ANTHROPOMETRIC INVESTIGATION IN GREAT BRITAIN AND IRELAND. 397 



the boys of St. Katheriaos School, into three divisions, according to 

 whether they possess a high, average, or low degree of mental ability, 

 and have compared these classes with one another. 



' From the post-mortem records I hope to obtain a formula, by which 

 the approximate braitv weight may be calculated from outside measure- 

 ments, and to ascertain the extent to which variation may occur from the 

 general mean, among cases which may be considered as normal. 



' 6. The majority of my measurements have been taken by one 

 assistant — a gentleman in whose accuracy and care I have every confi- 

 dence ; the post-mortem cases have been measured by a second assistant, 

 who is equally conscientious and trustworthy. 



' I have myself explained to them the method of using the instruments, 

 and the importance of accuracy in making and recording each measure- 

 ment. 



' 7. ^s far as they go, my figures appear to indicate a correlation 

 between large size of head and a high degree of mental ability, but there 

 are many exceptions to the rule. These I believe to be accounted for 

 chiefly by race differences. 



' The results which I have obtained in the diflferent groups appear to 

 accord very closely with one another, and with the results which have 

 been published by Alfred Binet in " L'Anne'e Psychol ogique," 1901 ; my 

 figures are, however, much smaller than those of Professor Binet, and 

 others which have been published in this country, and I shall therefore 

 most gladly co-operate with the Committee of the British Association by 

 allowing my figures to be used in combination with others, if desired. 



' With regard to modifications of my previous work, I should be 

 inclined to discard the longitudinal and transverse arcs of the head, in 

 preference for the " minimal frontal diameter " of Broca. The external 

 occipital protuberance is in many subjects quite indefinite, and all tape 

 measurements vary considerably with the condition of the hair ; so that 

 although I have taken these two tape measurements and the horizontal 

 circumference, I have not made use of them in estimating capacity. 



' I believe a few principal measurements such as the longitudinal, 

 transverse, and vertical diameters taken accurately will furnish far better 

 results than a large number of subsidiary measurements which enormously 

 complicate statistical work, and, at least for the purpose which I have had 

 in hand, are of very questionable value. 



' In investigating the influence of anj' particular characteristic, such 

 as mathematical or musical talent, a great deal of valuable time may be 

 saved by simply comparing the extremes with one another, and if the 

 average measurements have been found for the particular country or 

 district, I should regard it as unnecessary to measure the average 

 individuals in whom the characteristic in question was neither conspicu- 

 ously developed or altogether absent. 



' I think much may be gained by careful selection, and grouping of 

 the individuals measured, and that, on the other hand, there is a danger 

 in dealing with a large number of figures obtained from diflferent sources 

 of losing in the general mass, class, or type characteristics which may be 

 of great significance.' 



Gray, J., B.Sc, and Tochkr, J. P., P.I.C., have been engaged on 

 anthropometric work since 189.5. The measurements usually taken were 

 length and breadth of head, and height standing and sitting. The class 

 of people measured were farm labourers and artisans in rural districts iu 



