MAIZE. 101 



often over-emphasised to the detriment of inherent produc- 

 tive qualities. DeaUng with this, Hays and Garber comment 

 as follov^^s : — " Maize shows have accompHshed much in teach- 

 ing growers the characteristics of various standard varieties. 

 They have, however, over-emphasised the value of ear type 

 as a means of maize improvement. Much work has been 

 carried on with the view of determining the relation between 

 various ear and plant characters and ability to give high 

 yields. In general, no single character has been found to be 

 so closely related with yielding ability as to be of much value 

 from the standpoint of selection. This seems reasonable when 

 we realise that yield is the final result of many growth 

 characters. Too close uniformity of type probably tends to 

 reduce yield, for we have learned that self-fertilisation in 

 maize causes a marked decrease in growth vigour as compared 

 w^ith cross-fertilisation." Extensive experiments conducted in 

 Ohio, in which the yields from long and short ears, cylindrical 

 and tapering ears, ears having bare and filled tips, etc., were 

 compared, showed no appreciable gain in favour of any of 

 these characters. 



The specialised maize-breeder, working chiefly at experi- 

 mental stations, pursues more technical methods to improve 

 maize, and usually resorts to the ear-to-row method of selection 

 or combines the best qualities of various, strains through 

 hybridisation. 



The Eae-to-"Row Method. — The first step to take is to 

 label the most desirable parent plants in the field to be 

 selected from. Tt is wise to select at least 100 plants; care 

 being taken to select rohust plants growing in full competition 

 with other plants and having no environmental advantage. 

 When mature the desirable ears are selected, of which pro- 

 bably 75 per cent, will be discarded. Each ear is labelled and 

 described. Half of the grain from each is now planted in 

 separate rows, the remnant being carefully stored until the 

 following season. A test of this nature must be made in 

 duplicate, haviui^ every fifth row as a check. Beffinning with 

 a check, every fifth row throughout the test will he planted as 

 a check. Each series will thus contain 32 rows and the whole 

 test 64 rows. Eor the checks use good seed taken from the 

 shelled jrrain of the general crop. Rows twenty-two yards 

 loner with the plants every eisrhteen inches will be found a 

 useful lencrth. Care must be taken to have the same number 

 of plants in each row. 



