4,850 24-5 6,000 31-0 5,000 18-8 6,900 13-6 1 4,900 24-1 6,200 30-2 5,100 19-8 ' 4,950 24-0 6,300 30-0 5,200 21-0 i 5,000 24-3 6,400 28-6 5,300 22-0 5,100 25-3 6,500 27-2 5,450 22-7 5,200 27-0 6,700 24-5 5,600 22-8 5,300 29*0 6,800 23-5 5,800 21-5 5,400 29-0 Indigo and Gamboge (Plate 4). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave 1 Length. Intensity of Reflected Light. 4,300 12-2 4,800 30-8 5,400 34-0 6,000 24-2 4,400 15-5 4,900 32 -7 5,500 32-8 6,200 23-0 4,500 19-0 5,000 34-4 5,600 30-5 6,500 22-0 4,550 19-6 5,100 36-0 5,700 28-5 6,700 21-8 4,600 21-0 5,200 35 ’ 7 5,800 26-7 6,800 21-8 4,700 26-6 5,.300 35-2 5,900 25-3 54 t Antwerp Blue and Raw Sienna (Plate 4). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. "Wave Length, Intensity of Reflected Light. 4,300 9-5 4,900 16'0 5,500 17-4 6,100 10-3 4,500 12-3 5,000 17-0 5,600 15-8 6,200 9-7 4,600 12-3 5,100 18-0 5,700 14-7 6,500 8-3 4,650 12-7 5,200 18-3 5,800 13-3 6,700 7-8 4,700 13-3 5,250 18-5 5,900 12-2 6,900 7-3 4,800 14-8 5,400 18'0 6,000 Il-O Prussian Blue and Gamboge (Plate 4). Uneaded. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. A¥ave Length. Intensity of Reflected Light. AVave Length. Intensity of Reflected Light. 4,300 22-5 5,400 46-5 4,300 37-5 5,500 37-8 4,400 25-0 5,500 43-5 4,400 39-0 5,600 35-3 4,500 31-0 5,600 40-2 4,500 47-0 5,700 32-7 4,600 38-5 5,700 36-5 4,600 53 '6 5,800 30-4 4,650 40-5 5,800 32-5 4,700 51-4 5,900 28-0 4,700 43-4 5,900 29-5 4,800 50-2 6,000 26-3 4,800 48-8 6,000 27-5 4,900 50-0 6,200 25-0 4,900 51-7 6,100 26-2 5,000 49-5 6,500 24-6 5,000 53-0 6,200 25-3 5,100 47-0 6,850 24-4 5,100 52 ■ 7 6,400 24-8 5,200 45-0 5,200 51-3 6,850 23-0 5,300 42-5 5,300 49-0 5,400 40-2 55 ft Vermilion (Plate 5). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,350 6-75 5,500 11-5 5,900 59-0 6,600 90-5 4,800 6-75 5,600 15-0 6,000 78-0 6,750 84-0 5,000 7 ■ 5 5,750 31-5 6,200 97-0 7,000 72-5 5,300 9-5 5,800 40-0 6,500 94-5 Mercuric Iodide (Plate 5). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 3-0 5,000 3-7 5,800 29-5 6,400 74-5 4,500 2-0 5,200 5-5 5,900 54-5 6,500 75-4 4,600 4-0 5,400 6-8 6,000 66-5 6,600 75-8 4,750 5-8 5,500 8-0 6,100 70-0 6,750 76-3 4,900 4-2 5,700 18-0 6,200 71-7 7,000 76-3 4,950 3-5 5,750 22-0 6,300 73-5 t Carmine (Plate 5). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 37-0 5,200 31-5 5,700 41-0 6,400 75-0 4,500 38-5 5,300 33-0 5,800 50-0 6,500 76-0 4,550 39-0 5,380 34-5 5,900 59-0 6j600 75-0 4,700 36-5 5,450 33-5 6,000 67-0 6,800 72-0 4,800 34-5 5,500 33-0 6,100 70-5 5,000 32-0 5,550 33-25 6,200 72-5 5,100 3D0 5,600 34-0 6,300 74-0 56 * Indian Ked (Plate 5). Unfaded. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Liglit. Wave Length. Intensity of Reflected Light. 4,200 25-7 5,750 42-5 4,200 «-5 5,800 49-5 4,300 27-0 5,800 48-5 4,300 29 0 5,900 54-0 4,500 29-5 5,900 52-0 4,500 28-6 6,000 61-5 4,750 30-5 6,000 58-5 4,600 28-5 6,100 65-5 4,900 30-0 6,100 63-2 4,750 29-0 6,200 68-5 5,000 29-3 6,200 66-5 5,000 30-0 6,250 70-3 5,150 29-0 6,250 67-5 5,250 31-5 6,400 74-5 5,300 30-0 6,400 70-6 5,400 34-0 6,500 77-0 5,400 31-5 6,500 72-5 5,500 36-5 6,700 81-5 5,500 33-5 6,990 76-0 5,600 40-0 6,990 87-0 5,600 36-3 5,750 * 46-5 i t Crimson Lake (Plate 6). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,250 18-8 5,100 10'5 5,700 23-6 6,300 82-5 4,500 19-5 5,200 11-5 5,800 31-5 6,400 84-0 4,575 19-6 5,300 14-0 5,900 42-5 6,500 83-5 4,600 19-0 5,400 15-0 6,000 54-5 6,600 83-0 4,700 16-5 5,475 14-0 6,100 65-5 6,700 81-5 4,800 14-2 5,500 14'5 6,200 76-3 6,900 78-8 5,000 11-5 5,600 18'0 6,250 80-2 7,000 77-8 57 Kose Madder (Plate 6). Unpaded. Faded. Wave Length. Intensity of Reflected Lignt. I ! Wave 1 Length, j Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,200 30-0 5,600 22-5 4,200 45-5 5,400 34-0 4,300 31-0 5,700 32-5 4,300 45-5 5,500 38-0 4,400 32-0 5,800 49-0 4,400 45-0 5,600 42 ■ 5 4,500 31-0 5,900 58-0 4,500 45-1 5,700 48-0 4,600 27-0 6,000 65-0 4,600 46 -0 5,800 54-0 4,700 22-6 6,100 70-0 4,700 41-5 5,900 CO 4,800 20-5 6,200 73-0 4,800 37-3 6,000 63-0 4,900 19-8 6,300 77'0 4,800 34-2 6,100 67-5 5,000 18-5 6,400 78-6 4,950 34-0 6,200 71*5 5,100 18-0 6,500 79-6 5,000 32-7 6,300 74-2 5,250 17-2 6,700 80 0 5,100 32-0 6,400 75-0 5,400 18-2 5,200 33-0 6,500 73'8 5,500 19-0 1 5,300 33’ 1 6,700 69-4 t Scarlet Lake (Plate 6). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,250 36-5 5,000 25-5 5,700 41-0 6,300 81-5 4,400 34-0 5,150 25-0 5,800 50-5 6,400 80-6 4,500 32 5 5,200 25-2 5,900 62 0 6,500 80-0 4,600 31-7 5,300 26-0 6,000 76-0 6,800 77-0 4,700 30-5 5,500 29-5 6,100 82-0 4,800 28-2 5,600 34-0 6,200 82-0 * Venetian Eed (Plate 6). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Lengtli. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,200 22-5 5,000 30-5 5,700 58-0 6,200 87-5 4,300 25-0 5,200 31-5 5,750 61-0 6,300 90-5 4,400 27-7 5,300 32-5 5,800 65 '0 6,400 92-2 4,.’, 00 29-5 5,400 34-5 5,900 72-0 6,500 93-3 4,550 30-0 5,500 39-5 6,000 78-5 6,620 94-0 4,750 30 '0 5,600 47-0 6,100 83 '5 58 * Burnt Sienna (Plate 7). Wave Length. Intensity of Reflected Light. AVave Length. Intensity of Reflected Light. I AVave Length. Intensity of Reflected Light. AYave Length. Intensity of Reflected Light. 4,350 9-0 5,100 18-8 5,800 42-6 6,400 61-5 4,500 13-0 5,200 20-8 5,900 48-3 6,500 61-0 4,600 15-5 5,300 23-4 6,000 54-0 6,575 60-6 4,700 16-8 5,400 26-2 6,100 57-0 6,600 60-8 4,800 17-0 5,500 30-0 6,200 59-5 6,700 61-6 4,900 17-0 5,600 33-8 6,300 61-0 6,800 63-0 5,000 17-5 5,700 38-0 6,375 61-6 Brown Madder (Plate 7). Uhtaded. Faded. AYave Length. Intensity of Reflected Light. AYave Length. Intensity of Reflected Light. AYave Length. Intensity of Reflected Light. AYave Length. Intensity of Reflected Light. 4,350 33-6 5,600 46-0 4,350 64-5 5,600 74-9 4,500 34-3 5,700 47-5 4,500 66-0 5,700 74-5 4,700 38-0 5,800 49-4 4,600 69-0 5,800 74-0 4,800 38-8 5,900 61-8 4,700 75-0 5,900 76-0 4,900 38-0 6,000 54-5 4,775 79-5 6,000 81-2 5,000 37-5 6,200 59-0 4,800 79-0 6,100 87-0 5,150 37-3 6,300 60-5 4,900 75-3 6,200 89-6 5,300 38-5 6,500 63-0 5,000 74-0 6,300 89-0 5,400 40-4 6,700 64-4 5,200 74-5 6,500 90 '5 5,500 43-0 6,800 65-0 5,400 75-0 6,600 93-5 59 t Burnt Umber (Plate 7). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 34-5 5,000 45-6 5,600 62-0 6,200 75-8 4,450 36-0 5,100 48-0 5,700 66 ■ 5 6,300 76-8 4,650 36-0 5,200 50-2 5,800 70-5 6,500 79-0 4,700 37-3 5,300 52-0 5,900 72-0 6,600 80-6 4,800 40-0 5,400 54-0 6,000 ^ 73-4 4,900 42-8 5,500 57-5 6,100 74-7 t Purple Madder (Plate 8). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 15-2 5,000 16-0 5,700 24-0 6,400 49-5 4,400 16-2 5,100 16-3 5,800 26-5 6,500 51-5 4,450 17-2 5,200 17-2 5,900 30-7 6,600 53-0 4,500 16-7 5,300 18-4 6,000 35-0 6,700 54-0 4,600 15-5 5,400 18-8 6,100 39-5 6,900 54-0 4,750 15-6 5,500 20-0 6,200 43-5 4,900 15-8 5,600 22 '8 6,300 46-6 t Brown Pink (Plate 8). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,350 8-5 5,200 22-5 5,800 34-0 6,400 45-5 4,500 11-0 5,300 24-0 5,900 36-2 6,500 46-5 4,700 14-5 5,400 26-3 6,000 38-5 6,600 47-0 4,800 16-0 5,500 00 6,100 40-8 6,700 47-5 5,000 19-4 5,600 30-0 6,200 42-5 6,800 48-0 5,100 21-0 5,700 32-0 6,300 44-0 6,850 48-0 60 t Sepia (Plate 8). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of 1 Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 9-0 4,750 12-0 5,500 13-5 6,200 27-0 4,400 11-8 4,800 12-3 5,600 13-5 6,300 27-0 4,500 12-5 5,000 12-5 5,700 14-2 6,500 26-5 4,550 13-0 5,200 13-5 5,800 15-0 6,600 26-2 4,700 12-3 5,300 14-0 6,000 26-5 t Violet Carmine (Plate 8j. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. 1 Intensity of Reflected Light. Wave Length. Intensit.v of Reflected Light. 4,300 14-0 5,250 10*7 6,000 15-5 6,600 44-7 4,500 14-5 5,350 10-5 6,100 23-0 6,700 46-5 4,750 13-6 5,500 10-5 6,250 34-0 6,800 47-0 5,0C0 12-0 5,750 11-0 6,300 36-5 5,100 11-0 5,900 12-6 1 I 6,500 42-5 Crimson Lake AND A ntwerp Blue (Plate 9). Unfaded. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Ifght. Wave Length. Intensity of Reflected Iflght. 4,200 44-5 6,000 43-0 4,200 62-5 5,700 53-0 4,400 49-0 6,200 44-0 4,300 63-8 5,800 49-0 4,575 53-0 6,400 43-5 4,400 63-5 5,900 45-5 4,700 50-5 6,600 42-5 4,500 64-0 6,000 43-0 4,800 46-8 6,800 40-7 4,600 68-0 6,200 39-0 4,900 44-3 6,900 39-6 4,750 72-3 6,300 37-2 5,000 43-0 4,900 71-5 6,400 36-0 5,200 42-0 5,000 70-0 6,500 35-2 5,400 41-5 5,200 67-5 6,700 33-5 5,600 41-6 5,400 63-0 6,900 31-0 5,800 42-3 5,500 60-5 61 Rose Madder and Raw Sienna (Plate 9). Unpaded. Faded. Wave Length. Intensity of Reflected Light. w u ave Lerif?th. Intensity o*" Rellected Wave Length. Iiiteusiiy < of Relleetecl i Light. Wave Length. Intensity of Reflected Light. 4,300 29-0 5,500 35-5 4,300 29-5 5,400 35-5 4,400 29-5 5,550 36-0 4,400 30 -8 5,500 38-0 4, .500 31-0 5,600 37-0 4,500 32-5 5,600 45-0 4,600 33-0 5,700 50-0 4,575 33-5 5,700 54-0 4,700 32-2 5,800 63-0 4,700 31-6 5,800 63-0 4,800 31-5 5,900 81-0 4,800 30-5 5,900 68-0 4,900 29-2 6,000 76-0 4,900 30-0 6,000 71-5 5,000 26-5 6,100 79-5 5,000 28-0 6,100 74'2 5,100 25-0 6,200 82-0 5,075 26-0 6,200 76-0 5,200 29-5 6,300 83-6 5,100 26-4 6,400 78-4 5,300 33-5 6,500 85-0 5,200 3i -6 6,600 79 • 5 5,400 34-5 6,800 85'0 5,300 35-0 6,800 80-0 Indian Red and Indigo (Plate 10), Unfaded. Faded. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensify of Reflected Light. Wave Length. Intensity of Reflected Light. 4,200 28*5 5,900 34-5 4,250 37-5 5,600 54-6 4,500 33-5 6,000 34-6 4,400 41-5 5,700 56-0 4,600 35 '0 6,200 34-3 4,500 43 '6 5,900 58-3 4,700 36-0 6,300 33-6 4,700 45-5 6,000 59-0 4,750 35-5 6,500 33-0 4,900 47-0 6,200 60-7 4,900 34-2 6,800 33-0 5,000 47-5 6,400 62-3 5,000 33-4 5,200 49-2 6,500 63-0 5,200 33-0 5,300 50-0 6,700 64-6 5,500 33-5 5,400 51-2 6,900 66-4 5,700 33'9 5,500 53-0 62 Venetian Eed and Indigo (Plate 10). Umaded. Laded. Wave Length. Intensity of Refleeted Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 23-3 5,700 28-0 4,250 27-0 5,700 46*0 4,500 25 -5 5,800 28-0 4,300 28-0 5,800 49-5 4,700 27-8 6,000 28-3 4,500 31-5 5,900 52-0 4,800 27-5 6,200 29-3 4,700 34-2 6,000 53-0 4,900 27'0 6,300 30-0 4,800 35-0 6,200 54-5 5,000 26-8 6,400 31 -0 5,000 35'6 6,300 55-0 5,100 26-5 6,500 31-6 5,200 36-5 6,400 55 -6 5,200 26 -S 6,600 31-2 5,300 37-2 6,500 56-0 5,400 27-0 6,700 30-0 5,400 38-2 6,700 56-7 5,500 27-5 6,900 27-2 5,500 40-0 6,800 57-2 5,600 27-8 5,600 44-0 6,900 57-8 t Indian Yellow and Rose Madder (Plate 10). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Lengtii. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 11-0 5,200 11-5 5,800 41-0 6,500 73-5 4,500 12-6 5,300 12-6 5,900 49-5 6,600 74-0 4,700 13-8 5,400 14-8 6,000 57-0 6,800 73-0 4,800 12-8 5,500 17-8 6,200 68-0 4,900 11-6 5,600 23-2 6,300 71-4 5,050 10-8 5,700 31-8 6,400 72-6 63 t Gamboge (Plate 11). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Itensity of Reflected Light. 4,275 15-0 4,650 18-5 5,300 60-5 6,200 81-7 4,300 14-5 4,750 21-5 5,400 67-5 6,350 82-4 4,375 13-7 4,900 28-5 5,500 73-0 6,500 82-0 4,500 16-0 5,000 36 '.A 5,700 75'5 6,700 82-8 4,550 17-5 5,200 53-5 5,900 78-5 6,800 83-0 4,600 18-0 5,250 57-5 6,000 80-0 t Indian Yellow (Plate 11). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length, Intensity of Reflected Light. 4,250 18-5 4,900 54-5 5,600 77-7 6,300 89-0 4,300 21-5 5,000 58-5 6,700 79-3 6,400 9D5 4,400 27-0 5,100 62-5 5,800 80-3 6,500 93-5 4,500 32-5 5,200 66-0 5,900 81-0 6,600 95-0 4,600 38-5 5,300 69’6 6,000 82-6 6,700 96-3 4,700 43-8 5,400 72-6 6,100 84-3 6,900 98-0 4,800 49-0 5,500 75-4 6,200 86-5 t Cadmium Yelloav (Plate 11), Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Inten.sitv of Reflected Light. 4,200 21-5 5,000 42-0 5,600 88-5 6,200 92-0 4,400 22-5 5,100 49-5 5,700 91-5 6,300 89-5 4,500 23-5 5,200 57-5 5,800 93-5 6,400 86-7 4,700 28-0 5,300 65-0 5,900 94-6 6,500 84-0 4,800 32-0 5,400 73-0 6,000 95-0 6,750 77-0 4,900 37-0 5,500 80-5 6,100 94-0 64 * Yellow Ochre (Plate 11). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,375 15-5 4,900 25-5 5,500 67-5 6,200 78-0 4,500 21-5 5,000 29-3 5,600 75-5 6,300 77-5 4,550 21-6 5,100 34-0 5,750 79-5 6,400 77-5 4,600 21-0 5,200 40-5 5,800 80-0 6,550 77-5 4,750 2L7 5,300 49-0 5,900 79-5 4,800 23-0 5,400 59-0 6,000 79-0 * Chrome Yellow (Plate 12). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light, Wave Length. Intensity of Reflected Light. 4,350 30-0 5,000 68-5 5,600 94-5 6,400 91-5 4,500 30-7 5,100 76-5 5,750 89-5 6,500 9P0 4,600 32-5 5,200 80-5 5,870 87-0 6,700 89-5 4,700 37-5 5,300 86-5 6,000 89-5 7,000 87-7 4,7.50 44-5 5,400 91-5 6,100 91-0 4,800 51-5 5,500 95-0 6,200 92-0 4,900 61-5 5,550 96-0 6,300 92-0 t Naples Yellow (Plate 12). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 44-5 4,800 48-3 5,300 83-8 6,000 100-0 4,400 43-3 4,900 53-7 5,400 90-0 6,200 99-8 4,500 42-7 5,000 60‘0 5,500 94-0 6,500 97-3 4,600 43'0 5,100 67-5 5,600 96-5 6,800 92-0 4,700 45-0 5,200 76-0 5,700 98-2 6,850 90-5 65 t Aureolin (Plate 12). Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,250 15-0 4,800 25 -0 5,400 83-0 6,150 93-0 4,350 130 4,900 33-0 5,500 89-0 6,300 93-5 4,400 14-0 5,000 42-0 5,600 92-7 6,400 94-0 4,500 15-7 5,100 53-0 5,700 93 '5 6,500 94-6 4,600 16-2 5,200 65-0 5,800 94-0 6,620 96-0 4,700 190 5,300 75-5 6,000 93-5 * Five Shades of Raw Sienna (Plate 13). No . 1. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 10-0 5,000 25-5 5,600 58-0 6,200 72-4 4,400 14-5 5,100 28-5 5,700 64-0 6,500 73-2 4,500 16-5 5,200 32-0 5,800 67-5 7,000 73-5 4,700 19-0 5,300 37-0 5,900 69-5 4,900 23-0 5,500 50-0 6,000 71 -0 No. 2. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 19-5 4,900 34-0 5,500 61-5 6,200 80-5 4,400 24-0 5,000 36-0 5,600 68-5 6,500 81-0 4,500 27-3 5,100 39-0 5,700 73-2 7,000 81'3 4,600 29-2 5,200 42-5 5,800 77-0 4,700 30-5 5,300 47-5 5,900 78-5 4,800 32-0 5,400 53'0 6,000 79-5 A 54947. E GG Five Shades of Kaw Sienna— co7^^. No. 3. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 28M) 4;900 43-5 5,500 71-0 6,000 90-5 4,400 32-5 5,000 47-5 5,600 77-5 6,200 91-5 4,500 34-6 5,100 50-8 5,700 82-5 6,500 91-5 4,700 37-4 5,200 54-6 5,800 86-2 7,000 91-5 4,800 40-7 5,400 65-0 5,900 88-8 No. 4. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 35-0 5,000 57-0 5,600 87-5 6,000 94-5 4,500 40-5 5,200 65-5 5,700 90- 6 6,300 95-5 4,700 45*5 5,400 77-0 5,800 92'0 6,500 95-5 4,800 49-0 5,500 83-0 5,900 93-6 7,000 95 •5 No 5 . Wavf> Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Liglit. Wave Length. Intensity of Reflected Light. Wave Length . Iiiten.sity of Reflected Light. 4,300 39-5 5,000 61-5 5,600 93-2 6,200 98-0 4,; 00 44-5 5,200 72-3 5,700 95-0 7,000 98-0 4,700 50-0 5,400 85-0 5,900 97-3 4,300 53-2 5,500 90-0 6,000 98-0 G7 Three Shades op Prussian Blue and Chinese White (Plate 14). No. 1. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Lengtii. Intensity of Reflected Light. 4,300 49-0 4,700 51-0 .5,700 18-8 6,700 7-3 4,400 51-0 4,900 43-4 5,800 16-5 7,000 7-0 4,500 53-5 5,200 35 -0 6,000 12-8 4,550 54-0 5,400 28-3 6,400 8-4 4,600 53-5 5,600 22-0 6,500 8-0 No. 2. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Lengtli. Intensity cf Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 5U5 4,800 62-0 5,900 25-5 6,700 14-2 4,400 54-0 5,000 56-0 6,000 23-2 7,000 12-8 4,500 58-0 5,300 47 '0 6,400 17-0 4,600 62-0 5,500 40 '0 6,500 16-0 4,700 63-3 5,800 28-0 6,600 15-0 No 3. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. Wave Length. Intensity of Reflected Light. 4,300 61-0 5,000 71-6 5,900 41' 6 6,600 31 -O 4,400 65-0 5,200 66-8 6,000 39-0 6,800 29-7 4,600 71-5 5,400 61-0 6,100 36-5 7,000 29-0 4,700 73-5 5,600 53-5 6,200 35-0 4,800 74-0 5,700 49-0 6,400 32-4 4,900 73-5 5,800 45-3 6,500 31-6 68 The original curves were made on the scale given by the prisms used, but for convenience they have been converted into the normal wave-length scale. The tables which precede each diagram are the tables of the intensities of the ditferent colours of the spectrum reflected from the various pigments, the intensity of those reflected from a white surface being taken as 100. Thus if we turn to Plate I. and look at tlie curve for carmine, we find that in the red at wave-length 6,000 it reflects 67 of light ; that is, it reflects 67 °/q of the light which a white surface reflects. Again in Prussian blue at wave -length 4,600 it reflects 68 of light, which is equivalent to saying that it reflects 68 of that reflected from a white surface. It may here be well to point out that the true colour of the pigment can be ascertained by drawing a line parallel to the base of the curve and touching the lowest point in it. This abstracts from the curve of intensities all the white light reflected from the surface of the particles, leaving only the intensities of the rays which go to form the true colour. Curves of At page 18 has been given the curves of the light bluT ’ lass ’ transmitted through red, green, and blue glass at different parts of the spectrum. The following table gives the heights of the curves at any point of the spectrum. The light before its passage being taken as 100. Eed Glass . Green Glass . I Blue Glass . Scale 1 Number, j Intensity. Scale Number. Intensity. Scale Number. Intensity. 0-5 35 2 0-3 0-8 4-2 1 3.5 3 1-0 1-8 3-4 2 33-5 4 3-0 2-3 2-1 3 30 5 6-3 2-8 0-5 4 19 6 10-7 3-8 1-1 4-5 14 7*5 15-0 5-0 O'O 5 10 7 15-5 6-0 2-0 6 6 8 16-0 6-5 2-5 '1 3 9 14-0 7-0 1-0 7'S* 0 10 10 -.5 8-0 1-0 11 8-0 9-0 1-0 12 6’0 10-0 2-0 13 5-0 11-0 4-0 14 4-0 12-0 12-0 15 2-7 13-0 23-0 16 1-7 14-0 34-0 18 0-0 15-0 45-0 16-0 56-0 17-0 62-0 i 18-0 65-0 69 Appendix II. In reference to the calculations which have been given Calculations regarding the value of the intensities of the blue rays in of ihe intensity sunlight and skylight, the following results will be of of sunlight, ^c. interest. Experiment has shown that the white hot crater in the positive carbon pole of the electric light always emits the same quality of light with great constancy, and therefore is a very convenient standard of light to use, since the blue rays are present in great abundance in it. The light from this som’ce was by proper means reduced to the light of one candle, and at one foot distance from the photometer, the blue rays were found to have a value of 3‘627 units of an empyric standard. Sunlight on a surface held normally to the direction of the beam on the 21st May near midday, the sky being very clear, had a value of 19,900 of these same units, which is equivalent to saying that it had a value of 5,480 candles at 1 foot off the surface as measured by the blue rays. In text books it is said that sunlight has an illuminating power equal to about 5,600 candles at one foot distance from a surface, but it is not stated at what time of the day or year this value was obtained. As the proportion of blue rays in sunlight near midday in May is very nearly the same as in the standard light employed, it would appear that the text book value is almost identical with our estimate. The scale of units of work was obtained by taking em- pyrically that one candle (of the quality electric light) performed 3*627 units of work in 30 minutes. This makes sunlight of the quality of that measured near midday on the 21st May perform 19,900 units in the same time. The candle power in the above table was derived from the units of work performed during the whole of daylight. The following figure gives a measure of the candle power ofsun- of sunlight falling on a vertical surface in the same position light falling on as that in which the pigments were exposed on the 21st day of each month. The absorption by the atmosphere of these rays has been taken as *6 for each atmosphere traversed, an estimate which careful observations, carried on for a long period, shows to be the mean absorption value, one atmosphere being taken as the height traversed by a beam of light were the sun vertical. To prevent confusion in the figure the scale of the intensity has been multiplied by three, the unit being 100 ; hence at any time of day the candle power of the blue light measured from the curves can be obtained by dividing the ordinate of the 7u curve by three. Thus at 9 a.iu. oil the 21st of August the height of the curve is 7,860 ; this gives a cnndle power of sunlight at that day and hour of 2,620. Fig. XII. H0UR.AM. 5 6 7 B B 10 II ROON. ! 2 3 5 RM. The following table "gives the numerical results of the diagram. As before said these have been multiplied by three in the latter. Illuminating Value in Candles oe Sunlight falling ON A Vektical Subface facing 20° E. of N. ! 4 a.m. 5 a.ni. G a.m. 7 a.m. 8 a.m. 9 a.m. 10 a.m. 11 a.m. Noon. 1 p.ra. 2 p.m. 3 p.m. 4 p.m. Jan. 21 0 260 1,090 1,530 1,720 1,270 720 130 0 Feb. 21 0 280 1,390 2,300 2,820 2,740 2,340 1,420 631 59 Mai-. 21 0 213 1,350 2,350 3,100 3,320 3,240 2,630 1,740 910 150 Apr. 21 0 95 795 1,790 2,620 3,180 3,340 3,100 2,460 1,590 650 0 May 21 0 182 860 1,700 2,410 2,890 3,010 2,690 2,100 1,330 320 0 June 21 0 160 770 1,620 2,410 2,730 2,800 2,510 2,090 1,060 0 0 July 21 0 182 860 1,700 2,410 2,890 3,010 2,690 2,100 1,330 320 0 Aupt. 21 0 95 795 1,790 2,620 3,180 3,340 3,100 2,460 1,590 650 0 Sept. 21 0 213 1,350 2,350 3,100 3,320 3,240 2,630 1,740 910 150 Oct. 21 0 2S0 1,390 2,300 2,820 2,740 2,340 1,420 631 69 Nov. 21 0 260 1,090 1,530 1,720 1,270 720 130 0 Deo. 21 0 372 930 1,260 1,100 550 23 0 71 The amount of loss of illumination due to the surface being vertical and not normal to the direction of the beam of light will be shown by taking as an example August 21st. — 5 a.m. 6 a.m. 7 a.m. 8 a.m. 9 a.m. 10 a.m. 11 a.m. Aug. 21 180 1,550 3,000 4,040 4,640 5,180 5,460 Noon. 1 p.m. 2 p.m. 3 p.m. 4 p.m. 5 p.m. 6 p.m. 7 p.m. 5,480 5,460 5,180 4,640 4,040 3.000 1,5.50 180 The above tables are close approximations to the true values. It should be stated that the sun’s declination has also been taken as equal on the 21st of January and November, February and October, March and September, April and October, and May and November. This is not quite exact, but will make no practical difference in the results. The total units of work which sunshine would perform during the whole day, supposing the sky cloudless, is found I2C no 100 ao 80 70 CO so 40 3a 20 10 by taking the areas of these curves. Fig. XIII. shows the work in the units we have adopted, but multiplied by three 72 Value of the light in a gallery at SouthKensing- ton Museum. for every portion of the year, the units being the height of the curves. It will be seen that from the beginning of March to the end of September the work performed is very much greater than in the remaining portions of the year. Appendix III. The following table gives the mean light expressed in candle power at a distance of one foot from the object and measured by the blue rays falling on the wall of one of the South Kensington galleries. The values are deduced from measurements made during the whole day on various days in March, April, May, and June. The various galleries in the Museum differ very slightly in the quantity of light which falls on the walls. Value of Light in the Gtallery. Date. Mean Candle Illumina- tion. Units of Work during thewholeDay. Date. Mean Candle Illumina- tion. Units of Work during the wholeDaj'. March 23 3-57 364 May 1 21-4 2,640 24 6-25 605 2 18-0 2,218 25 9-53 965* 3 28-5 3,502 26 16-00 1,625 4 26-8 3,388 27 8-4 859 5 27-8 4,560 28 13-4 1,415 6 3-0 365* 29 13-0 1,396 15 20-0 2,628 30 4-0 420 16 22-5 2,925 31 10-3 1,121 17 36-2 4,740 April 1 4-4 480* 18 28-5 3,600 2 14-2 1,912 26 9-8 1,260 3 3-8 1,500 27 4-1 564* 4 4-8 523 28 10-8 1,400 5 8-4 921 29 16-0 2,180 6 8-6 938 30 26-3 3,600 7 10-6 1,167 31 19-2 2,626 8 3-7 430* June 1 16-0 2,220 9 15-7 1,708 2 17-4 2,420 10 10-5 1,153 3 8-6 1,200* 11 15-8 1,724 4 13-1 1,813 19 12-8 1,498 5 2-2 3,160 20 28-8 3,350 6 9-8 1,230 21 31-2 3,660 7 24-0 3,460 22 1-2 136* 8 23-4 3,300 26 26-5 3,230 10 27 36-8 4,200 11 ^ 11-7 3,300 28 14-1 1,789 12 28-0 4,200 29 4-0 487* 30 10-0 1,334 Those dates marked with (*) are Sundays, when the blinds in the galleries are closely drawn. 73 Appendix IV. The Science and Art Department invited some of the most distinguished artists who used water colours to furnish a list of the colours which they employed : 46 responses were received to the invitation, and the following is a tabulated list of the colours used. For convenience, a distinguishing number has been allotted to each of the artists from whom a list of colours has been received, and the colours used by each artist are indicated by means of this distinguishing number. LIST OF COLOURS INDICATmO THE NUMBER OF ARTISTS BY WHOM THEY ARE USED. Number of Artists by whom used. Colours. Distinguishing Number of Artist. BLACK. 1 Aspbaltum ... 27. 2 Black 13, 26. 3 Blue Black - - - 5, 20, 27. 1 Charcoal 17. 1 Graphito - - - 27. 11 Ivory Black 8, 12, 14, 27, 28, 30, 33, 38, 40, 42, 46, 14 Lamp Black - - - 1, 3, 4, 7, 10, 12, 18, 21, 23, 25, 31, 36, 39, 43. 1 Neutral Tint 42. 1 Persian Black 34. 1 Eoman Sepia 39. 30 Sepia - 1, 2, 3, 5, 6, 7, 10, 11, 12, 13, 15, 17, 23, 24, 25, 26, 27, 28, 30, 31, 34, 35, 37, 38, 39, 40, 42, 43, 44, 45. 1 Warm Sepia 32. BLUE. 6 Antwerp Blue 2, 4, 12, 18, 30, 38. 9 Azure Blue (or Ceruleum) 10, 16,22,23,28, 31, 39 (Newman’s) 40, 41. Ceruleum (or Azure Blue) [See above.] A 54947. 74 Number of Artists by whom used. Colours. Distinguishing Number of Artist. 45 BLUE — continued. Cobalt - - - - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1, 12, 13, 14, 8 Cyanine Blue 15, 16, 17, 18, 20, 21, 22, 23, 24, 25, 26,27,28,29,30, 31, 32,33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46. 7. 11, 22, 23, 27, 29, 33, 39. 5 French Blue - - - 5, 9, 12, 16, 31. 18 French Ultramarine [See under Ultramarine.] 24 Indigo - - - 3,4,6,10, 11, 12, 14, 16, 17,19,23, 2 Leicht’s Blue 24, 25, 30, 31, 34, 37, 39, 41, 42, 43, 44, 45, 46. 24, 25. 16 Prussian Blue 4,7, 8, 10, 12, 13, 14, 17, 21, 28,32, 36, 2 Smalt 39, 42, 43, 44. 15, 21. 15 Ultramarine (Eeal) 1, 7, 12, 14, 17, 18, 21, 27, 28, 35, 37, 18 „ (French) Ultramarine Ash - 39, 40, 42, 46. 3, 6, 7, 14, 15, 18, 20, 25, 28, 29, 30, 33, 34, 36, 38, 39, 40, 46. 2, 3, 4, 7, 12, 14, 15, 21, 22, 27,30, 33, 1 „ „ (Blue) - 34, 35, 40, 43, 45, 46. 11. 1 „ „ (Grey) - 11. 2 BEOWN. Bistre _ - . 34, 39. [See under Madders.] 33 Brown Madder 3 ,, Ochre [See under Ochres.] 41 Burnt Sienna [See under Siennas.] 18 „ Umber [See under Umbers.] 2 Cologne Earth 7, 12. 38 Eaw Sienna [See under Siennas.] 26 „ Umber - - - [See under Umbers.] 2 Turner Brown 11, 27. 33 Vandyke Brown 1, 2, 3, 4, 5, 7, 8, 9, 10, 12, 13, 15, 16, 5 GEEEN. Emerald Oxide of Chro- 17, 18, 20, 22, 23, 25, 26, 30, 31, 32, 35, 36, 38, 39, 40, 42, 43, 44, 45, 46. 3, 4, 15, 21, 32. 13 mium. Emerald Green - 5, 6, 7, 10, 12, 22, 24, 30, 33, 36, 38, 3S , 7 Oxide of Chromium 46. 3, 6, 15, 20, 30, 33, 38. 6 „ „ (Trans- 7, 23, 27, 36, 40, 46. 4 parent.) „ „ (Opaque) 7, 27, 36, 40. 5 Olive Green 6, 12, 29, 31, 32. 11 3 Terre Verte 7, 11, 22, 23, 30, 34, 35, 36, 40, 44, 46. Veronese Green 26,31, 40. 4 Cobalt Green 23, 26, 33, 34. 1 Transparent Green - 33. 5 Viridian - - - 10, 1.5, 20, 40, 41. 1 Miscellaneous kinds 40. Number of Artists by whom used. 75 Colours. GREY. Distinguishing Number of Artist. 3 3 1 G 4 10 1 1 9 34 1 5 2 3 1 3 9 3 34 1 1 4 12 31 4 Charcoal Grey - Pay lie’s Grey - Ultramarine Ash (Grey) 3, 14, 16. 34, 45, 46. [See under Blue.] RED. Brown Pink Carmine Crimson Lake Deep Madder - Deep Rose Indian Red Light Red Madder Carmine ,, Lake - Mars Red Orange Vermilion Permanent Crimson „ Scarlet Pink Madder Purple Lake - Rose Madder Scarlet ,, Lake ,, Vermilion Venetian Rod - Vermilion (Extract of) - 12, 19, 37, 39, 42, 45. - 9, 16, 32,39. - 13, 16, 24, 25, 31, 37, 39, 41, 42, 45. [See under Madders.] - 34. - 7, 12, 16, 29, 31,34, 35,38,39. - 1, 2, 3, 5, 6, 7, 9, 10, 1.5, 16, 17, 18, 19, 20, 21, 23, 24, 2.5, 26, 27, 28, 29, 32, 34, 35, 36, 38,39,40,42, 43, 44, 45,46. [See under Madders.] >> f> - 7, 33. - 24, 31, 46. - 23. - 9, 23, 37. [See under Madders.] - 8, 12, 24. [See under Madders.] - 9. - 12. - 5,11,12,45. - 4,7, 12, 13, 14,22,30,31, 33, 34, 40, 46. - 1, 2, 3, 4, 7, 8, 10, 12, 13, 14, 15. 17, 19, 20, 21, 25, 26, 27, 29, 30, 31, 31, 35, 36, 38, 39, 40, 41, 42, 43, 44. - 15, 18 (Fields), 22 (Fields), 28. YELLOW. 18 Aureolin 1 Burnt Roman Ochre 23 Cadmium 7 „ (p.ale) 8 „ (deep) - 1 „ Orange 4 Cdirome !2 Gamboge 1 Golden Ochre 11 Indian Yellow - 3, 7, 10, 11, 19, 20, 22,25, 27,29,30,31, 32,33,34,40,43,44. [See under Ochres.] - 2, 3, 4, 7, 9, 11, 12, 1.5, 18, 20, 22,26, 28, 30, 31, 33, 37, 38, 39, 40, 42, 44, 46. 3, 23, 25, 26, 27, 34, 43. - 3, 11, 23, 26, 27, 34, 36, 43. - 24. - 8,16,39,41. - 2,4, 6,7,9, 10, 12, 14, 15, 16, 17,21,22, 23, 24, 25, 28, 29, 30, 31, 32, 35, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46. [See under Ochres.] - 5, 10, 13, 16, 23, 31, 32, 34, 39, 44, 46. 4 54947. G 76 Numlierof Artists by whom used. Colours. Distinguishing Number of Artist. 22 YELLOW — continued. Lemon (or permanent) 2, 3, 5, 7, 9, 11, 15, 20, 21, 22, 23, 25, 26, 2 Yellow. Mars Orange 27, 30, 31, 33, 35, 36, 39 (Lemon and Newman’s Permanent), 44, 46. 32, 33. 1 „ Yellow - 27. 1 Matrise Yellow -. 28. 2 Naples Yellow - 16, 30. 1 Orange Chrome - 5. 1 11 „ Ochre - Roman Ochre > [See under Ochres.) 42 5 Yellow Ochre - „ Madder - J [See under Madders.] 1 „ Mars - 14. 1 VIOLET. Maddox Brown’s Violet - 30. 12 WHITE. Chinese White 3, 7, 10, 11, 12,14,22, 27,29, 30, 31, 35. 1 Windsor and Newton’s 37. 33 White. MADDERS. Brown Madder - 2, 4, 5, 6, 7, 10, 12, 14, 15, 17, 18, 20, 21, 1 Carmine Madder 23, 24. 25, 27, 28, 29, 31, 32, 33, 34, 35, 36, 39, 40, 41, 42, 43, 44, 45, 4 6. [See Madder Carmine.] 1 Deep Madder - 9. 1 Madder Carmine 43. 5 „ Lake 7,21, 23, 33, 39. 9 Pink Madder - 7, 9, 10, 12, 17, 23, 30, 35, 46. 13 Purple „ 8, 11, 17, 25, 27, 29, 32, 33, 34, 39, 42, 4 Bed - 43, 45, 46. 1, 17, 27, 28. 34 Rose „ - - 2, 4, 5, 6, 7, 8, 9, 11, 12, 14, 15, 16, 18, 4 Rubens „ - 19, 20, 22, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 40, 42, 43, 44, 45. 11, 23, 26, 40. 5 Yellow - 10, 19, 26,29, 32. 1 Madders (not otherwise 3. defined). 77 Nnmber of Artists by whom used. Colours. Distinguishing Number of Artist. OCHEES. 3 Brown Ochre 3, 27, 34. 1 Burnt Eoman Ochre 17. 1 Orange Ochre 33. 1 Golden Ochre 45. 11 Eoman Ochre 7, 12, 15, 17, 24, 25, 28, 29, 34, 38, 45. 42 Yellow Ochre 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 45, 46. 1 The Ochres (not otherwise defined). SIENNAS. 1. 41 Burnt Sienna 3, 4,5, 6, 7,9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44, 45, 46. 38 Baw Sienna 2, 3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 80, 32, 34, 35, 38, 40, 41, 42, 43, 44, 45, 46. 1 The Siennas (not otherwise defined). UMEEES. 1. 18 Burnt Umber 3, 7, 12, 13, 14, 15, 18, 21, 22, 24, 25, 27, 33, 34, 38, 39, 40, 43. 26 Eaw Umber 3, 4, 6, 7, 8, 12, 13, 15, 17, 20, 21, 23, 24, 25, 27, 28, 29, 30, 32, 33, 34, 38, 40, 42, 43, 45. 2 The Umbers (not other- wise defined). 1, 16. -I 1 1 \ r— \ 1 1 1 ^ 1 1 1 1 W White -5 Pq . ?■ c ■t rs s £ eS u •'P»}f--a&iLt}JQ % fn'^a DA^CER^ELD Lith 22 ., Bedford Covent Garden 16478,6 ^V]atjt 1=1 0) .3 Pq i s ^ ^ s; .2 I ^ ^ S ri? ^ 2^3 f^' c? Si PU .V I I I fi ?c ^ llViC) I I \>n''n DS42S/6 3S^' Plate Yll. £: . 5 i: cq pq cq < c4 c^ OANCLRriELD LiT H 22. B E D ^0 R D S^ CoV ENT Ca R D E S Plate VIII I S' cq ri: ■ K ss; e ■§ •2 S' k!^ h-* ^■i D(»,Ncs«FfeLo L't-. i?j?, Bro Plate IX. Dancerfislo- Lith, 22, Bcdford ST Covent Carden . W 2 S / e.eeJ Plate X. 1 .§ 'I ft?, 1 ^'U. ERFiri.D LiTH [jCDfOHD Co V EN T G AT' O L '- ■v-5. Plate XT. DANCERFtELD. L»TH- 22, BedfoRD S’!" CoVENT CaRDEN, 164-28(6. dS) 1 ) ■ ' - r* :■ ■ ■J - ! - Plate Xn. V .■ 2 ^ 5 . 2 : S'J C'' 1 ^ ^ ^ ^ ^ ^ ^ ^ 1 1 ^ |/6W l:VhUc s (U Oanccrfilild Lith.22, Bedtord Covent Garden. i 64 S > f :(683) - ''i - u Dancerfueld.Lith Z ?.. Bedford S'^ Covert Garden is 42< iie . 2 1 -1 GETTY RESEARCH INSTITUTE 3 3125 01212 5965 R E P O K T TO THE SCIENCE AND ART DEPARTMENT OF THE COMMITTEE OF COUNCIL ON EDUCATION ON THE V ACTION OF LIGHT OH WATER COLOURS. to of parliament Commantr of 1i|et ^siei^t];. LONDON: PRINTED FOR HER MAJESTY’S STATIONERY OFFICE, BY EYRE AND SPOTTISWOODE, PRINTERS TO THE QUEEN’s MOST EXCELLENT MAJESTT. And to be purchased, either directly or through any Bookseller, from liYRE AND SPOTTISWOODE, East Harding Street, Fleet Street, E.C., aud 32, Abingdon Street, Westminster, S.W. ; or ADAM AND CHARLES BLACK, 6, North Bridge, Edinburgh; or HODGES, FIGGIS, & Co., 104, Graiton Street, Dublin. 1888.