164 EEPOKT ON BONES FROM HARLYN BAY. 



old women by quite two-fifths of an inch. ; whence I conjecture 

 that these same okl women had lost about an inch and a half of 

 their adult stature. Similarly with the humerus. Ten men 

 over 70, and averaging 1683 mm. of corpse-length, had humeri 

 of 336'8 mm., while ten men under 47, and averaging 1683 

 mm., had humeri of 328'4 mm. only, /i.e., about ^-inch shorter, 

 which would correspond to a shortening of stature in the old 

 men of about 1| -inches. 



It is true, as Professor Pearson points out, that in ancient as 

 in modern burying-grounds, we often have to deal with a large 

 percentage of old j)ersons ; but this consideration does not seem 

 to me to be of much significance, seeing that the average stature 

 we desire to ascertain is not that of the immature or the decrepid, 

 but that of developed adults before they begin to decline. 



There is another drawback to Professor Pearson's system, 

 which it shares with all the others. It is based upon corpse- 

 length, and we do not — I may almost say we cannot — know what 

 is the true relation between corpse-length and living stature. 

 We know that a living man measures more in the horizontal than 

 in the vertical position ; but that is nearly all we know upon the 

 subject. Provisionally, we may adopt M. Manouvrier's estimate, 

 viz. : that the corpse measures about two centimeters, or -^-^ of 

 an inch more than the erect and living body. 



Pearson's plan is based on formulae of regression for the 

 organs concerned. I confess that I am not mathematician enough 

 to follow him always. But he is the only one of us who has 

 applied a competent knowledge of mathematics to the questions 

 dealt with; and he has certainly worked them out more 

 elaborately than any previous writer. On the qualifications 

 necessary to be applied to Eollet's material, such as the modes of 

 measurement, the thickness of cartilages, the dextro-sinistral 

 differences, the effect of drying on the length of bone, he is 

 extremely good. Manouvrier and he have worked out, both 

 what the former calls the paradox involved in the varying 

 proportion of increasing stature to bone-length, and the analogous 

 one affecting the stature of women. This second paradox consists 

 in the fact that whereas, speaking generally, in women as 

 compared with men, the body is long and the limbs are short, 

 nevertheless, any given length of femur usually indicates a 



