Ix the previous paper, one of u s has examined three special cases I,; apparent solution, namely, those of mercury, antimony, and arsenic sulphides. These liquids, which in many ways resemble ordinalmy colloidal solutions, were found to present a Peries passing from matter in a state of subdivision not too fine to allow of its observation under tlie microscope, to particley so fine as to diffuse in the manner of ordinary solutions. Arsenic sulphide, indced, was found capable in itself of presenting us with three diffcreiit types of solution, distinguished i n the paper as arsenious sulphide a, /I, and n/. Arsenic a is made u p of particles just visible under the microscope, arsenic /? contains 110 visible particles, whilst arsenic contains particles so minute as t o be diffukible. The case of mercury sulphide was specially interesting, as the sulphide, after precipitntion, is easily dissolved i n sulphnrett cld hydrogen water, and i n this solution the vibrating particles are clearly discernible under the microscope (see previous paper). We have here what appears to be a halfway stage towards true solution. The suggestion is a t once forced upon our minds that the state of pseudo-solution would gradually merge into that of true solution by carrying the subdivision farther and farther, till finally, perhaps, we may have the substance dissociated into ions. To trace the steps somewhat further would obviously be a task of no little interest. And on asking ourselves what solutions are most likely to fcrnish stages i n the phenomena of solution between those of arsenic y and those of true sdution, the case of ordinary colloidal solutions at once suggests itself. Thus t h e investigation into the nature of the sulphide solutions just described had not proceeded far before it was determined t o follow it u p with Home experiments on a variety of collo'idal solutions. Some of these have now been examined, and we have also obtained interesting evidence with regard to the transition from colloidal to crystalloidal solution, part of which we shall lay before t h e Society in thiR paper. Before describing our own experiments, it will be well to give a short account of previous work on the subject. Graham, i t will be remembered, supposed t h e non-dialysability of colloids to be due t o the size of their molecules, which he considered too large to pass through the pores of the membrane. A certain mystery has long hung about the facts of colloidal solution,