METHODS OF ANALYSIS. 221 



analysis of the clot, and from the quantity of this suhstance peculiar 

 to the serum, it would be very easy to calculate how much serum 

 (which must obviously also have been analysed) was inclosed in the 

 clot; if then we deduct the other constituents pertaining to the 

 serum (as determined by the analysis of it), from the quantity of 

 similar substances and the fibrin found in the clot, we should at 

 once obtain, by the simplest calculation, the quantity and the com- 

 position of the blood-corpuscles contained in 100 or 1000 parts of 

 blood. If this were the case, the problem would be completely 

 solved, but, unfortunately, neither in the pre-formed sulphates nor 

 in the organic matters can we find a substance which is entirely 

 excluded from the blood-cells. Hence, in all probability, we must 

 for ever rest satisfied with Schmidt's coefficient, as affording the 

 closest approximation ; but if other parts of the blood-analysis were 

 equally accurate, this coefficient would always afford highly correct 

 results. Physiology, and especially physiological chemistry, are 

 indebted for the most brilliant results to this ingenious combination 

 of Schmidt's. 



Schmidt's method of calculation in analysing the blood is very 

 easy of comprehension : we have the analyses of the clot and of 

 the serum, and the proportion (calculated from these data) of the 

 constituents of the whole blood : if we multiply by four the 

 number of the dry blood-corpuscles calculated by, Prevost and 

 Dumas' method, we obtain the quantity of fresh blood-cells, and 

 hence their ratio to the intercellular fluid. We now deduct, from 

 the analysis of the whole blood, the constituents belonging to the 

 quantity of intercellular fluid, and the remainder represents all the 

 substances belonging only to the blood-corpuscles. 



No one has yet attempted a quantitative determination of the 

 colourless blood-cells ; it is probable that we shall never arrive at 

 more than an average estimate of them. 



We have already spoken (in vol. i., p. 356) of the quantitative 

 determination of the fibrin, and pointed out that it is not to be 

 relied upon. We will here only add a few words regarding the 

 results of Hinterberger^s experiments, which show that we always 

 obtain less fibrin by whipping the blood than by washing the clot. 

 He is of opinion that the evaporation of the water occurring during 

 the coagulation of the blood, may be one of the causes of this 

 difference ; considering, however, the comparatively small quantity 

 of fibrin in the blood, the error arising from this cause would be 

 infinitesimally small, and in the best analyses of the blood would 

 be overbalanced by other errors of observation; thus, one of the 



