The data presented in this thesis indicate that that correlation could have been pushed as high as the _r_ between the intelligence of the pairs of brothers. In other words, a child could be made to resemble his brother as nearly in spelling ability as he did in intelligence. All disparity could be reduced to that of general intelligence. Then intelligence alone is inherited as far as the data here presented have any bearing on the matter in hand. The influence of environment is in this case a matter of no consequence, since the subjects all had the same schooling, and home influence does not as a rule teach children to spell; but the data are not irrespective of the influence of intelligence. INDICATIONS OF THE GARDEN CITY DATA Table 3 presents intercorrelations between IQ and quotients in the various subjects. The correlations are in each instance irrespective of chronological age since all quantitative indices are expressed as quotients. We have seen that they go up from September, 1918, to June, 1920. Every possible means was used to push these correlations to their limit, to remove all removable factors. We have seen that the data show here, as in Tables 7 and 8, that there is little association between traits which is not a result of differences in intelligence. Table 3 shows the same 48 children throughout. The _r_’s are not corrected for attenuation. Though the _r_’s are high throughout and go higher under special treatment, the association can still be more accurately registered by some attention to relation of the means and the S.D.’s. Two traits to be identical must have _r_ = 1.00 S.D._{_x_} = S.D._{_y_} and M_{_x_} = M_{_y_}. We have seen that the _r_ increases, M-M decreases and S.D.-S.D. regardless of sign decreases. (Tables 9, 10 and 11.) But as the S.D.’s of the Subject Quotients (though they do approach S.D. of IQ) sometimes go below the S.D. of IQ, we must know why. It is because the low IQ’s do better per their intelligence than the high IQ’s. We have seen above that the correlation between IQ and average of the Vocabulary, Reading, and Completion Subject Ratios is -.61 in November, 1918, and -.49 in June, 1920.