Notes

Burbidge (op cit., p. 77) discusses at length the taxonomic and nomenclatural difficulties in the treatment of T. violacea Harv. and T. simmleri Beauverd. He concludes that he has maintained specimens with ± entire corona lobes and ± plane perianth segments in T. violacea, and specimens with retuse corona lobes and concave perianth segments in T. simmleri. His treatment is in contradiction with the protologues of these names. T. violacea is described as having corona lobes that are obtuse, emarginate or bifid and perianth lobes with incurved margins (= concave?). The illustration agrees with the description, with the corona lobes clearly drawn with retuse apices. The description of T. simmleri describes the corona as urceolate, crenate-trilobed and emarginate, but no comment is made as to whether the perianth lobes are plane or concave. There is no type specimen of T. simmleri, just an illustration, and the locality is given as ‘Transvaal’. From the information provided by Beauverd, it is debatable whether T. simmleri should be sunk into T. violacea. In his account of Tulbaghia cytology, Vosa (1975) uses the illegitimate name T. cepacea (= T. simmleri, see Burbidge op cit.), and cites T. violacea in synonymy. If indeed these two are synonymous, T. violacea would take priority being the older of the two legitimate epithets.Cultivated in Uganda, Mengo District: Kampala, Makerere University Hill, in botany gardens, 5 May 1971, Lye 6026!There is quite a large difference in the size of the vegetative parts of the two specimens, the Tanzania specimen being smaller. This is probably due to the fact that the Ugandan specimen was cultivated.AFFINITY UNCLEAR Burbidge has named two collections from Tanzania (Shabani 76 and Bullock 1901) as ‘Tulbaghia sp. nov.?’. On careful examination of these two specimens, we too have found it difficult to decide where they should be placed, but we are of the opinion that they represent two different taxa. The two specimens differ in leaf characters, Shabani 76 has up to 8 long linear leaves (almost equalling the scape), whereas Bullock 1901 is almost devoid of leaves, and a couple of plants show few, greatly reduced leaves. The scapes of Bullock 1901 are more robust than the slender, wiry scapes of Shabani 76. The flowers show differences as well. Bullock 1901 was determined as T. aff. rhodesica R.E. Fr., and yet the corona is very well developed up to 5 mm in length (longer than the corona of T. cameronii) which completely discounts any relation to the aforementioned species (see under T. rhodesica R.E. Fr.). For the same reason, relation to either T. friesii or T. aequinoctialis (as suggested by Burbidge) can be ruled out. We would gladly place Bullock 1901 in the variable T. cameronii were it not for the few-flowered umbels that this specimen displays. The combination of umbel with few whiteflowers, and long orange coronas leads us to suggest that Bullock 1901 could be placed under T. leucantha. Burbidge explains that in the northern part of the range of T. leucantha, T. leucantha and T. cameronii overlap and that intermediates are found where the two are sympatric, probably representing hybrids (no collections are cited). It is therefore possible that our specimen is a hybrid. Previously, T. leucantha has been collected as far north as Zambia, and there is a specimen from Malawi determined as ‘possibly T. leucantha’, so it is not too unreasonable to assume that it may be present in the Ufipa district.Burbidge also determined Shabani 76 as Tulbaghia aequinoctialis, but T. aequinoctialis as is restricted to Angola (the most westerly distribution of any Tulbaghia ?) and Shabani 76 is from Mpemvi in the Buha district (T 4). This makes it highly unlikely that our specimen fits there. In Shabani 76 the umbel is few-flowered (2–4-flowered, and the corona is 2–2.5 mm long, not as marked as that in Bullock 1901 nor in T. cameronii, but far exceeding that of T. rhodesica). In view of its distribution, and the differences in vegetative and floral characters from sympatric species, it is likely that Shabani 76 represents a new taxon. However, the material is poor and the label inadequate to describe a new taxon; collection of further material is very desirable.