EPIDEMICS. 59 



winter and spring, the first infection may be heavy and widespread 

 and the chances for an epidemic may be increased in proportion. 

 Thus, the presence and unusual rustiness of barberries in any one 

 district and consequent abundance of secidiospores in that district are 

 favorable for a local epidemic, and an abundance of uredospores pro- 

 duced on winter grains in mild climates, or wintering in colder cli- 

 mates and then distributed by the wind, may have the same effect 

 over wider areas. That such wintering uredos and wind-blown 

 spores are usually present in sufficient quantities to give the rust a 

 good start is fairly well established. The multiplication and dissemi- 

 nation of these spores may extend over a period of several weeks and 

 may even be facilitated by periods of dry, windy weather under 

 temperature and moisture conditions in which germination will not 

 take place. 



Whether or not these spores cause infection after falling on the 

 grain depends upon various conditions. Sudden showers at this 

 time undoubtedly wash off many of the spores before germination 

 occurs, while fairly humid conditions and moderate temperatures 

 are not only favorable but almost absolutely necessary for infection. 

 Cool nights with an abundance of dew and humid, misty da}^s in 

 which the grain remains moist from 12 to 24 hours at a time are 

 exceedingly favorable and are far better than periods of excessive 

 rainfall, due to sudden showers, with periods of hot sunshine between. 

 Contrary to the general belief, moderately cool and even subnormal 

 temperatures are more favorable for spore germination and infection 

 of the grain than higher temperatures. Thus, in the excessive tem- 

 peratures which often occur in the Middle Northwest in July and 

 August, it is exceedingly difficult to produce rust infection by many 

 of the rusts even though moisture and other conditions are favorable. 



Marshall Ward (98, p. 233) has shown that in the case of the brown 

 rust of bromes, Puccinia dispersa Erikss., germination of the uredo- 

 spore will not take place at temperatures much above 26° to 27.5° C. 

 (78.8° to 81.5° F.) or below 10° to 12° C. (50° to 53.6° F.), will not 

 germinate at all at 30° C. (86° F.), and will produce maximum ger- 

 mination at about 20° C. (68° F.). The different species and varie- 

 ties or biologic forms of rusts vary somewhat in this respect, but 

 moderately cool temperatures are more favorable for germination 

 (and consequent infection) of the uredospore of most of them than 

 excessively high temperatures. Even after infection has taken place 

 excessive temperatures may inhibit to some extent the development 

 of the rust, while moderate temperatures will aid its development. 



The presence and germination of rusts being accounted for, it 

 remains to be seen when the grains are in the most receptive condi- 

 tion. In 1908 and 1909 the authors investigated this point for 



216 



