SMITH: STUDIES IN THE GENUS LUPINUS 3 
Vancouver and the adjacent islands have one isolated form, the 
Yakima Valley in Washington another. This latter, one local 
form of central Oregon, and one of western Nevada are the only 
ones peculiar to the country east of the Cascade-Sierra ranges. 
The more widely distributed form northward is apparently com- 
mon both to the waterways in Oregon east of the Cascades and 
also to southwestern Oregon and the northern counties of Cali- 
fornia. Central California is evidently the center of distribution 
of the group, and there many specialized and localized forms occur, 
from the coast well up into the Sierras. 
DIAGNOSTIC CHARACTERS 
Most of the characters ascribed by Sims and Bentham to their 
species are common to all members of the group, and cannot be 
used to distinguish between the two species. Agardh’s treatment 
of the species was mainly based upon the specimens seen by him 
in Lindley’s herbarium. A small-flowered plant collected by 
Douglas in California, as well as all the Chilean material there, he 
referred to Sims’s species. A more robust and longer-haired plant 
he accepted as Bentham’'s L. densiflorus, and this he acknowledged 
as having the aspect and nearly all the characters of L. microcarpus, 
sed robustior et bina Supe tits oabitie in Saher floribus obsitus. Ceterum 
petioli longiores, videntur. Unica autem dif- 
ferentia, quae ad speciem distinguendam adele videatur, e bracteis sumenda est. 
The yellow-flowered, short-pubescent form with long racemes 
of distinct verticels was set apart as his L. Menziesii. Three 
certain forms collected by myself in California were very readily 
determined as these three species of Agardh; but when I con- 
sidered other forms found by me, the original plates of L. micro- 
carpus and L. densiflorus, and also the conflicting determinations 
and opinions of various botanists, it seemed logical to conclude 
that the group needed special attention. 
Upon taking up the study, I soon found that reliable diagnostic 
characters are not as readily selected as I expected them to be. 
Many which at first recognition appeared to be all-sufficient for 
clear-cut distinctions finally suffered complete or partial abandon- 
ment after being carefully weighed. As a rule, the general aspect 
will appeal to the eye first and the general structure of the vege- 
