A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1219 



yields for each increase in size of seed pieces used. But here he also failed 

 to consider the factor of rate of planting. In another five-years experi- 

 ment, however, testing the efficiency of various spacings of seed, he has, 

 apparently unconsciously, furnished some much-needed information. This 

 test showed that with the same weight of seed planted per acre, the one- 

 ounce sets, planted twice as close as the two-ounce sets, gave greater 

 total, marketable, and net yields. 



Hume, Champlin, and Oakland (1914) compared large, medium, and 

 small seed pieces, eye frequency being constant, and observed an average 

 increase of total yield, in two varieties,. of 70.9 per cent in large seed pieces 

 and 55.5 per cent in medium seed pieces, over that from the small seed 

 pieces. Emerson (1907) conducted a very comprehensive and accurate 

 test of the relation of size of seed piece to yield under a constant rate of 

 planting. Planting eighth, quarter, and half tubers, 6, 12, and 24 inches 

 apart, respectively, he used 18 bushels of seed per acre in each plot. This 

 gave him the highest total yield per acre from the quarter-tuber pieces 

 and the lowest total yield from the half tubers. This, together with the 

 test by Zavitz (1916) previously cited, would indicate that with the same 

 rate of planting per acre, smaller pieces, down to one ounce, planted closer, 

 are likely to give larger yields than larger pieces planted farther apart. 



Schweitzer (1896), with twelve varieties in a one-year test, compared 

 the total yields and the yields of small tubers from one-eye, two-eye, 

 quarter-tuber, half- tuber, and whole-tuber pieces. This gave a variation 

 in rate of planting of from 4 to 87 bushels. His total yield increased with 

 each increase in size of seed piece and in rate of planting. Also, the per- 

 centage of small potatoes increased from 8.9 for one-eye pieces to 26.4 

 for whole tubers as seed. 



Harwood and Holden (1893) brought together a compilation of experi- 

 ments from thirteen stations designed to test the relation of size of seed 

 and rate, of planting, to yield. In summarizing the comparative value of 

 whole and of half tubers as seed, they showed that the results of a sub- 

 stantial majoritjr of these experiments were in favor of the whole tuber, 

 not only for total and marketable yield, but also for net marketable 

 yield and net value of the crop. Similarly, a majority of the compari- 

 sons of the half-tuber and the two-eye piece favored the former thruout. 

 In drawing such conclusions, it must be borne in mind that these dif- 

 ferences in yield were due to an increase in the amount of seed used 

 because of the larger size of the seed piece. These authors called at- 

 tention to the conclusions of the Ohio station, that " despite the fact 

 that whole potatoes give more small potatoes than one and two eye cut- 

 tings, it is also true that they give more large potatoes." 



The foregoing review of the question of large as compared with small 

 seed shows that few tests have actually proved any superior merit of large 

 seed, except as the amount used per acre was increased. The few tests of 



