Of the total number of delinquent boys, both working and non-working, under twelve years of age, 22.4 per cent were workers, while of those twelve to thirteen years old, 42.4 per cent were workers, and of those fourteen to sixteen years old, 80.8 per cent were workers. As comparatively few children under twelve years are at work, the fact that more than one-fifth of the delinquent boys in this age group are working children "becomes exceedingly significant." Of all children twelve to thirteen years of age, the great majority are not employed because of the fourteen-year age limit prevailing in all the states studied except Maryland; hence the larger proportion of working offenders cannot be explained by the influences of age. The increase of working delinquents above fourteen years is to be expected, because so many children go to work on reaching that age. Remembering that the proportionate excess of workers varies from two to nine times the ratio of non-workers, it is evident that this excess cannot be explained by a corresponding excess of orphanage, foreign parentage, bad home conditions or unfavorable age. As the report says, "It seems rather difficult to escape the conclusion that being at work has something to do with their going wrong."[116] The strongest argument against street work by children is to be found in the following table[117] of occupations pursued by the largest number of delinquents and giving the percentage of total delinquents engaged in each. As the report says, the following classification shows that the largest number of delinquent boys were found in those occupations in which the nature of the employment does not permit of supervision--namely, newspaper selling, errand running, delivery service and messenger service. Boys engaged in these occupations, together with bootblacks and peddlers, all work under conditions "which bring them into continual temptations to dishonesty and to other offenses."[118]