A 1984-85 study of Palau elementary students, who were pretested, instructed with a textbook and additional practice booklets, and posttested, revealed gains for every skill at every grade level. A 1986-87 study, using a series of 26 booklets representing a fairly complete math curriculum, showed that students using the experimental booklets gained more during the year than students using regular textbooks. A longitudinal study conducted between 1985 and 1992 showed decreasing scores over the period. Overall competence in mathematics at the sixth through eighth grade level was most highly correlated to recall of division facts. Students scored higher on the computational component than on the application component of the test. Evaluation of 1993 data indicated that girls in grades 2 to 8 outperformed boys on each of the subskills. Scores on the Stanford Achievement Test administered to some students in grades 4, 6, 8, 10, and 12 showed that girls outperformed boys at each grade level except 12. This performance is attributed to the fact that girls in Palau must go home and remain there after school while boys are free to roam about; hence, girls are more likely to do their school work. As boys get closer to graduation, they begin to take their studies more seriously and improve their performance. The outcomes of the studies strongly suggest that social factors are just as, or more important than genetic factors with regard to math performance. Sixteen data tables are attached. (JDD)