BIOMETRICS 



145 



number of men in percentages who exerted a given pull 

 (equal distances being marked off on the base line for equal 

 numbers of pounds, and in the same way equal distances of 

 the vertical line for each percentage). For instance, 2 per 

 cent, show a strength of pull of under 50 pounds, 8 per 

 cent, of between 50 and 60 pounds, 27 per cent, of between 

 60 and 70 pounds, etc. If we connect the tops of these 

 columns, we get a broken line representing the curve of 

 variation for the given measurements of strength of pull. 

 We see at once that the number of persons showing the 

 two extreme amounts of strength, the very weak and the 

 very strong men, is smaller than of those exhibiting a 



r-»o7o 



30 



20 



10 



•be. 10 20 30 ^ 30 60 70 80 90 100 



Fig. 62. — Diagram of Strength of Pull. (After Galton.) 



medium strength of pull, and that the curve gradually 

 slopes down towards both ends. 



If we have more exact measurements from a greater 

 number of people, the curve becomes a more regular flowing 

 one, as in Fig. 63, where variations in stature are repre- 

 sented, the result of 4,426 measurements of members of 

 Cambridge University of British extraction, recorded by 

 the Cambridge Anthropological Society. The base line 

 gives the stature in inches, the vertical line the number of 

 individuals exhibiting the different heights of stature. 

 Here the continuous line going through the points of 

 measurements plotted out in the diagram represents very 

 nearly a curve, which is identical with what is called in 



19 



