406 genus, and until this year had not noticed a peculiarity that 

 seems to exist in reference to the fertilisation of certain spe- 

 cies of this genus, which is this. The stamens slied their pollen 

 before the stigmas are ready for fertilisation, being tightly 

 closed together into a snire-like column while the pollen is being 

 shed. ITiis, of course, is a well-known feature in plant-fertila- 

 zation, its purpose being to secure cross fertilisation. But in 

 those flowers of Chelridopsis tuberculata and '<^, rostra ta and I have 

 seen this year, when the pollen is all shed then the stigmas com- 

 mence to separate and at the same time, or in one case just before 

 the stigmas began to separate, the petals begin to wither and are 

 ouite fader! while the stigmas seem to be quite fresh and receptive, 

 a neculiarity I have not noticed in other flowers, yet in some 

 SDecies the stigmas are radiately spreading while the petals are 



in a perfectly fresh condition. The pollen seems all to fall among 

 the stamens and is seen to be plentifully covering their filaments ; 

 therefore, I suspect that these plants are fertilised by very small 

 insects and mites, v-hich probably crawl through the mass of stamens 

 to get to the nectar-glands, and thus convey pollen from flov/er to 

 flower. 



The presence of a pair of leaves Hhe so-called bracts) upon 

 the pedicel above the basal pair has been thought to be a specific 

 C'aracter, but I find on a plant of C, tuberculata, cultivated by 

 Mr. ^ndean, some flowers are produced with and some v/ithout the 

 extra pair of leaves on their pedicels on different growths of the 

 same 



407 plant and growing in the same pot. Therefore, the specific value 

 of the presence of this pair of leaves is doubtful. . I have pre- 

 served specimens showing this variation for the Kew ^-^erbarium. 



I am unable to make a key to the species of ^heiridopsis, be- 

 cause I have no materiel adequate for the purpose. Therefore I 

 can only give a list of the species I refer to this genus, giving 

 reference to the original descript'ons and any subsequent figures, 

 the descriptions of later authors being omitted, as they are often 

 imperfect or erroneous J — 



C. bibracteata, K, E, Br. — Mesembryanthemum bibrac tea turn. 

 Haw. Syn. PI. Succ, p. 21? (1812). i^. S. Br. in ^ourn. Linn. 

 Soc. Bot., vol. XLV, p, 77, t. 6, fig. 14; and M. rostratum var. 

 brevibracte^ttim, Salm ^yck, Mesembryanthemum 3, fig. 7B. 



0. bifida, N, E, Br. — I'-esembryanthemum bifidum, ^w. ^^isc, 

 Nat. p. 29 (1803), K. E. Br., in Journ. Linn. ^oc. Bot.,^ vol. 

 XLV, p. 78, t. 9, figs. 34, 35; and M. multipunctatiam, ^alm.I>yck. 

 Hort. Dyck., p. 357 (1823), and ^'^semb. 3, fig. 6; Bgrger, ^esemb., 

 p. 256, fig. 55, IV. 



C, candidissima , N. E. Bp. (^ig., 211'. — Mesembryanthemum 

 denticulatum var candidissimuT:, Hgw. *^bs. ^semb. , p, 151 (l7S5) 

 Ann. S. Afr. Mus., vol. IX p. 142, t. 3, and ^'^. candidissimu, 

 N. E, Br. in Journ. Linn. Sqc. Bot., vol. XLV^ p. 79^ t. 7, figs. 

 25, 26. I'y figure et this place represents only a small, starved 

 Plant, as 1 find this species attains to two or three times the 

 dimensions there depicted, as is presented in ^''ig. 211 B, 



C. cernea, N. E. Bp. — Described later. 



