VARIETIES AND BREEDS 337 



similar robustness and general character of growth being CHAP, 

 placed next to each other VIL 



In previous years the trials had been conducted on check- 

 row principles, the seed being sown in hills 3 feet apart in 

 each direction ; but in this year that system was discarded 

 and the seed planted in continuous rows 3 feet apart. 



" The field was watched throughout the season, and any 

 blanks in the ' plant ' due to borer, faulty germination, or other 

 such accidental cause, carefully noted. Such blanks, of course, 

 affect the yield, and as they are the result of accident, and not 

 a characteristic of particular breeds, allowance must be made 

 for them in comparative experiments. These allowances, and 

 the fact that the most even stand in a field, even if not the 

 best part, is certainly not the worst, together make the cal- 

 culated yield per acre come out at a higher figure than is 

 actually obtained over large areas. This, however, though 

 giving a somewhat inflated appearance to the yield generally, 

 does not affect the relative positions of the different breeds." 

 In harvesting, two stretches, each 22 yards in length, were 

 selected out of the full length of the rows at places where the 

 stand seemed to be the most even for all breeds. The ears 

 from both rows of each breed were harvested from these two 

 selected areas, shelled, the grain weighed, and the result 

 calculated out to weight per acre, after the necessary additions 

 for blanks were made. This makes a total of four chains 

 length harvested for each breed. 



The character of the season should be taken into account 

 in comparing the yields. In 1905-6 two very severe droughts 

 were experienced during the growing period, viz., in January 

 and February, and it is probable that the medium-early 

 breeds were most affected, as these droughts took place at a 

 critical stage in their growth. Late breeds also suffered to 

 some extent, but the early breeds did not appear to have 

 been much affected. 



The year 1906-7 was ideal for securing heavy yields. A 

 good rainfall took place throughout the growing period, and 

 at no stage of its growth did the crop suffer either from 

 " drought " or a supersaturated soil. 



The year 1907-8 was too dry in February, a critical stage 

 in the crop's development, while in 1 908-9, though a sufficient 



22 



