74 
Journal of Agricultural Research 
Vol. XXIV. No. I 
mean soil temperature of about 25® to 20 days with a mean soil tem¬ 
perature of about 20°. It has also been repeatedly observed (ij, 8 ) 
that the plants which escape or survive the disease for the first month 
may remain healthy during the remainder of the season. 
These observations gave sufficient evidence to prove that soil temper¬ 
ature is an important factor in the development of yellows in cabbage 
plants transplanted into “sick’' fields. However, they do not satisfac¬ 
torily show what results might be obtained with younger seedlings 
planted at intervals throughout the season. Data upon experiments 
of this nature were considered of distinct value for comparison with 
results obtained in the greenhouse at different temperatures. There¬ 
fore, experiments were begun in 1917 for obtaining such data. Both 
the resistant Wisconsin Hollander and the susceptible Commercial Hol¬ 
lander were used. By beginning to plant early in the season and con¬ 
tinuing until late in the summer, it was possible to have each successive 
crop of seedlings exposed to a different range of soil temperature. It 
was the original plan to plant the seed at intervals of 7 to 14 days and 
record the final data 20 days later, as was done in the greenhouse, but 
for various reasons this schedule could not at all times be very closely 
adhered to. Consequently, the data obtained in these experiments are 
not altogether comparable with those obtained in the greenhouse. Even 
so, as will be shown later, the two sets of data coincide closely enough 
to justify the conclusion that under field conditions soil temperature is 
one of the chief limiting factors in the development of yellows in young 
seedlings as well as in the older plants after transplantation. 
Kxpkrim^nTai. methods 
The soil temperature was recorded in 1917 in the experimental plot 
at I inch below the surface throughout the growing season and at 4 
inches fora part of the season. In 1919 the temperature was recorded 
only at 4 inches below the surface. Fries kerosene bulb thermographs 
were used for this purpose. The instruments were checked once a week 
against a standardized thermometer to insure accuracy. 
The seed was sown in short rows and observations were made every few 
days for a period of 20 to 30 days. In a few cases more than 30 days 
intervened between observations. During the driest part of the season 
it was necessary to water the soil at the time the seed was planted in 
order to insure germination. The seedlings were counted soon after 
they emerged from the soil, and this number was used as a basis for 
calculating the percentage of disease. The results of the experiments 
of 1917 and 1919 are given in summarized form in Table XI; the percent¬ 
age of yellows and the range of the soil temperature are shown graphically 
in figures 7 and 8. In these figures is shown the correlation between the 
percentage of yellows and the curves which represent the range of mean 
daily sofi temperatures. The mean is an average of the temperature 
readings taken at 2-hour intervals during the day. This is only a rela¬ 
tive temperature, of course, because it gives no notion of the duration 
of the extremes, which is undoubtedly an important factor with diseases 
of this type. 
Expsrimbntai. data.— An inspection of Table XI and fibres 7 and 8 
shows that seedlings started early in the season may remain absolutely 
free of the ydlows disease for the first 30 days, while the successive 
later plantings, up to about August 15, show a progressive increase in 
