106 PROTEIDS. 



differently with reference to passing through membranes ; those 

 which pass through readily Graham denominated crystalloids; 

 while those that pass not at all or with difficulty, he called col- 

 loids, a term which does not necessarily imply that the substances 

 which are called by that name do not crystallize, for, as we have 

 seen, egg-albumin does crystallize, though salts are crystalloids 

 and egg-albumin a colloid. This principle is the same as 

 dialysis, or the separation of crystalloids from colloids in a 

 dialyzer, an apparatus constructed like the osmometer, already 

 described. 



As already stated, proteids, except proteoses and peptones, are 

 not diffusible, or, to put the statement affirmatively, are colloids 

 i. e., they do not readily pass through animal membranes ; and this 

 principle of dialysis is employed to separate them from crystal- 

 loids, such as sugar and salts. This may be demonstrated by 

 putting a saline solution of albumin and globulin, as, for instance, 

 blood-serum, in a dialyzer, the vessel outside containing distilled 

 water ; the salts diffuse, leaving the albumin and globulin behind. 

 The albumin, being soluble in water, remains in solution, and, being 

 colloid, does not diffuse ; the globulin, requiring the salts to hold 

 it in solution, is precipitated because the salts have diffused. 



Proteoses are diffusible, but less so than peptones ; protoproteose 

 more than deuteroproteose, and this more than heteroproteose, so 

 that the order of diffusibility would be peptones, protoproteose, 

 deuteroproteose, and heteroproteose. It is to be borne in mind 

 that diffusibility is a relative term, and that when peptones are 

 said to be readily diffusible the idea intended to be conveyed is, 

 that when compared with other proteids they exhibit this prop- 

 erty. If, however, they are compared with salts, their diffusibility 

 is low. 



The explanation formerly given to account for the non-diffusi- 

 bility of proteids was that they were not crystallizable ; but since 

 the discovery of the fact that some of them do crystallize, this 

 explanation is abandoned, and for it is substituted that of the 

 great size of their molecule. The molecular weight of some of 

 these proteids has been determined, and this confirms the theory 

 just enunciated: thus peptone, very diffusible, has a molecular 

 weight of 400 or less ; protoproteose, less diffusible, of 2467 to 

 2600 ; and deuteroproteose, still less diffusible, of 3200. 



Precipitation. As a class, proteids in solution are precipit- 

 able by certain reagents, of which the number is considerable. 

 Some of the principal precipitants are : Nitric acid, picric acid, 

 acetic acid with potassium ferrocyanid, acetic acid with excess of 

 sodium or magnesium sulphate or sodium chlorid, when boiled 

 with the solution of the proteid, mercuric chlorid, silver nitrate, 

 lead acetate, tannin, and alcohol. Tannin and alcohol will pre- 

 cipitate peptone, but most of the others will not. 



