Feb. 3,1923 Use 0} A Iternating Temperatures in Seed Germination 301 
KENTUCKY BLUEGRASS 
Figure 3 shows the average percentages and rates of germination of 
three lots of Kentucky bluegrass seed. Here the favorable effect of 
temperature alternations is strikingly evident and increases with the 
progress of the test. All of the low points in the graphs, except for the 
earliest count (seven days), represent germination at constant tempera¬ 
tures or with narrow (5 0 C.) alternations. A narrow alternation was 
more effective in a low temperature range (15 0 to 20° and 20° to 25 0 ) 
than in a higher temperature range (25 0 to 30°). This has been found 
to be true also with other seeds requiring temperature alternations for 
best germination. An alternation of 15 0 retarded germination in the 
early days of the test. The alternations 25 0 to 35 0 and 30° to 35 0 
were too warm for good results. 
It should be emphasized here that the beneficial effect of temperature 
alternations can not in any way be referred to a constant temperature 
equivalent to the mean temperature of the alternation or to the extreme 
temperatures of the alternation. Thus, temperature alternations with 
20 0 C. as either upper or lower extreme gave much better results than 
20 0 constant, and any alternation gave better results than either of the 
extremes of that alternation constantly maintained. Again, the ger¬ 
mination with the alternations 15 0 to 30° and 20° to 35 0 was approxi¬ 
mately three times as great as at their constant temperature equivalents, 
20 0 and 25 0 . 
At the same time it is possible to see an effect of mean daily tempera¬ 
ture supplementing the more important effect of the temperature changes 
as such. Since the higher temperature in each alternation was main¬ 
tained for approximately one-third of each day the mean temperature of 
each alternation can be represented by adding to the lower temperature 
one-third of the difference between that and the higher temperature. 
Table I shows the mean temperatures of the alternations which gave fairly 
good results, together with the average total percentages of germination. 
The germination increased regularly with increase in the mean temperature 
of the alternation up to 23 0 C. and fell rather abruptly after this mean. 
There was a similar correspondence between mean temperatures and the 
germination of Johnson grass seed, but not with any of the other seeds 
which required alternations of temperatures for best germination. 
Table I .—Temperature alternations , corresponding mean temperatures , and average 
percentages of germination produced in three lots of Kentucky bluegrass seed 
Temperature alternation. 
Approxi¬ 
mate mean 
tempera¬ 
ture. 
Average 
percentage 
of germi¬ 
nation. 
Temperature alternation. 
Approxi¬ 
mate mean 
tempera¬ 
ture. 
Average 
percentage 
of germi¬ 
nation. 
°C. 
15 tO 20. 
15 to 25. 
IS to 3°. 
20 tO 25 . 
°C. 
16 
18 
20 
21 
46 
48 
49 
52 
°C. 
15 to 35. 
20 tO 30 . 
20 tO 35 . 
2 S to 3S. 
°C 
22 
23 
25 
28 
58 
63 
56 
35 
Kentucky bluegrass is one of the kinds of seed with which Hartleb 
and Stulzer ( 12 ) secured much better germination at 30° C. than with 
an alternation between 20° and 30°. The results here reported, as well 
as the results of several other investigators, show that Hartleb and 
Stulzer’s results were unreliable and their conclusions false. 
