640 INTERNAL SECRETION ENDOCRINE GLANDS 



secretion is highly probable, it is not yet definitely settled whether 

 this secretion is formed in the organ as a whole, or only in the islets. 

 That lesions of the pancreas may be concerned in pathological 

 diabetes is well established, and it is of interest in connection with 

 the question we have just been discussing that in a certain number 

 of cases the changes observed have been in the islands (Opie) . And 

 in diabetes accompanying cirrhosis of the liver, which has usually 

 been considered to depend upon the hepatic changes, it has been 

 shown that in many, if not all, of the cases the pancreas is also 

 affected by a growth of connective tissue outside the acini (Stein- 

 haus). Some authors, indeed, have gone so far as to say that in all 

 cases of diabetes mellitus there is disease of the pancreas, but of this 

 there is no evidence. 



Ligation, or the establishment of a fistula, of the thoracic duct, 

 causes glycosuria in dogs. It is possible that this is really a mild 

 form of pancreatic diabetes, due to interference with the supply ol 

 the internal secretion of the pancreas, or of that part of it which 

 reaches the blood by the lymph- stream (Tuckett). 



Pfliiger long maintained that it is not the removal of the pancreas, 

 as such, but the section of certain nerves running into or through it 

 from the duodenum, which is the cause of the glycosuria. For, 

 according to him, when these nerves are divided or the duodenum re- 

 moved while the pancreas remains untouched, the result is the same as 

 if the pancreas itself had been excised. He imagined that these nervel 

 are ' antidiabetic ' that is, in some way oppose the production of 

 sugar while nerves coming from the so-called ' sugar centre ' in the 

 bulb (the centre assumed to be affected in the puncture experiment) 

 favour sugar production. Between these the normal balance is struck 

 in health; it is the upsetting of this balance by the crippling of the 

 duodenal fibres which is at the bottom of ' pancreatic ' diabetes. 

 But it has been shown that this hypothesis is without foundation. 

 Although in frogs removal of the duodenum does cause a certain degree 

 of glycosuria, this is not the case in dogs. And it has been demon- 

 strated clearly in the dog, by special experiments, that section of all the 

 old connections of a dislocated remnant of the pancreas does not cause 

 permanent glycosuria, whereas removal of the pancreatic tissue does. 



Sexual Organs. The influence of castration in preventing the 

 development of the sexual characters, and especially the physical 

 and psychical changes that normally occur at puberty, is also due 

 to the loss of the internal secretion of the generative glands, and 

 does not appear to depend at all upon the loss of nervous impulses 

 arising in these organs. In Herdwick sheep an outstanding sexual 

 difference is the presence of horns in the males, their absence in 

 the females. Removal of the testes from ram lambs arrests further 

 growth of horns forthwith and at any stage of development. The 

 retention of the epididymes, provided that the testes proper are 

 removed, does not alter the result of castration in the least. The 

 removal of one testicle slows horn growth without arresting it 

 (Marshall and Hammond). In partially castrated cocks it has been 



