PhotoHistory

From: PHOTOGRAPHY OF TO-DAY

By H Chapman Jones, 1913

PHOTO-MECHANICAL PRINTING

The problem of multiplying copies of photographs was present in the minds of many of the pioneers of the art. The preparation of a negative from which prints may be made by the photographic processes described in other chapters was a step in this direction. If a large number of prints is needed, as for publication purposes, it is usual to make duplicate negatives so that several prints may be simultaneously produced, but in any case the operation is tedious and the results comparatively costly, especially if a “printing-out” method is employed. It was realized nearly a hundred years ago that it was desirable to be able to prepare a plate by photographic means that would enable impressions to be obtained in an ordinary printing press, so that the accuracy and rapidity of photography might be combined with the facility of common printing methods. The elder Niepce did something, Fox Talbot did a good deal more of work in this direction, and in the early fifties there was realized a measure of success that was distinctly encouraging and that soon brought innumerable workers into the field. In one direction it was sought to photograph directly on to the block for the wood engraver instead of drawing upon it by hand, but this kind of work is now very nearly obsolete, for photography has superseded not only the draughtsman but the wood engraver too, except to an exceedingly small extent. Steel and copper plate engraving has followed wood cutting, and one may say in general language that all illustrations for book or similar work are now produced photographically. We do not propose to trace the history of the development of modern methods of illustration, for this is a very extensive subject, but rather to give some idea of the principles upon which the chief of them are based.

If a print is made from a negative by any photographic process, the result is generally called a photographic print or simply a photograph; but if from a photograph or negative a plate or a block is prepared and impressions are taken from this, the result is called a photo-mechanical print, because although it has a photographic origin, the print itself is prepared by mechanical means. We have seen that a carbon print is prepared by treating a pigmented film of gelatine with potassium bichromate, exposing to light under a negative, and then dissolving away with hot water those parts of the film that have not been acted upon by the light and therefore remain soluble. The darkness of the image depends entirely upon the thickness of the film that has been affected by light and so made insoluble. The dark shadows are represented by a comparatively thick layer of the pigmented gelatine, the medium tones by a thinner layer, and any parts of the image that are white have no deposit at all upon them, for there it has all been dissolved away. The image in fact is in relief, and the height of the relief is proportional to its darkness. This is a photographic print; but if an exactly similar relief is made by pouring melted pigmented gelatine into a mold and allowing it to set and dry, the print is a mechanical print, and if the mold is obtained by photographic means, it is a photo-mechanical print. The “casting” method of getting prints is called Woodburytype, after the name of the inventor. A gelatine relief is prepared in very much the same way as a carbon print is made, but in higher relief, and this when dry is placed with a sheet of lead in a hydraulic press, and the two are forced together until the relief is driven into the lead. Gelatine when dry is very tough and hard and gives a clean impression. The lead plate is printed from by pouring a pool of melted pigmented gelatine on it, putting the sheet of paper on top, and subjecting them to pressure in an ordinary printing press. The gelatine sets and the paper with the print upon it can be removed from the mold and hung up to dry. It is clear why the relief used in making the mold must be “higher” or thicker than an ordinary print, for it has when dry to produce a mold that will hold enough gelatine jelly to give the print, and the jelly shrinks very considerably as it dries.

In the “Stannotype” process the need for the hydraulic press is dispensed with. The gelatine relief is made as before and the mold is built up on it by passing it with a sheet of tinfoil between india-rubber rollers, so that the foil is pressed all over into close contact with its surface, and this tinfoil is then strengthened by the deposition of a metal upon it and backing it up with a suitable resinous body. Or the gelatine relief may itself be used as the mold by coating it with india-rubber cement and covering it with tinfoil pressed thoroughly down upon its surface.

There is one important feature of Woodburytype prints, namely that they are exactly analogous when finished to photographic prints made by the carbon process. They give a perfect gradation of unbroken tint, thus resembling photographic prints and differing from photo-mechanical prints produced by any other process. The “ink” is not what is usually understood by that term, but gelatine mixed with the required pigment. In all other mechanical processes printers’ ink is used, that is a mixture that has the general characteristics of an oil paint.

The “oil printing” methods we described earlier show that when a gelatine film is sensitized by means of potassium bichromate and exposed to light, that in proportion as the light acts the gelatine loses its power to absorb water and gains in its power to retain a greasy ink (or oil paint). If now the gelatine film is imagined as on a thick glass plate instead of paper, and the application of the ink to be done with a roller, then by putting a sheet of paper upon the inked-up plate and pressing them together, the paper will take the ink. The inking-up and the taking of impressions on paper may be repeated, and we have a photo-mechanical method of printing. This is in essence the process known as “collotype,” a most excellent method, though not so much used in this country now perhaps because the varying condition of the atmosphere with regard to its moisture affects any gelatine surface and introduces uncertainties that appear to be difficult to cope with. It will not be supposed that the preparation of a collotype plate is so simple a matter as just described. In the earliest attempts, about 1865, although the film was hardened, only some four or five dozen prints could be obtained from the plate before the surface deteriorated to a marked extent. But by degrees the adhesion of the film to the glass plate was improved and the process was perfected in other directions. It might be at first supposed that in this case there was a continuous film of ink thicker in the darker parts and thinner in the lighter. But printers’ ink is practically opaque, and although a little difference in intensity in the impression can be obtained by a slight variation of pressure, it is not possible in this way to get more than an exceedingly restricted range of gradation. And besides this, the ink would not “hold” well on a perfectly smooth gelatine surface. The fact is that the surface of the gelatine in a collotype plate is not continuous, but is broken up into innumerable little cracks, reticulated as it is called, and a part of the skill in the preparation of a collotype plate is shown in getting the reticulation of a suitable fineness. The coarser the reticulation is the more easy will the printing be, and a great deal too depends upon the character of the rollers and the manner of using them, for while a slow and heavy application will deposit ink, a quicker and lighter movement will remove it.

In any case a gelatine film requires to be treated with more care than a stone or zinc surface, and the greater care means slower work. In lithography, which is printing from a stone surface, the design or picture is obtained upon the stone in a greasy or waxy medium, and as the stone is slightly porous the material enters to a small extent into its surface and so becomes firmly fixed. By sponging the stone over with a solution of gum, this is slightly absorbed where the stone is bare and rejected by the greasy image. The conditions are now generally similar to those of a collotype plate; by passing an inked roller over it the image will “take” the ink while the moistened stone itself rejects it, and a sheet of paper pressed into contact will receive an impression. A finely ground zinc or aluminium sheet may be used instead of stone and the process is then called zincography or the equivalent. The drawing to be reproduced may be made directly on the stone, but more often it is made on a prepared paper and “transferred” to the stone or the zinc. So far these processes have no connection with photography, but it is possible to produce the greasy image by photographic means, and then we have photo-lithography or photo-zincography. The image may be produced directly on the stone or metal by coating it with a gelatinous material made sensitive with a bichromate, exposing to light under a negative and applying a thin coating of ink all over the surface. By washing with water and gentle friction, the soluble parts of the film with the ink on them are removed, and the image can be got into condition for printing from. But the more usual method is to prepare a “transfer” on paper coated with gelatine or gelatine and albumen by sensitizing it with a bichromate, exposing, and then applying ink to it as already described. The transfer is then laid upon the stone or metal sheet, pressed into contact with it, and the paper stripped off. There are very many variations possible at almost every stage of the process, and if the subject is other than a simple line subject, the method must be such as will cause the surface of the gelatine to reticulate in order that the gradation may be fairly represented. Photo-lithography and photo-zincography are largely used in the printing of maps, and one advantage of the latter method over the former is that the metal plates are lighter and much less bulky than the stones.

In collotype, photo-lithography, and similar processes the printing is done from a practically flat surface, and the distribution of the ink put upon it is determined by the character of the surface. It is true that there is sometimes a very slight relief, and this may be of a little assistance, but it does not constitute an essential feature in the process. There are two other main divisions of mechanical printing methods, and both of them are serviceable as photo-mechanical processes. In the first case those parts that are required to hold the ink are cut out or sunk below the surface, and in inking the plate these depressions are filled with ink while the surface of the plate is wiped clean before the paper is put upon it to make the impression; and in the other the design stands up higher than the remainder of the plate, which is cut away or otherwise lowered, and the ink is put on by a roller that touches only the projecting portions. The first is the character of a plate from which an engraving or etching is printed, while the second is sometimes referred to as a typographic method, because the types used in the printing of ordinary books and such documents have the letters and designs upon them standing up in relief. It might be thought that Woodburytype printing came in the first category, as there the picture is sunk and in printing it is filled with the ink. But Woodburytype is a process apart, the ink is not printers’ ink but a melted gelatine jelly with which the pigment is mixed, and the mold is not “inked” in the sense in which a printer inks a plate; the pigmented gelatine is simply poured upon it, so that when the paper is pressed on the top the jelly as it cools may set in the form determined by the mold. Printers’ ink cannot be used in this way at all; it does not set like the melted jelly, but merely gets hard by exposure to the air as oil paint does. A thick layer of it would spread and run after the paper was removed from the mold, and even if this were prevented it might require weeks or even months to harden. On the other hand, if it were attempted to ink up a Woodburytype mold as an engraved plate is inked, the process would fail, because in wiping the ink off the surface the cloth would go into the hollow parts and remove the ink there also. In fact there is no true surface in this case except round the edge and in the few small parts that are to be white in the print.

It is this very matter of surface that constitutes the radical difference between the Woodburytype mold and an etched or engraved plate that can be printed from in printers’ ink. The general surface of the plate must be preserved, and the lines or depressions that are to receive and hold the ink must always be narrow, and if a more extensive part of the plate is required to hold ink, the lines or depressions may be multiplied, but they must not on any account run into each other so as to make a depression of large area. In Woodburytype a thin layer of the jelly gives a light tint, but a thin layer of printers’ ink will give almost its full color, and to get a light tint with it, the ink must be alternated with blank spaces of bare paper. However closely a Woodburytype print is examined it will be found that the tones are continuous like the tones in a painting or an ordinary photographic print, but in a print from an etched or engraved plate the tones are all “broken,” the ink is in lines or dots, and it is sometimes necessary to look at the print from a little distance so that the lines or dots may merge into the inter-spaces and give the appearance of a continuous tint. It is possible to make the patches of ink so very small that the closest inspection will not reveal them: in such cases it only needs a magnifying glass or a low power microscope to show that they are really present.

If the subject from which the plate is to be made is a line subject, such as a drawing in pen and ink or an impression from a woodcut, the photographic production of the printing plate is simple in theory and fairly simple in practice, for in this case there are no half tones to trouble about. We will suppose that it is desired to produce a block to print with type that shall represent a diagram drawn in black lines on white cardboard. A sheet of zinc with a flat and suitable surface is coated with a thin film of albumen (white of egg) made sensitive to light by adding to it a solution of ammonium bichromate. The film is dried in the dark, pressed into close contact with a negative made from the original drawing, and then exposed for a short time to a strong light. Where the light acts upon the film the albumen is made insoluble, and it might be thought that it only remained to wash away the soluble albumen and put the plate into acid to dissolve away the surface of the exposed metal, to get the plate with the exposed lines raised on it ready for printing. This simple method would fail, because the insoluble albumen would not prove an effective protection to the zinc in the presence of the acid. The exposed plate is therefore first coated with a greasy (printers’) ink by means of a roller, then soaked in water and gently rubbed until the soluble parts of the albumen with the ink on them are washed away. We have now the lines in insoluble albumen with a thin coating of a greasy ink on them. By sprinkling finely powdered bitumen over the plate it will adhere to the greasy surface of the lines and may be dusted off the plain metal surface, and when gently warmed the bitumen will melt with the ink and afford a sufficient protection to the metal to allow of a preliminary attack with acid. But for further etching the lines need still more protection, so the washed plate is sponged over with a solution of gum, dried, and more ink is applied by the roller. The plate may now be returned to the acid that a little more of its surface may be dissolved away leaving the lines standing up in higher relief. But this action cannot be pushed far or the acid would dissolve the sides of the projecting lines of metal, and in time undercut and actually detach them. To prevent this a resinous substance is dusted on, it adheres to the greasy ink, and by gently heating the plate, it is melted and allowed to flow over onto the sides of the projecting lines of metal and protect them. By continuing these operations, the metal is gradually dissolved away where it is not protected, until the lines stand up sufficiently high to allow of the plate being printed from.

If the object or picture of which a printing block is required does not consist of simple black lines but of continuous tones of different darknesses, then a necessary preliminary to the making of the block is the translation of the even tones into an equivalent mixture of simple black and white. Suppose, for example, that there is a gray midway between black and white, we must have, instead of the continuous gray, a block that will print black lines or dots that shall collectively cover say half the space that represents the gray, so that when the print is viewed from a suitable distance the black dots or lines and the white spaces between them will mingle and appear like the gray that is desired. If the gray is darker, the black dots or lines must be thicker or more numerous so that there is more black and less white, and if the gray is lighter they must be thinner or less numerous. In mechanical methods as distinguished from hand engraving, dots are almost always used, and as the number of them is fixed, the problem resolves itself into the getting of larger dots or smaller dots to represent the darker and the lighter tones respectively. If the dots are of a regular shape, as they increase in size they will eventually touch each other, and instead of a white surface with black dots on it, there will be a black surface with white dots. These results are simply two phases of the same effect.

There have been many methods proposed for making this translation of true half-tone into black and white. It is easy to imagine a carbon print made with a white pigment, the original being represented by a relief that is higher in proportion to the darkness of its parts. If now a kind of brush with short india-rubber taper teeth instead of bristles is coated with black ink and carefully pressed against the white print, so that the india-rubber points just touch the deepest hollows or the lowest parts of the prints, there will be formed in these lowest parts small black dots because the points of the inked india-rubber teeth will just come into contact with the surface. But in those parts that stand up in higher relief the india-rubber teeth will be flattened against the surface and form larger dots of ink, and thus the sizes of the dots will vary exactly in proportion to the height of the relief and therefore in proportion to the darkness of the part of the original represented. The various continuous tones have thus been translated into a simple “black and white.”

The method that is used at the present almost to the exclusion of others is not quite so easy to understand as that just described and others of a similar character, although it is as effective and from a practical point of view superior. An ordinary photographic print of the object is photographed through a cross-lined screen placed just in front of the plate. The screen is covered with opaque lines ruled with mechanical regularity and in two directions, with a space between each line and the next about equal to the thickness of the lines, so that it may be regarded as an opaque screen with small transparent squares about the size of pinholes regularly disposed over it. For many years it was not clear why such a procedure as this should give the desired result, and while allowing that the action of the screen is complicated, there is little doubt now as to its primary action. At the same time that the lens gives an image of the picture on the plate through the screen, each hole in the screen gives an image of the lens or its diaphragm on the plate after the manner in which a pinhole will give an image, as we have seen earlier. But as the apertures of the screen are large in comparison with the image of the lens that they give, the little patch of light that falls upon the plate behind each aperture in the screen is brightest at its center and gradually fades off towards its margins. Where there is very little light, that is corresponding to the dark parts of the subject, there will only be enough light at the very center of the patch to give an image on development, while where a brighter light from the subject falls upon the plate the developable patch or spot will be larger. Thus the translation of the tones into dots of various sizes is effected, but the dots have margins that gradually get thinner and are therefore indecisive. The dots are made “cleaner” or more precise by applying a solvent of the image to the plate which removes the thin edges without materially affecting the denser central parts.

There are many matters that have to be taken into consideration in settling the coarseness of the screen, or in other words the number of dots in a straight line that would be included within a measured inch. For printing on a rough surface paper, such as is used for newspapers, and especially when the printing has to be rapidly done, the screen must be coarse so that the spaces between the dots may be comparatively large and stand a deep etching to prevent them from getting filled up with ink. But large dots are obtrusive and obliterate detail, and from this point of view the finer the dots the better. To get the best results from blocks made with a very fine screen the surface of the paper must be very smooth and the printing very carefully done. About sixty dots to the inch is suitable for a common paper, but a person with ordinary sight will see such dots when the print is held at the best distance for distinct vision. With one hundred and twenty to the inch the dots would be invisible unless the print were brought closer to the eye, and from this to about a hundred and fifty is commonly used for book illustrations. If very minute detail has to be reproduced the fineness may increase up to two hundred, or in exceptional cases to four hundred to the linear inch. With this last most persons would fail to see the dots even when their sight was assisted with a watchmaker’s magnifying glass.

For making the printing surface from the negative taken through a cross-lined screen, a sheet of copper of the required size and with a polished surface is coated with a solution of “fish glue” containing ammonium bichromate to render it sensitive. The commonest equivalent of fish glue is “seccotine,” which is well known to resemble a gum mucilage in remaining liquid when cold. The fish glue used is specially filtered to ensure a clean coating. The plate is gently warmed to dry the film, which is very thin, and then exposed to light under the negative. After exposure it is washed that the parts not changed by light may be dissolved away, and as what remains is hardly visible, being thin and colorless, the plate is put into a dye solution which stains the image and serves to show whether it is satisfactory. If all is well the plate is then heated to a pretty high temperature, hot enough to burn wood but not sufficiently hot to char the fish glue. This changes the coating into a very hard enamel-like substance, which will not only protect the metal during the etching, but is so much more able to resist the wear of printing than the copper itself, that it may be intact after sixty thousand impressions have been taken under conditions that ten thousand impressions would notably deteriorate the block if of bare copper, and this although the film is as thin as a soap bubble just before it bursts. To etch the plate it is put into a solution of ferric chloride, that is a compound of iron with the maximum amount of chlorine, and which in the presence of copper gives up some of its chlorine to it, and the chloride of iron that remains as well as the chloride of copper that is produced can be washed away. The plate now only needs to be mounted on a suitable wooden block to be ready for the printing press, but the simple process here described is often supplemented by a little further etching of some parts or a little touching up by hand.

Those illustrations that used to be called “copper plates” or “steel plates,” meaning impressions taken from copper or steel plates on which the picture was engraved, were produced by inking the plate all over, wiping the surface of the plate clean so that the ink remained only in the cut out lines and dots, and then taking the impression. These processes are now almost obsolete and their modern representative is called “photogravure.” The polished copper plate for this process has to have the picture put upon it as a “resist,” and it is then etched to form the depressions that are to hold the ink. The resist is a carbon print, and that as we know consists essentially of a film of gelatine that is graduated in thickness according to the darkness of the various parts of the original. But as in this case where the most etching takes place the most ink will be held and there will be the darkest part of the impression, we want the thinnest part of the carbon print to correspond with the darkest part of the subject, and the thickest part of the print with the lightest part of the subject that is, the carbon print must be a negative instead of as usual a positive. The common method is to prepare from the ordinary negative a carbon transparency, which is a positive, and from this to make another carbon print which will be the required negative and will be mounted on the copper plate for development. If this were done and the plate etched, the gelatine film which constitutes the carbon print acting as a resist and regulating the extent of the etching in the various parts of the plate, the image would be etched on the plate as a very shallow relief, but one which would not hold ink when the plate was wiped, for the ink would be wiped out from the shallow depressions just as it might be wiped out from the inside of a saucer. As explained in connection with the previous process the general surface of the metal must be retained, and etching restricted to very small areas of it, forming minute depressions or pits which will hold the ink. The “grain” necessary to produce this result is obtained by covering the surface of the copper plate, after it has been well polished, with minute particles of bitumen or resin, by shaking up the powdered material in a closed box, allowing it to stand for a short time that the coarser particles may settle, and then putting the plate in that the finer particles may rain down or subside upon it. By gently heating the plate, each little particle melts to a drop, and protects that portion of the surface that it covers from the subsequent etching. The “ground” may be laid by other means, as for example by spraying a solution of the substance upon the plate, but the result obtained is much the same. Upon the plate so prepared, the carbon print already described is mounted and developed in the ordinary manner, and when dry the plate is ready for the etching. A solution of ferric chloride is generally used, and its action, which takes but a short time, can be followed by the change of color that it produces of the metal surface. Where the gelatine film of the carbon print is thin, that is in the dark parts of the subject, the etching begins first and therefore finally produces the greatest effect, and as soon as the etching has begun where the thickest parts of the gelatine film lie, corresponding with the brightest parts of the subject, the plate is washed to stop the action, and the gelatine and bitumen are cleaned off by brushing with a solution of carbonate of soda. The plate is now ready to be printed from, unless it needs any touching up, but if many impressions are wanted it is usual to “steel face” it, by electrically depositing upon it a very thin film of the harder metal. The etching is of a very different character from that described in connection with other processes, indeed it may be more graphically described as a mere roughening of the surface of the metal. The roughening is more extensive and deeper where the etching liquid acts for the longest time, and in these parts when the plate is inked and wiped ready for taking the impression, the ink that is retained not only covers a greater proportion of the surface of the plate, but is held in greater quantity in the slightly deeper depressions. The gradation in the print depends upon both these circumstances.

It must not be supposed that all the applications of photography to mechanical printing processes have been described or even referred to in this section, for we have only aimed at giving the general lines upon which such applications are worked out. Each fundamental process is subject to numerous variations to adapt it to the different conditions of the many sorts of subjects that have to be dealt with, and to the results required. For example, photogravure, which seems so essentially a hand process and more suitable for the artist himself than for the mere printer, has been adapted to rapid printing by rotary presses. The subject of photo-mechanical printing is very wide and very technical, and will be briefly referred to again in connection with the photography of color.

PHOTOGRAPHIC TELEGRAPHY

As an addendum to this we may perhaps refer to a matter that seems to demand reference, though only a brief reference, because while it concerns photographs it is not a photographic but an electrical subject, namely the transmission of photographs by means of telegraphy. When we say that it is possible to send a photograph of a person or of an event in ten or fifteen minutes from Paris or Manchester to London or even from a much greater distance, the importance of the achievement may be realized. It means that photographs can arrive from a distance a day earlier than if they had to be carried by messenger, and this day saved may make all the difference between success and failure, as in transmitting the portrait of a criminal or the representation of an event of importance. In 1907, the Daily Mirror in London, L’Illustration in Paris, and the Lokal Anzeiger in Berlin installed apparatus designed for this purpose by Professor A. Korn, and a year later one was set up in Manchester. Since then the telegraphic transmission of photographs has been a regular practice for publishing purposes, and many improvements have been effected, notably by Mr. Thorne Baker.

There is only one current of electricity passing between the transmitting and the receiving station, and all that can be done with it is to increase, diminish, stop, or start it. It is out of the question to deal with the picture as a whole, and it is necessary to “telegraph” as it were just one small part of it at a time. To enable this to be done a suitable copy of the picture is bent upon and round a cylinder which revolves and moves very slightly longitudinally, so that a fixed point held against it would make a long spiral line that would cover practically the whole surface of the picture. By this continuous movement every part of the picture is brought in turn to the same place, and if the cylinder is of metal and the picture is prepared on a sheet of metal foil with a material that will not allow an electric current to pass, then as the cylinder revolves the current passes where the metal foil is bare, but whenever a part of the drawing comes beneath the point the current is stopped. The receiver consists of a similar cylinder covered with a sheet of paper that is so prepared that when an electric current passes through it a dark deposit is produced. The two cylinders are made to revolve at exactly the same rate, and thus all the spaces of bare metal in the drawing are represented at the distant cylinder by the deposit, and if the original is a negative, the distant reproduction is a positive. There are other ways that serve to translate the light and shadow of a picture, piecemeal, into variations of an electric current, and from the varying current to reproduce the picture, but the main principle involved is always analogous to that described.