302 METABOLISM DURING WALKING. 



Figure 41 gives 5 curves of the duration of the pulse-cycles on Feb- 

 ruary 29, each point, as in figure 40, being the average of 2 cycles. 

 Curve A represents pulse-cycles with the subject standing in the second 

 period of the day, and exhibits like variations to those referred to in the 

 discussion of other curves. In the corresponding curve A of figure 40, 

 although the individual pulse-cycles were measured, the time-intervals 

 on the photographic paper were small, and it was suggested that some of 

 the irregularities in the curve may have been due to errors of measure- 

 ment. In curve A of figure 41, however, the time-interval was 

 lengthened and this source of error was thus largely removed, but the 

 same variation in the pulse-cycles for standing is apparent. The 

 maximum variations during this record are as large as 0.4 second, with 

 0.3 second as the greatest difference between two successive points. 

 In curve B, the time-interval was not lengthened and the errors in 

 estimating the fractions of a second are therefore greater. The stand- 

 ing portion of the curve, as in the curves earlier discussed, shows gross 

 variations in the cy }le duration. On the transition to walking there is 

 a lengthening of pulse-cycle duration which is directly at variance with 

 the other records given in figures 39 and 40. The question may fairly 

 be raised if the time of transition were correctly indicated on the 

 photographic paper and if it may not have been a second or two later. 

 As in the preceding figures, the major portion of the rise in the curve 

 had taken place by the measurement of the fortieth or fiftieth cycle, 

 with an average cycle duration at that time of 0.5 second. Thereafter, 

 the change in the duration was but slight, reaching an approximate 

 value of 0.45 second at the end of the curve, which covers a time-inter- 

 val of 65 seconds. 



After the subject had been walking 26 minutes, a short record was 

 made (curve C) in which the tune-intervals on the record were 

 lengthened by increasing the speed of feeding the paper to the camera. 

 It may be noted that in this curve (C), the duration of the pulse-cycle 

 had further shortened to 0.36 to 0.39 second, and there is less variation 

 between the readings. This is in harmony with curves D and E of 

 figure 40, which illustrated the regularity of the pulse-cycle duration 

 after a continued period of exercise. Curves D and E in figure 41 

 will be considered subsequently, as they represent records taken when 

 the man was standing after walking. 



From these measurements of the durations of the pulse-cycle in the 

 transition from standing to grade walking, there is evidence that the 

 standing pulse varied widely between successive cycles; that the inter- 

 val preceding walking is likely to be influenced by psychical effects 

 due to the anticipation of the starting of the treadmill; that most of the 

 rapid shortening of the cycle duration after the beginning of walking 

 occurs within 25 or 30 cycles, or about 15 to 20 seconds, and is over 

 within 1 minute ; and that the shortening of the duration thereafter is 

 very gradual and may continue for a period of 25 to 30 minutes. 



