THE FLORA OP AMOORLANP. 131 



whilst, at the same time, sectional divisions are amply sufficient for call- 

 ing attention to the greater importance of some of the specific distinc- 

 tions as compared with others. 



Plagiorhegma, Maxim., is, under our view, a second species of Jef- 

 fersonia, with which it forms a very well characterized natural genus. 

 The undivided leaves and more oblique dehiscence of the capsule than in 

 the original J. dipliylla, are excellent specific distinctions, but would 

 not, in any other instance, be admitted as generic. The flowers are un- 

 known; for the two loose ones picked up from the ground, and described 

 by Maximowicz, were found not to tally with the appearances of the 

 young capsule, and consequently were at first concluded to be abnormal, 

 but afterwards admitted (p. 460), to have probably belonged to some 

 other plant. We mention this as a caution to travelling collectors 

 against a not uncommon practice of hastily concluding that flowers or 

 fruits picked up under a tree or herb must have fallen from it — a source 

 of many an error in descriptive botany. 



Hylomecon, Maxim., " has the habits and flowers of Stylophorum, 

 with the fruit of Chelidonium" This fruit, however, the author had 

 not himself seen, but states it, on the authority of the natives, to be two 

 inches long, and slender — a very insufficient authority for establishing 

 a genus, when there is nothing in the enlarged ovary (as correctly 

 figured) to indicate this elongation ; for, might not the natives have con- 

 founded it with the true Chelidonium majus, stated to be very common 

 in the country ? It is true, that in Hylomecon, as well as in the nearly 

 allied Dicranostigma, Hook, and Thorns., from the Himalaya, the ovary is 

 2-merous, whereas in the two American species of Stylophorum, it is 

 usually 3-4-merous ; but even in these, the carpels are sometimes re- 

 duced to 2 ; and by uniting the Himalayan as well as the Mantschurian 

 plants with the American, we have a natural and well-characterized 

 genus of four species. 



Phellodendron, Rupr. belongs to a group allied to Xanihoxylon, but 

 differing chiefly in the opposite leaves, the scarcely imbricate or some- 

 times valvate corolla, and the superposed ovules. There are several 

 south and east Asiatic species, which, from having been usually com- 

 pared with Xanthoxijlon only, have been proposed as genera under the 

 names of Lepta, JBoymia, Philagonia, and Megabotrya ; but which, on a 

 general review, show that there is nothing in character or habit to se- 

 parate them (as well as the opposite-leaved species published as Xan- 

 thoxyla), from the old-established genus Evodia, with which no one 

 seems to have thought of comparing them. 



Maackia, Eupr. differs from the North American Cladrastis, Eaf., in 

 its small crowded flowers, with a shorter calyx, more obtuse at the 

 base, and rather shorter pod ; the other characters, the foliage and habit, 

 are nearly the same ; and although a very distinct species, we cannot see 

 sufficient grounds for separating it generically from the rT. American 

 plant. The stamens are figured as monadelphous ; they are, however, 

 in a flower we examined, free as described. 



Sckizopepon, Maxim., a cucurbitaceous plant, cannot well be judged 



