In the course of a longitudinal study that addressed the questions of how children learn to comprehend what they read and how they learn science concepts, many other research questions emerged. The original longitudinal study investigated how children learn to comprehend what they read. Three school districts with fairly stable student populations, reputations for overall high student achievement, and a willingness to commit fully to the 6-year project, participated in the study. Analysis was completed on two cohorts of children (approximately 625 students) assigned to about 40 teachers. Fourteen different research questions emerged in the kindergarten through second-grade work. Among the surprising answers to these questions are the following: (1) the length of the school day did not predict kindergarten children's school achievement; (2) individual teachers' behaviors were quite stable both from morning to afternoon and from year to year; (3) whole-class reading instruction produced the greatest gains in student achievement in reading; (4) teachers' instructional practices mediated children's performance on criterion-referenced measures; and (5) elementary-grade science textbooks were quite considerate to students. (Five figures and seven tables of data are included; 62 references are attached.) (Author/RS)