1 ’OTATO I N VKSTIG A TION S 
29 
to accentuate the value of the larger seed to a greater degree than 
is shown in the experiments where the planting was done with a 
machine. In each of the two years there were a few hills which were 
badly diseased as the result of attacks of Rhizoctonia and Fusarium. 
These hills, which Avere abnormally small, were dug separately and 
discarded. The yields per acre were calculated on the basis of the 
total area planted. 
Since the amount of seed required to plant a given area varies 
in proportion to the size of the seed piece planted it follows that 
,the larger the seed used the greater will be the cost per acre, which 
must be taken into account in determining the economy of the use 
of the different sizes of seed. To offset this increased seed require- 
ment for the larger sizes the amount of seed used for each size was 
: subtracted from the calculated yield, giving Avhat is termed the 
net yield. The results are shown in the 3rd, 5th, 8th, 11th and 
15th columns of Table 12. Within each of the three groups the 
number of stems, the number of tubers and the weight of tubers 
per hill increased as the size of the seed piece increased. Consider- 
ing only the marketable tubers aa^ find the same general trend in 
the case of the number and weight and the net yield per acre though 
a few exceptions are noted. The most significant point in connec- 
tion with the net yields is the well defined decrease where seed 
pieces weighing less than two ounces were used. Above that size 
the differences are either very small or inconsistent. The per cent 
of marketable tubers, as regards both number and weight, varies 
inversely as the size of the seed piece, the more uniform gradation 
being found in the yields obtained from the whole seed. 
In Table 13, Avhich is summarized from Table 12, are shown 
comparisons where seed pieces of the same weight are taken from 
tubers of different weights. In each of the first three groups, where 
Avhole seed is compared with cut seed pieces of equal weight, the 
Avhole seed produced a greater number and weight of tubers, includ- 
ing both the total and marketable, and a larger net yield after 
deducting the weight of the seed used. A comparison of the results 
obtained from the halved and quartered seed, which appears in the 
first and last groups, indicates the superiority of the seed cut in 
halves over that Avhich is cut in quarters. 
EXPERIMENTS OF 1918 
This experiment Avas conducted under practical field conditions 
with the planting and digging done by machinery instead of by 
hand as in former years. The same variety, the Rural New Yorker, 
