•l 



48 APPENDIX. 



J 



on the peculiar features of each of these groups will be found in 

 the pages referred to. 



I now proceed to review in succession the value of the cha- 

 racters selected as the discriminating marks of the subtribes^ in 

 the arrangement followed by M. Dunal. There does not appear to 

 me sufficient reason for separating the genus Triguera as a sub- 

 tribe distinct from the Solanea. It is certainly a well-marked 

 genus, possessing prominent characters^ and differs only from 

 the other genera of the latter subtribe in the slightly oblique 

 form of its bell-shaped corolla ; but^ like others of the Solane(B, its 

 border has five equal and regular lobes^ and agrees with them in 

 aestivation j it has also five equal stamens^ supported on a ring 



as m Cyphomandra, but this ring is more free from the tube of 

 the corolla ; its anthers open by apical pores, as well as by lateral 

 slits, as in some sections of Solanum j in the structure of the 

 ovarium, its style and stigma, in its fruit, its placentation, its 

 seed, and its embryo, there is nothing different from what we 

 frequently meet with in Solanum itself, M. Dunal, on the author- 

 ity of Cavanilles, states the fmit to be 4-locular, each cell pro- 

 ducing only two seeds, which are superimposed. I found the 

 fruit to be distinctly 2-locular, being divided by a single mem- 

 branaceous dissepiment, with two or three seeds in each cell 

 fixed, as in Solanum, to fleshy placentse adnate to the dissepiment! 

 The seeds are reniform, compressed, large in proportion to the 

 size of the fruit ; but their paucity in each cell is a test of no 

 value, for I found in Withania only a solitary seed perfected in 

 each cell. There is not therefore a single character in Triguera, 

 except the small obliquity of the tube of the corolla, that is not 

 met with in other genera of the Solanea*. 



Among the subtribes Solanece and Atropinece of M. Dunal, we 

 find genera placed heterogeneously together, without regard to 

 uniformity of character, and totally irrespective of the most im- 

 portant feature of aestivation. Thus, among the Solanea, which 

 possess a valvate sestivation, is placed the genus Nicandra, with 

 a corolla resembling that of a Convolvulus, the lobes of its border 

 possessing a decidedly imbricated sestivation. The want of at- 

 tention to this last-mentioned important character has in the 

 same manner, led to the confused association of several genuine 

 sections : thus, among the Atropines, we find the very natural 

 group of the Jaborosea, distinguished by a tubular corolla, which 



* I have obsen-ed in several other cases an equal degree of obliquity in 

 the corolla. Among these may be mstanced Hyoscyamus pictus where it 

 IS quite as oblique and gibbous as in Triguera : the same fact is depicted in 

 the plate given of Hyoscyamus mger, in Nees's Gen. PI. Fl. Germ. figs, b, 

 6 and 7. ^ 



