Sphaerotheca 77 



knots of similar structure. Tiie abortive branchlets have exces- 

 sively numerous buds all infected by the insect, and covered by 

 the fungus. The conidial stage is found associated with the 

 perithecia and sometimes even extending out on the twigs to 

 the under side of the leaves. The perithecia are found in the 

 spring but do not mature their spores till late fall or winter." 



The more obscure cells of the outer wall of the perithecium, 

 the tendency of the inner wall to separate completely from the 

 outer, and the smaller average size of the perithecium and ascus 

 seem to be the only characters separating the present plant from 

 5. humiili, and are hardly important enough, by themselves, to 

 give specific rank. There is some evidence, however, which points 

 to the possibility of S. phytoptophila being distinct as a " biological " 

 species, and it is from these considerations, viz : that it is possible 

 that the present plant is a species closely allied morphologically to 

 S. huniuU^ but yet physiologically so dependent on its special host 

 as to be unable to grow elsewhere, that I have kept the two 

 distinct. ■ ' ' 



In the specimens I have seen of ^. phytoptophila the Inner wall 

 of the perithecium, composed of colorless, angular (often hexag- 

 onal) cells with refractive walls, shows a marked tendency to 

 separate completely from the outer wall. When a ripe perithecium 

 is burst by pressure, causing the expulsion of the ascus, the inner 

 wall, often in the form of a hollow oval body split at one side, is 

 forced out at the same time (see Fig. 124). If the perithecium 

 which is opened is immature, the inner wall is frequently pressed 

 out unburst, completely enclosing the ascus. In 5. ///////////.the 

 inner wall very rarely, if ever, presents this appearance, usually 

 closely adhering in fragments to the outer wall, or becoming more 



or less absorbed. The small size of the ascus and perithecium 

 could probably be matched in certain forms of S. hnuudi. 



The following observations of Halsted's (155) made on living 



M 



examples of 5. phytoptophila are of great interest : *' At this 

 season of the year (January) all traces of the mildew are absent 

 from anything except the infested or abnormal branches, and 

 upon these the perithecia are limited to the bud-scales, with a 

 particular preference shown to the lower portion of the scale. 

 Upon further study, the buds infested were found to be much 



