46 MISSOURI BOTANICAL GARDEN. 
fluence of nigra. One specimen identified by Heller as 
nigra, has the leaves of nigra, but the fruit of longipes. 
Another specimen identified by him as longipes venulosa, 
has fruit like a specimen from Watson, Mo., collected and 
named S. nigra by Mr. Bush. Specimens from N. Mexico 
show greater departure from the typical forms, and appear 
to approach nearer S. nigra. Of course in a case like this 
it is impossible to make any statement with assurance, 
on account of the limited accessible material. Specimens 
from Apalachicola Bay and elsewhere in Florida agree well 
with Washington and Missouri specimens. Those from 
Washington have shorter pedicels on an average than the 
others here mentioned, the latter presenting in general the 
widest divergence from nigra and being the purest examples 
of S. longipes. 
While there appear no material differences between all 
the forms which have been united under S. occidentalis as 
exhibited in the United States and S. Wardi, Bebb, it can 
not be questioned that S. nigra, Marsh. is distinct. This 
distinction is presented to us by a number of prominent 
characters. First, 8. longipes, Shuttlw. is a much smaller 
tree, the largest observed being rather under 30 ft. high 
and 9 in. in diameter at base, though usually much smaller. 
S. nigra, growing on the same spot, may attain double this 
size or more. Second, the bark of the trunk and larger 
branches is deeply cross-checked and firm, not flaky and 
shaggy as is nigra when old. This, once seen, serves to 
distinguish it from all other native willows even when bare 
of foliage. Third, the intense, whitish glaucous under 
surface of the leaves, together with the usually tomentose, 
young, very leafy shoots (though tomentum of the 
twigs is almost equally marked in some examples of S. 
nigra), enforces recognition even at a distance, or from a 
swiftly moving train. The impression thus made is assisted 
by a massiveness of foliage due to a vigorous growth of 
young shoots from the preceding year’s branchlets, the 
ends of which are, at least in its more northern limits, 
4 
