359 



I. DUPATYA FLAVIDULA (Mx.). 

 Eriocatilon flaviditliim, Mx. FI. ii. i66 (1803). 



PcepalantJms flavidiihtSy Kunth, Enum. iii. 532 (1841). 



Dnpatya flavidiila, Kuntze, Rev. Gen. PI. 745 (1891). 



Leaves i to 2 inches long, three to five-nerved, linear-subu- 

 late, floccose at base and smooth or sparingly pubescent above. 

 Scapes numerous, five-sulcate, pubescent, 4 to 12 inches high. 

 Sheaths longer than the leaves, obliquely fissured, slightly inflated 

 at the summit, pubescent like the scape. Heads 2 to 3 lines in 

 diameter. Involucral scales straw colored, scarious, smooth, shin- 

 ing, oval or ovate, obtuse, somewhat hairy at base. Receptacle 

 pilose. Bracts very thin, white, linear, rounded or pointed at 

 the apex, about as long as the flowers, slightly hairy, often obsolete. 

 Flowers about \y^ line high, trimerous, long pedicellate. Sterile 

 flowers — outer perianth segments woolly at base, obovate or trun- 

 cate and pilose at the apex. This encloses the inner part of the 

 perianth consisting of a smooth, delicate, white ; campanulate. 

 Somewhat three- toothed tube ; stamens three, slightly exserted. 

 Fertile flowers — outer perianth segments distinct, hairy at base, 

 linear, acute, smooth, white, upper similar but much narrower, 

 enclosing the ovary and connate over it nearly to the top. Style 

 three-parted, forming three stigmas. Seeds sparingly and ob- 

 scurely costate when mature. Roots spongy, scarcely nodose. 



Kornicke (Linnaea, 27, 590) under the name Eriocaiilon flavi- 

 duluin, Mx., following Pursh (Fl. i, 92) and Elliott (Bot. ii 566) 

 states that two plants have been sent from North America under 

 this name and that he regards Kunth's P. flavidiilus as some- 

 thing distinct from the plant of Michaux. That which he de- 

 scribes is undoubtedly something distinct and is clearly an 

 Eriocanlon, but, so far as I can judge, it corresponds very nearly, 

 if not quite, to E. articidatum. The plant of Elliott is also, I 

 think, that species. Michaux distinctly calls his species piibern- 

 lent and the scapes aggregated and five-striate, while his other 

 characters correspond very well with our plant. There is not, 

 so far as ascertained, any other in the habitat given by him, 

 " Carolina " that bears such characters. 



Low sandy pine barrens, So. Va. to Florida. March-July. 



