Tradescantia multibracteata M. Ferrarese, Büneker & Canto-Dorow, 2017

Büneker, Henrique Mallmann, Ferrarese, Marina Deon & Canto-Dorow, Thais Scotti Do, 2017, Tradescantia multibracteata, a new species of Commelinaceae from Rio Grande do Sul, Brazil, Phytotaxa 308 (1), pp. 89-96 : 90-95

publication ID

https://doi.org/ 10.11646/phytotaxa.308.1.7

persistent identifier

https://treatment.plazi.org/id/03F687F9-FFF5-5451-FF10-FD96A605FCDB

treatment provided by

Felipe

scientific name

Tradescantia multibracteata M. Ferrarese, Büneker & Canto-Dorow
status

sp. nov.

Tradescantia multibracteata M. Ferrarese, Büneker & Canto-Dorow View in CoL , sp. nov. ( Figs. 1A–B View FIGURE 1 , 2A–F View FIGURE 2 ).

Species morphologice proxima Tradescantia crassula , differt aspectu longiori florenti, laminis foliaribus canalatis, generaliter longioribus, lineari-lanceolatis, inflorescentia typus cincinno-triplus et pedicellis et sepalis glabris.

Type: — BRAZIL. Rio Grande do Sul: Júlio de Castilhos, saxícola, às margens do rio Toropi, próximo à antiga Usina de Quebra Dentes , fl., 29°23’19.78”S 54°01’11.26”W, 26 September 2014, H. M. Büneker 355, R. C. Pontes & L. Witeck (holotype HDCF!; isotype SMDB!) GoogleMaps .

Herbs saxicolous, rosette ca. 25 cm diam. when sterile, 30–50 cm long when fertile, decumbent. Roots robust, little branched. Underground stem rhizomatous, robust, whitish. Above ground stem erect, cylindrical, succulent, green, glabrous, with internodes spaced by 5.5–7.0 cm when fertile. Leaves 7–11, laxly and spirally-alternate arranged, green; sheaths 1.7–3.8 × 1.5–4.7 cm, white-green, ciliated, flexible hyaline trichomes ca. 3 mm long; blades 6.5–37.5 × 1.8–2.5 cm, linear-lanceolate, channelled, succulent, green, slightly discolored, adaxially and abaxially glabrous, the last leaf with a slightly lighter color than the adaxial, apex acute to attenuate, margin with sparse hyaline trichomes ca. 1 mm long; midvein and secondary veins adaxially inconspicuous, abaxially conspicuous at the base. Synflorescence terminal, composed of axillary and terminal inflorescences ( Fig. 4 View FIGURE 4 ); inflorescence structured with main florescence and/ or main florescences at paraclade of 1st order, (1–)2–3(–5) inflorescences per leaf axil; main florescences consisting of a pedunculate triple-cincinni fused at the base (three cincinnus sub-opposite to pseudo-verticillate disposed); peduncle 1.6–6.0 cm long, complanate, green, glabrous; supernumerary bracts usually present, 0–2 per triple-cincinni, smaller than those of the cincinni bracts, but with similar shape and color; cincinni bracts 10–26 × 3–9 mm, unequal among themselves, lanceolate to ovate, green, cymbiform, slightly keeled, uneven, slightly inflated, glabrous, base truncated, not saccate, apex acute to acuminate; triple-cincinni 7–24-flowered; bracteoles ca. 1.6 mm long, ovate, hyaline, glabrous, base truncate, apex acuminate, margin erose. Flowers bisexual, actinomorphic, pedicellate; pedicel 1.0– 8.3 mm long, glabrous, green; sepals 5–7.0 × 2.8–3.0 mm, free, oval-lanceolate, green, glabrous or rarely with up to three deciduous trichomes at apex, keeled, slightly cucullate, base truncate, apex acute-attenuated, margin hyaline; petals 4.0–8.0 × 2.8–5.0 mm, oval, white (white-pinkish according Aona & Machado 958), glabrous, apex acute-attenuated, slightly cucullate; stamens subequal; filaments 2–3 mm long, subcylindrical, white, provided with moniliform hyaline trichomes in the basal portion; anthers yellow, reniform, divergent, rimose; connective 0.9–1.0 × ca. 0.4 mm, rhomboid, yellow; pistil 5.7–7.7 mm long; ovary 1.4–1.7 × ca. 1.3 mm, ovoid-globular, glabrous, white, 3-loculate, locule 2- ovulate; style 4.3–6.0 × 0.3–0.4 mm, sub-cylindrical, glabrous, white; stigma punctate. Capsule 2.9–3.7 × 1.9–2.7 mm, elliptic, light brown to brown at maturity, glabrous. Seeds 1.1–2.3 × 1.3–1.4 mm, 1–2 per locule, ellipsoid.

Specimens examined (paratypes): — BRAZIL. Rio Grande do Sul: São Martinho da Serra, Salto do Guassupi , fl., 29°25’21.84”S 54°00’48.00”W, 4 October 2015, H. M. Büneker 492, T. S. Canto-Dorow & M. D. Ferrarese ( HDCF, SMDB) GoogleMaps ; Jaguari , fr., 27 December 1985, J. N. C. Marchiori 149 ( HDCF, Fig. 5B View FIGURE 5 ) ; ca. 12.5 km ao norte de Jaguari na BR-287 em direção a Santiago , fr., December 2005, L. Y. S. Aona & M. C. Machado 958 ( UEC, photo!) .

Etymology: —The epithet refers to the remarkable high number of bracts in the inflorescence. From Latin multo = many, bractea = bract.

Distribution and habitat: — Tradescantia multibracteata is a saxicolous plant found on rocky cliffs along the Toropi river and its tributary, the Guassupi river (municipalities of Júlio de Castilhos, Quevedos and São Martinho da Serra) and also on rocky outcrops in the municipality of Jaguari, in the central region of the state of Rio Grande do Sul, Brazil ( Fig. 3 View FIGURE 3 ).

Conservation Status: — Tradescantia multibracteata is endemic in the central region of Rio Grande do Sul, with small populations on cliffs and rocky outcrops above the banks of the Toropi river and its tributaries, an area extremely threatened by the planned impoundment area from small hydroelectric plants in construction ( Marchiori et al. 2014). It can also be found in a similar environment in the region of Jaguari. Overall, the area of occupancy (AOO) of this species is 1559 km 2 and only five populations are known. Therefore, we ranked this new species as Vulnerable (VU) according to the B2ab (ii, iii, iv) criteria by the IUCN (2016).

Taxonomic observations and discussion: —Based on floral morphology T. multibracteata fits relatively well within the current circumscription of Tradescantia sect. Austrotradescantia (sensu Pellegrini et al. 2017), the florescence structure of T. multibracteata however is unique within this section, and it is rather difficult to analyze. Based on Panigo et al. (2011), the synflorescence ( Fig. 4 View FIGURE 4 , Sy) of T. multibracteata is always terminal, composed of axillary ( Fig. 4 View FIGURE 4 , In - a) and terminal ( Fig. 4 View FIGURE 4 , In - t) inflorescences. The inflorescence is structurated with main florescence (main florescence=inflorescence ( Fig. 4 View FIGURE 4 , MF=In), characteristic of most species of the genus Tradescantia ) and/or main florescences at paraclade of 1st order ( Fig. 4 View FIGURE 4 , In - a). They are usually two or three inflorescences per leaf axis. The main florescence consisting of a pedunculate ( Fig. 4 View FIGURE 4 , P) triple-cincinni ( Fig. 4 View FIGURE 4 , MF) fused at the base (three cincinnus ( Figs. 4 View FIGURE 4 , pF; 2F) subopposite to pseudoverticillate disposed). Being the main florescences probably unheard of for Tradescantia , conventionally it is a double-cincini and is triple-cincinni in T. multibracteata . Aside from that, the presence of supernumerary bracts (0–2 per triple-cincinni ( Fig. 2 View FIGURE 2 Fbs)) is unique within T. sect. Austrotradescantia , but well-known in species from T. sect. Cymbispatha (Pellegrini 2015; Pellegrini et al. 2016, 2017) and T. sect. Mandonia ( Grant 2000; Pellegrini 2015; Pellegrini et al. 2017). Despite this difference, we believe that T. multibracteata is closely related to the remaining species in the T. crassula group (sensu Pellegrini 2015, 2016), due to its erect habit, succulent leaves, generally with inconspicuous secondary veins, cincinni bracts non-saccate at base, pistil longer than the stamens and preferentially inhabiting open areas and rocky outcrops. In particular, T. multibracteata is morphologically similar to T. crassula sensu Pellegrini (2015) . They have a similar vegetative habit (only spirally-alternate vs. distichously or spirally-alternate), but the new species is larger than T. crassula when in bloom (30–50 cm vs. 10–46 cm), the leaf blades are channelled (vs. subflat) and generally longer (6.5–37.5 cm vs. 4.4–12.8 cm) and linear-lanceolate (vs. elliptic to broadly elliptic, ovate to broadly ovate to obovate or rarely lanceolate). Also they differ in features of their main florescences; in T. multibracteata they consist of a triple-cincinni (vs. double-cincinni) and glabrous pedicels and sepals (vs. hairy).

Recently, one of the specimens we designate here as paratype of T. multibracteata , i.e. J.N.C. Marchiori 149 at HDCF ( Fig. 5B View FIGURE 5 ), has been placed under T. valida by Pellegrini et al. (2017). These authors reported the rediscovery of T. valida Brückner (1932: 510) and used the specimens J.N.C. Marchiori 149 and L.Y.S. Aona & M.C. Machado 958 (specimen we have also studied and classified as T. multibracteata ) as supporting evidence for the rediscovery of T. valida . Until now T. valida is only known from its holotype deposited at B ( Fig. 5A View FIGURE 5 ). Brückner (1932), when describing the species, indicated that the plant was sent to him by Professor Dr. Mevius and that it was cultivated at the Botanical Garden of the University of Münster ( Germany). A note stating that the plant came to the garden along with epiphytic orchids purchased from Brazil was sent to Brückner along with the specimen ( Fig. 5A View FIGURE 5 , note in German, lower left hand side corner). Despite this evidence, Pellegrini (2015) indicted that T. valida is endemic to Bolivia and morphologically close to T. seubertiana Pellegrini (2016: 80) . Specimens from Ballivia, Beni, Bolivia collected by J.C. Solomon 13916 (LPB, MO, NY, US n.v.) were used to support this distribution. It is evident when comparing the type specimen of T. valida with J.N.C. Marchiori 149 (see Fig. 5 View FIGURE 5 ) that these plants belong to two morphologically very distinct species. Some of the most obvious differences between T. multibracteata in relation to T. valida (sensu Brückner 1932) are the distance between internodes (5.5–7 cm vs. 3–4 cm), the shape, color and length of the leaf blades (linear-lanceolate, green, up to 37.5 cm long vs. oblong, vinaceous, up to 9 cm long), the type of main florescences (triple-cincinni vs. double-cincinni), the color of the pedicel and indumenta (totally green and glabrous vs. vinaceous and sparsely hairy), the sepals indumenta (glabrous or rarely with up to three deciduous trichomes at apex vs. base sparsely hairy and fasciculated-hirsute apex) and the petal color (usually totally white vs. pink). After studying the description of T. valida published by Pellegrini et al. (2017) it appears these contrasting characteristics were combined to produce their re-description of T. valida .

We believe the taxonomic status of T. valida is still far from being resolved and the description prepared by Pellegrini et al. (2017) should be revised and based on detailed examination of voucher specimens besides J.N.C. Marchiori 149 and L.Y.S. Aona & M.C. Machado 958 as these two belong to T. multibracteata . Furthermore, we noticed a mismatch between the phenological stage reported in Pellegrini et al. (2017) for J.N.C. Marchiori 149 and the actual specimen. The specimen J.N.C. Marchiori 149 consists of two fertile exsiccates ( Fig. 5B View FIGURE 5 ), but unlike what is indicated by Pellegrini et al. (2017), they do not have any flowers but only capsules. These are in advanced stage of maturation and most of them are already opened. In fact, it is very unlikely Pellegrini et al. (2017) have actually seen this specimen. First, to the best of our knowledge, none of the authors in Pellegrini et al. (2017) have visited the Herbário do Departamento de Ciências Florestais (HDCF); where the first author of this study works. Second, there are no records (over the last three years) stating the Commelinaceae collection at the HDCF has been visited by a researcher. And third, the specimen in question has not been loaned or digitized nor has an image been made available on-line (personal communication from L. Denardi, curator of the HDCF herbarium). And lastly, Pellegrini et al. (2017) made no comment whatsoever in the discussion of T. valida (p.15) about the label placed on the specimen sheet indicating its paratypus status ( Fig. 5B View FIGURE 5 ).

Based on the discussion above we believe T. valida has not yet been re-discovered and due to the lack of other material to link the holotype of T. valida with we deem it is pertinent to consider this name as nomen dubium. However, and similar to Pellegrini (2015), we believe that T. valida (sensu Brückner 1932) is morphologically similar to T. seubertiana and T. umbraculifera Handel-Mazzetti (1908: 204) . They share similar leaf blade shape (oblong to lanceolate) and cincinni bracts shape.

H

University of Helsinki

M

Botanische Staatssammlung München

R

Departamento de Geologia, Universidad de Chile

C

University of Copenhagen

L

Nationaal Herbarium Nederland, Leiden University branch

SMDB

Universidade Federal de Santa Maria

T

Tavera, Department of Geology and Geophysics

S

Department of Botany, Swedish Museum of Natural History

J

University of the Witwatersrand

N

Nanjing University

Y

Yale University

UEC

Universidade Estadual de Campinas

Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF