lO 



Muhlenbergia, Volume 7 



back and parallel above, flaring below; wings strongly inflated, 

 open and exposing the keel for nearly its whole length; keel but 

 moderately curved, ciliate toward the purple apex: mature pods 

 25 mm. long, 5 mm. wide, 8-seeded: seeds 2.5 mm. long, 2 mm. 

 wide, pale, marbled and mottled with deep brown. 



The type, in the herbarium of A. A. Heller, xs Heller 662'/^ 

 collected April 25, 1903, on sand hills near the Marine Hospital, 

 San Francisco, California, distributed as Liipimis micraiitJms. 



In 1907 I again collected this species in sand near Ocean 

 View, San Mateo county, and distributed it as L. bicolor^ sug- 

 gesting in Muhlenbergia 2: 292, that San Francisco "may really 

 be the type locality," basing my opinion on Piper's remark (Cont. 

 U. S. Nat. Herb. II: 353) that "the species seems to be com- 

 mon in California, and extends into Oregon, but no specimens 

 from Washington have been seen." But this theory of mine 

 would have to be abandoned, even without having re-collected 

 bicolor at Vancouver, Washington, the type locality, for Douglas 

 did not visit San Francisco until four years after it was pub- 

 lished. 



Fi.t!;ure 4. Lupimis sabic/osus, flower and calyx, natural shape 

 Between 1903 and 1907 I had studied the description and 

 plate of L. bicolor^ and came to the conclusion as did Professor 

 Smith, recently expres.sed in this journal (6: 137), that the sand 

 loving plant of the San Francisco peninsula is identical with L. 

 bicolor. I still held that view until a few days ago, and had in- 

 cluded it under my notes on bicolor iox this paper. But a care- 

 ful study of the dissected floral parts of the two shows them to 



