348 UREDINEAK. 



present; the palisade layer is doubled, and rupture of the 

 epidermis takes place ; chlorophyll-formation is suppressed, the 

 cell-sap becomes yellow, and starch tends to accumulate. 



P. dispcrsa may cause serious damage to wheat and rye ; P. 

 rubigo-vcra, also on barley and oats. The spore-patches are 

 found on stalks and leaf-sheaths more than on the lamina. The 

 mycelium may hibernate in grasses, so that the fungus is not 

 dependent on the aecidial stage ; for this reason the disease is 

 not easily combated. 



P. glumarum Eriks. et Henn. Golden-rust. This species, 

 hitherto generally included under P. rubigo-vera (D. C.) has been 

 separated by Eriksson and Henning. 1 Experimental infection 

 on Boragineae gave negative results. 



Eriksson distinguishes the following specialized varieties of 

 this species : 



A. Definite (and undoubtedly distinct). 

 1 . Var. tritici on Triticum vulgare. 



'2. Var. hordei on Hordeum vulgare (somewhat uncertain). 



3. Var. dyini on Eli/ nuts arcnarius. 



4. Var. agropyri on Triticum rcpens. 



B. Not sharply defined : 



5. Var. sccalis on Secede ccrculr. 



The uredospore-sori are lemon -yellow in colour, and form 

 lines on the leaf-blade which may run together and reach a 

 length of 10 mm. The teleutospore-sori form long, fine, brown 

 or black lines : the sori are divided into numerous chambers, 

 each enclosed in a circle of curved brown paraphyses. The 

 spores germinate in the autumn of the same year. The pro- 

 mycelium is yellow till the spores are abjointed ; in this way 

 it is distinguished from P. dispcrsa. 



P. poarum Xiels. (Britain). Uredo- and teleutospores on 

 /'"it. According to Xielson, the aecidia occur on Tussilayo, 

 /V/V.s/7r.s, and Adcnostylcs. Fentzling (lor. <//.) lias described 

 certain anatomical changes which accompany deformations due 

 to the aecidia. 



P. phlei-pratensis Eriks. et Henn. This has a hibernating mycelium 

 which produces uredospores continuously on I'kleum and probably also 

 on Fi'xtuca. Aecidia have not as yet been observed. 



' Eriksson and Henning (loc. clt. ). 



