322 CLVi. CYPERACEiE (clarke). [Cj/perus. 



Upper Guinea. Gambia, Mackenzie- Skues ! Ni»er Territory : Quorra 

 (Niger) River, Vogel, 19 ! Nupe, Barter, 620 ! Gold Coast : Accra, Don, 7 I Ansell I 

 Vogel ! 



:bower Guinea. Lower Congo : Maladi, 150-300 ft., Hem, A, 210 ! Gaboon: 

 Jardin ! Congo Free State : Kisantu on the Inkissi River, Gillet, 1289 ! Angola 

 Pungo Aridongo ; Condo Quisonde, 3500 ft., Welwitsch, 6903 ! Huilla, near 

 Humpata, 6000 ft., Welwitsch, 6904! Newton, \4.\ German South-west Africa : 

 Dammaraland, Een ! 



Soutb Central. Congo Free State : Lusambo, Duchesne, 18 ! 29 ! 



XtEozaml). Dist. Zanzibar, Stuhlmann, 94 ! Taylor I British Central Africa : 

 Urungu ; Fwambo, Carson^ 20 ! 



Also in South- Africa, where this type form is rare. 



1 have described above the Upper Guinea plant, the type of Vahl, and distinct 

 enough from C. compactus, Lam., by the broad flattened spikelets, few to a 

 head. 



Var. pseudonivea, C. B. Clarke. Spikelets 3-13 to the head, ^ by ^—1 in., 

 somewhat compressed, but turgid. — C. pseudoniveus, Boeck. in Verhandl. Bot. Ver. 

 Brandenb. xxix. 45. 



JmO-wbt Guinea. German South-west Africa : Amboland ; Olukonda, Schinz, 

 376 ! Dammaraland, Uen ! 



AXozaml). Dist. British Central Africa : Urungu ; Fwambo, Carson, 67 ! 

 Nyasaland ; Manganja Hills, Meller ! and without precise loeality, Buchanan, 1424 ! 

 1425! 1451! Rhodesia: Zambesi Valley ; ^oxumo., Menyharth, 1172! Leshumo 

 Valley, Eoluh ! 



Frequent in South Africa. 



The type of Boeckeler's Cyp. pseudoniveus is Schinz, 376, which Buchanan, 

 1424 ! 1425 ! matches. Several of the numbers cited for this variety were formerly 

 referred to C. compactus, Lam., and I see no good line between this var. pseudo- 

 nivea and C. compactus. 



14. C. ochrocephalus, C. B. Clarke in Durand d- ScKinzy 

 Conspect. Ft. Afr. v. 571, not of Steud. Glabrous. Rhizome seen 

 9 in. long by \-\ in. in diam., several arising from one point, cylindric, 

 perfectly straight, rigid, spreading on the ground, and rooting beneath 

 nearly their whole length" covered above by shining scales J- J in. long. 

 Stem 1 from the end of each rhizome, 4-12 in. long. Leaves 4 by J 

 in. Head 1, J-J in. in diam., of 50 spikelets, yellow; bracts 3, the 

 lowest 1-h in. long, similar to the leaves, dilated at the base. Spikelets 

 (young) \-\ in. long, 3-1-fiowered ; lowest flower 2-sexual, second 

 flower male. Glumes elliptic-oblong, much inrolled, smooth. Style- 

 branches 3, long, linear.— K. Schum. in Engl. Pfl. Ost-Afr. C. 122 ; 

 Durand & Schinz, Etudes Fl. Congo, i. 291. C. ohtiusifloms, var. 

 stylo 2-fido, Ridley (errore) in Trans. Linn. Soc. ser. 2, Bot. ii. 132. 

 Rynchospora ochrocephala^ Boeck. in Flora, 1879, 5C8 ; Ridley in 

 Trans. Linn. Soc. ser. 2, Bot. ii. 164. 



Soutb Central. Lunda : Kimbundu, Pogge, 412 ! 



This extraordinary species K. Schumann proposes now (I.e.) to arrange in Mariscus, 

 probably because of the few flowers to the spikelet. I have seen no ripe spikelets 

 and leave it here for the present ; the rhizome is exceedingly like that of Cyp. 

 angolensis. 



I 



I 



