RUST-RESISTANT VARIETIES OF WHEAT 43 



the wall between two epidermal cells (Plate VII, 3). In these swell- 

 ings, which appear very much like very young appressoria, the pro- 

 toplasm often aggregates more densely than in the other parts of the 

 tube. The germ tube may often follow the epiderm for considerable 

 distances, sometimes for the length of 15 epidermal 'cells, before a 

 definite appressorium is formed. Nearly always when a stoma is 

 reached the tube forms a very definite swelling (Plate VII, 1) which 

 constitutes an appressorium. The protoplasmic contents of practically 

 the entire germ tube are concentrated in this appressorium. It dips 

 down into the stomatal opening and a fine protoplasmic process is sent 

 through to the substomatal space. Here a definite swelling takes 

 place, forming the substomatal vesicle (Plate VII, 4 and 5). Appar- 

 ently, in some cases, the vesicle develops no further. In the great 

 majority of cases, however, the protoplasm aggregates in it and 

 infection-threads are sent out. Very often these infection-threads 

 follow closely along under the epiderm cells and send small knoblike 

 or sometimes flattened haustoria into the host cells. Branching then 

 takes place among the cells of the leaf, many haustoria being sent out 

 and the hyphae developing very rapidly. The threads may grow 

 directly across the substomatal space and branch between the mesophyll 

 cells (Plate VII, 6). This does not seem to be the usual method, 

 however. Nuclear division takes place rapidly to keep pace with the 

 growth of the hyphae. Frequently a number of nuclei are found in a 

 single cell. The entire hypha often retains its protoplasm for a con- 

 siderable length of time. Ordinarily the' tip only remains densely 

 protoplasmic while the rest of the hypha becomes much vacuolated. 

 Very long hyphae are often found growing very vigorously but appar- 

 ently not sending out haustoria. They seem to be in the nature of 

 distributive filaments. The fungus does not seem to spread very far 

 from the point of infection. When large areas of the leaf are involved, 

 a number of points of entry can nearly always be found. On the 

 fourth day an infected area is usually well filled with much branched 

 hyphae. About this time wefts begin to be formed, the protoplasm 

 aggregating in the tips of the hyphae. A dense mass of filaments is 

 formed just beneath the epiderm and the epidermal cells are wedged 

 apart. The large, upright fungus cells, which show the binucleate 

 condition very clearly, begin to form uredospores. This is well under 

 way on the fifth day and by the sixth or seventh day the epiderm has 

 been completely ruptured while spores are being shed in great pro- 

 fusion. It is interesting to note that the host cells, which are often 

 half filled with large knoblike, filamentous or coiled haustoria, are 

 usually still quite healthy at the time pustule formation begins. 



The whole appearance of both fungus and host during the first 

 few days after infection indicates a fairly perfect relation between 



