MH. G. BENTHAM OX EUPHOUBIACEJS. 243 



mmal or leaf-opposed, whilst in Exccecaria they are more gene- 

 rally axillary, and according to Mueller the seeds are constantly 

 strophiolate in Sebastiania and without strophioles in Exccecaria. 

 This character requires further investigation in many species; but 

 m the meantime it is better to follow Mueller's distinction than 

 to add a number of synonyms by transferring so many Sebas- 

 tiania to Exccecaria. With regard to the subdivision of Sebas- 

 tiania, the section Ditrysinia, the North- American 8. ligustrina, 

 with its larger less-divided calyx, connects in some measure Se- 

 bastiania with Stillingia, of which it has the habit but not the fruit, 

 and with Sapium. Microstacliys and Eusebastiania, as limited by 

 Mueller, and Sarothrostachys and Adenogyne of Klotzsch, forming 

 part of Mueller's Gusso?iia,are well-characterized sections ; but I 

 should exclude Gymnanthes from the genus. The species of 

 Exccecaria, which I should limit to Mueller's sections 5, 6, and 7, 

 have been much confused according to the state in which the 

 specimens have been described. Spirostachys of Sonder, forming 

 Mueller's Exccecaria qfricana and E. melanosticta, appears to me 

 to be identical with the typical E. agallocha, with the generic 

 character taken from the form of the male amentum when in 

 young bud ; and Baker's Stillingia lineata, var. densiflora, scarcely 

 differs from the same Exccecaria in foliage only ; it is very different 

 in inflorescence &c. from the true Stillingia lineata. Tcenosapium 

 is separated in the ' Prodromus ' from Exccecaria on account of a 

 lateral flattening of the style, a character which can scarcely be 

 more than specific. Sclerocroton ellipticum, Hochst., is certainly 

 not Sapium indicum, to which Mueller refers it, but, from our 

 specimens, is merely a different state of Sclerocroton reticulatum, 

 Hochst., or Exccecaria reticulata, Muell. Arg. A few American 

 species described by Mueller or by G-risebach appear further to 

 connect Exccecaria with Sebastiania ; but the specimens are as yet 

 insufficient to exhibit all their seminal or other characters- 11. 

 Gymnanthes, a genus of about ten tropical American, chiefly West- 

 Indian species, reduced by Grisebach to Exccecaria, by Mueller 

 to Sebastiania, appears to me to be well distinguished as originally 

 proposed by Swartz. The male flower, as in the four next 

 following genera, is absolutely without any perianth ; or the calyx 

 is very rarely represented by a minute scale, which may be a bract. 

 One species {Exccecaria polyandra, Griseb.) has been rejected from 

 the order by Mueller, he having once fouud a hermaphrodite 

 flower. On carefully looking over the flowers of our specimens, 



