SMITH: STUDIES IN THE GENUS LUPINUS 5 
group. The verticels are regularly crowded in L. luteolus and 
L. horizontalis, usually well separated in L. microcarpus and 
L. subvexus; but in L. densiflorus is found a full range of variations 
from densely crowded to quite remote. The floral bracts vary 
much in length, usually becoming reflex-withering as the flower- 
buds open; but in one variety of L. subvexus they are rather con- 
spicuous for their tardiness in reflexing. Their length seems to be 
too inconstant to be of much diagnostic value. 
With few exceptions, I have not been able to accept size and 
color of petals for specific distinctions, though such are abundantly 
used in the varietal characterizations. At the free edges above, 
the keel petals are usually ciliate, often densely so, and a tendency 
to be ciliate at the free edges below is more or less pronounced in a 
few of the forms herein designated as varieties; but in L. luteolus 
this ciliation below is as dense and prominent as that above, 
usually extends backward onto the claws, and seems to be worthy 
of special notice. Variation in size, shape, apex and base of 
banner; ciliation, length, and width, but not detailed form, of 
wings; size, shape, and curvature of keel—are all worthy of con- 
sideration in the delineation of varieties. ‘The pubescence of the 
calyx usually agrees rather closely with that of the peduncles, 
petioles, etc. The lower lip, I am persuaded, yields good varietal 
characters as to length, dentation, and inflation; but the upper 
lip, at least in many cases, varies too much within the flowers of a 
given raceme to be trusted for any distinctions. The presence or 
absence of bracteoles is usually a decided uncertainty; but their 
presence may be diagnostic in L. horizontalis and in Elmer’s L. 
glareosus. 
The size, relative thitkness, surface, color, and marking of the 
seeds—when more of same have been collected and properly identi- 
fied—will probably be found of value in the varietal, if not in the 
specific, classification; but in L. luteolus again, I find a seed char- 
acter seemingly distinctive and peculiar to that species. The 
mature pods show much variation in size, but otherwise give little 
evidence of deserving much consideration. 
PATHOLOGICAL MATERIAL 
Aboud specimens are occasional, and care should be exer- 
cised to avoid them in the designation of varieties. Field work 
