14 MAL-NUTRITION AND THE TEACHER'S 



It is almost entirely on this material that Mr. Yule has 

 based his elaborate memoir on the Association of Attributes 

 in Statistics. 



Now Mr. Yule has condemned the L.C.C. investigation on 

 the ground that it is ' an excellent example of the danger 

 of purely verbal classification', while he accepts Dr. Warner's 

 investigation as one in which ' continuity of standard was 

 ensured '. 



But among the multitude of categories dealt with by 

 Dr. Warner there are some which are capable of quantitative 

 measurement, and particular attention must be drawn to his 

 treatment of these characters. In discussing size and shape 

 of head, for instance. Dr. Warner remarks : — ' Of all defects 

 in development, abnormalities of the cranium appear to be 

 the most important, being the most numerous and having 

 the highest pathological co-relations. The size and probable 

 volume of the brain is a point of first class importance, and 

 the size of the cranium is in children a fair indication of the 

 size of the brain.' [Report^ p. 20.) 



Let us therefore examine in detail Dr. Warner's treat- 

 ment of cranial abnormalities, and let us confine our atten- 

 tion to large and small crania which could have been directly 

 measured. On p. 72 of his Report we find the following 

 description of these categories : 



' A. I. Cranium defective includes any defect in size, 

 form, proportions, or ossification of the cranium. A given 

 case may come under more than one of the classes below. 

 As to a standard of normal size : In a well-developed child 

 of good potentiality, the head circumference at 9th month 

 is 17 J inches, at 12 months 19 inches, at 7 years 20 to 

 21 inches. This is rather a high standard of size. 



* A. 2. Cranium large. A head of 22 inches circum- 

 ference or over may be considered large in a schoolchild ; 



