208 A Monograph of the Erysiphaceae 



Galiinn. Curiously enough, this form in Europe has been referred 

 by all botanists to E. communis i^E. polygoni), although it undoubt- 

 edly belongs to E. circlioraceanim, of which it appears to be a 

 form approaching E. galeopsidis. In all the European specimens 

 I have examined the perithecia have contained asci without any 

 trace of spores, and the habit, size, shape and number of asci, 

 etc., further show that the fungus does not bfelong to E. polygoni. 

 Moreover, in Turkestan specimens on Galium boreale and other 

 species of the genus, and in American ones on Galium aparine 

 and G. irifiorum the asci are regularly bisporous with apparently 

 ripe spores. For this reason American mycologists have placed 

 the form under E. cichoraccarum. The shape of the haustoria in 

 the form on Galium appears also variable. These, although 

 small, and in this respect very unlike those of E. galeopsidis, are 

 sometimes, although rarely, distinctly lobed ; in other specimens 

 they resemble those of E. cichoraccarum. It appears, therefore, 

 from the above cases, that E. cichoraccarum is sometimes similar 

 to E. galeopsidis in not producing spores on the living host-plant. 

 The difference in the size and shape of the haustoria in E. 

 cichoraccarum and E. galeopsidis is, as a rule, striking. The 

 haustoria of the former species (examples on about thirty different 

 host-plants were examined) are small and simple, those of the lat- 

 ter are much larger, much lobed irregularly, or often more or less 

 reniform (deeply bi-lobed) in shape. Among some American ex- 

 amples, however, I was surprised to find in two specimens named 

 E. cichoraccarum on Eupatorium ageratoides (one from Madison, 

 Wisconsin, Sept., 18S2, L. H. Pammel) in Professor Earle's 

 herbarium; the other from Oregon, Illinois, Sept. 12, 1888 (in the 

 Herbarium of the University of Illinois), haustoria of exactly the 

 same size and shape as those on examples of E. galeopsidis on Gale- 

 opsis. Unfortunately, both these specimens are apparently rather 

 young, and the asci contain no spores. It is, therefore, impossible 

 to say whether we have in this case a form of E. cichoraccarum 

 with lobed haustoria, or whether it may not possibly be that of E. 

 galeopsidis, hitherto supposed to be confined to Labiatae and Chelone 

 among Scrophularineae, sometimes occurs on Compositae. In an- 

 other case the fungus is undoubtedly E. cichoraccarum. This is the 

 specimen on Sonchus an^ensis in Syd. Myc. March. 3051, and in 



