384 



On titc Luiis of J nheritam-e in Man 



direct coefficients. We Iiave tlius, if it were neeiled, still t'iirther eviilence that 

 the original estimates of the slreiigtli "f ln-redity were far too low. 



TA ULK VI. 

 Gross Farentul Heredity Coefficients. 



Oue of the inost difficult poiiits to be sure about is the theoretieal reiati'iuship 

 which is to be e.xpected between tiic iiiteusities of direct and cross iiiheritauce. If 

 (i) all Organs and charactens were inlierited at the sanie rate, and (ii) the organic 

 correlations in younger and older generatious were the same, and (iii) the 

 variabilities of tliese generations, as measnred by their coefficients of variability, 

 were the same, then it follows that tlie moan of two corresponding coefficients of 

 cross hercdit}' is the product of the coefficient of direct herodity into the organic 

 correlation *. But none of tlic three conditions stated above is accurately fultillod, 

 as we have seen, in the pn-seiit inatcrial. Notably wo tind sensible divergence 

 froni the first. We niay possibly atteiiipt to allow for the lii^st distiubiiig factor in 

 the foUowiiig niauuer: the cro.ss-correlatious siiuuld vanish (d) wheii tiie direct 



ß. S. I'roc. Vol. 62, p. 411. 



