January, 1900. Bruncken— Distribution of Some Trees, etc. 39 
of growing between the swamps and river margins. Both are 
under water at certain periods of the year. But along the rivers, 
this period is considerably shorter ; the water covering the soil is 
running, and well mixed with air; the soil itself is a mixture of 
fine mud, sand and coarse gravel. On the other hand, in the 
swamps the water is shallow, stagnant and grows very warm dur- 
ing the spring and early summer. The soil is either a very fine 
muck or a mass consisting of imperfectly decomposed vegetable 
matter. Such soil is known to be very poorly aerated. The 
chief defect, therefore, of a positon in a swamp as compared with 
one along the river shore, seems to lie in the direction of air sup- 
ply. This may account for the narrow leaves. In the absence of 
a sufficient supply of oxygen to the roots, the plant is compelled to 
reduce the vigor of its life processes. This is done by reducing 
the leaf area, through which these processes are carried on. An- 
other striking peculiarity of all swamp-growing willows seems to 
point in the same direction. While the river willows are either 
trees or shrubs composed of a few stout shoots(5), the swamp 
willows of all species have a very large number of slender shoots, 
and are never trees. The advantage of this form of growth may 
lie in the increase of bark surface, through the lenticels of which 
oxygen is taken in. 
There is one more station for w^illows in this region which de- 
serves mention. That is the shore of Lake Michigan. As far 
as I am aware the two species so characteristic of places along the 
coast where dunes are found (S. adcnophylla and S. glaucophylla) 
are not found within the limits covered by this article. But more 
or less vigorous specimens of 6^. fluviatilis are found in many 
places along the sandy beach. In some places also .S'. nigra is 
seen, but seems to be apt to take places where clay from the steep 
bank (6) has been washed down on the sand, especially in spots 
where a spring keeps the ground wet. In similar places there are 
also a few specimens of 5^. hicida and 6". discolor, and I have 
found one 5^. alba. All willows along the lake shore are young 
individuals. This must be either because a change has recently 
occurred which gives them a chance to occupy a field from which 
they were formerly excluded, or because the conditions are so 
unfavorable that all invaders perish before they have become old 
and large. I am unable to say which reason is the true one. 
5. The willows aptly illustrate the difficulty of defining a tree. One of the 
commonest lorms is one where two or three stout, ascending shoots, each several 
inches in diameter, rise to a height of twentv feet or more. According to the rule 
adopted in Sargent's Silva, these would have' to be called shrubs; but who has the 
heart to refuse them the name of trees ? 
6. The shore of Lake Michigan in this locality is formed by a steep clay bank 
rising to a height of one hundred feet in some places. Between this and the water's 
edge there is a beech of more or less breadth, composed sometimes of coarse gravel 
or even boulders, but more generally of sand. 
