IN STUDYING QUESTIONS OF CHEMICAL MECHANICS. 295 
emanates individually from all, if the liquid is chemically homo- 
geneous, or only from a certain number of them, distributed 
uniformly among the others, if it is not. This proposition will 
be sufficiently evident to geometricians, by the very symmetry 
which I have just expressed. It is verified experimentally in 
all its physical consequences by a multitude of results, one of 
the most evident of which is the exact proportionality observed 
between various thicknesses of one and the same active liquid, 
and the magnitude of the total deviations which it impresses on 
the plane of polarization of the same luminous ray, whether em- 
ployed in continuous or discontinuous columns. The identity of 
the ray is necessary, because the effect produced is generally 
unequal on rays of different refrangibilities. In all diaphanous 
liquids hitherto observed, with the exception of those of which 
tartaric acid is one of the elements, the deviations produced by 
one and the same thickness in the planes of polarization of 
different simple rays are very nearly, although not exactly, reci- 
procal to the squares of the lengths of their fits, which renders 
them relatively stronger for the most refrangible rays. But in 
liquids of which tartaric acid forms a part, especially where it is 
not very intimately combined with other bodies, the dispersion 
of the planes of polarization follows wholly different laws, which 
may even be varied almost at will by modifying the temperature 
of the system, or the proportion of its elements, or the nature of 
the substances in contact with the acid. 
4. The permanence of the optical action exercised by the 
liquid masses in a state of repose, as well as of relative mobility 
of their particles, which is a fundamental condition of the pre- 
ceding principle, is verified, so to speak, without preparation, 
in the natural state in which perfectly fluid masses are observed. 
For we may very well presume that, in the midst of the external 
circumstances by which they are surrounded, the particles 
composing them are as little fixed as those of gases. But we 
may prove this fundamental fact by means of the following 
apparatus. In a large tube terminated by thin glasses with 
parallel faces, I inserted laterally a metallic rod, bearing at 
its inner extremity a plane metallic diaphragm, having in its 
centre an annular opening; so that on turning this rod in va- 
rious directions, by its projecting extremity, I was able suddenly 
to agitate the liquid with which the tube was filled. But these 
