1921] NATURAL SCIENCES OF PHILADELPHIA. 495 



two-ridged within on anterior side, the orifice open; the two post- 

 erior lobes 2-2.5 mm. long, united about one half length, their free 

 portions upcurved-reflexed; the thrae anterior lobes 4-5 mm. long, 

 united at base; the corolla externallj^ finely glandular-pubsscent^ 

 internally slightly pubescent over bases of anterior lobes; white^ 

 within throat, especially median to each lobe, with a few violet 

 lines. Anther -cells divaricate, oblong, intensely violet, .5-.6 mm. 

 long, glabrous, opening throughout. Sterile filament sUghtly 

 exserted, flat, not wider distally, moderately bearded distally with 

 yellowish hairs. Capsule at least 6 mm. long, ovate-acuminate in 

 outline, glabrous, not seen mature. 



Type, sandstone and shale cliffs, Ozark, Franklin County, Ark- 

 ansas, collected in flower May 29, 1920, F. W. Pennell 10616; 

 in the Herbarium of the New York Botanical Garden. 



Along river cliffs in southern Ozarks of western Arkansas. Dis- 

 tribution should be studied, and constancy of characters of dis- 

 tinction from P. arkansamis noted. These seem to be well-marked. 



Arkansas. Franklin: Ozark, E. J. Palmer 8143 (M), Pennell 

 10616 (A, C, H, J, K, L, M, O, P, S, T, U, Y). Also seen from Frisco 

 Springs, "Northwest Arkansas," Wells 49 (U). 



16. Pehstemon pauciflorus Buckley. 



Penstemoii pauciflorus Buckley, in Proc. Acad. Nat. Sci. Phila. 1861 : 461. 

 1862. "Past [=Post] oak woods south of Fort Belknap, [Texas, S. B. 

 Buckley, May, 1860-61]." Fort Belknap was in Young County on the 

 Brazos. The description inapplicablie to any Penstemon, and at variance 

 with the plant to which the name is now applied in these points: segments 

 linear-subulate, corolla-tube constricted to scarcely a line in width and 

 sterile filament glabrous with cilated apex. Fortunately, Dr. A. Gray had 

 seen Buckley's material in the two collections to which he sent specimens, 

 the herbarium of the Philadelphia Academy, and that of Ehas Durand. 

 In his criticism of Buckley's proposed species [in Proc. Acad. Sci. Phila. 

 1862: 165. 1863] he tells us that Buckley confused, vvith a genuine Pen- 

 stemon, Phlox pilosa L., and that most of the description applies to the lat- 

 ter! Also that the Penstemon, "a slender form of P. puhescens," has a 

 sterile filament ''heavUy bearded," a feature particularly true of the pres- 

 ent species. The specific name should certainly apply to the Penstemon 

 component. As there appears to be but one species of this genus charac- 

 teristic of post-oak woods in Texas, and that one has these features, 

 the Penstemon element of Buckley could scarcely fail to have been the 

 plant now considered. As no Penstemon was supplied to the Academy 

 herbarium the iypQ must be the plant of the Durand Herbarium, but 

 either this admixture was cuUed out by Durand from his material, which 

 is now in the Herbarium of the Academy of Natural Sciences of Phila- 

 delpha, or else it is included in the portion of his herbarium which went 

 to Paris. Accordingly it seems best to redescribe the species. 



Stems scarcely clustered, frequently but one from a root, slender, 

 3-7 dm. tall, minutely pubescent over entire surface, green or purp- 

 lish. Leaves membranous, green, slightly paler beneath, finely 

 pubescent to nearly glabrous; the lowest narrowed to petiole-like 

 bases, the longest ovate to oblong, 3-10 cm. long, obtuse, crenate- 



