July 1906] Better Classification of the Uredinales 
151 
side the peculiar urediniospores also teliospores with verrucose, 
globoid cells and fascicled pedicels, for which Puccinia Pruni- 
spinosae is a good illustration, and still another group of genera 
having teliospores borne one or more on free pedicels, and the 
spores often flattened above and below, for which Uromyces 
brevipes, the rust on Rhus, is representative. 
Having removed these groups of genera related to Rave - 
nelia we shall still have left species with urediniospores of the 
usual appearance, but with sori surrounded by numerous para- 
physes. Among these we shall find a group of genera with pecu¬ 
liarly tuberculate teliospores having lateral germ pores, clearly 
related to Phragmidium, and still another group of genera in 
which the teliospores possess a Hygroscopic layer between the 
outer and inner parts of the wall, clearly related to Uropyxis. 
This latter group is still further separable into genera with 
lateral pores like Uropyxis, or with apical pores like species 
of Puccinia having subepidermal pycnia. 
Having now removed a large number of species from the 
parallel genera Puccinia and Uromyces, and segregated them 
into groups of genera related variously to Ravenelia, Phrag¬ 
midium and Uropyxis, let us look at what remain, all of which 
have flask-shaped pycnia arising from beneath the epidermis. 
We can easily discover here two groups of genera, one hav¬ 
ing indefinitely extended aecia and colorless teliospores, germ¬ 
inating in the sorus as a rule, of which Puccinia evadens found 
on Baccharis is a representative, and the group of genera hav¬ 
ing definite aecia and colored teliospores, embracing all that 
is left of the genera Puccinia and Uromyces, of which most 
grass and sedge rusts, Puccinia Helianthi, etc., are representa¬ 
tives. 
Having now segregated the species usually placed under 
Puccinia and Uromyces into seven groups of genera with affini¬ 
ties extending through the whole length of the Pucciniaceae, 
let us resolve these several groups into their respective genera. 
In order to do this it is necessary to take a glance at the prob¬ 
able scope of the influences which have determined the devel¬ 
opment of the genera in the Uredinales. It seems highly prob¬ 
able that in general the influences which have acted to limit 
and shape the species and also the genera of higher plants, such as 
temperature, humidity, elevation, natural barriers, succession of 
seasons, etc., have also had similar effects upon the species of the 
rusts. In addition to these a set of influences have been brought to 
bear by virtue of their strict parasitism, which are scarcely to be 
paralleled in any other group of plants. This is shown in limit¬ 
ing the species to certain' orders, genera, or even species of 
hosts. How far the host has reacted upon the rust to modify 
its form and structure is difficult to decide, but that such action 
has occurred there seems to be no occasion for doubt. This 
