COMPOUND ANIMAL SUBSTANCES. 93 



however be occasionally present, when, even after the addi- 

 tion of a solution of salt, sugar, or iodine no blood-corpuscles 

 can be detected; in this case the latter are in a state of 

 pei'fect solution. 



When the fluid is of a dark brown, or blackish-red colour, 

 hfematin is the colouring constituent. If the fluid be of a 

 clear brown or yellow colour, hsemapheein is almost sure to be 

 the origin of the tint, especially if any taste of bile be percep- 

 tible.- Biliphsein will also communicate a yellow, brown, or 

 greenish-brown colour ; in this case there is frequently a bitter 

 taste, and on the addition of nitric acid, there is always a 

 change of coloiu' into green or blue, and yellow. 



Qualitative analysis. 



Having poured the fluid into proper test-glasses, we carry on 

 our investigations in the following manner : 



1. If, on the addition of very dilute hydrochloric acid, a 

 precipitate be thrown down, we see whether it will dissolve in 

 an excess of the test.' Assuming that the solution is effected, 

 ferrocyanide of potassium is added ; if this test instantly throws 

 down a white or yellow precipitate, one or more of the protein- 

 compounds (enumerated in i) are present. 



In order to ascertain which of the protein-compounds 

 has yielded these indications,^ a portion of the fluid is boiled : 

 if it become turbid, and if the turbidity commence and be 

 most distinct at the surface, or if the fluid coagulate, then 

 albumen is present ; in this case nitric acid and bichloride of 

 mercury will throw down copious precipitates. If the fluid 

 become turbid on the application of heat, and the coagulum 

 assume a red tint, then globulin and hpematin are also present, 

 althoixgh the microscope may have failed in detecting blood- 

 corpuscles : in this case, however, the fluid is always of a 

 rather pink or reddish tint. 



If the fluid does not coagulate on the application of heat, 

 casein, or one of the caseous substances must be present. 



> If very dilute hydrochloric acid he employed, the albumen will not be precipi- 

 tated. (See p. 18.) I prefer hydrochloric to acetic acid, because the latter throws 

 down pyin with the protein-compounds. 



2 Fibrin is recognized by its spontaneous separation, and need not be sought for in 

 the manner indicated in the text. 



