Graffenrieda goldenbergii, Lima, Laíce Fernanda Gomes De, Baumgratz, José Fernando Andrade, Lughadha, Eimear Nic & Santos, João Ubiratan Moreira Dos, 2016

Lima, Laíce Fernanda Gomes De, Baumgratz, José Fernando Andrade, Lughadha, Eimear Nic & Santos, João Ubiratan Moreira Dos, 2016, Two new species of Graffenrieda (Melastomataceae, Merianieae) from the Amazon Rainforest, Phytotaxa 267 (1), pp. 77-83 : 80-82

publication ID

https://doi.org/ 10.11646/phytotaxa.267.1.8

persistent identifier

https://treatment.plazi.org/id/03C187F4-FF89-4607-FF16-B645FADBFAAE

treatment provided by

Felipe

scientific name

Graffenrieda goldenbergii
status

sp. nov.

Graffenrieda goldenbergii View in CoL L.F. Lima, Baumgratz, NicLugh. & J.U.Santos sp. nov. ( Figure 2 View FIGURE 2 )

Type: — BRAZIL. Acre. Mâncio Lima, Parque Nacional da Serra do Divisor , Serra do Moa , trilha para a cachoeira do formoso ( Igarapé do Anil ), 13 November 2007 (fl.), R. Goldenberg et al. 962 (holotype RB!; isotype LABEV, UPCB!) .

Graffenrieda goldenbergii resembles G. tristis but differs in the pulverulent indumentum of the rachis, pedicel, hypanthium and calyx, in the abaxial leaf lamina which is lepidote, with trichomes 0.2 mm and caducous, in the calyx with irregularly valvate dehiscence, in the obovate petals with asymmetrically rounded apex and in the costate fruit with hypanthium and calyx membranaceous and outer calyx lobes inconspicuous.

Treelet ca. 6 m tall. Rachis, pedicel, hypanthium and calyx pulverulent with an indumentum of white trichomes which are early caducous; leaf lamina moderately lepidote abaxially, with an indumentum of brownish trichomes, axils of the acrodromous veins barbate, trichomes caducous or persistent. Branches flat or cylindrical. Leaves isophyllous, petiolate; petiole 1–2 cm long, semicylindrical; lamina 5–11 × 3–5.5 cm, elliptic to obovate, apex acute to cuspidate, base acute, margin entire, with 3 basal acrodromous nerves, of which the marginal pair very inconspicuous. Inflorescence a thyrsoid of corymbiform cymes, ca. 15 cm long, terminal, sessile; bracts early caducous, not seen; bracteoles ca. 0.5 mm long, narrowly triangular to linear. Flowers 4-merous; pedicel 1–3 mm long; hypanthium ca. 2.5 × 2 mm, shortly tubular, membranaceous, calyx irregularly valvate in bud, forming 2–3 lobes, tube ca. 0.5 mm long, inner lobes 1.5–2 mm long, membranaceous, triangular to broadly triangular, apex acute or obtuse, outer lobes inconspicuous, reduced to teeth; petals ca. 5 × 4 mm long, white, obovate, apex asymmetrically rounded. Stamens equal in shape and form, yellow, filaments ca. 3 mm long, anthers ca. 3 mm long, slightly falcate, thecae not rugose, with a ventral terminal pore, connective prolonged ca. 0.3 mm below the thecae, spur ca. 0.3 mm long, acute; ovary ca. 2 × 1.5 mm, 3- locular, ovate, apex truncate, costate, free inside the hypanthium, style ca. 8 mm long, stigma punctate. Capsule of the velatidium type, 4.2–5 x 2.5–4 mm, brown, membranaceous, costate, not sulcate; seeds 1–2 × 0.2–0.5 mm, narrowly obpyramidal to cuneately obpyramidal.

Distribution and habitat:— The first record of this species is from Peru, San Martin Department, Province and District of Lamas, to the north of Santo Antonio de Cumbaza. The first collection made in Brazil was in Acre state, near the border with Peru, in the Moa mountain range which is part of the Divisor mountain range, within the Parque Nacional da Serra do Divisor, a federal conservation unit.

Etymology:— The epithet “goldenbergii ” honours Dr. Renato Goldenberg, of the Universidade Federal do Paraná, Brazil, who is a specialist in the Melastomataceae family and collected the type specimen.

Paratype:— PERU. Departament San Martin, Province de Lamas, District Lamas, norte de Santo Antonio, 2–4 Km ao longo do rio Cumbaza, 02 Octuber– 04 November 1937 (fr.), C.M. Belshaw 3521 (K!).

Conservation status:—Data Deficient (DD). There are only two collections of G. goldenbergii , and one of them was recently collected in the Serra do Divisor National Park, in the Amazon Forest of Acre state, Brazil. However, at the moment, we can not assess the extinction risk of this species, because there are scarce data about biological, ecological and species distribution, and it is necessary to undertake fieldwork to increase new information.

Discussion:— Graffenrieda goldenbergii is characterised by the caducous indumentum of the rachis, pedicel, hypanthium and calyx which is pulverulent and white, and that of the abaxial surface of the leaf lamina which is moderately lepidote, with trichomes ca. 0.2 mm diam., also caducous, by the calyx with irregularly valvate dehiscence, by the obovate petals with apices asymmetrically rounded and by the membranaceous, costate fruit which are never sulcate nor mucronate at the apex.

It resembles G. tristis ( Triana 1871: 71) L.O. Williams (1963: 564) in the barbate axils of the acrodromous nerves on the abaxial leaf surface, in the morphology of the leaves and stamens, and in the tetramerous flowers with white petals. However, G. tristis is distinguished by the following combination of characters: indumentum of the branches, both surfaces of the leaf lamina and rachis densely lepidote, with trichomes ca. 0.5 mm diam., early caducous only on the adaxial surface of the lamina but persisting on the median acrodromous vein; indumentum of the pedicel, hypanthium and calyx glandular-granulose and persistent; calyx with calyptrate dehiscence; petals oblong with acute apex; and fruit markedly costate-sulcate, thickly coriaceous and 4–5 mucronate apically.

Among the species with calyx dehiscence irregularly valvate, G. caudata Wurdack (in Maguire et al. 1964: 152) is the most similar. However, it differs in having leaf lamina caudate at apex with 5 suprabasal acrodromous nerves and a bilocular ovary. Among the species with regularly valvate calyx dehiscence, G. goldenbergii is most similar to G. obliqua Triana (1871: 71) , which is distinguished by leaf laminas with 5 acrodromous nerves, the inner pair being asymmetric and 5–10 mm from the leaf base, as well as by the mode of calyx dehiscence.

The passage of seventy years between first collection of this new species and its description here is much longer than typical for a Brazilian species ( Sobral & Stehmann 2009). This extended interval can be explained by two factors: the first collection, which was made in the Peruvian province of San Martin, bore fruit but no flowers and remained undetermined for 31 years until J.J. Wurdack recognized it as a Graffenrieda . Secondly, another 39 years elapsed before a second collection was made, with flowers and in Brazil. The low collection densities in and between the areas from which these two collections were made make it difficult to discern whether the current known distribution is a genuine disjunction or simply an illustration of the need for more intensive sampling of the Amazon flora.

UPCB

Universidade Federal do Paraná

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