220 NORTHEAST EXPERIMENT FARM. 
series of check plots, whose yields may be plotted on paper, 
and the ‘‘value’’ of each intermediate plot thus obtained by 
means of a curved line, it is easy to figure the yield of 
each variety in per cent of the check row. If the aver- 
age yield of all the check plots is adapted as the stand- 
ard yield, and the yield of each variety corrected by mul- 
tiplying this figure by the percentage for the variety, the 
results must be as nearly accurate ascau be obtained. The 
tests for the first five years were not conducted on this prin- 
ciple. During this time some 30 kinds were tried. In 1901, 
many new varieties were obtained and the best of old kinds 
saved. The comparison both of these and the new kinds is 
thought to be so much more accurate for the two years, that 
figures for previous yields are omitted. Table CX gives the 
average yield for two years of 48 varieties, showing the 
total yield and the yield of marketable tubers, on which the 
rank is based. The main crop potatoes have been planted 
between the 23rd of May and 1st of June depending on the 
season. They are dug between Sept. 25th and Oct. 10th. 
The very late kinds do not ripen before frost, and where 
other things are equal preference should be given to the 
second early or medium late varieties. 
Where there is so much new land to be brought under 
cultivation, potatoes will usually be planted on such soil, 
and with good results. There is a certain amount of work 
to be done in breaking and harrowing, and the planting and 
cultivation of potatoes helps to work down the surface. On 
such land there will be very little if any scab, even from 
scabby seed. But no crop, not even potatoes, will give best 
ylelds the first vear, nor until the soil is well rotted and 
worked up. This was well illustrated in 1902. A field 
which had had the timber cut from it in ’98, and had since 
been closely pastured by sheep, was broken late in the fall 
of 1901. In the spring it was cross plowed and planted to 
potatoes, yielding 175 bushels per acre. A strip along one 
end of this field had been plowed early in the spring of 1901, 
sown to oats and pastured off by sheep during the summer. 
This was plowed with the rest in spring of 1902, and the 
yield on the piece was at the rate of 342 bushels per acre or 
nearly twice as much. A part of this difference may be 
