BLOOD. 



24.') 



In all otlier methods of analysing the blood in which the water 

 is determined by a separate process, and the dried residue is 

 used for further investigation, an error in its estimation will 

 simply increase the absolute quantity of the solid constituents, 

 without disturbing their relative proportions. But in the ap- 

 pHcation of theu' method it is easy to see that each per-centage 

 of retained water not only increases the absolute quantity of 

 the solid constituents to the amount of 1% but also the weight 

 of the corpuscles, not only by the addition of the retained water, 

 but also by the weight of the residue of the serum, due to an 

 equal quantity of water, and which amounts to 1*1^. 



Moreover, the supposition of Andral and Gavarret, that the 

 humidity of the clot should be considered as serum is totally 

 devoid of foundation. The corpuscles cannot be supposed to 

 swim in the plasma as dry molecules, and it has not been 

 proved that the fluid, with which they are filled, is the fluid of 

 the serum. 



These observations are sufficient to show that Andral and 

 Gavarret's method, and my own, give somewhat different re- 

 sults: the difi'erences, however, are not very material, and are 

 easily explicable on the grounds already stated. 



The changes which the composition of the blood may expe- 

 rience in its various pathological conditions, are either de- 

 pendent upon the quantity of solid residue generally, or upon 

 the changed relative proportions that the various proximate 

 constituents bear to each other. 



If we assume the composition of healthy blood, (as deduced 

 from the mean of my analyses) to be represented by 



the following difi'erences will be found to occur among; the 



specimens of diseased blood which I have analysed, 

 quantity of — 



The 



