In order to determine whether the slight inequalities in interval length have any considerable effect on the results, we have calculated the relation between the length of interval and amount of improvement. The coefficient of correlation by the method of rank differences is equal to .03. The small number of cases renders the unreliability of correlation very great, but we can at least say that there is no consistent relationship between improvement and time interval, within the narrow limits here set. We are probably justified in taking twelve months as the interval, since such was the case in eight out of the fifteen test cases, while the greatest variation above this made was four months, and below it, one month. The gains in height after twelve months are shown in Table XVI. The average gain of test group in excess of control, is only .08 inches, and the median .5 inches. Variability is about the same as at the end of six months, P. E. +- .28. The average is only .29 P. E., but the median is a little larger, 1.79 P. E. If these measures are compared with the results after the first period, we have: 6 months 12 months Average of gains of test group in excess of control .16 .08 Median of gains of test group in excess of control .2 .5 P. E. of difference .16 .28 Average in terms of P. E. 1.00 .29 Median in terms of P. E. 1.25 1.79 There seems to be little gain in height after the first period. Test cases 1 and 11 each show a gain of 3.7 inches after fifteen and twelve months respectively, compared with gains after six months of 1.6, and 1.9 inches. But case 2 in the control group, makes still greater comparative gain, +1.8 inches after six months and 4.6 inches after seventeen months. In this case there are almost six additional months for the child to grow, which may account for the larger gain. Control case 1, however, may be compared with his partner, mentioned above, since the interval between tests was the same for both. This boy grew 1.4 inches in six months, and 3.3 inches after 15 months. This is practically equal growth with test case 1. Control case 11 also shows relatively great growth during 12 months, +3.2 inches, whereas the growth in six months was only 1.3 inches. Out of the test group, 7 cases gained more in the first period of six months, than in the second, while only 6 gained more in the second than in the first. Of the control group, 7 cases made more than half of their total gain during the second six months of the total twelve months' period. Since this is true, it seems likely that whatever increase in growth we find during the second half of the twelve months' interval, may be explained by incidental causes, and that so far as actual gain in height is considered, there is no further effect from the operations, after six months.