558 
G. R. BISBY 
west; in the Uromyces, over a greater area of North America, but 
limited to fewer hosts. Uromyces plumharius also shows less vari- 
iation than does Puccinia Epilohii-tetragoni; this is rather to be 
expected. An apparently complicating factor is that of the occurrence 
of this Puccinia upon various species of the genus Epilobium in some 
part of every continent, doubtfully excepting Africa. This is, however, 
in line with other relations of these rusts to the genus Epilobium, as is 
noted elsewhere. The variable American hosts are followed by variable 
rusts; this variability is further indicated by the fact that species of 
Uromyces occur only in America. Indirectly, it seems to the writer, 
the sharp difference in host genera attacked by the two above rusts 
reflects a- considerable accuracy of taxonomic arrangement of hosts. 
Puccinia luxurians and Puccinia Krookii, as stated, may be in- 
cluded with Puccinia Epilohii-tetragom. 
The relations of Puccinia Veratri, with aecia upon Epilobium, and 
of the other species of this group, has been indicated, and scarcely 
needs further comment. Puccinia Veratri has a more extended dis- 
tribution than has Puccinia Epilohii-tetragoni. What conclusions 
are to be drawn from the comparatively greater frequency, but more 
limited distribution, of Puccinia Epilohii-tetragoni, as compared with 
the evident rarity, yet, in America, broader distribution of the short- 
cycled Puccinia Epilohii and Puccinia scandica, the writer is not 
prepared to say. 
Into the third group may be placed Puccinia Jussiaeae, Puccinia 
Circaeae, Puccinia gigantea, Puccinia Fuchsiae, and Puccinia Peckii. 
The teliospores in this group are easily distinguishable from those of 
the group mentioned in the preceding paragraphs, being longer and 
narrower, often paler. Here, too, this similarity exhibited is close 
enough through all the species of this group to suggest a relation. 
Puccinia Peckii bears the same relation to this group that Puccinia 
Veratri does to the preceding group. As Puccinia Veratri is more 
extensive in range, in North America, than is Puccinia Epilohii' 
tetragoni, so Puccinia Peckii has also a more extended range than has 
Puccinia Jussiaeae. The hosts of the two latter species are in no 
cases identical, in spite of an evident relationship. This third group 
lacks, as far as known, a representative with uredinia upon the 
Onagraceae. • While, of the short-cycled species in this group, Puccinia 
gigantea and Puccinia Fuchsiae appear to be more rare, Puccinia 
Circaeae is common. 
