During the twenty years which have elapsed since the first interpretations were set forth there have been modifications of each hypothesis, in the direction of mutual reconciliation. This has come about through extended researches by many inquirers, furnishing additional data. III. CORRELATION OF ABILITIES IN VARIOUS GROUPS Some of the significant studies of correlation made since Spearman and Thorndike proposed their conflicting interpretations, have been cited in the appended list of references. Two samples of the results of these studies are herewith presented. The first is from Simpson’s study of mental tests given to two groups of adults, chosen respectively from the opposite extremes of competency, as shown by social-economic success. One group was composed of successful professional educators. The other was composed of unskilled laborers and unemployed men. The table on page 18 shows how the traits measured cohere among these individuals. The coefficients are positive, in the majority of cases highly so. The second sample is from Weglein’s study of standing in school subjects, among high school pupils. Bearing in mind that, if no mutual relationship exists among the abilities considered, coefficients of correlation will approach zero, it is clear that there is decided positive, but not perfect, correspondence. The wider the range of competence tested, the greater the correspondence found. High school pupils (from among whom those having very little ability for the subject matter taught have already been eliminated) show smaller coefficients than do the persons measured by Simpson. If all adolescents in existence were obliged to study the subjects listed by Weglein, and if the resulting grades were then correlated, the coefficients would be notably higher than those recorded. TABLE FROM SIMPSON PEARSON COEFFICIENTS OF CORRELATION (CORRECTED FOR ATTENUATION) Correlations of abilities in two selected groups, and in the two treated as one group. In the case of each test the heavy-face figure given first is for the Good and Poor together, divergences being measured from the median of the 37 individuals. The second figure is for the Good group, divergences being measured from its median. The third figure is for the Poor group.