222 SMITH: STUDIES IN THE GENUS LUPINUS 
L. pallidus Brandegee, Zoe 4: 203. 1893. 
L. micensis Jones, Proc. Cal. Acad. I. 5:630. 1895. 
L. desertorum Heller, Muhlenbergia 2: at 7005... 
L. Agardhianus Heller, ibid. 7: 13. 
Heller has shown that Nuttall’s L. gracilis, of the Gulf States, 
has priority over L. gracilis Agardh, hence the second name of the 
above list is properly replaced by the last one given. Agardh very 
obligingly contrasts his two species on one and the same plate, and 
careful comparison of his two descriptions discovers no marked 
differences other than those shown by his drawings. The odd 
fact is, that he classifies L. concinnus in his group ‘‘B”’ (Pilosi) 
and his L. gracilis in his group ‘“‘E” (unlettered and unnamed, but 
plainly provided for in his arrangement). 
This group is primarily Californian, ranging from Monterey 
County to northern Lower California, just two forms having an 
extensive range into the Mexican Plateau region, as far north as 
southern Utah, east to southern New Mexico, and south into 
Sonora. 
The pubescence varies from long-villous to subappressed, and 
the habit from erect and simple to much branched with the lower 
branches widely spreading. Some of the axillary branches are 
occasionally reduced to mere racemes, and even more often to 
one or two axillary flowers, which develop normal pods and seeds: 
hence the character, ‘‘racemes . . . sessile in the axils,’ ascribed 
by Watson to his L. Orcuttii. The flowers are usually small 
7-9 mm. long, but conspicuously larger in a form of the San — 
Bernardino and the San Jacinto Mountains. The petals vary 
from white and yellowish to a bright blue or a rich reddish purple, 
variously designated by authors. Few other characters seem to 
be marked enough to justify their use in diagnoses, and I must 
admit that the six forms which I am able to recognize are so 
closely related to each other that their relationship seems to me 
to be best expressed by classifying them all as varieties of one 
species. I cannot appreciate color alone as of sufficient value to 
warrant specific distinctions, and the variable pubescence here, 
as in L. densiflorus, L. sparsiflorus, etc., seems to prohibit positive 
lines of separation. 
