158 Dr. Arxott's Notice of the Species of Salvadora. 



probable that it merely escaped notice from being erect and short, as in 

 S. oleoides; and if so, the only distinction between this last and Dr. 

 Wight's new'species rests on the broad or narrow leaves. 



Now the question again arises, what is the true S. persica ? and which 

 of the Indian ones approaches it most? We have two principal com- 

 ])etitors for the honour, but before assigning the preference to either, we 

 must examine what Garcin himself says on the subject : — 



1st. The corolla which he describes in place of the calyx is " divided in 

 four lobes, which, as soon as they spread open, turn outward, and roll 

 backward on themselves;" the real calyx was probably small and concealed 

 by the revolute corolla. 2d. The stamens are about the same length as 

 the lobes of the corolla. 3d. " This plant is woody ; it gi-ows sometimes 

 into a tree, sometimes into a shrub, and sometimes into a bush;" " that of a 

 larger^ sort of a shrub is what it most frequently grows into; it produces a 

 number of boughs without order, and very tufted branches, which most 

 commonly hang down to the ground." 4th. The length of the leaves, 

 ■which varies ou the same branch, '' is generally from one inch and a-half 

 to two inches and a-half, and their width is from nine lines to an inch a 

 little below. the middle in each, which is the widest part." 5th. The 

 flowers " are small, and disposed in clusters ou the tops of the shoots." 

 6th. " The berry is of the shape and size of a gooseberry, of three or four 

 lines in diameter: at first it is of a pale green, then a bright purple, and 

 in its maturity of a dark red; each berry is supported on a strong thick 

 pedicle attached to a small bunch." The seed " is as large as a grain of 

 hemp seed, that is, about two lines in diameter, but sometimes less." Garcin 

 adds, " it delights in the hottest and driest places, such as those adjacent 

 to the Persian Gulf, and perhaps more so than palm trees, wherefore I 

 doubt of there being any growing in the countries that lie to the east of 

 the gulf, and accordingly I have met with none, either in the neighbour- 

 hood of Surat or in the kingdom of Bengal, where there are regular rainy 

 seasons every year. I should rather believe it is more likely to be found 

 in the deserts of Africa, on this side of our tropic, these being proper 

 places for it, and where it rains seldomer than in any other part of the 

 globe." The inhabitants of the gulf call the plant Tchuch. 



Now, all the Indian species of Salvadora may be divided into those 

 with short erect, and those with reflexed lobes to the coroUa ; and if, as 

 I conjecture, S. StocJcsii of Wight belongs to the former (Dr. Wight having 

 seen it only in fniit, in which state the corolla of S. oleoides is also scarcely 

 perceptible), then the only Indian ones that can be compared with Gar- 

 cin 's plants are the S. persica of Wight, and the S. p>ersiai of Roxburgh. 

 As to the former, Wight describes it with " narrow elliptic-lanceolate 

 leaves," but figures it with leaves liuear-oblong, or narrow oblong-lanceo- 

 late, (the largest being 1\ inches long and only 4i lines wide,) which does 

 not accord with the original species of the genus. Again, Wight says, 

 " the berries of the Persian plant are described as yellow or black, those 

 of the Indian one are red,"' while he himself does not know the colour of 



