PHANEROGAMS — PROFESSOR BAYLEF BALFOUR. 5 



1. P. longisiliqua, Dene, in Ann. Sc. Nat. ser. 2, iv. (1835), 69 ; Boiss. 

 Flor. Orient, i. 157 ; Oliv. Flor. Trop. Afr. i. 62 ; Fourn. in Bull. Soc. Bot. de 

 France xi. (1864), 56. 



F. stylosa, T. Anders, in Journ. Linn. Soc. v. (I860), Suppl. 1. 



Matthiola stylosa, Hochst. et Steud. in herb. Schimp. Arab. sect. i. m 860. 



Socotra. Common on the limestone plains. B.C.S. n. 584. Schweinf. n. 

 224. 



Distrib. Arabia, Nubia. 



Decaisne first briefly described this species thus : — " ramis foliisque incanis, 

 siliquis pedunculatis cernuis linearibus (2 poll, longis, 2\ lin. latis)," — from 

 imperfect specimens collected by Bove" on the plains of Yemen. T. Anderson, 

 when he wrote his florula of Aden, had not seen Decaisne's plant, but on the 

 strength of a remark by Walpers (Eepert. i. 139) regarding it, " non hujus 

 generis esse videtur," he made of the Aden plant a new species. Oliver (loc. cit.) 

 describes the species, adding " I have not seen type specimens of this plant ; the 

 name is taken from Schweinfurth's Nubian distribution. It appears doubtfully 

 distinct from F. Hamiltonii, Royle Illustr. 71." 



A type specimen is now in Kew Herbarium and, though it is imperfect, 

 enables me to determine that Decaisne's species is a good one. From F. 

 Hamiltonii, Royle, it is distinguished by its much larger flowers, pods, and non- 

 capitate stigmas. 



F. linearis, Dene., another African species, which Hooker and Thomson (in 

 Hook. Flor. Brit. Ind. i. 140), consider identical with F. Hamiltonii, Royle— (I 

 cannot agree with them, but with Fournier regard it as a distinct species, recog- 

 nisable by the longer, not capitate but bifid style and more slender stigma) — is 

 also separated from F. longisiliqua, with which it has close affinity, by its smaller 

 flowers and shorter and relatively broader pods, — characters which T. Anderson 

 also recognised as specific. A plant in Kew Herbarium marked " F. heliophila, 

 Bunge (Iter Persicum) " seems to me doubtfully distinct from our plant. 



The Socotran specimens are remarkable for their very large flowers. 

 These sometimes attain a length of over § of an inch. The plant is very 

 common on the plains, especially about Galonsir. Schweinfurth marks his 

 specimens " fl. livido." I found flowers varying in colour from a pink to that 

 livid or violet shade seen in, for example, the common Malcolmia maritima. 



2. F. prostrata, Balf. fil. in Proc. Roy. Soc. Edin. xi. (1882), 500. 



Herba prostrata ; foliis obovatis obtusis saepe apiculatis crassis strigosis ; siliquis linearibus. 

 Prostrata lignosa ramis cinereis pilis adpressis dense vestitis. Folia obovata obtusa ssepe 



apiculata crassa strigosa f-l£ poll, longa £— \ poll. lata. Alabastri oblongi. Flores 



breviter pedicellati. Sepala lineari-acuta strigosa. Petala dimidio calycem superantia. 



Stylus bifidue. Siliqua linearis | poll longa \ poll, lata ; stylum in fructu non vidi. 



Semina uniseriata compressa alata. 



