366 
Journal of Agricultural Research 
Vol. XXV, No. 9 
In 1917, and again in 1922, Hordeum murinum was found to be one of 
the most common hosts for stripe rust along the Pacific Coast in southern 
California. Very seldom was any other host found infected, although 
other possible hosts often were growing in close proximity. 
EXPERIMENTAL METHODS 
The greenhouse experiments in the study of specialized varieties of 
Puccinia glumarum , as well as in the testing of wheat varieties for suscep¬ 
tibility to stripe rust, were carried on at Corvallis in a small wing of the 
greenhouse of the Oregon Agricultural Experiment Station. The methods 
used in greenhouse inoculations were, with certain modifications, like 
those employed by Stakman and Piemeisel {18) in their stem-rust studies. 
Every precaution was taken to guard against chance infections. 
Seedlings were used with few exceptions, and were inoculated on the 
first or primary leaf. Plate 1, A and B, shows the method of preparing 
the seedlings for inoculation and the devices used for obtaining favorable 
conditions for infection. In every case the plants in a given pot were 
divided into two groups before any inoculations were made. This was 
done usually by pulling up or cutting close to the soil those in a line 
through the center of the pot. A piece of string was then laid on the soil 
between the two groups of seedlings. One half of these were inoculated 
and the other half left as control plants. In only one or two instances 
did any infection develop on the controls, and then only on a plant 
immediately adjacent to the inoculated ones. In such cases the results 
were not considered. 
Inoculations were made by carefully transferring fresh urediniospores 
to the upper side of the leaf by means of a small scalpel. Melchers (13) 
found that he could obtain satisfactory infection on wheat with Puccinia 
graminis tritici E. & H. by inoculating the under side of the leaf. This 
did not prove to be the case with Puccinia glumarum tritici. Several trials 
were made by taking wheat plants of the same age and variety and inocu¬ 
lating part on one side of the leaf and part on the other. Only a very few 
infections resulted from the inoculations made on the under side, while 
nearly 100 per cent of those inoculated on the upper side of the leaf 
became infected. 
At first, bell jars and battery jars were used for incubation chambers. 
Eater, however, shallow tubs covered with a window sash were used to 
good advantage. After inoculation, the pots were placed in a tub con¬ 
taining about 2 inches of water, covered, and left in this moist chamber 
for 48 hours. They then were removed to the greenhouse bench. The 
various strains of the rust were kept in separate compartments in the 
greenhouse, and were isolated by means of partitions made of fine cheese¬ 
cloth. 
RESULTS 
Inadequate greenhouse space made it impossible to work with a 
large number of collections of stripe rust. All of those studied proved to 
be Puccinia glumarum tritici . Each infected wheat readily, rye slightly, 
barley very slightly, if at all, and oats not at all. Table I shows the 
results of inoculations made on various varieties of barley and rye with 
urediniospores from wheat. As barley has been found heavily infected 
in the field several times, it appears that there must be a distinct special- 
