122 THE JOUKNAL OF BOTANY 



receptacle. In the submersed leaves it is unlike both species, for 

 in R. Lenonnandi all the leaves are nearly uniform in shape and 

 texture, and in II. pdtatns the submersed leaves are repeatedly 

 divided into capillary segments, and there are rarely any leaves of 

 a transitionil character ; whereas in 11. Hiltoni, while none of the 

 submersed leaves are truly capillary, almost all of them are transi- 

 tional in character, some of the specimens — that figured, for in- 

 stance — presenting a series of gradations from the repeatedly and 

 very deeply divided lowest leaves with linear segments to the 

 cordate-orbicular floatinoj leaves. 



NOTES ON AFRICAN STEROULIACE^. 



By Edmund G. Baker, F.L.S. 



The following notes have been made during a revision of the 

 African Sterculuicem at the British Museum with Dr. Schumann's 

 recently published Monograph." 



Melhania. 



Of this genus twenty-five species are described. There seems 

 confusion with regard to M. ffiiquensis Bolus and M. Rehmamii 

 Szyszyl. In Journ. Bot. 1898, p. 5, I indicated that I thought 

 these species identical ; but Dr. Schumann expresses his dis- 

 agreement with this view, and places them in different groups. 

 M. griqnoisis was established by Mr. Bolus in Journ. Linn. Soc. 

 xxiv. p. 172 (1887), on Burchell no. 2050, from Kloof Village, 

 Asbestos Mountains, and on a plant from Griquatown, Mrs. Orpen, 

 Herb. Bolus, no. 6045. Mr. N. E. Brown in a note (/. c) added 

 several plants, among them Uehmann no. 5220. Dr. Schumann, in 

 PlantcB Marluthiaim (Engler, Bot. Jahrb. x. p. 41, July, 1888), 

 subsequently described as M. fjriqwmis Bolus, Marloth no. 1132, 

 from West Griqnaland ; but neither here nor in his monograph 

 does he cite any other plants upon which Bolus himself established 

 his species. I have carefully re-examined Burchell no. 2050 at Kew, 

 and still consider it inseparable from M. Relvnanni. The bracts of 

 the epicalyx in Burchell's specimen are not, however, linear-subu- 

 late, but ovate. Relnnann no. 5220, the type of M. Re/nnanni, is 

 among the plants referred to M. (jriqiiensis by Mr. Brown in his 

 note additional to Mr. Bolus's original description. 



Dr. Schumann does not retain M. BwchelUi DC. The type is 

 Burchell no, 2417, and is in the Kew Herbarium ; the leaves are 

 much broader than in M. prostrata DC. — the lamina is 7-9 cm. 

 long and 7 mm. to 2 cm. broad, and closely tomentose above, not 

 glabrescent. M. Damarana Harvey, a species omitted from my 



* Monographieen afrikanischer Pflanzen-Familien iind •Galtungen. Heraus- 

 gegeben von A. Engler. V. — SterculiaceaB Africanee. Bearbeitet von K. Schu- 

 mann. Leipzig: Engelmanii, 4to, pp. 140, tt. xvi. 



