Saxicolella sp. A

Cheek, Martin, Molmou, Denise, Magassouba, Sekou & Ghogue, Jean-Paul, 2022, Taxonomic revision of Saxicolella (Podostemaceae), African waterfall plants highly threatened by Hydro-Electric projects, Kew Bulletin 77 (2), pp. 403-433 : 424-427

8. Saxicolella sp. A

Syn. Sapicolella nana auct. non Engl. (GBIF.org)

Prostrate herb with long, ribbon-like roots (black in life), infrequently bifurcating. Shoots sessile, forming 1-flowered leafy rosettes along margins of the root in rows, stem not visible. Leaves 2 – 6 per rosette, linear, flattened, as long or longer than the flower, entire, very rarely bifid at apex, apex acute to obtuse, stipules not seen. Spathellum caducous, sometimes visible as a broken ellipsoid remnant. Pedicel cylindrical, stout and white or pink in life, as long as ovary and nearly as broad. Stamen with filament 1.5 – 2 times as long as ovary, white to pink. Anthers with cells white, hemiglobose, isodiametric, semi-latrorse, each about as wide as filament. Gynophore absent. Tepals not seen. Ovary sessile narrowly ellipsoid, dull red, longitudinal lines not visible apart from commissure, orbicular in plan view (not laterally compressed). Styles 2, purple-red, erect stout cylindric, apices obtuse. Fruits 8-ribbed (Tepier et al. 2221).

RECOGNITION. Differing from Sapicolella nana in that the root is long and ribbon-like with the shoots lacking visible stems, and arising along the margins of the root in rows (vs disc-like, crustose, the shoots with visible stems, arising from the centre of the root in a cluster); the leaves entire, linear, and not trifid from a point c. 1.5 mm from the base; the ovary sessile (the staminal filament inserted at its base), not with a distinct gynophore; fruit 8-ribbed (not 6-ribbed).

DISTRIBUTION. Gabon, Ogooué River and its affluent the Ivindo.

SPECIMENS EXAMINED. GABON. Ogooué-Ivindo Province, Ivindo National Park, Kongou waterfall area, islet in the middle of the Ivindou R., 00°17ˈ22"N 012°25ˈ22"E, 355 m alt. fl. 8 Feb. 2018, Tepier, Niangadouma & Akouangou 2221 (BRLU, LBV, MO all n.v. photo); Haut-Ogooué Province, Boumango, Ogooué R., 02°16ˈ27"S 012°28ˈ20"E, 382 m alt., fl. 5 Aug. 2019, Nguimbit with Boupoya, Ikabanga, Kaparidi 26 (BRLU spirit only n.v.); Ogooué-Lolo Province, Chute Sessengué sur l’ Ogooué, 00°39ˈ19"S 012°50ˈ05"E, 222 m alt., fl. 21 July 2019, Boupoya with Nguimbit, Kaparidi 1952 (BRLU spirit only n.v., photo.).

HABITAT. Rocks in rivers sometimes (e.g. near Kongou Falls, Boupoya 1952) with other species of Podostemaceae , probably Ledermanniella (Boupoya 1952 – 1955, BRLU images); 222 – 382 m alt.

CONSERVATION STATUS. Once this species is formally published it will be possible for it to be Red Listed, most likely as Endangered EN B1ab(iii) since three locations are known with threats and extent of occurrence is calculated as 881 km 2. Although the three points are widely separated, they plot in a straight line forming a nearly linear polygon. Tepier 2221 is stated to be “frequent on rocks on Ivindo River” but it is close to the Kongou falls. In 2007 the President of Gabon announced that a hydro-electric dam will be built at the Kongou falls inside the Ivindou National Park to support the Belinga Iron Ore project ( Wikipedia, Kongou Falls, 2020). When implemented it can be expected to result in extinction of this species at this location as such projects have caused extinction of Podostemaceae elsewhere in Africa ( Cheek et al. 2017b). While threats of this nature have not been found for the other two locations indicated, there is no scope for complacency. Six new large hydroelectric projects are planned in Gabon ( Makoni 2020) and the total rises to 29 if potential and smaller projects are included ( Cutler 2019). Apart from these, falls and rapids are the ideal sites for placing smaller hydro-power projects needed for future development projects. The location of Nguimbit 26 is close to an undeveloped iron ore project in Congo- Brazzaville, while Boupoya 1952 is about 10 km Eof Lastoursville, one of Gabon’ s major towns.

PHENOLOGY. Collected in flower in Feb. and July and August (dry seasons).

VERNACULAR NAMES. None are recorded.

NOTES. The first author came across this taxon when doing final checks for additional records of the genus on GBIF.org before finishing the manuscript for this paper. The three specimens were identified as Sapicolella nana respectively by Rutishauser, Bidault & Rutishauser, Bidault & Mesterhazy. An excellent image of Tepier et al. 2221 and ten for Boupoya 1952 are available online through GBIF and Tropicos, (http://www.tropicos.org/ ImageDownload.aspx?imageid=100597656 and http: // www. tropicos. org / ImageDownload. aspx?imageid=100891262 respectively) but none for Nguimbit 26 which is included on the assumption that it matches the two specimens which had been named identically.

Inspection of these images shows several characters that differ from those of S. nana which are given under “Recognition” above. No scale was available, so the description is made without giving dimensions, entirely on the basis of the online photos. Therefore, a full formal description of this taxon is still advisable, including observation to confirm the expected unilocularity of the ovary. Since there are limitations to studying specimens only from images which can be misleading, Sapicolella sp. Ais treated as unplaced here. While it appears to belong to subgenus Butumia , it is indeed similar to S. nana . It is possible that when the specimens themselves are studied, the dissimilarities with S. nana will be reduced and that Sapicolella sp. Awill be reduced to synonymy. Yet this seems unlikely given the number of qualitative morphological differences enumerated above, sufficient to result in separation at subgenus level. The specimens themselves could not be reviewed on loan because of technical difficulties due to their transfer from BRLU to ZT (Bidault pers. comm. to Cheek June 2021).

Nguimbit 26 was collected at a point so close (<3 km) to the border with Congo- Brazzaville that it maps just inside that country on Tropicos. Since the river concerned extends far into that country it is very likely that the species occurs there and that it is not endemic to Gabon.

Gabon is seeing an upsurge in description of new species with 27 published in 2019, second only after Cameroon in tropical Africa, many of these being rare, range-restricted and endemic ( Cheek et al. 2020b), joining those already documented narrow and threatened endemics as such Diospyros rabiensis Breteler (Breteler 1993), Whitfieldia purpurata Heine and Whitfieldia rutilans Heine ( Grall & Darbyshire 2021), some of which are already extinct such Pseudohydrosme buettneri Engl. ( Cheek et al. 2021a). as