Du. T. tiiomso:n' on two new genera of composite. 34^3 



tata brevitcr S-dentata. Antliera^ longe caudatae, eaudis longe 

 plumosis. Stylus basi bulbosus, breviter exsertus, ramis bre- 

 vibus oblorfgis, margine et dorso fimbrilliferis. Aehenium teres, 

 erostre, dense villosum. Pappus subbiseriulis, setis rigidulis 

 scabris. Frutex Himalaicus alternifolius j capitula ad apices 

 ramorum, corjmbosa. 



CaTAMIXIS BACCHAROI0ES, 11. Sp. 



Habitat in collibus subtropicis Siwalicis Himalayae occideutalis prov. 

 Garhwal. 



Ramuli teretes, adpresse sericei vel glabreseentes, dense foliosi. Folia 

 li-2-j)ollicaria, obovato-spathulata, petiolo piano basi dilatato semiam- 

 plexicauli, coriacea, glabra, grosse sinuato-dentata ; superiora ramosa ; 

 suprema ad corymbi raraos, bracteiformia^ linearia, integra, Corymbus 

 compositus. Capitula gracilia, vix |^-pollicaria. 



I found excellent specimens of tbis very remarliable plant in 

 -Dr. J. L. Stewart's collection of West Himalayan plants sent to 

 Kew some years ago. In arranging tbe Herbarium of tbe late 

 -Dr. Falconer, I found specimens Avitli leaves only, but no flowers ; 

 and Mr. Bcntbam bas been good enough to show me a specimen 

 HI his Herbarium, collected by Mr. Edgeworth, and ticketed 

 ^' BaccJiaris oUgantJia, DC, in Herb. Eoyle, in Prodromo omisaa." 

 Tlie original Eoylean specimen I have not succeeded in finding. 

 This specimen is in very young bud, far too immature to show 



the generic characters. All these specimens are doubtless from 

 the 



same region. 



The affinities of Catamixis point in many directions. Hence 

 the name (KUTafitlis^ mixture). Theligulate corolla would techni- 

 cally place it in Cichoracese, with which suborder, however, it has 

 evidently no close affinity. The habit is that of a shrubby 

 Saccharis, PlucJiea, or Pteronia, and the lobes of the style are 

 very asteraceous. The bulbous base of the style points rather to 

 Mutisiaceae; and I am induced, after long hesitation, to place 

 it there, from its very evident close relationship to Leucomeris, 

 also a West Himalayan genus of one species. In the number of 

 the florets, the long and feathery tails of the authors, the silky 

 achenes, and, to a great extent, in habit the two agree. The co- 

 rolla, however, is very diff'ercnt ; but that of Lcitcomcris, being 

 quite regular, is exceptional in the suborder, ^lionlii Leucomeris 

 DC removed elsewhere, on a general revision of Composita?, Cata- 

 "^ixis must necessarily follow it wherever it goes. 



I-INK. PBOC. — BOTAKT, TOL. IX. 2 A 



