years, and had resulted in the discovery of some facts of the highest 
scientific interest and greatest practical importance. These experiments 
were made with very simple apparatus ; the first he used consisting of 
a small jar containing the solution for difi'usion, placed within a larger 
one into which the difi'usion was to take place. The solution in the 
phial is thus allowed to communicate freely with the water in the jar, 
and diffuses itself into the water against its own gravity. The extent 
to which this goes on varies with solutions of different salts, and is 
ascertained, after a given time, by estimating by evaporation or other¬ 
wise, the amount which has passed into the outer jar. In his later 
experiments he simply uses a jar divided into sixteen parts, fourteen 
of which are filled with water, and the whole quantity is made up by 
introducing, by means of a pipette to the bottom of the jar, the proper 
quantity of the solution to be tested. The results arrived at were 
given in detail and illustrated by tables. By this process chemical 
affinity seems to be overcome; but the important fact is, that whilst 
all crystalline substances possess a high but varying power of diffusion, 
such substances as gelatine, albumen, &c., are almost devoid of this 
property, and this has led Mr. Graham to divide bodies into two 
classes in relation to their diffusiveness. He calls the former Crystal¬ 
loids, the latter Colloids. It was found that a crystalloid will diffuse 
in a colloid nearly as well as it will in water, which a colloidal substance 
refuses to do, being obstructed as well by a thin layer as by a large 
mass. Paper sized with gelatine, starch, and a variety of similar 
substances, produces the same results ; but the material which is the 
most useful for this purpose is the recently-introduced “parchment 
paper.” When the separation of a crystalloid from a colloid is effected 
by diffusion through a septuan of this kind, it is called Dialysis. The 
apparatus used in the process is called a Dialyser, and is made by 
stretching a piece of parchment paper over a hoop of gutta percha, and 
retaining it in its place by an outer loop. It is thus like a tambourine, 
and being put to float on water in a suitable vessel, the liquid is poured 
in; it is pervious to crystalline, but impervious to colloidal substances, 
the former passing into the outer liquid, and the latter being retained 
—a separation of the two is in this manner effected. After considering 
some matters of scientific interest connected with the subject, the 
paper concluded with a detail of the applications of dialysis. One 
of the important uses is its employment for the separation of animal 
and vegetable poisons from the organic matter with which they are 
usually combined in toxicological analysis, thus placing in the pos¬ 
session of the chemists a ready, certain, and expeditious method of 
