138 BOTANICAL GAZETTE [AUGUST 
5. It is not low temperature that kills invading mesophytes, 
but the edaphic physiological aridity prevailing in the central 
zone, which decreases the absorption of water by roots at a time 
when transpiration and the growth of the plants demand a greater 
physiological soil-water content. 
6. The topographical distribution of plants in the bog is also 
affected by relations in regard not to low temperatures, but to 
the uneven physiological water content and the physical condition 
in the peat substratum. 
THE DIFFERENCES BETWEEN AIR AND SOIL TEMPERATURES 
We proceed now to a brief consideration of the question whether 
_the differences between air temperature and that of the soil are 
sufficiently marked during the growing period to prove a factor 
in the selection of plants for bog areas. To show this relation, 
data on the corresponding minimum and maximum air temperatures 
for the period under investigation have been added to table VI. 
The records were taken from the Climatological Service of the 
U.S. Weather Bureau at Pataskala, some 35 miles (0.56 km.) 
distant from Buckeye Lake. They represent fairly nearly the 
conditions at Cranberry Island on the corresponding period. 
Additional data are found also in table VIII. Upon comparison 
it will be seen that during July and August, the months which 
proved most critical for the cultivated plants grown in the bog 
area for experimental purposes, the shoots of plants were in an 
atmosphere varying between 7° and 35° C., while the roots and 
rhizomes were at temperatures varying between 16° and 27° C., 
ie., within a range of temperature differences not less than 9° 
and not more than 19° C. For a growing season lasting from May 
5 to October 1, the average date of the latest and earliest killing 
frost, the actual differences between the temperature of shoots 
and roots amounted to 34°5 and 21° C. for each of the absorbing 
and transpiring organs respectively in the central zone, i.e., within 
a range of 29° C. It is seen that rapid and passing changes of air 
temperatures and the occasional extremes do not affect the sub- 
stratum temperatures. Only average effects prevail and the 
