Sphaerotheca 55 



a few forms of 5. ImninU, e. g,, specimens on Gcum album, the 

 cells here and there become largely and slightly irregular in shape, 

 and I have found contiguous cells to measure lo/i and 2Z fi wide, 

 although in the specimen in question the cells averaged under 

 20 fx. Nevertheless, on the whole, with the few and rare excep- 

 tions noted above, the difference in the size of the cells is constant 



and strongly marked, so that there is usually no difficulty in re- 

 ferring at once any specimen, by means of this character, to either 

 S, hinmdi or its var. fidiginea. The difference can be well seen 

 by comparing together the two commonest examples of these 

 plants, viz, S, kuvinii on the hop, with S, hiuuu/i, var. fuUginea on 

 the dandelion. 



I do not find the other separative characters put forward by 

 Burrill to hold good. Identically formed appendages are found 



in certain examples of the two plants, and we can only say that as 

 a rule 5. hii)}iuli has longer, straighter, darker colored appendages, 

 while in the var. fuliginca they are, as a rule, shorter, paler, and 

 more tortuous. But to this rule there are certainly numerous 

 exceptions, and we find forms of 5. humuli^ e, g.^ on Gcinn album, 

 Rubus^ Gilia linearis^ Geranium^ etc, which have appendages iden- 

 tical in all respects with those usually characteristic of the var. 

 fiiligmea, from which these examples 



r 



nnea, from which these examples can be separated only by 

 the small size of the cells of the perithecium. Similarly, certain 

 forms of the vz.r. fuUginea, e. g,, those on Microseris tcnella, Senccio 

 lugens, etc., have long, deeply colored appendages, very closely 

 approaching those of typical 5. Jinmnli. 



The size of the ripe spores in the two plants is difficult to ascer- 

 tain, as the spores are formed very tardily, and in many of the 

 most interesting specimens it is frequently the case that no ripe 

 spores are to be found. As far as I have been able to observe, how- 

 ever, the difference in size given by Burrill does not appear to be 



constant ; 



fuH 



exceed the length given by this author, viz, 15//; in the speci- 

 men in Syd. Myc. March. 297 (on Veronica longifolia) they fre- 

 quently measure 25 // long. 



There remains, therefore, only the difference in the size of the 

 perithecium, and this, while affording a fairly reliable character, 

 cannot, I think, be considered as one of more than varietal impor- 

 tance. 



