Maidenhair Ferns, Adiantum, are Indeed Monophyletic and Sister to Shoestring Ferns, Vittarioids (Pteridaceae)

Across the tree of life, molecular phylogenetic studies often reveal surprising relationships between taxa with radically different morphologies that have long obscured their close affiliations. A spectacular botanical example is Rafflesia, a holoparasite that produces the largest flowers in the world, but that evolved from tiny-flowered ancestors within the Euphorbiaceae. Outside of parasitic lineages, such abrupt transformations are rarely seen. One exception involves the “maidenhair ferns” (Adiantum), which are quintessential ferns: beautifully dissected, terrestrial, and shade loving. The closely related “shoestring ferns” (vittarioids), in contrast, have an extremely simplified morphology, are canopy-dwelling epiphytes, and exhibit greatly accelerated rates of molecular evolution. While Adiantum and the vittarioids together have been shown to form a robust monophyletic group (adiantoids), there remain unanswered questions regarding the monophyly of Adiantum and the evolutionary history of the vittarioids. Here we review recent phylogenetic evidence suggesting support for the monophyly of Adiantum, and analyze new plastid data to confirm this result. We find that Adiantum is monophyletic and sister to the vittarioids. With this robust phylogenetic framework established for the broadest relationships in the adiantoid clade, we can now focus on understanding the evolutionary processes associated with the extreme morphological, ecological, and genetic transitions that took place within this lineage. Keywords—Epiphytes, gametophytes, molecular phylogeny, rate heterogeneity. Early molecular phylogenetic analyses of ferns (Hasebe et al. 1994, 1995) inferred several unexpected associations that had not previously been suspected. Most of these newly recognized relationships, which subsequently drew considerable attention, have stood the test of time. A prominent example is the monophyly of the heterosporous water ferns in the Marsileaceae and Salviniaceae (Hasebe et al. 1994; Rothwell and Stockey 1994; Pryer et al. 1995; Pryer 1999). Another surprise emerged within the Pteridaceae, grouping Adiantum (maidenhair ferns) together with the vittarioids (shoestring ferns) in a well-supported clade now referred to as the adiantoids (Schuettpelz et al. 2007). Adiantum and the vittarioids could not be more morphologically or ecologically disparate. In coarse morphology, their conspicuous sporophytes look nothing like one another. The leaves of Adiantum are typically broad and finely divided, whereas those of vittarioids are almost always simple and strap-like (Fig. 1A; Tryon and Tryon 1982; Kramer 1990). Fertile Adiantum leaves are uniquely distinguished by their sporangia borne on, and limited to, false indusia, whereas vittarioid sori occur on the laminae (Crane et al. 1995). These groups also display major differences in the morphology of their gametophytes, although these are less obvious to the naked eye (Fig. 1B). The gametophytes of Adiantum, like those of most ferns, are determinate and heart-shaped, with a distinct midrib and broad wings (Nayar and Kaur 1971). They are generally ephemeral to short-lived (months) and are incapable of vegetative reproduction. Vittarioid gametophytes, on the other hand, are indeterminate and ribbon-like. They can be exceptionally long-lived (years) and can also reproduce asexually via propagules called gemmae (Atkinson and Stokey 1964; Farrar 1974, 1985). In addition, Adiantum and vittarioids occupy two dramatically different niches. Whereas the cosmopolitan genus Adiantum usually occurs on shady forest floors, vittarioids generally grow as epiphytes, colonizing tree trunks and canopies of tropical rain forests. The differences between these two groups also extend to their genomes. Although most Adiantum species have diploid chromosome numbers of n = 29 or n = 30 (Löve et al. 1977), nearly all vittarioids studied are n = 60 (Löve et al. 1977), suggesting that at least one genome duplication event occurred early in the evolutionary history of this lineage. Because of their distinctive simplified morphology, vittarioids have until quite recently been regarded as a distinct family, the Vittariaceae (Tryon and Tryon 1982; Kramer 1990). However, in recent years, studies have not only identified a close relationship with Adiantum, but also suggested that perhaps vittarioids may even be nested within this genus (Prado et al. 2007; Schuettpelz and Pryer 2007). These analyses of plastid data have further revealed yet another dissimilarity between Adiantum and the vittarioids: a striking difference in branch lengths. Vittarioid branches are extraordinarily long relative to Adiantum, or to any other fern lineage, for that matter. As a consequence, branch support across the vittarioid topology is consistently robust, whereas there is mostly weak support among species of Adiantum, especially for the backbone nodes that lie far deeper than the tips (Schuettpelz and Pryer 2007; Schuettpelz et al. 2007). While it is clear that Adiantum and the vittarioids together compose a robust clade, there are still unanswered questions regarding the monophyly of Adiantum and the evolutionary history of this genus, as well as that of the vittarioids. For example, are there correlates in the morphology, ecology, and life history of vittarioid ferns that may be contributing to their faster rate of molecular evolution? Here we review recent studies and analyze new plastid data, and find strong support for the monophyly of Adiantum and for other deep divergences within adiantoids. This robust phylogenetic framework will permit us, in future studies, to explore the evolutionary processes that resulted in this extreme morphological, ecological, and genomic makeover