lo; The Protei<l* of Sfruin. [June 19, 



ous than has hitherto been supposed, my next object was to find 

 .pome distinction between these varieties of serum-albumin other than 

 that of heat-coagulation. It must, however, be remembered that the 

 mere difference of temperature of coagulation is an important one ; 

 in fact, one of the chief differences between fibrinogen and serum- 

 globulin is that the former coagulates at 56 C. and the latter at 

 75 C., and jet the function of these two bodies is vastly different. 

 In the want of knowledge of the chemical constitution of the proteids, 

 one has to be content at present with mere physical properties. 



My object, however, was to find certain salts which would precipi- 

 tate these varieties of serum-albumin. In the same way that Ham- 

 marsten discovered that magnesium sulphate will precipitate serum- 

 globulin from serum, and will precipitate nothing else, so it was 

 hoped that certain other salts would precipitate one variety, and that 

 one only, of serum-albumin. 



On this point I am at present still engaged. A large number of 

 salts have been examined, and these experiments constitute the second 

 category spoken of at the commencement of this paper. 



The most important result obtained has been with sodium sulphate. 



When sodium sulphate is added to saturation and in excess to 

 serum, and shaken for many hours, no precipitate occurs. 



When magnesium sulphate is used a dense precipitate occurs ; this 

 is filtered off, and then the filtrate is saturated with sodium sulphate; 

 another dense precipitate occurs.* If the filtrate saturated with 

 magnesium sulphate be shaken with sodium sulphate in excess for 

 eight or nine hours, the serum-albumin is found to be completely 

 precipitated; and the filtrate is found to be absolutely free from 

 proteid, and is clear and colourless. f 



If this precipitate be washed with water saturated with magnesium 

 sulphate and sodium sulphate, and then dissolved by adding water, 

 the resulting solution will be found to contain proteid, which can 

 again be separated by heat-coagulation in the way previously stated. 



This experiment has been performed with sheep's serum, and has 

 been confirmed on rabbit's serum. 



Potassium acetate was found to act similarly to sodium sulphate; 

 adding it to saturation to serum already saturated with magnesium 

 sulphate produced complete precipitation of all forms of serum- 

 albumin. Potassium acetate differs from sodium sulphate in also 

 producing a precipitate in simple serum. Here it precipitates both 

 globulin and serum-albumin. It thus resembles in its action potas- 

 sium carbonate, which is known to precipitate all the proteids. 

 Potassium acetate is a difficult salt to work with, as it deliquesces in 



* Schafer, " Journal of Physiology," vol. iii, pp. 181-187. 



t I have since ascertained that if serum is saturated with the double sulphate of 

 sodium and magnesium, all the proteids are quickly and completely precipitated. 



