626 VITALITY AND EFFICIENCY WITH RESTRICTED DIET. 



of the morning record being 2.3 oscillations. This difference of ap- 

 proximately two oscillations applies throughout the records, the morn- 

 ing records always being the higher for Squad A. With Squad B, if 

 we compare the evening of November 3 with the morning of November 

 4, the scores are, respectively, 65.2 and 66.9 oscillations, a difference 

 of 1.7 oscillations in favor of the morning. Throughout the measure- 

 ments for Squad B, the morning tests give a higher score by approxi- 

 mately 1.5 oscillations. The two series of finger movements are thus 

 in entire contrast to the results found in the strength of grip test in 

 which the evening performance was the better. (See p. 585.) 



The individual averages in tables 184 and 185 call for no special 

 comment. The group results are plotted in figure 120. Squad A 

 shows a continual decrease from October 28 to December 9. In the 

 evening records, December 8 also showed the poorest record, although 

 the curves were fairly regular up to that time. It may be assumed that 

 the score for October 28 is a little high if the curve for the evening records 

 is taken as a basis, for on the previous evening, October 27, the score 

 was higher than it had been previously by about 1 oscillation. No 

 finger movements were taken on the morning of December 20 on 

 account of the electrocardiograms. There is a high point in the curve 

 for January 13 which corresponds well with that found for the evening 

 of January 12; following this there is a decline on January 27 which 

 is also found on January 26. No records could be taken on the last 

 morning, as the time was given up almost entirely to the walking 

 experiment. The records for Squad B, also shown in figure 120, 

 demonstrate a gradual decline from November 4 to January 14, the 

 first food-reduction experiment. On January 20, the second experi- 

 ment which came within the food-reduction period, there is a pre- 

 cipitous drop of 4 oscillations from the preceding record of 65.9, which 

 is a decline of 6 per cent. Finger movements could not be taken on 

 the final morning, January 26, on account of the walking experiment. 



The finger movement records were counted in blocks of 2 seconds.^ 

 For example, the records of Gul in the evening experiment of September 

 29 for the five 2-second intervals,^ were 18.6, 16.7, 16.7, 16.4 and 

 16.2 oscillations in each succeeding 2-second period, with a total of 

 84.6 oscillations for the 10 seconds; (see table 182). Comparison of 

 the results in such short and successive intervals may be counted 

 on to show the influence of fatigue which, however, in the 10-second 

 test, does not develop to an unpleasant degree. The records, tabulated 

 as illustrated for Gul on the one date, are wonderfully uniform in the 

 amount of decrease from one 2-second period to the next; in conse- 



^The tables for finger movements given by Miles, Carnegie Inst. Wash. Pub. No. 266, 1918, 



p. 86, are in this form. 

 ^These 2-second intervals were accurately marked on the record by the timed jump spark 



from the recording point (see p. 187). 



