1 8 MAL-NUTRITION AND THE TEACHER'S 



some cases they are significant with regard to their probable 

 errors. Professor Pearson concluded that there is a slight 

 correlation between intelligence and size of head, but that 

 this correlation is so small that it would be absurd to 

 endeavour to predict the intellectual ability of an individual 

 from his or her accurate head measurements. It is clearly 

 impossible to detect such slight differences by the rough 

 and ready methods adopted by Dr. Warner. 



Now although the mean values differ to such a small 

 extent, it is possible that mental dullness might be asso- 

 ciated both with very large and with very small heads. We 

 must then compare the variability of the ' very dull ' with 

 that of all the children. These results are also given in 

 Table H, and it will be seen that only in two cases out of 

 eight is the difference more than twice its probable error, 

 and even in these cases we find a possibly significant 

 difference among the girls associated with a non-significant 

 difference among the boys. Dr. Warner's conclusion that 

 mental dullness is closely associated with size of head may 

 be rejected, therefore, as arising solely from the methods 

 employed. 



His treatment of shape of head deserves further mention. 

 Among cranial * defects ', Dr. Warner includes cases where 

 crania are 'dolichocephalic'. Now 'dolichocephalic' is a 

 perfectly definite term. It is applied to heads whose 

 cephalic index is less than 75. Dr. Warner, however, 

 applies it to 'Heads long in antero-posterior diameter', 

 i. e. he applies the term not to an mdex, but to a lengthy 

 and further does not say within what limits crania are to be 

 regarded as dolichocephalic. It is interesting to note that he 

 applies this term to only 81 children out of 100,000, to 

 •13 % of the boys and to -026 % of the girls. Giving the 

 usual meaning to the term, Professor Pearson in the investi- 



