LIFE HISTORIES OF RUSTS. 27 



of the original material) the wheat rust on barley had lost from the 

 original material 0.63 /« in Avidth (practically negligible) and 2.32 fx 

 in length, while the barley rust on wheat had gained 0.21 n in width 

 (again practically negligil)le) and 2.61 /« in length. If these measure- 

 ments are compared with those of the original material it will be 

 seen that the wheat rust on barley has decreased in spore size to 

 almost exactly that of the original barley rust and the barley rust on 

 wheat has increased in spore size to nearly that of the original rust 

 on wheat, as follows: 



Original wheat rust 18. 15 /i by 31. 33 /z. 



Barley rust after 10 months on wheat 17. 67 /i fcy 31. 12 /i. 



Original barley rust 17. 46 ju by 28. 51 /t. 



Wheat rust after 10 months on bariey 17. 52 /^ by 29. 01 /z. 



Although these differences are not great, they seem sufficient to 

 indicate that the host plant exercises not only a decided physio- 

 logical and biological reaction upon the parasite but that it may, 

 even in such a short period as one year, exert a measureable effect 

 on the morphology.^ It has already been shown (p. 17) that wheat 

 rust if first transferred to barley may be transferred to oats with 

 considerable ease, thus showing the physiologic change going hand 

 in hand with the morphologic change. 



GENERAL SUMMARY OF CONCLUSIONS DERIVED FROM EXPERIMENTS ON BIOLOGIC 



FORMS. 



In summarizing, the following points in regard to biologic forms of 

 rusts of cereals may be emphasized: 



(1) Puccinia graminis tritici Erikss. and Henn. (stem rust of wheat), F. graminis 



hordei F. and J. (stem rust of barley), P. graminis secalis Erikss. and Henn. 

 (stem rust of rye), and P. graminis avenae Erikss. and Henn. (stem rust of 

 oats) are undoubtedly biologic forms of the same species Puccinia graminis 

 Pers. 



(2) These forms are not entirely confined to their respective hosts, but vary in 



range in part according to the host plants they have been recently inhabiting. 



(3) P. rubigo-vera tritici Carleton (leaf rust of wheat) and P. ruhigo-vera secalis Carle- 



ton (leaf rust of rye) are more highly specialized than the corresponding 

 stem rusts. 



(4) P. graminis hordei (stem rust of barley) has ordinarily the widest range, while 



Puccinia simplex Erikss. and Henn. (leaf rust of barley) and P. rubigo-vera 

 secalis (leaf rust of rye) have more restricted ranges. 



(5) Under very favorable conditions, particularly after first transferring to barley, 



all the stem rusts can be carried successfully to the other cereals. 



(6) WTien the rusts are transferred to uncongenial hosts and produce pustules on 



these, the pustules are almost invariably minute and weak, producing com- 

 paratively few spores. Some pustules apparently never open. The con- 

 genial hosts of each rust may be summarized as follows: 

 P. graminis tritici (stem rust of wheat) on wheat and barley. 



' Evans (50, p. 461) has shown previously that many of the biologic forms of the genus Puccinia can be 

 distinguished by slight differences in morphology of the early uredo mycelium, particularly in the forma- 

 tion of the substomatal vesicle. 

 •2\Q 



