CLASSIFICATION OF PYRENOMYCRTES. 31 



Hereafter it may be that a species witli septate sporidia would have 

 to be excluded if artificial are to give way to natural affinities, and 

 secondary features to be promoted to the exclusion of primary. In 

 like manner, Dothidea has always been regarded as a very natural 

 though less perfectly characterized genus, but unfortunately the 

 sporidia are more variable, and hence eight genera in three groups, 

 represent the hyaline simple spored, the hyaline uniseptate, and 

 the coloured septate spored. If the proposed scheme is to be fully 

 carried out, it must be considerably augmented, and there is no 

 reason why, if pressed to its logical conclusions, nearly every 

 species should not have a claim to be regarded as the type of anew 

 genus. In Professor Saccardo's scheme there are seven typical 

 groups — 



1. Allantosporte. 5. Phragmosporje. 



2. Hyalosporaj. 6. Scolicosporse. 



3. Phaiosporfe. 7. Dictyosporte. 



4. Didymosporfe. 



It is presumed that the same tei'ms represent the same things in 

 all cases, but that is not practically the case, for mPerisporiacce we 

 find that Hyalosporce are simple (page 1), in SpTiceriacece they 

 become 1-3 septate (page 5), in Perisporiacece globose, ovoid, or 

 oblong ; in Sphceriacece ovoid, fusiform, or oblong ; in ILjpocreacece 

 ovoid-cylindrical, and sphteroid in HystcriacecB. From the scheme 

 it would appear that the above groups represent — 



1. Sporidia sausage-shaped, almost colourless. 



2. Sporidia hyaline, simple, or 1-3 septate. 



3. Sporidia simple, coloured. 



4. Sporidia bilocular. 



5. Sporidia coloured, septate. 



6. Sporidia filiform. 



7. Sporidia muriform. 



These characters are not definitely stated as applicable to all the 

 families, bi;t under some a diagnosis is given which is not identical 

 with that in others. (Compare pp. 1 and 5 ; 3 and 9 ; 2 and 7.) 

 It is not so much with this grouping of genera that we are con- 

 cerned, although that is clearly open to criticism, as with the pri- 

 mary features of the genera themselves, and here we cannot imagine 

 that any practical mycologist could possibly place together as 

 nearest allies, Cajmodium elongatum, Sphceria herbamm, Sphceria 

 obduceiis, Cucurbitaria Berberidis, Valsa fenestrata, and Vaha ves- 

 tita, and yet these are the tyi^es of six consecutive genera, com- 

 posing one group. All sense of affinity or relationship must be 

 wholly obscured by infatuation for the one idea of conformity in 

 shape, colour, and septation of the sporidia. One seems prepared 

 to encounter almost anything after such a notion of botanical affi- 

 nities, and even Microthijriiun microscopicum^ with Sphce)-ia puncti- 

 formis and Sphceria epicymatia in juxta-position is only accepted as 

 additional evidence of the fact, so often exemplified in the world, 



