BUD SELECTION. 35 



that progress in makinjc: useful new varieties would have been very 

 much slower. It was practicable to select choice plants from the field 

 of flax, because each seedling jilant stands out distinct on its single 

 main root, while it is not practical to thus select in case of wheat, 

 which sends up numerous culms from each seed, the culms from two 

 or more seeds often being interwoven. 



While the general principles of breeding are the same throughout, 

 their application must vary greatly, according to the habit of the 

 species dealt with. One plan is best with open-fertilized species, 

 another with close-fertilized species. Species freely cross-pollinated 

 by the wind may require different treatment from those occasionally 

 crossed hy insects. Varieties propagated mainly or wholly by cuttings 

 require different plans from varieties propagated by seeds. Annuals 

 often require a different method of selection from that used with per- 

 ennials. In case of varieties used in one locality which are jiroduced 

 from seeds or bulbs grown in another place, the breeding requirements 

 differ from those where the seeds are bred in the same environment 

 under which the crops are to be grown. The whole range of special 

 means and requirements for breeding the thousands of varieties for 

 many localities having different conditions can hardly be more than 

 touched upon. Further on a somewhat detailed plan for breeding- 

 several crops will be given as examples, showing how the systematic 

 improvement of a plant may be worked out. 



BUD SELECTION. 



While efforts should be mainly directed to a choice among seedling 

 Individuals, yet the variation among the buds, branches, spikes, or 

 heads of the ijlants is sufficient to warrant general and in some cases 

 special attention. Wliile we are warranted in i)lanting the seeds of 

 wheat singly in the hill, that we may judge or test the weight and 

 quality of the croj) of seed from each individual plant so as to secure 

 the best i^lanf, we should also secure the seeds from the best spikes. 

 And in case of many flowers multiplied l)y cuttings, where bud varia- 

 tion is sometimes considerable, marked improvements may l)e made 

 by bud selection. Each branch, leaf, and flower springs from a single 

 bud, and, though bud variations do not generally breed true to tj^pe, 

 they so often do that the selection of marked bud variations is well 

 worth attention. Sometimes a limb, a flower, or other new l)ud 

 feature does not breed true at once when reproduced. IJy growing 

 many plants from the peculiar ])ranch the same bud variation may 

 appeal' on one or more plants, and upon following up nature's hint 

 workable ore may be discovered deep down in the mine. Sometimes 

 we find seminal variations, and more frequently bud variations, wliich 

 breed true, but as a rule we secure them wavering, as if staggering 

 under a load of inherited tendencies which are trying to reduce them 

 to the level of their species, variety, or family. As the special scliool 



