596 VITALITY AND EFFICIENCY WITH RESTRICTED DIET. 



ment showed a reflex time lengthened to 42 a. The latency was also 

 longer than normal in the next three experiments, October 27, 

 November 10, and November 24. On December 8 it was the same 

 value as the normal, but on the succeeding four dates it was longer, 

 conspicuously so on January 26. Squad B began with a longer 

 latency than A. The first two experiments separated by 1 month 

 show the same latency, 37 <r. At the next date this was lengthened 

 to 44 0-, a rather conspicuous and unaccountable change.^ On De- 



SquadA 

 Squad B 



Fig. 111. — Patellar reflex averages. 



h, latency; A, amplitude, and S, refractory period separation. The scale at the left is for 0.001 

 second, that at the right is for millimeters in the case of A and for 0.01 second for S. Solid 

 lines represent Squad A, and broken lines Squad B. 



cember 15 and January 5 the latency was slightly shorter than on the 

 previous days, October 6 and November 3. On January 13 it was 

 at the initial level, that is, 37 <j. This was the first low-diet date. 

 On the next two, January 19 and 27, the latency was longer than 

 at any other point save November 17. 



The curves for the amplitude of the reflexes tend in each case to be 

 the complement of those for the latency. When the latency is long 

 the amplitude is small, and vice versa. Squad A shows a decrease in 

 the amplitude for October 13, 27, and November 10. This accom- 

 panied the increase in latency. Squad B likewise shows a decrease in 

 amplitude on November 17 with their associated latency of 44 a, but 



^See discussion of November 18 in relation to speed of eye movements, page 621. 



