juneis, I9IS Pv£cinia Graminis 195 



EXPERIMENTAL METHODS 



The plants used for inoculation were grown in 4-inch pots in the green- 

 house. When they were 6 to 8 days old they were inoculated with an 

 ordinary flat inoculating needle, special care being taken not to injure 

 the tissues in any way. After inoculation the pots were placed in pans 

 containing a little water and then covered with bell jars for 24 or 48 

 hours, or, in a few cases, longer. The inoculated portions were placed 

 in killing fluid at periods ranging from 48 hours to 12 days after inocula- 

 tion. Flemming's weaker solution, medium chromo-acetic acid, and 

 picro-acetic acid were used at various times. The material was em- 

 bedded in paraffin in the usual manner and was cut from 6 to ii/z thick 

 and then stained. For staining, Harper's modification of the triple 

 stain and Gram's stain, with a counter stain of eosin, gave the best 

 results. 



HISTOLOGICAL DETAILS OF HYPHAL INVASION 



It is probably superfluous to call attention to the fact that when nor- 

 mal infection, such as occurs, for example, when oats are inoculated 

 with P. graminis from Dactylis glomerate, takes place, the host cells 

 remain at least apparently normal for a considerable length of time. 

 In the early stages of infection, although the leaf tissues in the infected 

 area may become yellowish in color, it is clearly evident to the naked 

 eye that no extensive and rapid killing of tissues is taking place. When 

 such areas are carefully examined in section, the host cells very fre- 

 quently appear entirely normal. (PI. XXVIII, fig. I.) The hyphae may 

 grow very vigorously, send haustoria into the cells, and branch profusely 

 without destroying the chloroplasts or in other ways injuring the cell. Even 

 after pustule formation has begun, many of the cells just at the edge of 

 the pustule, where the hyphae are massed in great numbers, still retain 

 their chloroplasts and are apparently normal in other respects (PI. 

 XXVIII, fig. 2). It not infrequently happens that cells just under a uredi- 

 nial sorus, even when the fungus has sent numerous haustoria into them, 

 still retain a number of chloroplasts and seem to have suff'ered no serious 

 injury. Of course, dead cells are found in a heavily infected region, but 

 in no case does the fungus seem to kill quickly and sharply the cells with 

 which it comes in contact. Normal infection has been described and 

 illustrated quite completely by Ward (9), Evans (2), and the writer (8). 



When examination is made, however, of the tissues of a plant that has 

 been inoculated with a rust form which does not grow in it normally, 

 very sharp differences are observable within 48 hours after inoculation. 

 Even in the very early stages it is very evident that normal infection 

 is not taking place and that there is a comparatively violent action and 

 reaction between the plant and the parasite. 



The sequence of events in resistant or immune forms is very nearly the 

 same in the different forms studied. The germ tubes form appressoria 



