(VoL. 3 
62 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
A soil thermograph was installed in the experimental plot; 
the bulb was placed six inches below the surface of the soil, 
and records were kept covering the growing period. "These 
show that the temperature of the blaek elay loam, such as is 
found in Racine and Kenosha counties in Wisconsin, 18 сош- 
paratively high, the minimum temperature of the soil rarely 
falling below the minimum for the air, and the maximum tem- 
perature of the soil, because of the lag, often exceeding that 
of the air, especially on cold or cloudy days (fig. 15). Тһе 
same relation between the main attack of the disease and 
temperature is apparent, although, because the high tempera- 
tures were maintained throughout July and August as they 
had been the previous summer, the percentage of disease also 
increased over a longer period than in 1912. The total per- 
centage of the disease in this year was less than in the pre- 
vious years, because plants from the resistant strains were 
counted with the control, the totals from the entire plot being 
used. 
Experimental results.—The experiments to show the rela- 
tion between temperature and the attack of the fungus were 
started in 1913. For this phase of the investigation the 
plants were grown in uniformly diseased soil in two different 
greenhouses, one of which was kept as near 25°C. as possible 
and the other at 15-20°C. Im the first experiment three flats 
of infected soil and one of greenhouse soil (uninfected) were 
placed in the warm house, and one flat of infected soil and 
one of uninfected greenhouse soil were placed in the cool 
house. The flats were planted on October 4, 1913, with two 
hundred seeds in each flat. The steam was turned on Novem- 
ber 18, at which time the plants in all the flats appeared nor- 
mal in their development. On November 25, however, yellows 
appeared in the flats of infected soil in the warm house. The 
plants in uninfected soil in both houses remained healthy. 
Figures 16-18 give an idea of the range of temperatures in 
the two houses for the entire period during which these experi- 
ments were made. 
On December 6 the above experiment was repeated; four 
flats of infected soil were planted, and two placed in each 
