Rubus rosifolius Smith (1791: 60)

Beek, Abraham Van De, 2021, Rubi Capenses: a further contribution to the knowledge of the genus Rubus (Rosaceae) in South Africa, Phytotaxa 515 (1), pp. 1-71 : 10

1. Rubus rosifolius Smith (1791: 60) . Type :— MAURITIUS. Commerson-Thouin 680 (holotype LINN [http://linnean-online.org/43735/])

Heterotypic synonym:— Rubus trifoliolatus Suessenguth (1950: 54) . Type:— SOUTH AFRICA. Western Cape: Franschhoek Pass , 17 November 1946, J.W.H. Giess 1391 (holotype M-0214132).

Notes: —Plants of R. rosifolius which are similar to the type of R. trifoliolatus , are also found in other regions, see e.g. J.J.S et Rant 607 (P03144942), collected in Indonesia on 1911.09.19. Suessenguth’s specimen is the only more recent discovery of R. rosifolius in the Western Cape.

Many older collections are from Table Mountain (see Sochor et al. 2018 for cited specimens). The species was not found there more recently.

Another interesting specimen was found in the collection of Zeyher in P (Zeyher 398 [5378], P02579477) under the name R. heptaphyllus Opiz (1822: 269) . It was collected at Witteboom, near Cape Town, on 11 August 1826, and mentioned in Drège (1843). Drège (1843) mentions two other localities of R. heptaphyllus (Groot Drakenstein/ Paarlberg and Keiskamma/Buffelsrivier), but these specimens could not be traced. The sample from Witteboom is poor, but it seems to be identical with R. rosifolius .

Sochor et al. (2018) and Sochor (2019) provide good descriptions and pictures of R. rosifolius .

Distribution: —The species has a wide distribution around the Indian Ocean and on the Pacific Islands. Plants in South Africa are usually considered as introduced ( Gustafsson 1934, Henderson 2011, Sochor et al. 2018, Sochor 2019). Kalkman (1981: 109) doubts this opinion because of the general distribution of R. rosifolius and R. niveus Thunberg (1813: 9) . Kalkman’s suggestion is confirmed by a specimen that has not been cited in the given literature. This specimen in the collection of Drège (Zeyher 9537, 22 January 1827, P04154180; see also Drège 1843: 59) was gathered in the Camdeboo Mountains at 4000 ft. It is hard to imagine that an introduced species would grow there at that time. Instead of the hypothesis of introduction, a natural distribution is more plausible. The Indian Ocean currents often bring material from eastern regions to the coasts of the western islands and to southern Africa. Plants, which have their origin in the Far East, can reach Africa in this way and spread through the continent. Of course, they will have their main distribution in the coastal areas of the east coast. The presence of R. niveus in southern Africa must also be viewed from this perspective. Its occurrence is not in typical introduction areas, but in natural habitats in the mountains.