—.* 
NOTES ON SALIX LONGIPES. 
winter-killed. The resultant is a remarkably different 
expression in external outline from that caused by the more 
siender, somewhat drooping withes of nigra. Fourth, 
the twig can not be snapped off at its junction, being in 
marked contrast with the exceedingly brittle base of the 
nigra twig. The well-matured twig is gray with a brown- 
ish tinge, and often. pubescent into the second year, while 
those less matured are reddish-brown and smooth. Fifth, 
the base of the leaf varies from the acute, through round, 
to the auricular form, the latter a feature not observed, to 
my knowledge, in any other willow. In nigra, the variation 
is at most only from the acute to the obtuse or roundish 
form. Sixth, the leaves are larger, the upper surface of a 
paler green; stipules larger, generally obtuse, seldom 
pointed, as they invariably are in nigra; aments mostly 
longer; fruit larger, discoloring a darker brown; pedicels 
longer; anthers a paler yellow. Seventh, S. longipes is 
exceedingly chary as to habitat, selecting almost invariably 
rocky, or at least gravelly portions of the stream, avoiding 
the alluvial stretches as if they were poisonous. This 
statement is confirmed by the independent observations of 
both Mr. Bush and Mr. Eggert, and also by Chapman. 
Nigra, on the other hand, occupies every situation or soil 
wherever it finds sufficient moisture. The former, though 
less hardy, appears to be a more vigorous grower. These 
appear to comprise all the important distinctions. 
It was supposed, and so stated, that the time of flower- 
ing is about 10 days (2 to 3 weeks in Silva) later than of 
nigra, but this year’s observation, in the vicinity of St. 
Louis, shows no material difference. A comparison is 
somewhat difficult, since this same season a variation of 
about 20 days in the blossoming of individual trees of nzgra 
was noticed. Nevertheless, it has to be admitted that so 
much in this matter of flowering depends upon the varying 
character of the seasons, that nothing short of a series of 
observations would prove conclusive. 
Again, in Mr. Bebb’s original description he puts the 
5 
