HUMAN DIABETES 533 



when the patient is receiving no carbohydrate whatever in the 

 food. In the early stages glycosuria may appear only after the 

 taking of carbohydrates, so that it may be entirely suppressed 

 by proper diet. Later, however, as the power to utilize sugar 

 becomes still more impaired, the demand of the tissues for sugar 

 becomes so great that carbohydrates are formed in large 

 amounts from proteids, and perhaps also from fats. At the 

 same time the sugar given by mouth passes unappropriated 

 through the tissues, and the greater part of it, sometimes all, 

 reappears unchanged in the urine. As with experimental pan- 

 creatic diabetes, the power to utilize levulose is retained longer 

 than for dextrose, but eventually even the levulose is largely 

 lost, after being partly converted into glycogen. 



The power to oxidize substances other than dextrose is at 

 first apparently unimpaired, but later the general oxidative 

 capacity is reduced, and we find large quantities of uuderoxid- 

 ized products of metabolism appearing in the urine, especially 

 the " acetone bodies." The presence of these organic acids in 

 the blood in large amounts leads to diabetic coma, which is in 

 most respects an aeid intoxication. (This matter is discussed 

 fully under the topic of " Acid Intoxication/' Chap, xviii.) 

 Undoubtedly, other toxic substances also accumulate, and con- 

 stitute an important, if subordinate, cause of the toxic manifes- 

 tations of the disease. Failure of oxidation of fats is perhaps 

 responsible for the frequently observed accumulation of large 

 quantities of fat in the blood lipemia. (Discussed under 

 " Fatty Metamorphosis," page 344.) The failure of oxidation 

 of sugar is so marked that it is impossible to cause the glyco- 

 suria to be reduced greatly, at least in severe cases, by hard 

 muscular exercise. Thus, a patient who is excreting sugar on 

 a carbohydrate-free diet may climb a mountain or do other 

 severe work which requires normally the burning up of 80 to 

 100 grams of carbohydrate, and yet continue to excrete nearly 

 or quite as much sugar as he did before. This indicates that 

 all the energy available for the diabetic must come from fats 

 and proteids, because of the inability to utilize sugar under even 

 the most extreme conditions. 



In any case, the pathology of human diabetes usually re- 

 sembles that of diabetes following experimental pancreatectomy 

 in its chief features, and the problems are quite the same as 

 those which have already been discussed in connection with the 

 experimental disease. We are entirely uninformed as to the 

 parts played by defective glycogenesis and by defective oxida- 

 tion of sugar, independent of a preliminary conversion into 



