Sansevieria sumbawangana M.Burkart, Constantine, Sikawa & Yinger, 2025

Burkart, Michael, Kavula, Kened Alfred, Constantine, Immaculate Kileo, Mollel, Neduvoto Piniel, Piniely, Luciana Naftal, Sikawa, Robert Augustino & Yinger, Barry R., 2025, Sansevieria bangalalana sp. nov. (Asparagales, Asparagaceae), close to extinction in the wild, and five other narrowly endemic and threatened species of Sansevieria from Tanzania previously unknown to science, European Journal of Taxonomy 1026, pp. 65-106 : 94-98

Sansevieria sumbawangana M.Burkart, Constantine, Sikawa & Yinger sp. nov.

urn:lsid:ipni.org:names:77370781-1

Figs 1, 20–24, Table 3

Diagnosis

This capitate Sansevieria is characterised by its peculiar leaf form ( Fig. 20), the extraordinary leaf width, the massive rhizome of up to 70 mm diameter, the almost uniform, very dark colouration of the flat leaves, and the protrusions on the fruits. The width relationship of the perianth lobes is also uncommon – in most species of Sansevieria , the outer lobes are a little bit narrower than the inner or of the same width, not wider as in S. sumbawangana sp. nov.

Etymology

Sansevieria sumbawangana sp. nov. is named for Sumbawanga Town, where this species was found for the first time.

Type material

Type

TANZANIA • Rukwa Region, Sumbawanga, Malangali Ward ; 1850 m a.s.l.; 5 Mar. 2020; B. Yinger & R.A. Sikawa YS 0317; holotype: NHT [ 000001154 ] .

Living specimens ex holotypo cultivated at TSF and POTSD .

Paratype

TANZANIA • Mbeya Region, Mbozi District ; 1612 m a.s.l.; 20 Oct 2021; B. Yinger & R.A. Sikawa YS 0930; NHT [ 000001158 ].

Living specimens ex paratypo cultivated at TSF .

Description

Plant rhizomatous, stemless, vegetative height 416 mm, 2–5 leaves per shoot†† ( Fig. 20A). Rhizome subterranean but close to soil surface, runners short, to 100 mm long††, rhizome diameter 70 mm, inner rhizome cortex brownish orange 165B. Leaves ( Fig. 20) dispersed in all directions, upright to spreading, very stiff, to 1 m long and to 210 mm wide††, lanceolate, widest part above the middle, distal part rather shortly narrowed to a leathery 20–25 mm long tip, basal part without petiole, openly u-shaped to almost flat, central lamina ca 20 mm thick††, almost uniformly dark green, adaxial surface smooth, abaxial surface very smooth, both with many (ca 25) longitudinal, wavy fissures from base to tip ( Fig. 21), leaf edges somewhat wavy, narrowly red-brown and whitish. Inflorescence ( Figs 21–22) capitate, terminal from a fully leaved shoot; axis 385 mm long overall; peduncle 300 mm long, 20 mm in diameter, green with lighter tiny dots and lines, bearing 7 bracts to 78 mm long; some bracts dry, the fresh ones coloured greyish purple; flowering part 85 mm long, flowering head 236 mm in diameter excluding anthers and styles, very dense, 2–3 flowers per partial inflorescence; bracts of partial inflorescences 25–30 × 7–14 mm, herbaceous, lanceolate-elliptic, one- to five- but mostly three-nerved, no extrafloral nectar seen on them. Flowers large, longitudinally striate, on inarticulate, green pedicels 10 mm long, corolla outside coloured yellow green with brown dots on the tube, lobes coloured white inside, tube 90 mm long, lobes 32 mm long, overall flower length 122 mm, tube 3 mm wide at the narrowest point, 4 mm across the ovary, 5 mm across the distal part of closed flowers shortly before anthesis, inner lobes 3 mm, outer lobes 4 mm wide, all strongly curling back at anthesis, closed anthers 5 mm long, open anthers 4 mm long, filaments thread-like (not flattened), 36 mm long, style straight, 115 mm long excluding ovary, 38 mm exsert, i.e., 2 mm exceeding the filaments. Flowers opening at dusk, wilting in the morning, self sterile. Fruits (only seen unripe#, Fig. 23) rugose and with protrusions at their distal ends. Seeds unknown.

Ecology and distribution

There are presently two locations only where this species has been collected, including the type location. These are in the Rukwa and Mbeya regions, respectively, both in SW of Tanzania ( Fig. 1). The site of YS 0930 was on rocky slopes and in ravines under a largely closed canopy of trees and shrubs, which would completely shade the plants during the rainy season, but at the time of the visit at the height of the dry season let in plenty of light ( Fig. 24). YS 0317 was collected in a strongly conversed area on a termite mound beside a recently built house ( Yinger & Sikawa 2023c: fig. 19). Both sites are at rather high elevations, 1850 resp. 1612 m.

Taxonomic remarks

We are not aware of any species in the subgenus Capitulatus , to which this species clearly belongs, that appears to be closely related to S. sumbawangana sp. nov. A comparison with S. kirkii , S. bhitalae R.H.Webb & L.E.Newton and S. sinus-simiorum Chahin. , the only other species of the subgenus Capitulatus that are similar in overall size, is given in Table 3 .

The bracts of partial inflorescences of capitate Sansevierias often have several nerves (e.g., S. kirkii , S. fischeri , S. bhitalae ), as is the case in S. sumbawangana sp. nov. Species from the other subgenera mostly have single-nerved bracts (but see, e.g., S. rosulata T.G.Forrest ; Forrest 2017).

Tentative threat assessment

Endangered: EN D.

The type location of this species is in an area that has been almost completely converted to agriculture and housing estates. There is almost no natural vegetation remaining in the area, although quite a few termite mounds remained. The plants there were persisting beside a termite mound close to several houses. Although we searched the surrounding area, we did not see any other plants of this species. There are three plants remaining at the type location.

We saw more plants in an area that is mostly undisturbed, on rocky slopes and in ravines within the drainage of Lake Rukwa (collected under YS 0930). Although it is risky to do a field identification without flowers, we believe that we have seen at least two more colonies in the vicinity of Lake Rukwa, comprising at least 200 plants. Although this area is not formally protected, it does not seem to be immediately threatened. The location and topography make it an unlikely site for farming or housing.

Regarding these circumstances, the appropriate category of threat for S. sumbawangana sp. nov. is EN (IUCN criterion D, <250 mature individuals) according to our data.

Parts of the side of the lake opposite where we found our collection are already within a reserve that might become part of an expanded national park. Its vegetation is quite different from the west side; there was only S. bhitalae s. lat. found.

In the TSF living collection there are several plants from other locations which, based on their foliage, might belong to this species, but they have not flowered yet. This option can lead to a reduction in the threat status of S. sumbawangana sp. nov.