Mar. 17, 1919 Fusartum-Blight of Potatoes under Irrigation 



289 



other five groups. Originally in numbering rows in the field the even and odd fol- 

 lowed naturally, so that in the tables the parallel adjacent rows are most easily desig- 

 nated by the terms "even" and "odd." In the tables the "even" rows are read 

 vertically — that is, the designating letters are placed across the top, and the aggregate 

 totals for each class across the bottom; the "odd" rows are read horizontally, the let- 

 ters at the left and the aggregate totals at the right. However, as noted before, many 

 of the twins differ individually in their performance, so that in order to express this 

 deviation the total of one class in one row is split up according to the numbers required 

 to express the reciprocal in the other row. A concrete example is given in Table i A. 

 In the even row there are 352 H, 4 HD, 5 DH, i D, 13A, and 12 O plants, a total of 387. 

 In tlie odd row there are 320 H, 5 HD, 7 DH, i D, 23 A, and 31 O plants, a total of 387. 

 In comparison with 352 H plants in the even row, twin for twin in the odd row, 296 

 plants are also H, while 4 are HD, 7 DH, i D, 20 A, and 24 O. Turning about, it is 

 seen that, of the 320 H plants in the odd row, twin for twin in the even row, 296 are 

 H, 3 HD, 3 DH, I D, II A, and 6 O. The same system follows for the other classifi- 

 cations. It is revealed, then, how nearly alike twin seed pieces perform, for where 

 they are alike the numbers appear in either the HH, HDHD, etc., squares down to 

 00. Differences are shown when they appear elsewhere. 



Table I. — Aggregation of seed piece performance of Irish potatoes 



A. — WISCONSIN PEARL 

 [i equals 0.258398 per cent of 387] 



B. — INFECTED WISCONSIN PEARL 



[i equals 3.70370 per cent of 27] 



C. — CERTIFIED WISCONSIN RURAL NEW YORKER 

 [i equals 0.223225 per cent of 444] 



