CEREAL IMPROVEMENT HYBRIDIZATION 215 



by American investigators, and a very exhaustive list 

 belonging to each cereal is given by Tschermak (Fru- 

 wirth, 1910, pp. 119-121, 195-197, 248-253, 332-334). 



225. Discrimination in cereal breeding. Much loss 

 of time is avoided if the cereal breeder will form in mind 

 a purpose or ideal, and then work definitely to that end. 

 Just one important quality should be sought in the same 

 cross. Permitting others to be considered will only 

 interfere and delay the work, unless some very striking 

 form appears which is seen at once to promise something 

 of unusual additional value for a particular purpose. If, 

 for example, it is attempted to add, by crossing, rust 

 resistance to a wheat already good in many other respects, 

 that quality only should be kept in view. Manifestly 

 the most difficult thing, requiring critical knowledge and 

 good judgment, is the selection of the parent varieties 

 and then the individual parents. To give ample oppor- 

 tunity for variation, numerous crosses should be made 

 of the same parent varieties. The crosses themselves are 

 only starting, points. The greater part of the work is the 

 discriminating selection from the progeny. It is highly 

 important that all breeding be done in the district where 

 the new strains are to be distributed, in order that the 

 breeder may work with a full knowledge of the conditions 

 and give the varying progeny an opportunity to fit the 

 environment. 



226. Method of cross-pollination. The parent plants 

 having been selected, it will depend chiefly on the cir- 

 cumstances of flowering which shall be chosen as the 

 female parent. If no other conditions need be considered, 

 the one ripening later should be chosen. The essential 

 tools needed in artificial crossing are forceps, preferably 

 self-closing, and small scissors. Supplies of paper bags and 



