OF ATTRIBUTES IN STATISTICS. 297 



chance deviation from the positive would be, roughly, twenty-one to four, or say five 

 to one only not overwhelming odds by any means. 



As in the general case of complete crossing versus self-fertilisation, the differences 

 between species are, however, almost certainly significant. It may be true that 

 "crossing of different flowers on the same plant is always, on the average, better than 

 pure self-fertilisation," but the closeness with which the law holds good will vary in 

 different species. 



VI. ILLUSTRATIONS continued. 

 (B.) The Association of Defects in Children and Adults. 



51. The material on which the following investigation is based is drawn almost 

 wholly from the " Report on the Scientific Study of the Mental and Physical con- 

 ditions of Childhood,"* issued by a committee with representatives from the 

 British Medical Association, the British Association, the Charity Organization 

 Society, &c. Before 1892 the same work was in the hands of other committees of 

 those bodies, and in 1897, a " Childhood Society " was formed to carry it on. Two 

 series of investigations have been made under these committees, the first from 

 1888-91, the second from 1892-94. As I understand, the whole of the observations 

 in the first period, and the great majority of them in the second, have been made by 

 Dr. FRANCIS WARNER. 



For the complete description of the method of observation, &c., I must refer to the 

 report itself. A very large number of schools were visited (Board Schools, Poor 

 Law Schools, Voluntary Schools, &c., most, but not all, in London), and the children in 

 them examined individually for the presence or absence of certain defects, of which 

 the main classes were (using Dr. WARNER'S notation), t 



A. Defects in development of the body or its parts ; in size, form, or proportioning 



of parts. 



B. Abnormal nerve signs ; certain abnormal actions, movements, and balances. 



C. Low nutrition, as indicated by the child being thin, pale, or delicate. 



D. Mental dulness. The teachers' report as to mental ability was added to the 



record of each child registered, and those stated to be below the average in 

 ability for school work were registered as " Dull." 



These main classes of defects observed were the same in both investigations, in 

 each of which 50,000 children were observed. The returns given are, however, 

 somewhat more detailed for the later investigation, the material being sub-divided, 

 for instance, into school standards. I have used material from both. 



52. The whole of this mass of observations was made, as stated, on children not 



* Published by the Committee, Parkes Museum, Margaret Street. Price 2s. &d. 

 t Report ; pp. 12-13. A fuller description of the signs is given on pp. 13-16 and pp. 72 el seq. 

 VOL. CXCI V. A. 2 Q 



