MISCELLANEA. 



Note on the Influence of Change in Sex on the Intensity of Heredity. 



By frank E. LUTZ, M.A. 



(1) In Professor Kail Pear.son's mcnioir, On tha Inhentance of Eye Colour in Man* it 

 is stated on page 117 " that the youuger gencration is more highly correlated with an ascendant 

 or collatcral of the s;ime than of the opposite sex," and again page 115, "that change of sex 

 weakens the intensity of heredity." It secnied worth while pursuing this point further and 

 grouping togetlier all the available evidence as ftvr as eye-colour in man is concerned. Aocord- 

 ingly, Mr tJalton having again niost kindly put his eye-colour data at Prof. Pear.son's 

 disposal, I have worked out the correlation between great-gi'andparents and oflspring, which 

 completes as far as ancestral correlation is concerned the whole of the available records of this 

 sort. I hope shortly to examine in the same way a large mass of data which I have gathered 

 concerning coat-colour in dogs of the l'ointcr breed. 



I u.se the Ibllowing schenie of notation for pedigree : 



0(J 1$ Oftspring 



, mczr ^ 



2 (J 3 9 Parents 



_i_ 



~i r- -1 



4(J 5? es 79 Grandparen ts 



, 1 T I 1 1 I ' n I ' -1 



8cJ 99 lOcJ 11? 12(J 13? US 15? Great-grandpareuts 



Thi.s only difi'ers from JIr Galton's widely accepted schenie in the introduction of for the 

 niale oftspring to distinguish theui from 1, the female offspring. /■„, will mark the correlation 

 between a pair of relatives one of whom is in class «, the other in ulass s' ; and a reference to 

 the above scheme will at once show us which pair of relatives is being dealt with. Thus ;•„, ,._, is 

 the correlation of male great-grandchildren with the father of the maternal grandfather, and 

 so on. 



(2) Unfortunately the data for great-grandparents is not very extensive. We had altogether 

 only 2996 cases, or about 187 for each of the 16 possil.>le great-gi-andparental correlations. This 

 is distinctly too few considering the magnitudc of the probable errors which occnr when we de;d 

 with quantities not quantitatively measurablet. Accordingly I paid no attention to the 

 Separation of sex in the oftspring but grouped my great-gi'andparental data into four clas.se.s : 

 (a) those in which there is no change of .sex between great-grandparent and oflspring, (6) those 

 in which there is one change of sex, (c) those in which there are two ohanges, (d) those in which 



* Phil. Trans. A, Vol. lü.5, p. 102 et seq. 

 t Phil. Tr,u,.'<. A, Vol. l'.tr,, p. 42. 



