In the seven decades since Leopold's groundbreaking 1939 study, there has been no longitudinal study covering more than two years of a Japanese bilingual subject's development. Despite the lack of longitudinal research, however, we have been broadly informed by the veritable outpouring of research on a short-term basis since the late twentieth century. This longitudinal study tracks a Japanese-English bilingual from childhood (4;09) into late adolescence (19;01). Data collected orally were analyzed in terms of fluency, accuracy, complexity, and vocabulary. By continuing through the late acquisition phase, we were also able to observe the development of the subject's growing ability to employ narrative structure, examining her nondominant language, English, from both the linguistic and narrative perspectives, to determine whether the nondominant language develops in a similar manner to monolinguals and/or to bilinguals who are acquiring two linguistically close languages. The results indicate that, even when exposure to the nondominant language is limited and the two languages are typologically distant from each other, the development of the nondominant language is similar or identical to that of a monolingual in core linguistic areas, and that a child acquiring two languages simultaneously is able to develop two separate languages--partially supporting the Separate Development Hypothesis. This support is not robust, however, because our subject showed some idiosyncratic errors and a unique narrative style, possibly due to L1 influence. The results also imply that extensive and intensive exposure to the nondominant language at or before a certain age may be essential to reach a level similar to monolinguals. (Contains 1 figure and 12 tables.)