THE METABOLISM OF THE CARBOHYDRATES 659 



thesia. Such observations have been made on rabbits, and a few on man 

 himself. By collecting blood from the ear veins of rabbits, it has been 

 found that, after giving from two to ten grams of glucose by stomach, 

 the glucose concentration of the systemic blood begins to rise in fifteen 

 minutes, attaining a maximum in about an hour and then returning to 

 the normal level in about three hours. 



Similar results have been obtained by examination of the venous blood 

 in man. After giving 100 grams of glucose by mouth, for example, there 

 is commonly an increase in blood sugar amounting to from 30 to 34 per 

 cent of the normal and lasting for from one to four hours. The existence 

 of this postprandial hyperglycemia, as we may call it, indicates that the 

 sugar-retaining powers of the liver and muscles are not sufficiently de- 

 veloped to prevent the accumulation of some of the absorbed sugar in the 

 systemic blood. Whenever this increase exceeds a certain limit, some of 

 the sugar begins to escape through the kidney into the urine, producing 

 glycosuria postprandial glycosuria. The percentage of blood sugar above 

 which glycosuria occurs is, in the case of man, probably about 0.10 to 0.11 

 gm. per cent. After damage to the kidney, as in nephritis, or in long-stand- 

 ing cases of mild diabetes, the percentage may probably rise considerably 

 higher in the blood without evidence of glycosuria. 



Value of Blood Examination in Diagnosis of Diabetes. The determina- 

 tion of the amount of ingested carbohydrate required to bring about post- 

 prandial glycosuria constitutes, as we have already seen, the so-called 

 assimilation limit for sugar, which is often taken as an index of the sugar- 

 metabolizing power of the organism. It is evident, however, that the time 

 of onset, and the extent and duration of postprandial hyperglycemia must 

 serve as a more certain index of the sugar-retaining power of the liver 

 and muscles; and now that a simple and rapid clinical method exists 

 (Lewis-Benedict method) for the accurate determination of sugar in small 

 quantities of blood, there is no reason why this index should not be used 

 for the detection of failing powers to metabolize carbohydrate. 



In no disease, probably not even in tuberculosis, is it more important 

 than in diabetes that an early diagnosis should be made. Thus, if we find 

 that the postprandial hyperglycemia after a certain amount of carbo- 

 hydrate develops to an unusually high degree and persists for an unusual 

 length of time, we are justified in curtailing the carbohydrate supply so as 

 to hold these values down to their level in normal individuals. It is almost 

 certain that the first sign of diabetes is an unusual degree and duration 

 of postprandial hyperglycemia. At first the excess of sugar leads to no 

 damage and it is insufficient to cause any evident glycosuria, although it is 

 quite likely that if the urine in such individuals were collected at very 

 frequent intervals after eating carbohydrate-rich food, glucose would be 



