A rereading of the American scientific literature on sex determination from 1902 to 1926 leads to a different understanding of the construction of the Mendelian-chromosome theory after 1910. There was significant intellectual continuity, which has not been properly appreciated, underlying this scientific "revolution." After reexamining the relationship between the ideas of key scientists, in particular Edmund B. Wilson and Thomas Hunt Morgan, I argue that, contrary to the historical literature, Wilson and Morgan did not adopt opposing views on Mendelism and sex determination. Rather, each preferred a non-Mendelian explanation of the determination of sex. Around 1910, both integrated the Mendelian and non-Mendelian theories to create a synthetic theory. One problem was the need to avoid an overly deterministic view of sex while also accepting the validity of Mendelism. Morgan's discovery of mutations on the X chromosome takes on different significance when set in the context of the debate about sex determination, and Calvin Bridges's work on sex determination is better seen as a development of Morgan's ideas, rather than a departure from them. Conclusions point to the role of synthesis within fields as a way to advance scientific theories and reflect on the relationship between synthesis and explanatory "pluralism" in biology.