Commelina obliqua Vahl (1805: 172)

Hassemer, Gustavo, Ferreira, João Paulo Ramos, Funez, Luís Adriano & Aona, Lidyanne Yuriko Saleme, 2016, Identity and typification of Commelina vilavelhensis (Commelinaceae), and typification of C. robusta and C. scabrata, Phytotaxa 260 (2), pp. 144-156 : 145-152

Commelina obliqua Vahl (1805: 172)

≡ Commelina erecta var. obliqua ( Vahl 1805: 172) Clarke (1881: 181)

Lectotype (designated by Hunt 1994, referred as “type”):— FRANCE. Ex horto Celsii, Ventenat s.n. (C 10009563!). Figures 1 and 2 .

= Commelina monticola Seubert (1855: 264)

Holotype:— BRAZIL. MιΝΑඌ G*ú*©Αιඌ: 1840, P. Claussen 420 (BR 21451010, photograph!). Figure 3 .

= Commelina robusta Kunth (1843: 52)

Lectotype (hereby designated):— BRAZIL. F. Sello s.n. (B 100296353, photograph!). Figure 4 .

= Commelina scabrata Seubert (1855: 266)

≡ Phaeosphaerion persicariifolium var. scabratum ( Seubert 1855: 266) Clarke (1881: 137)

Lectotype (hereby designated):— BRAZIL. BΑ*üιΑ: C.F.P. von Martius s.n. (M 274913, photograph!). Figure 5 .

= Commelina vilavelhensis da Maia et al. (2012: 294–295) , syn. nov.

Lectotype (hereby designated):— Figure 1 in da Maia et al. (2012).

Distribution: —Neotropical, occurring from southern Mexico to Uruguay and northern Argentina, including Trinidad and possibly some other Caribbean islands ( Hunt 1994). We could not find any record of this species from Chile.

Habitat: —Forested and open areas, and also in ruderal (human-impacted) environments.

Conservation status: —Least Concern (LC). This species has a wide distribution and is common throughout its range.

Observations: —Because the protologue of C. obliqua does not mention the collection or the herbarium where the type was deposited ( Vahl 1805), a holotype was not designated for this name. Considering that Martin Henrichsen Vahl (1749–1804) worked in Copenhagen, and the label of one sheet at C (C 10009563; Figure 1 ) matches the information provided by Vahl (1805) (“Ex horto Celsii”) and is identified as C. obliqua ( Figure 2 ), we believe that this sheet must be original material of C. obliqua . Therefore, the first work that specified the collection where the type of C. obliqua was deposited, which we believe to be Hunt (1994), effected the lectotypification of this name.

Nevertheless, we must highlight that Hunt (1994) presents the type (“Tipo”) of this species as “Cult. en Uppsala” (cultivated in Uppsala). This information has been followed in important databases such as Tropicos (http://tropicos. org). However, a careful reading of the label of the type (“Ex horto Celsii”) leaves it clear that this plant originated from Jacques Philippe Martin Cels’ (1740–1806) garden, which was located in Paris, France, not in Uppsala, Sweden. We are here accepting the lectotypification of the name C. obliqua by Hunt (1994) despite this important error.

The name C. robusta , a synonym of C. obliqua , needed to be lectotypified because its protologue ( Kunth 1843) does not indicate a collection where any type-materials have been deposited. However, as most of Friedrich Sello’s (1789–1831) collections are at B herbarium (or were, before it was bombed in 1943), we were able to find the typespecimen of this name. The name C. scabrata needed to be lectotypified because, despite its protologue ( Seubert 1855: 266) very clearly specifying M herbarium as the deposition place of the type, we found that there were two sheets at M (M 274913 and M 274914) corresponding to plants collected by Carl Friedrich Philipp von Martius (1794–1868) that are original material of this name. Therefore, both these sheets were to be considered syntypes of C. scabrata . We designate here the best and more complete of these sheets (M 274913) as the lectotype of C. scabrata .

During our revision of the synonyms of C. obliqua we noticed that some important databases such as Tropicos and Flora do Brasil 2020 (http://floradobrasil.jbrj.gov.br) erroneously present the name “ Heterocarpus obliquus (Vahl) Hassk. ”, supposedly a homotypic synonym of C. obliqua Vahl (1805: 172) . However, a careful reading of the protologue of H. obliquus leaves it clear that this name is a new combination for the basionym C. obliqua Buchanan-Hamilton ex Don (1825: 45) , which is a later homonym of C. obliqua Vahl (1805: 172) and a synonym of C. paludosa von Blume (1827: 2) . Therefore, H. obliquus (Buchanan-Hamilton ex Don 1825: 45) Hasskarl (1870: 7–9) is a new combination for an illegitimate later homonym.

The situation of the name C. vilavelhensis is far more complicated, however. This name was validly published in 2012, being authored by Daniela Corrêa da Maia, Armando Carlos Cervi and Rosângela Capuano Tardivo. Its protologue cites five collections: the type (D.C. da Maia 41; the holotype at UPCB, and two isotypes, one at HUPG and the other at MBM) and five paratypes (D.C. da Maia 28, 29, 30, 31, 32; all supposedly deposited atUPCB). This article includes illustration of the new species ( Figure 1 in da Maia et al. 2012), but no photographs. It is important to highlight that Figure 1 in da Maia et al. (2012) makes reference to the sheet D.C. da Maia 42 (UPCB), which is not even included as examined material.

The article in which the name C. vilavelhensis was published ( da Maia et al. 2012) is clearly part of the results of Daniela Corrêa da Maia’s masters dissertation ( da Maia 2006, which was never published), where C. vilavelhensis is presented still unnamed, as “ Commelina sp. ”. The description and distribution of the new species in both works are the same; however, there is an important difference in the examined material of this species: da Maia (2006) cites the following: D.C. da Maia 28, 29, 31 (presented twice, for plants collected in different Brazilian states!), 32, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 53 —all supposedly deposited to UPCB. The same illustration as Figure 1 in da Maia et al. (2012) is presented, also referring to the sheet D.C. da Maia 42 (UPCB).

However, during our visits to MBM and UPCB herbaria (in 2013, 2014, 2015 and 2016) we have learned that not a single one of the specimens cited in da Maia et al. (2012), or in da Maia (2006) for “ Commelina sp. ”, was ever deposited to any of these herbaria. We also contacted the curator of HUPG herbarium (Rosângela Capuano Tardivo, who is the third author of C. vilavelhensis ), who confirmed that these specimens were not deposited to this herbarium as well. When contacted, Daniela Corrêa da Maia, the first author of C. vilavelhensis , informed that she has not seen the plants since 2006, when she finished her Master course at Universidade Federal do Paraná. Unfortunately, the second author of C. vilavelhensis and former curator of UPCB herbarium, Armando Carlos Cervi (1944–2014), is no longer living. Because all original specimens of this name were found to be missing, we have no option but to designate the illustration in the protologue ( Figure 1 in da Maia et al. 2012) as the lectotype of this name (§9.12 of the Botanical Code).

With a careful reading of the description of C. vilavelhensis , in addition to the examination of its illustrations, we concluded that this name is a synonym of C. obliqua ( Table 1). This is supported by the characters presented by da Maia et al. (2012) in their description of C. vilavelhensis such as: “lamina lanceolate; spathes sessile, ovate, isolate or in groups of 2–5, apex acute to obtuse, lateral margin fused only at the base of the peduncle; petals three, two well developed, dorsal, reniform unguiculate, apex rounded, light blue, and the other one much smaller, inconspicuous, whitish; stamens three; anther elliptic, rimose, basifixed, yellow; staminodes three, antheroids yellow” (translated from Portuguese); and also the asymmetric base of the leaves, omitted in the description but evident in the illustrations ( Figure 1 in da Maia et al. 2012).

When presenting their diagnosis of C. vilavelhensis, da Maia et al. (2012) compare their new species with two other species, C. erecta and C. benghalensis . However, they compare the three species using morphological characters highly variable within species of Commelina , completely unsuitable for the intended comparisons. In their identification key to separate C. vilavelhensis from C. erecta and C. benghalensis, da Maia et al. (2012) only use floral characters, thus completely omitting other taxonomically important characters such as the shape, insertion pattern and indumentum of leaf sheath and lamina. One of the most useful characters to separate C. erecta from most other species of Commelina in South America is the presence of an auricle at the junction of the lamina with the sheath, which e.g.

is absent in C. vilavelhensis and C. benghalensis . Based on all information available in da Maia et al. (2012), i.e. the description, the illustration and the identification key, we believe that C. vilavelhensis matches the morphology of C.

obliqua, and not C. erecta or C. benghalensis (see Figure 6 ). This last species ( C. benghalensis ) is morphologically so different to C. erecta and C. obliqua , by its having ovate leaves and subterraneous, cleistogamic flowers (see identification key below), that we believe that there is no risk of it being confused with plants matching the protologue of C. vilavelhensis .

Furthermore, the identification key provided along with the description completely omits C. obliqua (it includes only C. benghalensis , C. erecta , and C. vilavelhensis ), a species then already well known to have a wide distribution in Paraná and Santa Catarina states, southern Brazil. All this evidence, in addition the evident poor care dispensed by the authors in their publication of da Maia et al. (2012) supports our hypothesis that these authors confused a few collections of C. obliqua for a narrow endemic new species.

Material examined: — ARGENTINA. MιඌιοΝ*úඌ: Froulena, San Antonio, 500 m, 18 April 1949, E. Petersen & J.P. Hjerting s.n. (C). BRAZIL. BΑ*üιΑ: Abaíra: Caminho de Levi-Serrinha, 13°19’S, 41°51’W, 13 December 1993, W.

Ganev 2604 (CEPEC, HUEFS); Água Quente: Vertente Norte, vale a NW do Pico das Almas , margem do riacho, 13°30’00”S, 41°58’59”W, 30 November 1988, R.M. Harley 26507 (CEPEC) ; Carinhanha: Médio do São Francisco, a 1 km do Rio Carinhanha , 14°18’41”S, 43°48’32”W, 28 December 2007, M.L.S. Guedes 14061 (HUEFS) ; Contendas do Sincorá: FLONA, 50 m da sede, via Rio Combuca , 13º55’18”S, 41º07’08”W, 366 m, 6 February 2015, L.Y.S Aona et al. 3841 (HURB) ; Ibicoara: Chapada Diamantina, caminho para a Cachoeira do Buracão , 13º18’56”S, 41º09’38”W, 957 m, 13 October 2007, M.L. Guedes et al. 13822 (ALCB) ; Jeremoabo: Margem do Rio Vazabarreis , 10°04’30”S, 38°28’51”W, 4 June 2008, K. Mendes 183 (HUEFS) ; Rio de Contas: estrada para Campo do Queiroz, próximo ao Rio Brumadinho , 13º31’22.8”S, 41º53’47.8”W, 1195 m, 20 May 1999, R. Romero et al. 5862 (UEC) ; Una: Faz. Piedade, Rod. S. José / Una , a 9 km do entroncamento / BR-101, 15°17’36”S 39°04’31”W, E.B. Santos 184 (CEPEC). C*úΑ*© Crateús : Serra das Almas, 22 February 2000, L.W. Lima-Verde 1151 (EAC) ; Crato: Flona do Araripe , 27 May 1999, L.W. Lima-Verde 1478 (EAC) ; Quixeré: Chapada do Apodi, Manga Grande-Faz. Mato Alto , Grupo Apodi , 14 May 1997, L.M.R. Melo 95 (EAC). Dιඌτ*©ιτο F*úƊ*ú*©ΑĿ: Brasília: Granja do ipê, córregooCapãPreto, 20 May 502, V.F.Paiva 502 (HEPH, HURB) ; Brasília: Estação Ecológica do JBB, Córrego Cabeça de Veado , 15º52’00”S 47º51’00”W, 1025 m, 22 October 2003, L.C. Milhomens et al. 306 (HEPH, HURB). Eඌ*ssỉ*©ιτο SΑΝτο: Linhares: Reserva Natural Vale, 15 April 1999, D.A. Folli 3395 (CVRD, HURB). MΑτο G*©οඌඌο: Barra dos Bugres: Posto Fraternidade, 6 June 1998, R. Loureiro s.n. (UFMT 18217) ; Cuiabá: estrada Cuiabá-Chapada dos Guimarães, km 20, lado direito, 21 April 1978, M. Macedo & A. Assumpção 831 (UFMT). MΑτο G*©οඌඌο Ɗο SUĿ: Piraputanga / Aquidauana : Pesqueiroda Serra , 20º27’52.5”S, 55º29’15.1”W, P. Castilho 18 (CGMS). MιΝΑඌ G*ú*©Αιඌ: Caldas: 18 March 1866, A.F. Regnell III-1281 (C) ; Carangola: Rio Carangola , 400 m, 25 October 1988, L.S. Leoni 484 (HURB) ; Carmo do Rio Claro: Poço das Virgens , 20º54’42.9”S, 45º59’09.9”W, 781 m, 12 October 2007, A.R. Silva & E.L. Siqueira 480 (FLOR 46994) ; Faria Lemos: Cafarnanum “ECCB”, 940 m, 19 March 2006, R.L. Sousa 64 (HURB) ; Fervedouro, Córrego Água Limpa , 26 April 2006, L.S. Leoni et al. 6466 (HURB) ; Itamarandiba: Parque Estadual Serra Negra , 26 January 2012, D.J.P. Gonçalves et al. 452 (UEC) ; Jaboticatubas: Parque Nacional Serra do Cipó, Alto do Palácio , 16 June 2009, M. Carvalho- Silva et al. 609 (HEPH, HURB) ; Juiz de Fora : L. Krieger s.n. (HURB 12189) ; Lagoa Santa : s.d., J.E.B. Warming s.n .

(C); 30 September 1863, J.E.B. Warming 1047 (C); 15 November 1864, J.E.B. Warming 1067 (C); Patrocínio: Fazendas DATERRA, 12 June 1998, F.T. Farah & C.A. Freitas 185 (ESA); São Roque de Minas: Parque Nacional da Serra da Canastra, estrada São Roque de Minas-Sacramento, próximo à guarita de Sacramento, 21 March 1998, P.T. Sano et al. 971 (HURB, SPF); São Tomé das Letras: APA São Thomé, Vale do Canta Galo, próximo a cachoeira das panelas, 1050 m, 2 April 2015, P.S.P. Sampaio & A.F.P. Sampaio 1025 (HUSC); Viçosa: sítio, 23 May 2000, L.H.L. Freitas s.n.

(HURB 8116, VIC 24801). PΑ*©ΑΝǍ: Curitiba : Parque Iguaçu, 4 February 1987, R. Kumrow & J. Cordeiro 2875 (FLOR 27304) ; Londrina : Mata do Godoy, 30 May 1985, L.A.C. Rodas et al. s.n. (FUEL 825) ; Matinhos : Caioba, 17 June 1863, Unknown Collector 547 (C) ; Sapopema : Faz. Lageado Liso , 11 March 1989, M.C.M. Marques et al. s.n. (FUEL 6764, HURB 522). P*ú*©ΝΑ*§*ôU*öο: Recife : Cidade Universitária ,usCampda UFPE, Riacho do Cavouco , próximo ao ITEP, Várzea do Rio Capibaribe, 08º03’06.6”S, 34º57’14.6”W, 12.5 m, 18 September 2006, M.S. Leite 154 (FLOR 40197). Rιο Ɗ*ú JΑΝ*úι*©ο: Itatiaia: Parque Nacional do Itatiaia,Lago Azul, próximo da estrada, 22º15’00”S, 44º34’00”W / 22º28’00”S, 44º45’00”W, 650 m, 15 June 1997, J.M.A. Braga et al. 4152 (HURB, RB) ; Nova Friburgo : Macaé de Cima, 900 m, 19 January 1999, L.O. Anderson et al. 99/12 (UEC) ; Parati : Fazenda São Lourenço, 23º19’42”S, 44º39’25”W, 17 November 1993, E.B. Martins et al. 29410 (HURB, UEC) ; Petrópolis : June 1863, Unknown Collector s.n. (C). SΑΝτΑ CΑτΑ*©ιΝΑ: Benedito Novo: margens do Rio Benedito,6 2º47’05.53”S, 49º25’01.99”W, 15 March 2008, E. Crestani s.n. (FURB 8447) ; Blumenau : Ilhas fluviais do Rio Itajaí-açu , APA ( Área de Proteção Ambiental ) das Ilhas Fluviais, Floresta Ombrófila Densa, 26º52’31.82”S, 49º05’21.99”W, 10 m, 28 April 2014, A.N. da Silva et al. 1373 (FLOR 52463) ; Concórdia : Lajeado dos Pintos, 27º10’47”S, 52º01’11”W, 768 m, 20 January 2009, A.L. Gasper et al. 1193 (FURB, HURB) ; Florianópolis : Ilha de Santa Catarina, Rio Vermelho, 10 m, 16 November 1984, F.A. Silva et al. 211 (FLOR 33548) ; campo em restinga, 10 m, 29 March 1985, M.L.R. Souza & D.B. Falkenberg 718 (FLOR 33549); Passos Maia : Baia, 26º49’12”S, 52º01’01”W, 1075 m, 6 April 2009, S. Dreveck & F.E. Carneiro 817 (FURB) ; Rio do Oeste : Sumidor, 27º10’48”S, 49º51’36”W, 340 m, 15 March 2011, A. Korte 6313 (FURB, HURB) ; Rodeio : Bairro São Pedro Velho, 26º55’29”S, 49º24’19”W, 121 m, 20 November 2012, L.A. Funez 1264 (FURB) ; São Bento do Sul : CEPA, 30 March 2007, A.L. Gasper & L. Sevegnani 95 (FURB) ; São Francisco do Sul : Praia do Ervino, Parque Estadual do Açaraí, 26º23’36”S, 48º36’38”W, 2 m, 22 February 2011, A. Korte 5764 (FURB, HURB) ; Xaxim : Sangra do Tigre , 27º00’00”S, 52º33’36”W, 616 m, 28 March 2011, A. Korte 6441 (FURB). S*ºο PΑUĿο: Águas da Prata: ca. 500 m do Distrito de São Roque da Fartura, brejo na beira da estrada para Vargem Grande, 21º51’00”S, 46º45’02”W, 1170 m, 11 January 1994, V.C. Souza et al. 4994 (HURB, SPF) ; Barra do Chapéu : Propriedade da Dona Pedrina, no remanescente de floresta de araucária, 25º17’26”S 49º0’24”W, 807 m, 24 August 2008, R.T. Polisel et al. 506 (HURB, SPSF) ; Campinas : Sousas, Ribeirão Cachoeira, 22º50’S, 46º55’W, 700 m, 3 August 2006, L.T. Vieira & P.L. Sims 388 (UEC) ; Campos do Jordão : São José dos Alpes, 1700–1900 m, 13 February 1996, L. Amaral Jr. s.n. (HURB 12346) ; Ibiúna : Parque Estadual de Jurupará, Cachoeira do França, 27 April 1993, R. Esteves 66 (HURB, SPSF) ; Itapeva : Estação Experimental, 24º02’46”S, 49º04’19”W, Divisa da Estação Experimental com a ponte FEPASA-FES, 29 September 2010, R. Cielo-Filho et al. 1153 (HURB, SPSF) ; Itirapina : beira do Córrego do Feijão, 22º12’S, 47º45’W, 18 June 1995, M.C. E. do Amaral et al. 95/75 (UEC) ; Jaguariúna : CEMA, Centro Educacional Municipal Ambiental, 22º40’81”S, 46º58’64”W, 17 March 2005, V.L.G. Brito 1 (HURB, UEC) ; Jundiaí : Serra do Japi, sentido bairro Eloy Chaves, próximo à represa do DAE, 10 January 1997, E.R. Pansarin 125 (HURB, UEC) ; Miracatu : Fazenda Irereí , Serra de Paranapiacaba, BR-116 350, 24º03’S, 47º13’W, 20 April 1994, J.R. Pirani & R.F. Garcia 3158 (HURB, SPF) ; Paulo de Faria : Estação Ecológica de Paulo de Faria, 19º55’S 49º31’W, 15 March 1995, V. Stranghetti 475 (UEC) ; Piracicaba : mata estrada de Limeira (DIR), 16 April 1993, K.D. Barreto et al. 296 (ESA) ; Porto Ferreira : Parque Estadual do Porto Ferreira, trilha de uso extensivo, 21º49’S, 47º25’W, 520–600 m, 25 March 2010, J.A. Lombardi & A.P.C. Oliveira 7682 (HRCB, HURB) ; São Lourenço da Terra : Reserva Particular do Patrimônio Natural Paiol Maria , Trilha das Canelas a partir da entrada pela Trilha Ibira-Mirim, 23º49’31.5”S, 46º56’43.8”W, 733 m, 15 April 2010, M.N. Saka & B. Franco 69 (HRCB, HURB) ; São Paulo : campus da Universidade de São Paulo , Instituto de Biociências , campinho com remanescente dos “campos do Butantã”, 22 December 1999, D. Sasaki 21 (HURB, SPF). DENMARK. Copenhagen: plant grown at Copenhagen Botanical Garden from living material collected by Eugenia Bordas on 5.6. 1969 in Cerro Tobatí , Paraguay, 26 November 1987, J.P. Hjerting s.n. (C). FRANCE. Ex horto Celsii, Ventenat s. n. (C 10009563). PARAGUAY. MιඌιοΝ*úඌ: Santiago: Estancia La Soledad, dense wood on boggy ground, 24 April 1961, T.M. Pedersen 5966 (C) .

Figures

FIGURE 1. Image of the lectotype of Commelina obliqua (Ventenat s.n., C 10009563).

FIGURE 2. Image of the label of the lectotype of Commelina obliqua (Ventenat s.n., C 10009563).

FIGURE 3. Image of the holotype of Commelina monticola (P. Claussen 420, BR 21451010).

FIGURE 4. Image of the lectotype (hereby designated) of Commelina robusta (F. Sello s.n., B 100296353).

FIGURE 5. Image of the lectotype (hereby designated) of Commelina scabrata (C.F.P. von Martius s.n., M 274913).

FIGURE 6. Photographs of Commelina obliqua (A and C) and C. erecta (B and D). A. Flower of C. obliqua showing the anther of the central stamen with ellipsoid thecae and the yellow anthers of the lateral stamens. B. Flower of C. erecta showing the anther of the central stamen with falciform thecae and the white-bluish anthers of the lateral stamens. C. Base of a leaf of C. obliqua showing the characteristic brown-reddish trichomes, the non-auriculated apex of the sheath, and the asymmetric base of the lamina. D. Base of a leaf of C. erecta showing its whitish trichomes, the auriculated apex of the sheath, and the symmetric base of the lamina. Photographs by L.A. Funez (A, C and D) and João de Deus Medeiros (B).