112 



Scientific Proceedings (99). 



steam distillation as ordinarily conducted in a general analysis 

 recovers but a small fraction of the total amount present, it is 

 obvidifs that only a few milligrams are to be expected. In several 

 experiments I was able to recover, by steam distillation for a period 

 of two hours, only 6 to 8 mg. from 100 mg. actually added. 



In view of these facts, it is evident that the Vitalli Tornani 

 test is to be used only when one is in a position to sacrifice prac- 

 tically all the material for this test alone. In a general unknown, 

 however, this cannot, of course, be done, as a large part of the dis- 

 tillate is needed for testing the presence of various other volatile 

 poisons. 



The problem, then, resolves in finding a reagent which, first, 

 is extremely sensitive to chloral, second, one for which only a small 

 fraction of the suspected material need be used and, third, one 

 which will react with chloral in a manner different from the re- 

 action with chloroform and, fourth, if possible, one in which for- 

 malin will not interfere with the result. 



A study of the following substances was made under different 

 conditions of alkalinity (sodium hydroxide, sodium carbonate, 

 calcium hydroxide). 



Resorcin 



Alpha naphthol 



Beta napthhol 



Orcin 



Phloroglucin 



Hydrochinon 



Pyrocatechin 



Phenol 



Cresols 



Thymol 



Pyridine 



Picric acid 



Amino acids and boiling 

 Fatty oils and dehydrating agents 

 Yellow ammonium sulphide 

 Calcium sulph-hydrate 

 Sodium thiosulphate 

 The result reveals only two reactions which would serve the pur- 



