20 THOMAS: INFECTION OF APIUM GRAVEOLENS 
ten spots per leaf were measured. Thus fora plant of five leaves 
fifty spots were measured. 
In the case of the first series kept in the dark room before » 
inoculation, the most marked increase in the size of the spots 
was on the youngest leaves, which were put out partly or entirely 
while the plants were in the dark room. On these the spots at 
times exceeded in diameter those on the oldest leaves. In the last 
series of plants the increase in size apppeared to be proportionate 
for all the spots. It has been noted that when infected plants 
are placed in the inoculation chamber, for forty-eight hours, a 
zone of at least one half to one millimeter surrounding each spot is 
broken down. That the mycelium of Septoria should advance 
this distance in so short a time does not seem probable. It 
appears rather that the weakened tissues of the host plant succumb 
where the fungus is already present. 
RELATION OF TEMPERATURE TO INFECTION 
It has been recognized since the late blight disease began to be 
studied that it is more severe in the early autumn than during mid- 
summer. I have found this to be true in the greenhouse as well 
as in the field. Several experiments have been performed to test 
the relation of this condition to temperature. Plants were 
inoculated uniformly and divided into two groups which were 
kept through part or all of the incubation period at temperatures 
TABLE XI 
EFFECT OF MAINTAINING INOCULATED PLANTS AT DIFFERENT TEMPERATURES 
THROUGHOUT THE INCUBATION PERIOD OF TWENTY-ONE DAYS 
Number of infections per plant 
Mean ay, temperature PUNE 2% "ER SEU CRIT reece SEN aR Rarer a a RR 
Fifa j4 | s| 6 | 7 | 8] 9 | 1 | Average 
SRN aeahencm ata cam seeuncmmiirer ot ee eee ROTM TY ees esl! ceidiomal ood } 
Rae Ce ee. 383 | 272 ntl tay 189 377} 98 133 | 209 269 253-7 
sa Cup ste ts--} ag! 86 [ taol, $2) 3} alae! ya} teal. 37] 79-4 
differing from 7 to 13. degrees Centigrade (mean average). Five 
sets of from ten to twenty plants each were inoculated. The 
infection develops more rapidly at higher temperatures but later 
counts show usually no striking difference and the individual 
plants vary widely. One set (TABLE XI) showed a marked 
though not altogether consistent difference in counts. However, 
