According to the latest views of physiological chemists, the innumerable fine particles of milk are identified with the caseinogen. For instance, Halliburton regards these particles as caseinogen and nuclein which are obviously not fat-globules; and Stirling unhesitatingly accepts the view of the particulate nature of this proteid, adducing as proof the fact of the absence of all particles from the filtrate of milk through a porous clay-filter.

I used a “Berkefeld” diatomaceous filter (holding 150 c.c.) said to “sterilise” water. On immersing it in milk, and by creating a partial vacuum inside it, there was obtained an exhaustion-filtrate perfectly limpid and free from all particles. Unless sterilised by heat, a vigorous growth of fungi appeared in it, there being abundant nourishment for them, seeing that it contains lactose, chlorides, phosphates, sulphates, calcium, and a proteid (undoubtedly the lact-albumin), but is incapable of clotting with rennet. Its reaction is faintly acid. Caseinogen cannot, then, be held to be “dissolved” in the plasma, in the ordinary sense of a “solution,” else this pressure-filtrate would be coagulable; and only as a last resource should we fall back upon the rather unsatisfactory hypothesis that it may be present as molecules too large to pass through the pores of a diatomaceous filter.

There is in milk undoubtedly, besides what we have been accustomed to call the oil-globules, a very large number of extremely minute particles or granules exhibiting Brownian movement, and it is these that are held to be caseinogen exclusively, while the globules are said to be only fat.