22 Uredospores. 



interfered with, as in the case of ithe teleutospores on the stem. But 

 the uredo may become inured to unfavorable conditions, such as drought 

 or cold, ;unl <\irrv on the life of the species, independent of the teleuto- 

 spore. This is well seen in Pucdnia poarum, for in both Europe and 

 America it has been found on the leaves of Poa pratensis even after the 

 melting of the snows, and in Australia it occurs on Poa annua throughout 

 the winter months, the rust disappearing with the withering of the host, 

 which generally happens early in October. An extreme case seems to 

 have been reached in P. vcxans, Farl., where, in addition to the ordinary 

 uredo, there is a specialised form to which the special name of amphispore 

 has been given, which is thick- walled, strongly papillate, and only ger- 

 minates after a period of rest. 



It is quite common for the uredo-layer to be attacked by the parasitic 

 fungus, Darluca filum, Cast., so much so that it has been found upon 24 per 

 cent, of the species of Pucdnia. It is somewhat unfortunate that Dr. Cobb" 

 has confounded this parasite with spermogonia, producing spermatia, for 

 in referring to peach rust, he writes: "I frequently find among the 

 uredospores of a pustule of this rust, small black pycnidia, producing a 

 multitude of two-celled spores, which, when placed in a moist chamber, 

 often bud and multiply after the manner of yeast plants, but which occa- 

 sionally produce a mycelium. These two-celled bodies have, as I have 

 on several occasions publicly remarked, no slight resemblance to the so- 

 called spermogonia of several species of Aecidium." Although spermatia 

 are well-known to be unicellular, yet Carleton 2 quotes this authority for the 

 statement that ordinary germ-tubes are produced in the germination of 

 spermatia as well as in the other spore-forms. This rust parasite is very 

 commonly distributed, attacking the mycelium and probably checking the 

 development of spores. It occurs on aecidia, uredo, and teleuto-layers, 

 and is recorded on Uromyces (9), Uromycladium (i), Pucdnia (22), Phrag- 

 midium (i), and Aecidittm (2). 



Paraphyses most commonly occur in connexion with the uredosori, and 

 are found in Australian species of Puccmia, Melampsora, and Phragn:idi::m 

 as well as in Uredo. 



Occasionally they arise in both uredo and teleuto-sori, as in Pucdnia 

 magnusiana and P. purpurea, and sometimes they are variable in their 

 presence as in Pucdnia poarum, where Plowright found none in Britain, 

 although they are common in Australian specimens. In Phragmidiuin 

 subcortidum not only are the uredosori provided with paraphyses, but like- 

 wise the aecidial patches, since they are without a surrounding membrane. 



The following are the known Australian species, with paraphyses in 

 tneir uredosori : Pucdnia lotii, P. magnusiana, P. poarum, P. ' pruni, 

 P. pnrpurca; Melampsora hypericorum, M. lint; Phragmidium barnardi, 

 P. subcortidum; Uredo kuehnii, and U. spyridit. 





