Host and Parasite in certain Diseases of Plants. 



441 



He also f ound that the conidia germinate, by developing a germinal 

 tube only at a given period, and not at any indefinite time they may 

 happen to be sown : consequently they are unable to infect the host 

 unless they happen to be on the proper spot at the right time. 



Now Kiihn showed long ago* that the infection, of the cereal by 

 Ustilago can only occur near the " collar " of the young germinating 

 seedling, and although differences have arisenf as to the exact spot 

 in this region which alone can be infected, there is one point on which 

 all are agreed, namely, that the germinal tube can only enter the 

 young actively growing tissues in that region. Immediately the first 

 internode is completed, the seedling is proof against infection in the 

 open. J 



That it is really a matter of the age and condition of the tissue 

 is beautifully demonstrated by Brefeld, who showed that if the 

 conidia are forced into the bud by means of a syringe, so that they 

 can germinate in contact with the embryonic cells at the growing 

 point, infection may be ensured at any time. 



But far the most interesting point about these fungi is that when 

 the germinal hypha is once inside the tissues it goes on growing with 

 them, keeping between the cells. Although we have almost no infor- 

 mation as to details here, it can hardly be doubted that the chief 

 agent in maintaining the balance of position in this case is the living 

 substance in the cells of the host ; but I know of no explanation for this 

 beyond the general one, that so long as the cells of the internodes are 

 actively performing their normal functions, the hypha3 have to be 

 contented, so to speak, with a sort of suppressed existence in the 

 intercellular spaces and middle lamella of the walls. 



True, when the young fruits begin to fill out, the mycelium accom- 

 plishes a sharp revenge by destroying the whole fruit ; but it is 

 obvious that the relations which determine the epidemic or sporadic 

 character of the disease are those between the tissues and the germinal 

 hypha? and young mycelium, and there are great variations in these 

 matters, even in the group of the Ustilaginece itself. § 



Very different again must be the relations in detail between host and 

 parasite in the case of those Uredinece which cause epidemic leaf diseases, 

 especially those which form haustoria,|[ and it is almost impossible 



* ' Krankh. d. Kulturpflanzen,' 1859, p. 46. 



f Cf. Wolff, ' Brand des G-etreides,' Halle, 1874 ; Hoffmann, ' Bot. Untere 1866 ; 

 ' Ueber den Flugbrand,' p. 206 ; Kiihn, " Beobachtungen ii. d. Steinbrand d. 

 Weizens" ('Oesterr. Landw. Wochenschr.,' 1880). 



X This no doubt explains the fact that in a wet spring nearly all seeds with spores 

 attached become infected, because the tissues remain in a youthful condition longer 

 than in a dry season. 



§ 'E.g., contrast the behaviour of Protomyces, JEntyloma, and Ustilago Maidis, 

 for instance, with that of most other Ustilaginece. 



|| In all cases where haustoria are developed the mycelium enters into a peculiar 



VOL. XLVII. 2 K 



