Myriopteris

No single morphological character divides Myriopteris, as presently circumscribed, from the other cheilanthoids. Convergent evolution in arid environments is thought to be responsible for widespread homoplasy in the morphological characters traditionally used to classify this group.[1] While small, bead-like ultimate segments are associated with the genus, they only appear in about 40% of its species,[2] and appear in some cheilanthoids outside the genus as well. Cheilanthes sensu stricto bears 32 spores per sporangium in sexual species and 16 in apogamous species; with the exception of a few species of Notholaena, Myriopteris and the other cheilanthoids bear 64 spores per sporangium when sexual and 32 per sporangium when apomictic. Myriopteris can also be separated from Cheilanthes s.s., although less reliably, by spores bearing crests or wrinkles (rather than spines or bumps) and a lack of enlarged vein endings (rather than prominent hydathodes).[1]

The three subgroups of the genus, informally called the alabamensis, covillei, and lanosa clades, likewise lack unique defining features. The small, bead-like ultimate segments are found in the core subclade of the covillei clade and also in M. gracilis, which is deeply nested in the alabamensis clade.[3] Circinate vernation (the unfolding of new leaves as fiddleheads) is found throughout the lanosa clade and also in M. wrightii, the most basal member of the alabamensis clade.[3] Most species have round rachises, although early-diverging members of the alabamensis and lanosa clades have rachises deeply grooved on the upper surface and flattened rachises shallowly grooved near the frond tip, respectively.[2] Leaf indument (hairs and scales) is highly diverse across the genus and a key feature in species identification.[2]

Myriopteris covillei has large, prominent scales beneath the leaf.

The base chromosome number for the genus appears to be x=30, except for a portion of the alabamensis clade, where x=29; the latter is typical for other cheilanthoids. Most species are either sexually reproducing diploids or apogamous triploids, with the exception of M. lendigera and some specimens of M. microphylla and M. scabra, which are sexually reproducing tetraploids.[4]

The genus was first described in 1852 by A.L.A. Fée, who separated it from Cheilanthes proper by the presence of red hairs among the sporangia and a scarious (hardened) indusium formed from the leaf margin. He typified it on Myriopteris marsupianthes. Fée described the division of the leaf into numerous small, beadlike segments, all capable of bearing spores,[5] which may have led him to choose the name Myriopteris; "myrio-" means "very many"[6] and "pteris" means "fern". John Smith recognized Myriopteris in his Cultivated Ferns of 1857, noting the "minute, orbicular or cuneiform, concave" ultimate segments typical of species in the genus.[7] However, most authors until the 21st century preferred to include the genus in Cheilanthes.[8]

The development of molecular phylogenetic methods showed that the traditional broad circumscription of Cheilanthes is polyphyletic. Many of the morphological characters that have traditionally been used to separate the cheilanthoid ferns into genera, including Cheilanthes, are homoplasious; that is, they have appeared independently in unrelated groups, probably as a result of convergent evolution in arid environments.[1]

A molecular phylogeny of 157 cheilanthoid species by Michael D. Windham et al. in 2009 revealed seven well-supported clades within the group. One of these included a number of mostly North American species usually placed in Cheilanthes sensu lato. They informally referred to this group as the myriopterids, Myriopteris being the most senior genus whose type specimen was included in the clade.[9] Further phylogenetic analysis by Eiserhardt et al., in the course of a study on cheilanthoid evolutionary radiation in the Cape Floristic Region, also supported the existence of the myriopterid clade and showed that Cheilanthes rawsonii, an African endemic, was deeply nested in it, its closest relative being C. parryi.[10] After more extensive sampling within the clade, Amanda Grusz and Windham revived Myriopteris in 2013 and provided names in it for all species in the myriopterid clade.[1]

Further molecular studies by Grusz et al. confirmed the monophyly of this group, and showed that its species can be divided among three well-supported clades. These were informally referred to by the epithet of a prominent species in the clade, without formal taxonomic rank; they are the alabamensis clade (A), covillei clade (C), and lanosa clade (L).[11]

The genus Cheilosoria was described by Conde Vittore Trevisan in 1877 to accommodate Cheilanthes allosuroides, now M. allosuroides, and a few other species of Cheilanthes with long sori along the veins and relatively unmodified false indusia.[12] That genus was lectotypified on C. allosuroides by Edwin Copeland in 1947.[13] If the alabamensis clade were to be treated as a separate genus, Cheilosoria would be the senior name for it.[14]

The following species (including two hybrids) are those recognized by Grusz & Windham in 2013, with some additions from the Checklist of Ferns and Lycophytes of the World (version 8.11).[15] Letters in parentheses following the scientific names indicate which of the three clades the species belong to, if known.

Myriopteris parryi growing in a rock outcrop in Death Valley.

The greatest diversity of species occurs in Mexico.[16] Myriopteris species are mostly confined to the Americas, ranging from southwestern Canada[17] to southern Chile. M. rawsonii is endemic to Namibia and South Africa.[16] It is most closely related to M. parryi,[10] a species of the Sonoran and Mojave Deserts.[4]

Like either cheilanthoids, Myriopteris species occupy dry habitats, growing on rocks or soil.[18]