“…Testing relatively large unselected groups of school children Bradford, 11 Chapman and Wiggins, 11 Lentz, 9 Cattell, 7 Fraser Roberts, Norman, and Griffiths, 7 Moshinsky, 10 and Burt, 28 have obtained correlations from -.19 to -.33 between intelligence and number of children in the family. In certain instances where special groups were tested, as the gifted by Terman 18 and those from a slum clearance area by Dawson,4 the obtained correlations were within this range. However lower negative correlations were generally obtained with the highly selected groups, such as the coal miners' children (Sutherland 12 ), special groups of private, preparatory, and secondary fee-paying children (Fraser Roberts et al, 1 Moshinsky 10 ), and college students (Himmelweit 8 ).…”