THE SEED 27 



In other words, if one waits until toward the close 

 of the season before selecting fruits for seed, the 

 tendency will be for the succeeding crop to be later 

 in ripening, and if this is followed up for a few 

 years, a naturally early variety will be transformed 

 into a late variety. At first thought, one might 

 naturally take it for granted that such would be the 

 case, and yet I have been unable to find any well- 

 authenticated data, coming from actual experi- 

 ments, on the subject. In reply to an inquiry, Dr. 

 B. D. Halstead, of Rutgers College, New Brunswick, 

 New Jersey, says : " I do not know where to turn 

 for the information you desire. The whole subject 

 is well worthy of prolonged study, as only a long 

 series of generations can give the required data for 

 the deducing of a rule." 



The same list of questions was addressed to Prof. 

 P. K. Blinn of Rocky Ford, Colorado, and here is 

 his reply : " In regard to developing early maturity 

 in cantaloupes, I have found in my work, that the 

 individual .plant is the unit of variation, and should 

 be the unit of selection ; that is, if seed of a number 

 of individual plants from even a relatively pure 

 variety be saved separate and planted in a compara- 

 tive test under uniform conditions, there are apt 

 to be marked contrasts in time of development as 

 well as many other variations which offer the op- 

 portunity for selection of any of the desired traits. 

 The more careful and systematic the breeding or 

 seed selection has been, the less the variations will 

 be and the more uniform the product. As a general 

 proposition, the early selected seed will tend to 

 produce the earlier crop, for the reason that the 

 early selection will obviously include more of the 



