H.— ANTHROPOLOGY. 145 



The last point to which I wish to draw your attention is head .shape. 

 As you know, the anthropologist usually thinks of skulls in terms of their 

 length and breadth, and certainly he has gained a great deal of useful 

 information in the past from this cranial index, his so-called sheet-anchor. 

 Lately, however, he has felt that something more is needed, and specialists 

 in craniology have piled up such a mass of arcs, indices, coefficients, and 

 angles that few are able to criticise their fellow- workers, because few are 

 able to understand what their fellow-workers are doing. 



Unfortunately we cannot claim that the results have kept pace with 

 the growing complexity of the methods, and if we are ever to interest the 

 non-specialist, and to induce him to add to our knowledge by using the 

 enormous mass of material which comes in his way, we must devise some 

 system simple enough to be grasped by any educated person and yet 

 more valuable than the mere record of the length and breadth of the head. 



The reason why the cranial or the cephalic index is not enough is that 

 it treats the head as if it were a structure of two. instead of three, dimen- 

 sions. To use a homely simile, it is like giving the length and breadth of 

 a box and then expecting the hearer to grasp what that box is like. 

 We have hundreds of thousands of records of the length and breadth of 

 heads, but very few of their height. Even when the height is recorded we 

 set it down and try to visualise it by comparing it with the length — just 

 as we compare the breadth of the skull with its length. 



In other words, we use the length as though it were a constant with 

 which we could compare the variable breadth and height, though we 

 know that the length may be just as variable as either of the other 

 diameters. 



I want to submit to you that, if we use all three dimensions — length, 

 breadth, and height — together, a standard will be gained which roughly 

 will represent the size of the skull, and with this each dimension may 

 be compared, and a proportional index for each established. The most 

 accurate method, no doubt, is to take the product of the three dimensions 

 and then to extract the cube root and multiply it by three. The result 

 of this is a standard by which the length, breadth, and height of the skull 

 may be divided, and in this way proportional indices obtained which will 

 bear a definite relation to the size of the skull. Unfortunately the process, 

 though soon learned, is tiresome and needs a logarithm table, which is 

 not always to hand. 



A much simpler, and for all practical purposes an equally valuable 

 method of gaining proportional indices is to add together the length, 

 breadth, and height of the skull, and then to divide each dimension by the 

 sum thus obtained. This gives a series of indices which are, on an average, 

 •006 lower than those which the cube-root system supplies ; but in no 

 case does this alter the relative position of any of the series of British 

 skulls tabulated in the accompanying list. My colleague, Dr. Mulligan, 

 has been kind enough to prepare a second table which shows how nearly 

 the results of the two methods correspond. 



Before considering these tables, which I think give a more rational 



and coherent view of British craniology than could be obtained from the 



old cranial index, though I am glad enough to use that too, let us take 



as an example the skull of a Saxon, fately dug up at Bidford-on-Avon. 



1(127 1. 



