Typification of some species names in Elaphoglossum section Polytrichia ( Dryopteridaceae ) from Brazil

The fern genus Elaphoglossum has received a great deal of attention in Brazil over the last two centuries. Nevertheless, many of the early names remain inadequately typified. In this paper, the nomenclature of some Brazilian species of Elaphoglossum sect. Polytrichia is discussed under the rules and recommendations of the International Code of Nomenclature for algae, fungi, and plants. Lectotypes are designated for the following names: Acrostichum amplissimum Fée; A. apodum Kaulf. var. sprucei Baker; A. glaziovii Fée; A. lindbergii Mett. ex Kuhn; A. prestonii Baker; Elaphoglossum spannagelii Rosenst.; and Elaphoglossum ulei H. Christ. Most of these types were collected in Brazil during the 19th century and are now preserved in several European herbaria.


Introduction
During the preparation of a taxonomic revision of Elaphoglossum Schott ex J. Sm.(Dryopteridaceae) in Brazil, I found several scientific names that needed to be lectotypified.Many of these names were first published in Acrostichum L., which formerly included all ferns with acrostichoid sori.It is realized nowadays that such an assemblage results in a paraphyletic group containing species now placed in distinct families (see Smith et al. 2006).Within Brazil, Elaphoglossum has received a great deal of floristic attention (e.g., Fée 1869;1873;Baker 1870;Rosenstock 1906Rosenstock -1907;;Alston 1958;Brade 1961;1965;Sehnem 1979;Windisch & Kieling-Rúbio 2010), but many names have yet to be typified.Brade (1966) studied some types that were collected by Glaziou in Brazil and are now preserved at the Herbarium of the National Museum of Natural History in Paris, France (herbarium P).He cited many of these collections as "typus, " in what was then an effective lectotypification, but without designating a specific specimen as a lectotype.This situation is covered by Article 9.17 of the International Code of Nomenclature (McNeill et al. 2012), which provides for the narrowing of a lectotypification from such a "first-step" designation of a gathering to a "second-step" designation of a single specimen.Rouhan & Cremers (2006) also lectotypified some names from Brazil.
In this paper, I clarify the names of certain species of Elaphoglossum sect.Polytrichia H. Christ described from Brazil.The members of this section can be recognized morphologically by the presence of subulate scales and absence of hydathodes on the leaves of adult sporophytes (Mickel & Atehortúa 1980;Matos & Moran 2013).

Material and methods
This study was largely based on the examination of types and other specimens from the following herbaria: AAU, ALCB, B, BHCB, BM, BR, CEPEC, CGE, COL, CR, F, GH, HB, HBR, HUEFS, INB, INPA, JBSD, K, LE, LIL, LSCR, M, MBM, MEXU, MO, NY, P, PH, PI, R, RB, S, SP, SPF, UC, UPCB, and US (acronyms according to Thiers 2013).Many of these herbaria are currently in the process of databasing their collections, and in several cases the sheets are provided with a barcode label and number serving as a unique identifier.Whenever possible, these numbers were used here to refer to a particular specimen of a gathering.Most of the information about botanical publications, authors, collectors, dates, and types was taken from Urban (1906) and Stafleu & Cowan (1976-1988).In some cases, I included additional information about collecting localities in the nomenclatural paragraphs.Such information is enclosed in square brackets and was inferred from reliable external evidence other than the protologue.There are four sheets of Glaziou 2436 at P. From among these, I select P00249820 as the lectotype because it has Fée's original label and a paper tag with the collection number (n. 2436) attached to the petiole.It also has a handwritten note suggesting that the specimen was used to prepare the illustrations for the original publication (Fée 1869: Tab. 6).
Sheet number P01398966 bears the same collection number as the type (i.e., Glaziou 2436), but should not be considered part of the original material for two reasons: First, the label has a different collection date (i.e., 1870), indicating that the specimen was collected after the name was validly published.Second, this specimen was probably never examined by Fée, as suggested by a label stating that it was transferred to P from the "Herbarium Cadomense" (CN) on 6 November 1974.
Sheet number P01604287 of Glaziou 2059 should not be considered an isolectotype for the same reasons presented above for P01398966 (see under Acrostichum amplissimum Fée).
In tentatively transferring Acrostichum glaziovii Fée to Elaphoglossum, Brade (1961) did not provide a full and direct reference to the basionym and its place of valid publication.Being made after 1 January 1953, this new combination was not validly published (McNeill et al. 2012: Art. 41.5).After five years, however, the same author validly published the new combination E. glaziovii (Fée) Brade (Brade 1966) by indicating the basionym, as well as its author and place of valid publication.This infraspecific taxon (as "Var.ß. minor") was described from a single specimen, Gardner 5925 (BM).This specimen must, therefore, be regarded as the holotype (McNeill et al. 2012: Art. 9, Note 1).The sheet contains an entire plant, with two sterile leaves and one fertile leaf attached to a short creeping stem.= Elaphoglossum hybridum (Bory) Brack., U.S. Expl.Exped., Filic. 16: 69. 1854.

Acrostichum prestonii
Baker stated that Acrostichum prestonii was described partly from dried specimens collected by Glaziou in Brazil and partly from living material cultivated at Kew Gardens.I found three sheets annotated as A. prestonii at K: two of them collected by Glaziou 5367, and one (K000501503) undoubtedly vouchered from a cultivated plant.The cultivated plant, which was brought from Brazil by Rev. T.A. Preston, represents another species, Elaphoglossum glaziovii (Fée) Brade, and should be disregarded as a type.Although Glaziou's collection number was not mentioned in the protologue, Baker indicated Glaziou 5367 in a subsequent publication (Hooker & Baker, Syn. Fil., 520. 1874).
Working with a different set of Brazilian collections sent by Glaziou, Fée (1873) described Acrostichum insigne based on the same type of A. prestonii Baker.By definition, A. insigne Fée is a superfluous name and must be rejected (McNeill et al. 2012: Art. 52.1).Consequently, the more recent combination Elaphoglossum insigne (Fée) Brade (Brade 1956) is illegitimate because it is based on a superfluous name.Acrostichum insigne Baker (1877)  There are four sheets of Glaziou 5367 at P, including one with an original tag.The type collection of Acrostichum insigne Fée was cited in the protologue as Glaziou 5369, but this was a typographical error.Indeed, Glaziou 5369 is cited again on the next page of Fée's publication, as the type of A. omphalodes (Fée 1873: 7).Other duplicate specimens are found at B, K, and S. All of these are part of the same gathering and should be considered syntypes.There are two duplicates of the original material at Kew, both being sterile but complete (i.e., leaves attached to a stem).Between them, I select the herbarium sheet that has three leaves (K000994040) as the lectotype of A. prestonii.Rosenst., Hedwigia 46: 153. 1907  Although not mentioned in Taxonomic Literature II (Stafleu & Cowan 1983), many types of ferns described by Rosenstock are now deposited at S. The S material no.S-R-1881 is selected as the lectotype of Elaphoglossum spannagelii, even though it has teratological (i.e., lobed) leaves.The sheet bears an authentic label noting "Elaphoglossum Spannagelii Ros.n. sp." and stamped with "Dr.E. Rosenstock, Gotha", indicating that it belonged to Rosenstock's personal herbarium.Duplicates of this collection are kept at HB, HBR, R, and UC.

Elaphoglossum spannagelii
The UC specimen (UC443383) has an authentic label printed with the heading "E.Rosenstock Filices brasilienses".It is annotated as "Spannagel 2", but all other information in the label matches Spannagel s.n., including the collection date (i.e., 1905)  There are two sheets of Ule 327 at P and one at B. One of the specimens from P (P00249984) has an authentic stamp indicating that it once belonged to "Herb.H. Christ, " and for this reason I have selected it as the lectotype of Elaphoglossum ulei.
The specimen at B (B200070088) is a mixed collection containing three gatherings.Only Ule 327, which is composed of 10 plants in a small paper envelope, corresponds to the protologue.Therefore, it is designated here as an isolectotype.
, who did not locate the types, in his treatment of Elaphoglossum for Peru.
Spannagel s.n., also collected in Lages, SantaCatarina, but on March 1905; and(3) Wielewski s.n., collected in Lucena, Paraná [which is now the municipality of Itaiópolis in Santa Catarina], in 1903.Among them, Spannagel s.n. is the only one that was not mentioned in the protologue.Nevertheless, all of its specimens have an authentic label headed "Dr.E. Rosenstock, Filices austrobrasilienses exsiccatae -n.3".Duplicates of this gathering are preserved in several herbaria in Europe and the United States, and should be regarded as syntypes.