160 Dr. Arnott's Notice of the Species of Salvadora. 



tine, and Egypt, are all the same; and, moreover, shows the accuracy of 

 Garcin's suggestion, that his plant was more likely to be found in Egypt 

 than in India. 



The only other species which requires notice is one from Senegal, called 

 by Zuccarini aS*. paniculaia, and on it I can throw no light whatever. I 

 am even ignorant where Zuccarini has published an account of it. Dr. 

 Wight considers it to be the same as Koxburgh's plant, but whether he 

 has seen Zuccarini's memoir, or judges only from the name, which is cer- 

 tainly more applicable to Roxburgh's than to the Arabian or Egyptian 

 species, I have no information. Delile says of his S. persica, that it is 

 found over the whole north of Africa, from Egypt to Senegal; and Guille- 

 min states in the Diet. Class. d'Hist. Nat. that he had received specimens 

 of S. persica from Le Prieur from Senegal ; although, therefore, more 

 information is required on this point, there appears to me a presumption 

 that the Senegal species is also the Egyptian one, and consequently that 

 the true S. jjevsica, or the plant of Garcin, extends from the Persian Gulf 

 to the west coast of Africa. 



The conclusion, then, which I draw is, that the S. persica of the Per- 

 sian Gulf and the countries to the west, and consequently the mustard- 

 tree of Scripture, is not the plant of Kcenig, Vahl, Roxburgh, Royle, or 

 Decaisne, which, to distinguish it, I shall call S. Koenigii ; but that it is 

 very nearly allied to S. persica of Wight, which for the same reason may 

 be designated S. Wighlii ; it differs chiefly from this last by the much 

 greater breadth of the leaves, and perhaps also by the larger fruit. 



Without giving any decided opinion as to what are species or what 

 varieties, (which I can scarcely do without having more materials at my 

 disposal,) I pi'opose to arrange the different forms of the genus as follows: — 



A. Corolla persistent; its lobes much longer than the short campanu- 

 late calyx, reflexed ; stamens protruded. 



1. S. persica (Garcin) ; leaves oval or ovate, racemes rather short with 

 close nearly sessile flowers. — Cissus arborea Forsh. S. paniculata Zucc. ? 

 — Hab. Persian Gulf, Arabia, Palestine, and northern Africa. 



2. S. Wightvi; leaves linear-oblong or narrow oblong-lanceolate, racemes 

 rather short with close nearly sessile flowers. S. persica WigU Icones, t. 

 1621.— Hab. Scinde. 



3. 8. Kcenigii; leaves from linear to ovate, racemes lax, flowers dis- 

 tinctly pedicellate. — 8. persica Vahl 8ymb. t. 4; Lam. 111. t. 81; Roxb. 

 Cor. t. 26; Royle; Decaisne in Jacquem. Voy. iv. t. 144, /. S. indica, 

 Wight III. t. 181. Embelia grossularioides Kmn. Embelia Burrnanni 

 Retz. Obs. Bot.fasc. 4. p. 23. Riviua paniculata Linn. — Hab. From 

 near Cape Comorin, at the south extremity of the peninsula of India, to 

 the northern Circars, and thence to Delhi and northern India. 



B. Lobes of the corolla short and erect (or deciduous ?) ; calyx cam- 

 panulatc. 



4. 8. Stocksii (Wight); leaves oval or ovate, racemes rather short, with 

 shortly pedicellate flowers. — Wight Icones, t. 1621. B. — Hab. Scinde. 



