I say _permanent_ juices with intent, in order to exclude the albumen, which being coagulable at the lowest cooking temperature is not permanent. It is one of that class of bodies to which Graham gave the name of colloids (glue-like), such as starch, dextrin, gum, &c., to distinguish them from another class, the crystalloids, or bodies that crystallise on solidification. The latter diffuse and pass through membranes by diosmosis readily, the colloids very sluggishly. Thus a solution of Epsom salts diffuses seven times as rapidly as albumen, and fourteen times as rapidly as caramel. The difference is strikingly illustrated by the different diffusibility of a solution of ordinary crystalline sugar and that of barley-sugar and caramel, the latter being amorphous or formless colloids that dry into a gummy mass when their solutions are evaporated, instead of forming crystals as the original sugar did. Some of the juices of meat, as already explained, exist between its fibres, others are within those fibres or cells, enveloped in the sheath or cell membrane. It is evident that the loose or free juices will be extracted by simple diffusion, those enveloped in membranes by exosmosis through the membrane. The result must be the same in both cases; the meat will be permeated by the water, and the surrounding water will be permeated by the juices that originally existed within the meat. As the rate of diffusion—other conditions being equal—is proportionate to the extent of the surfaces of the diverse liquids that are exposed to each other, and as the rate of diosmosis is similarly proportioned to the exposure of membrane, it is evident that the cutting-up of the meat will assist the extraction of its juices by the creation of fresh surfaces; hence the well-known advantage of mincing in the making of beef-tea.