Now in a gelatine sol we have the necessary conditions for a system in which the compressed water bears an unusually large ratio to the total, owing to the enormous surface developed by the minute particles of the disperse phase (amicrons) and to the unusually wide zone of compression surrounding each particle caused by the strongly hydrophile nature of gelatine. It should be pointed out that these zones of compression do not involve any abrupt transition from the zone of non-compression, the layer nearest the particle is under the greatest pressure, and the concentric layers under less and less pressures, the actual compression being thus an inverse function of the distance from the particle. Now if there be a gradual increase in the concentration of the sol, the time will come when these zones of compression begin to come in contact, and the system will then show a considerably increased viscosity. With further increase in concentration the zones of compression will overlap throughout the system, and when the layers under considerable pressure are thus continuous, the whole system will acquire a rigidity much greater than water and approaching that of a solid body. This is a gelatine gel, or "jelly." With increasing concentration the jelly becomes increasingly rigid, and if it be eventually dried out under suitable conditions it forms what is practically solid body--gelatine--which, however, still contains from 12 to 18 per cent. of water. It will be clear that, in the case of gelatine jellies (_e.g._ of 3-10 per cent. strength), an increase in temperature will cause an increase in the kinetic energy of the particles and effectively reduce the zones of compression. Indeed, they may be reduced to such an extent that they are no longer in contact, and the rigidity due to the continuous contact of the layers of great compression will then disappear; as we say usually, the jelly melts. On cooling, the decreased kinetic energy of the water molecules results in the return of the state of compression, with rapidly increasing viscosity and eventual gelation; as we say usually, the jelly sets. Neither of these changes takes place at a definite temperature (like a melting-point), and in "melting" (solation) or in "setting" (gelation) the temperature-viscosity curve is quite continuous. By various arbitrary devices, however, approximate melting and setting points may approximately be determined. The results also vary somewhat with the concentration of the gel or sol. Gels between 5 and 15 per cent. strong melt about 26°-30° C. and set at 18°-26° C. On this view, we must regard a gelatine gel as a continuous network of water under great compression, and in this network are zones of still greater compression, which surround the particles of the disperse phase--the gelatine itself, and zones of less compression which in a weak gel, at any rate, have a compression equal to or much the same as the normal state of compression in water.