414 
Journal of Agricultural Research 
Vol. XXVI, No. 9 
infection then spread centripetally as well as centrifugally, and it fre¬ 
quently happened that the majority of the plants reestablished inside 
the old areas during the fall and winter were killed by this retroactive 
invasion. 
The map shown in Figure 4 shows the location of activity in the 9- 
hectare field during the seasons of 1921 and 1922. It will be noted that 
with the great majority of rings the diseased area of the succeeding 
year seems to be merely an extension of the centrifugal spread of the 
previous season. Some of the spots, however, showed no renewed infec¬ 
tion in 1922. The largest areas, which showed indications of having 
been active longest, seemed to show a tendency toward immunity. 
Fig. 4.—Map of 9-hectare alfalfa field, Sacaton, Ariz., showing areas infected with rootrot on September 
30, 1921, and on June 22, 1922. Note the circular form maintained by most of the isolated centers of in¬ 
fection and the position in which the disease activity recurs at the beginning of season. 
RATE OF SPREAD OF ROOTROT CIRCLES 
In the literature there seems to be no record as to the rate of spread 
of rootrot infection. Duggar (i) states that the most rapid spread which 
he observed was in a field of irrigated alfalfa, and Pammel (4) notes that 
areas of considerable extent in cotton fields may be affected in one 
season; but no actual measurements are recorded. On September 30, 
1921, the diameters of three circular spots in an alfalfa field were measured, 
and photographs made of the areas. They were later measured on June 
26 and September 30, 1922. Photographs were taken of the three spots 
on the dates named in the legends for Plates 4, 5, and 6, and approximate 
measurements made. The rates of spread of the three areas are given 
in Table IV. It will be observed that the greatest spread was in circle 
No. 2, where the increase in diameter was 14.4 meters in one year, but 
this was caused by its merging with a neighboring spot. 
