1921] HARRINGTON—GERMIN ATION 330 
this chamber being slowly heated during the forenoon, and cooled 
either slowly or rapidly as desired during the afternoon. When 
only one chamber was used the heating was accomplished by means 
of a properly adjusted gas flame below the chamber, and the 
cooling by means of a graduated stream of cold water in the top 
of the water jacket. 
The species investigated may be divided into two groups: (1) 
those whose seeds germinate well at any constant temperature from 
17.5° to 22.5°C., and also with temperature alternations; (2) those 
whose seeds require a temperature cooler than 20° C. for complete 
germination. Some of the samples in each group contained many 
dead seeds, or seeds incapable of germination at any temperature. 
Results 
Although a direct comparison between the tests made during 
the two periods (1912 and 1914) is impossible, the results of all 
the tests can best be discussed together. They will be considered 
from three standpoints: (1) the effect of alternating versus con- 
stant temperatures; (2) the effect of the different temperatures 
upon the germinating capacity; and (3) the effect of the different 
temperatures upon the rapidity of germination. 
ALTERNATING VERSUS CONSTANT TEMPERATURES 
All the species included in the investigation, with the possible 
exception of petunia, germinated as completely and as quickly with 
a favorable constant temperature as with any alternation of tem- 
peratures. It should be remembered, however, that taking the 
seeds out of the chambers to count those germinated introduced 
a brief change of temperature which may not have been entirely 
without effect. The influence of this brief temperature change, if 
it has any, would be greater when the germinated seeds are counted 
every day or two as in these experiments, than if they were counted 
less frequently.? 
? While the use of an alternation of temperatures does not seem to be necessary 
for satisfactory germination of the kinds of seeds treated in this paper, it is very 
desirable, and in some cases imperatively demanded, with many other kinds of seeds. 
This subject will be treated in an article to appear shortly in the Journal of Agricul- 
tural Research, 
