60 ERYTHEA. 



■t— t— t-Pod turgid but not membranous-inflated, more or less coria- 

 ceous, sub-terete; flowers several to many, capitate or epicate. 



++ Somewhat glandular-viscid ; pod half 2-celled. 



Aragallus viscidus (Nutt.) Greene, 1. c. Oxytropis visdda 

 Nutt,; T. & G. 1. c; Gray, 1. c. A rare plant but excellent speci- 

 mens have been secured as follows: Wallace Creek, near Garfield 

 Peak, July, 1894, 669; in the same locality, July, 1898, No. 5016; 

 Gros Ventre River, Aug., 1894, No. 928. Specimens from Great 

 Falls, Mont., by R. S. Williams, No. 619, though not typical must 

 be referred here. 



■♦-*++ Neither glandular nor viscid ; pod nearly 2-celled. 



Aragallus gracilis. Tufted, the branches of the caudex short ; 

 scapes few to several, erect, rather slender, stouter in fruit, lightly 

 pubescent, somewhat striate, 2-5 dm. high (including the spike): 

 leaves several from the crowns, mostly erect, shorter than the scapes 

 (about half as long as the scape in fruit) ; leaflets numerous, 10-15 

 pairs, mostly oblong-lanceolate, from sparsely soft- pubescent to 

 lightly canescent, 15-25 mm. long; inflorescence spicate, elongated 

 and somewhat open in fruit; bracts rather conspicuous, often equal- 

 ing or exceeding the calyx, silky, the hairs sometimes short and 

 nigrescent; calyx short-cylindric, its tube scarcely twice as long as 

 its linear lobes, pubescence similar to that of the bracts; corolla 

 ochroleucus or more distinctly yellow, about 15 mm. long, and 

 twice as long as the calyx ; wings obliquely truncate or emarginate 

 at apex ; pod pubescent, oblong-ovate, about 20 mm. long, semi- 

 membranous, soon splitting the calyx, nearly two-celled. 



That this is a valid species I am convinced, but I was at first 

 inclined to refer it to the Oxytropis campestris spicata Hook, of T. 

 & G. Fl. N. A. I, 341. From doing so I am deterred by the 

 statements that in that var. the flowers and leaflets are somewhat 

 remote, and further by the fact that Dr. Gray, in his revision of 

 Oxytropis, Proc. Am. Acad. XX, 6, says, " var. spicata seems rather 

 to belong to 0. Lamberti." A. gracilis is certainly more nearly 

 allied to A. campestris. 



This species is not to be confused with any other in the Rocky 

 Mountains unless it be with A. alhijiorus, from which its less 

 tufted habit, its slender scapes, its smaller and more distinctly 



