reminiscence, while in Age-group 2 reminiscence occurred only for the girls. The degree of reminiscence as well as its frequency of occurrence are thus to some extent a function of sex. SUMMARY AND CONCLUSIONS Partial learning is obviously one of the essential conditions of the phenomenon of reminiscence. The phenomenon was exhibited only by the two younger age-groups, and for the connected type of material. The greater degree of reminiscence and reminiscence for the longer intervals of time were manifested by the youngest age-group. Within these age-groups, the subjects who learned the least amount of material exhibited the greatest degree of reminiscence. Moreover, these age-groups showed the greatest amount of reminiscence when the material was scored for correct verbal reproduction. Reminiscence occurred more frequently with girls than with boys, and girls displayed a greater degree of reminiscence when it was present in both sexes. So far we have merely stated the conditions under which reminiscence occurred without making any attempt to explain it. Any suggested explanation must take into account the fact that reminiscence may be a special case of the more general law that the percentage of retention varies with the material, the method of scoring, the amount learned, and the age and the sex of the subjects. It may be suggested that the inverse correlation between the percentage of retention and the age of the subjects is due to the fact that the younger children learned a smaller proportion of the material presented, and, hence, the opportunity for reminiscence was greater. Reminiscence, therefore, may be a function of the degree of learning rather than of some other factor connected with mental growth. The percentages of retention, however, will necessarily exhibit a spurious inverse correlation with the learning scores inasmuch as these percentage values were derived by dividing the amounts retained by the learning scores. The degree of retention may be a function of the extent to which the material was spontaneously revived during the interval for which retention was tested. It is possible that reminiscence for the poetic material was manifested by the younger children because its subject-matter and rhythmical construction naturally appealed to children of this age. Possibly these same children would not have exhibited reminiscence for other types of connected material. The greater degree of reminiscence manifested by the girls may, likewise, be due in part to their greater interest in this type of material, or to the absence of various distracting conditions that would naturally inhibit the spontaneous arousal of the material during the interval. The fact that the children exhibited the greater degree of reminiscence when the material was scored for correct verbal reproduction may be due in part to their greater training in verbatim methods of memorizing. The division of the age-groups on the basis of the amount learned does not necessarily represent differences in learning ability. Some of the subjects in the lowest third of the age-groups made poor scores on the first reproduction because of unfavorable chance conditions that obtained at the time of learning. Likewise, chance conditions necessarily favored certain of the individuals of the highest third of these agegroups. According to chance, many of the subjects in the lowest third would make relatively better scores in the second reproduction, while the individuals of the highest third would be likely to make relatively poorer records in the second test. This factor of chance may thus account in part for the greater reminiscence exhibited by the poorer learners. These explanatory concepts are offered merely as suggestions. In all probability the factors accounting for reminiscence are many and various. Our experimental procedure was designed, primarily, to ascertain some of the factual conditions of the phenomenon rather than to explain it. REFERENCES 1. BALLARD, P. B. Obliviscence and reminiscence, Brit. J. Psychol., Monog. Suppl., 1913, I. 2. BEAN, C. H. The curve of forgetting, Arch. of Psychol., 1912, 20, No. 3. 3. BROWN, W. To what extent is memory measured by a single recall, J. Exper. Psychol., 1923, 6, 377-82. 4. EBBINGHAUS, H. Über das Gedächtniss, Leipzig, 1885. (Eng. Tr. by Ruger and Bussenius, Teachers' College, Columbia University, 1913.) 5. Finkenbinder, E. O. The curve of forgetting, Amer. J. Psychol., 1913, 24, 8–32. 6. HUGUENIN, C. Reviviscence paradoxale, Arch. d. Psychol., 1914, 14, 379-83. 7. LUH, C. W. The conditions of retention, Psychol. Rev., Monog. Suppl., 31, No. 3. 8. NICOLAI, F. Experimentelle Untersuchungen über das Haften von Gesichtseindrücken und dessen zeitlichen Verlauf, Arch. f. d. ges. Psychol., 1921, 42, 132-49. 9. RadossawlJEWITSCH, P. R. Das Behalten und Vergessen bei Kindern und Erwachsenen, Leipzig, 1907. 10. STRONG, E. K. The effect of time-interval upon recognition memory, Psychol. Rev., 1913, 20, 339-72. RACIAL DIFFERENCES IN THE INTELLIGENCE OF SCHOOL CHILDREN BY FLORENCE L. GOOD ENOUGH Institute of Child Welfare, University of Minnesota The widespread interest in tests of intelligence which has sprung up in this country during recent years has led to a number of studies of the comparative performance of various immigrant groups and their immediate descendants. No attempt will here be made to review these studies in detail; but the general nature of the results of them with respect to children of school age has been summarized in the accompanying list. For the greater part these studies are based upon American-born children; although it is probable that in some instances foreign-born children have also been included. The figures presented are representative of the findings of a number of other investigators whose results have not been included. They also correspond rather closely to the scores made by the different nativity groups in the Army Intelligence Tests as reported by Brigham (3). It should be emphasized that figures such as these have reference only to the representatives of the various nationality groups who have immigrated to this country, and that selective immigration may be in large measure responsible for the results which have been obtained. That these immigrant groups and their immediate descendants markedly differ in their performances in the ordinary type of intelligence-test is a fact which appears to have been reasonably well established. Two theories have been offered to account for these differences. The first ascribes the inferior showing made by the South Europeans and the negroes to such post-natal factors as inferior environment, poor physical condition and linguistic handicaps. The second point of view, while it |