8 LECTURE I. 



probabilities, and in no case should it be credited as if resting upon facts 

 established experimentally. Analogies in many cases are without doubt 

 very valuable, and often form the skeleton upon which we can build further. 



We receive a new impulse and gain a new point of view with every 

 advance made by pure chemistry concerning substances of physiological 

 interest. Our task is to utilize each discovery thus made and to give it a 

 strictly objective test with regard to its application to the processes taking 

 place in the tissues. Here again we meet all too frequently with hypotheses 

 which are stated as facts. We hardly need to mention how extraordinarily 

 restraining the direct amalgamation of these entirely different elements is 

 for a healthy progress in the knowledge of chemical processes in the animal 

 organism. For these reasons Wohler's experiment proves positively 

 merely that when benzoic acid is introduced into the system it causes an 

 increased elimination of hippuric acid. It must remain an open question 

 as to whether the benzoic acid introduced stands in direct relation to the 

 other acid or merely indirectly causes its formation. In this particular 

 case, however, the latter case seems scarcely probable, although we must 

 make such a limitation unless we propose to draw our conclusions beyond 

 the realms of fact. 



We must now mention an important aid which the chemist makes 

 use of constantly in his experiments, which are often indirect in nature. 

 We refer to the control experiment. It is clear that there is nothing 

 to be gained by merely feeding an animal with benzoic acid and deter- 

 mining subsequently the amount of hippuric acid eliminated. We must 

 first learn how much hippuric acid the animal in question eliminates under 

 normal conditions. If the experiment is to be made convincing, the amount 

 of hippuric acid contained in the urine of one and the same animal fed 

 uniformly must first be determined and this continued for several days. 

 Then for a time a little benzoic acid should be added to the food, which 

 otherwise must remain qualitatively and quantitatively the same as before, 

 and again the hippuric acid be determined in the urine. If now the experi- 

 ment be continued for another period of several days in which no benzoic 

 acid is fed to the animal, then, if the whole experiment is consistent, it 

 will be possible to determine whether the benzoic acid stands in any 

 relation to the elimination of hippuric acid. 



The uncertainty of the significance of experiments made with animals 

 is in many cases greatly increased by the fact that individual variations 

 often play an important part. Many contradictions to be found in the 

 literature are due solely to the fact that the experiments were not carried 

 out long enough. We must not only require that such experiments should 

 be carried out in a single individual for quite a length of time, but in dif- 

 ferent individuals of the same animal species as well. It is, furthermore, 

 of great value to make experiments with different species of animals, for 



