622 VITALITY AND EFFICIENCY WITH RESTRICTED DIET. 



The average variability of the two squads for right and left move- 

 ment is presented in figure 118. Squad A appears to show a gradual 

 increase in the variability, which is high for December 9, January 13, 

 and January 27, except that on the latter date the variability for R 

 was decreased. In the case of Squad B the variability is larger on 

 November 18 than for the other two normal dates. It also rises at 

 the point of January 20 during the low-diet experiment. Squad B 

 shows no considerable increase in the variability at their first low-diet 

 date. This would appear to agree, also, with Squad A. Since the 

 first date for Squad A, October 28, was during the early part of the 

 low-diet period and showed a variability of nearly 7 per cent, we judge 

 this to be about as small as could reasonably be expected. 



Associated with the prolonged reduced diet the eye movements show 

 a slower speed in the case of Squad A amounting in the extreme to ap- 

 proximately 5 per cent for averages of movements right and left. With 

 this there is also an increase in the variability. The results for this 

 measurement found on Squad B confirm the direction of those observed 

 with Squad A. 



(15) SPEED OF THE FINGER MOVEMENTS. 



In this muscle coordination test the number of complete oscillations 

 of the finger which the average individual can perform in 10 seconds 

 ranges between 55 and 75. In the normal series of 1917 finger-move- 

 ment records were obtained on 61 of the men, the records were photo- 

 graphic, and were extended for only 8 seconds. The average for the 

 group of 61 normal men in the 1917 series on whom such records were 

 taken was 52.1 complete oscillations in 8 seconds. The records made 

 in the evening for Squads A and B are given in tables 182 and 183. 

 The average performance on the normal date (September 29) for Squad 

 A was 68.7 oscillations; the average for the first experiment with Squad 

 B was 66.0 oscillations. The foregoing values are for records 10 

 seconds long. If we count only the first 8 seconds so as to make them 

 comparable with the normal series of 1917, the initial normal results 

 are 55.7 and 53.7 oscillations for A and B, respectively. Thus the two 

 groups of men agree fairly well with each other and with the larger 

 normal group. With Squad A the individual normal averages range 

 from 52.3 oscillations for Tom to 84.6 oscillations for Gul This last 

 figure is unusually high. Gul has previously been noted as a rather 

 short, extremely intense individual. With Squad B the range for the 

 first experiment for the individual subjects is smaller, being from 58.2 

 for How to 72.5 for Wil. The members of A show individual averages 

 exclusive of the first (normal) experiment, which range from 56.5 oscil- 

 lations for Moy to 77.5 oscillations for Gar. The average for the 10 

 subjects for the nine evening experiments is 65.1 oscillations. Squad 

 B show individual averages for their five normal evening experiments 

 which range from 58.6 (How) to 75.2 (Van), with a total average of 



