Feb. 3,1923 Use of Alternating Temperatures in Seed Germination 323 
Table III .—Germination of celery , Kentucky bluegrass, and tomato seed with 
II temperature alternations 
Alternation number. 
Approximate maximum tem¬ 
perature (°C.). 
Approximate mean temperature 
(°c.). 
Temperature at end of ic hours 
(°C.). 
Number of hours below 22.5 0 C.. 
Number of hours at 20° C. 
* [Celery: 
Pi .2 5 lots of good seed. 
3 lots of poor seed. 
Kentucky blue grass: 
7 lots. 
Tomato: 
, 4 lots. 
!?*«, 2 
a; &p.5< 
§p*f s 
s S & 
> O bJO 
_ 
I 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
2 7 - 5 
28. 5 
29-5 
33 -o 
28. 5 
33 -o 
28.5 
29-5 
32.5 
.023 
59 -o 
21-5 
21.5 
22.5 
22. 0 
22. 5 
24. 0 
23-5 
23-5 
25-5 
24-5 
24.5 
20 
20 
20 
20 
21-5 
22. 0 
22.3 
22. 2 
23.0 
22. 7 
23.8 
19 
19- 
16 
18 
14 
12 
11 
II 
10 
10 
8 
15 
*5 ’ 
12 
15 
4 
3 
3 
3 
2 
2 
(a) 
78 
76 
78 
74 
73 
72 
66 
67 
67 
43 
39 
33 
35 
36 
35 
27 
24 
20 
26 
16 
65 
68 
64 
64 
63 
61 
59 
58 
60 
59 
50 
73 
7 i 
68 
66 
59 
® The temperature in this alternation sometimes did not fall to 20° C. in the entire 24 hours. 
For best germination of the three kinds of seed here considered the 
temperature after its initial rise to a higher degree should fall at least 
as low as 21.5 0 C. within 15 hours from the time the heating was begun, 
should be below 22.5 0 for at least 12 hours of the day, and should be 
as low as 20° for several hours each day. These results could be 
attained by the method of transferring between two germinating cham¬ 
bers at fixed temperatures even if the seeds were kept as long as 10 
hours in the warmer chamber. While such alternations were not used 
in this investigation the indications, both from the work on tomato seed 
recorded in this paper (Table III) and from Vanha’s work (21), are 
that the longer period at the higher temperature would be injurious. 
Altogether the investigation revealed an unexpectedly large degree of 
tolerance to a wide range of conditions in temperature alternations on 
the part of celery and bluegrass seed. 
The method of transferring between two germinating chambers at 
fixed temperatures is much simpler to operate than heating and cooling 
a single chamber, on account of the difficulties in the way of accurate 
temperature control in the latter case; it involves more uniform condi¬ 
tions throughout an entire chamber and from day to day, and it gives 
at least as good results. It is therefore decidedly preferable to the other 
method. 
3. Rapidity of germination 
Celery and bluegrass seed germinated most rapidly with those alterna¬ 
tions of temperature which also gave most complete germination and 
least rapidly with those alternations which gave least complete germina¬ 
tion. Here, as in the earlier comparisons between alternations and 
constant temperatures (see p. 301), the special effect of a favorable 
alternation of temperatures predominates over the effect of low or high 
mean temperatures. This is in marked contrast to the temperature 
responses of constant-temperature germinators. 
Parsley seed germinated more rapidly with alternation No. 5 than 
with a warmer alternation or with either more rapid or less rapid cooling. 
All other kinds of seed germinated more rapidly with the warm than 
with the cool alternations. The differences were related to the rate and 
extent of cooling, the mean temperature, and the proportion of each 
day at a comparatively warm temperature, and were not related to the 
maximum temperature occurring in the alternation. Usually germination 
