CLASSIFICATION OF VEGETABLE PROTEINS 77 



solutions, many are not precipitated until their solutions are nearly 

 saturated with this salt. The distinction, therefore, which is at present 

 almost universally made between globulins and albumins, based on 

 the complete precipitation of the former at half saturation of their 

 solutions with ammonium sulphate, cannot be applied to the vege- 

 table globulins. 



The animal globulins are all coagulated by heating their solutions 

 to various temperatures. Most of the seed globulins are only imper- 

 fectly coagulated by heating their solutions even to boiling, and some 

 of them can be thus heated for a very long time without showing any 

 apparent change. 



A number of the vegetable globulins can be readily obtained in 

 crystals ; some of them crystallise with difficulty, and nearly all of those 

 which do not crystallise are obtained in the form of minute spheroids. 

 Crystallisation can be effected in various ways. Globulins which crys- 

 tallise readily are usually precipitated by dialysis in well-formed crystals. 

 Such are edestin from the hemp-seed, excelsin from the Brazil-nut, and 

 the globulins of the squash-seed, the flax-seed, the oat-kernel and the 

 castor-bean. The globulin of the castor-bean is commonly obtained 

 by dialysis in the form of spheroids, but frequently mixtures of 

 spheroids and crystals result, although sometimes completely crystalline 

 separations occur. Phaseolin, from Phaseolus vulgaris^ frequently pre- 

 cipitates on dialysis in the form of minute spheroids, mixed with a few 

 octahedral crystals which sometimes are of unusual size and very per- 

 fect form, but completely crystalline preparations of this globulin have 

 never yet been obtained. Crystals of such globulins as crystallise 

 easily can generally be obtained by diluting their sodium chloride solu- 

 tions with water heated to 50 or 60 until a slight turbidity forms. 

 After warming the diluted solution until this turbidity disappears and 

 then allowing it to cool slowly, the protein separates in well-developed 

 crystals. 



Schmiedeberg (457), who obtained crystals of the globulin from 

 the Brazil-nut, by treating its solution with magnesia and evaporating 

 it slowly, considered the crystals of this substance to be the magnesium 

 salt of the protein. This view of the nature of the crystals of this 

 globulin has been generally accepted, and is frequently found, in the 

 current literature dealing with this protein, although Osborne (302), 

 many years ago, showed that much more perfectly crystalline prepara- 

 tions of this globulin than are described by Schmiedeberg are obtained 

 by simply dialysing its faintly acid saline solutions in running water. 

 These crystals are unquestionably salts of the globulin with the acid 



