Normal and diabetic blood sugar curves have been shown in Part I at some length, so that in this paper miscellaneous conditions of altered carbohydrate metabolism are condensed into the least possible space. As the aim is conciseness and convenience of reference rather abrupt transitions are unavoidable. The test meal was 100 gm. glucose, except in a few cases, whose inclusion was thought legitimate because the curves were, in general, as high as after the usual dose. Table 2 supports those who have asserted the relative frequency of hyperglycemia in renal glycosuria. For example, in the one hour column, which presents the largest number of figures for analysis, i. e., sixty-seven, as much as 13 per cent. exceeded 0.16 and 40 per cent. exceeded 0.11. On the other hand, the diagnoses are not adequately proven in all the original articles; hence the data are reproduced to indicate possibilities rather than