810 



MEMOIRS NATIONAL ACADEMY OF SCIENCES. 



[Vol. XV, 



white men and 2,877 negroes who were free from the disease. The comparison included both 

 alpha and beta scores and involved the following numbers : 



The percentage distributions of weighted scores for the white alpha groups, 501 infected 

 and 4,792 noninfected, were as follows: 



The difference is too large to be explained as the chance result of sampling. The ratio 

 of the difference of the means to the standard deviation of that difference is 7.96, which means 

 that the chances are infinitely small (less than 1 in 10,000) that as great a difference could 

 occur by chance. 



The percentage distributions of scores for the 131 infected and 823 noninfected white men 

 who took beta were as follows : 



Chances that as great a difference could result from sampling are 48 in 1,000. 

 The percentage distributions of scores for the 74 infected and 1,817 noninfected negroes 

 who took alpha were as follows : 



The corresponding comparison for the 56 infected and 1,060 noninfected negroes who 

 took beta: 



In the case of the negroes, both in the alpha and beta group, the chances are very great 

 (826 in 1,000) that a difference as large as that found could have resulted from errors of sampling. 

 This is in marked contrast with the findings for white men, and the results may mean that 

 hookworm disease has a more serious effect on whites than on negroes. 



There is available also a report from Camp Lee on the scores in examination a of 97 white 

 men infected with hookworm disease. Of 121 who appeared for psychological examination, 

 20 per cent were found to be illiterate, as against 18 per cent illiterate for the regiment as a 



