The mean of the actual gains in the second period exceeds that of the first. Again the second group of results is more variable, decreasing the reliability. There seems, however, to be a definite increase in the net gain of the test group during a second six months' period. Some individual cases may be cited. The greatest gain after six months is 8 units in the test case, matched by an equal gain of 8 units in the control group. After twelve months, the test group shows one gain of 19 units, the highest gain in the control group being 13. Six cases in the test group, and 13 in the control had lost at the end of six months, but after twelve months, all but 2 of the test cases showed a gain, and all but 5 of the controls. In 10 test cases out of the total 13, more than half of the gain occurred during the second six months. In the control group, six of the cases made more than half of their gain during the second six months, and the second interval gains of the other 7 cases exceeded the 50 per cent mark by so little that they may be accounted for by chance. These results seem to indicate a slight but actual increase in the net gain of the test group during the second six months of the experiment, and an accompanying growth in the variability of these gains. It will be remembered that the results described in the previous chapter show no gain in strength of grip as a result of operation. Comparison of the 13 cases tested after the second interval, with the 16 cases at the end of the first, gives results as follows: TABLE XVIII Gain in Grip, Second Retest, 13 Pairs N[16] Test Group (A) Control Group (B) Test 1 Test 2 Gain Test 1 Test 2 Gain A-B 7 7 6 -1 9 14 5 -6 19 22 20 -2 15 18 3 -5 2 9 10 1 10 15 5 -4 27 9 6 -3 8 9 1 -4 13 26.5 28.5 2 22 27 5 -3 10 12.5 12 -.5 10 12 2 -2.5 16 19 18.5 -.5 10 10 0 - .5 3 12.5 13 .5 14 14.5 .5 0 14 16 17 1 22 22.5 .5 .5 21 16.5 15 -1.5 19 17 -2 .5 1 13 18 5 11 15 4 1 23 11.5 15 3.5 14.5 16 1.5 2 11 11 11 0 11.5 7.5 -4 4