In the first of these methods, we are so far from being able to calculate the force of the chemical powers called into play, that Berthollet was induced to deny the existence of chemical power in the various phenomena of solution and precipitation of saline substances, and according to him what is called insolubility in a body is merely the result of its strength of cohesion, an entirely physical property.

When the intervention of caloric is required, the effects are still more complicated, as they vary according to the intensity of the heat employed, and the time its action is exerted; besides, the chemical action, when it does take place, is frequently so instantaneous that it is impossible in our present state of science to imagine any means by which it might be measured. In the combination of the three bodies, iodine, bromine and chlorine, with the metals, however, most of these objections cease to exist, or may be easily avoided. As their vapors combine with the metallic surfaces at the ordinary temperature, they are all of them in the same circumstances in that respect; and if the temperature should be required more elevated, the gasiform state of these substances, iodine not excepted, enables us to submit the metals to be experimented upon all at the same time to the same influence. If, therefore, it were possible to reduce the metallic substances into fine powders the particles of which were of the same dimensions, by acting upon them with either of these vapors, an idea might be formed of the affinities which produce their binary compounds by the increased weight acquired by the powders in this process; but the difference which exists in the physical properties of the various metals would preclude the possibility of any near approach to accuracy in this mode of proceeding; but by acting on the polished metallic surfaces, as in the preceding experiments, all the advantages offered by the process with the powders are included, whilst several of the difficulties are removed.

As the film of the compound augments, it undergoes the various changes of color which take place in all transparent films, thus affording a means of ascertaining the absolute thickness obtained in different circumstances, when it would be difficult to detect the slightest difference in weight by means of the most delicate balance. The depth of this coating may be ascertained when either the index of refraction of the compound itself is known, or if the angle of polarized light is given by means of the law discovered by Sir David Brewster, between the tangent of the angle of polarization, and the index of refraction. The most convenient way which occurred to me of performing these experiments, was the employment of a bell-glass within which some iodine is fixed at the top; this apparatus being placed over the metal to be acted on, the experiment may be watched in all its progress, and the action can be retarded or accelerated at pleasure by varying the interval of the iodine from the metal, or by interposing at some distance from its surface a disc of paper so as to cause the vapors of iodine to pass through it. Bromine may be made use of likewise by pouring a few drops of it over some carded cotton, and using it in a similar manner with the iodine. In respect to chlorine, it is most convenient to disengage it slowly by dropping a little sulphuric acid upon some chlorinated lime.

In illustration of the objects of this mode of experimenting, I will aduce some of the results it has given me with various metals. Some of the experiments below were performed before I had the idea of watching the progress of the combination through a transparent medium; they are therefore less exact than they might otherwise have been: but I have preferred stating them as I had inserted them in my note-book before I had conceived any idea as to their probable utility in the elucidation of chemical affinity, and when I intended them for other purposes, which I shall hereafter explain.

Iodine with Silver and Copper.

1st change.Silver—pale gold.
"Copper—assumes a darker red.
"Silver—blue.
2nd do.Copper—blue.
"Silver—white.
3rd do.Copper—white.
"Silver—yellow.
4th do.Copper—yellow more extended than on the silver.
"Silver—Orange.
5th do.Copper—Red.
"Silver—blue, bluish-red.
"Copper—red, with a tinge of green on some parts.
"Silver—greenish blue.
"Copper—red, tinged with green.
"Silver—green.
"Copper—orange.
"Silver—yellowish green.
"Copper—orange tending to red.
"Silver—yellowish green.
"Copper—orange-red.
"Silver—red.
"Copper—dull green.
"Silver—red.
"Copper—green.
"Silver—deep green.
"Copper—dull red.

Bromine with Silver and Copper.

5th change.Copper—sensibly darkened.
"Silver—unchanged.
"Copper—deep red.
"Silver—unchanged.
"Copper—red, blue.
"Silver—pale gold.
"Copper—white, orange of the 2d order.
"Silver—yellow.
"Copper—green of the 1st order, red 3rd order.
"Silver—blue.

Chlorine with Silver and Copper.

The affinity of chlorine with silver is much inferior to that which it possesses for copper.