156 Mr. Bentham on the Structure and Aj^nities 



minosœ. Linnseus placed it next to Cicer, Jussieu between Ononis and Antfii/Ilis. 

 De Candolle in classing the Leguminosœ strictly according- to the embryo and 

 the germination, as far as known, places Aracliis amongst his Geoffroyœ, or 

 Papillonacecv with combined stamina and a straight embryo; at the same 

 time, well aware of the little real affinity between Aruchis and the arbores- 

 cent Geoffroyœ, he suggests the probability of its forming with l^oandzeia a 

 small distinct tribe. ï^oandzein has, however, since been discovered, both by 

 Ernest Meyer and by Decaisne, to have a curvet! embryo, and has been 

 removed to Phaseolecc, and Arachis remains alone amongst genei'a with 

 which it is connected by the single character derived from the embryo. 



The very obvious resemldance of the flowers with those of Stylosantftes 

 first led me to suppose that it might be better placed amongst Hedijsareœ, 

 where the peculiar foliage of Arachis is not uncommon ; and a closer ex- 

 amination of the structure of the organs of fructification in both genera has 

 furnished some remarkable peculiarities which seem still further to connect 

 them. 



The perfect flowers of Arachis, accurately described by De Candolle as to 

 the calyx, corolla, stamina, ovarium and style, have appeared to me to be 

 constantly sterile ; they have, indeed, a perfect ovarium with two or three 

 ovules, but I have always observed it to fall off with the calyx, and the 

 legumes, as far as I have been able to ascertain from dried specimens of 

 several species in various states, arise constantly from female flowers of a 

 very different structure. These have neither calyx, corolla, nor stamina, but 

 from between two bracteolse, similar to those which are found at the base of 

 the sterile flowers, proceeds a stift' rigid stipes or torus, which is speedily 

 reflexed and elongated, and is terminated by wiiat appears to the naked eye 

 a sharp point. Examined under a glass this point discloses at its extremity 

 a truncated, somewhat concave and dilated stigma ; and within it will be found 

 a cell, within which are lodged two or three ovoid, anatropous ovules placed 

 transversely one above the other. After fecundation, when the extremity has 

 nearly reached the grounil, it begins to swell, but remains continuous with 

 the stipes or torus, without any articulation even at the maturity, when the 

 leo-ume is usually broken oft' with more or less of laceration. 



The perfect flowers in Stylosanthes are precisely similar to those of Arachis, 



