Arthur : Amphispores of grass and sedge rusts 



39 



globose uredospores, with rather thin brownish yellow wall, finely 

 verrucose, and with about six scattered germ pores. 



The amphispores are similar, but with very much thicker walls, 

 deeper colored, somewhat more coarsely verrucose, and with per- 

 sistent pedicels. They are very rarely collected, and are only 

 known to the writer from Colorado, on what is probably Sporoholiis 

 airoides. The type of Uromyces scaher was given as ''an unknown 



," and has not been seen by the writer, but is clearly the 

 amphisporic stage oi this species. The species ranges from Mon- 

 tana and South Dakota to Texas and Arizona. 



5. Puccini A Cryptandri Ell. & Barth. {Uromyces shmilans 

 Pk.) rarely presents any other form of uredo than the amphispore. 

 If the usual thin-walled uredospore does occur as a clearly marked 

 form, it has been rarely collected. There is no collection of the 





(i 



1, 





* J 



IE 



f 1 



V 



X2 



6 



Figure 5. Puccinia Cryptandri E. &: B.; on Sporobehis cryptandnn^ etc. 

 Central United States. IIj Uredospores. X, Amphispores. 



Figure 6. Uromyces Rottboelliae Kx\ki,\ on RoitboelUa speciosa. Northern India, 

 alt, 2ICO m. II, Uredospores. X, Amphispores. 



species in the writer's herbarium e^athered earlier in the season 



t> 



than the middle of August. Spores from a collection made 

 August 21 are figured as the summer form of the spore. The 

 walls are thin, and the spore appears paler and less firm than in 

 collections made in October and November. The usual autumn 

 collections have the characteristics of amphispores. The sori are 

 dark cinnamon-brown, with apparently persistent spores, although 

 in transferring them to the microscope they separate readily from 

 the pedicels. The type of Uromyces sinudans has been examined, 

 and the amphisporic character of the spores is especially marked. 

 Some experimental evidence that these are resting spores, and 



