Let us see what the rough preliminary estimates on the basis of school retardation would indicate for the Minneapolis delinquents. We may disregard the upper limit of 14 years since compulsory attendance in Minnesota for backward pupils continues until age 16. For the limits of five and four years of retardation in school below the standard of 7 years in the first grade we would have estimates of 2.6% to 6% of deficiency among the ordinary cases of delinquent boys and 14.7% to 23.1% among the ordinary delinquent girls. Among the recidivist group of boy offenders 3% to 11% would be below these borderlines. Among the Glen Lake School group 12% are four years or more and 4% five years or more retarded. This last is to be compared with our judgment on the basis of individual examinations with the Binet scale in which we concluded that 2% were presumably deficient and 5% doubtful as to deficiency. The estimates on the basis of school retardation are somewhat too large. This would certainly be true for older delinquents. In as much as the laws for compulsory school attendance usually do not enforce attendance after 14 years of age, it would probably be better generally to treat all over 14 years of age as if they were of this age at the time of leaving school. This limiting age of 14 checks more closely with the mental examination records reported by Williams (_149_) and Ordahl (_41_) for groups of delinquents in the California state schools. With her unselected group of 88 women at the Bedford reformatory, Weidensall found that 39% had not completed the fifth B grade (_60_, p. 23). This is not far from the estimate of presumable deficiency among such inmates on our borderline with the Binet scale. Considering the actual years of school retardation relative to years of attendance, so far as she was able to discover, and adding the 8 who never attended school, we have 20% five or more years retarded in school and 28% four or more years retarded (_60_, p. 251). She says further regarding the bi-modal distribution of ability which she found among her group: “The division which alone served to separate the better from the poorer subjects was that of the grade completed upon leaving school. Those who had accomplished the completion of at least 5B grade formed a curve which paralleled very closely that of the Cincinnati girl of fifteen, while those who had not succeeded in passing 5B comprised the majority of those who collected at the poorer mode of the Bedford 88 curves. Throughout, the grade completed has proved to be more often a measure of our subjects' ability to progress in school, less often a measure of their opportunity to attend school.”