16 
C. Experiment Station 
With the exception of a few preliminary tests made in a gas oven, all 
lots of seed have been treated in an electric oven provided with a constant 
temperature regulator. This oven which has been kept in a constant tem¬ 
perature incubator room during fall, winter and spring, and in the open 
laboratory during the summer, lias maintained itself within one degree 
of the desired temperature. During the tests with seed of the 1923 and 
1924 crops, this oven was equipped with a temperature recording device 
in order that the operator should be informed of any deviation from the 
desired temperature due to temporary failure of the heating current. All 
tests were discarded in which significant variations from the desired tem¬ 
peratures had occurred. A considerable number of tests made in the 
early part of this study had to be discarded because of an unobserved 
source of error which made them untrustworthy and which should be 
avoided by persons desiring to do work requiring the use of definite tem¬ 
peratures. This source of error arises from the fact that the difference 
in temperature between two positions on the same oven shelf may be 
surprisingly large. This became apparent when several lots of seeds which 
had been heated concurrently but for different periods of time on the same 
shelf of the oven gave unexpected differences in percentage of germination. 
The lotes heated for the shortest periods showed strong reduction in germi¬ 
nation while those heated for the longest periods showed no reduction. 
When the temperatures of the different positions on the oven shelf were 
tested it was found that there are differences of as much as 10° C. between 
certain positions which are not more than 2 inches apart. This situation 
is represented in Figs. 1 and 2, which show the temperatures at certain 
indicated positions on the oven shelf when the oven is being operated at 
two different temperature levels. The small square indicates the locus 
of the oven thermometer. The circles show the temperatures found at the 
positions of the other thermometers when the oven thermometer stands 
at the temperature recorded in the square. The temperatures at one end 
and the back of the oven are too high while those at the other end and the 
front are too low. Obviously, when several lots of materials are being 
heated at the same time, and, presumably, at the same temperature, they 
are actually being subjected to temperatures which may differ sufficiently to 
introduce serious errors. After this variation in the temperature of dif¬ 
ferent positions on the oven shelf was discovered, the different lots were 
heated singly and always in the same place and a thermometer was kept in 
direct contact with the seeds in order to make certain that they were 
actually being subjected to the temperature desired. 
As cotton seed comes from the gin, it contains, as a rule, a goodly 
number of seeds possessing defective embryos and consequently incapable 
of germination. The greater part of these imperfect seeds may be recognized 
by their light weight, small size, and shrunken coats. In order to exclude 
undue variations in percentage of disease and germination which might arise 
from the presence of faulty seeds, only such seeds were selected as appeared 
capable of germination. 
Unless otherwise specified, fifty seeds were used in each lot and the per¬ 
centage of diseased seedlings is based on the number of seeds in each lot. 
