HYBRIDIZING AND CROSSBREEDING 193 



have studied the problem carefully and followed it up 

 systematically over a long period. 



It has been a source of comment that while the breeding 

 of Carnations has been carried on for many years, there 

 has not as yet appeared a true bred Carnation, one that 

 will reproduce itself from seed. This can perhaps be 

 accounted for in two ways. In the first place, the per- 

 petual flowering Carnation of today is so different from 

 the original from which it sprang, and the change has 

 been so gradual, there remains a strong tendency to revert. 

 Then, too, the breeders have not concentrated their efforts 

 on the production of a true bred strain. The custom has 

 been to infuse the blood of any variety possessing certain 

 qualities in which the strain may have been deficient, 

 consequently the ancestry of most varieties is so complex 

 that no definite results could be relied upon when a cross 

 was made. 



Unexpected colors and other characteristics crop out 

 after several generations, and it is only by close breeding 

 over a long period of years that such variations can be 

 eliminated, as we have demonstrated in our own work. 

 For years we have confined the bulk of our crossing to 

 certain colors, separating the colors into five classes, 

 namely white, flesh pink, deep pink, red (including crim- 

 son), and yellow. We cross a white variety with another 

 white, a red with another red, and so on. As a result, 

 we get each year a larger per cent, of true colored seedhngs. 

 But as stated above, there will occasionally crop out a 

 color that has not appeared in the pedigree for several 

 years. For instance, we have two seedHng varieties which 

 are identical (deep flesh pink) in color. One came from 

 two intense scarlets, the other from two pink varieties. 

 The influence of the parents can be seen, however, in that 



