Note 32 — I. EXPERIMENTS WITH PROFESSIONAL MEN.

All foodstuffs were analyzed from large samples, to diminish as much as possible the errors of analysis. Nitrogen was determined by the Kjeldahl-Gunning method, the figures given being the average of closely agreeing duplicate analyses.

While nitrogen was thus determined in every sample of food by direct chemical analysis, the fuel value of the food was calculated mainly by use of the data furnished by the Bulletin issued from the U. S. Department of Agriculture, Office of Experiment Stations. No. 28