286 OBSERVATIONS ON THE NATURAL FAMILY 



Polygamia segregata, from finding that this had been 

 already done by Solander, whose name {Cartodium) , how- 

 ever, he did not think it necessary to adopt, and with 

 whose generic character he probably was not acquainted. 

 In his own he very erroneously states that there is no par- 

 tial involucrum, and hence perhaps M. Labillardiere entirely 

 overlooked Craspedia when he established his Bichea from 

 a nearly related species of the same genus. That such is 

 the case I have long since briefly noticed,^ and have ascer- 

 tained by a comparison of the specimen of Craspedia 

 uniflora in George Eorster's Herbarium with Bichea glauca 

 of Labillardiere, and other species of the same genus which 

 I have observed in New Holland. 



M. Labillardiere's character of Bichea is essentially cor- 

 rect. Tt is well to remark, however, that his general invo- 

 lucrum is formed of the bracteae subtending and in equal 

 number with the outer partial capitula; and that the 

 general receptacle has no other palese than the analogous 

 bracteae of the inner capitula. It is the more necessary to 

 take this view of the structure, as I have found in New 

 Holland a nearly related genus {Calocephalus), which differs 

 from Craspedia and Bichea in the want of these bracteae, 

 as well as in the partial receptacles being without paleae, 

 and in the rays of the pappus being plumose only in the 

 upper part. I have also another genus of this tribe {Leu- 

 cophyta) from the same country, difl^ering from Calo- 

 cephalus in having a general involucrum consisting of a few 

 short bracteae, in the squamae of its partial involucra being 

 107] concave and bearded at top, and in the rays of its 

 pappus being plumose through their whole length, as in 

 Craspedia, from which it is distinguished by the want of 

 paleae on the partial receptacles, and very remarkably in 

 habit. 



I have selected the foregoing genera as having been 

 either published under diflerent names, or, as it appears to 

 me, unnecessarily subdivided. In this extensive class it 



' III Prodr. Flor. Nov. Holl. p. 555. 



