THE INTERNAL SECRETION OF THE PANCREAS 431 



of the thoracic duct at the neck, or the diversion of the chyle 

 by means of a fistula, was followed by permanent glycosuria, 

 which was unaffected by either the fasting state or by feeding 

 with carbohydrates. The fact that the thoracic duct is not the 

 only channel by which the chyle is carried off, explains the 

 negative cases. 



The results of these experiments led me to conclude that a 

 substance is contained in the chyle which influences the meta- 

 bolism of sugar in the organism. I resumed this line of in- 

 vestigation in association with Offer in 1907, and we discovered 

 that the instillation of adrenalin into the conjunctiva produces a 

 mydriasis, which, according to Lowi, is characteristic of pancreatic 

 diabetes. We discovered, further, that the chyle in the thoracic 

 duct possesses an inhibitory action in regard to adrenalin, and 

 that adrenalin glycosuria may be reduced, or even abolished, by 

 the subcutaneous injection of chyle, as well as by intensifying 

 the normal lymph secretion by means of lymphagogue substances. 

 It would appear from this that the substance contained in the 

 chyle which influences the metabolism of the carbohydrates, is 

 identical \vith the pancreatic hormone. The pancreatic substance 

 which possesses the property of inhibiting the formation of sugar, 

 reaches the circulation by way of the lymphatics. 



Such being the case, then, it is to be expected that the in- 

 jection of chyle from the thoracic duct of normal dogs into the 

 veins of dogs without pancreas, will be followed by a diminution 

 of the glycosuria. As a matter of fact, positive results were 

 obtained in one case by Lepine, though Falta's results were 

 negative. 



In my later experiments, the action of the chyle in reducing 

 glycosuria after pancreasectomy, w~as very clearly shown, the 

 best results being obtained in cases where, owing to the presence 

 of small remnants of pancreatic tissue in situ, the derangement of 

 metabolism had not attained the maximal. On the days upon 

 which large quantities (200 to 500 c.cm.) of chyle from healthy 

 dogs were injected subcutaneously, the animals showed a diminu- 

 tion of the quotient D :N to 1.5 or 1.2; while upon days when 

 no lymph was given the quotient reached 1.8, and later even 2.1. 

 It should be noted in passing that these results suggest the 

 therapeutic employment of chyle, or of lymphagogue substances, 

 in those cases of clinical diabetes which suggest pancreatic in- 

 adequacy, but not pancreatic suppression. 



The analogy between the symptomatology of true clinical 

 diabetes and the experimental pancreatic form is so far-reaching, 

 that it is not surprising that \ve frequently encounter the opinion 

 that the metabolic disturbances of clinical diabetes are also pan- 

 creatogenic in their origin and are the expression of diminished 

 or internal secretory activity on the part of the pancreas or its 

 suppression. It should be borne in mind, however, that a 



