310 BIOT ON THE EMPLOYMENT OF POLARIZED LIGHT 
upon light; this is to diffuse them by dilution in an inactive 
liquid which does not modify them chemically; and this will 
also offer the advantage of evidencing the same kind of action 
in a great number of substances usually solid and opake, which - 
cannot be liquefied by heat without the risk of altering them, or 
which, if they are transparent in a solid state, are at the same 
time crystallized, which developes in them phzenomena of polar- 
ization peculiar to this kind of aggregation, by which the effects 
of their molecular power are modified, or even completely dis- 
similated. 
14. For greater simplicity, let us at first admit that the sol- 
vent employed does not enter into actual combination with the 
constituent groups of the active substance, and that it offers to 
them only a space in which they can distribute themselves uni- 
formly. Let E be the weight of the inactive liquid employed, P 
the weight of the active substance mixed in it, which may be 
indifferently liquid or solid; and for shortness let us make 
— P . 
See ae 
« will represent the ponderable proportion of the active sub- 
stance contained in each unit of weight of the mixture. Let us 
designate, as before, by [«] the deviation which this substance 
would produce on a simple ray selected as type, if it could be 
observed isolated, in the state of disaggregation, through the 
unit of thickness, and under a density equal to 1. Then, 
the density of the mixture being 8, and the observation being 
made through a tube of a length /, the total deviation which it 
would produce on the type ray, under these circumstances, if 
it were entirely formed of the active substance, would be [a] /8, 
as before. But the deviation « really produced will be less than 
this, in the proportion of « to 1; that is to say, proportionally to 
the relative mass of the active groups actually contained in the 
total mass of the length /, under the density 6. Thus it will be 
only [a]/e8. Then designating it by «, we shall have, by elimi- 
nating [«] from the equation, 
ot a(P+E 
(a) =f sor fa eo ee 
The second member of this new equation is composed wholly 
of observable quantities. On calculating it according to experi- 
ment, for different values of its elements /, «, 8, and a, it ought 
