32 THE RUSTS OF GRAINS IN THE UNITED STATES. 



teleutospores IVoiii Agropyroti repens Beauv. and Poa pratensis L. 

 would give rise to the aecidia on Berberis and, in 1865, that aecidio- 

 spores from Berberis sown on rye would produce uredospores and 

 later teleutospores. He did not stop here, but kept at work on other 

 rusts, and in 1865 showed that Puccinia coronata has its secidium on 

 Ehamnusfrangula and P. ruhigo-vera its secidium on Lycopsis arvensis. 

 From this time on, life-history work on the Uredinese has made rapid 

 strides and the connection of one secidial form after the other has 

 been discovered by such men as Oersted, Fuckel, Magnus, Schroter, 

 Wolff, Rostrup, Winter, Nielsen, Reichardt, Hartig, Rathway, 

 Cornu, Plowright, Farlow, Barclay, Thaxter, Eriksson, lOebahn, 

 Arthur, Holway, Kellerman, and others.^ 



In 1884 Plowright produced successful infection on Berberis ^\-ith 

 rust from teleutospore material from Agropyron repens sent by 

 Arthur from the United States, and Bolley produced successful 

 infection on barberry with teleutospore material from wheat in 

 1889 (l,p. 395). Carleton (30, p. 54) produced successful infection 

 on barley from secidiospores from Berberis. These experiments 

 have been repeated by Arthur time and again and have been per- 

 formed by the authors. They prove that the Puccinia gmminis in 

 America is the same species as the P. gmminis in Europe, although 

 the physiological speciaUzations and consequent biologic forms are 

 different in the two countries. 



More work has been done on Puccinia graminis than on any other 

 rust fungus, and its relationship to the aecidium on several species of 

 Berberis has beenproved repeatedly by Eriksson. Species of Mahonia 

 have also been proved to be secidial hosts of tliis rust (85, p. 234; 



13, p. 96). 



The relationships of Puccinia graminis on oats and rye to the secid- 

 ium on species of Berberis and the relations of P. coronata to the 

 secidium on species of Rhamnus have also been demonstrated, while 

 the secidial forms of the other leaf rusts are not known. This has 

 led again to the mooted question whether or not the secidial stage 

 is necessary in the life liistory of rusts, and, if not absolutely nec- 

 essary, what function the secidial stage fills. 



FUNCTIONS OF THE ^CIDIUM. 



General Discussion. 



As early as 1882 Plowright (85, p. 234) questioned whether the 

 secidium is an essential stage in the life history of rusts and grasses, 

 and gave as his principal reason for raising this question the apparent 



1 For citations of literature see Plowright (84), 1889; Erilcssou and Henning (39), 1894; Klebahn (63), 

 1904; McAlpine (70), 1900; Arthur (2, 4-11), 1899-1909. 

 J 10 



