Proceedings of Sevexteentii 'Nohmat, Institute 249 



should be closer in the row than when these factors are absent, in 

 order to secure the greater percentage of tubers of medium weight. 

 It has been my experience with growers that they desire large 

 tubers which usually yield heavily, are easily harvested, and have 

 brought regular market prices when delivered at the shipping 

 point. If the above considerations are correct it would seem that 

 the practical way to meet the standards for table stock is to use 

 a mechanical grader for size, and secure the type and quality by 

 sorting and selecting the seed stock used to produce the crop. 



The buyer or grower of seed stock should have in mind qualities 

 other than appearance. Every grower should desire first of 

 all a large yield per acre, because this usually means a small cost 

 per bushel for production, and his profit is the net margin be- 

 tween his cost and sale price, times the quantity. He should have 

 the same sum left for profit if he cleared 10 cents per bushel on 

 a 400-bushel per acre crop that he would have if he cleared 20 

 cents per bushel on a 200-bushel crop, but in practice he usually 

 nets more per bushel on the 400-bushel crop and makes four times 

 the profit on the 400-bushel crop over the 200-bushel one. There- 

 fore it would seem that the first and most important factor to 

 consider is hereditary characteristics for a large production. To 

 obtain a large production because of climate, soil, fertilizer or a 

 favorable season, does not add to the value of the tuber for seed 

 purposes. Any one who has done selection work and isolated pure 

 lines for a number of years, has learned that different lines will 

 vary more than 100 per cent in yield, owing to seasonal and 

 fertility conditions ; but that, side by side, one line will more than 

 double the yield of another year after year no matter what the 

 conditions. This must be due to the inherent characters possessed. 

 During the past year I have helped harvest lines in their second 

 and third year of development, which showed within an extreme 

 range of four tubers diff'erence in counting 100 hills; also that 

 the shape was as uniform as the set. 



To give methods or results of our selection work would make 

 this article too long, but it is safe to say that any variety or group 

 contains enough variations so widely different that a grower can 

 secure any desirable character by isolating a hill showing these, 

 until enough seed is secured to plant his entire field. Many hills 



