THE BLOOD. 61 



at _|_ 75 to + 90 C. By the careful addition of acid the coagula- 

 tion temperature may be lowered ; by the addition of alkali it may 

 be raised. In blood-serum from certain animals and in human 

 transudations HALLIBURTON found the coagulation to take place 

 on heating to the following temperatures: + 70 to 73 C.; 77 to 

 78 C.; and 82 to 85 C. He therefore considers the serum 

 albumin a mixture of three albumins, at, fi, and y, which coagulate 

 at the three points mentioned. In cold-blooded animals he found 

 only the albumin a. 



The serum albumin differs from the albumin of the white of 

 the hen's egg in the following particulars : it is more laevogyrate ;. 

 the precipitate formed by hydrochloric acid easily dissolves in an 

 excess of the acid; it is much less insoluble in alcohol; and lastly 

 it acts differently inside of the organism. If egg-albumin is intro- 

 duced into the blood system it passes into the urine, while the 

 serum albumin does not. A solution of serum albumin positively 

 free from mineral bodies has never yet been prepared. A solution 

 as poor as possible in salts does not coagulate either on boiling 

 or on the addition of alcohol. After the addition of a little 

 common salt it coagulates in both cases. 



In preparing serum albumin, first remove the globulins by 

 saturating with magnesium sulphate at about -}- 30 C., and filter 

 at the same temperature. The cooled filtrate is separated from 

 the crystallized salt and is treated with acetic acid of about 1#, 

 The precipitate formed is filtered, pressed, dissolved in water 

 with the addition of alkali to neutral reaction, and the solution 

 freed from salt by dialysis. The serum albumin may also be sepa- 

 rated from the filtrate saturated with magnesium sulphate by 

 adding sodium sulphate to saturation at about -f- 40 C. The 

 pressed precipitate is also in this case dissolved in water and the 

 solution freed from salt by dialysis. The albumin may be obtained 

 in a solid form from the dialyzed solution either by evaporating 

 the solution to dryness at gentle heat or by precipitating with 

 alcohol, which must be removed quickly. In the detection and 

 quantitative estimation of serum albumin, the filtrate from the 

 globulins which have been removed by magnesium sulphate is 

 heated to boiling, after the addition of a little acetic acid if neces- 

 sary. The simplest way is to consider the difference between the 

 total albumins and the globulins as serum albumin. 



