Jones—Relation of Soil Temperature to Disease in Plants. 441 
evident, however, that more satisfactory results could he secured 
by the use of small unit tanks instead of those united in a perma¬ 
nent series (text fig. 3). These units, which are termed the 11 "Wis¬ 
consin temperature tanks’’, can then be assembled in any number 
Text Fig. ,3. Section of the improved Wisconsin temperature tank, 
showing construction and arrangement of apparatus. The soil cans (ca.) 
are suspended in the water from the rigid cover (co.). Water for cooling 
and filling the tank is run in through water inlet (wa.i.). The alternating 
110-volt heating current is carried through heavy wires (he.c.) from 
switch (sw.) to relay (re.) and thence to the heating unit (he.u.) en¬ 
closed in the water-tight copper tube (he.) The direct current operating 
the relay is carried from storage batteries through line (ba.c.) from 
thermostat (th.) to the magnet on the relay (re.). 
or arrangement desired to meet the needs of a particular experi¬ 
ment. Naturally much attention has been given to the various 
possible methods of maintaining a constant water temperature and 
also to the size and character of the pots or containers in which 
the experimental plants are grown. This has also included the 
matter of special methods for securing uniformity in soil moisture 
and aeration, as well as devices for covering or insulating the free 
