132 BENTHAM ON PAPILION'ACE^ AND C/ESALPINlEiE. 



distinctive character,* it does not appear to me to be necessary 

 to break up any really natural group. In all those cases 

 where the general distinction between PapilionacecB and 

 CasalpiniecB is most decided, this character also is the most 

 evident; and although many Sophorece on the one hand, 

 and several LeptolobiecE on the other, approach mutually to 

 each other in point of aestivation, these two tribes are also 

 evidently allied to each other in many other points. The 

 only genus, where the aestivation has been hitherto observed 

 to be really variable or doubtful, is Leptolobium itself, 

 which may be considered in many respects as a connecting 

 link between the two sub-orders, besides tbat it is scarcely 

 yet sufficiently known to be assured that it is in fact a natural 

 genus. 



There is another secondary point of view in which a cha- 

 racter should also be considered, when relied upon for the 

 separation of large groups of plants, that is, its artificial merit 

 in assisting us in their practical arrangement; and for this pur- 

 pose, two great requisites are, freedom from ambiguity, and 

 facility of observation. The undue importance formerly at- 

 tached to easy and artificial characters appears, of late years, 

 to have induced some botanists to run into the opposite ex- 

 treme, and almost to prefer minute and difficult ones ; but 

 surely, when two characters are equally natural, a preference 

 should be given to the most evident and consequently the 

 most useful ; and here, it does appear to me, that the aestiva- 

 tion is at once the most natural and the easiest to ascertain. 

 Few indeed, if any, are the cases where the opening of the 

 bud will not at once give decided evidence of the aestivation 

 of the petals ; but among the embryos of SophorecB, Dalber' 

 giece and Cfssalpinieae, there are numberless species where it 

 would be difficult to say, whether the curvature is or is not 

 sufficient to distinguish them from Orthoblastae. 



The following are the characters by which I would distin- 

 guish the three great divisions of Leguminosffi : — 



• I reckon any aberration from the papilionaceous aestivation as non- 

 papilionaceous. 



