14c Classification. 



It may also be chosen because it is a perfect store-house of nourishment 

 provided bv the young embryo in a form easily assimilated, so that the 

 parasitic fungus steps in and utiUzes for its own reproduction what was 

 originally intended for the propagation of the species of the host-plant. 



The most prominent feature of the smuts is their reproductive bodies 

 or spores, and they form the basis of our classification, taken in conjunction 

 with their mode of formation and germination. In fact, so important a 

 systematic character is the mode of germination of the spore that it 

 is used to divide the group into two main divisions, and in cases 

 where this has not been observed, there is often room for doubt as 

 to the exact position of the species. The spore, on germination, 

 puts forth a germinal tube, which may either divide transversely and 

 form lateral as well as terminal conidia, or it may remain at first 

 undivided and produce a crown of conidia at the apex. Too much 

 stress must not be laid on the midivided germinal tube, for it is the whorl of 

 conidia at the apex which is the distinctive feature. This is the basis of the 

 main division of the order into the two families of JJstilaginacece. and 

 Tilletiacece, and it shows how necessary it is for classificatory purposes to 

 determine the mode of germination of the spore. There is general unanimity 

 as to the division of the smuts into two families, but as to the limits of the 

 genera there is difference of opinion. The one touches the other at so many 

 points that there is always room for individual judgment in determining 

 whether a genus shall be circumscribed or made to include a number of 

 outlving forms, and what characters are to be regarded as essential. 



Thus Cornu^ has separated out from the genus Ustilago those forms 

 which, instead of producing their spores more or less simultaneously, develop 

 them in succession in spore-bearing hyphae arising from a so-called fertile 

 stroma or persistent mycelium which surrounds a central columella, and 

 for such species he has constituted the genus Cintractia. Wherever this 

 feature is clearly shown of basipetal formation of the spores as in C. axicola, 

 there, as a matter of convenience, the genus is adopted, although some would 

 not consider it of sufficient importance to merit generic distinction. 



Again, in addition to a central columella, there are some forms closely 

 allied to Ustilago which are provided with a definite fungus membrane 

 enclosing the spores, and which opens in various ways to allow them to 

 escape. On account of this pecuHarity, De Bary^ has placed them in the 

 genus Sphacelotheca, as illustrated in S. hydropiperis. But the fungus 

 membrane enveloping the spores shows every gradation from that of 

 Sorosporium reilianum, where it is partly composed of plant cells at the base 

 of the ovary and entirely fungus cells at the top, to that of Ustilago (Sphace- 

 lotheca) hydropiperis, where the false membrane forms a special receptacle, 

 splitting at the top to allow the escape of the spores. Besides, this membrane 

 enclosing the spore-layer also occurs in species of Cintractia (C crus-galU) 

 and Sorosporium {S. panici-miliacei) , so that the distinction is not suffi- 

 ciently definite, in my opinion, to justify the formation of a new genus. 

 Sometimes the plant tissues are hollowed out into cavities, and the simple 

 spores develop there, thus constituting the monotypic genus Melanopsichium. 



In such a genus as Sorosporium, where the spore-balls readily separate 

 into their constituent spores, it is not always easy to separate it from 

 Ustilago, especially when the mode of germination is the same. In order to 

 be certain in some cases it is necessary to examine the spores in the early 

 stages of their formation. 



In Thecaphora and Tolyposporium the spore-balls are rather permanent, 

 and it is mainly by the structure and germination of the spores that the 

 difference is determined. 



