Diceratostele gabonensis Summerh., Bull. Misc. Inform. Kew 1938: 151 (1938).
Roots ca. 0.5 cm in diameter, wiry, sparselly branched. Stem 36-60 cm tall, erect, very stiffy, glabrous, with short aerial roots in the lower part, branching from axillary buds below the inflorescence, clad in strawy or medium green sheaths. Leaves 5-10; petiole 3.5-8 cm long, stout, canaliculate; blade 8-18 cm long, 2-6 cm wide, elliptic, apiculate, plicate, relatively stout to papery, glossy bright green with 3 impressed nerves above, slightly paler below with 3 prominent nerves. Inflorescence 1-4 per plant, up to 4 cm long, 10-15-flowered, terminal or axillary, rachis pubescent. Flowers inconspicuous, tubular, resupinate, white to greenish. Floral bract up to 15 mm long, ovate-lanceolate, acute, pubescent. Ovary up to 10 mm long, twisted, erect, pubescent. Dorsal sepal 11-20 mm long, 2-3.5 mm wide, narrowly obovate to elliptic, obtuse with or without short apiculus, densely pubescent outside. Petals 10-17 mm long, 2-3.3 mm wide, similar in shape to lateral sepals, with longitudinal pubescent crest outside. Lateral sepals 11-18 mm long, 2-3.5 mm wide, obliquely obovate to elliptic, subobtuse, with longitudinal crest outside, densely pubescent outside. Lip sessile, constricted near the middle, canaliculate; hypochile 4.8-9 mm long, 3.8-6 mm wide, trullate, side lobes obtuse, with large, fleshy, flat callus in the centre; epichile 6-7 mm long, 4.5-5 mm wide, obovate-elliptic, obtuse, flabellate and crispate along margins, with five longitudinal crestes in the centre. Gynostemium 7 mm long.
Riverine forests, damp forests. January-April.
Not in cultivation
This remarkable new genus seems to be properly placed in the subtribe Sobraliinae. Among African orchids it is very similar vegetatively to Corymborchis but the anther is placed at the apex of the column and appears to be operculate. Unfortunately no pollinia have been found in the few flowers available for examination. Mansfeld has recently (Notizbl. Bot. Gart. Berl. 13, 666-667 : 1937) advocated the transference of the two genera Rolfea Zahlbr. And Neobartlettia Schltr. from Tropidiinae to Sobraliinae and from examination of both genera I entirely agree that this should be done. Diceratostele is undoubtedly closely related to these two genera but differs from both in the remarkable flattened horn-like appendages at the top of the column as well as in the general structure of the lip. It is, however, interesting to note that many species of the genus Sobralia itself possess to a much less extent similar appendages at the apex of the column. Sobralia differs from Diceratostele in many ways, of which the most obvious are the inflorescence and the size of the flowers. The fundamental structure of the column and anther in the genera mentioned above as well as in Elleanthus is relatively constant and quite different from that in Corymborchis and Tropidia. Since Mansfeld has also placed Arundina and Thunia, both Asiatic genera, in the Sobraliinae the discovery of an African genus belonging to this subtribe is of considerable importance.
W. & WC. Trop. Africa
Orchidaceae of West-central Africa, vol. 1, Gdansk University Press Gdansk 2010