CHEMICAL TOXICOLOGY 277 



quantity of available material, the nature of the poison 

 Buspected, and the degree of delicacy of its detection. It is 

 always wise to reserve a portion, preferably about one 

 quarter, for confirmatory analysis, or to replace accidental 

 loss. When the analyst has at his disposal the whole 

 stomachs of a horse, sheep, ox, or pig, he may consider 

 that he has carte blanche, for the material suffices for more 

 than one complete analysis. Quite otherwise when he has 

 a small animal such as a bird, cat, or small dog to deal 

 with. In such case he must use discretion in the taking of 

 parts for the various operations. 



Apparatus. — It is the common practice in medico-legal 

 work on the human subject to advise perfectly new apparatus 

 for each- analysis. The apparatus used ought to be in good 

 condition ; thus porcelain dishes must have an unimpaired 

 glaze ; but the dogma of new vessels is somewhat extreme. 

 To have used a set of apparatus in an analysis which has 

 yielded negative results is the best proof of its cleanliness. It 

 is an elementary maxim of the trained chemist to thoroughly 

 clean all apparatus in such wise as to meet the object of the 

 intended analysis. Thus a flask and condenser intended to 

 be used in Eeinsch's test will, as a matter of routine, be 

 cleansed by boiling hydrochloric acid ; a dish designed for 

 the nitric acid solution of parts will be boiled out with that 

 acid ; flasks intended for the alcohol extraction of organic 

 poisons will be cleaned with a mixture of strong sulphuric 

 acid and potassium chromate after the laboratory attendant 

 has ' cleaned ' them. No chemist ever uses a piece of 

 apparatus which he has not personally cleaned. Possibly 

 if the Courts thoroughly appreciated the fact that it is 

 a chemist's business to keep his apparatus clean, and not 

 to get muddled, and mix the specimens or reagents, less 

 weight would be attached to the general preliminary pre- 

 cautions. But, on the other hand, it is essential that the 

 person who issues the report should himself either have 

 performed, or personally supervised, all the operations. 



Reagents. — All reagents designed for toxicological analysis 

 must be of proved purity. The condition of reagents and 



