CIHM ICIMH Microfiche Collection de Series microfiches ({Monographs) (monographles) Canadian institute for Historical iMicrorsproductions / Institut Canadian da microraproductions historiquas 1994 Tachnical and Biblioflraphic NotM / Nottt ttchniquM et bibliographiqiMs The Institute has attemptad to obtain tha bast original copy availabia for filming. Faaturas of this copy which may ba bIMiographically uniqua, which may altar any of tha imagas in tha raproduction, or which may significantly changa tha usual mathod of filming, ara chacfcad balow. □ Coloured covers/ Couverture da r juiaur a Covers damaged/ Couverture andommagia □ Covers restored and/or laminated/ Couverture restauria et/ou pallicuMe n n Cover title missing/ Lc titra da couverture manque Coloured maps/ Cartas giographiques en oouleur Coloured ink (i.e. other than blue or Mack)/ Encre de couleur (i.e. autre que bleuc ou noire) Colot' -d plates and/or illustrations/ it/ou illustrations en couleur Bourn .h other meterial/ Relie avec d'autres documents Tight binding may cause shadows or distortion along interior margin/ La reliure serrie peut causer de I'ombre ou de la distorsion le long de la marge interieure D n Blank leaves added during restoration may appear within the text. Whenever possible, these have been omitted from filming/ II se peut que certaincs pages blanches ajouties tors d'une restauration apparaissent dans le texta, mais, lorsque cela ctait possible, ces pages n'ont pas M filmies. Additional commenO:/ Commcnuircs iupplimentaires: L'Institut a microfilm^ le meilleur exemplaire qu'il lui a M pouible de se procurer. Les details de cet exemplaire qui sont peut4tre uniques du point de vue bibliographique, qui peuvent modifier une image reproduite. ou qui peuvent exiger une modification dans la mithode normale de f ihnage sont indiques ci-dessous. □ Coloured pages/ Pages de couleur □ Pages damaged/ Pages endommagies □ Pages restored and/or laminated/ Pages restauries et/ou pdlicultes I /I Page* discoloured, stained or foxed/ LlJ Pages dicolories. tachetto ou piquMs □ Pages detached/ Pages ditachies rTj Showthrough/ Ll_J Transparef«ce □ Quality of print varies/ Quallte inigale de I'impression □ Continuous pagination/ Pagination continue □ Includes index(es)/ Comprend un (des) index Title on header taken from:/ Le titre de I'en-tlte provient: □ Title page of issue/ Page de titre de la livraison □ Caption of issue/ Titre de depart de la livraison D Masthead/ Generique (periodiqucs) de la livraison This item is filmed at the reduction ratio cheeked below/ Ce document est filme su taux de riductton indiqua ci-dessous. 'OX UX 18X n ; 12X 1CX 20X 22X 26X 30X 24X 28X D 22 1 The copy filmed har* hai bean reproduced thanks to the generosity of: National Library of Canada L'axemplaire filmt fut reproduit grice A la gtnArosit* de: BibliothAque nationale du Canada The images appearing here are the best quality possible considering the condition and legibility of the original copy and in keeping with the filming contract specifications. Las images suivantes ont M reproduites avac ie plus grand soin, compte tenu de la condition at de la nettet* de i'exempiaire fllmA. et en conformity avac lea conditions du contrat de fllmage. Original copies In printed paper covers are filmed beginning with the front cover and ending on the last page with a printed or illustrated impros- sion. or the back cover when appropriate. All other original copies are filmed beginning on the first page with a printed or illustrated impres- sion, and ending on the last page with a printed or illustrated Impression. Les exemplaires originaux dont la couverture en papier est Imprlmte sont fllmte en commenpant par Ie premier plat et en terminant soit par la derniire page qui comporte une empreinte d'impresslon ou d'lliustration, soit par Ie second plat, salon la cas. Tous les autres exemplaires originaux sont filmte en commenpant par la premlAre page qui comporte une empreinte d'impresslon ou d'iliustration et en terminant par la derniire page qui comporte une telle empreinte. The last recorded frame on each microfiche shall contain the symbol — »• (meaning "CON- TINUED"), or the symbol ▼ (meaning "END"), whichever applies. Maps, plates, charts, etc., may be filmed at different reduction ratios. Those too large to be entirely included in one exposure are filmed beginning in the upper left hand corner, left to right and top to bottom, as many frames as required. The following diagrams illustrate the method: Un des symboles sulvants apparaitra sur la derniire image de cheque microfiche, selon Ie cas: Ie symbols — ► signlfle "A SUIVRE". Ie symbole ▼ signlfle "FIN". Les cartes, planches, tableaux, etc.. peuvent Atre fllmte A des taux de rMuction dlffirents. Lorsque Ie document est trop grand pour Atre reproduit en un seul clich*. 11 est fiimA A partir de Tangle supArleur gauche, de gauche A droite. et de haut en bas, en prenant Ie nombre d'Images nteessaire. Les diagrammes sulvants illustrant la mAthode. 1 2 3 1 2 3 4 5 6 MICROCOPY RESOLUTION TEST CHART (ANSI and ISO TEST CHART No. 2) 1.0 Im ■ 2.8 msD MBI §Sm M22 13^ IK JUm 1& b. ., ■tuu M25 wmm M 2.2 2.0 1.8 1.6 ^ /APPLIED IM/^GE Inc 1653 East Main Street Rochester. New York 14609 USA (716) 482 - 0300 - Phone (716) 288 - 5989 - Fox Color Values in ^ Monochrome r ^ t^ A LECTURE ON Orthochromatic Photography Delivered before the Canadian Institute February J6th, J90J, by J JOHN S. PLASKETT, B.A., Department of Physics, University of Toronto /j TORONTO, Canada With the Compliments oj the Author, Color Values in Monochrome by Photography It is a matter of common ( vperience among all photographers, that the print from a negative taken on an ordinary plate Joes not represent, even approximately, colors in their true values. Everyone has noticed, for instance, that in general, in photographs, blue, a comparatively dark color, is repre- sented by a tone several shades lighter than that produced by the much brighter color, yellow. We have grown so accustomed to this defect of photography that a photo- graphic print which correctly represents the color values of the object, looks strange and, at first glance, untrue. It is, however, not a matter of common knowledge that this de- fect of photography can be overcome, and that colors can be represented in their true values, or, what is a better term, luminosi- ties in a monochromatic print. Lisxht and ^" ^^® ^^^^ place, it is neces- ^ < sary to investigate tJie nature of color and the nature of the effects produced by it, or rather by what causes it, on the photographic plate. It seems hardly necessary to state that color, as a purely physical process, —3— mm^^^'w :m^^^f does not exist, but that it is a physiological eflFect peculiar to the eye; that, without the eye, there would be neither liglit nor color, but merely an exceedingly rapid and, at the same time very minute vibration or quivering- of that hypothetical medium, filling all space, which is called the ether. It is this vibration, acting on the retina, that is the primary cause of the sense of sight ; while the vari- ous colors tnat we see are produced by dif- ferences in the magnitudes and, at the same time, in the frequency of these vibrations ; the ether waves that produce the red and orange colors being longer than those that give rise to violet and blue. Roughly speak- ing the sensation of red is caused by waves in the ether about a-.'.oTF inch long, green by waves TTirioo inch lor ,', and violet by waves ffoiTolJ inch long. Similarly the effects on a photographic plate are due, not to any par- ticular co'^'- or colors, but to ether vibrations of certain definite wave lengths, some of these producing, when acting on the retina the sensation of color, while others, which strongly affect the plate, have no action on the eye. If throughout this paper, therefore, certain colors are spoken of as acting on the plate, it is to be understood as referring to the particular ether vibrations that excite those color sensations. Color There are three constants gen- Constants 7''^ """if '" "^fT^ '°'°'" = runty, Hue and Lummosity. —4— Tv.aK^«inMM^iBiii^3i&ii»'«aiiir7v»tt*"^^'«'%'"rity ot a color refers to its freedom from admixture with any oilier color and is always judjfed with reference to the spectrum, whose colors, when it is properly formed, are absolutely pure. A spectrum, appioximately pure, can be formed by project inj^, by means of the lantern, an image of a narrow slit in a piece of metal in the slide carrier on the screen, and interposing^ in the path of the rays a prism or prisms. The narrow band of white hg;t is broken up or decomposed by the The prisms into a broad, brilliantly- colored band whose appearance spectrum is familiar to all. The general arrangement and spacing of the colors can be seen by reference to the upper, left-hand corner of Fig. i, pages 24-25, which is a dia- gram of the particular spectrum used. The lines drawn across the diagram and lettered are the principal absorption lines of the solar spectrum, and answer the purpose of indicat- ing, when their positions are known, the wave length and color of the light at any part of tiie spectrum. The lengths of the ether waves decrease as you go from the line A to the line H or from red to violet, and any par- ticular color can be exactly defined by giving the length of the wave producing it. Purity of color, then, is due to light of, approxi- mately, a single wave length, represented of course by a very narrow band in the spec- trum, and not to the superposition of waves of different lengths which would produce an —5- !■• ^ , itnpure color. All natural colors Natural ' „ .. are, |j;enerally speaking, more VrfOiOrS or less impure, and are produced by the mi^i e of pure colors ; but all colors appear pure tO the eye which has not the power of esthnating, unaided, the purity of a color, or of resolving impure or mixed colors into their constituents. A color wliich, to the eye, appears pure, and as nearly as possible equal to the spectrum color, may be com- posed of nearly all the spectrum colors. This can be shown by interposing colored glasses in the path of the beam of light pro- * « , ducing the spectrum. The Absorption ^j^^^^ .^ ^^ ^^^ ^^^ ^^ absorb, opCCtra partially or wholly, some of ihe colors, allowing the others to be fully transmitted, and this is illustrated in black and white by the diagrams in Fig. 2. The rectangular figures are intended to represent the spectrum of white light, while the shaded portion represents the part transmitted by the particular glass considered, and the unshaded part what is absorbed. The .esultant color of the glass by transmitted light is evidently that produced by the union of the spectral colors transmitted. The hue of a color is the name given to wh r OBANtiE GLASS ^' GREFN CI. A;,; BLUE CLASS METHYL VIOLET ■^' .. .i .r'':;A I AURANTIA METHYL VIOLET AND AURANTIA Fir. i- luminosity. Thus the hue of blue glass if blue, although, on analysis, it is found to contain not only blue but all the other colors of the spectrum, in a greater or less degree, blue and violet predominating. It is the combination of these pure colors in a certain proportion that produces the blue hue whtfch is, of course, not pure. Luminosity J,^^ '"'^"'^y °f - color is ' the relat've brightness of that color with reference to ny standard, and is evidertly the only attribute of color that can be rendered in black and white. Photo- graphy, being a process in monochrome, can not render contrasts of color or hue, but only contrasts of luminosity or light and shade. That is to say if there are two hues, red and blue, for instance, of the same luminosity, a photograph or any reproduction in mono- chrome should represent them as of exactly the same shade of grey. The truth of this statement will perhaps be more evident to you, if you ask yourself which should be the darker. Ordinarily, however, a photograph will not represent them as of the same tint, and our object is to see if we can not obtain, in a photograph, an exact representation of the luminosities of the object or objects de- picted, irrespective of their colors. Evidently, then, it is necessary, in order to make ac- curate comparisons, to measure the relative luminosities of colors. Luminosity ^^ *^ ^^ once seen on observ- r .1 ing the spectrum, that cer- tain of the colors, such as opecirum orange and yellow, are much brighter or more luminous than any of the others. The actual brightness, or luminosity, or light intensity of the —8— various colors of the pure prismatic spectrum has been accurately determined by many in- vestigators, Frauenhofer,Crova and Lagarde, Abney and others. Captain, or as he now is, Sir William Abney's values were taken and plotted out as ordinates or vertical lengths above the spectrum diagram of Fig. i, pages 24-25, before referred to. Thus, for each par- ticular wave length or corresponding color, the value of the luminosity was taken, and a length proportioned to that value was set out in the vertical line through its position on the spectrum diagram. The resultant curve, drawn through these points, shows graphically the luminosity at any part or of any color of the spectrum. It is at once seen, from the figure, that greenish yellow is the brightest color, the values shading down rapidly on either side to the red andblue. Luminosity ?^ '^^" "°*' °^ course, be r o. . inferred from this that, in * natural or pigment colors, VrfOlOrS greenish yellow is at vays the brightest ; in fact it may be so shaded with black as to be less luminous than a red or blue. In order to obtain the luminosities of colors, other than the pure spectrum colors, which are in actual practice seldom used, it is necessary to measure them. There are several methods of doing this, perhaps the simplest being to compare the luminosity of the color to be measured, with the luminosity of the grey produced by the —9— mixtureof black and white. A disc of Wack and a disc of white card of the same diameter, which were slit along a radius to allow over- lappings, were placed on the same axis as a disc, of greater diameter, of the color to be measured and rapidly rotated. The eye de- termines, almost instinctively, whether the color is brighter or darker than the grey produced. When it is judged that they are equally bright the luminosity of the grey, and hence of the color, is determined from the relative proportions of black and white pro- ducing the required grey. If the rotation apparatus has a device for changing the pro- portions of black and white during rotation, the measurement is much more easily and accurately made. By this method, and with the assistance of Mr. A. H. Abbott, 8. A., of the Psychological Laboratory of the Univer- sity, who has had considerable experience in photometry, the luminosities of a set of colors, for which I am also indebted to Mr. Qj|« Abbott, were determined. These CU ♦ colors, pasted on black card- ^ *'* board, were used as a test object, and a photograph of this test chart is reproduced in Fig. 3. It will at once be seen, from the card being represented white, that the figure is a negative of the chart. In that case, if the plate correctly rendered luminosities, the brighter the color the darker should be its representation. In fact the densities of the rtrips should be proportional to the lengths of the radial lines which repre- F'Sr- 3-— Color Chart on Cramer " C " Plate. sent, graphically, the measured luminosities of the colors. The negative, of which Fig. 3 is a reproduction, was made on an "ordi- nary " (Cramer '* C ") plate, and it shows, at a glance, the correctness or rather the incor- rectness of the representation. The brightest colors, Nos. 9 and 10 (yellow and yellowish green), instead of being represented by the heaviest deposit, have only slightly affected the plate, while some of the darker colors, notably the blue (No. 17), have the greatest action on the plate as represented by the densest deposit. li > I! . PhotOgfraphs ^^^ explanation of .this of the curious result is easily Spectrum "^T* '^^^" ^°" ^''^'"'"^ ^ a photogfrapl. of the spec- trum on such a plate. Fig. 4 is a reproduc- tion of the negative and has superposed, in addition, the spectrum chart and luminosity curve previously referred to. The positions of the colors and of the absorption lines are shown by the latter, and it can at once be I!!! R ED |0?'0> J GY-; GRF.ttl Fig: 4.- Photograph Spectrum on Cramer " C" Quartz Prism. seen which of the spectrum colors are im- pressed on the plate. It is found that the waves giving rise to the red, orange, yellow, and yellow-green colors have no eflf.-t on the plate ; while waves giving rise to th ae and violet sensations, and other waves _ .,ed the ultra violet, beyond the limit of the visible specirum, produce very marked deposits on the plate. This peculiar behaviour of the plate under the spectrum at once affords an — 13 — explanation for Ihe defective renderings of color values in the photograph of the color chirt, and will also explain the peculiarities of color rendering so common in ordinary photographs. Color Sensitive Plates The evident remedy for such a state of affairs is to produce a plate which is sensitive to all the visible spectrum colors, and, moreover, which is most sensitive to those which are most luminous, and in direct proportion to their luminosity. For if we can photograph the spectrum correctly, we can also photograph any other object correctly, whose colors are always composed of spectral or mixtures of spectral colors. And, in fact, we can reproduce natural colors more satis- factorily than spectrum colors, for the reason that, although the dominant hue of the object may be one that, if pure, would not affect the plate, still other colors composing this domi- nant hue may be active and help to produce a truer rendering than if the colors were pure. »¥•* T « 1 A plate, sensitive to wave The Ideal , .. .. . „ lengths correspondmg to all X late the spectral colors in proportion to their luminosity, and insensitive to wave lengths longer or shorter than these, has not yet been discovered, nor, if an opinion may be expressed, will it ever be discovered. We have plates now, however, which are a de- cided improvement over the ordinary type, — «3— ^. B.*r-'i and which, by the aid of a certain artifice, \yill produce exceedingly close approxima- tions to the ideal result. Historical ^^* ori^^inator of this later type Of . * of plate was, undoubtedly, the OKCtcn late Prof. H. W. Vogel, of Berlin, who, in 1873, observed that collodio- bromide plates, stained with a yellow dye, were much more sensitive to the yellow band of the spectrum than unstained plates, and who, after further investigation, announced that if certain red or yellow dyes were used to stain these dry collodion plates, their sen- fitiveness to yellow and green was much increased. This discovery created quite a sensation in the photographic world, and the action of a large number of the coal tar dyes was investigated by scientists such as Eder, Abney, Bothamley, Ives and others. The introduction of the gelatino-bromide dry plate, and its marvellous qualities sorbed photo- graphers' attention for some time, and ortho- chromatism (from two Greek words meaning " correct color "), as it was called, was allowed to lapse, and was apparentl> fc-got- Commercial '^"\,.^"^ '" .SSa-'Sj Tailfer p _ and Clayton at Paris patented :S a method for making the gelatino-bromide dry plate color-sensitive by adding eosin to the emulsion before coating. This method, improved as regards the keep- ing qualities of the plaies, was introduced —14— St' .: into England by B. J. Edwards & Co. in 1884. . , In America, John Carbutt in- American ^ugurated the commercial man- ufacture of color-sensitive plates in 1886, while later the Cramer Co. purchased the Edward's process and so improved it that, according to their claim, the use of color screens was largely avoided. They called their productions isochromatic (equal color) plates, and these plates, which give very good results, can be obtained in Toronto. Many of the American dry plate manufactur- ers have now added orthochromatic plates to _ < the list of their products. In French France Lumi^re Bros, have in- troduced a remarkable series of plates sensi- tised for varying portions of the spectrum. Series A, sensitive to yellow and green ; series B, sensitive to yellow and red ; and series C or panchromatic sensitive to red, yellow, and green. In this connection it must be noticed that all orthochromatic plates, as well as ordinary plates, are decidedly too sensitive to blue and violet. In England _, ^, f among several commercial CngflfStl brands of orthochromatic plates, besides Edward's, already referred to, I will only mention two, the Ilford chromatic and the Cadett spectrum, as these can be obtained in Toronto, The Ilford chromatic is a very clear working and brilliant plate, with which excellent results can be obtained ; its prin- cipal disadvantage is its slowness, which "«.';- i«?is:i.?si3«*i: ' ■■*5.- * : / :^4ii} ' I I , practically debars its use, especially with screen, wherever there is movement. ^ The Cadett Spectrum Plate is re- markable in two ways: for its extreme rapidity, being: by far the fastest ortho- chromatic plate made, and for its sensi- tiveness to all the spectrum colors, with the exception of a narrow band in the extreme red. Messrs. Cadett and Neall have issued light filters for use with the Spectrum plate, which enable a true record of color luminosities, with the exception of the dark red before mentioned to be obtained. The '^ *^*" "°^ ^®' *^ y®^' definitely Thco explained why the silver salts ' in the film are rendered sensi- tive to light vibrations of longer wavelength, by the addition of certain dyes. Two theories have been propounded to account for it. According to the physical theory, the light vibrations absorbed by the dyes are transferred to the silver ^alts, thus rendering them sensitive in the region of absorption. In support of this theory is the fact that each dye sensitises in that portion of the spectrum which it absorbs. The chemical theory states that the coloring matter is decom- posed under the action of light, the complex molecule of the dye giving rise to products, which act as reducing agents on the silver salts. In support of tiiis theory, in likeman- —16- I ■:{■:' _:ji-'?5i^'~:-^/!JK£- ner, is the fact that the best sensitisers are those dyes most unstable under the action of light. vir A^f jf f There are, as will have been Methods of ,- a e .u i • . • • ^ ^ ^ noticed from the historical Sensitising sketch, two methods of orthochromatising, or, more simply, sensitis- ing plates. The first method is by immers- ing an ordinary plate in a bath of the dye solution, eosin, erythrosin, and cyanin, being the dyes most commonly employed. The second method is to incorporate the dye with the emulsion in the process of manufacture. Practically the same effects can be obtained, but, aside from the troublesome nature of the first process, plates prepared by bathing have not good keeping qualities, from one to four weeks being the limit ; while plates pre- pared by the second method keep nearly as well as ordinary plates. In preparing plates by the bath method only a very weak solu- tion of the dye is needed, one part in from 3,000 to 25,000 parts of water ; and, to increase the sensitiveness of the plates, although reducing their keeping qualities, a little ammonia may be added. The plates are immersed for one or two minutes, as a rule, and are then dried in the dark. T> ,* J I prepared some Cramer " C " plates by immersion, sensi- r^lates tising them with eosin, with erythrosin, and with cyanin, and ihese plates were then used to photograph the spectrum. —17— if it is desired to obtain exact quant itive results from a test of this nature it is neces- sary to form a pure spectrum on the plate by means of a spectroscope. This could not be arranged without special apparatus, so a spectrum, formed as previously described, was used and a photograph of this taken in Spectrum ^ .-•^'"era. The results ob- T>, , tamed by this method, rhotOgraphs although not giving an exact measure of the amount of light action of the various parts of the spectrum, clearly indicate the effects produced on the different plates, and, when placed in the form of dia- grams, enable comparisons to be made. From the negatives thus obtained curves were drawn, in the same manner as the luminosity curve before refened to, except that the length of the vertical lines was taken as proportional to the light action on the plate, or, what is the same thing, to the density of the resulting ntn-ative. The GraohfCal ""^^"'^^"^ curve and the lumin- M ffirt/4 °^'*^ curve were then drawn Method above a figure showing the positions of the absorption lines, and evi- dently, the more nearly these curves coincide, the better will be the color renderings of the plate. In the upper, right-hand corner of Fig. 1 is the diagram representing the effect on a Cramer "C " plate, the negative itself being reproduced in Fig. 4, and at the other side of ihe figure is another diagram from the same brand. The former was taken —18- from a spectrum produced by a quartz prism, which transmits the ultra violet waves, while jflass prisms, used in the latter, and in all the others, absorb them. It will be observed, from the diagrams of the bathed plates, that the general effect is to render them sensitive to the longer waves without appreciably diminishing their sensitiveness to the shorter or blue and violet waves. CommerCfal The other diagrams shown Ortho- '" ^*^- ' were drawn, as will chromAt'r ^® ^^®" ''^°'" '^® titles in the figure, from negatives of the spectrum on three different Plates Fig 5.— Photogrraph of Spectrum on Cramer Iso. Med. brands of commercial orthochromatic plates, the Cramer Isochromatic of medium speed, the Ilford Chromatic, and the Cadett Light- ning Spectrum. These are, so far as I know, the only makes obtainable in Toronto. To avoid choosing any pprticiilar brand for experiment, the three were investigated, as impartially as possible, and the results will be —19— presented with no more comment than seems necessary for explanation. Fig. 5 is a repro- duction of a negative of the spectrum on a Cramer Isochromatic medium plate, and it can at once be seen how much better an effect is produced than on a Cramer ordinary plate. The greatest density now is obtained between the D and E lines, in the greenish- yellow, nearly at the point of greatest lumin- osity. The diagrams in Fig. i, pages 24-35, show quite a similarity, especially between the Ilford and Cramer, the Cadett, how- ever, being more sensitive to the orange and red waves, o ' 4a. These curves show that, Sensftiveneis ,., . ., . ..,. . _ although the addition of a to Olue dye or dygs tj, J he emulsion, before or after coating, has the eflfect of renoering the plates sensitive to vibrations causing sensations of green, yellow, and in some cases, red, there is still too great sensi- tiveness, by far, to the vibrations giving rise to blue and violet sensations, to correctly render color uiminosities. There are two WW .« J r methods of overcoming this Methods 01 ..a, ,. c » u ^ • • .. ^ ditficulty : hr-,t, by dimmish- LiOfreCtingf \„g (he sensitiveness of the plates to the blue and violet waves, second, by diminishing the action of the blue and violet waves on the plate. A correction, or even an approach to a correction, by the first method has not yet been attained, nor does it appear likely to be attained. The second method, however, offers a simple means, by absorbing part of the blue and violet waves, of overcomiii); the difticulty. It will be remembered that the effect of interpoHing colored glass in the path of the beam pro- ducing the spectrum was to absorb, partially or wholly, some of the spectrum colors. If then a colored glass or film can be obtained, which will absorb most of the blue and violet and all the ultra violet waves, and if this be placed in the path of the beam of light enter- ing the camera, near the lens being the most convenient place, the luminosity renderings should be materially benefited. ^1 A device of this nature is called a color screen or ray filter, and OCfCCns its appearance by transmitted light is, in general, yellow or orange. The absorption diagram of orange glass in Fig. 2 represents, in a general way, the absorption caused by color screens. Screens may be composed of colored glass, stained gelatine or collodion film, or a liquid solution contained in a thin glass cell with parallel sides. They may be placed in front of, or behind the lens of the camera, or directly in front of the plate ; or the same object would be attained by illuminating the subject with yellow or orange light. The dyes commonly employed for staining the films are some of the coal tar series, the principal ones being aurantia, aniline yellow, naphthol yellow, brilliant yel- low and others, while chromate and bichro- mate of potassium in solution are generally used for liquid screens. — 31 — m-9 Fig. 6.— Photograph of Spectrut. on Ilford Chromatic and Screen. Eff«''t of ^^® effect produced by r' 1 " Q employing^ such screens Cooler :>CfCCns ^hen photographing the spectrum is shown in Figs. 6 and 7, which are repfoductions of the negatives obtained. Fig. 6 is a negative on an Ilford chromatic plate using a liquid screen of a yellow color, while Fig. 7 is a negative on a Cadett spec- trum plate using the Cadett Absolutus screen, which is of a reddish orange color. Again Fig. I pages 24-25, represents, diagrammati- cally, these results, with two others, and shows G , BLUE B V : ViOLCf ^*g- 7 —Photograph of Spectrum on Cadett Spectrum .ind Absolutus Screen. —33 — «-r *ir as» the decided advantage of using screens. The two curves are brought nearer to coincidence, especially when using the Absolutus screen, and the effects obtained thus approach very closely the ideal result. ^1 r Evidently, the yellower or more orange the screen the more blue OCrecnS ;§ cut out, and it is easily possible to get a screen so deep in color as to render the blues too dark in the print, while, if not deep enough, they wil! be too light. It will be noticed, also, in the diagrams of the un- screened plates, that each brand has a dif- ferent shape of curve, and one screen will not, therefore, suit all plates. To obtain the best results, the screen should be so adjusted to the plate as to absorb waves having too vigorous an action, and to transmit freely ^1 r* J 4.4. those with too feeble an 1 nc ^aaett action. Up to the present Screens Cadett and Neall are the ONLY makers of orthochromatic plates who have attempted to supply screens correctly adjusted. They have issued two screens {' : use with the Spectrum plate, the Gilvus and the Absolutus. The Gilvus increases the ex- posure about five times, and tlie Absolutus about forty times. This increase of exposure is due to the absorption of the most chemi- cally active waves, and applies, of course, in a greater or less degree to all color screens. It is claimed that the Gilvus screen gives correct rendering of all color luminosities but -33— x ll 1 ^1 ^^./T ■ ^ .■-. ^ Iliii ,y all (f/^ a ff 4 f MJl-Jg'M-'jyl t' '.-*-f^f^egg|r£ -24— Fiffi •!■ »■" 1^/.*^ *ii'«" ii>. 'J. * I icilAIIEl|-C- "CWAKTZ [fKISM I — ' T-t*»tXTlM?,cnfw ' r — »5- the red, and the Absolutus gives correct values to all colors but a narrow band in the extreme red. The Ilford Co. also supply two screens of colored glass, a light and a dark, increasing exposure about three and six times respectively. The screens of course, are not adjustable, and no claim is made as to color renderings. The Cadett screens are, I be- lieve, made of four stained films, two gelatine and two collodion sealed in optical contact between two pieces of optically worked glass, and hence allow considerable variations in Absorption ^^^ absorptions produced. Spectra of "^^ese absorptions, and that o J of the liquid screen used, are OCreens indicated, in a rough way, by the shaded part of the rectangle below the diagrams in Fig. i. On comparing this with the curve above, and with the curve for the same plate without a screen a very good idea of the function of the screen may be ob- tained. 'Ill 111 IP I liii ilii Method of ^^^ ^^"^^ curves will also A J, serve to indicate the method Adjastment ^o be pursued in adjusting screens to plates. Take a photograph of a pure spectrum through the screen to be tested on the plate to be tested, measure the densi- ties of the different parts of your negative, and plot a curve to correspond. It can at once be seen, on comparing this curve with the luminosity curve, what colors have too vigorous an action, and the screen can be — ab— ISKi'.^W'BK changed to absorb more of that color. It is essentially a method of trial and error, and, owing to the tedious photometric measure- ment of densities, is not suited for commercial purposes. Abney's Captain Abney has devised CoIq, ^ simple instrument, which Sensitometer forThl" "'" ''"''"'■""^'- tor the same purpose. He takes, usually, four pieces of colored glass, red, yellow, green and blue, which transmit approximately pure colored light. The luminosity of the light transmitted by each is accurately measu^-d, and behind the three lightest are introduced rotating sectors, or patches of developed grey of such density as to reduce their luminosity to that of the darkes . Then red, yellow, green and blue light, all of equal luminosity, will be trans- mitted, which should give, on a pUte behind a correctly adjusted filter, patches of equal density. As the glasses need only be very small, the whole apparatus, with the plate to be tested, can easily be placed in a quarter plate printing frame, and hence is quickly and cheaply tested. If the patches are not of equal density, and this can readily be judged by the eye, the filter must be changed to absorb more of the color giving too dense a patch, or less of the color giving too light a patch. This, as before, is a method of trial and error, but with the advantage over the former of avoiding the tedious measurement —27— ^m ]^r Wi' 7C <'*if 0' f of densities, and of overcoming possibly errors due to improper exposure or develop- ment, while the results obtained with screens so adjusted, should be equally good. Results ^^® results obtained in photo- e graphing the spectrum decidedly the Spectrum orthoch'-omatic show advantages of over ordinary plates, the advantages of screened over unscreened orthochromatic plates, and finally the advantages of scientifically screened orlhochromotic plates over all others. In photographing colored objects of all kinds we will see that the same results obt£tin as in the photographs of the spectrum. This is what is to be expected when it is con- sidered that the colors of all objects are either spectral colors or mixtures of spectral colors. ja *. p Asa test object, a color chart Results from , . . was construe) ed by pastmg VrfOlOr Chart colored paper on a sheet of black cardboard. Twenty-four different hues, intended to represent spectral colors as closely as possible, were arranged on the circumference of a circle and four different shades of grey inside. At the two lower corners of the card ten other colors were arranged in two squares. As a guide to the colors arranged on the circumference I may say that No. i is red. No. 5 orange. No. 9 yellow, No. 13 green. No. 17 blue. No. 21 violet, and the intermediate strips are inter- mediate in hue. As before mentioned, the -28— 1/' ifO radial lines around the circumference repre- sent, by their lengths, the luminosities of the colors, and, in the same way, the lines arranged between the two colored squares represent the luminosities of the strips com- posing the squares. It must be distinctly remembered that all the figures of this chart are reproductions of negatives, and hence, in a correct rendering of the color luminosities of the strips, the densities of the patches of grey representing the strips should be exact- ly proportional to the lengths of the lines ^^ 4, beside them. Fig. -i, already Ordinary . ^ • '. ' mentioned, is a negative on a Plates Cramer " C " plate and imme- Fig. 8- Photograph Color Chart en Cramer Iso Med. diately shows the very faulty record given by ordinary plates. Three other brands of ordinary plates were tried, and, although the — 29 — Photogrraph of Color Chart on Ilford Chromatic. Fig:. lo. Photograph of Color Chart on Cadett JSpectrum. -30- differences were not very marked, the Cramer was undoubtedly the best. Fig. 8 is Oftho- ^^^of" •f>eneg:ative on a Cramer chromatic ]T '"^'^'""' ^^^^^' ^''«' 9 a" Plates ^ Chromatic and Fig. lo a Cadett Spectrum. These three, although giving a much better result than ordinary plates, are still far from perfect. The patches representing the blue and violet strips are mt.ch too dense, while the yellows Advantage ^""^ "o»«le"se enough. When of Screens Tr "'^ '^^ "^""^ '^'"^^^ ^' betore, a screen to absorb some of the blue and violet, the improvement >s decidedly marked and in three of the remaining figures the values of the circular set of strips are as nearly as can be judged by the eye, approximately correct. Figs. 1 1 and 1 2 are on Ilford Chromatic plates with the Ilford light screen and a liquid screen of a full yellow color respectively. Figs. 13 and 14 are on Cadett Spectrum plates with the G.lvus and the Absolutus screens respectively. The two lower squares form a severer lest of the qualities of plates and screens than the circle, and it is with these that the final comparison must be made. Composite '^ '** evident, from the compari- Nature of ''°" °^ spectrum and color chart Colors "^&at'V'-s on ordinary plates, that the values given by the latter, although far from correct, are not so faulty as we would be led to expect from the photographs of the spectrum. The cause of —31 — Iff. II. Photojfraph of Color Chart on llford and Ilford Screen *• ■.! Fi^. 12. Photogrraph oi Color Chart on UforC, and Bichrome Screen. —32— ""■W^ r'.iMP'T^I- a.; . «...j>':«apuMBWe Photosrraph ot Color Chart on Cadett Spictrum and Uilvus Screen. Fib:. 14. Phot-graph of Color Chart .,.. Cndclt Spectrum and Ahsolutus Screen. this IS evident when the colors are analysed spectroscopically. h will be found that they do not reflect one or two colors only, but, generally speaking:, all the colors but one or two, which are, of course, complementary to the dominant hue. The composite nature of these colors, and indeed of all colors in nature, •sanaid lo their coriect rendering. If the colors of natural objects were pure spectral colors, the ordinary plate, being sensitive only to the blue and violet, would not be affected by j^reen, yellow, orange or red, and, as a consequence, they would alike be ren- dered, in the print, as one mass of black without form or detail. Colored '^ '* * natural transition from PictufM *^* pigment colors on the test ^"'" chart to those used in paintings and other colored pictures, and similar effects will be noticed in their reproduction. In all colored pictures, especially old paintings or pictures containing deep brown or other dark tints, the use fa screen, comparatively deep in color, is necessary to obtain the values of the different colors, and indeed it seems to be impossible to obtain detail in the shadows without such a screen. In the subject chosen for reproduction (no figures ; slides shown in original lecture) this is shown in a marked degree. The difference between the results obtained on ordinary and on orlhochromatic plates without a screen is not very marked but when screens are used the improvement —34- fc '■".^m "* very noticeable. Not only i« ,he effect much more brilliant, and free fron, the flatness which characterize, the former, but also de- tail iH obtamed in the browns and reds which .« entirely absent in the unscreened pictures. Moreover ,t is quite evident from the result, that a deep screen (the - Absolufus") jjives much hr ,r rendering than a plain full yellow, or orangre screen, and that, by the use of a properly adjusted screen, effects completely satisfying to the eye can be obtained. Flowers ^^^ '''°'°" °^ nature, on the other hand, are chiefly of in- terest in this connection as they are present m flowers, fruit, etc., and in landscapes. Flower photography is another branch of the art ,n which orthochromatic plates offer de- cided advantages over ordinary plates. The use of a screen, however, i. not so indispens- able ,or . ,e best results as in the case of pictures. Indeed, if there are no blue or violet flowers, better results will often be obtained without a screen, as its use. espe- cially If at all deep in color, seems to intensify he yellows, causing them to appear too white LllITT- " ''""^- An exception must be made to this statement in the caseof.he Absolutus screen, which, when used with the Spectrum plate, always possesses a decide ' advantage. A screen is necessary, if there are blue or violet flowers, to cut down their actinic value and to prevent them appearing tr.. f 'V^f^''"'- These points are illus- trated in the slides (no figures) of crimson. —35— P-nk, and yellow roses, and violcc-. The or l>ochron,a.ic pla.es wi.l.ou, a sceen give both ,he volets are by far ,00 light. The use of a yellow screen, although bringing the volets to their correct tone, ntakes the yellow roses too light, and the only method of over ad u d b, „„, „, .,.^ „,^„_^^^^ ^ Landscapes ;s '^^'^ '^^^ and. to most photo- p.-aphers, the most important l2Tl "' ■"•""-'"P"^. 'andscapes', also exh b„s .„ comparative results the advant- ages of color sensitive plates. At this time of the year, wmter. when landscapes contair ■ttle else but black and white, orthochroma" UC pla.es naturally, do not show such marked advantages over ordinary plates as ■n s.nnng, summer or autumn. Thev still sufficient to recommend their use. These r».nts are : first, used with or without a screen better detail in the snow can be secured, second, parts of the landscape uncovered with snow show much more ^e a'i screen, than on ordinary plates, third cloud ,„ be obtained without special car^ as the landscape, and here also a screen is an advantage. But it is chiefiy in the o h reasons that the advantages of orthochroma- -36- fied ihal no one who makes a fair .rial of or..,ochroma>ic and ordinary p.a.e. 1 he an,e .nhjec. „i„ be ever a^ain con.en. Juh the faulty records „f ,he la„er. These scenes, are dlustra.ed by slides, are: im- proved rendering of the various shades of green .n the foliage, the genera, brightening ing?„ t" "" " """ "' ""^ "^y- 'he clear^ ■ng up of an excess of haze obscuring the d-tance. and, finally, „,e p„,,ibiliTy „f Qouds "blaming, especially when a and l»„H '"'•'^'''^ ''"■"" '* "'«d, clouds and landscape m the same negative. Even wdl produce a very natural tone in the skv gradually lightening as it approaches tfe honzon, and any clouds, even very light and fleecy ones, will show in the printf as i„ nature, as white on a darker background Screens ^''^ foregoing remarks indicate the use „f '" ""^ ""'•'""'J' °f <^»«'. the use of a screen is necessary if true color values are desired. In ,he reproductio^ of P^mentsinany form a screen is necessary and .t ,s generally desirable in other subjects as flowers and landscapes. If these attl; objects contain yellows and greens Zy, the use of a screen would not cause a marked d,«^^rence in the result, but if there are Wue' ~.:. 't;„rd::;rr";cf ••''""^ "'"- vn their actinic value to cor- —37— f respond with their luminosity value is neces- sary. Varieties ''^'* regards the kind of filter, of Screens *''^ ""^^^ desirable, if it could be obtained of the correct tint, would be a piece of optically worked colored glass. But, as will readily be seen, it is practically impossible to obtain a suitable color. The liquid screens, such as the Bausch & Lomb ray filter, present the ad- vantage that their color can be varied by altering the strength of the solution, but possess the drawback that the liquid evapor- ates knd requires replenishing, and that the inside of the cell gets dirty and requires cleaning, which is troublesome to accomplish. Screens made of stained gelatine or collodion films, eit^^er coated on glass or sealed be- tween two glasses, can be obtained of any desired tint and can consequently be cor- rectly adjusted. There is no liquid to be spilled and evaporate, but care has to be taken to keep them from the light when not in use, as there is danger of fading. Cadett '" working with the Cadett spec- _ trum plates, it would be advis- OCreens able to use the screens they supply, as much better results would be ob- tained than with any screens not specially adjusted. The Gilvus can be used for all, or nearly all landscape work, and does not in- crease the exposure sufficiently to prohibit instantaneous work. The Absolut us should -38- be used where absolutely correct values are required, and when the increase of ex josure permitsr. If both can not be obtained, the Gilvus would probably be the most generally useful for amateur^. For use with the Ilford, Cramer and other orthochromalic plates it is useful to possess two screens, a yellow and an orange, the yellow answering for ordinary landscape work, and the orange when there are fleecy clouds to be reproduced, or in Unadjusted Ph°t08:'-aphing pictures. The Ilford screens, light and dark, OCrCCns ^^de of colored glass, would answer for most purposes, and are quite con- venient to use. Whatever screen is used, however, it is essential that the glass or glasses of which it is composed be perfectly plane and parallel, or distortion, and conse- quent loss of definition, will inevitably result. ManipolatiOn ^^^°^^ concluding a few of OrthocLfO- ••^'"^'^^^^ about the working . , Tif A * °^ orthochromatic plates matlC Plates ^j^ht be of service to be- ginners in the method. It has often been urged against their use that they are trouble- some to manage in development, and of poor keeping qualities. Orthochromatic plates, certainly, can not be exposed so freely to the regular ruby light as ordinary plates without danger of fog, but ample light can be used, especially in the latter stages of development, to properly watch progress ; and while, per- haps, not keeping so well as ordinary plates —39— Exposure their good qualities are preserved much longer than is generally imagined ; for in- stance, a box of Ilford Chromatic plates, which I purchased nearly a year ago, yields excellent negatives without any sign of de- terioration. When using slow orthochromatic plates, such as the Ilford Chro- matic and the Cramer Iso. Med., ample ex- posure is required. For, while they seem to possess great latitude as far as over-exposure is concerned, the evil effects of under-exposure are far more marked than is the case with ordinal^ plates. In my experience with these two makes not one plate has been over- exposed, although, as I thought, ample ex- posure was always given. The Cadett Spec- trum plate, as is natural with a very fast plate, has not so great latitude, and care must also be taken not to over-expose. As a guide to the exposure required, if we take the typical case of an ordinary landscape on a bright day in summer, at three o'clock in the afternoon, and if the lens be stopped to f-ii, the Ilford Chromatic would require an ex- posure of at least iV second, the Cramer Iso. Med, 1^2 second, the Cramer Iso. Inst. -3V second, and the Cadett Spectrum ^^- to j^jj second, depending on the speed number of the emulsion. If a yellow screen were used these exposures would be increased four or five times, and the slow plates could not be used ^or instantaneous work, even with an aperture of f-8. ^ —40— \ IS Dark Room '^''^ developing light that . , j^ >*afe to use in working orlho- i^ISfntS chromatic plates depends, of course, on the plate. Plates that are insensi- tive to red can be worked in a deep ruby light without fear of fog, but plates sensitive to red can not be safely exposed to such a light. For plates of the former class, of which the Ilford and Cramer are examples, the regular ruby and orange glass may be used, with the illuminant turned down low, and the plates exposed to the direct rays as little as possible during the early stages of development ; when development is nearing completion, however, the light may be turned up to ex- amine progress without danger of fogging the plate. For plates of the latter class, of which the Cadett Spectrum is an example, develop- ment will either have to be conducted in darkness, or with a light which transmits a color to which the plate is insensitive. Cap- or tain Abney recommends two _ , ^ gelatine films, formed from dry l^IS^ntS plates fixed and washed, one stained deeply with methyl violet, the other with brilliant yellow, and bound together film to film with lantern slide binding strips. The absorption spectra of methyl violet and aurantia, which is similar to brilliant yellow, are shown in Fig. 2, and it will be observed that, while methyl violet transmits a narrow band in the extreme red and some blue and violet, aurantia transmits everything but blue and violet. If the two are combined the —41— sf result W.1I be. as shown in the lower diagram v to transmit only the extreme red. and snch a combmat.on. if carefully used, should be safe w.th the Spectrum plates. Sanger Shepherd recommends three films, of methyl violet naphthol yellow, and aurantia. and for a com- bination still more safe naphthol yellow aurantia. brilliant green, and fuchsm. The Cadett safe light, which can be purchased ready to use, will probably prove mo.t suitable for amateur's use, and is. I believe, composed of a film sta.ned with methyl violet and two colored glasses, yellow and yellow green Even when usmg these safe lights it would do "o harn, to keep the tray covered, during de- velopment, as much a« possible. Developers ^^® developers to be used . ^.«. ^°'' ^''thochromatic plates do not differ materially from those ordinarily employed. In fact any developer suitable tor an ordinary plate can be used on an orthochromatic plate. There is one point to be guarded against, however, when working with slow plates such as the Ilford and Cramer Medium, and that is to use only about one half the usual quantity of the developing agent or reducer, pyro, metol, e.konogen, or whatever is used, with the usual quantity of alkali, and to make up the Excess of ''^'*"^« with water. It will be Contrast ^"""'^ *^^^ '^ ^ developer of full the negative will be too dense and contain —42— too much contrast to give harmonious resuUs. This peculiarity, the great density giving powers of orthochroriiatic plates, should be of great advantage for any subject whose contrasts it is wished to increase, for instance a landscape flatly lighted, or on a cloudy day. For such an effect the proportion of develop, ing agent might be increased with advantage. Very fast plates, such as the Cadett Spec- trum, do not require tins precaution and in fact the developers for these plates call for a larger proportion of reducer, or rather for a smaller proportion of alkali. The object of this, no doubt, is to prevent chemical fog which is more likely to occur with fast plates, but in my experience I had no difficulty with fog, and the development proceeds in exactly the same manner, only more slowly, as r ith ordinary plates. Necessity for ^" ^^^"^ ^^^ manipulation of c . orthochromatic plates differs cspenence only in the points mentioned from that of ordinary plates and no difficulty should be experienced in their successful handling. It must be remembered, however, that any change in materials or methods in photography requires experience to become familiar with the new conditions imposed, and hence that perfect results should not be expected to be attained with the first box of plates. —43- Conclusion '" "•""cl'-sion I can only hope we'la. th. r •"'"',' '""'' ""deeded, as wej a. the hm.ts of a single lecture allow '" "'f'"^ you familiar with the n,ain po „T,' about orthochro«,ati.m and the correcret denng of color values. 1 ,ru„ alto T. :::: ::- r -''"' -"" »■■- ---i,?: ;-.ou „ii, ...:f'.:-'j^z,:::z::: the ordinary plate, and will ,h„s l,as,e„The day when its faulty record win be repTa ed X^,Z,r ™"deri„g of color lu„,i„„sities given bv the orthochromatic plate. ^ —44— The same careful attention in all detail, it given to the other Brands of Plates manufac- tured by Messrs. Cadett & Neall, Limited, Ashtead, Surrey, England, and the Trade can be supplied by MESSRS. EAKINS & FERRIS, Sole Importers, Toronto, Canada. Ordinary Plates Professional Plates Special Rapid Plates Lightning Plates Spectrum Plates Safe Lights Screens 'limwsfAiiESiijme'iSftsm.wm--^ . v»«»rram0svw "»fP IftT^'MiCi*--', -" ,1,- «• J»fl^^ai«f."^vJj"-?' - -^ 'i^yn '