MENISPEKMACE^. 23 



II. Pachygone^;. — Seed with a fleshy exalbuminous embryo. 

 (All other characters of Cocculece) (9 genera). 



III. Chasmanthere^:. — Seed with a thin albumen, concavo-convex 

 from within outwards, surrounding an embryo with thin foliaceous 

 laterally divaricated cotyledons. (All other characters of Cocculea 

 and Pachygonece (10 genera). 



IV. Cissampelide^e. — Flowers not regularly trimerous, their 

 anthers united on top of a common column. Carpels solitary excen- 

 tric (3 genera). 1 



In each of these series the genera are distinguished by the essen- 

 tially variable characters presented by the total number of verticils 

 of the perianth and androceum, and the presence or absence of a 

 corolla, the definite or indefinite number of the stamens, and their 

 freedom or union, the number of carpels, the form of the fruits, and 

 especially of the stones and their internal projection ; the straight 

 curved or hippocrepiform direction of the long axis of the seed 

 and embryo, the form of the albumen and its continuity or seg- 

 mentation into ruminate lobes ; the equality or extreme inequality 

 of the two cotyledons. 2 



The structure of the stems in this order has often been described 

 on account of its great anomalies. The variation often found 

 in matters of detail in passing from one genus to another, 

 do not seem to affect the general type of histological structure 

 usually ascribed to the stems of Menispermads. This structure is 

 such that Lindley was formerly inclined to separate this order from 

 the Exogens, recognising at the same time their striking analogies 

 with those of the Aristolochiads. Decaisne, taking up the study of 



1 These plants might he classed by their male all those observed in the flowers of both sexes, 

 flowers, and this is what Bent-ham & Hookeb And this has been done to some extent by Miebs 

 have tried to do (Gen., 32). But as yet, not who proposes (in Ann. Nat. Mist., ser. 3, xiii. 

 only have as many difficulties been found in 122) to divide Menispennacece into seven tribes: 

 rendering these characters practically available 1. Heteroclinece (nearly the same as our Chas- 

 es, with those drawn from the carpels, but their mantherece) ; 2. Anomospermece (the single genus 

 use is just as artificial. Thus, it is quite certain Anomospermum) ; 3. Tiliacorea (which we con- 

 that to separate, because of the seed-structure, fine to Tiliacora and Chaimantherd) ; 4. Hyp- 

 genera with fundamentally similar male flowers, serpece (Limacia) ; 5. Leptogoneee (including 

 sucli as Cocculus, Pachygone, and Hcematocarpus, Assampelideee and some of the Cocculea:) ; 6. 

 is a most artificial course. No less so is it, how- Pl-atygonece (formed chiefly of the genus Cucculus 

 ever, to group together from characters taken and its numerous sections which are raised to the 

 from the androcenm alone, genera differing so rank of genera) ; and, 7. Pachggonece (Pachygone 

 widely in the organization of the female flower and Sarcopetalum). 



as, for instance, Stephania, Sarcopetalum, and - Miees thinks, moreover, that the nervation 



Aspidocarya. Evidently we should view at the of the leaves in this order may form a differential 



same time both, not merely single characters character of great value, 

 taken from the flower of one sex alone, but of 



