1906.] 



Chloroform in the Blood of Animals. 



453 



In this case the corpuscle control suffered no loss, hut that of the plasma 

 was marked. Possibly this may be due to the fact that in this experiment 

 the separation of the corpuscles and plasma was much more complete. In 

 this case the chloroform was not completely eliminated from the corpuscles. 

 The loss suffered by the control and chloroform plasma was practically the 

 same, so that probably little or no chloroform was present. 



Experiment 20. — In order to prove conclusively that the loss of chlorine 

 noted in the control samples in the two previous experiments was really due 

 to the evolution of free hydrochloric acid on heating, a cat was anaesthetised 

 with ether, and bled. About 18 grammes of the blood were mixed with 

 a saturated solution of sodium oxalate (neutral) in the proportion of 1 c.c. 

 to 47 of blood and centrifugalised. 12 - 9 grammes of plasma and 4 - 9 grammes 

 of corpuscles were obtained. The samples were then placed in desiccators 

 and dried in vacuo over sulphuric acid at the temperature of the laboratory. 

 In each desiccator was suspended a small tray of pure crystals of copper 

 sulphate to absorb any hydrochloric acid that might be evolved. After the 

 samples were dry the copper sulphate crystals in each case were dissolved 

 in water and tested for chlorine. We were, however, unable to detect with 

 certainty any trace of hydrochloric acid. It must be remembered, however, 

 that the quantities evolved from the amounts of blood taken would be in 

 any case exceedingly small. The dried samples of plasma and corpuscles 

 were then placed in small flasks. Each flask was connected with a larger 

 flask, partly filled with strong sulphuric acid, by means of a fairly wide tube 

 containing crystals of pure copper sulphate, and the air was pumped from 

 the apparatus. The flasks containing the dried matter were then heated for 

 four or five hours on the water bath. At the end of the experiment, the 

 copper sulphate crystals in each tube were dissolved in water and tested for 

 chlorine. 



Each solution was found to give a slight but quite definite precipitate with 

 silver nitrate, which careful tests proved to be silver chloride. The precipi- 

 tate appeared to be somewhat larger in the case of the plasma than in the 

 case of the corpuscle. The precipitates were, however, too small to admit of 

 being accurately weighed. 



From these experiments it is therefore evident that dried blood loses traces 

 of chlorine at temperatures between 80° and 100° C. 



It is somewhat difficult to explain these rather unexpected results, but it 

 may be that the loss of hydrochloric acid is due to the interaction of the 

 sodium phosphate with the sodium chloride of the blood at the temperature 

 of the experiment. This hypothesis is in accordance with the fact which we 

 have before noticed that mixtures of common phosphate of soda and salt lose 



