152 
M. cordifolium, Thunb., “ Fl. Cap..” p. 417. 
One sheet of this isa collected in a » wood near Leekoe River, in the Humansdorp ss 
Division. It is— . 5 | 
M. cordifolium. Linn. f.. °* Suppl.,” p. 260 (1781), and is identical with Zeyher No. 2621, 
M. crassifolium. Thunb., ` Fi. Cap.,” p. 421. 
There is one sheet of this containing two specimens, for which Thunberg gives the 
locality : hills in Zwartland, Malmesbury Division. These specimens belong to— +g 
M. filicaule, Haw., ` * Mise.. Dee ae, . 
M. crassifolium, Thanb., ` Prodr., 90 (1800), not of Linnaeus. 
Thunberg’s specimens are identical with MacOwan, `` Herb. Aust.-Afr.” (1731), from 
the foot of Devil’s Peak. 
Sonder in © FI. Cap..” Vol. H. p. 407. refers M. crassifolium, Thunb., to M. repians, Ait., 
but that is a very different species. with larger, prominently dotted (not smooth) leaves and 
much larger yellow or white (not rosy) flowers. 
M. debile, Haw., also placed by Sonder as a synonym of M. reptans, Ait., is likewise 
totally distinct from the latter, and 1 believe it to be the plant distributed by Schlechter 
under No. 9241 as M. filicaule (which it certainly is not) from French Hoek. 
M. criniflorum, Thunb., ` Fl. Cap..” p. 411. 
One sheet containing four good specimens, collected on sandhills near Capetown. 
These specimens agree with those of MacOwan. “ Herb. Austr.-Afr.” (1745), from near 
Sea Point. Their correct synonymy is 
M. criniflorum, Linn. f., `` Suppl..” p. 259 (1781); Thunb., ` Prodr.,” p. 88 (1800). 
M. limpidum, Ait., ` Hort. Kew.,” ed. 1. Vol. IT, p. 182 (1789). 
M. spathulatum, Willd., ` Sp. Pl.” Vol. IL. p. 1025 (1799), not of Thunberg. 
M. tricolor, Willd., ` Hort. Berol.,” Vol. 1. p. 22, t. 22 (1803), not of other authors. 
(V.B.-Willdenow’s work was issued in parts. so that the date 1816 on the title-page is 
very misleading. that being the date when Volume 11 was issued as a whole, but it was 
issued in parts previously.) 
The colour of the flower varies. 
Sonder and Berger have erroneously quoted JM. tricolor, Willd.. as belonging to 
M. pyropaeum, Haw. They also erroneously quote Houttuyn as being the author of 
M. eriniflorum, whilst Thunberg (" Fl. Cap.,” p. 411) ) attributes the name ` M. capense ” 
to Houttuyn, but neither of these names are to be found at the place quoted in Houttuyn’s 
work ; the plant referred to being M. pugioniforme, Houtt., ` Nat. Hist.,” Deel IL, t. 53, 
separatelv issued as Handleid., `` Planten tot de Kraidkunde,” Vol. IX, p. 32, t. 53, which 
represents a form of M. pomeridianum, Linn., in fruit, with capsules having 16-20 valves 
and cells, whilst M. criniflorum bas only 5 valves and cells to its capsules. But authors 
have paid little heed to frait-characters in this genus. 
M. crystallinum, 'Thunl.. ` Fl. Cap. p. 413. 
There are two sheets of this. containing specimens that appear to be M. erystallinum, 
Linn. The locality given for it by Thunberg is near the Sundays River, but sheet 1 is 
labelled as being from a plant cultivated in Upsala Botanic Garden. 
M. decussatum, 'Thunb., ` Fl. Cap...” p. 414. . 
One sheet with two specimens upom it. collected on dry Jills near the Olifants River, 
towards the north, in the Karroo below the Bokkeveld. Its synonymy is— 
M. brachiatum, Ait., °° Hort. Kew.,” ed. 1, Vol. IT, p. 191 (1789), and Haw., “* Misc.,’” 
p. 58. 
M. decussatum, Thunb., `` Prodr.,” p. 88 (1800). 
