vii.] DISCUSSION OF THE DATA OF STATURE. 103 



relation between the major and minor axis of the 

 ellipses was changed 3 per cent. ; and their inclination 

 to one another was changed less than 2 . 1 



It is obvious from this close accord of calculation 

 with observation, that the law of Error holds through- 

 out with sufficient precision to be of real service, and 

 that the various results of my statistics are not 

 casual and disconnected determinations, but strictly 

 interdependent. 



I trust it will have become clear even to the most 

 non-mathematical reader, that the law of Eegression 

 in Stature refers primarily to Deviations, that is, to 

 measurements made from the level of mediocrity to the 



1 The following is a more detailed comparison between the calculated 

 and the observed results. The latter are enclosed in brackets. The letters 

 refer to Fig. 11 : 

 Given 



The "Probable Error "of each system of Mid- Parentages = 1-22 

 inch. (This was an earlier determination of its value ; as already said, 

 the second decimal is to be considered only as approximate.) 

 Eatio of mean filial regression == . 

 " Prob. Error" of each Co-Fraternity = 1'50 inch. 



Sections of surface of frequency parallel to XY are true ellipses. 



(Obs. Apparently true ellipses.) 

 MX : YO = 6 : 17 '5, or nearly 1 : 3. 



(Obs. 1 : 3.) 

 Major axes to minor axes = J 7 : J 2 = 10 : 5-35. 



(Obs. 10 : 5-1.) 

 Inclination of major axes to OX = 26 36'. 



(Obs. 25.) 

 Section of surface parallel to XZ is a true Curve of Frequency. 



(Obs. Apparently so.) 

 " Prob. Error ", the Q of that curve, = 1.07 inch. 



(Obs. 1-00, or a little more.) 



