314 Chloroform in the Blood of Ancesthetised Animals. 



Table VI. 



Number of Weight of 

 experiment. blood. 



Chloroform by 

 Carius. 



Chloroform by 

 Nicloux. 



, { 



grammes. 

 4'579 

 4 -1432 



14 9269 



13 -879 



0319 

 -0327 



-0334 

 -0359 



Conclusions. 



1. The method of chloroform determination, based on Dumas' reaction, 

 though not possessing a very high degree of precision as an exact chemical 

 method, as is apparent from a comparison of Tables I and II, is sufficiently 

 accurate for most practical purposes and, as used by Dr. Mcloux, is a simple 

 and rapid method, and one capable of giving satisfactory results for the 

 estimation of small quantities of chloroform in air and in simple solution in 

 such liquids as urine, etc. 



2. In the case of the blood of anaesthetised animals, if the blood for 

 analysis is allowed to clot before admixture with acid alcohol, the results by 

 Nicloux' method are uniformly too low. An inspection of Tables III and V 

 will show that results concordant amongst themselves are obtained when 

 sensibly equal quantities of blood are taken, but, as might be expected, the 

 figures decrease the larger the amount of blood used. In cases where smaller 

 amounts of blood are taken and, consequently, the amounts of chloroform 

 dealt with are very small, the results are often near the truth ; but this, we 

 think, is largely due to the inherent difficulty in the method as regards the 

 determination of the end-point;. 



3. In cases where the clotting of blood was prevented by oxalate, the 

 results more nearly approach those given by the methods we have previously 

 adopted, but still are somewhat lower. 



If we take the results given in Table Y, the average percentage of 

 chloroform indicated by our method is O0319, and by that of Mcloux in the 

 case of oxalated blood is 0*0259. In our method we are satisfied that the 

 average maximum error is not more than 5 per cent., generally considerably 

 less. We find the difference between the results given by our method and 

 that of Nicloux' method averages 19 per cent. 



4. In the single experiment quoted with hirudin, the figures by the two 

 methods are remarkably concordant. 



