NORTH AMERICAN SPECIES OF PUCCINIA 



ON CAREX 1 



Frank D. Kern 



In spite of an accumulation of considerable information con- 

 cerning the North American sedge rusts most mycologists have 

 regarded this group as especially difficult. This is perhaps due 

 to the fact that the telia and teliospores do not usually present 

 diagnostic characters. In the characterization of these species 

 it has been necessary to look for other features, and as has been 

 the case in several other groups of the rusts, it has been found 

 that the urediniospores have furnished reliable characters in many 

 instances. The aecial connections as brought to light by cultures 

 are also important. The presence of amphispores in some of 

 these species, the correlation of some species with forms usually 

 regarded as belonging to another genus, and the splitting of other 

 species into races lend general biological interest to these studies. 



In order that the results may be made as available as possible to 

 those wishing to collect, or determine specimens of the group, con- 

 siderable attention has been given to the preparation of keys and 

 indices. The key relating to the aecial forms is somewhat unique 

 and yet it is believed that even a cursory glance will reveal the 

 way in which it may be useful. It is the idea that anyone having 

 an aecial stage on a known genus of host should be able to make 

 out whether or not it belongs to a Carex rust, and if so, to locate 

 the species without difficulty. No host bears the aecia of more 

 than one species. Altogether thirty genera serve as aecial hosts 

 distributed among ten families, only one of which is monocotyl- 

 edonous. Eight orders are represented. The following table 

 shows the distribution of the aecial hosts. 



i Read before the Botanical Society of America at the New York meeting, 

 December 30, 1916. Contribution from the Department of Botany, The 

 Pennsylvania State College, No. 10. 



205 



