AND THE COMPLEMENTARY REFLECTED. 155 



632. This important proposition I prove in the following way : I take a plate, A G, 

 Jig. 99, three inches by four, and by partially screening its surface, while in the act of 

 iodizing, with a proper piece of flat glass, I produce upon it five transverse bands, b, 

 c,d,e,f; the fifth, f, which has been longest exposed, is of a pale lavender colour; the 

 fourth, a bright blue ; the third, a red ; the second, a golden yellow ; and the first, un- 

 iodized metal ; the object of this arrangement being to expose, at the same time and 

 on the same plate, a series of films of different colours and of different thickness, and to 

 examine the action of the rays impinging on them, and the rays reflected by them. 



633. Having prepared a second plate, B, and iodized it uniformly to a yellow, I 

 deposited it in the camera, and now placing the first plate, A G, so that the rays 

 coming on it through the window from the sky shall be specularly reflected to the 

 object-glass of the camera, and the image of A G form upon B, I allow the exposure to 

 continue until the yellow of A G is beginning to turn brown ; then I shut the camera 

 and mercurialize both plates. 



634. In consequence of what has been said (625), it will be readily understood, that 

 of the bands on A G, the first one, which is the bare metal, does not whiten in the mer- 

 cury vapour ; the second, which is yellow, mercurializes powerfully ; the third, which 

 is red, is less affected ; the fourth, which is blue, still less ; and the fifth, which is lav- 

 ender, hardly perceptibly. 



635. But the changes on B, which have been brought about by the rays reflected 

 from A G, are precisely the converse; the band, which is the image of b, is mercurial- 

 ized powerfully ; that of c is untouched, and absolutely black, d faintly stained, e whiten- 

 ed, and/ mercurialized, but little less than b. 



636. It follows from this, that a white stripe on B corresponds to a black one on 

 A G, and the converse ; and for the depth of tint of the intermediate stripes, those of 

 the one are perfectly complementary to the corresponding ones of the other. 



637. By the aid of these results, we are now able to give an account of the varia- 

 bility of sensitiveness in photogenic preparations ; the yellow iodide of silver is exces- 

 sively sensitive, because it absorbs all the chemical rays that can disturb it, while the 

 lavender is insensitive, because it reflects them. Under this point of view, sensitive- 

 ness, therefore, is directly as absorption, and inversely as reflexion. 



638. The superiority of DAGUERRE'S preparation over common sensitive paper may 

 now be readily understood. It absorbs all the rays that can affect it, but the chloride 

 of silver, spread upon paper, reflects many of the active rays. The former, when pla- 

 ced in the camera, gives rise to no reflexions that can be injurious ; the latter fills it 

 with active light, and stains the proof all over. Hence the Daguerreotype has a sharp- 

 ness and mathematical accuracy about its lines, and a depth in its shadows, which is 

 unapproachable by the other. Moreover, the translucency of the white chloride of sil- 

 ver, as well as its high reflecting power, permits of particles lying ont of the lines of 

 light being affected, the luminous agent being diffused in the paper. 



639. The fact, therefore, that a given compound remains unchanged even in the 

 direct rays of the sun, is no proof that light cannot decompose it ; it may reflect or 

 transmit the active rays as fast as it receives them. It results from this, that optical 



