46 Albany Museum Records. 



size and colour of the corolla. Thus the vay\ incurvata Bak. can 

 be produced by growing seedlings of the typical form in the shade, 

 the var. suhferox, Spreng, is met with near Grahamstown mixed 

 with the more prickly form. I have already stated my belief 

 (on page 44) that A. Galpini, Bak. must be referred to^ this^ 

 and I think A. suj^raJaevis, Haw, must also be looked upon as 

 one of i+s varieties. Though the Flora Capensis states that the 

 leaves of A. supralacvis are ensiform, this statement is 

 scarcely borne out by Salm-Dyck's illustration {A loe, sect. 

 xxvii, fig. ()). The slight twist in the older leaves which make 

 them appear sub ensiform can also be noticed in ^. /erar, Mill. 

 There remains only the smaller size of the flowers which, how- 

 ever, can scarcely be looked upon as a specific distinction. The 

 flowers of ^./(^roa*, Mill, are near Grahamstown usually re-d, but 

 sometimes yellow and some received from Uitenhage were 

 cream-coloured. 



63 (327). A. africana, Mill. — This species is not uncommon 

 near Grahamstown, and extends almost to the level of the sea 

 near Port Alfred. It flow^ers in midwinter. Though the flowers 

 are usually yellow, I have noticed some plants with red flower- 

 buds. The tube in our specimens is | the length of the perianth. 

 Though the leaves are usually as described in the Flora Ccq)ensis, 

 one frequently meets specimens which are very prickly both on the 

 face and the back, and this happens to be the case with most of the 

 specimens cultivated in Grahamstown, while amongst the cultivated 

 specimens ot A.fero.r, Mill, the leaves have usually only very few 

 prickles on the back and none on the face. This, however, is due 

 to selection and not to the effects of cultivation. I must express 

 my surprise that in the Flora Capensis, A . a/ricana is placed 

 between A.ferox and A. sujtralaevis. The latter, to say the least, 

 are very closely allied and have not the curved corolla of A. 

 africana. The key on p. 304 should therefore be slightly 

 amended. I may take this opportunity of stating, that in some 

 other places it is also somewhat misleading. 



()5 (327). A. rupestris, Bak. — Fine specimens of this species 

 are growing in the Capetown Municipal Gardens. They are said 

 to have originally come from Namaqualand. 



CG. (32<H). A. variegala., L. — This old favourite of Eui-opean 

 garilens is represented in the Herb, of the Albany Museum fj:om 

 {Sheldon, Cape Colony, where it is common. It seems to be 



