136 PITTOKIA. 



typical species as a Loins; but it was just this character 

 upon which Mr. Beiitham placed most reliance w^hen essay- 

 ing to establish Hosaclda as a genus.^ At a later date he 

 abauJoiieil that as a feeble character, believing that he had 

 found better marks of a genus in the structure of the corolla. 

 He then claimed that Ilosackia might be known by the space 

 intervening between the claw of the banner and those of the 

 other petals^ by the partial coherence of the wings wnth the 

 keel, and their spreading away from it.^ He appears still 

 later to have become aware that not more than two or three 

 out of the twenty-five species of Ilosaclia he had seen, dis- 

 played the adnate and spreading wings ; and so he eventually 

 left the genus as distinguishable from Lotus by nothing but 

 the remoteness of the claw of the banner from those of the 

 wings and keel.' There is not a group except our first, in 

 which this character does not fail utterly in a greater or less 

 number of the species. Some authors in Europe have con- 

 tended for a genus Dorycninm, based on Linmean species of 

 Lotus, on account of such a difference in the insertion of the 

 petals, together with a difference in the shape of the keel at 

 its apex. l^Thether Donjcnium, resting on such characters, 

 can be other than an artificial genus may be left to the 

 judgment of European botanists. It is clear to us who know 

 them that our American Loti can never, be separated into 

 natural sections on such grounds. 



The typical species of our fourth group are rather numer- 

 ous, and bpar one mark not found in any of the others, 

 namely, an articulation of the pedicel with the common 

 petiole, by which joint the mature pod, usually short, few- 

 seeded and indehiscent, falls to the ground, calyx and all, 

 w^hen ripe. This character was discovered and first announced 

 by the present writer.* But he had not then duly considered 



' Bot. Keg. XT, under 1. 1257. 

 ^ Trans. Linn. Soc. xvii. 363. 

 ^ Genera Plantarum. i. 490. 

 *BnlLCalif. Acad, ii 145. 



