COMPOSITION OF THE BLOOD 339 



The Proteins of blood-serum consist of an admixture of albumins 

 and globulins. It is quite uncertain how many different proteins the 

 blood-serum (or plasma) may contain, but certain fractions can be 

 readily distinguished from one another. Among the globulins the 

 "Insoluble Globulin" or "Euglobulin" may be readily separated by 

 simple dilution of the blood-serum with from ten to twenty volumes of 

 distilled water, followed by acidification with dilute acetic acid or with 

 a stream of carbon dioxide. The same fraction separates out on sub- 

 mitting blood-serum to dialysis. An additional globulin fraction, the 

 so-called Pseudoglobulin remains in solution, but may be coagulated 

 by half-saturation of the serum with ammonium sulphate, and there 

 are indications that this substance, in turn, is not a single chemical 

 individual The Albumin fraction, which is not coagulable by half 

 saturation with ammonium sulphate, may also not improbably consist 

 of a mixture of proteins. Thus from the serum of the horse, but only 

 with great difficulty from other sera, a Crystalline Serum Albumin may 

 be obtained by first removing the globulins by half-saturation with 

 ammonium sulphate and then adding more ammonium sulphate until 

 coagulation of the albumins just begins, and allowing the mixture to 

 stand for some time. Only a portion of the albumin is deposited in 

 crystalline form, however, and we are uncertain whether the portion 

 which does not crystallize merely represents the quantity requisite to 

 saturate the liquid with crystallizable albumin, or whether it represents 

 a different protein. 



In addition to the albumins and globulins the blood often contains 

 very small amounts of Proteose, and also a glucoprotein, termed sero- 

 mucoid which yields glucosamin on hydrolysis. It is present in blood- 

 serum only to the extent of from 0.2 to 0.9 parts per thousand. 



It has been noted by a large number of investigators that the relative 

 proportion of globulins to albumins in the blood-serum may present 

 remarkable abnormalities in persons or animals afflicted with certain 

 Infections. Normally the globulins are always less abundant than the 

 albumin fractions, so that the ratio globulin albumin is always less 

 than unity. In animals or human beings infected with Streptococcus 

 or Staphylococcus , however, the ratio may be much more than unity, 

 the globulins in some instances amounting to as much as eighty or 

 ninety per cent, of the total proteins. The question of the origin of 

 this remarkable change is of course one which is of great importance to 

 our understanding of the mechanisms by which the organism protects 

 itself against infections, more particularly since, in the case of Diph- 

 theria at least, the Antitoxins resulting from infection or immunization 

 have been found to be associated with the globulin-fraction of the 

 serum. 



The older analyses aiming at the solution of this problem were 

 subject to very great errors and uncertainties, because of the compara- 

 tively large volumes of blood which were required for a single analysis. 

 The proteins were coagulated by alcohol, dried and weighed, while 



