Grias purpuripetala S.A.Mori, J.D.García-Gonz., S.Angel & C.Alvarado
-
Description
Author: Scott A. Mori, Juan D. García-González, Sonia P. Angel, Cinthia Alvarado & Xavier Cornejo.
Type: Colombia. Nariño: Municipio de Barbacoas, Vereda El Barro, Corregimiento de Altaquer, Reserva Río Ñambí, 1350 m alt., 27 Apr 2009 (fl), S. P. Angel et al. 282 (holotype, COL; isotypes, COL, PSO).
Description: Pachycaulous, monocaulis, understory trees, 3-10 m x 6-15 cm, the trunk cylindrical, not buttressed. Bark without fissures or scallops, appearing smooth but on close inspection somewhat rough, the outer bark thin (<1 mm thick), the inner bark 5-10 mm thick, pale yellow. Stems glabrous, the leaf-bearing stems to 30 mm diam. Leaves: petioles to ca. 50 mm long, semi-elliptical in cross section, glabrous; blades oblanceolate, somewhat spathulate in some leaves, 100-200 x 42-75 cm, chartaceous to coriaceous, glabrous, with scattered dark green punctuations abaxially, the base attenuate and usually obtuse to sometimes rounded, the margins entire, the apex abruptly acuminate. Venation brochidodromous, the secondary veins in ca. 45 pairs in largest leaves, the tertiary veins finely percurrent, the higher order venation obscure. Inflorescences cauline, short racemes, with 1 to 5 flowers, the rachises glabrous, ca. 10-15 mm long; pedicel ca. 2 mm long below articulation, 6 mm long above articulation, glabrous, subtended by narrowly ovate bract, 5-7 x 3-3.5 mm, with 2, small (ca. 2 x 0.5 mm), caducous bracteoles inserted ca. 5 mm above articulation. Flower buds globose; mature flowers 5-6 cm diam. when petals oriented upward as seen on tree, ca. 12.5 cm when artificially spread horizontally; hypanthium glabrous; calyx splitting in 2-3, somewhat cucullate lobes at anthesis, the lobes 12-14 x 9-16 mm, glabrous, rose to light purple; petals elliptic to somewhat oblong, ca. 60 x 40 mm, dark purple, oriented with axis of flower at anthesis (i.e., not spreading), and curved inward at apex; androecium globose, forming carnose tube, ca. 20 mm high, externally rose-colored, divided into 2 chambers, arching from base to apex abaxially, the lower chamber erect or nearly so adaxially, the upper chamber arching, adaxially concave, abaxially convex, with 34-31 stamens, inserted in more-or-less three levels, one level along rim, another level just below rim, and another level just below that, the filaments tapered into anthers (i.e., not constricted), the outermost 5-6 mm long, the connectives expanded, the anthers oblongoid, ca. 3 mm long, with introrse dehiscence, slightly apiculate at apices of thecae; ovary 3-locular, with 2-3 ovules per locule, glabrous and obconical at summit,, the summit more-or-less truncate, with an intrastaminal disc ca. 0.1 mm high, and a shallow depression between disc and style, the style absent. Fruits widely fusiform, ca. 8.5 x 6.5 cm (dry), with 8 longitudinal visible but not prominent ribs (dry), the mesocarp not seen. Seeds ca. 50 x 40 mm.
Common names: Colombia: Hoja cuero (S. P. Angel et al. 282), huevos del gallo.
Distribution: Known only from the Ñambi cloud forest reserve in Nariño, Colombia on the western slopes of the Andes.
Ecology: A common understory tree of pretmontane wet forest. Although we noted numerous individuals of this species, only a single fertile plant was encountered in late April when the type was collected.
Phenology: Trees as short as 3.5 m have been collected fertile. Collected in flower in Apr and Jul and in fruit in Apr and Dec. The manager of the Nambi Resesarch Station, Maurício Flores, told us that this species is fertile year around.
Pollination: The pollinators are unknown but this species is hypothesized to be bat-pollinated for the following reasons a) flowers were seen on the ground at 2:00 p.m. and 11:00 a.m. when we walked to the station on 23 April and left the station on 24 Apr 2009, respectively suggesting that the flowers are nocturnal, b) the flowers are unusually large for the genus which suggests a large pollinator such as bats, c) the observation of what we judged to be bat feces were found under the tree, and d) the discovery of an androecium of a flower that was partially eaten in a manner suggesting that bats were responsible. The presence of an intrastaminal disk and a circular depression between the disk and the base of the style suggest that nectar is produced as a pollinator reward. In addition, the tubular androecium is also capable of holding large quantities of nectar. Knudsen and Mori (1996) have noted that the floral aroma of Grias is consistent with beetle pollination but there have been only casual observations of beetles in the flowers of Grias. If nocturnal beetle pollination does occur in this genus, that would have set the stage for the evolution of this presumably bat-dispersed species. Some other species, such as G. neuberthii, have yellow flowers, a color often associated with bee pollination. With such little knowledge and so many potential pollinators, Grias is a prime candidate for a phylogenetically-based pollination study.
Dispersal: No observations have been recorded but the mesocarp is probably eaten by animals. Observations are needed on the size, texture, and color of the mesocarp, the type of seed coat, and the way the seeds are attached to the fruit (i.e., a description of the funicle and the presence or absence of an aril).
Predation: No observations recorded.
Field characters: The pachycaulous, monocaulous growth form; very large leaves grouped at the ends of stout branches; well-developed and carnose androecial tube with reflexed stamens; apiculate anthers; large flowers with purple petals; and broadly fusiform fruits with distinct ridges make this species easily recognized in the field. The dark purple flowers separate it from all other known species of Grias.
Taxonomic notes: The pachycaulous growth form is found in all species of Grias but other species are often branched , especially when they are the size of the flowering individual we collected; thus, it would not be a surprise if further exploration revealed branched plants. We hypothesize that the flowers are nocturnal based on observations of fallen flowers in the late morning and early afternoon. All petals observed were erect, i.e., oriented upward along the long axis of the flower. Our observation that the ovaries are 3-locular is based on a single cross section of an ovary and the presence of three stigmatic lines in another flower; but, because the ovaries of other species of Grias are mostly 4-locular this needs to be confirmed with additional observations. This species is most closely related to G. theobromicarpa of northern Ecuador from which it differs in the dark purple instead of white flowers. Both of these species have expanded connectives, a feature not found in other species of Grias.
Conservation: Because G. purpuripetala is known by few collections from the type locality and nearby within an area of less than 5.000 km2, we suggest that this species be assigned the IUCN conservation status of endangered, EN B1abiii (IUCN, 2001).
Uses: The Awá Indians are reported to smoke the leaves to make them more pliable for use as blanket when they spend nights in the forest.
Etymology: The species epithet refers to the dark purple color of the flowers.
Source: Mori, S.A., J. D. García-Gonzállez, S. P. Angel & C. Alvarado. 2010. Grias purpuripetala (Lecythidaceae), a new purple-flowered species from southern Colombia. Britonia 62: 105-109.
Acknowledgements: We are grateful to B. Angell for allowing us to use her line drawing to illustrate the characters of this species.
- Sorry, no descriptions available for this record.
