364 Limiean Society. [March 7? 



Melianthus being usually considered as an anomalous form of Zy- 

 gophyllece, the question presents itself, to what natural class this last 

 order is to be referred ? Is it also to be admitted simply as a tribe 

 of Rutaceas (as defined by the Jussieus), or shall we follow Mr. Ro- 

 bert Brown in considering it as an independent order ? The author 

 not only declares in favour of the latter opinion, but expresses his 

 belief that while, on the one hand, Diosmece (including Rutece, Dios- 

 mea proper, Zanthoxylece and Aurantiacece), together -with. Simarubece 

 and MeliacecE, constitute a natural class, so, on the other hand, Zygo- 

 phyllecE, Oxalidece, Connaracea, Legiiminosa and Moringea are closely 

 connected into one group, not only by their general structure and 

 fades, but by the common tendency of their compound leaves to 

 periodical sleep, or occasionally to movement under an irritating in- 

 fluence, a physiological phsenomenon connected with the structural 

 fact of the articulation of the foliole with the petiole on which it 

 moves. 



Neither of the two natural classes just mentioned admits, in the 

 opinion of the author, the new order of Melianthece. 



The pinnate leaves, irregular flowers, excentric and incomplete 

 disc placed outside of the stamens, the quaternary proportion * of 

 these organs in contrast with the quinary division of the calyx, the 

 occasional cohesion of two of the sepals, the close analogy of the 

 follicular capsule of Diplerisma with that of Cardiospennuin, and of 

 the coriaceous fruit and arillate seeds of Bersamece with the corre- 

 sponding parts mPaullinia, and the fact of a species o{ Natalia being 

 justly named PaMZ/i/zzo/c^e^, are the points by which the close affinity 

 of Melianthece with Sapindacece are traced out. Thus by the know- 

 ledge of very recent materials (^Bersama and Natalia being both but 

 lately discovered) are confirmed the views which Adanson expressed 

 upon the affinities of Melianthus, when, in his otherwise rather hete- 

 rogeneous family of Gerania, he placed that singular genus between 

 Cardiospermum and Geranium. 



After some other general considerations, the author concludes with 

 a review of the geographical distribution of Melianthece, the most 

 striking fact mentioned being the occurrence of Melianthus Hima- 

 layanus, Wall., in the mountains of northern India, while its only 

 congener, the well-kno-wn Melia?ithus major, L., does not exceed the 

 limits of the flora of the Cape of Good Hope. 



* Bersama must here be excepted, because of its five stamens. 



