46 AMERICAN POMOLOGY. 



axioms of breeders of animals, assumes that crosses, as of 

 distinct races, will not be so likely to produce good results, 

 as a system of breeding in-and-in, persistently carried out. 

 This plan he recommends, and alludes to the quince and 

 mulberry as suitable species to operate upon, because in 

 them there are fewer varieties, and therefore less liability 

 to cross-breeding, and a better opportunity for breeding 

 in-and-in. He also reminds us of the happy results which 

 follow the careful selection of the best specimens in gar- 

 den flowers and vegetables, combined with the rejection 

 of all inferior plants, when we desire to improve the char- 

 acter of our garden products, and he adopts the views of 

 certain physiologists, which, however, are questioned by 

 other authorities, to the effect that violent or decided 

 crosses are always followed by depreciation and deteriora- 

 tion of the offspring. 



The whole communication referring to these experiments, 

 which are almost the only ones, so far as I know, which 

 have been conducted in this country to any extent, to 

 verify or controvert the Van Mons' theory, is very inter- 

 esting, but it is easy to perceive that the experimenter, 

 though apparently very fair, and entirely honest, has been 

 fully imbued with the truth and correctness of the propo- 

 sition of Van Mons, that the first ripened seed of a nat- 

 ural plant was more likely to produce an improved va- 

 riety, and that this tendency to improvement would ever 

 increase, and be most prominent in the first ripened seeds 

 of successive generations grown from it. 



The theory of Van Mons I shall not attempt in this 

 place to controvert, but will simply say that nothing 

 which has yet come under my observation has had a ten- 



