The affinities ofPrototaxiteshave been debated ever since its fossils, some attaining tree-trunk proportions, were discovered in Canadian Lower Devonian rocks in 1859. Putative assignations include conifers, red and brown algae, liverworts and fungi (some lichenised). Detailed anatomical investigation led to the reconstruction of the type species,P. logani, as a giant sporophore (basidioma) of an agaricomycete (= holobasidiomycete), but evidence for its reproduction remained elusive. Tissues associated withP. taitiin the Rhynie chert plus charcoalified fragments from southern Britain are investigated here to describe the reproductive characters and hence affinities ofPrototaxites. Thin sections and peels (Pragian Rhynie chert, Aberdeenshire) were examined using light and confocal microscopy; Přídolí and Lochkovian charcoalified samples (Welsh Borderland) were liberated from the rock and examined with scanning electron microscopy.Prototaxites taitipossessed a superficial hymenium comprising an epihymenial layer, delicate septate paraphyses, inoperculate polysporic asci lacking croziers and a subhymenial layer composed predominantly of thin-walled hyphae and occasional larger hyphae.Prototaxites taiticombines features of extant Taphrinomycotina (Neolectomycetes lacking croziers) and Pezizomycotina (epihymenial layer secreted by paraphyses) but is not an ancestor of the latter. Brief consideration is given to its nutrition and potential position in the phylogeny of the Ascomycota.This article is part of a discussion meeting issue ‘The Rhynie cherts: our earliest terrestrial ecosystem revisited’.