( 4 ) 



dots (M and M 1 ) in a direct line. This will give the "bimalar 

 breadth," and should be entered in column 24. 



Then measure with the same instrument the distance from either 

 of the orbital dots (M or M 1 ) to the nasal dot (N), and double 

 the result. This will give the " naso-malar breadth," and should be 

 entered in column 25. It will perhaps be the safest plan to test 

 the distance from each dot (M and M 1 ) to the dot N before 

 doubling. 



Figure 5 shows the points for the index on the living 

 subject. 



The index is 



Naso-malar breadth x 100 

 Bimalar breadth 



to be entered in column 26. 



Vertical proportions of the head. — These are the only measure- 

 ments which present any serious difficulty, and after several 

 experiments I hope that a mode of overcoming this difficulty has 

 been discovered. The measurements are taken with the graduated 

 T — square {Equerre cephalometrique), and the smaller steel sliding 

 scale or the wooden triangular slide. Their accuracy depends 

 upon the subject's head being exactly upright, and being kept in 

 that position while the measurements are going on. There appear 

 to be two recognised methods for placing the subject's head in an 

 upright position. The first, devised by Dr. Barclay in 1803, consists 

 in making the subject hold with his teeth a flat plate of metal 

 mechanically levelled. Topinard discusses this plan and condemns 

 it as too complicated. For use in this country it is open to the 

 further objection that unless all the subjects operated on at the same 

 time belong to the same caste and sub-caste, the plate of metal 

 would have to be continually washed in deference to caste prejudices. 

 It also appears to me that if a man has got a plate of metal between 

 his teeth, the height from the top of his head to the bottom of 

 his chin cannot be correctly measured, and will in practice vary 

 considerably. The second method, which Topinard prefers, "consists 

 in" directing the subject to look steadily at the horizon, and in 

 correcting the position of his head if by accident or through 

 nervousness he does not look straight before him in the natural 

 manner." In this manner, Topinard adds, the head will be adjusted 

 in accordance with the plane of vision, and will necessarily assume 

 a correct position for the purpose of measurement. 



We must, I think, take it on Topinard's authority that the 

 head can be correctly placed by following these instructions. We 

 are met, however, by the further difficulty that after the correct 

 position has been ascertained the subject cannot keep his head 

 absolutely still, and that every movement, however slight, materially 

 affects the measurements. Having got the correct position, we want 



