The genus Psilocybe

The genus name Psilocybe is attributed to (Fr.) P. Kumm., published in 1871. Its Index Fungorum record, number 18418, gives the name status as legitimate and carries the nomenclatural comment "Nom. cons. et typ. cons., see Art. 14": both the name and its type are conserved. The record does not name the conserved type species. Niskanen and colleagues give it as Psilocybe semilanceata.

Conservation is a deliberate exception rather than an ordinary outcome of the rules. Article 14.1 of the Code provides lists of conserved names "In order to avoid disadvantageous nomenclatural changes entailed by the strict application of the rules, and especially of the principle of priority", and adds that conserved names "are legitimate even though initially they may have been illegitimate". Article 14.2 states the purpose plainly: "Conservation aims at retention of those names that best serve stability of nomenclature."

That matters for reading everything below. The name Psilocybe stayed with the psychoactive species because a decision was taken about which usage would disturb the literature least, not because priority happened to fall that way.

Psilocybe was once a much larger and more mixed assemblage. Ramírez-Cruz and colleagues report that Psilocybe sensu lato numbered between 277 and 300 species, citing Guzmán (2005) and Kirk et al. (2008), and that molecular analyses by Moncalvo et al. (2002) and Matheny et al. (2006) showed it to be polyphyletic: two separate genera under one name.

A nomenclatural proposal by Redhead et al. (2007) resolved which group would keep the name. Since its acceptance, Psilocybe has been applied to the clade of psychoactive species, and Deconica to those former species of Psilocybe and Melanotus that lack hallucinogenic compounds. Ramírez-Cruz and colleagues record the scheme as approved by Barrie (2011), McNeill et al. (2011) and Norvell (2011), and confirm Deconica as monophyletic within Strophariaceae sensu stricto.

The division is chemical and phylogenetic before it is morphological. Ramírez-Cruz and colleagues state that besides the bluing reaction, no single macroscopic or microscopic character can be used to separate the two genera, and caution that some species of Deconica darken to blackish, which should not be mistaken for a bluing reaction.

The older, wider use of the name is still physically present in collections. Bradshaw and colleagues sequenced 71 fungarium specimens filed as Psilocybe and found 52 to be Psilocybe sensu stricto, 20 of them type specimens. Of the remainder, 14 belonged to Deconica, one to Kuehneromyces, three formed a monophyletic group of uncertain placement within Strophariaceae, and one could not be assigned to an existing genus at all.

Four further specimens were consistently incongruent with the taxa they clustered with and were re-examined microscopically and redetermined, as Psilocybe hoogshagenii, Psilocybe cyanescens and, for two of the four, Psilocybe caerulipes. The authors describe misidentification as a common issue in fungaria and in Psilocybe specimens in particular.

This is the practical reason a reference of this kind records which source a name came from and when it was read, rather than treating a determination as settled because it appears on a label.

Ramírez-Cruz and colleagues describe the family placement of Psilocybe and Deconica as long debated, noting that Guzmán (1983) and Singer (1986) both treated Psilocybe sensu lato in the Strophariaceae. Their own analyses recovered Psilocybe sensu stricto as monophyletic in the Hymenogastraceae, with Deconica in Strophariaceae sensu stricto.

Niskanen and colleagues reshaped the family-level classification of the suborder Agaricineae using a phylogenomic approach, and describe Agrocybaceae as a new family whose currently included genera are Agrocybe and Psilocybe alone. They set out the alternatives they weighed, and chose three families over one because keeping the group together "would lead into an entity with significant variation, especially since the Agrocybe/Psilocybe clade stands out as being morphologically different".

This reference follows Agrocybaceae, and the species accounts say so. Where a taxonomic database still returns Hymenogastraceae for these names, the accounts record that as the state of that database on the date it was read, not as a competing published treatment.

Two principal lineages are recovered within Psilocybe, labelled clades I and II. Ramírez-Cruz and colleagues found them with three nuclear markers; Bradshaw and colleagues recovered the same early divergence from 2,983 single-copy gene families, with identical topology from a partitioned supermatrix and from summary coalescence of individual gene trees.

In the marker phylogeny, clade I contains a tropical group that splits into the "cordisporae", "mexicanae" and "zapotecorum" subgroups, alongside a temperate group of Psilocybe caerulipes, P. hispanica, P. pelliculosa and P. stuntzii distributed in Europe and North America. Clade II mixes tropical and temperate species with larger basidiospores, 10 to 14 micrometres long, and resolves into a group holding P. serbica and P. mescaleroensis close to the P. cyanescens complex, and a second holding P. cubensis, P. subcubensis, P. thaiaerugineomaculans and P. thaiduplicatocystidiata.

Bradshaw and colleagues estimated the stem age of Psilocybe at 67.61 million years and the crown age at 56.43 million years. They also state that their dating relied entirely on secondary calibrations, and that the absolute dates "should still be considered tentative estimates until primary calibrations are available or corroborating evidence is provided". The dates are reported here on those terms.

Psilocybin is produced by four core genes, PsiD, PsiK, PsiM and PsiH, which occur together in a biosynthetic gene cluster. Bradshaw and colleagues found two orders of those genes across the genus, and the boundary between them is the same ancient split. Clade II carries the order PsiD, PsiM, PsiH, PsiK, first reported from P. cubensis and P. serbica; clade I carries PsiD, PsiK, PsiH, PsiM. The first pattern appeared in 17 of their specimens and the second in 35.

PsiD, PsiK and PsiM were topologically congruent with the organismal phylogeny, which the authors read as vertical inheritance within the genus. PsiH was the exception, with duplicated copies in both clades, and copies in P. zapotecorum, P. mexicana and P. tampanensis truncated and so not functional.

Ancestral state reconstruction placed wood decay as the ancestral ecology of the two clades, with a shift to soil after the split and two independent specialisations on herbivore dung. The authors qualify this themselves: critical species are missing from the phylogeny, including the dung-dwelling P. natalensis and the early-diverging, non-psilocybin-producing P. fuscofulva, either of which may substantially alter the reconstruction.

Nineteen taxonomic sections were proposed for Psilocybe sensu lato by Guzmán (1983, 1995) and Guzmán et al. (2007a). Ramírez-Cruz and colleagues sampled 11 of them and report that most sections based on morphological features are not supported by their analyses. Section Semilanceatae in the sense of Guzmán (1983) is not monophyletic, and one clade of clade II groups species originally assigned to three different sections.

The characters the sections were built on do not track the tree. Basidiospore shape in frontal view and spore wall thickness, both commonly used to characterise sections in Psilocybe and Deconica, are reported as homoplastic, with spores bearing angles phylogenetically dispersed across both genera. Chrysocystidia, which had been thought diagnostic at family level, are entirely absent from Psilocybe sensu stricto and evolved on at least two occasions within Strophariaceae sensu stricto.

Species boundaries within the genus are unsettled too. Ramírez-Cruz and colleagues state that they suspect P. cyanescens and P. subaeruginosa represent the same species, on the basis of high genetic similarity and shared micromorphology.

Bradshaw and colleagues put the number of currently accepted species of Psilocybe sensu stricto at approximately 165. At the time of their study, the psilocybin gene cluster was known for five of them, and only 24 had been included in multi-locus molecular phylogenetic datasets.

Their own work reports phylogenetic relationships for 39 species, and a combined dataset of 67 unique taxa. Across all publicly available sequence data they found 40 species represented, about 24 per cent of the genus, and no species for which every gene region came from a single voucher specimen.

The molecular framework that the paragraphs above rest on therefore covers a minority of the named species. It is strong where it reaches and silent elsewhere, and that is the honest shape of current knowledge about the genus.

Three editorial decisions follow from the above. Family placement follows Agrocybaceae, the most recent published treatment that explicitly includes Psilocybe, while older placements retrieved from taxonomic databases are retained in the accounts as dated source history rather than dropped. Accounts are not organised by section, because the sections are not supported by the phylogenies this reference cites. Disagreements about rank and circumscription are recorded on the account they affect, and marking one resolved records an editorial choice about which treatment to follow, not the disappearance of the scientific disagreement.

This overview is a summary of published work on the genus, not new taxonomic research. It proposes no names, ranks or combinations, and the species index is not a checklist of Psilocybe: it holds the accounts researched so far, against a published estimate of roughly 165 accepted species.