their dilution a tri-nitro or di-nitro cellulose is said to be formed, one of which is the explosive gun-cotton, insoluble in ether and alcohol, whilst the other, though inflammable, is readily soluble in a mixture of these two solvents. When collodion is poured on a glass plate it leaves on drying a hard transparent film which under the microscope is slightly reticulated. Before drying, the film is gelatinous and per fectly adapted for holding in situ salts soluble in ether and alcohol. Where such salts are present they crystallize out when the film is dried, hence such a film is only suitable where the plates are ready to be immersed in the silver bath. As a rule, about five grains of the soluble cotton are dissolved in an ounce of a mixture of equal parts of ether and alcohol, both of which must be of low specific gravity, "725 and 805 respectively. If the alcohol or ether be much diluted with water the cotton (pyroxylin) precipitates, but, even if less diluted, it forms a film which is " crapey " and uneven. Such was the material with which Le Gray proposed to work, and which Archer actu ally brought into practical use. The opaque silver plate with its one impression was abandoned ; and the paper support of Talbot, with its inequalities of grain and thick ness, followed suit, though not immediately. When once a fine negative had been obtained with collodion on a glass plate the image showing high lights by almost complete opacity and the shadows by transparency (as was the case, too, in the calotype process) any number of impressions could be obtained by means of the silver-printing process introduced by Fox Talbot, and they were found to pos sess a delicacy and refinement of detail that certainly eclipsed the finest print obtained from a calotype nega tive. To any one who had practised the somewhat tedious calotype process, or the waxed-paper process of Le Gray with its still longer preparation and development, the advent of the collodion method must have been extremely welcome, since it effected a saving in time, money, and imcertainty. The rapidity of photographic action was much increased, and the production of pictures became possible to hundreds who previously had been excluded from this art-science by force of circumstances. We can merely give an outline of the procedure, referring the reader for further information to the manuals of photography. A glass plate is carefully cleaned by the application of a detergent such as a cream of tripoli powder or spirits of wine (to which a little ammonia is often added), then wiped with a soft rag, and finally polished with a silk handkerchief or chamois leather previously freed from grease. A collodion containing soluble iodides and bro mides is made to flow over the plate, all excess being drained off when it is covered. A good standard formula for the collodion may be taken to be as follows, 55 grains of pyroxylin, 5 oz. of alcohol, 5 oz. of ether ; and in this liquid are dissolved 2J grains of ammonium iodide, 2 grains of cadmium iodide, and 2 grains of cadmium bromide. When the collodion is set, i.e., when it is in a gelatinous condition, the plate is immersed in a bath of nitrate of silver a vertical form being that mostly used in Britain, whilst a horizontal dish is used on the Continent a good formula for which is 350 grains of silver nitrate with 10 oz. of water. The plate is steadily lowered into this solu tion without pause, and moved in it until all the repellent action between the aqueous solution of the silver and the solvents of the collodion is removed, when it is allowed to rest for a couple of minutes, after which period it is taken out and placed in the dark slide ready for exposure in the camera. After undergoing proper exposure the plate is withdrawn, and in a room lighted with yellow light the developing solution is applied, which originally was a solu tion of pyrogallic acid in water restrained in its action by the addition of acetic acid. One of the old formula; 825 employed by Delamotte was 9 grains of pyrogallic acid, 2 drachms of glacial acetic acid, and 3 oz. of water. The image gradually appears after the application of this solu tion, building itself up from the silver nitrate clinging to the film, which is reduced to the metallic state by degrees. Should the density be insufficient a few drops of nitrate of silver are added to the pyrogallic-acid solution and the developing action continued. In 1844 Hunt introduced another reducing agent, which has continued to be the favourite down to the present time, viz., ferrous sulphate. By its use the time of neces sary exposure of the plate is reduced, and the image de velops with great rapidity. A sample of this developing solution is 20 grains of ferrous sulphate, 20 minims of acetic acid, with 1 oz. of water. This often leaves the image thinner than is requisite for the formation of a good print, and it is intensified with pyrogallic acid and silver. There are other intensifies used to increase the deposit on a plate by means of mercury or uranium, followed by other solutions to still further darken the double salts formed on the film ; but into these it is not necessary to enter here. Such intensifying agents have to be applied to the image after the plate is fixed, which is done by a concentrated solution of hyposulphite of soda or by cyanide of potassium, the latter salt having been first introduced by Martin and Gaudin in 1853 (La Lumiere, 23d April 1853). Twenty-five grains of cyanide of potassium to one ounce of water is the strength of the solution usually em ployed. The reaction of both these fixing agents is to form with the sensitive salts of silver double hyposulphites or cyanides, which are soluble in water, not, as is often considered to be the case, to merely dissolve the silver salt itself. It may be well to remark that the utility of bromides in the collodion process seems to have been recognized in its earliest days, Archer (1852) and Bingham (1850) both mentioning it. We notice this, since as late as the year 1866 a patent-right in its use was sought to be enforced in America, the patent being taken out by James Cutting in July 1854. Positive Pictures by the Collodion Process. In the infancy Positive of the collodion process it was shown by Mr Home that collodion a negative image could be made to assume the appearance liro of a positive by whitening the metallic silver deposit. This he effected by using with the pyrogallic acid developer a small quantity of nitric acid. A better result was obtained by Mr Fry with ferrous sulphate and ferrous nitrate, whilst Dr Diamond gave effect to the matter in a practical way. Mr Archer used mercuric chloride to whiten the image. To Mr Hunt, however, must be awarded the credit of noticing the action of this salt on the image, in his paper in the Philosophical Transactions of 1843. The whitened picture may be made to stand out against black velvet, or black varnish may be poured over the film to give the necessary black background, or, as has been done more recently, the positive pictures may be produced on japanned iron plates (ferrotype plates) or on japanned leather. This process is still practised by some photographers, and from the number of ferrotype plates sold the number of portraits taken by it must be still very large. Moist Collodion Process. From what has been stated Moist above it will be seen that for the successful working of the collodion collodion process it was necessary that the plate should P ro ess be exposed very shortly after its preparation ; this was a drawback, inasmuch as it necessitated taking a heavy equipment into the field. In May 1854 Messrs Spiller and Crookes published in the Philosophical Magazine a process whereby they were enabled to keep a film moist (so as to prevent crystallization of the silver nitrate) several days, enabling plates to be prepared at home, exposed in the field, and then developed in the dark room. The plate XVIII. 1 04
