tort] FULLER—EVAPORATION AND SUCCESSION 199 
This association is unique in the dominance of conifers, but is 
also notable for the comparative abundance of its undergrowth, 
although many species have decidedly xerophytic characters. 
That it is a comparatively short-lived association is evident from the 
presence of seedlings of Quercus velutina, the dominant tree of the 
succeeding association, very early in its history. Comparing the 
graph of its evaporation with that of the cottonwood dune (fig. 4), 
it will be seen that it is much lower, never reaching 20 cc. per day, 
and is subject to less violent fluctuations. Its maxima and minima 
are nearly synchronous with those of the cottonwood dune. The 
maximum evaporation rate is 17.5 cc. per day, the minumum falls 
below 4 cc., and the average for the season of 178 days is 11.3 cc. 
daily. 
Proceeding inland from the lake shore, the pines gradually 
decrease in numbers, and the black oak, Quercus velutina, becomes 
more plentiful, until at a distance of about 500 meters south of 
the last station it forms an almost pure stand with only occasional 
trees of white oak, Quercus alba. The shrubby undergrowth con- 
sists principally of Prunus virginiana, Rosa blanda, Viburnum 
acerifolium, Vaccinium pennsylvanicum, Ceanothus americanus, and 
seedlings of Quercus velutina and Q. alba. Among the herbaceous 
members of the association are Smilacina stellata, Lupinus perennis, 
Tephrosia virginiana, Lithospermum canescens, Asclepias tuberosa, 
Helianthemum canadense, Polygonella articulata, and Aster linarii- 
folius. In this oak dune association four stations were placed 
within a range of 100 meters; no. 6, on a fixed dune 15 meters high, 
well covered with the oak forest; no. 7, on a slope at an altitude 
of about 8 meters; and nos. 8 and 9, on the general floor of the forest 
some 5 meters above the level of the lake waters. All were about 
equally exposed and shaded. No. 6 was established on May 6, 
and the other stations on May 26. Station no. 9 was subject to 
so many interruptions that no report of its evaporation is presented, 
but the graphs from the other three (fig. 2) show a very close agree- 
ment, with differences corresponding directly to their elevation. 
A maximum of nearly rg cc. per day occurred in May during 
the second week of the record, before the trees were in full foliage. 
The absence of leaves would largely account for this excessive rate, 
