A Sketch of his Work as a Craniologist. 27 



length to height by indices. This plan is adopted in distinguishing 

 between round and long, or flat and high heads. Such a method 

 is simple, and in practice has been found useful in classifying 

 skulls and in distinguishing races. It is obvious, however, that 

 two skulls differing considerably from one another both in capacity 

 and shape might have the same length-breadth and length-height 

 indices. If we examine the median longitudinal arc of the 

 vaulted portion of the skull we find that it passes from the root 

 of the nose upwards and backwards to the vertex forming two 

 curves, with their convexities directed forwards and upwards, it 

 then turns downwards and backwards to the most posterior part 

 of the skull, and finally forwards and downwards to end at the 

 posterior margin of the large hole at the base of the skull through 

 which the brain becomes continuous with the spinal cord. Now, 

 the form of these curves from the nose to the vertex, and from the 

 vertex to the back of the head, may differ greatly in two skulls 

 which have the same length and height. Thus, in one the fore- 

 head may be high and protruding, the roof of the skull be directed 

 nearly horizontally for a considerable distance and then descend 

 abruptly to the occiput, while in the other the forehead may be 

 low and retreating, and the longitudinal arc only gain the same 

 height as the other skull for a very short distance before it begins 

 to descend again. The problem was, and indeed still is, How can 

 we best give numerical expression to these differences ? The plan 

 often adopted of measuring the length of the three portions of this 

 arc, viz., frontal, parietal, and occipital, gives the respective share 

 the bones so named take in the formation of the arc, and hence 

 may be supposed to show the relative development of the anterior, 

 middle and posterior parts of the vault of the skull. In some skulls, 

 however, owing to irregularity in, or disappearance of, the lines of 

 union between these bones, their respective lengths cannot be 

 definitely ascertained, and in any case such measurements do not 

 show the contour of the arch. For this purpose I believe that the 

 best method yet devised is that of Grattan's. By means of his 

 craniometer the skull is firmly fixed in position by passing two pegs 

 into the external openings of the ears and pressing another screw 



