Fungal Taxonomy II:
The Higher Fungi

© Peter Werner
Original publication: Mycena News, December 2003

In my last article, I discussed how advances in molecular systematics had revolutionized our understanding of relationships among the fungi, and between fungi and other organisms. I also discussed some recent discoveries about the relationships of the basal fungi (the so-called "primitive" or "lower" fungi). In this article, I will discuss how molecular systematics has advanced our understanding of relationships among the fleshy ascomycetes and basidiomycetes that we are so familiar with.

The Ascomycota and Basidiomycota together form a monophyletic clade that is, in turn, a sister clade to the Glomeromycota, the newly discovered phylum that includes the VAM fungi and their relatives. Some long-accepted ideas about the relationships among ascomycetes and basidiomycetes have been confirmed by recent molecular data. For example, it has long been thought that the Ascomycota and Basidiomycota are more closely related to each other than they are to other fungal groups, sharing such traits as a spitzenkörper on the growing hyphal tip, and, unlike other fungal groups, completely lacking coenocytic hyphae. This relatively close relationship has been borne out repeatedly in molecular analysis.

The relationships within the Ascomycota were assumed to more or less correspond with the morphology of the fruiting structure, hence all ascomycetes with apothecia (a cup- or saddle-shaped fruiting structure) were classed as Discomycetes, all ascomycetes with perithecia (flask-shaped fruiting structures) were Pyrenomycetes, and so forth. Its now clear that this classification does not hold up well for at least several groups of ascomycetes. Cup-shaped ascocarps evolved independently at least twice, for example, and there is even evidence that the yeasts independently evolved at least twice. The majority of fleshy mushroom- and truffle- forming ascomycetes are found within one taxon, the Pezizales, which are distinguished by an operculate opening at the tip of each ascus. (This trait has been lost in the truffles, which, like most hypogeous fungi, are incapable of forcible spore discharge.)

Within the Basidiomycota, it has also long been hypothesized that the rusts and smuts form the more basal clades, and that the homobasidiomycetes (fleshy basidiomycetes) and the jelly fungi form a clade that is probably closest to the smuts. Basidial morphology pointed to the hypothesis that the homobasidiomycetes form a monophyletic clade derived from the jelly fungi. These ideas about basidiomycete evolution have largely been confirmed by molecular analysis.

However, when we examine the subject of the evolution of the homobasidiomycetes, we see an area where many of our old ideas are being subject to radical revision. Like early hypotheses about the ascomycetes, the different morphologies of homobasidiomycetes were presumed to each correspond to a distinct taxonomic group. Hence, agarics, gasteromycetes, polypores, and so on were each thought to be distinct monophyletic lineages. Detailed microscopic and developmental studies of many of these fungi have revealed that many outwardly similar taxa in reality had very different underlying morphologies.

In the last several years, molecular analysis has revealed that many of these morphological states have in fact evolved numerous times. Agarics are a good example - while the majority of agaric species belong to one clade, the euagarics, it is clear that there are at least four other separate evolutionary lines that have produced agaric species. The most notable group is the Russulales (Russula and Lactarius), who's closest relatives are woody resupinate fungi like Stereum. Hence, Russula, which for all outward appearances looks as much like an agaric as, for example, Tricholoma, in fact represents a completely separate line of evolution from a polypore-like ancestor to an agaric morphology. Similarly, Panus, Lentinus, and Lentinellus each represent separate lines of evolution from closely-related woody fungi. The gasteromycetes also represent at least three separate evolutionary lines; the Lycoperdaceae turn out to be very close relatives of Agaricus and Lepiota, Scleroderma and Pisolithus are related to the boletes, and earthstars and stinkhorns are relatives of Gomphus and Ramaria. Woody fungi, such as polypores, turn up in five of the eight major evolutionary lines of homobasidiomycetes; this type of morphology may have evolved many times independently, or it may be an ancestral state in several of the evolutionary lines of fleshy fungi.

The relationships between the various clades of fleshy basidiomycetes is still, for the most part, not well resolved. There is a vague pattern that shows the club, coral, and cantharelloid fungi branching off early in the evolution of fleshy basidiomycetes. The russuloid fungi and their woody resupinate relatives branched off from the ancestors of the euagarics and the boletes some time after this. However, the close relationship of the euagarics and boletes is now very well-established. The two are sister taxa, descending from a relatively recent common ancestor. (Whether this ancestor was agaricoid or boletoid is unclear.)

The boletoid clade is notable for its extreme morphological diversity. This clade contains not only boletes, but also several groups of gastroid and hypogeous fungi, several groups of agarics, and even Serpula, the dry-rot fungus. These varying morphologies are scattered throughout the boletoid clade, and surprisingly, not all boletes are directly related to other boletes. Boletus and Suillus are on widely separate lines of boletoid evolution, with Suillus being a close relative of Rhizopogon and the gophidiaceous agarics. Boletus, for its part, shows closer affinity to several species of Paxillus than to Suillus. The bolete genus Gyroporus is very close to a gastroid clade that includes Pisolithus and Scleroderma. Rounding out this extreme morphological diversity are number of paxillaceous agarics that are found scattered at different points throughout the boletoid clade.

The relationships we are discovering within the euagarics are no less surprising and represent a complete revision of how we have viewed agaric "Families" in the past. I will cover this topic in my next article.

Further reading: