168 CIRCULATING FLUIDS: 



with ether, alcohol, and boiling water. Tlie ultimate resi- 

 due consists of fibrin, blood-corpuscles, and albumen; by 

 deducting the aheady determined weight of the fibrin and 

 blood-corpuscles, we obtain the weight of the albumen. Ether 

 takes up the fat ; alcohol, certain extractive matters, and lac- 

 tates ; boiling water, certain extractive matters, chloride of 

 sodium, &c. The serum (the quantitative relation of which 

 to the clot is known) is gently boiled, by which means the 

 albumen is coagulated, and all moisture is removed by eva- 

 poration. 



The dried residue is pulverized and treated with boiling 

 water, which leaves albumen and fat unacted upon ; the latter 

 of which may be now taken up by ether. 



The water dissolves the salts, certain extractive matters, and 

 some fat, or fatty-acid compounds. 



The watery solution must now be CA^aporated, and the residue 

 treated with alcohol, which takes up the chlorides of sodium 

 and potassium, the lactates, extract of flesh, and perhaps some 

 fat, if any happens to be present. 



An objection may be raised against this method, that the 

 separation of the blood-corpuscles from the serum is not suffi- 

 ciently perfect. 



The complete removal of the serum is a matter of very con- 

 siderable difficulty, in consequence of the formation of a dried 

 surface, at those parts of the clot which are in contact with the 

 paper, by which means a check is opposed to the egress of any 

 moisture from the interior portions. Indeed, the moist clot 

 can only be perfectly freed from hsematoglobulin with difficulty, 

 and with the loss of some fibrin; if it were thoroughly dried, 

 the difficulty would be confined to the washing out of the 

 blood-corpuscles. But when fibrin remains for a considerable 

 time in Avater, a small portion of it is dissolved, and a part of 

 it is transformed into a viscid mass, consisting of very minute 

 microscopic granides, Avliich are not easily Avashed out. When 

 all the blood-corpuscles are not inclosed by the coagulated 

 fibrin, the serum assumes a reddish tint in consequence of their 

 presence; they must then be taken into estimation with the 

 serum. In most cases, anatyses made in this manner would 

 yield too high a number for the blood-corpuscles ; in some few 

 cases the assigned number Avould be too small. 



