ISOLATION AND PREPARATION 15 



by the previous treatment, and the different proteoses separated from 

 each other by the methods commonly employed for effecting such a 

 separation of proteoses of animal origin. 



The complete and quantitative separation of globulins, albumins, 

 and proteoses is a matter of difficulty. By dialysis heteroproteose 

 may be precipitated with the globulins, and, owing to the widespread 

 occurrence of proteins in seeds which in the free state are soluble in 

 water but as salts behave like globulins, more or less of these proteins 

 may be present in the solution after coagulating the albumins, and 

 consequently become mixed with the proteoses. The separation of 

 the above-mentioned proteins soluble in the free state, as well as the 

 globulins, would be comparatively easy if either of these could be 

 completely coagulated by boiling their solutions, as is the case with 

 animal globulins. Unfortunately these proteins are very imperfectly, 

 if at all, coagulated by heat, and therefore cannot thus be separated ; 

 in fact many of the so-called seed globulins closely resemble hetero- 

 proteoses in their solubility. 



The precipitates of globulin obtained by the preceding process are 

 united and dissolved in dilute sodium chloride solution and, after 

 filtering, are reprecipitated by dialysis. If a sufficient quantity of 

 globulin separates it may be further purified by fractional precipitation 

 with ammonium sulphate, according to Hofmeister's well-known 

 method. 



By extracting the seeds of the cereals with water only a small 

 proportion of the total proteins are removed and this consists of a 

 little proteose, a few tenths per cent, of albumin, and a very little 

 globulin. Many of the leguminous seeds, such as peas, vetches, 

 beans (except Phaseolus], lentils, and soy beans, yield relatively much 

 protein to water which, on adding a little acetic acid, and passing 

 carbonic acid through the extract, or on standing until the extract 

 becomes slightly acid, is largely precipitated. This precipitate, when 

 freshly formed, is soluble in sufficiently concentrated saline solutions, 

 and thenceforth behaves like globulin and for convenience is commonly 

 treated as globulin according to the current arbitrarily employed methods 

 of classifying proteins. From other leguminous seeds, such as the 

 lupines, very little more than traces of protein are extracted by water. 



Water extracts much of the protein from many of the oil seeds, 

 when freshly ground, such as the Brazil-nut, the almond, and the hazel- 

 nut, but from many of the others it extracts but little. The aqueous 

 extracts of such of the oil seeds as yield much protein to water give 

 abundant precipitates when treated with carbonic acid, and this method 



