86 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
In three experiments, soil temperature alone was varied. The plants 
were grown in soil which was first sterilized and then inoculated with 
F. lycopersici. The temperature of the soil was controlled to secure a range 
from 14 0 to 35 0 C. The period of growth at these temperatures was four 
to six weeks. The optimum soil temperature for the disease was found to 
be about 28° C., though at 31 0 it developed almost as virulently. At soil 
temperatures of 33 0 or above and of 21 0 or below, the disease was practi¬ 
cally inhibited. 
Two experiments were conducted in which both air and soil temperatures 
were controlled. For this work the air in three greenhouse compartments 
was maintained at temperatures designated as cool (about 17 0 C.), warm 
(about 27 0 C.), and hot (about 33 0 C.), respectively. Three of the Wis¬ 
consin soil-temperature tanks were placed in each house and regulated to 
hold the soil temperatures, likewise, at 17 0 , 27 0 , and 35 0 C., thus permitting 
nine combinations of air and soil temperatures. 
Air temperature was found to be as effective in controlling the appear¬ 
ance of the disease as soil temperature. 
In only two of the nine combinations of air and soil temperatures did 
the disease make a rapid development, these two combinations being warm 
air (27 0 C.) and warm soil (27 0 C.), and hot air (33 0 C.) and warm soil 
(27° C.). 
If the soil was kept too cool (17 0 C.) or too warm (35 0 C.), the disease 
did not develop, even with optimum air temperature. 
If the air was kept too cool (17 0 C.) and the soil temperature was opti¬ 
mum (27 0 C.) for the disease, heavy infection occurred in the root and ex¬ 
tended up into the basal portion of the stem. The plants continued to grow 
thriftily, however, and there were no external symptoms of the disease. 
The temperature conditions of soil and air most favorable for the 
disease, as determined in tanks, are a soil temperature of about 27 0 , and an 
air temperature, after the fungus has established itself in the stem, of about 
28° C., with short periods when the temperature suddenly rises to an 
excessively high point (33 0 or 34 0 C.). 
Evidence accumulated during the course of these experiments leads to 
the belief that the wilting and death of plants attacked by the Fusarium 
wilt disease is due not to mechanical plugging of the xylem bundles, but 
rather to toxic action. 
New York Agricultural Experiment Station, 
Geneva, New York 
LITERATURE CITED 
1. Brandes, E. W. Banana wilt. Phytopath. 9: 339-389. 1919. 
2. Caldwell, J. S. The relation of environmental conditions to the phenomenon of per¬ 
manent wilting. Physiol. Res. 1: 1-56. 1913. 
3. Edgerton, C. W. Tomato wilt. La. Agr. Exp. Sta. Bull. 115. 1920. 
4. Free, E. E. Symptoms of poisoning in plants. Johns Hopkins Univ. Circ. 36: 394. 
1917. 
