46 RELATION OF PARASITE TO SUBSTRATUM. 



another host. Such heteroecious parasites may be, however, also 

 luonoxeiious ; for example, Mclam^ysora Goeppcrtiana has its 

 teleiitospore-form only on the cowberry, its aecidium-form 

 only on the silver tir. On the otlier hand, ChnjHomijxa 

 rkododendri frequents several species of Ehododendron, while 

 the aecidia occur only on Picca cxcelsa ; Cronartium asclcjjicidcum 

 comes on both Gentiana and Cynanchum, the aecidial stage 

 only on Finns sylvestris. With Gymnosporanginm clavariacforme 

 this condition is reversed, the teleutospore-form occurs only on 

 Junipcrus co)anumis, the aecidial on various species of Cratacyns 

 and other genera. 



The effect of various substrata on the development of any 

 fungus may Ije most conveniently investigated : {a) on facul- 

 tative parasites and saprophytes, {h) on polyxenous species of 

 fungi, (c) in cases where the fungus inhabits essentially different 

 organs or tissues of the same host. 



The most obvious effect of the substratum is presented during 

 the germination of spores. The spores of most parasites ger- 

 minate in water. Those of certain smut-fungi, especially in 

 tlie fresh condition, will not germinate at all, or only to 

 a very limited extent in water, whereas they will do so 

 immediately and unanimously on being offered a nutritive 

 solution. Tilktia, a genus of Ustilagineae, behaves, however, 

 in (piite the reverse way, it germinates only in water, and 

 refuses to do so in nutritive solutions. Hartig found that the 

 spores of dry-rot {Mcrulim) would neither germinate in water 

 nor in tlie usual nutritive solutions, but that they did so at 

 once on adding alkalies to the water, such as those supplied 

 by addition of urine. Very characteristic is the behaviour 

 of these spores, which only germinate in contact with their 

 host-plants, like many Chytridieae ^ (Synchytrium), as well as 

 Complctoria and Protomyccsr Others again send out germ-tubes 

 which remain small and soon die away if an immediate 

 opportunity of penetration into a host is not presented. De 

 Bary states this to be the case with swarm-spores of Cystopas, 

 Peronospora nivea, Erysipheae, etc. Amongst the Uredineae, the 

 germ-tubes are short-lived ; they will penetrate into almost any 



1 De Bary, Morphology and Biology of the Fuiigi, cliap. vii. 



- An exactly parallel case is presented by the seed of Orobancheae, wliich 

 germinate only in contact with the roots of their host (Koch's "Orobanchen," 

 Heidelberg, 1887). 



