New Yorke AGRICULTURAL. EXPERIMENT STATION. 183 
In another experiment made the past season four similar appli- 
cations of paris green increased the yield about 46 bushels per 
acre through partial prevention of blight.°° 
CONCLUSIONS. 
Soluble bordeaux, as used in this experiment, has slight value 
as a preventive of potato blight (Phytophthora), but is far infer- 
ior to soda bordeaux and the regular lime bordeaux of the 1-to-8 
formula. Probably it is useless to experiment with dilutions 
greater than that used in this experiment. 
Soda bordeaux proved inferior to lime bordeaux. It is not to be 
recommended, at least until further tests® have been made. Its 
only advantage is that it does not clog the spray nozzles so readily. 
It has the disadvantage that arsenite of soda can not be used with 
it safely except by the addition of lime. 
POTATO TROUBLES IN NEW YORK IN -1904. 
In New York the season of 1904 was cool, with an abundance 
of rain, making conditions generally favorable for potatoes, ex- 
cept in the matter of blight. Owing to the protracted wet weather 
in the spring, planting was much delayed in some localities. 
Many fields were planted from June 5 to 15. A large propor- 
Details of this experiment will be given in a subsequent bulletin. 
The soluble bordeaux has been tested also by Director Woods of the 
Maine Station with similar results (Me. Exp. Sta. Bul. 112:8-12). His 
conclusions are as follows: “The soluble bordeaux of equal strength to 
regular bordeaux mixture cost much more, both in materials and labor, 
than regular bordeaux mixture. The yields were smaller and the quality 
inferior from the plots sprayed with soluble bordeaux. For both of these 
reasons its use is not recommended.” 
It should be stated that only carefully conducted experiments are of any 
value in solving this problem. Such experiments as farmers make are not 
sufficiently accurate. The difference in efficiency between soda bordeaux and 
lime bordeaux is certainly not great—not more than a few bushels per 
acre—and there are several other factors which, unless eliminated from the 
experiment, would make the results unreliable. Only two such factors need 
be mentioned: (1) Inequality of soil. Soil varies so much that the yields 
of adjacent plats treated as nearly alike as possible often vary. by 25 to 
50 bushels per acre. Hence, wide plats are objectionable. The plats should 
be long and narrow (preferably consisting of a single row) and repeated 
several times as in the above experiment (Page 179) and in the Station 
ten-year experiments (Page 91.) (2) Thoroughness of spraying. Great 
care must be taken that the soda bordeaux and lime bordeaux plats are 
sprayed with equal thoroughness. Even a slight variation in the thorough- 
ness of spraying may materially affect the yield. 
In the report of the Irish experiments cited by Prof. Fraser few details 
are given, making it impossible to determine whether the experiments were 
properly conducted. : 
