2 BULLETIN 629, L T . S. DEPARTMENT OF AGRICULTURE. 
the stem rust of oats, Puccinia gramirds avenae Erikss. and Henn. 
(PL I, fig. 1, and PL III). These rusts are both rather common in 
the United "States. The stem rust probably causes the greater loss 
in the Northern States and the crown rust in the Southern States. 
THE CULTURE OF CEREAL RUSTS IN THE GREENHOUSE. 
Obligate parasitism is a well-known characteristic of the rust 
fungi, and all attempts to grow them in artificial media have failed. 
They are easily cultured on the living host, however, and, as pointed 
out by Carleton (2), 1 it is possible to make many interesting studies 
.by this method. Carleton described the method of inoculation used 
by him and gave suggestions concerning work with rusts. 
Evans (4) has tested the rust resistance of oat varieties in the 
greenhouse and concludes that the Indian varieties are far more sus- 
ceptible to the crown rust than to the stem rust. 
Melhus (8) has described and illustrated the apparatus used and 
has given his methods in culturing parasitic fungi on the living host, 
including notes on culture work with the crown rust of oats. 
Fromme (5) has published a comprehensive paper dealing with 
the culture of cereal rusts in the greenhouse, in which he briefly 
reviews the work of previous writers and describes in some detail 
his own methods, particularly the results obtained with the crown 
rust of oats. 
Stakman (12) briefly describes culture methods used in his green- 
house studies of biologic forms of the cereal rusts and calls attention 
to the effects of temperature, humidity, and light on the incubation 
period. In a second paper (13) he has briefly described similar 
methods used in the additional study of cereal rusts on plants grown 
in the greenhouse. 
Melchers (7) has suggested the use of galvanized-iron tubs with 
window-glass covers as moist chambers. He also advocates wetting 
the leaves to be inoculated by the " finger-rubbing " method instead 
of with an atomizer and keeping the pots bearing the rusted leaves 
in shallow pans of water, to avoid the necessity of overhead watering. 
EXPERIMENTAL METHODS. 
In the greenhouse experiments of the writer, methods similar to 
those above cited were employed. For the infection studies on seed- 
ling oat plants 4-inch pots were used, and 5-inch pots for the plants 
inoculated at heading time. Greenhouse potting soil of approx- 
imately the same make-up was used throughout the work. For the 
studies of plants at heading time four see'ds of a variety were sown 
in each pot. When the plants were 6 to 8 inches high they were 
thinned to two in a pot, the number usually reserved for subsequent 
inoculation. 
a The serial numbers iu parentheses refer to " Literature cited," p. 16. 
