The relation of self-regulation measured prior to school entry to developing math and reading ability in prekindergarten through the second grade was examined in a prospective longitudinal sample of 1,292 children and families in predominantly rural and low-income communities in 2 regions of high poverty in the United States. Direct assessments of executive function, effortful control, and stress response physiology (indexed by resting levels of cortisol and alpha amylase obtained from saliva) were measured at child age 48 months and parents and teachers reported on children's effortful control using temperament rating scales at child age approximately 60 months. Math and reading ability, as measured by the Woodcock-Johnson III applied problems and letter-word subtests, respectively, were measured at prekindergarten through the second grade. Effects for self-regulation measures were seen primarily for initial level and to some extent growth in both mathematics and reading, even when controlling for family demographic characteristics that represent relevant selection factors into higher levels of both self regulation and academic achievement. These effects persisted for mathematics but not for reading with the inclusion of child cognitive abilities, vocabulary, and speed of processing measured in prekindergarten, concurrent with the first time point for the academic measures. Results are interpreted as indicating a role for self-regulation in learning ability generally, likely through support for attention and reasoning abilities that are most specific to the assessment of mathematics in this analysis. Implications for instruction and for assessment and the best ways to support the development of early math and reading ability for children at risk for school failure are discussed.