1 \)TATO I N VESl’IGATION S 
33 
not be checked up by the stand since no counts were made of the 
number of plants in each roV. The separation into marketable and 
cidls was made with a sorter using a 1%-inch screen for the Russet 
Burbank and a 2-inch screen for the Peachblow. The sorting was 
done February 7 and 8. The averages of the yields from the four 
sources of seed are shown in the 13th and 17th columns of Table 12 
’ and the 10th, 13th and 16th columps of Table 13. 
While the results are somewhat conflicting, considered as a 
whole, the general tendency is toward an increase in yield of market- 
able tubers corresponding with the increase in size of the seed piece 
planted. After deducting the amount of seed used the net yields 
from the smaller sizes are slightly larger in the case of the whole 
■: seed though the differences are small, while with the cut seed there 
! is in general an increase with the larger sizes. Comparing whole 
and cut seed' where seed pieces of the same weight were used we 
find a significant increase in favor of the whole seed. The two com- 
1 parisons of seed cut in halves and in quarters show a smaller yield 
from the seed cut in halves in the case of the two-ounce seed pieces 
I and a larger yield where the 1%-ounce seed pieces were used. The 
differences, however, are so small as to have little significance. 
GENERAL SUMMARY OF SIZES OF SEED 
C Considering the results of all the size of seed experiments as a 
whole, we find the superiority of whole seed clearly shown. This 
( form of seed has consistently produced a better stand, a larger total 
' yield and generally a larger yield of marketable tubers than cut 
' seed. Whole seed was also found to produce a greater number of 
tubers per hill which, however, were in most cases of smaller aver- 
' age size than those grown from cut seed. This feature may be 
utilized to advantage in controlling to a certain extent the character 
of the crop. For example, the Rural New Yorker normally sets a 
j small number of tubers, some of which usually develop into over- 
sized stock. The use of whole seed, which would tend to increase 
the number and decrease the size of the tubers, would undoubtedly 
result in a crop of greater uniformity and better quality than that 
grown from cut seed. On the other hand, the Charles Downing, a 
f variety in which the percentage, of small tubers is so large as to be 
considered an undesirable characteristic, would, it is believed, pro- 
duce more profitable crops if cut seed were used, provided as good 
a stand were secured from this kind of seed as from Avhole seed. 
A study of the influence of the size of seed piece on the yield 
