364 BOTANICAL GAZETTE [DECEMBER 
however, that so little is known about the sunlight relations of 
plants that attempts at close correlation are for the present un- 
profitable. 
TEMPERATURE (fig. 5).—Temperatures of both air and soil 
show similar seasonal variations, with high points in July, and a 
falling off in August and September, interrupted by a brief period of 
high temperatures during the first few days of the latter month. 
The figures up to July 1 can be accepted only as approximations, 
since they are absolute maxima and minima for periods of more than 
a week, instead of mean daily maxima and minima. For this reason 
too much significance must not be attached to the greater spread 
between maxima and minima. Although a greater spread did un- 
doubtedly exist, it probably was not so great as the thermometer 
readings at the ends of the periods would indicate. 
So far as any significance may be attached to the figures before 
the period of daily readings began, the influence of the diminution 
of general exposure through the leafing of the trees seems to be at 
work here also, for during May the maxima both for air and soil 
show as marked a spread as that shown for the warmer months of 
July and August. This spread is especially noticeable in the data 
for soil maxima," inasmuch as it amounts to eight or ten degrees 
throughout May, falls to a point at one of the June readings, and for 
the rest of the season never exceeds three degrees. These results 
agree fairly well with those of McDoucGa tt (14), who found a con- 
sistent seasonal variation of about 4° F. between the soil tempera- 
tures of typical upland and lowland stations in Illinois forests. 
Topographical differences seem to have some influence also, 
although the correlation is not so clear here as it is in the case of the 
atmometric and radio-atmometric data. It may be remarked, 
however, that the high lying stations (typified by no. 7) show the 
highest maxima both for air and soil, and the low lying stations the 
lowest maxima. On the other hand, the lowest minima for the 
air are obtained at the high lying stations and the highest minima 
at the lower ones. This may be due, among other things, to the 
denser leaf covering and the greater amount of underbrush at the 
* Because of the close agreement between the readings for all the stations after 
July 1, it is not thought worth while to present the soil temperature data in detail. 
