Oct. i, 1920 Fusarium-B light (Scab) of Wheat and Other Cereals 27 



considerably with the condition of the seed used. When light, shriveled 

 seed is sown on infested soil, or when such seed is inoculated by being 

 dipped in a suspension of conidia and then sown on sterile soil, the 

 seedlings will succumb to the attack of the parasite much more rapidly 

 than when healthy seed is used. Abundant watering of the plants also 

 increases to some extent the rapidity of the attack. 



In general, under greenhouse conditions, the first symptoms of root 

 infection appear not earlier than the seventh day after sowing. Infec- 

 tion is usually abundant after the fourteenth day. When naturally 

 infected seeds have been used on sterile soil the symptoms of root infec- 

 tion may appear even before the seventh day. 



On heads. — In head infection there is much less variation in the incu- 

 bation period. In damp weather, the period that elapses between 

 inoculation and the appearance of the first symptoms (water-soaking) 

 varies from three to six days. In dry weather, symptoms of infection 

 may not appear until after the first rain, or if heavy dew falls during the 

 night and lasts for the greater part of the forenoon, symptoms of infection 

 may appear from five to eight days later. 



The rapidity with which the blight infection spreads from the point 

 of infection to the rest of the head varies greatly. It varies considerably 

 with different individuals and depends much upon the kind of weather. 

 On healthy, vigorous, and more succulent plants the infection spreads 

 much more rapidly than on plants of average vigor. Moist and cloudy 

 weather, followed by warm and clear weather, greatly accelerates the 

 rapidity of infection and killing, yet even under such conditions the 

 infection may be restricted on many heads to a single spikelet, the rest 

 of the head remaining healthy and developing perfectly normal, plump 

 kernels. 



For the study of the rapidity of the spread of the disease from the 

 point of infection, heads showing primary infection were located daily 

 and marked with tags so that they could be located again. Heads so 

 tagged were examined every two or three days and the changes recorded. 

 In this way the effect of the various factors affecting the rapidity of 

 blight infection and killing were studied. The following are typical 

 records of some infected heads, made in 1918: 



N 1009, July 11, 1 spikelet infected. Infection at base of head. 



July 14, 4 spikelets infected. 



July 17, whole head killed. 

 N 101, July 11, 5 spikelets infected. Infection at middle. 



July 14, 8 spikelets infected. 



July 17, whole head killed. 

 N 1038, July 9, third spikelet from bottom infected. 



July 14, 4 spikelets infected. 



July 17, Whole head killed. 

 N 1039, July 9, 1 spikelet infected. Infection at middle. 



July 14, 4 spikelets infected. 



