Apr., 1922] 
RAINES — VEGETATIVE VIGOR OF THE HOST 
201 
with an aecidium. The uredospores of this form are markedly less rounded 
and the teleutosori are much longer than in the case of the wheat and rye 
rusts as I have found them. The rust on Elymus has, however, been classed 
with P. rubigo-vera DC. 
Puccinia poculiformis: 
On Cinna arundinacea, at Jerome Avenue, New York City, September 15 (III). 
An Experiment to Determine the Time and Method of First Appearance of 
the Rust, in which the Source and Condition of the Seed 
as Possible Factors are also Tested 
In investigations of a physiological nature on the cereal rusts it is de- 
sirable that the probability be established that, for the variables being 
compared, the source of the fungus being worked with is constant. This 
point in experimental method assumes particular importance, in investiga- 
tions such as are the subject of the present paper, from the recent demon- 
stration of the variation in physiological properties of strains of rust from 
different locaHties. Assuming that the rust endemic in any locality is 
fairly constant in its behavior (whether the constancy is due to actual 
genetic purity, or to an admixture of strains in constant proportions), a 
possible source of error would be the seed transmission of the rust, making 
the source and condition of the seed a factor in determining the nature of 
the rust. Critical evidence on the question of the role played by the seed 
in the first appearance of rust on cereals in early summer is still lacking. 
As against the feeling of necessity, almost, in the minds of some investi- 
gators, of the assumption of seed transmission of the rusts in explaining 
certain phenomena in the epidemiology of these diseases, must be counted 
the inability to demonstrate with certainty a means of transmission, or to 
observe seed transmission of cereal rusts under controlled experimental 
conditions. 
To test the possible role played by the seed in determining the first 
appearance of rust under field conditions in the vicinity of New York, 25 
patches of wheat, 8 patches of barley, one patch of rye, and one patch of 
oats were planted with seed from widely different sources, and of varying 
age and condition on a plot of ground near the New York Botanical Garden. 
The plantings were made on high, well drained land which had not been 
cultivated since 191 2, when it had been put to corn. The nearest plot of 
cereal was a field of oats about one quarter mile away, and there were no 
other grain fields within a distance of one mile at least. It is not thought 
that this single experiment is of very great significance as to the general 
question of possible seed transmission of rust. It is of interest, however, 
to include such inferior seed as that planted in the first three plots in a test 
as to the time of appearance of the rust. 
The varieties planted, the source of the seed, and the date the rust first 
appeared on the plants are shown in table 2. 
