Ferns are foreign. I feel always at some distance from them. To help prepare my lectures on ferns this week, I went to the library for F. O. Bower’s multivolume monograph The Ferns, which dates from the 1920s. While some of the ideas in The Ferns may be dated, Bower wrote well, and he was sharply insightful. I like also the simple line drawings that illustrate the volumes, and I scanned several to use in my lectures. The tone of the fern story is improved, I think, by discussing the most recent phylogenetic trees, which use DNA sequence data to reconstruct relationships among ferns, in the context of slightly yellowed, but simple and clear drawings from Bower’s books.
Since Bower’s time, the idea of a fern has decayed. The old assemblage of ferns consisted of those plants that had fronds—mostly leaves that were lobed or divided like a feather, a form described as pinnate—and had sporangia on the lower surface of the leaves, although the moonwort ferns were always an oddity when it came to their sporangia. In the old assemblage of ferns, there were those recognized to have delicate sporangia, which were poised on thin stalks and had a thin wall to surround the spores, and others characterized by relatively massive sporangia. Those with delicate sporangia were sometimes called the ‘true ferns’ in contrast to those that had ‘indelicate’ sporangia, the moonworts and marattiid ferns
Over the past decade, a group of fern biologists, led by Kathleen Pryer, has produced a series of fascinating phylogenetic studies of ferns. Their results have shown that the ‘true ferns’ are not the closest relatives of either the moonworts or the marattiid ferns. The moonworts and marattiid ferns have fallen away from the ‘true ferns’ to lie among the plants once called fern allies. Pryer and her co-workers have found that the closest living relatives of the marattiid ferns may be horsetails, and the moonworts find their closest relatives to be the oddities sometime known as whisk ferns, the genera Psilotum and Tmesipteris.
Fossil relicts tied to the true ferns, horsetails and marattiid ferns suggest that these were already independent evolutionary lineages over 340 million years ago in the time we call now the Upper Devonian. Missing from the assemblage of fossils that reach deep into the history of land plants are the whisk ferns and moonworts. The extant whisk fern Psilotum has a form marvelously similar to those found generally among the earliest vascular plants of the Devonian, but Psilotum has no known fossil record of its own. The moonworts, the sister lineage of the whisk ferns, have only a relatively shallow fossil record—one composed largely of forms present still today. Where are those fossils of moonworts and whisk ferns?
Pryer and her colleagues have used DNA sequence differences to estimate that moonworts and whisk ferns shared a common ancestor as recently as 310 million years ago, a date somewhat later than the evolutionary divergences that have given us ferns, horsetails, and marattiid ferns. Did that common ancestral lineage of the moonworts and whisk ferns have such a generalized form, so lacking in distinctive attributes, that we can’t see its links to those relatives that are alive today?
* * *
Among the publications by Kathleen Pryer and her colleagues, I’ve used especially the following two recent papers for background:
Pryer, K.M., A.R. Smith and J.E. Skog. 1995. Phylogenetic relationships of extant ferns based on evidence from morphology and rbcL sequences. American Fern Journal 85 (Special Issue): 205-282.
Pryer, K.M., E. Schuettpelz, P.G. Wolf, H. Schneider, A.R. Smith, R. Cranfill. 2004. Phylogeny and evolution of ferns (monilophytes) with a focus on the early leptosporangiate divergences. American Journal of Botany 91: 1582-1598.