The purpose of this research was to develop methods to study the effects of context on children's comprehension and production of temporal terms such as "before,""after,""next,""then," and "but first." A longitudinal study, using naturalistic and traditional laboratory methods, and three cross-sectional studies were conducted with children between the ages of two-and-a-half and four. Most of the studies examined aspects of the Semantic Feature Hypothesis (the only stage theory of lexical acquisition that has been extensively tested with children's understanding of temporal terms). The main findings from both the longitudinal and cross-sectional studies were: (1) children understand temporal order through a sentence's inherent meaning; (2) although children will perform better on act-out sentences conjoined by "before," the Semantic Feature Hypothesis is inadequate in explaining the results; (3) some children will use a main-clause-first strategy in acting out temporal sentences, whereas other children will use a first-clause- first strategy; and (4) children use many routes to lexical mastery. The longitudinal analyses support the methodological claim that repeated measures on the same children yield a much richer picture of lexical and conceptual development. Appended are sections on the following topics: (1) spontaneous usage of searched temporal terms; (2) copies of test forms as administered; (3) unfamiliar temporal connectives: a test of the first-clause-first strategy; (4) a comparison of young children's comprehension and memory for statements of temporal relations; and (5) lexical development: a mini-longitudinal approach. (Author/NCR)