180 CIRCULATING FLUIDS : 



state. The globulin usually occurs in analyses of tlie blood as 

 a sulphatCj and as sucli I liave abvays estimated it. It is of 

 a grayisli-wliite colour, forms a brownish solution in water, 

 and on incineration leaves an ash, more or less abundant in 

 iron. 



If after the separation of hsematin (in the manner already 

 described), and after being washed in alcohol, the globulin 

 retains a red tint, it must be again treated with a lukcAvarm 

 mixtui'e of sulphuric acid and alcohol, as before, wdiich dis- 

 solves the hsematin that had remained attached to the glo- 

 bulin. It must then be repeatedly washed with alcohol, until 

 it no longer exhibits any acid reaction. 



Hmnatin. From the remarks which have been made re- 

 specting the albumen and the globulin, the reader may con- 

 clude that the hsematin cannot always be determined with 

 exactness ; I conceive, however, that with all due care, the 

 error in the determination of the hsematin should be very 

 trifling in 100 parts. It by no means necessarily follows that 

 hsematoglobulin should under all circumstances contain a con- 

 stant proportion of hsematin. Moreover, if the fat has not 

 been previously entirely removed, a certain qviautity may be 

 associated with the hsematin. If the hsemaphsein is separated 

 from the hsematin by means of warm alcohol, the fat dis- 

 solves simultaneously with the former of these colouring mat- 

 ters, and remains closely connected with it. If the alcohol 

 used for the separation of the hsematin from the globulin is 

 not suflBciently strong ; and if, after the saturation of the sul- 

 phuric acid with ammonia, a sufficient time is not allowed 

 for the sulphate of ammonia to separate, a portion of this 

 salt will pass through the filter, and become mixed wdtli the 

 hsematin upon the evaporation of the alcohol. If this is the 

 case, the salt may be easily recognized in the hsematin by its 

 crystalline form ; and it must be extracted with water. It is 

 always advisable to use strong alcohol, and to allow the satu- 

 rated solution to stand for some hours before it is filtered. 



Hamaphmn. The determination of this constituent is 

 somewhat uncertain and difficult, on account of the minute 

 proportion in which it exists. It is occasionally found to 



