K. Pearson and A. Lkk 



:57!) 



I tliink \ve iiiay t'airly take tho intensity (jf iiihcritance for measurable 

 characters in man to be ■•tG, or cvcn for rougher work '5. It may bo as well to 

 put on rccord here the princiijal results for hercdity in tho dii'ect line so far 

 rcafhed. I oinit the results obtained in niy niemoir of 1895*, for I consider iny 

 presc'iit data to replacc that series. 



TM'.LE V. 



ParenUtl Inlieritauce in Differoü Species. 



I consider that this table contains the inost reliable data \ve yet have collected 

 and reduced for parental influence on offspring. 



The general mean of the whole of these series is '48 and so far as we have yet 

 gone, we may I think conclude, that : 



(n) Tliere is no reason for supposing parental heredity to be stronger in one 

 species than a second. 



{h) Its values lie between •42 and '52 and clustor round •48. 



Thus for most practical purposes we may assume parental heredity for all 

 species and all characters to be approximately represented by a correlation of 'ö. 



In the conrse of the past 8 years many cases of parental inheritance have been 

 worked out by the biometi'icians associated with me at University College, sorae of 

 the most important of these are still unpublished, others have been replaced by far 

 more reliable data ; in further cases we kuow that the material was doubtful, e.g. 

 the cephalic index iov fathers and children of the North American Indians, or sire 

 and offspring in the Basset Hounds. In such cases better material has been 

 sought and our first results modifiedf. But in the present controversial phase of 



♦ Phil. Trans. Vol. 187, p. ■253 et seq. 



+ For example the greyhounds have shown that anomalie,s of the Bas.set Hound results were peculiar 

 to the material, the cephalic index is inherited quite normally when we test it on material with reliable 

 parentage, etc. etc. 



