SIZE OF POTATO SETS. 3 
whole potatoes of known average weight in comparison with halved 
tubers of known weight showed an increase in favor of the Avhole 
tubers of more than 37 per cent, while 1-eye sets from the apical and 
basal portions (seed and stem ends) of the tubers showed an increase 
of more than 123 per cent in favor of the basal-eye sets. As might 
be expected, the large whole tubers from which all but one eye had 
been removed outyielded the large 1-eye sets from the basal parts of 
large tubers. The increase in this case was nearly 49 per cent. One 
of Anderson's conclusions (7, p. 14) as a result of this study was 
" that the weight of the crop is always in some measure influenced by 
the weight of the seeds planted." In a later portion of his paper 
Anderson says (7, p. 27) : 
Although it appears from the foregoing experiments that the weight of the 
crop was always greatest where the seeds planted were the most weighty, yet it 
would be too rash in us from thence to inferthat as great a crop could in no case 
be obtained from the same extent of ground if it were planted with small potatoes 
or small cuttings as if it were planted with large ones. For it is possible that the 
more bulky seed would require a greater space to nourish them properly than 
small ones; so it is not impossible that if these small seeds were planted at a 
proportionally smaller distance the crop might equal that obtained from the 
larger ones. 
The foregoing deductions by Anderson express a truism that has 
been apparently overlooked by many later investigators who have 
studied this subject. 
Wimpey carefully selected 48 bushels of sets from a large quantity of 
seed. The tubers selected ranged in size from that of a large pullet's 
egg to that of a goose egg. Fourteen bushels were planted whole 
on 60 poles of land, or 0. 375 acre; the remainder, 34 bushels, were 
cut and planted on 265 poles of land, or nearly 1.64 acres. The 
whole seed yielded 66 bushels and the cut seed 3i2 bushels. On the 
basis of area planted the whole seed should have yielded 70.64 bushels. 
Disregarding the larger quantity of whole seed used for a given area, 
the percentage of increase of cut seed over whole seed was only 7 per 
cent. If, however, the comparison is made on the basis of total yield 
minus the seed used, the cut seed outyielded the whole seed by more 
than 21 per cent. There is one feature of Wimpey 's experiment which 
may have had something to do with the poorer showing made by the 
whole seed. He states in his paper that he cut the larger tubers, 
indicating that the smaller sized ones had been used in the whole-seed 
plat, This lack of uniformity in the size of the tubers in both cases 
probably exerted an influence on the resultant yield. Wimpey seems 
not to have thought it of sufficient importance to be worth considering. 
Cochran studied the relative merits of large, medium, and small 
sized sets. His results support Anderson's deductions that the larger 
the set used, within reasonable limitations, of course, the larger the 
yield. The large sets yielded 280 pounds; the medium, 249.5 pounds; 
and the small, 168 pounds. As the weight of seed planted is not 
stated, the actual net yields can not be computed. 
King says that " experience proves that the best seed (cuttings or 
sets) is to "be had from the largest potatoes of every kind. A good 
set, part of a large potato, all other things equal, wall naturally produce 
a stronger and better plant than part of a small potato." 
Barrell obtained a very much greater yield from large than from 
small seed. 
