Precipitins and Precipitable Substances 97 



favour of the action not being fermentative, and I do not doubt but that 

 Biondi did not work with a sufficiently large series of dilutions, for if he 

 had he would have convinced himself to the contrary, as can be readily 

 done 1 . Oppenheimer and Michaelis (18, vn. 1902) consider the compo- 

 nents in serum which lead to precipitation to be constituents of albumin, 

 not merely attached thereto. In a later paper Michaelis (9, x. 1902, 

 p. 734) states that he found precipitins in that fraction of serum which 

 is precipitated by half saturation thereof with ammonium sulphate, 

 namely, in the globulin fraction. He attempted to further isolate the 

 precipitin by repeating the fractional precipitation with ammonium 

 sulphate and found that most of the precipitin came down with the first 

 fraction (0 and 30 /o saturation), the second fraction (precipitated by 

 30 and 50 / saturation) only containing traces of precipitin. The 

 precipitable substance behaved similarly. He finds that anti-ox 

 globulins do not precipitate serum-albumin. Corin (1902) considers the 

 active principle which leads to the formation of precipitin to be para- 

 globulin, and that therefore it would be best to treat animals with 

 paraglobulin solutions to obtain antisera, such as he prepared for the 

 dog. He finds the precipitin bound up with the paraglobulin, and this 

 can be separated and dried, the dried powder being brought into clear 

 watery solution and used directly for testing. 



According to Obermayer and Pick (1902) the very rapid appearance 

 (within 15 minutes) of precipitins in the circulation of animals which 

 have received an intraperitoneal injection of precipitating antiserum, 

 indicates that the precipitin is non-albuminous. They found the 

 precipitins to be contained in the euglobulin fraction, both in rabbit 

 and horse immune serum, obtained from the goat and rabbit respect- 

 ively. In the hope of determining which of the constituents of egg- 

 albumin are concerned in causing the formation of precipitin in corpore, 

 they treated rabbits with subcutaneous injections of egg-albumin and 

 tested the action of the antiserum thus obtained upon egg-white and its 

 constituents. They state that normal rabbit serum produces an appre- 

 ciable precipitum in egg-albumin solutions 2 , this being due especially 

 to the action of dysglobulin on the serum, the action of other constitu- 

 ents being much less marked. They consider that the egg-white 

 precipitins do not constitute an integral part of the albuminous sub- 



1 An exteDsive series of tests made by Mr Strangeways at my suggestion during the 

 past year, which have not been hitherto published, distinctly show Biondi to have been 

 wrong. 



a Strength not stated ; see normal precipitins, p. 150. 



N. 7 



