Alice Lee 
211 
Cuttinsr off all bones over 45"5 we find 
= 2-6392, (Z = 2-6264, 
leading to -^/^i = '3826. 
Hence h' = -984, ->/r, -783 and s^r, = 1-195. 
These provide for the non-trimcated population. 
Mean 42-48, s.D. = 2-056, i\^=104, 
which are in still better agreement with Parsons' constants for the 105 bones than 
the constants for the 85 bones were for their series. It would appear therefore that, 
if we suppose all bones under 45 female and use our Tables, we get results in 
reasonable accordance with Parsons', and possibly by a theoretically more justifiable 
method than endeavouring to sex the bones above 44 and below 48 from other 
characters. 
We have considered from the same aspect the character breadth of lower 
articular end of femur. Pai-sons' distribution of 89 female femora is as follows 
(p. 257): 
Breadth in mm. 
00 
61 
62 
63 
64 
65 
66 
67 
68 
69 
70 
71 
73 
74 
75 
Frequency 
1 
5 

e 
8 
6 
17 
13 
13 
8 
6 
3 
1 
1 
1 
In this Table he has only one bone in excess of the numbers on which he 
bases his means on pp. 250 — 1. If we truncate at 69-5, i.e. reject all bones over 
69 mm., we find 
= 3-9803, 2-9058, 
and -i/ri = -47l4. 
Hence we deduce 
/t' = -5295, f,= -977, f,= 14^26. 
These lead to 
/t = 1-503, (7 = 2-839, i\r=98-4, Mean = 68-00 mm. 
The actual values given by Parsons' distribution above are 
o- = 2-571, i\^=89, Mean = 67-54 mm. 
Thus the agreement is not nearly so good as for the diameter of the head ot 
the femur, being about 10 wrong in a and iV. It should give as good a result 
if the method were quite satisfactory, for the bones have been sexed by the 
diameter of the head, and the limit 44 mm. for diameter of the head corresponds 
fairly closely to 69 mm. for the breadth of lower articulation. 
As this paper is not intended as a discussion of Parsons' data, to which we 
hope again to return, we will only deal with one more illustration of the use of 
