56 BULLETIN 727, U. S. DEPARTMENT OF AGRICULTURE. 
OVERWINTERING. 
Previous references have been made to the subject of overwintering, 
and it now remains to present the evidence relative to this phase of 
the problem. Sheldon (46, pp. 127-137) noted that anthracnose of 
watermelons was. more severe where the crop was grown on the same 
ground or near the same ground used the previous year. Taubenhaus 
(51) made similar observations in Delaware and states that a 6-year 
rotation is thus necessitated. 
In the course of the present work, much observational evidence has 
been accumulated during disease-survey trips, in which, however, 
only the statements of growers were available regarding the previous 
history of the fields. Concerning the 17 fields in which anthracnose 
was found near Sparta, Wis., in 1915, it was learned that in the case 
of 5 of these, cucumbers were grown on the same soil in 1914 and that 
in the case of 3 others the 1914 crop was adjacent to the 1915 field. 
The coldframe field near Norfolk, Va., previously mentioned, where 
anthracnose was epidemic in 1917, was in cucumbers in 1916 and the 
disease was present that year as well. 
At Madison, Wis., anthracnose was prevalent-in fields 1 and 2 in 
1916. In field 1 a-late epidemic left numerous diseased fruits as 
well as vines, which were plowed under late in the fall. In 1917, 
cucumbers were again planted in these fields, as well as the four men- 
tioned in connection with the seed treatments. Field 2 was planted 
in part with seed treated with mercuric chlorid and in part with 
hand-thrashed seed from cucumbers free from anthracnose. Seed 
treated with mercuric chlorid was employed for all of field 1 except 
for one set of plats used to test out a variety of seed treatments,. 
Among the latter were untreated control rows, but these were planted 
with Iowa seed which was proved to be free from anthracnose by 
the Michigan field trials previously described. The introduction of 
anthracnose into either of these fields with the seed is therefore 
unlikely. 
In field 2 no anthracnose developed in 1917. In field 1, where 
much more diseased material was plowed under in 1916, anthracnose 
appeared, not in ‘‘original centers,” but as rather scattered infected 
plants in atleast three areas where diseased material was known to 
have been abundant the year before. No anthracnose developed in 
the three fields where treated seed was planted in soil not previously in 
cucumbers. The recurrence of anthracnose in field 1 under these con- 
ditions is taken as rather convincing evidence of overwintering in the 
soul. 
In the course of inspection of the fields in the cooperative seed- 
treatment tests it became further apparent that the disease over- 
winters in the soil. It has been indicated in Table VIII that all of 
the outbreaks of anthracnose among the Indiana fields planted with 
a 
