96 BOTANICAL GAZETTE [august 



some 



Arthur's (i) new classification is based upon the assumption 

 that all of the spore forms — spermatia^ aecidio-j uredo-, and teleuto- 

 spores^ and sporidia — were present in the ancestors of our rusts, and 

 he leaves untouched the question of the origin of this multiplicity of 

 spore forms. Reasoning from this assumption, the forms with ab- 

 breviated life-cycles arise by a process of elimination. In this way 

 any portions of the sporophyte and perhaps the whole gametophyte 

 might be successively suppressed. 



Endophyllum has perhaps arisen in this way. In it a uninucleated 

 mycelium produces binucleated aecidiospores which apparently have 

 no uredo host. The spores germinate by a sort of promycelium 

 which bears four uninucleated sporidia. These sporidia again infect 

 the aecidial host, producing again a mycelium of uninucleated cells. 



If we assume on the other hand that the rusts evolved from lower 

 fungi, and that the various subgenerations of the sporophyte are 



w^ay old 



forms eliminated, quite a different classification is suggested. The 

 gametophyte is universally regarded as the primitive generation, and 

 in its earliest appearance was relatively simple. In the pteridophytes 

 the sporophyte increases in importance until the gametophyte is 

 finally the more inconspicuous. 



I incline to the view that the lepio- and micro- forms, in which the 

 teleutospores or spermatia and teleutospores only occur and are borne 

 on mycelia with uninucleate cells, are the primitive rusts. We have 

 in them the gametophyte bearing the gametes and producing the 

 fusion cell. A very short outgrowth of this cell now bears the teleuto- 

 spores in which the sporophyte generation has its end. The first 

 modification of this life-history would then appear in a further develop- 

 ment of the sporophyte. This might be brought about by the fusion 

 cell producing a mycelium directly rather than the basal cell normally 

 produced, or it is possible that the basal cell should produce temporary 

 spores, which might carry this sporophytic mycelium to other tissue, a 

 result which would be altogether favorable on account of the exhausted 



condition of the host in the neighborhood of the fusion cell. Spores 



■ 



formed in this way would conform to our definition of primary urcdo- 

 sporcs, which it would thus appear are to be considered as the first 

 added spore generation, and the hrachy- form would thus be the first 



