Dr. Arnott on some neiv or rai'e Indian Plants. 153 



After the fall of the segments of the corolla, its annular base 

 around the ovarium, bearing on its inside the persistent Ala- 

 laments and scales, presents so much the appearance of a cup- 

 shaped fleshy torus or disk (such as is to be seen in some spe- 

 cies of Celastrus), that for some time 1 felt disposed to view it 

 as such, and that the petals might be distinct and attached 

 to the back or edge of this ring ; but on examining several 

 flowers in various stages I can perceive no trace whatever of 

 an annulus, so long as the segments of the corolla were not 

 broken off. There is therefore no distinct disk ; but I am far 

 from denying it a torus, and in a sense somewhat different from 

 what usually exists in the Corolliflorae. In most of that group 

 of orders the petals are soldered together by their margins, 

 and the filaments of the stamens, although apparently epipe- 

 talous, are decurrent, and may have their insertion traced to 

 the same point as that of the corolla ; there is besides some- 

 times a hypogynous disk, or glands. In Microtropis, however, 

 the stamens are not at all decurrent, and therefore it is not 

 improbable that there is a disk or torus, with w hich the lower 

 part of the petals is completely incorporated, and that the sta- 

 mens and scales are inserted on the upper or inner surface of 

 the disk. Under this point of view Microtropis might be left 

 in Celastrinece, and the principal objection to &uch would arise 

 from the supposed torus being hypogynous, while in those 

 species of Celastrus in which a cup-shaped torus is to be seen 

 the torus is adnate to the bottom of the calyx, the margin 

 only being free. In Celastrinece, moreover, the tube of the 

 calyx is shallow and broad, in Microtropis it is small, if indeed 

 any can be said to exist, for the sepals appear almost quite 

 distinct. 



Nearly all the genera referred to Celastrinece and Ilicinece 

 require careful revision, several of them being much at variance 

 with the characters of the orders in which they are placed. 

 Professor Lindley inserts all the section AquifoliacecB of De 

 Candolle in one order, reserving Celastrinece for the others. 

 Mr. G. Don, in his edition of ^ Miller's Dictionary, or General 

 System of Gardening,^ refers Cassine, Nemopantlies, and a few 

 others of DeCandolle's Aquifoliacem to Celastrinece, and My- 

 ginda, Ilex, Pi'inos, and some new genera of Blume's to Hi- 



