1 86 I'KRFOiniAXCE OF THE EXPERIMENT 



selected. If, in such a case, the sample of organ + 

 serum shows a markedly stronger blue coloration, 

 then the case has to be looked upon as positive in 

 respect to the reaction. Should, however, the differ- 

 ence in the intensity of the coloration be very small, 

 the experiment has to be performed again, using a 

 less quantity, say i c.c., of serum. It would then be 

 possible to ascertain whether decomposition had 

 taken place or not, as the serum sample would be 

 negative. 



The appearance of the reaction should on no 

 account ever be determined by artificial light. Again, 

 it is not advisable to compare the test-tubes in their 

 stands, but each one should be taken out separately, 

 and examined against white paper by both trans- 

 mitted and reflected light. 



Breaches of this rule are very often committed. 

 Many reactions are declared positive which, when 

 thoroughly investigated, show not the slightest 

 coloration. If a sample is marked as being just 

 perceptibly positive, then a number of other samples 

 should be changed about in the hand, and, only if 

 the same sample can be unhesitatingly picked out as 

 showing a coloration, should one's judgment con- 

 cerning the reaction be relied on. 



Difficulties are only experienced with reddish and 

 yellowish-brown tones, but these have no relation 

 whatever with the ninhydrin reaction. They can 



