94 R. H. A. PLIMMER. 



In view of the uncertainty of the earlier results of Vasilieff and of 

 Walt her on the qualitative adaptation of the pancreatic juice to the 

 food-stuffs present in the intestine, the adaptation of the pancreas to 

 lactose, described by Weinland and Bainbridge, acquired great 

 importance, as presenting direct evidence of the adaptation of the 

 pancreas to the food. The question of the accuracy of their observations 

 must therefore largely influence our attitude to the views put forward 

 by Pawlow, with regard to the correlation of the total activities of the 

 pancreas with the nature of the food. At the suggestion of Prof. 

 Starling I have therefore submitted the whole question to a fresh 

 investigation. 



The credibility of the conclusions arrived at by Weinland and 

 Bainbridge depends on the trust which can be placed on their methods. 

 Those employed by Weinland have been justly criticised by 

 Bainbridge. In Weinland's experiments an infusion of the pancreas 

 was treated with a given proportion of lactose, and then placed in the 

 incubator with the addition of toluol for some hours. The presence or 

 absence of lactase in the infusion was judged by (1) submitting the 

 mixture to the action of Saccharomyces apiculatus, which, while fer- 

 menting glucose, has no action on lactose : (2) by estimating the 

 rotatory power of the solution after separation of the proteids it 

 contained by means of Briicke's reagent. In no case were control 

 observations carried out on mixtures in which the pancreatic infusions 

 had been boiled, but the presence or absence of change was judged of by 

 comparing the rotation with the rotation which would be expected from 

 the amount of lactose added to the fluid. (3) In a certain number of 

 cases the osazones were prepared from the mixture. For the identifica- 

 tion of these osazones Weinland trusted chiefly to their microscopic form, 

 no determination being made either of the melting point of the purified 

 crystals or of their content in nitrogen. 



Of these three methods W T einland placed most reliance on the 

 polarimetric. In his dilute solutions the differences of readings that 

 he observed were well within the limits of experimental error. Two 

 minutes might make a difference of 10/o inversion. Moreover, if any 

 proteolysis had taken place in the solution amino-acids would be formed 

 and would not be precipitated by Briicke's reagent. In the strongly 

 acid filtrate obtained from this reagent, nearly all the possible amino- 

 acids which might be present would have a dextro-rotation, and 

 therefore might account for the whole of the increased rotatory power 

 observed in these various experiments. The presence of dextrose in the 



