(2942) Proposal to conserve the name Dicksonia bipinnata (Dennstaedtia bipinnata, Mucura bipinnata) (Dennstaedtiaceae) with a conserved type

Abstract

(2942) Dicksonia bipinnata Cav., Descr. Pl.: 274. 1802 (ante 2 Mar), nom. cons. prop. Typus: [Puerto Rico?], Ventenat (B-W barcode B -W 20165 -01 0; isotypus: US No. 1148434 [barcode 00745250] [fragm.]), typ. cons. prop. The Dennstaedtiaceae is a medium-sized family of ferns with puzzling nomenclatural history and issues, either by the numerous species conceptions and rank adoptions within the bracken ferns – for example, the epithet of Pteridium arachnoideum (Kaulf.) Maxon has also been used at three infraspecific ranks under three different species – or by the conservation proposals to maintain nomenclatural stability in the family. There are at least two species with conserved names (Pteris arachnoidea Kaulf. and Hypolepis nigrescens Hook. – see App. IV of the ICN; Turland & al., in Regnum Veg. 159. 2018), and, among its 11 currently recognized genera, three names are conserved or proposed for conservation: Dennstaedtia, nom. cons. prop.; Microlepia, nom. cons. prop.; Pteridium, nom. cons. (see https://naturalhistory2.si.edu/botany/codes-proposals/). Within the last decade, phylogenetic works have demonstrated the paraphyly of the Dennstaedtiaceae genera, such as by Perrie & al. (in Austral. Syst. Bot. 28: 256–264. 2015), Schneider & al. (in Perspect. Pl. Ecol. Evol. Syst. 18: 70–78. 2016), Shang & al. (in Molec. Phylogen. Evol. 127: 449–458. 2018), and Schwartsburd & al. (in Molec. Phylogen. Evol. 150: 106881. 2020). Shang & al. (l.c.) were first to organize the species of subfamily Hypolepidoideae into monophyletic genera, describing the new genus Hiya H. Shang and slightly altering the previous circumscription of Hypolepis Bernh. Just recently, Triana-Moreno & al. (in Taxon 71: 688–690. 2022, 72: 20–46. 2023) dealt with subfamily Dennstaedtioideae, making substantial changes in generic circumscriptions (especially in Dennstaedtia s.str.), resurrecting Sitobolium Desv. from synonymy, and describing the new genus Mucura L.A. Triana & Sundue to accommodate two Neotropical species previously known as the “alate Dennstaedtia”: M. bipinnata (Cav.) L.A. Triana & Sundue and M. globulifera (Poir.) L.A. Triana & Sundue, with the former taken as type. But just when the systematics of Dennstaedtiaceae appeared to be finally resolved, a new nomenclatural issue came along: there is a serious historical problem with the type of Dicksonia bipinnata Cav. (and thus, with the type of the newly described Mucura), that needs to be exposed and solved. We explain it below and propose a reasonable solution. Dicksonia bipinnata was described by Cavanilles (Descr. Pl.: 274. 1802), based on a specimen sent by Ventenat, without location data. According to Stafleu & Cowan (in Regnum Veg. 94 [TL-2, 1]: 471. 1976), in 1802 Cavanilles was already established at the Madrid botanical garden (MA), and so his types are expected to be there. During our search, we found a Ventenat specimen that matches Cavanilles's description and has the name “Dicksonia bipinnata” on the label: MA No. 213495 (available at: http://161.111.171.57/herbarioV/visorVCat.php?img=MA-01-00213495). This specimen consists of part of a leaf, representing a bipinnate-pinnatifid plant. This specimen has no adaxial wings that connect an axis onto the next order of axes, which is one of the main characters defining the newly described Mucura (= “alate Dennstaedtia”), and helps to easily diagnose it. In other words, the type of Mucura does not represent the description of Mucura provided and intended by Triana-Moreno & al. (l.c. 2023: 40); it is a specimen of Dennstaedtia s.str. Dicksonia bipinnata should, in fact, be treated as a synonym of Dennstaedtia dissecta (Sw.) T. Moore, if we strictly follow the rules of the Code (Art. 7 and 11). By extension, the newly described Mucura becomes a synonym of Dennstaedtia s.str., and the two alate species of Dennstaedtia s.l. will need to be placed in a new as-yet-unnamed genus, with one of them requiring a new specific epithet. But, Triana-Moreno & al. (l.c. 2023) are not entirely responsible for this confusion. The puzzle started in 1810, when Willdenow described Dicksonia adiantoides Humb. & Bonpl. ex Willd. (Sp. Pl. 5: 488. 1810). Willdenow (l.c.) did not explicitly cite specimens, but he cited three locations (Caripe, Hispaniola, Puerto Rico), a Plumier plate (Traité Foug. Amér.: t. 30. 1705), and Dicksonia bipinnata Cav. in synonymy. Christensen (in Dansk Bot. Ark. 9: 26. 1937) and Maxon (in Proc. Biol. Soc. Wash. 51: 39–40. 1938) determined that Caripe referred to a Humboldt & Bonpland specimen at B-W (barcode B -W 20165 -02 0, available at https://herbarium.bgbm.org/object/BW20165020), Hispaniola referred to Plumier's plate (l.c.), and Puerto Rico (and the synonymy of D. bipinnata) referred to a Ventenat specimen at B-W (B -W 20165 -01 0, available at https://herbarium.bgbm.org/object/BW20165010), with a fragment later forwarded to US (barcode 00745250). The Ventenat specimen at B-W is also a part of a leaf, but representing a tripinnate-pinnatifid plant. This specimen has, indeed, the typical adaxial, connecting wings present in the “alate Dennstaedtia” (or in Mucura, as its description proposes). In conclusion, the Ventenat specimens at MA and B-W do not correspond to the same taxonomic entity; they are not the same species; and they are even in different genera sensu Triana-Moreno & al. (l.c. 2023). In our interpretation, they should not be considered as duplicates. As Christensen (l.c.) suggested, the Ventenat specimen at MA would have come from a plant cultivated in European botanical gardens, whereas Maxon (l.c.) believed the B-W specimen to really come from Puerto Rico. It is very probable that Willdenow (l.c.) believed his Ventenat specimen (at B-W) to be a duplicate of the MA specimen, since he synoymyzed Dicksonia bipinnata under his newly described Dicksonia adiantoides – this led Maxon (l.c.) to speculate that Willdenow's change of epithet, from bipinnata to adiantoides, was due to its morphological inaccuracy. The misinterpretation of the identity of these two Ventenat specimens (at B-W and MA) has persisted since the works of Christensen (l.c.) and Maxon (l.c.). Christensen (l.c.) suspected the specimens to be duplicates and suggested Dicksonia adiantoides to be synonymyzed under Dicksonia bipinnata, since the latter name had priority. Maxon (l.c.) accepted Christensen's ideas and formally synonymized Dicksonia adiantoides under the newly combined Dennstaedtia bipinnata (Cav.) Maxon, although he admitted the Ventenat specimen in Willdenow's herbarium to be “almost certainly a part of the original collection” [of Dicksonia bipinnata]. Having seen only the fragment of the B-W specimen at US, Maxon (l.c.) failed to indicate which of these specimens he considered the type (B-W or MA). Tryon, who made the first taxonomic revision of Dennstaedtia in a modern format (in Contr. Gray Herb. 187: 23–52. 1960), added further to this confusion. Tryon (l.c.: 40–41) accepted the name Dennstaedtia bipinnata, citing the specimen at MA as the holotype (although not seeing it), citing the specimen at B-W as an isotype (neither seeing it), and citing the specimen at US as a fragment of the isotype (the one specimen he actually saw). Tryon (l.c.: 40–41) also considered Dicksonia adiantoides a synonym of Dennstaedtia bipinnata. Furthermore, Tryon (l.c.: 40–41) circumscribed Dennstaedtia bipinnata as having tripinnate-pinnatifid laminae and adaxial, connecting wings. It appears that the circumscription of Dennstaedtia bipinnata proposed and consolidated by Tryon (l.c.), based on a fragment of the “isotype” at B-W, has been followed by every other pteridologist who treated Dennstaedtia in nearly every Neotropical flora. For example, Proctor for Jamaica (Ferns Jamaica: 168–169. 1985) and Puerto Rico and the Virgin Islands (in Mem. New York Bot. Gard. 53: 100–101. 1989), Tryon & Stolze for Peru (in Fieldiana, Bot. 22: 99. 1989), Cranfill for the U.S.A. (Florida) (in FNA Ed. Comm., Fl. N. Amer. N. Mexico 2: 200. 1993), Moran for Mesoamerica (in Davidse & al., Fl. Mesoamer. 1: 152. 1995), Smith for the Venezuelan Guiana (in Berry & al., Fl. Venez. Guiana 2: 50. 1995), Navarrete & Øllgaard for Ecuador (in Nordic J. Bot. 20: 327–329. 2000), Mickel & Smith for Mexico (Pterid. Mexico: 245. 2004), Murillo-Pullido & al. for Colombia (Pterid. Colombia: 85. 2008), Schwartsburd & al. for Bolivia (in Phytotaxa 332: 253. 2017), Schwartsburd for Brazil (in Flora do Brasil 2020, http://floradobrasil.jbrj.gov.br/reflora/floradobrasil/FB90915), and so on. Most of these authors cited the specimen at MA as the holotype and the specimen at B-W as the isotype (with a fragment at US), but none of them has ever seen the material at MA. In fact, neither Christensen (l.c.) nor Maxon (l.c.) had seen the specimen at MA, as they admitted. The confusion culminates in the description of Mucura. Triana-Moreno & al. (l.c. 2023) cited as type of their new genus “Mucura bipinnata (Cav.) L.A.Triana & Sundue (≡ Dicksonia bipinnata Cav.)”. Although they did not specify any type specimen, it must be that of Cavanilles. Despite the long confusion of the Ventenat specimens at MA and B-W as being duplicates (from Willdenow, l.c. to Triana-Moreno, l.c. 2023), there is absolutely no evidence for the B-W specimen to be original material of Dicksonia bipinnata. Cavanilles had no apparent contact with this specimen; it disagrees with the protologue (it is tripinnate-pinnatifid!); and it is probably from a different gathering than that of the MA specimen. Therefore, the typification cited by Tryon (l.c.), “holotype: MA, isotype: B-W” (or, lectotype and isolectotype in current understanding – Art. 9.10) is flawed. As far as we know, Dicksonia bipinnata has only one type specimen: the holotype at MA. On the other hand, since the conception of Dicksonia bipinnata (≡ Dennstaedtia bipinnata, Mucura bipinnata) has been built on this false “isotype” at B-W, and has been largely adopted over the last 85+ years, culminating in the recent description of a new genus (Mucura), we understand that a strict application of the nomenclatural rules of the ICN will be disadvantageous. Thus, in order to maintain nomenclatural stability for the Neotropical species commonly known as Dennstaedtia bipinnata (now Mucura bipinnata) (Art. 14), and to prevent loss of the newly described Mucura, we propose the conservation of Dicksonia bipinnata with a conserved type: the Ventenat specimen at B-W (with a duplicate/fragment at US). If approved, the Ventenat specimen at MA would no longer be the type. NTLP, https://orcid.org/0000-0002-3145-8183 PBS, https://orcid.org/0000-0002-5305-9300 We thank John McNeill and John Wiersema for their suggestions, editing, and handling of our manuscript. NTLP thanks FAPEMIG (for the Doctoral grant) and IAPT (research grant 2020). PBS thanks FAPEMIG (for the research grant APQ-00935-22).