48 BOTANICAL GAZETTE [JANUARY 
ent. The entire absence of forest relics over most of the prairie 
makes it extremely improbable that the entire surface of the 
county was ever forested. The level till plains between the stream 
systems and the moraines were probably prairie even at the time 
of greatest forest advance. The immigration of the forest was 
restricted to the two lines of greatest physiographic diversity, the 
stream valleys and the moraines. 
e must now account for the removal of this large body of 
forest from the moraine, and for the persistence of the small 
remainder in a few outlying tracts like the one at Bur Oak Grove. 
Examination of the conditions in the grove will suggest the reason, 
which is substantiated by other observations elsewhere in the 
county. 
Along the western margin of the grove some of the ridges are 
still forested, while others are under cultivation. Examination of 
the vegetation along the roadsides on the cultivated ridges shows 
on some of them such typically forest plants as Aster Drummondit, 
Silene stellata, Hedeoma pulegioides, and others. It is evident 
from the flora that these ridges were originally forested also. On 
some other ridges these species are entirely absent, and the roadside 
vegetation consists of typically prairie species, as Andropogon 
furcatus, Sorghastrum nutans, Panicum Scribnerianum, Silphium 
integrifolium, Petalostemum violaceum, and Parthenium integri- 
folium. Evidently these ridges were originally prairie. By this 
method of observation of the relic plants, the exact boundaries 
of the grove can be determined. In this way it becomes evident 
that, in every case, those ridges which are or were forested are 
protected on the west by a conspicuous slough, while the prairie 
ridges extend west without interruption out upon the open prairie. 
Since the forested part of the grove is exclusively on the ridges, 
it is clear that the whole forest was protected on the west side by 4 
series of sloughs. The prevailing winds are also from the west, 
and prairie fires driven to the eastward by a west wind were unable 
to cross the slough into the forest. It may be concluded, accord- 
ingly, that prairie fires were the chief and probably the only agent 
in the removal of the forest from the moraines and other places 
where it was not properly protected by a water barrier. The grove 
