480 Journal of Agricultural Researck voi. x, N&.9 



pallidum subvar. pyramidatum, H. spontaneum K. Koch, Hystrix patula 

 Moench., Lolium italicum R. Br., L. perenne L., L. temulentum h., Phleum 

 pratense L., Secale cereale L. 



Inoculated but not infected: Agropyron caninum (L.) Beauv., A. 

 desertorum Schult., A . elongatum Host., A . imbricatum Roem. and Schult., 

 A. mtermedium Beauv., A. repens (L.) Beauv., A. smiihii Rydb., A. 

 tenerum Vasey, Bromus inermis Leyss., B. pumila, Cynodon dactylon 

 (L.) Pers., Cynosurus cristatus L., Danthonia spicata (L.) Beauv., 

 Desckampsia flexuosa (L.) Trin., Elymus virgimcus L., Festiica rubra L., 

 Hordeum jubatum L., H. spontaneum K. Koch, H. vulgar e pallidum Ser. 

 H. vulgare pallidum subvar. pyramidatum, Poa compressa h., Savastana 

 odorata (Iv.) Scribn., Triticum vulgare Vill. 



Carleton (6, p. 63-64) gives the following as hosts for stemrust of oats: 

 Avena sativa patula, A. sativa orientalis, and A. sativa nuda (cultivated 

 varieties and Dactylis glomerata and Arrhenatherum elatius. The fol- 

 lowing he considers as probable hosts: Avena fatua, A. hookeri, A. 

 pratensis, A. sferilis, Koeleria cristata and Lolium perenne. The writers 

 have shown that A. fatua, and Koeleria cristata are undoubtedly hosts and 

 that Lolium perenne can be infected, although whether or not it is com- 

 monly a host has not been determined. Carleton's results and those of 

 the writers are in general agreement with those of Eriksson (11, p. 601) 

 in Sweden, although the same species were not used in inoculation 

 experiments in all cases. Jaczewski (18, p. 353), on the other hand, 

 arrived at different conclusions as a result of his work in Russia. The 

 most striking difference is that h€ gives Dactylis glomerata as immune, 

 whereas in this country it is one of the very common hosts. Eriksson 

 and Henning also give it as one of the hosts in Sweden. It would be 

 interesting to knov/ whether the rust is really different in Russia or 

 whether for some reason D. glomerata did not become infected in the 

 limited number of trials which Jaczewski made. 



A number of investigators have called attention to the versatility of 

 the oat stemrust. There is no doubt that it can infect many different 

 grasses. 



It is interesting to notice the varying degrees of virulence of the rust 

 on different hosts. It is extremely virulent on such hosts as Avena 

 sativa, A. fatua, and Dactylis glomerata, while on Alopecurus pratensis 

 it is moderately virulent, on Holcus lanatus a little less virulent, on 

 Agrostis alba still less virulent, on Hordeum vulgare and Secale cereale it 

 develops very imperfectly, and on Triticum vulgare it scarcely ever even 

 produces flecks. 



The stemrust of oats is similar to P. phleipratensis and P. graminis 

 agrostis in its ability to infect plants in such varying degrees. (See PL 

 57-59-) The other biologic forms do not seem to behave in this way. 

 The wheat stemrust and the rye stemrust sometimes infect plants weakly, 

 but, as a rule, they either cause heavy infection or none at all. 



