404 Arthur and Kern : Peridermium 



graphic study of the North American species up to the present 

 time, yet there has been a steadily increasing accumulation of 

 knowledge regarding them, well indicated in Farlow and Seymour's 

 indexes, referred to above, in 1896 seven species being recognized, 



and in 1905 fifteen species. 



. The present paper describes twenty-seven species of Perider- 

 mium, ranging from Mexico to Alaska, and from the Atlantic to 

 the. Pacific coasts, and also three species not yet found in America, 

 but which doubtless occur, as the telial forms are abundant. Some 

 important characters are used in the diagnoses not hitherto em- 

 ployed for American forms, such as those derived from the pres- 

 ence and form of pyenia, the structure, especially the cross-section 

 view of the peridium, and the thickness of the wall of the spores. 

 The characters of the peridial cells and of the pyenia have been 

 obtained by making thin vertical sections of the leaf. A piece of 

 the leaf is allowed to soak for a few minutes in boiling hot water, 

 then placed between pith, and sectioned with a sharp razor and 



steady hand. 



Many recent, as well as earlier, mycologists have treated the 



species of Peridermium under Aecidium. But that method, it seems 

 to us, more obscures than illuminates the subject. Both names as 

 ordinarily used belong to form-genera. The species are all aecia 

 (aecidia) of species belonging to telial genera, but our knowledge 

 is yet insufficient to properly assign them. The species in the 

 great majoriy of cases which have been described under the form- 

 genus Aecidium are the aecia of pucciniaceous species, while the 

 species described under Peridermium with scarcely an exception 

 are the aecia of melampsoraceous species, using these terms in the 

 Dietelian sense. The species of Peridermium with rare or possi- 

 bly no exceptions can be distinguished from other aecial species 

 by both host and structural characters, and there seems no good 

 reason for submerging them under the more general form-genus 

 Aecidium. 



The genus Peridermium as here understood embraces all aecial 

 forms possessing peridia, inhabiting the Pinaceae and Gnetaccac. 

 By this definition those aecia on gymnospermous hosts having no 

 peridia, usually classed under Caeoma, are excluded, for example, 

 Caeoma Laricis belonging to the poplar and willow rust, C. 



