44 PROCEEDINGS OP THE AMERICAN ACADEMY 



three genera have been made from it, it seems incapable of division 

 even into well-marked subgenera. The principal available characters 

 are given by tlie pappus ; but this is apt to be reduced, or even to dis- 

 appear, in some specimens of almost every group. Pectidopsis was 

 founded on an awuless state of a not uncommonly l-2-a\vned sjiecies. 

 Pectidium, with the anomalous Pedis imberbis (of similar habit, but a 

 perennial), along with the two species of Heteropcctis, Avliich are 

 annuals with the ordinary Pedis habit, may be associated by the char- 

 acter of the two or three corneous awns. In P. pundata these are 

 not always completely smooth and naked ; an occasional upturned 

 denticulation has been observed. Allowing that in every section the 

 pappus may in the same species be sometimes obsolete or reduced to 

 a crown, the genus may be fairly well disposed of under three sec- 

 tions : — 



1. EuPECTis. Pajipus paucipaleaceus, vel pauci-aristatus aristis 

 setiformibus, nunc ex aristis et paleolis paucis vel definitis constans, 

 plerumque uniserialis. 



2. Pectothrix. Pappus (saltcm fl. disci) multisetosus, intequalis, 

 plerumque biserialis, setis interioribus validioribus quandoque aristi- 

 formibus inferne sensim latioribus nee vero paleaceis. 



(even the paleae longer) than the akenes. No. 688 of Fendler's Venezuelan col- 

 lection is a very similar plant, except for the sliorter peduncles, and may be 

 probably referred to the same species. The palesfi and awns of the pappus are 

 not rarely connate into a tube. 



* * Segmenta foliorum angustissima, integerrima, vclhinc inde lobulata. 



T. CORONOPIFOLIA, Willd. Euum. Suppl. CO (sine char.) ; Jacq. f. Eclog. i. 118, 

 t. 80. T. angustifolia, HBK. Nov. Gen. & Spec. iv. 194. There are cultivated 

 specimens in the Berlin herbarium, and the original of Humboldt at the Paris 

 Museum. In both, as I learn from comparisons kindly made for mc, tiie longer 

 palea3 of the pappus are about one third the length of the akene ; the " lonyiores" 

 in Kunth's description is therefore a mistake for breviores. I possess a single 

 indigenous specimen, collected by Graham in Mexico, which belonged to the 

 herbarium of John Stuart Mill. 



T. FiLiFOLiA, Lag. Nov. Gen. & Spec. 28. T. multijida, DC. The name of 

 Lagasca may somewhat confidently be restored for this low species, interme- 

 diate between the preceding and the following, winch better deserves the name. 

 Like it, tlie longer paleae of the pappus are strictly aristiform. Tlio divisions of 

 the leaves are more numerous, shorter, and less slender ; the heads not so 

 slender, mostly very short-peduncled, and disposed to be clustered. 



T. MiCRAXTiiA, Cav. Ic. iv. .11, t. 3o2. Tliis is an anisate-scented little spe- 

 cies, paniculately much branched, slenderpeduncled (except as to the later 

 heads) ; the attenuate-filiform leaves with only 3 or 5 divisions, or some of the 

 lowest and uppermost entire. It reaches the United Slates. 



