TROPICAL AMERICAN COMPOSITAE. 31 



all the features important for classification. It appears to be very 

 different from any species known and may well have published record. 

 It has somewhat the habit of a Si/mphi/opappus, but the pappus- 

 bristles are capillary and distinct to the base, showing no tendency to 

 become firm or to be connate into a ring. It may here be mentioned 

 that recent examination of Symphyopappus leads to the belief that it 

 is an exceedingly weak genus of very doubtful distinctness and little 

 taxonomic value. Its distinctions from Eupatorium, both technical 

 and habital, break down completely. 



E. sciAPHiLUM Robinson, Proc. Am. Acad. liv. 256 (1918). This 

 interesting species, originally collected in Dept. Antioquia, Colombia, 

 by Kalbreyer, appears to have been rediscovered by Dr. Pennell in 

 the shrub zone, alt. 2800-3000 m., below. Paramo de Chaquiro, Dept. 

 Bolivar, Colombia, no. 4349 (X. Y.). Dr. Pennell's plant while 

 agreeing with the original material in all the more essential points 

 differs in having the leaves crowded, somewhat smaller (about 4 cm. 

 long) and less rigidly coriaceous, the contour being elliptic-oblong 

 rather than spatulate-oblong. In the presence of close agreement in 

 the inflorescence, involucre, florets, achenes, pappus, pubescence, 

 etc., it is probable that these dift'erences are largely the result of indi- 

 vidual environment or are at most only of a formal nature. 



E. sciATRAPHES Robinson, Proc. Am. Acad. liv. 257 (1918). Among 

 the plants recently collected in Venezuela by Prof. H. Pittier is a 

 Eupatorium in bud. It is labeled as follows: "Xiquibao. Fls. pale 

 lilac. Medicinal, pectoral. Around Caracas: altitude 800 to 1000 

 meters. Cultivated. June, 1918. //. Pittier, no. 7882." In habit, 

 foliage, pubescence, as well as in the details of inflorescence and florets, 

 so far as shown, this plant so closely coincides with the Santo Domin- 

 gan E. sciatraphes that its specific identity seems practically assured. 

 As in the original material the number of florets in the closely fastigiate 

 heads shows considerable variability, ranging in the Venezuelan 

 material from 4 to 8, while in the Santo Domingan specimens exam- 

 ined it ranged from 8 to 13. More mature material of the Venezuelan 

 plant would be essential to prove with absolute certainty the 

 identity, yet the highly characteristic habit, texture, lucidity, and 

 nervation of the leaves, etc., give fairly conclusive evidence. It is 

 to be observed that the Venezuelan material is from cvdti\ated stock, 

 so there is as yet no evidence that the plant is indigenous on the 

 continent. 



E. (§ Eximbricata) simulans, spec, nov., ut dicitur herbaceum 

 vel fruticosum 1-2 m. altum; caule tereti griseo-brunneo juventate 



