Taxithelium juruense ( Broth . ) Broth . ( Pylaisiadelphaceae ) an endangered Brazilian endemic , with notes on the genus Taxithelium in Brazil

(Taxithelium juruense (Broth.) Broth. (Pylaisiadelphaceae) an endangered Brazilian endemic, with notes on the genus Taxithelium in Brazil). Since its original description, several collections of this species were reported in diff erent states of Brazil. During a revision of the genus, it was confi rmed that the geographic range of this species is restricted to the state of Acre, all other reports being based on misidentifi cations due mostly to the highly polymorphic nature of Taxithelium planum (Brid.) Mitt. Here, we present a more detailed study of T. juruense with characters that will help to distinguish it from T. planum. Th e misidentifi cations have obscured the real conservation status of this species, which should be endangered (EN). Key and illustrations of the Brazilian species of Taxithelium are provided.


Introduction
Taxithelium Mitt. is probably one of the most widespread moss genera in the Tropics.Th e genus is best represented between 30° N and 20° S, with most species occurring in Southeast Asia, especially the Malesian region (Damanhuri & Longton 1996;Ramsay et al. 2002).In the Neotropics it is represented by six species being T. planum (Brid.)Mitt.by far the most common species (Damanhuri & Longton 1996;Câmara 2011).
Taxithelium was fi rst recognized by Spruce (1867), but the name was only later validly published by Mitten (1869) in the tribe Sematophylleae.Brotherus (1925) placed the genus in the family Sematophyllaceae, and many taxonomists have subsequently followed this treatment (Vitt 1984;Buck & Vitt 1986;Tan & But 1997;Buck & Goffi net 2000;Ramsay et al. 2002).However, Taxithelium lacks the collenchymatous exothecial cells, long rostrate operculum, and inflated alar cells that are otherwise diagnostic for the Sematophyllaceae.Recent molecular studies (Buck et al. 2000;Tsubota et al. 2001a,b) show that the Sematophyllaceae sensu latu includes two sister clades: the core sematophyllaceous taxa and the so-called "the Brotherella lineage." Based on these results, Goffi net & Buck (2004) described the new family Pylaisiadelphaceae for the " Brotherella lineage." Although this group lacks an obvious morphological synapomorphy, Goffi net & Buck (2004) are followed here and Taxithelium is included in the Pylaisiadelphaceae.
Th e genus has been defi ned by the presence of multiple papillae disposed in series on the lumina of leaf cells (hence, Tax-"taso"= arranged and thelion = nipple), but there are some species in the Old World that have lost the papillae during evolution (Câmara & Kellogg 2010).Other characters that are useful in recognizing the genus in the New World are the complanate branches with leaves having an alar region with cells not nearly as well diff erentiated as in most Sematophyllaceae sensu latu.Th e genus has recently being revised worldwide and recircumscribed with two subgenera and twenty species (Câmara 2011).
Th e Brazilian species of Taxithelium were treated by Buck (1985) when he revised the genus for Brazil.He provided keys, maps, illustrations and descriptions.As expected since then, many other records have been cited elsewhere in Brazil, giving the impression that T. juruense has a wider distribution with many disjunctions.Th is was most likely due to misidentifi cations and caused much confusion.Th is may also lead to a wrong assessment of its conservation status.
In order to help solve this problem, aft er studying the type material and virtually all the Taxithelium collections known from Brazil, we were able to determine a more precise geographical range as well as provide a description of morphological traits that would enable its diff erentiation from other species within the genus, thus helping to avoid superfl uous records and to know better its conservation status.

Material and methods
Th is study is a consequence of a worldwide revisionary monograph for the genus and is based on the investigation of loans totaling 6200 specimens from 30 herbaria (B, BM, BR, CANB, DUKE, E, FH, G, H, INPA, JE, L, M, MG, MICH, MO, NICH, NY, NSW, PC, PHS, S, SING, SINU, SP, TSN, UB, UPS, US, W), including types for all Taxithelium species (see Câmara 2011 for more details on types).Specimens were also studied during our visits to herbaria RB, UFP, HUEFS, CEN, IBGE and HEPH.Specimens were re-hydrated in boiling water and then mounted in Hoyer's solution (Anderson 1954).All observations and measurements were made on mounted material.

Taxonomic treatment
Taxithelium juruense (Broth.)Taxithelium juruense is endemic to Brazil and grows as an epiphyte on twigs in lowland forests.It is only known to occur in the state of Acre.Aft er looking at virtually all plants named as T. juruense in the world, the authors concluded that the many citations from elsewhere in Brazil are misidentifi cations, mostly of T. planum, which is a highly variable species with many ecological variations.Buck (1985) treated this species and provided keys and illustrations.
Taxithelium juruense can be diff erentiated from T. planum by its smooth perichaetial leaves.Also the stem and branch leaves in T. juruense have revolute margins and a very poorly developed alar region (fi gure 1 and fi gure 2).Another useful character is the presence of papillae on the uppermost apex cells, while in T. planum those are usually absent.
It has also been confused with Hypnella pallescens (Hook.) A. Jaeger, even though both are pluripapillose.Hypnella (Pilotrichaceae) is a very distinct genus (in fact, it belongs to another order-Hookeriales), due to the presence of double costae with papillae that are sometimes bifurcated, which is unknown in Taxithelium (Câmara & Kellog 2010).
We believe that all material from foreign herbaria formerly misidentifi ed have been annotated; the same is true for Brazilian herbaria UB, SP, HUEFS, INPA, MG.Taxithelium juruense has a very restricted range (with only two localities known), and it has not been collected again in over 40 years, even with collecting projects happening in the region, such as "Florística e Botânica econômica do Acre, Brasil" (Floristics and economic botany in Acre, Brazil).Th is project started in 1992 in collaboration with the New York Botanical Garden and Universidade Federal do Acre and has recently made collecting trips in Juruá Mirim and Moa rivers, but no T. juruense was found.Even so, according to the IUCN (2001, 2010) criteria, T. juruense would fi t into the category Data Defi cient (DD).
However, due to bryophyte particularities, Hallingbäck et al. (1998) have published "Guidelines for applications of the revised IUCN threat categories".Following these guidelines T. juruense should be considered as an endangered species (EN), according to criteria "B" (restricted area of occupancy, recently recorded in one 10 km x 10 km square and found up to two localities/severely fragmented and in decline).
We adopt this more appropriate approach, since this species may well be extinct from nature or have an even more restricted area of occupancy.
Even though there is a red list for the threatened fl ora in Brazil (http://www.biodiversitas.org.br/floraBr/), Taxithelium juruense is not listed there.We strongly recommend its inclusion.It would be the fi rst threatened moss in the Amazon basin.
Other species of Taxithelium from Brazil: As stated above there are two other species of Taxithelium occurring in Brazil, neither one is threatened so far; they are T. planum (Figure 1A, B) and T. pluripunctatum (Figure 1D).Th ese species belong in two diff erent subgenera.Th e report of T. portoricense R.S. Williams for the Brazilian Amazon by Lisboa & Ilkiu-Borges (1997) was based on a misidentifi cation of T. pluripunctatum, (the paper itself shows a picture of T. pluripunctatum).Taxithelium portoricense is endemic to the Caribbean islands (Câmara 2011).
Taxithelium planum is a very common species known to occur widely from southern North America to South Brazil and also Africa.It has been misreported from Asia because it resembles the Asian species Taxithelium nepalense (Schwägr.)Broth.As stated above it is a highly variable species; leaf shape variation can be seen in Fig. 1A,  B. Another useful character is its somewhat developed alar region when compared to the other two (Figure 2). In

Key to the Brazilian species of Taxithelium
Conservation status: Taxithelium juruense is endemic to Brazil and is known to occur only in two localities in Acre state: on the Jurua-Miry and Moa rivers.Th e fi rst collections were 200 made by the naturalist Ernest Ule 2295, 2307 (H), between 1901 and 1902 on the Juruá Mirim River (originally spelled Juruá-Miry).Even though the type collection examined at herbarium H does not contain exact data on locality, duplicates from that expedition deposited at NY clarify that it happened in the municipality of Porto Valter.Later, aft er 60 years, Prance 12541 (NY) recollected the species by Moa River, municipality of Mâncio Lima, about 50 km from Porto Valter.

Figure 1 .
Figure 1.Diff erences in leaf shape from the three species of Taxithelium from Brazil.A-B-T.planum; C-T.juruense; D-T.pluripuncatum.
Brazil Taxithelium planum is distributed widely in the states of AC, AL, AM, AP, BA, DF, ES, GO, MA, MG, MS, MT, PA, PB, PE, PR, RJ, RO, RR, SC, SP, TO.It ranges from 0-1090 m.Taxithelium planum is differentiated from T. pluripunctatum by the presence of foliose pseudoparaphyllia and ovate complanate leaves.It also has papillae of a baggy shape (Câmara & Kellogg 2010) whereas T. pluripunctatum has conical papillae.Taxithelium pluripunctatum is a strictly Neotropical species known to occur mostly from Central America to northern Brazil in the states of AM, BA, PA, PE, RR.It is a lowland plant known to occur in Brazil from 0 to 200m.

Figure 2 .
Figure 2. Diff erences among the alar region in the three species of Taxithelium from Brazil.A-T.juruense, B and C-T.planum; D-T.pluripuncatum.