In the first division of this work, all the details of the original processes are given with considerable minuteness, and the vignette heading to that section exhibits all the apparatus required for even the improved modern practice.

The following remarks by M. Daguerre on polishing and preparing the plates, from the Comptes Rendus of March 13, 1843, should be carefully attended to, as the preliminary process upon which the success of every subsequent state depends.

" Since the publication of my process, I have not been able to occupy myself much with it. The investigations to which I devoted myself have been in an entirely new direction, and the experiments which they required were analogous with the preceding ones, only inasmuch as they were made on a metallic plate. However, I have lately been so much struck with the unequal results which the impressions generally present—even those of persons who are especially occupied with them—that I determined to seek some means of remedying this serious inconvenience, which I attribute to two principal causes.

" The first relates to the operation of polishing, which it is physically impossible to effect without leaving on the surface of the plate traces of the liquid and of the other substances used in this operation : the cotton alone which is employed, however clean it may be, is sufficient to leave a film of dirt on the silver. This first cause constitutes a very great obstacle to the success of the impression, because it retards the photogenic action by preventing the iodine from coming in direct contact with the silver.

"The second consists in the alterations of the temperature of the air with which the plate is in contact, from the first operations to the mercurial operations. It is known that when a cold body is surrounded with warmer air it condenses its moisture. To this effect must be attributed the difficulty which is experienced in operating in a humid medium, especially when we come to the mercurial operation, which requires, to raise a suitable vapour, a temperature of 122° F.

" This vapour, which first heats the air contained in the apparatus, produces on the metal a dew which weakens the image. It is very evident that this humid layer is very injurious; since if, for example, the plate, on leaving the camera obscura, be breathed on two or three times, the mercurial vapour can no longer cause the impression to appear.

" The water which is condensed, even at the slightest difference of temperature between the surface of a body and the surrounding air, contains in solution, or in suspension, a non-volatile substance, which might be called atmospheric dust; and as soon as the equilibrium of temperature is established between the air and the surface of the body, the humid vapour which was condensed on it is volatilized, and depositing on it the dust which it contains, goes on to be re-saturated in the air with a fresh quantity of this impure substance.

" In order as much as possible to neutralize this effect, the temperature of the plate may be kept higher than that of the surrounding air, during each of the operations. But it is impossible to cause this heat to reach to 122° F., in order for it to be of the same temperature as the vapour of mercury, since, if the plate be exposed to that degree of heat after the operation of light in the camera obscura, the image will be altered.

" I first tried to absorb the humidity of the air in the mercurial box by the usual means, such as lime, etc. ; but these means are insufficient, and only complicate the process, without giving a good result. Another means which has been proposed consists in vapourizing the mercury under the pneumatic machine ; by this process, truly, the dew on the plate is avoided, but the pressure of the air, which is indispensable to the impression, is suppressed. The results thus obtained, also, are always wanting in purity.

"The following is the process at which I have stopped, because it is very simple, and because it obviates the two inconveniences above mentioned; that is to say, it frees the silver as much as possible from all dirt or dust, and neutralizes the humidity produced by the elevation of temperature in the mercurial box. By the first of these two effects it increases the promptitude, and by the second it renders the lights much whiter (especially by the application of M. Fizeau's chloride of gold) : these two effects are always certain. The promptitude given by this process is to that hitherto obtained as 3 to 8 : this proportion is accurate.

" This process consists in covering the plate, after having polished it, with a layer of very pure water, and heating it very strongly with a spirit-lamp, and in afterwards pouring off this layer of water in such a manner that its upper part, where the dust which it has raised floats, does not touch the plate.

" It is necessary to have a frame of iron wire of the size of the plate, having at one of its angles a handle, and in the middle, on the two opposite sides, two small cramp-irons, to retain the plate when it is inclined. After having placed this frame on a horizontal plane, the plate is placed on it, which is covered with a layer of very pure water, and putting as much water as the surface can retain. The bottom of the plate is afterwards very strongly heated, and very small bubbles are formed at the surface. By degrees these bubbles become larger, and finally disappear ; the heat must be continued to ebullition, and then the water must be poured off. The operator should commence by placing the lamp under the angle of the frame where the handle is ; but, before removing the frame, this angle must be very powerfully heated, and then, by gradually removing it by means of the handle, the water immediately begins to run off. It must be done in such a way that the lamp shall follow, under the plate, the sheet of water in its progress, and it must be only gradually inclined, and just sufficient for the layer of water, in retiring, not to lose in thickness ; for, if the water were (fried up, there would remain small isolated drops, which, not being able to flow off, would leave on the silver the dust which they contain. After that, the plate must not be rubbed : very pure water does not destroy its polish.

" This operation should be performed only just before iodizing the plate. Whilst it is yet warm, it is placed in the iodizing box, and, without allowing it to cool, it is submitted to the vapour of the accelerating substances. Plates thus prepared may be kept one or two days (although the sensibility diminishes a little), provided that several plates be placed opposite to one another, at a very short distance apart, and carefully enveloped to prevent change of air between the plates.

" The plates cannot be too well polished. It is one of the most important points to obtain a fine polish ; but the purity often disappears when substances which adhere to the surface of the silver are used,—such as the peroxide of iron, which has been very generally made use of for giving the last polish. This substance, indeed, seems to burnish the silver, and to give it a more perfect polish ; but this polish is factitious, since it does not really exist on the silver, but in fact on a very fine layer of oxide of iron. It is for this reason that there is required for polishing them a substance which does not adhere to the silver; pumice, which I recommended at the commencement, leaves less residue.

" As regards the liquid to be employed : in the first operations nitric acid of five degrees must be employed, as I stated in the first instance ; but for the last operations it must be reduced to one degree.

" The polishing with oil and the heating may be suppressed.

" I take the opportunity afforded by this communication to lay before the Academy the following observations, which I owe to experience :— " The layer produced by the descending vapours of the iodine and of the accelerating substances forms with silver a more sensible compound than is obtained with the ascending vapours. I make this observation only to lay down a fact, for it would be difficult to employ descending vapours, on account of the dust which might fall during the operation, and from stains.

" The resistance which light experiences in passing through a white glass is well known. This resistance is even greater than it appears, and may be attributed not only to the dust which is left on the glazing in cleaning it, but also to that which is naturally deposited on it. The object-glass of the camera obscura is certainly in the same case. To ascertain this, I put the object-glass in cold water, which I boiled ; I knew that it was impossible to remove it without the sides. This operation had, therefore, no other object than to raise the temperature of the glass to 212° F. C., and I then immediately poured on the two sides of the object-glass very pure boiling water to remove the dust. By operating directly with the object-glass, thus cleansed, I still further increased the promptitude. This means presents too many difficulties to be put in practice ; only care should be taken to clean the object-glass every day.* " The atmospheric dust, which is the scourge of the photogenic images, is, on the contrary, favourable to images which are obtained by contact or at a very short distance. To be convinced of this, we have only to clean the two bodies which we wish to put in contact with the boiling water, as I have just indicated, and to keep them both at the same temperature as the air; there will then be no impression, which evidently proves that these images have no relation with the radiation which gives photographic images".

* Professor Stokes has recently confirmed the truth of this by some very conclusive experiments.