474 VITALITY AND EFFICIENCY WITH RESTRICTED DIET. 



than it was on the normal day. Thus with all but Ham, Liv, and 

 Sne, slight falls in the respiration rate occur at the end of the reduced- 

 diet period. The three subjects named show positive rises, par- 

 ticularly Liv. Since Lon was not at the Laboratory on the normal 

 day (January 6) we have for him only a record for reduced diet. The 

 respiration rate on this day was not far from the average found with 

 Squad A, but distinctly lower than that found with most of the other 

 men in Squad B. The three very low rates of 9, 9, and 8 with Fis, Har, 

 and Tho at the end of the reduced-diet period have special interest. 

 These are lower than any rates observed with Squad A in the standing 

 position and even lower than those observed with the men of Squad A 

 in the lying position at the end of the research. No explanation for 

 these low values is at hand other than changes in diet. 



RESPIRATION RATE DURING WALKING. 



Finally, we have rates obtained inamediately prior to, during, and 

 after walking. Inasmuch as the subjects during the walking experi- 

 ments were confined inside an air-tight metal box, it was impracticable 

 to use the ordinary methods of graphic registration to determine the 

 respiration rate. Recourse was thus had to the special technique 

 outlined on page 130 of this report. Although in a preliminary test 

 it was possible so to adjust the technique for the individual subjects 

 as to obtain most satisfactory records, in the actual experiments we 

 were able to adjust the subject for only a moment, and the experi- 

 ment was then immediately begun without opportunity for subse- 

 quent adjustment. As a result, much to our regret, a considerable 

 number of the records are illegible. This frequently happened with 

 the records of a particular subject, indicating that the difficulty was 

 in the adjustment for that especial man. 



The respiration rate is commonly recorded in respirations per minute, 

 and we have followed this custom in presenting the previous data. 

 In the records with the treadmill chamber, however, we were unable 

 to count the number of respirations, as was possible from the graphic 

 records obtained from the pneumograph around the chest in the 

 experiments with the respiratory-valve apparatus, or to record the 

 rise and fall of a spirometer as in the experiments with the portable 

 respiration apparatus. It was necessary to depend here upon short 

 photographic records; consequently the actual number of respiration 

 cycles counted was, at times, rather few and the few records had to 

 be multiplied to bring them to the per minute basis. While the pulse- 

 rate may be advantageously counted for 30 seconds or even as low as 

 20 seconds, and raised to values per minute without serious error, 

 it is a legitimate question as to whether or not it is possible to count 

 4, 5, or even less respiratory cycles, determine the length of time photo- 

 graphically to 0.01 part of a second, and then raise this to a rate per 



