Miscellanea 
233 
Thus the best fitting curve to the series is an unsyrametrical curve of Type IV. and calcu- 
lating the constants of the curve in terms of the moments, its equation is found to be 
-/ = 19-57 1 + 
-4-8529 4 . 5 744tan-i 
22-319, 
The origin of the curve is at a pulse of 52-99. 
For the case of the Breathing distribution we find 
4-7243 . (See tig. 1.) 
-1608, 
/x 2 = 5-5183, 
/* 3 =+ -7574, 
^4 = 110-4248, 
ft = -0034, 
ft= 3-6262, 
k =+-0021, 
Mean breathing = 15-839 per minute, 
Standard Deviation = 2 - 349. 
This time we see that the best fitting curve is symmetrical and of Type II. Calculating the 
constants we have as the equation to the curve 
/ r 2 \ -7-2908 
3^46-04 (l+ 6Wo7 ) • (See fig. 2.) 
The series for this case is very irregular and it would not be possible to find any curve which 
would give a very close fit. Doubtless this is largely due to the small number of observations, 
but again the nature of the irregularities suggests that there may possibly be some unsuspected 
source of heterogeneity in the data. 
My thanks are due to Miss H. Gertrude Jones who has very kindly drawn the diagrams. 
TABLE I. 
Stature in cms. 
Oh 
45 
49 
53 
57 
61 
65 
69 
73 
77 
81 
82— 85 
86— 89 
90— 93 
94 
98 
102 
106 
42- 
46- 
50- 
54- 
58- 
62- 
66- 
70- 
74- 
101 
105 
109 
Totals 
1 
2 
6 
9 
13 
12 
7 
3 
3 
1 
57 
1 
6 
14 
12 
24 
14 
6 
6 
1 
3 
2 
1 
10 
15 
10 
13 
6 
3 
5 
1 
67 
1 
3 
4 
10 
4 
5 
2 
2 
1 
32 
Totals 
2 
4 
11 
36 
47 
53 
45 
22 
15 
11 
5 
253 
Two additional observations of Pulse fall into the groups 58 — 61 and 66 — 69 respectively. 
They do not appear in this table as the corresponding Stature was unknown. 
Biometrika viii 30 
