Mr. J. Miers on the genus Duboisia. 439 



subsequently I pointed out the features that separate it from the 

 Scrophulariacece, and suggested its true position in the system 

 among the Atropacece, in the tribe Duboisiece {huj. op. iii. 165). 

 Since the description of the typical plant, now forty- three years 

 ago, no other species has been known, and that was called 

 D. myoporoides by Mr. Brown, on account of the similarity of its 

 habit to Myoporum. This genus, indeed, serves to connect the 

 Atropacece with the Myoporacece, as at present limited, through 

 Disoon, which has a monopetalous corolla with five equal seg- 

 ments, having an imbricated aestivation, with the same peculiar 

 involution of the margins, as in the Duboisiea. Like Antho- 

 troche it has didynamous stamens, with similarly formed anthers, 

 only that they are introrse : it has also a bilocular ovarium, but 

 each cell has only a solitary suspended ovule ; its fruit is also 

 drupaceous and bilocular ; of its embryo nothing is known : 

 should it even have a superior radicle, as is most probable, its 

 ordinal tendency would even then appear to lean more towards 

 the Scrophutariacece than to the Myoporacece. The same may 

 be said of Nesogenes, judging from the description given of its 

 structure. The chief distinction between the Scrophutariacece 

 and the bilocular section of the Myoporaceae consists in the dif- 

 ferent dii'ection of the embryo, but this character is of little 

 value, as it arises merely from the more pendent or ascending 

 position of the ovules, and in both cases the radicle points alike 

 to the hilum. We must remember that exceptional cases of this 

 kind occur in Scrophulariacecje, for instance in Leptorhabdos, 

 Melampyrum and Tozzia, which also have only two suspended 

 ovules, where sometimes only a single seed becomes perfected, 

 and where from its pendulous position the radicle is superior, 

 contrary to the usual character of the family. Under such views, 

 the ordinal tendencies of Leptorhabdos and Disoon appear to 

 point in the same direction, from which the baccate fruit of the 

 latter would not exclude it, because, although a rare occurrence 

 in Scrophulariacece, this does sometimes occur, as in Halleria, 

 &c. : in Atropacece it is more frequent. Consequently it would 

 be more consistent to refer to Scrophulariacece all the genera of 

 the Myoporacece possessing a bilocular ovarium, where the ovules 

 are attached to a simple dissepiment, and to confine the limits of 

 the Myoporacece to those genera where the dissepiment is so 

 greatly produced and introflexed as to produce four distinct 

 cells, and often other pseudo-cells. The latter, according to the 

 views of most botanists, offer a structure closely approaching 

 that of the Verbenacece and Borraginacece (the Echial alliance of 

 Prof. Lindley) : the former clearly belong to one of the orders 

 of the great Solanal alliance as above suggested : the distinction 

 in point of structure is considerable and manifest. In habit, 



