For some years the Seed-Testing Laboratories of the Bureau of Plant Industry has been using daily alternating temperatures for the germination of certain seeds. In these alternations the lower temperature is used about 18 hours and the higher temperature about 6 hours of the day. Alternating temperatures are found especially favorable for the germination of the seeds of certain grasses: Johnson grass, Holcus halepensis L. (Sorghum halepensis Pois.) ; bluegrass, Poa pratensis; and Bermuda grass, Capriola dactylon (L.) Kuntze. For Johnson grass 30 C. for 18 hours and 45 C. for 6 hours of the day have been found to be the most favorable alternation. Sudan grass, Holcus halepensis sudanensis (Piper) Hitchc. (Andropogon halepensis sudanensis Piper), which is closely related to Johnson grass and very similar to it in most respects, germinates readily at constant temperatures. During the last three years (1916-1918) an extensive physiological study of the behavior of the seeds of Johnson grass and Sudan grass has been made in order to get some light on the reasons for the difference in the requirements of the two regarding alternate temperatures for germination. Incidentally such a study has thrown much light upon delayed germination and the physiology of germination in general. Naturally seeds of several other grasses and of a number of other families have been drawn into the investigation. The present paper reports one phase of this extensive comparative study of the physiology of germination.