DR. THOMSON ON THE GENUS HEMIOYMNTA, GRTFP. 127 



are five stamens alternate with the lobes of the corolla, and a four- 

 celled ovary with one solitary ascending ovule in each cell. The 

 style is twice bifid ; and the fruit, still immature, is rostrato-cuspi- 

 date and " calyce cupuliformi semicinctus." 



Lieut. Beddome's plant, however, differs from Grifiith's descrip- 

 tion in its alternate (not opposite) leaves, and is an undoubted 

 species of the genus Cordia, as now generally understood. Still, 

 the agreement of every other character is so complete, that I can- 

 not but think that Grriffith has been misled by imperfect speci- 

 mens to regard the subopposite arrangement of the leaves, so 

 commonly seen close to the cymose inflorescence of many species 

 of Cordia, as a constant character ; and I feel equally certain that 

 the supposed opposition of the leaves, and a certain general resem- 

 blance in foliage to Omelina, and in flowers to Tectona, further 

 led him to consider this plant Yerbenaceous, notwithstanding the 

 striking character of the division of the styles, to which he never- 

 theless directs attention as indicating an affinity with Cordia. 



Turning to the described species of Cordia, I find a plant de- 

 scribed by Wallich (Roxb. Fl. Ind. ed. Carey & Wall. ii. 329 ; 

 WaU. Cat. 897; C Wallichii, Don, Syst. ; D.C. Prod. ix. 479) under 

 the name of C. tomentosa, with which Mr. Beddome's plant should 

 be compared, not only on account of a general agreement in the 

 brief character assigned to it, but because it was described by 

 Wallich from Heyne's collection, and may therefore not impro- 

 bably have been -obtained from the province of Malwah. As no 

 specimens of Wallich' s C. tomentosa exist in India, I have sent a 

 small specimen of Mr. Beddome's plant to Dr. Hooker, with a 

 request that he should compare it with the Wallichian herbarium 

 and append the result to this note. 



Should my conjecture prove unfounded, there are two other 

 species of Cordia with which I think our Cordia will prove allied, 

 and with which it should be compared, before it can be considered 

 as new. These are, C ovalis, E. Br., ? D.C. Prod. ix. 479, and C. 

 amplifolia, Alph. D.C. Prod. ix. 481. This last is said to produce a 

 valuable timber, agreeing in this point with the Jubbulpore plant. 



There can be little doubt but that all these plants are closely 

 allied ; and the African origin of the last two affords an additional 

 proof of the curious resemblance of the flora of Western India to 

 that of Eastern Africa, as has already been pointed out in the 

 ' Plora Indica ' (Intr. p. 113), and which is strengthened by every 

 addition to our knowledge of the botany of the province of Mal- 

 wah especially. 



