Feb 3 ,1923 Use of AIternating Temperatures in Seed Germination 329 
Table IV .—Comparison of temperature conditions giving most rapid germination in the 
chambers with those giving earliest seedling production in the field 
Kind of seed. 
Bean. 
Beet. 
Muskmelon. 
Onion. 
Parsley. 
Pea. 
Squash. 
Tomato. 
Watermelon 
Place of test. 
Daily range of 
temperatures 
giving most 
rapid germina¬ 
tion or earliest 
seedling 
production. 
Number 
of hours 
below 
22X 0 c. 
/Chamber. 
°C. 
20 to 33 
10 
\Soil... 
18 to 32 
12 
/Chamber. 
20 tO 33 
10 
/Soil. 
18 to 32 
11 
/ Chamber. 
20 tO 33 
10 
/Soil. 
20 tO 40 
9 
/Chamber. 
20 tO 33 
10 
/Soil. 
14 to 30 
16 
/Chamber. 
20 tO 28. 5 
14 
/Soil. 
17 to 32 
12 
/Chamber. 
20 tO 33 
10 
/Soil. 
16 to 33 
14 
/ Chamber. 
20 to 33 
10 
/Soil. 
23 to 41 
3 
/Chamber. 
20 tO 28. 5 
11 
/Soil. 
18 to 32 
11 
/Chamber. 
20 tO 33 
10 
/Soil. 
20 tO 40 
8 
Number 
of tem¬ 
perature 
altera¬ 
tion.** 
9 
9 
9 
9 
5 
9 
9 
7 
9 
° See figure 17. 
CONCLUSIONS SUGGESTED BY THE FOREGOING COMPARISONS 
Altogether Table IV and Figure 20 show a close correspondence be¬ 
tween chamber temperatures and soil temperatures which gave most 
prompt germination of the seeds used. Undoubtedly a similar corre¬ 
spondence exists between the temperature conditions best fitted for arti¬ 
ficial germination tests of Bermuda grass and Johnson grass seed (fig. 
5 and 6 and text p. 304-305) and the soil temperatures during the period in 
which they germinate in nature in the warm climate of the places to 
which they are native. 
The results of the comparison suggest: 
1. For general use in conducting germination tests an alternation of 
temperatures covering a somewhat wider temperature range than the 
customary alternation between 20° and 30° C., for instance an alternation 
between 20° and 32 0 C. However, if the alternation is secured by the 
method of transfer between two germinating chambers at fixed tem¬ 
peratures, probably better results would be obtained with many kinds of 
seed by the use of a temperature range less than that characteristic of 
conditions in the soil, where the temperature changes are less rapid and 
the period during which the temperature is near either extreme is shorter. 
In view of these considerations and the additional fact that the warmer 
alternations Nos. 9 and 10 though giving more rapid germination in* many 
cases, were less favorable in other cases than the cooler alternations, an 
alternation similar to No. 3 seems to meet the theoretical requirements of 
approximating with sufficient accuracy what may be termed optimal 
field conditions. 
2. For use in germination tests of onion seed an alternation of tempera¬ 
tures covering a cooler range than that of any of the alternations with 
which it was tested in this investigation. 
3. For the germination of seeds of the cucurbits an alternation of tem¬ 
peratures covering a warmer range than any of those investigated. 
Whether or not any advantage would be derived from any of these 
suggested changes in method remains for further experiment to show. 
