EXPERIMENT STATION RECORD. 



Vol. XIV. November, 1902. No. 3. 



The second International Conference on Plant Breeding and Hybridi- 

 zation, which was recently held in New York City, exceeded in attend- 

 ance and interest the previous meeting, held in London in 1899, and 

 served to show the large interest in the subject in this country. An 

 extensive programme of papers was presented, but only a portion of 

 them were read. These discussed the principles of plant breeding 

 and their application, as well as giving the results of years of work 

 and observation in the production and propagation of improved varie- 

 ties of plants. 



In the theoretical discussion of the papers there was an almost uni- 

 A^ersal acceptance of MendeFs law regarding the appearance of domi- 

 nant and recessive characters in the later generation of hybrids. This 

 law, although announced in 1865, has only recently been given wide 

 publicity through its publication in various journals (E. S. R., 13, 

 p. 744). It was the consensus of opinion that it is the best available 

 working hypothesis for the plant and animal breeder, and that it seems 

 to stand the test of expei'ience to a remarkable degree. The methods 

 of Mendel were commented upon, and in a number of papers his 

 conclusions were reaffirmed relative to the necessity of large num- 

 bers of individuals in breeding experiments and the continuation of 

 the investigations through many generations, in order that the results 

 may be of permanent value. The futility of indiscriminate crossings 

 and the necessity of working with pure strains or races was shown by 

 abundant examples. A hybrid produced from a mixed ancestry is 

 very liable to be inferior to either or both of its parents, unless by a 

 long sj^stem of cultivation the characteristics of the parents have 

 become definitely fixed. 



In all kinds of breeding experiments it is necessary to adopt an 

 ideal and adhere closely to it, rejecting for the time all secondary vari- 

 ations that may appear. If these seem very promising, thej^ can be 

 cultivated independently of the main investigation, but nothing should 

 divert the breeder from continuing to follow to its conclusion the line 

 of experimentation, which should be well formulated at the start. 



205 



