335 here named C, coniaunis. Not I-. pugioniforme, Linn. 



South Africa I Locality and collector not stated, but see 

 note belov;. 



The above description is^ translated from Haworth s account of 

 of the plant he described as i-. pugioniforne, but w^hich on account 

 of it^ shorter duration, angiilar and rough pedicels, calyx-lobes 

 longer than the petals and v;ith fuscous membranes, cannot be the 

 same as '-. puglonif orme, Linn. 



IIav7orth states that this species is nearly tv.dce as large as 

 and more glaucous than his M. capitatum (which - G. pugioniformis) , 

 and that its flov;ers open earlier in the morning but do not last so 

 long. 



Ar>parently no type or figure of the original plant has been 

 r^reserved, and as the locality from v/hich it came is unknov/n, this 

 species cannot be identified v;ith absolute certainty. But I have a 

 branch of a plant that was sent to me by Dp. i'^ughan Brown fromthe 

 region of Saldanha Boy, vhich in habit its angulete and slightly 

 rough pedicels and fuscous membranes of its calyx-lobes corresponds 

 with Hav/orth's description Of M. pugioniforme, and as the region is 

 one from v/hich plants were introduced at the end of the eighteenth 

 century, it is not impossible that Hcworth's plant may have come 

 from that area. Therefore, in the absence of any evidence to the 

 contrary, I propose that this Saldanha Bpy plant be considered as 

 being the type of G. capensis, until proved not to be so, and give 

 the following account of the c'ried fruiting material I have? — 

 "Flowering br- nches anperently much like those of G. communis and 

 more or less rngular. Leases very shrivelled, but one or tv/o appear 

 to have been acutely trigonous, -ciowers 5 or 6 at the end of the 

 branch. ledicels 3-5 inches long, distinctly angular at the upper 

 part, and slightly rough to the touch. Gapsule T^ig. 138) from 

 edffe to edge of rim 1-1^ inch in diameter, shallowly hemispherical 

 beneath, with ridges radirting from the redicel, v/here they sre 

 almost wing-like, to the sinuses betv-een the cakyx-lobes,end on the 

 top with about 15 valves raised into a cone 7-9^ lines high and 6-9 

 lines in daneter at its base. Seeds 2/3-line long, ellipsoid and 

 distinctly much longer than thick, smooth, brown, spotted with 

 darker bro^^n. 



N. E. Brov>'n 

 (To be continued.) 



MESEI','!BRYANTHEI.IUT,: , 

 Gard. Chron. III.. 91'. 104. 1932. 

 (Gontinued from page 335, Vol. XG.) 



COKIGOSIA. 



104 5, G. pulliloba, N.E.Br.— (Fig. 42). Plant apparently like 

 C. communis in habit, size and general a^^pearance, v/ith decumbent 

 flowering branches un to a foot long, terete, smooth. Pedicels 

 3-?: inches long, terete, smooth, not angular, ^lowers apparently 

 very similar to those of C, communis in si-e and colour, but ^dth 

 the membranous m-?r£rins of the cal:'rx-lobes drrk fuscous, not pallid. 

 Petals shorter than the cal;'rx-lobes. Stigmas about 15. Ggpsule 

 from edge to edge of the rim pn inch or rather more in diameter, 

 an^ the ocne of carpels about six line in height, with the valves 

 p-rooved on the back. 



Gppe division: Near J'^lmesbury ^oe.6. through Riet ^ley, V-'olley 



