Drosera biflora Willdenow

Gonella, Paulo Minatel, Sano, Paulo Takeo, Rivadavia, Fernando & Fleischmann, Andreas, 2022, A synopsis of the genus Drosera (Droseraceae) in Brazil, Phytotaxa 553 (1), pp. 1-76 : 21-23

3. Drosera biflora Willdenow in Roemer & Schultes (1820: 763). Figures 8b, 9f, g

Lectotype: — VENEZUELA. Amazonas: in arenosis humidis Rio Atabapo (Orinoco), Humboldt & Bonpland 1100 [lectotype B-W06247- 010!, first-step lectotype designated by Duno de Stefano (1995), second-step lectotype designated here; isolectotypes B-100272056!, P-00679674 !].

≡ Drosera pusilla Kunth in Humboldt, Bonpland & Kunth (1823: 390), nom. illeg., nom. superfl.

Perennial, acaulescent or forming short, etiolated stems in submerged habitats. Leaves semi-erect, with geniculateinvolute vernation, distinctly petiolate, spatulate, adaxial petiole surface glabrous, abaxial petiole surface sparsely eglandular-pilose, petiole about as long as the lamina, lamina obovate; stipules rectangular in outline, divided into several laciniae from near the base. Scape erect at the base, glabrous, 15–30 mm long, 1–3(–7)-flowered; pedicels usually exceeding half length of the sepals, 1.7–3(–4) mm long; sepals oblong-elliptical (narrowly obovate), united only at the very base, glabrous or sparsely glandular-pilose mostly at the base, apex rounded (obtuse), reflexed in fruit, 3–5 × 1.7–2.0 mm; petals white; gynoecium 3-carpelate, styles bifurcated at the base. Seeds ovoid, testa foveate.

Illustrations: — Kunth (1823: Tab. 490, fig. 1—habit and details, as “ Drosera pusilla ”).

Distribution: — Venezuela, Colombia ( López 2007) and Brazil (North: AM; Fig. 8b).

Habitat: —Wet areas of savanna vegetation, in wet white sandy soils (campinarana). Data about elevational range is scarce, but based on the known localities it is restricted to low elevation sites around 40 to 90 m a.s.l.

Phenology:— Collected with flowers in March, May, June, September, October, November and December, suggesting it may be found in flower year-round, unless submerged during the rainy season.

Conservation status: —Data Deficient (DD). Drosera biflora grows in remote and sub-sampled regions, presenting a large area of suitable habitats where it may occur. Based on the scarce available data on its distribution in Brazil, Colombia and Venezuela (AOO= 16 km 2, EOO= 19,535 km 2), it is not possible to properly evaluate its conservation status. Drosera biflora is not known to occur inside any protected area.

Notes: — Drosera biflora is vegetatively most similar to D. capillaris Poiret (1804: 299) and D. esmeraldae ( Steyermark 1952: 244) Maguire & Wurdack (1957: 335) , but presents more uprightly held leaves (Fig. 9f, g; vs. decumbent in D. capillaris and D. esmeraldae ) and shorter inflorescences ( Table 3 ). From D. capillaris , D. biflora is distinguished by pedicels that are longer than half the sepal length (vs. shorter), longer oblong-elliptical to narrowly obovate sepals that are united only at the very base (vs. shorter, ovate, united up to half length), and seeds with foveate testa (vs. papillose). With D. esmeraldae , D. biflora shares the similarly long pedicels and a foveolate seed testa (the latter however has more foveate seed ornamentation), but is easily distinguished by shorter, usually few-flowered scapes (vs. longer, multiple-flowered scapes), and longer sepals with rounded apex (vs. shorter sepals, with acute to cuspidate apex).

We here adopt a stricter concept of D. biflora , different from the broad circumscription presented by Correa & Silva (2005) , which included D. esmeraldae under its synonymy. Drosera biflora , D. capillaris , and D. esmeraldae present a very similar vegetative morphology, yet do not seem to be closely related, as suggested by the divergent seed morphologies. These three taxa can be easily distinguished by reproductive characters such as pedicel length, sepal shape and indumentum, as well as seed ornamentation ( Table 3 ).

The morphologically closest taxon to D. biflora seems to be D. amazonica : both species share a similar habit and ecology, foveate seeds and sepals which are horizontally patent to reflexed in fruit. Drosera biflora differs from D. amazonica by its well-developed peduncle (vs. flowers subsessile), a glabrous or glandular-pilose scape and calyx (vs. scape and calyx eglandular-pilose), and the glabrous to sparsely glandular, oblong-elliptical sepals with obtuse apex (vs. sepals eglandular pilose, very narrowly oblanceolate to subulate, with acute apex).

Sepals patent to reflexed in fruit constitute a diagnostic character never cited before for D. biflora (a character it shares with D. amazonica ). It was observed in all fruiting specimens of D. biflora examined and is a stable character for distinguishing it from similar species.

Rivadavia et al. (2009) treated D. biflora as synonymous with D. capillaris , however, based on the small, fewflowered variant of D. capillaris , which is common in Brazilian lowland wet savannas of Roraima state.

Roemer & Schultes (1820) named D. biflora using Willdenow’s handwritten species name and diagnosis on a herbarium specimen (nomen in schedis), also referring to that (“ Reliq. Willd. MS. ”), rendering Willdenow as the sole author of the name [ICN Art. 46.3 (viz. Ex. 15); Turland et al. 2018]. This is also the case with D. tenella Willdenow in Roemer & Schultes (1820: 763), which is here considered a synonym of D. capillaris . For these two names, two specimens of the type are found at B (one at the general collection B and another at Willdenow’s collection B-W) and a duplicate is also found at P, making necessary the designation of a lectotype. Duno de Stefano (1995) cites a holotype at B, constituting an inadvertent lectotypification and requiring a second-step lectotypification (ICN Arts. 7.11, 9.10, and 9.17; Turland et al. 2018). In both cases, we designate the specimen from Willdenow’s collection (B-W) as the lectotype because it contains the species name and diagnosis in Willdenow’s handwriting.

Kunth (1823) described D. pusilla based on a different duplicate of the same type specimen of D. biflora , and further mentions the latter name as its synonym, rendering D. pusilla the status of an illegitimate and superfluous name (ICN Art. 52.2; Turland et al. 2018).

Specimens examined: — BRAZIL. AMAZONAS: Barcelos , rio Marari , afluente do rio Aracá, acima da comunidade Bacuquara, 35 m, 13 August 2011, Lima et al. 7237 ( RB) . Igarapé Tibuiari , afluente do rio Uaupés perto do local chamado Monte Cristo e próximo da confluência com o rio Negro , 22 November 1987, Kawasaki 246 ( INPA, NY). Prope San Carlos, ad Rio Negro , April/ May 1853, Spruce 2997 (K, W). Rio Içana , Tunuí, 24 October 1947, Pires 713 ( IAN, INPA, NY). São Gabriel da Cachoeira, Rio Curicurierí (Curicuriari), base of Serra Cujubí, 22 January 1948, Schultes & López 9637A ( GH) .

Material outside Brazil examined: —VENEZUELA. Amazonas: Rio Negro, at base of Cerro Cucuy, 02 March 1944, Baldwin Jr. 3205 (F, K, NY, P, US).