334 



SLEEP-WAKEFULNESS CYCLE 



by one long period of sleep, the percentages 

 varying from 7 to 17, but fully two-fifths of 

 the sleep periods in Section IV were of the 

 single continuous type. Dividing the sleep 

 into two equal parts was not a popular pro- 

 cedure, but Sections I and II chose that 



HQO/f5 



Fig. 2. Mean percentages of the entire personnel 

 of the DOGFISH who were eating, reading, writ- 

 ing, or playing games during every hour. Data 

 are based on ten 24-hour periods. Note the dis- 

 parity in scale between the upper and the two 

 lower parts. 



arrangement over twice as often as Sections 

 III and IV. Taking a nap, in addition to 

 one long sleep period, seems to have been the 

 favorite solution for the three watch sec- 

 tions, and even Section IV indulged in a 

 single additional nap nearly as often as in 

 only one uninterrupted sleep. Finally, there 

 were some cases where three or more periods 



of sleep occurred in a 24-hour period (13 

 percent for the entire personnel). 



The mean duration of sleep ashore was 

 estimated at about 8 hours per night (Table 

 I). The estimates of the total duration of 

 sleep per 24 hours on the submarine aver- 

 aged 8| hours, but data indicate that the 

 mean duration was underestimated by more 

 than half an hour. Peculiarly, the mean 

 durations of sleep were closest for Sections 

 III (9.16 hours) and IV (8.96), and more 

 than one hour apart for Sections I (8.56 

 hours) and II (9.58 hours). Perhaps the 

 ability of Section II to command the usual 

 night hours of sleep (0-8), and also take an 

 afternoon nap, is responsible for the highest 

 mean duration of sleep for this section. The 

 ranges of the mean total durations of sleep 

 for individuals were about the same in the 

 four sections, actual extremes being, in 

 hours, 7.22-10.44, 7.39-11.28, 6.11-11.28, 

 and 6.78-11.50. Single continuous sleeps of 

 12-14 hours' duration were not uncommon 

 in Section IV, the only one in which that 

 length of sleep was permissible. Comparing 

 each individual's actual mean duration of 

 sleep with his own estimate of that duration, 

 it was found that two-thirds of the men 

 underestimated and one-third overstimated 

 that figure. 



Meals and Recreation 



The tendency of wakefluness to be con- 

 centrated during the afternoon and evening 

 hours was reflected in the incidence of food 

 intake and recreational activities. Regular 

 meal hours (breakfast before 8, dinner be- 

 fore 12, and supper before 20 o'clock) stand 

 out as spikes in the low curve of food intake 

 (Fig. 2). Whereas fully 80 percent of the 

 personnel showed up for supper, only 65 

 percent ate dinner, and less than 55 percent 

 got up or remained up for breakfast. Other- 

 wdse there was always a spurt in eating at 

 the time of, or shortly after, watch changes, 

 particularly by those who had previously 

 missed the regular meal. Reading and, to 

 a much smaller extent, writing were also 



