METHODS OF ANALYSIS. 223 



Corne,* on the influence of motion on the diminution of the fibrin, 

 and the deductions they have drawn from them, are entirely 

 dependent on errors in their mode of analysis. It is certainly 

 impossible altogether to avoid these errors ; for if we carefully 

 collect all the insoluble substances passing through the linen filter, 

 the number which we obtain for the fibrin is too high. The 

 following is, we believe, the best, although it is by no means a 

 perfect method of collecting all the insoluble matter which collec- 

 tively goes under the name of fibrin : whipt blood must be strongly 

 watered, and the flakes allowed to deposit themselves, (this deposi- 

 tion is, however, often very imperfect ;) the fluid, as far as it has 

 become clear, is to be drawn off, and the turbid residue, with the 

 coarser clots, to be repeatedly shaken with water, and the clear 

 fluid to be removed till water ceases to be at all coloured by it ; 

 afterwards, if it be possible,a paper filter must be used, through which 

 the fine flakes cannot penetrate, in place of a linen one ; previously, 

 however, heating the fluid with an equal volume of spirit. (When 

 the fluid has once become colourless, no more coagulable sub- 

 stance is generally found in the fibrin.) The fibrin may then be 

 tolerably easily collected on the filter, and it should afterwards be 

 washed with boiling spirit. We cannot even in this way accu- 

 rately determine the fibrin ; but we know what we have to deal with, 

 and that if we always determine the fibrin in excess, it is far less 

 dependent on pure casualties than if we had used linen filters, or 

 had partially watered the blood ; in one case any calculation is 

 impossible, in the other we know that there is always an excess of 

 fibrin and a slight loss of blood-corpuscles. Our knowledge of 

 the error leads us to hope that we may subsequently learn to avoid 

 it : accuracy and patience are indeed indispensable in this mode of 

 determining the fibrin. Moreover, in this somewhat circumstantial 

 operation, the blood does not readily undergo putrefaction, in con- 

 quence of the frequency with which the water is changed. 



For the method of determining the albumen in the serum, we 

 must refer to the observations already made in p. 339 of the first 

 volume. In regard to the special application of those remarks 

 to the analysis of the blood, we need only add that it is always 

 important, besides determining the albumen in the serum, to 

 determine also the coagulable matter contained in the clot, or 

 the clot when free from fibrin, or in the clefibrinated blood, as a 

 controlling check which is indispensible in blood-analyses; so as in 

 some degree to combine the methods of Becquerel and Rodier, or of 

 * Compt. rend. T. 30, p. 110. 



