In seeking to account for the transition of certain liquids to rigid jellies three main lines of explanation have been put forward, namely, (1) a micellar theory according to which the process consists in the separation of a dissolved or dispersed substance in the form of interlacing fibrils which are generally supposed to be associated with some of the solvent, any free solvent remaining enmeshed in the network so produced; (2) a theory which assumes a modification of the properties of the solvent in the neighbourhood of the disperse phase so as to produce rigidity of the whole system, without requiring any particular spacial arrangement of the solid units; and (3) an emulsion theory, which regards jellies as emulsions of two partially miscible liquid phases possessing an interfacial tension. Of these three theories , the last-named fails to provide a theoretically satisfactory explanation of the mechanical properties of jellies; it would only account for rigidity when the disperse phase occupied the major part of the total volume, and is not supported by any direct experimental evidence. The micellar theory, on the other hand, is supported in certain cases—for example, in that of sodium stearate in water, where a solid phase separates in the form of tenuous crystals—by direct ultra-microscopic observations, and a similar fibrillar structure is often assumed when no such arrangement is actually visible. It has also been shown that some of the elastic properties of jellies of gelatin and of cellulose acetate are best accounted for by this theory. Within the last few years an important class of reversible jellies, chiefly metallic oxides, has been investigated, and their properties have been extensively studied and described in a series of papers published from the Kaiser-Wilhelm Institut at Dahlem. In one of the earliest of these papers, describing some ultramicroscopic observations of gelatinising ferric oxide sols, it was concluded that the formation of the jelly was not caused by any actual contact between the particles of ferric oxide, which remained more or less uniformly distributed in the liquid, the only apparent difference between the jelly and the sol which gave rise to it being the absence of brownian movement in the former. This observation, if correct, is evidently consistent with the second theory and definitely negatives the first. Further, measurements of the viscosity, at different rates of shear, of a number of sols, some of which are capable of gelatinising, and of suspensions of starch grains in an indifferent organic liquid can be explained by assuming the existence of a layer of liquid adherent to the surface of the particles, its thickness being diminished as the rate of shear increases. Finally, Porter and hedges have pointed out that their observations on the distribution of suspended particles of gamboge under gravity can be expressed in terms of Porter's general formula of osmotic pressure if the effective volume of the particles is taken to be considerably greater than that actually measured. Apart from the possibility that a number of different explanations may be needed to account for the production of various kinds of jellies, it may be noted that, up to the present, direct observation has pointed to a fibrillar structure in some soap jellies and to an absence of structure in certain jellies formed from oxide sols. The object of the present paper is to show that sols which owe their stability to an electric charge may be expected to give rise to jellies when that charge is reduced, and to present some experimental evidence for this view.