Piate LXXVI. 
VIOLA arzorescens, L. 
Natural Order VIoLACEz. 
Gen. Cuar.—See description of Plate LIV. Part III. 
Spec. Coar.— lowers round in outline; peduncle glabrous or puberu- 
lent; bracts exceedingly minute. Sepals triangulari-lanceolate, acuminate, 
membranous and ciliate at edge, appendages about } length of limb, 
subtriangular, irregularly toothed. Petals, the upper two very broadly 
obovate, on short claws; the two lateral broadly and obliquely obovate, 
beardless; lowest petal ovate, shorter than the lateral petals, entire or 
subemarginate at apex, spur short, rounded, scarcely exceeding calyx- 
appendages, hairy within. Stamens, the two lowest ones spurred, spur 
short, decurved, linear-compressed, bent back against one of the anther 
cells. Stigma rounded, orifice very small, scarcely projecting. Capsule 
oblongo-obtuse, glabrous. eaves linear-lanceolate, sometimes having a 
short tooth on either side above the middle, glabrous or puberulent;_sév- 
pules linear, puberulent. Stems, almost woody, forming dense tufts 
decumbent below. Plant perennial, flowering in October, destitute of 
stolons. 
Viola arborescens, Linn, Sp. Plant. p. 1825; Gren. et Godr. Fl. de Fr. 
i. 182; Woods, Tour. FI. p. 40. 
Haszitat.—St. Cyr (department Var), on the neck of the promontory 
known as Les Baumelles, where I gathered this plant in profusion, and in 
excellent condition, on October 28th, 1868. 
Remarks.—-In order to find the nearest relations of this curious and 
rare violet, we must look for Viola decumbens, Linn. fil. at the Cape, and 
V. arborea, Forsk. in Yemen, both of which closely resemble V. arbores- 
cens, L. In France, besides the St. Cyr station, V. arborescens grows near 
Toulon (Gren. et Godr.) ; on the Montagne de la Clappe, near Narbonne ; 
and at Ste. Lucie. In Spain it is found near the shore along the whole 
eastern line of coast, attaining an altitude of 1000 to 2500 feet ;* also 
in Portugal (Nyman) and in Algiers ; while a marked variety having much 
broader and many toothed leaves (y. serratifolia of De Candolle’s Pro- 
dromus, a synonym of V. suberosa, Dsf.) grows also in Algiers, and near 
Mogador and Tetuan. 
I am puzzled to account for the fact that M. Boissier, and all the col- 
lectors in Spain and Algiers whose specimens I have seen, give March or 
April as the date of flowering, and a specimen in the Kew herbarium 
* Boiss. Voy. Esp. ii. p. 71. 
