SMITH: STUDIES IN THE GENUS LUPINUS 393 
scattered, glabrous above or more or less hairy; peduncles variable 
in length, the short capitate or subcapitate racemes exceeding 
the foliage or surpassed by same; flowers crowded, I0O-II mm. 
long, commonly few; pedicels 1-2 mm. long, villous; calyx bracte- 
olate, villous, upper lip 4-7 mm. long, bifid, the divisions slender, 
lower lip 5.5-7 mm. long, three-toothed; petals blue or purple, 
narrow, banner 8-10 mm. long, 3-5 mm. wide, usually acute at 
apex, wings 6-8 mm. long, 2-3.5 mm. wide, oval or oblong, keel 
straight, 6-8 mm. long, about 3 mm. wide; pods rhombic-ovate, 
about 10 mm. long; seeds about 2 X 2.5 mm. 
Watson soon recognized the affinity of his supposed perennial, 
L. Kingii, and his annual, L. Sileri, and reduced the latter to 
synonymy. This viewpoint has been followed by Coulter (Man. 
Bot. Rocky Mts. 273. 1885), Coulter and Nelson (New Man. 
Rocky Mt. Bot. 469. 1909), Rydberg (Fl. Colo. 100, 1906; FI. 
a SS ag a 
fem 
3 1 2 See 5 
G. 44. LUPINUS Kincn Wats. 1. Korstion & Baker 131 (FS); 2. C. A. Purpus 
bie aay, 3. S. Watson 234 (US); 4. C. F. Baker 439 (US); 5. M. E. Jones 5641 
(US). 

Rocky Mts. 469. 1917), and Heller (Muhlenbergia 6:71. 1910). 
Heller and Rydberg have also referred Greene’s L. capitatus 
to Watson’s species. Wooton and Standley (Contr. U.S. Nat. 
Herb. 19: 340. 1915), however, have recognized L. Kingii and 
L. Sileri as distinct species, each represented in New Mexico, and 
have segregated L. argillaceus as a new species. My study of the 
specimens listed below prevents my recognition of more than one 
species here, though as least one variety probably deserves at- 
tention. I can not see specific characters in the variations in 
length of peduncles, or in the extremes of types of branching 
(widely divaricate to ascending) here; but the more sharply de- 
fined variation in pubescence seems to be of sufficient definition 
for use in varietal recognition. Greene attributes to L. capitatus 
