Obviously these artificial phenomena have nothing to do with living substance. Yet if we grind up a living muscle with some sand in a mortar we do destroy something. The muscle could be made to contract, but after disintegration this power is lost. We have certainly destroyed a structure, or mechanism, of some kind. But, again, the paste of muscle substance and sand still possesses some kind of vital activity, for with certain precautions it can be made to exhibit many of the phenomena of enzyme activity displayed by the intact muscle fibres, or even the entire organism. Mechanical disintegration, therefore, abolishes some of the activities of the organism, but not all of them. If, however, we heat the muscle paste above a certain temperature, the residue of vital phenomena exhibited by it are irreversibly removed, so that heating destroys the mechanism. This we can hardly imagine to be the case (within ordinary limits of temperature at least) with a physical mechanism, but again a mechanism which is partly chemical might be so destroyed. We see, then, that protoplasm possesses a mechanical structure, but that all of its vital activities do not necessarily depend on this structure. The full manifestation of these activities depends on the protoplasmic substance possessing a certain volume or mass, and also on a certain chemical structure. If living protoplasm has a structure, and is not simply a mixture of chemical compounds, what is it then? Two or three physico-chemical concepts are at the present time very much in evidence in this connection. When the substances known as _colloids_ were fully investigated by the chemists, much attention was paid to them by the physiologists, so that life was called “the chemistry of the colloids,” just as after the investigation of the enzymes it was called the “chemistry of the enzymes,” and when the discovery of the relative abundance of phosphorus in cell-nuclei and in the brain was discovered, it was called the “chemistry of phosphorus.” Colloids (_e.g._ glue) are substances that do not readily diffuse through certain membranes, in opposition to crystalloids (_e.g._ solution of common salt) which do readily so diffuse. They form solutions which easily gelatinise reversibly, that is, can become liquid again (glue); or coagulate irreversibly, that is, cannot become liquid again (albumen); which have no definite saturation point; which have a low osmotic pressure (and derived properties), etc.; and the molecules of which are compound ones consisting of combinations of the molecules of the substance with the molecules of the solvent, or with each other, that is, they are molecular aggregates.