Two male plumage morphs, ‘white’ and ‘rufous’, coexist in the Madagascar paradise flycatcher Terpsiphone mutata. There has been enduring debate about whether this dimorphism in male plumage represents seasonal dichromatism, delayed plumage maturation, or genetic differences. We present data from a nine‐year study monitoring plumage changes in 119 individually colour‐banded males (430 male‐years). Our data show that paradise flycatchers are not seasonally dichromatic, and that although males show delayed plumage maturation, the rufous morph is not simply a precursor to white plumage, as previously thought. Individual males followed irreversible developmental pathways to ‘rufous’ or ‘white’, and could be reliably assigned to these distinct phenotypes by their second year. ‘White’ males adopted definitive plumage by the age of three years, whereas ‘rufous’ males did so between the ages of three and six years. The fixed nature of these morphs suggests that a genetic basis for the dichromatism is likely. However, variation in the timing of definitive plumage acquisition in rufous males could involve condition dependence and be environmentally influenced.