184 SALMON: SUPPLEMENTARY 
often more angularly bent, than the appendages of S. ‘‘Castaguet.” 
On opening the perithecia, 2 to 4, or rarely 5, asci are found. In 
Professor Magnus’s examples 2 to 3 wasthe usual number. Very 
rarely, apparently only a single ascus occurs, but as a rule a sec- 
ond ascus—often empty or more or less aborted—is to be 
found. In the rare cases in which only a single ascus is observ- 
able, this form of &. polygont is easily separable from S. “Castag- 
net,” not only by the nature of its appendages, but also by the 
more or less ovate (not subglobose) ascus, containing (in this form) 
only 2-3 spores, and not 8 as in Sphaerotheca. Moreover, in all 
such cases, the examination of several perithecia invariably shows 
the existence in most of several (2 to 5) asci, and so leads us to 
conclude that the perithecia containing only a single ascus are ill- 
developed examples. Quite similar specimens of this small form 
of E. polygoni occur on species of Geranium, and have been com- 
monly confused with S. humult. 
Brannon (8) has reported, in the case of examples of Sphaero- 
theca “‘ Castaguet Lév.”’ on a species of Erigeron, the finding of 
two or three perithecia, each containing two asci, and observes: 
‘In these unusual forms one ascus was much larger than its com- 
panion, but not as large as the ascus existing alone in a perithe- 
cium. A few of these unusual forms might lead to the question- 
ing of what has, heretofore, been considered a strong generic 
difference between Sphaerotheca and Erysiphe.’ This occasional 
occurrence of more than one ascus has also been noted by other 
observers in species of Sphaerotheca ; in two instances I have found 
perithecia of S. mors-uvae (Schwein.) Berk. & Curt. containing 
three asci (see monograph, 73). We see, therefore, that in rare 
instances Sphaerotheca may approach Erysiphe in possessing more 
than one ascus; on the other hand perithecia of the small form of 
Erysiphe polygoni on Caltha palustris, mentioned above, may V¢Ty 
rarely contain only a single ascus. It is clear, therefore, that 
Erysiphe on one side shows an affinity with Sphaerotheca; on the 
other, through &. sorti/is it claims relationship with Microsphaer@ 
through MW. astragali and M. Baumleri. 
Brannon (8) has reported the occurrence of Z. polygont on a 
species of As¢er and considers that this host had probably rece! 
the mildew from neighboring plants of Ranunculus recurvatus whi 
