SIZE OF POTATO SETS. 13 
Bruce (18, p, 10-11) in summarizing his experiments comparing 
whole and cut seed concluded that when the same weight of seed is 
planted on a given area cut seed will give a larger yield of merchant- 
able stock, hut that it does not give a greater total crop than is 
obtained with whole seed. 
Close and White (26, p. 161) found that larger yields were ob- 
tained from egg-sized tubers of both the Irish Cobbler and Green 
Mountain varieties than from walnut-sized tubers. Marble-sized 
tubers gave the smallest crop. 
In a six-years' study of the comparative yields from large, medium, 
and small sized whole tubers by Zavitz (104, p. 102), the average 
production per acre from the three sizes was 338, 274, and 201 
bushels, respectively. In 1916 the same author (105, p. 43) reported 
further experimental data in which a comparison had been made 
between single-eye sets varying from one-sixteenth of an ounce to 
2 ounces in weight. The figures he gives show a progressive in- 
crease in yield from 46.2 to 132.7 bushels per acre in favor of the 
heavier set. Progressive increases were also obtained from sets of 
the same weight but with eyes ranging from two to five to the set. 
Whole potatoes were also found to be superior to two halves or four 
quarters planted in the same hill. Sheppard and Churchill (87) 
at the Edgeley, Dickinson, and Williston substations studied the 
value for seed purposes of large and small whole tubers and of 2-eye 
sets. They obtained for a 2-year period an average yield of 138.7 
bushels of marketable tubers per acre from the large tubers, 92 
bushels from the small whole tubers, and 101 bushels from the 2-eye 
sets. 
Henshaw (48) demonstrated that medium-sized tubers, 1\ to If 
inches in diameter, were best. With tubers of this size it would 
require from 15 to 18 hundredweight to plant an acre if they were 
spaced 28 by 14 inches apart. In two of the four seasons in which 
the tests were made cut seed gave the best results. In the other two 
seasons whole seed proved superior to cut seed. In experiments 
conducted by Bliss \l7 , p. 23) in three counties in Iowa in 1909, 
large whole tubers proved superior to halves, quarters, 2-eye sets, 
and 1-eye sets. The average yield of merchantable stock from the 
whole tubers was at the rate of 125.7 bushels per acre; halves, 100.1 
bushels; and quarters, 93.2 bushels; while 2-eye and 1-eye sets 
yielded 85.8 bushels. Somewhat similar results are reported by 
Schollander (81, p. 28) in a summary of the results of a 5-year test 
of large and small whole tubers, large halves, and of 3-eye, 2-eye, 
and 1-eye pieces. The 5-year averages show that the yield of mer- 
chantable tubers was greatest from the large 1-eye sets. _ Large whole 
tubers gave the greatest total yield. The order of yields of mer- 
chantable tubers was (1) large 1-eye sets, (2) large whole tubers, 
(3) large 3-eye sets, (4) large 2-eye sets, and (5) small whole tubers. 
In total yiefd the order was (1) large whole tubers, (2) 1-eye sets, 
(3) small whole tubers, (4) 3-eye sets, and (5) 2-eye sets. Hume 
(49) compared sets averaging 1.4 ounces in weight with those weigh- 
ing 0.8 ounce and obtained an increased yield of 27.6 per cent in 
favor of the larger sized set, The total yield per acre from the larger 
sets was at the rate of 180.6 bushels, while that from the small sets 
was 141.5 bushels. 
