New York AGRICULTURAL EXPERIMENT STATION. 281 
fluence production. The results in some cases are inconsistent, 
owing to the lack of uniformity in the soil of the farm and the 
use of the oblong-plat method of comparison; but so often were 
the tests repeated and so great was the care used to recognize and 
to eliminate or neutralize disturbing factors that the conclusions, | 
where these are definitely stated, have been, in the main, confirmed 
by later tests by other Stations. 
In trials of cultural methods and fertilizer tests on Station soil, 
the handicap of uneven productivity rendered the results with po- 
tatoes almost as inconclusive and valueless as those with other 
crops. In a few lines, however, the indications uniformly pointed 
to the same conclusion, showing that the factor under considera- 
tion was one of real importance in potato growing. In fertilizer 
tests on Long Island most striking and uniform results were secured 
during several years’ trials; and a means of economizing in the 
growth of the crop was plainly indicated. Unfortunately this 
method seems out of line with the general practice on the Island, 
and growers hesitate to give it a fair trial; so that the tests have 
not influenced practice as their accuracy and clear-cut teaching 
merit. 
In treatment of blight and rot, the Station work has materially 
benefited potato growers; not only in the State but in all potato- 
growing sections. This factor in potato culture is discussed at 
length on pp. 147 to I5I. 
Character of seed.— Experiments* made in early years of Station 
history, and repeated, with modifications, after a lapse of twenty 
years, indicate a decided advantage from the selection of tubers for 
seed from the productive hills of the parent crop. In 1884 seed 
thus selected outyielded seed selected from unproductive hills in 
seven cases out of nine when large tubers were compared and in 
eight cases out of nine when the comparison was between small 
tubers of the two classes. In nearly half the cases the small tubers 
from productive hills gave better yields than large tubers from 
unproductive ones. In 1885 the tests showed a decided advantage 
for the seed from heavy yielding hills. In 1887 the comparison 
was extended to seed from 116 varieties, but the area devoted to 
each test was small. In two-thirds of the individual comparisons, 
the advantage was with the seed from productive hills, and the 
average gain was fourteen bushels to the acre. In this same year 
Rpts. 3:301-305 (1884); 4:232-235 (1885); 5:148, 149 (1886) ; 6:78-86 
(1887): Syllabi, Normal Institute Lectures, 1905, 1906. 
