IX— X] 



Introduction 



XXV 



Values of the "generalised probable error" coefficients are given in Table X 

 for n=l to 11, and by means of a smooth curve the results may probably be 

 extended to ?? = 1 5. The values found for this extension are : 





n=12 



71 = 13 



7! = 14 



n=15 



Xo 



3-367 



3-513 



3-654 



3-791 



Illustration (i). Let us consider long bone data for Frenchmen. 1 = F= femur, 

 2. = H = humerus, 'S = T = tibia, 4 = i2 = radius*, then by formula (xviii) p. xxiv : 



R = 



1, -8421, -8058, -7439 



•8421, 1, -8601, -8451 



•8058, -8601, 1, -7804 



■7439, -8451, -7804, 1 



Further in cms : 



m, = 45-23, 0-1 = 2-372, 

 mi, = 38 01, 0-0=1-538, 

 »i3 = 36-81, 0-3 = 1-799, 

 m, = 24-39, 0-4 = 1-170. 

 What is the chance that the following individual rnay be considered French ? 



i?"= 36-97, J?' =26-82, r = 30-56, i?' = 2068. 

 The deviations in terms of their standard deviations are : 



x^ = {F' - m,)la, = - 3-482, x, = (//' - m^)lo; = - 4-059, 

 x, = {T'- m,)/a, = - 3-474, a:, = (R'- m,)ja, = - S'lTl. 

 Further : 



% = 3-7810, 



R^ 

 R 



= 20231, 



-if 22 



ll 



Ru 

 R 



Ru 



= 6-5496, 

 = 1-1404, 



R 



= 4-3406, 



R 



'y' = 0-2130, 



Ru 

 R 



R23 

 R 



= 3-6508, 



= 2-1946, 



„ =2-3175, ^ = 0-6842. 

 Whence ;;^2= 16-741,035 and % = 4-0916, 

 n is even, hence : I^[N = — "ULLtl^ > = ^2-77 



m. 



* For particulars of these length measurements the reader must consult B. S. Proc. Vol. 61, 

 pp. 343 et seq. and Phil. Trans. Vol. 192, A, p. 180. 



B. d 



