Summary. 13. Agrocybe gasteromycetoides from Wyoming, North America, is described as new; relationships between some agaricoid and gasteromycetoid fungi are discussed. 14. The new names Agrocybe dennisii, A. hesleri, and Conocybe acutoconica are proposed. Twenty-two new combinations are proposed in the genera Agrocybe, Bolbitius and Conocybe. 13. A XEROMORPHIC MEMBER OF THE FAMILY Whilst examining material of the Bolbitiaceae in various United States herbaria, particularly that in the University of Michigan, a great deal of information on the family was amassed, and amongst all the new species which will ultimately have to be described the fungus considered herein deserves rather special treatment. It is of particular interest when comparison is made with a recent account of hypogeous cortinarii published by Thiers & Smith (1969). Although the fungi they described were considered true members of the genus Cortinarius, the characters exhibited have assisted our interpretation of the links between Cortinarius (Agaricales) and Thaxterogaster (Gasteromycetales). In the same way the present fungus contributes to an understanding of the relationship between the genera Agrocybe and the rather heterogeneous Galeropsis. The fungi described by Thiers & Smith were considered to possess in their genotypic composition an arrestment factor for basidiocarp expansion, probably best considered as causing a slowing down of the expansion of certain tissues of the basidiocarp, resulting in the stipe elongating only slightly or not at all, and the veil becoming thick and tough. In the fungus described below, the cessation of growth is apparently limited to the margin of the pileus and although a well developed stipe is formed, the pileus, during its paravelangiocarpic development, adheres tightly to it. Differentiation continues within the closed pileus to form the pileal trama, hymenophoral trama etc., and results in a series of gills enclosed in a 'peridium'. The gills or tramal plates ultimately separate one from the other and from the stipe by expansion of the lamellar gap; some of the gills also become bridged by veins or are finally costate. There is little doubt that this fungus would rightly fit into the old concept of the genus Secotium. However, the collector made a very significant observation: once the basidiocarps had been gathered some accidentally broke, and, surprisingly, a spore-print typical of an agaric was formed. Thus the hiatus between those characters associated with the agaricoid and gasteromycetoid basidiocarps becomes reduced. Indeed just as Thiers & Smith considered their fungi to be true agarics, it is proposed to treat the fungus present in the genus Agrocybe by virtue of its hymeniform pileipellis, colour of the spore-print, structure of the hymenophoral trama and morphology of the basidiospores. Indeed in gross morphology the fungus described below resembles members of the Agrocybe semiorbicularis-pediades