ARTHUR AND MAINS: GRASS RUSTS 415 
to separate and appear like one-celled spores in rows, the envelop- 
ing outer layer of the spore-wall being invisible, having been swollen 
and partly dissolved by the potash. However, if sections are 
stained with eosin, and‘.the spores pressed apart, it is easy to 
discern a delicate outer layer of wall, swelling in water and be- 
coming thicker above than at the sides of the spore (Fic. 2B). 
The presence of this hygroscopic layer explains why the cells 
remain in chains, even after macerating in potash, and doubtless 
why the cells adhere laterally with so much tenacity in an ordinary 
mount. In P. pallescens there is no such hygroscopic layer. 
In the above paragraphs comparison has been drawn especially 
between three tropical rusts of similar but extreme development, 
which are found on grasses having slight relationship, according 
to the classification used by modern agrostologists: Puccinia 
pallescens on Tripsacum, the first genus in Hitchcock and Chase’s 
“Grasses of the West Indies,’’ P. phaksopsoroides on Olyra, the 
fifty-second genus in the same work and Uredo ignava on Bambos, 
the one hundred and tenth and last genus of that work. The 
telia of the last species have not yet been found, but it is con- 
fidently believed that when discovered they will resemble those of 
the other two species named. 
PURDUE UNIVERSITY, 
AFAYETTE, INDIANA 
