THE METABOLISM OF CARBOHYDRATES 851 



B cells. Changes have also been found in the islets of Langerhans in fatal 

 cases of diabetes in man. It seems therefore probable that what we may 

 term, for lack of a better word, the antidiabetic functions of the pancreas, 

 are associated with and dependent on the integrity of the islets of Langerhans. 

 (5) DIABETES IN MAN. In its severer forms the diabetes of man 

 resembles very closely that produced in the dog by total extirpation of 

 the pancreas. The output of urine is largely increased and the frequency 

 of micturition is often the first symptom noticed. On examination the 

 urine, though light in colour, is of a high specific gravity, 1030 to 1035, 

 and may contain from 5 to 10 per cent, of sugar. The appetite is largely 

 increased, but in spite of the large amount of food taken the body wastes. 

 The excessive quantity of fluid lost by the body gives rise to a constant 

 thirst. The patient may die after some months or years in a condition 

 of diabetic coma. Warning of the onset of this condition is given by the 

 rise of ammonia in the -urine and by the appearance of oxybutyric and 

 diacetic acids. The breath may smell of acetone, and this substance may 

 also be present in the urine. On the other hand, the diabetic state is 

 attended by diminished resistance of the tissues to infection. A pimple 

 may become a carbuncle ; a slight sore on the foot may give rise to a rapidly 

 spreading gangrene of the lower extremity ; tubercular infection of the 

 lungs spreads rapidly to the whole organ so as to stimulate pneumonia. 

 The patient may thus die of some such intercurrent infection before the 

 onset of coma. In a few cases the pancreas is found to be atrophied or 

 diseased, but in the large majority no marked pathological change is to be 

 observed in this organ. Yet the condition is essentially similar to that 

 which occurs in pancreatic diabetes. The radical defect is the inability, 

 relative or complete, of the organism to assimilate carbohydrate. We 

 may find all grades between such cases and those in which there is still a 

 considerable power of assimilation. In order to determine the grade of the 

 disorder, it is usual to give a test diet with a certain proportion of carbo- 

 hydrate, e.g. 100 grm. of bread with meat, bacon, eggs, butter, green 

 vegetables, cheese, lettuce, coffee and wine. If the urine remains free 

 from sugar on this diet, the diabetes is mild in character. More bread 

 may then be added to the diet from time to time until sugar appears in 

 the urine and the limit of tolerance for carbohydrate has been reached. 

 In many cases the sugar will disappear from the urine on the administration 

 of a diet consisting entirely of proteins and fats. When this has been 

 effected, carbohydrates may be added in small proportions to the diet until 

 the limit is found at which the assimilatory powers of the patient are reached. 

 It seems that administration of any carbohydrate in excess of this limit is 

 )f disadvantage to the patient and hastens the progress of his disorder, 

 icn the power of assimilating carbohydrates is entirely abolished, the 

 )gnosis is almost absolutely fatal. This point may be determined in 

 70 ways. In the first place, a patient with no power of carbohydrate 

 dmilation will continue to excrete sugar in the urine on a pure protein- 

 it diet, and the D : N ratio will be 2-8 or higher. Information may also 



