296 
Journal 0} Agricultural Research 
Vol. XXIII, No. s 
Haberlandt (10) and Eidam (5) have suggested temperature effects 
upon the water intake of seeds as a possible cause of the favorable effect 
of temperature alternations upon germination. This hypothesis seems 
hardly tenable, since certainly some and probably most seeds whose ger¬ 
mination is favored by temperature alternations take up at any tempera¬ 
ture more water than the minimum required for germination. Von 
Liebenberg (16) suggested that the reserve materials made available at 
any given temperature were wholly or largely used up in respiration, but 
that a slight surplus, becoming soluble at higher temperatures, is avail¬ 
able for growth when the temperature is lowered with consequent reduc¬ 
tion in the intensity of respiration. Gassner (7) considered the effect of 
temperature alternations upon the germination of Chloris ciliata to be 
entirely a matter of oxygen relations. In line with Gassner’s conclusion 
Vanha (21) called attention to the fact that differences in temperature 
between the different parts of the seed, the germinating bed, and the 
outer air following a sudden temperature change cause different gas densi¬ 
ties which may set up lively gas movements leading to removal of C0 2 
and renewal of oxygen—conditions favorable to increased respiration and 
probably to germination. 
Probably none of these hypotheses can furnish a complete explanation 
of the effect of temperature alternations, but they point the way for an 
attack upon the problem. A detailed study of the respiration and 
internal changes of the seeds at different temperatures would probably 
help in its solution. The respiratory quotient of dormant apple seeds 
decreases with decreasing temperature, and vice versa. If this is true of 
dormant embryos generally, there may be at low temperatures an accumu¬ 
lation and actual metabolism of oxygen in a form which becomes immedi¬ 
ately available for the inception of growth processes when the seeds are 
placed at the higher temperature. This hypothesis is in some respects 
the converse of that of Von Liebenberg (16), but it does not stand in 
opposition to that of Vanha (21) or to that of Gassner (7). 
The purpose of the present paper is not to discuss the possible explana¬ 
tion of the effects of temperature alternations upon germination but 
merely to show somewhat more in detail the nature and extent of such 
effects. 
EXPERIMENTAL WORK 
Several years ago seeds of carrot (Daucus carota L.), celery (Celeri 
graveolens (L.) Britton), parsley (Petroselinum hortense Hoffm.), and 
parsnip (Pastinaca saliva L.) were tested for germination at 15 0 , 20°, 
253°°> and 35 0 C., and with every possible alteration between these 
temperatures in which the higher temperature was maintained 7^ hours 
and the lower temperature 16^ hours of each day—a total of 5 constant 
temperatures and 10 temperature alternations. The temperature alter¬ 
nations were accomplished by transferring the trays with their load of 
seeds from one germination chamber to another and back again each day. 
Under nearly all of these 15 temperature conditions, germination tests 
of seeds of the following grasses were made: Timothy (Phleum pratense 
L.), awnless brome grass (Bromus inermis Leyss.), perennial rye grass 
(Lolium perenne L.), Italian rye grass (Lolium multiflorum Lam.), 
meadow-fescue (Festuca elatior L.)> orchard grass (Dactylis glomerata L.), 
redtop, (Agrostis polustris Huds.), Kentucky bluegrass (Poa pratensis 
L.), and Bermuda grass (Capriola dactylon (L.) Kuntze). 
