REPORT OF COMMITTEE ON PUBLICATION 143 



cavity is filled by the irregular mass (PI. I, fig. 5). In other cases the 

 diameter is hardly more than that of the primary spore (PI. I, fig. 1). In 

 this case the shape is usually pyriform. The size varies from 6 to 22 by 

 17 to 27/z. 



As has been indicated above there arise from the secondary spore one 

 or more strands of mycelium which radiate in different directions into the 

 tissue of the leaf. These do not average more than 2 to 3/i in diameter. The 

 exact time elapsing before the haustoria are produced has not been deter- 

 mined but in all material examined by the writer they are not formed on the 

 secondary spore as is figured by Istvanffi, nor are they developed from the 

 mycelium immediately. 



The incubation period varies with the weather, the variety, the season 

 of the year and the portion of the plant affected. Istvanffi states that during 

 the spring the incubation period is longest and that it gradually grows 

 shorter until the early part of August after which it again lengthens. The 

 periods as he determined them are May, 15 to 18 days, June, 11 to 14 days, 

 July, 6 to 7 days, August, 5 to 6 days and late August to September, 9 to 11 

 days. On the fruit the period averages from 2 to 5 days longer. 



The varieties Delaware, Niagara, Catawba and Clinton were inoculated 

 during the summer of 1913. It was found that on the first three, which are 

 relatively more susceptible than the last, that the period of incubation was 

 from 7 to 12 days, while on the Clinton under the same conditions, the 

 period was 20 days. 



The difference exhibited in the incubation period on the Delaware and 

 Clinton begins to be evident in the earliest stages of infection. Penetration 

 of the germ-tube through the stomatal opening occurs in 3 to 5 hours in the 

 leaves of the Delaware, while on the variety Clinton penetration apparently 

 does not take place until after about 20 hours. In about 36 hours there is a 

 considerable formation of mycelium in the former variety while in the latter 

 the secondary spore has barely germinated (PI. I, figs. I and II). If this 

 retarding of growth continues it easily explains the great difference in the 

 incubation periods. 



Following the method previously described by the writer (1913), numer- 

 ous infection experiments were performed during the summer of 1913. Spore- 

 laden drops of water were placed at all points on the upper and lower sur- 

 faces of the leaves. Notwithstanding the fact that the conditions under 

 which the inoculations were made, were as nearly identical as it was possible 

 to secure, in some cases the inoculations were on the upper and lower sur- 

 faces of different leaves in the same inoculation chamber, it was never 

 possible to produce an infection through the upper surface of the leaf. The 

 Eur- pean investigators report a certain percentage of successful infections 

 on this surface. Istvanffi has shown that this is true because stomates 

 occur along the mid-rib and the principal veins and about the tips of the 

 leaves, while on the lower surface they are very thickly and uniformly 

 scattered over the surface. According to our observations on several differ- 

 ent varieties there are no stomates on the upper surface of the leaves of 

 American varieties. This would explain why it was not possible to infect 

 the upper surface of the leaves. 



Istvanffi has shown that there are a few stomates on the ovary of the 

 flower of vinifera varieties and that they do not increase in number. When 



