The time at which the maximum effect appears is likewise of great importance. An examination of the figures given in the last column shows that in all but a few instances the highest value appeared in the first 2 hours. The most notable exceptions to this are the experiments with Dr. R., February 21, 1907, with the maltose-dextrose mixture; H. R. D., April 21, 1906, with bananas and sugar; and A. H. M., April 10, 1907, with popcorn.

The general conclusions from this series of calorimeter experiments would therefore be that the ingestion of pure or nearly pure carbohydrate produces a positive increase in the metabolism which for short periods, at least, may amount to almost 35 per cent, and that this increase nearly always takes place in the first 2 hours of experimentation. An examination of tables 101 to 122 shows clearly that results obtained in the periods subsequent to the first 2 hours of the experiment give very little, if any, evidence as to the nature of the metabolism, save that a persistent increase in the carbon-dioxide production is usually found. It is of particular significance, however, that in the majority of the experiments direct calorimetry shows definitely an increment in the heat production due to the ingestion of carbohydrates; we can therefore consider this fact as established. The value of this known fact will be more apparent when an analysis is attempted of the intermediary processes involved in the metabolism of carbohydrate.

On the other hand, we are not able from these calorimeter experiments to determine with great exactness the time relations between the ingestion of carbohydrate and the metabolism. While they bring out the fact that the maximum effect of carbohydrate ingestion appears in the first 2 hours and that thereafter practically no effect is noted in the majority of instances save in the production of carbon dioxide, experiments made with shorter periods are absolutely essential for a more careful analysis of the relationship. For these determinations in shorter periods recourse must be had to the long series of observations in the respiration experiments with carbohydrates, from which an estimation may be made by indirect calorimetry of the course of the metabolism after carbohydrate ingestion.