As an undergraduate student back in Panama we took a class by Noris Salazar Allen on bryophyte morphology and systematics. One of the final projects consisted of collecting species of bryophytes. One of the commonest plants collected by all students was the cloud forest inhabitant Dumortiera hirsuta. It grows in wet places and close to streams, in close association with Monoclea gottschei and sometimes with the hornwort Nothoceros vincentianus. It is widely distributed and rather common in tropical America. It is also found in North America, Africa, mainland Europe and England, although unlike in my homeland, the plant is considered rare in England (see Porley 2013)
|Dumortiera hirsuta, habitat of Central American phenotypes (left and center). |
One of the glabrous morphotypes (left) and the velvety phenotype (center).
Enlarged view of the dorsal papillae of the velvety phenotype (right).
Dumortiera is a rather conspicuous plant with a large thallus up to 20 cm long (Gradstein et al. 2001), deep green colour, lacking pores, sometimes with vestigial air chambers and typically with hairs along the thallus margins (hence the name hirsuta). On the ventral part of the thallus the rhizoids are concentrated along the middle line and resemble a midrib. The male receptacles are nearly sessile while female receptacles are on an elongate stalk. There are no known forms of asexual reproduction. The circumscription of two subspecies (the glossy D. hirsuta subsp. hirsuta, typified by material from Jamaica and the velvety D. hirsuta subsp. nepalensis, typified by material from Nepal) has been contentious, a recurrent theme in widespread species. The two subspecies have been traditionally separated by the dorsal smooth surface of the thallus, with few (if present then rather spherical) or absent papilliform cells in D. hirsuta subsp. hirsuta, while subspecies nepalensis has vestigial air chambers and papilliform ovoid-pointed cells on its dorsal surface (Schuster 1992).
rule consensus tree from maximum parsimony analysis |
of the four locus matrix (rbcL, rpoC1, psbA-trnH and
the ITS region) from Forrest et al. (2011).
Recently, Forrest et al. (2011) sequenced 50 accessions of Dumortiera, mostly from Panama and Costa Rica, but with representatives from continental Europe, England, Ireland, Nepal, India, Malaysia, Mexico and the USA. Surprisingly (or not, considering the cryptic species mentioned in the last blog) there are two main clades, a central American one (Panama/Costa Rica) consisting of the plants that have a velvety appearance (which have been previously referred to subs. nepalensis) and a widespread clade that contains plants from Central America and the rest of the world. An English Dumortiera accession is related to French and Irish collections.
The morphological part of Forrest et al.’s study describes three main phenotypes within the species: (A) mostly glabrous thalli with remnants of air chambers and a few dorsal papillae; (B) glabrous thalli with remnants of air chambers only at the apex. Both these morphotypes have a hairy thallus border with abundant, distinctive marginal trichomes. The third morphotype (C) is characterized by velvety thalli with remnants of air chambers filled with dorsal papillae of variable sizes, without marginal trichomes. The velvety Dumortiera (subs. nepalensis) is from Nepal and the study shows that it is not genetically related to the Central American plants.
Based on Forrest et al.’s work there are at least two well-defined species within Dumortiera, one restricted to Central America, and one with a more widespread distribution. However, confusing the issue somewhat, the morphological features do not seem to be useful to circumscribe the widespread clade! A monograph is really needed to relate morphology to these robust molecular lineages.
A recurrent theme in bryophyte systematics is that widespread species have highly structured genetic clades that are not associated with clear-cut morphological characters. The implications are not only for taxonomists to untangle this problem, but also for conservation. What needs protection, restricted clades or widespread ones? It is possible that further sampling (for example within Africa and South America) will uncover more distinct lineages. Even in countries with really well-known floras, such as the UK, where Dumortiera hirsuta is listed as vulnerable, there are no specific conservation actions. In the tropics the situation is far more complex. A potentially exciting worldwide phylogeographic study (along with a taxonomic treatment) is required to sort out whether Dumortiera is a worldwide traveller that originated in the tropics, how many species there actually are within this lineage, and how to promote the conservation of this complex thalloid liverwort.
Thanks to Noris Salazar Allen, Jose Gudiño and Laura Forrest for pictures and comments.