AND DIABETES 165 



appear, there is still an efficient amount of renal 

 substance. To give precision to the test it is usually 

 wise to catheterize both ureters, and to analyse the 

 urines separately, 



{d.) By variations in the blood-content of CO^. — 

 Apparently either a marked rise or a marked fall in 

 the amount of carbon dioxide in the blood may 

 produce glycosuria. Prolonged deep breathing, for 

 instance, reduces the CO,^ in the blood to a very low 

 figure, and not infrequently transient glycosuria 

 results. This has been adduced as the explanation 

 of its occurrence after violent anger or after anaes- 

 thetics. Strong saline injections may also induce 

 glycosuria. 



(e). The ductless glands and diabetes. —There is 

 probably an important and complicated relationship 

 between the ductless glands and glycosuria. So far 

 we only know a few isolated facts. Thus, adrenalin 

 injection causes glycosuria of pancreatic type ; in 

 hyperthyroidism sugar may appear in the urine and 

 the toleration limit of glucose may be very low (less 

 than 100 grams may set up glycosuria) ; on the other 

 hand, partial excisions of the pituitary gland raise 

 the toleration for glucose above the normal 150 

 grams. 



( / ). Pancreatic diabetes. — Removal of the pancreas 

 in dogs or other animals if complete induces fatal 

 diabetes exactly corresponding to severe cases of the 

 disease in man ; a sub-total removal brings about a 

 milder type of the disease. The symptoms are 

 improved by pancreatic grafting. If less than four- 

 fifths of the pancreas is taken away, no glycosuria 



