294 BIOT ON THE EMPLOYMENT OF POLARIZED LIGHT 
ventured upon it; and in fact I do not hesitate to say, that, 
thus regarded, it could not have been solved. For the experi- 
ments which I shall relate will prove that definite combinations, 
proceeding by intermittent proportions of simple multiples, are 
wholly exceptional cases in the series of chemical reactions ; that 
they are isolated, so to speak, from the continuity of those re- 
actions by the intervention of forces or of circumstances which 
interrupt that continuity. So that, to endeavour to ascend from 
thence to the general mechanism of the material systems from 
which these intermittent combinations are separated, or which 
are transformed into them by a sudden convulsion of all their 
parts, would be to place ourselves in a more unfavourable posi- 
tion than that of the geometrician who should seek to deduce 
the nature of a curve or of a surface from the mere knowledge 
of its individual points, without any notion of the general con- 
ditions which characterize the continuity of its course. Now it 
is this character of continuity which I proceed to establish in 
the phenomena I propose to consider. 
3. For this purpose, conformably to all physical and chemical 
inductions, I shall consider bodies permanent, of sensible dimen- 
sion, as composed of disjointed particles, having, in their inap- 
preciable minuteness, individual configurations and properties, 
and kept at a distance by forces of whose nature we are igno- 
rant, but whose existence is manifested by the property which 
all sensible masses have of contractibility and dilatation, within 
certain limits, without decomposition. Taking, then, such a 
system in the state of relative mobility of its parts which cha- 
racterizes perfect fluidity, I shall apply to it a principle which 
may be established by mathematical demonstration, and verified 
in all its deductions by experiment; the following is the pro- 
position. 
When a perfect liquid, physically homogeneous and of uni- 
form temperature, being traversed perpendicularly by a ray po- 
larized in a given plane, impresses on the plane of polarization 
of this ray deviations of constant direction, which preserve the 
same direction and the same amount, in the relative state of agi- 
tation or of repose of the particles, such an effect could only 
be produced mechanically by an action which the geometrical 
elements of the liquid mass would exercise, in all directions, 
with an equal energy. Its dissymmetry around the normal can 
only result from a molecular power peculiar to the imagined 
material groups which constitute this mass,—whether this power 
