42 A STUDY IN CEREAL RUSTS 



in the cells, but that they became disorganized within a very few days. 

 He attributed this to defective symbiotic relations between host and 

 parasite. Miss Marryat (1907) showed that Puccinia glumarum when 

 grown on a semi-immune host plant killed local areas of the host, sent 

 out but few haustoria, and never developed any but small or abortive 

 pustules. 



It is a matter of common observation that in dealing either with 

 cereals uncongenial to a given biologic form of Puccinia graminis or 

 with varieties of wheat resistant to Puccinia graminis tritici flecks are 

 often visible after inoculation, but no pustules, or only small ones, are 

 produced. Examples of this are shown in Plate I, A and Plate VI, A. 

 All degrees of this killing can be observed. The more readily the rust 

 infects a plant the less likely are these dead areas to appear. When 

 Puccinia graminis tritici is put on Minnesota No. 163 wheat, pustules 

 are formed in great abundance, but the leaf tissues remain alive for a 

 long time. When, on the other hand, resistant forms such as Kubanka 

 2094 or Khapli are inoculated, areas of the leaf are killed outright ; and 

 if pustules are formed at all they are very small. In extreme cases of 

 incompatibility such as is found between Puccinia graminis avenae 

 and wheat, the leaf area involved is usually so small that no indications 

 of it can be seen with the unaided eye. 



HlSTOLOGICAL DETAILS OF INFECTION 



In order to determine the behavior of germ tubes in susceptible 

 and nearly immune forms, leaves of Minnesota No. 163 and of Khapli 

 were inoculated with Puccinia graminis tritici. Minnesota No. 163 was 

 also inoculated with Puccinia graminis avenae. They were then placed 

 in a pan of water under bell jars for 48 hours. Leaves were selected 

 and killed every 24 hours, beginning with the first day. 



For killing, aceto-alcohol, medium chromo-acetic acid, and Flem- 

 ming's weaker killing-fluid were used. The leaves were embedded 

 in the usual manner and sectioned from five to ten microns thick. 

 For staining, Haidenhain's iron-alum haematoxylin and orange G, 

 Erlich's haematoxylin, Gram's stain and eosin, Delafield's haematoxylin 

 used according to Durand's method, fuchsin and methyl green and the 

 safranin, gentian violet, orange G. combination, used according to 

 Harper's modification of Flemming's method, were used. The last 

 named gave the best results. 



INFECTION OF MINNESOTA NO. 163 



At the end of 24 hours many of the spores have usually germ- 

 inated, sending out long germ tubes, although some apparently 

 germinate later. Two tubes may be sent out from the same spore, 

 but usually one develops more vigorously than the other. The tube 

 usually follows the epiderm closely; swellings are often found above 



