Ilypoestes.^ xcviii. acanthaoe.*: (clarke). 251 



Smith ! Usambara ; Umba Valley, Smith ! Muoa, Hoist, 2987 ! 3075 \ Masbeua, 

 Hoist, 3492 ! 8858 ! Kwa Msbuza, Hoist, 9148 ! Usaranio ; Ukwcre district, 160 ft., 

 Stuhlmann, 8386 ! Ugalla River, Boehm, 279 ! Lower Plateau, nortli of Lake NvHsa, 

 Thomson 1 Lake Tanganyika, Cameron! Kavala Island in Lake Tanganyika, Carson! 

 Portuguese East Africa : Lower Zambesi : Lupata, Kirk ! Lower Valley of the River 

 Shire, Metier! British Central Africa : Urungu; Fwambo, Carson, 7jO\ !i3 I Sutt ! 

 Nyasaland; between Mpata and Tanganyika Plateau, Whyte ! Tanganyika Plateau at 

 Fort Hill, Whyte! North Nyasa, Whyte! Nyika Plateau, 6000-7000 tl., Whyte ! 

 Kondowe to Karonga, 2000-6000 ft., Whyte! Mount Chiradzulu, MeHer, 4000 ft., 

 Whyte! Shire Highlands, 5z<cAa??a», 116 ! 465! Mount Mlanje, WV/^e .' Rhodesia: 

 Leshunio Valley, Holiib, 816 ! 



Also in Arabia, Socotra and South Africa. 



I have been unable to sort this species into varieties. The plants named by 

 Lindau H. latifolia, Hochst. {Hoist, 8858 ; Volkens, 1652) are identical uitli those 

 named by him H. verticillaris, Solander (Hoist, 3492, 914-8), from the same 

 locality : but they are not exactly H. latifolia, Hochst., which is founded on Kotschy, 

 296, and has ovate glabrous leaves 4 in. broad. H. latifolia, Nees, has leaves rather 

 larger than usual (attaining 4 by 1§ in.), but not diftering in form from thost- of 

 H. ve7-ticillaris. H. mollis, T. Anders., from the Congo, is remarkably like the 

 typical Cape H. verticillaris. The most distinct form (included in the above li^t of 

 numbers) is Schiveivfurth's 128, collected in Gallabat (there are several examples like 

 it from the Mozambique region), named by Schweinfurth & Lindau H. latifolia, 

 from which it appears to differ considerably ; the leaves are narrowly elliptic, much 

 acuminate, the sepals are united scarcely \ their length, the 2 outer bracts are united 

 scarcely \ their length, and with many spreading long white hairs. Nees says (in 

 Linnaea, xv. 376) that this species " varies by having 4 didynamous stamens." 1 

 have never met with 4 stamens in any plant of the Justicioid series. 



Var. § glabra, S. Moore in Journ. Bot. 1880, 363. Almost glabrous, except the 

 inflorescence. Leaves attaining 1 by \-\ in. Bracts pubescent, and also witli some 

 scattered white longer hairs. 



XiO-wer Guinea. Angola, Welioitsch, 5059 ! 



S. Moore is probably right in regarding this as a depauperated state of H. vi ,fi- 

 cillaris ; but it is perhaps as well entitled to be made a species as any of the species 

 which follow. 



10. H. Preussii, Lindau in Engl. Jahrh. xx. 4S. Leaves up to f) 

 by IJ in., elliptic, long acuminate, pubescent on the nerves and with 

 scattered hairs, very shortly narrowed into the petiole ; otherwise as 

 H. verticillaris, R. Br. 



ITpper Guinea. Cameroons : Buea, 2500 ft., Preuss-, 'too \ 

 This certainly does not match any of the numerous examples above referred to 

 H. verticillaris, but it is difficult to say in what its specitic difference exists. Lindau 

 says the " bracteoles are adnate to the 2-toothed involucre " (which he states of some 

 other species of Hypoestes). There is no union between the bi act coles and bracts; 

 but, as in m'dny AcanfhacecB, the bracteoles cling or stick somewhat to the bract which 

 is external to them. 



17. H. violaceotincta, Limlau in Engl. Jahrh. xxiv. :V2'.\. In- 

 florescences axillary and terminal, very compound panicled. Spikelets 

 small ; two outer bracts hardly exceeding j in. in length ; otherwise as 

 H. vei'ticillaris, R. Br. 



Upper Guinea. Sieira Leone : Bafodeya, Scott-Elliot ! Togolai.d : Misahuhc. 

 JSaumann, 476 ! 



