COMPARATIVE VIEW. 157 



reduction to a single species i.s even confinncd hy the mul- 

 titude of varieties tliat exist ;^ by nearly the whole of these 

 varieties being destitute of seeds ; and by the existence of 

 a plant indigenous to the continent of India, producing 

 perfect seeds; from which, therefore, all of them maybe 

 supi^osed to have sprung. 



To these objections to the hypothesis of the i)lurality of 

 species of the Banana, may be added the argument referred 

 to as contributing to establish its Asiatic origin ; for we an! 

 already acquainted with at least five distinct species of 

 Musa in equinoctial Asia, while no other s})ecies has been 

 found in America ; nor does it appear that the varieties of 

 Banana, cultivated in that continent, may not etpially Ije 

 reduced to Musa sapientum as those of India : and lastly, 

 it is not even asserted that the types of any of those sup- 

 posed species of American Banana, growing without culti- 

 vation, and producing perfect seeds, have any where been 

 found." 



That the Bananas now cultivated in equinoctial Africa, m 

 come originally from India, appears to me equally probable, 

 though it may be allowed that the Ensete of Bruce'^ is 

 perhaps a distinct species of this genus, and indigenous 

 only to Africa. 



^ Musa sapientum, Uox. Corom. fab. 275. 



^ M. Desvaux, in a dissertation on the genus Musa {in Joiini. iJe Boianiqiic 

 appl. \o\. 4!, p. 1), has come to the same conclusion respecting the original 

 country of the cultivated Banana, and also that its numerous varieties arc 

 reducible to one species. In this dissertation he takes a view of the floral en- 

 velope of Musa peculiar to himself. The perianthium in this genus is generally 

 described as consisting of two unequal divisions or lips. Of these, one is divided 

 at top into five, or more rarely into three segments, and envelopes the other, 

 which is entire, of a difi'crent form and more petal-like texture. The en- 

 veloping division M. Desvaux regards as the calyx, the inner as the corolla. 

 It seems very evident to me, however, that the deviation in Musa from the 

 regular form of a Monocotyledonous flower, consists in the confluence of the 

 three divisions of the outer scries of the perianthium, and in the cohesion, more 

 or less intimate, with these of the two lateral divisions of tlic inner series ; the 

 third division of this series, analogous to the labellum in the OrchidciT, being 

 the inner lip of the flower. This view seems to be established l)y the several 

 modifications observable in the dillcrent species of Musa itself, especially in 

 M. super /ja or]{o\huv^h, {Plants of Coromaud. IJ, tab. -223), and in t lie flower of 

 Musa figured by Plumicr, {Nov. Gen. t. 3J'), but still more by the irregularity 

 confined to the inner series in Strelitzia, and by the near approach to regularity, 

 even in this scries, iu Kavenala (or Urania), both of wliicli belong to the same 

 natural order. •* Travels, vol. 5, p. 3G. 



