EARLY IDEAS ON INBREEDING AND CROSSBREEDING 11 



miles apart for a number of years were planted together in alternate rows. 

 All of one stock grown in this field was detasseled and thus it could not be 

 self fertilized but could produce only hybrid seed. The tasseled stalks of the 

 other lot would still be pure bred as there was no foreign pollen to contami- 

 nate their ears and they could again serve as a parent to a hybrid. A small 

 amount of the first parental stock which furnished the detasseled stalks was 

 grown apart for future hybridization. The hybrid seed was planted, and 

 produced the main crop. Beal increased his yield by this method by as 

 much as 151 exceeds 100. This method and these results, it should be 

 emphasized, were published in 1880. 



E. Lewis Sturtevant, the first director of the New York Agricultural 

 Experiment Station, made a number of studies of corn hybrids starting in 

 1882. His findings are interesting and important but not directly applicable 

 to heterosis. Singleton (1935) has called attention to this work and to the 

 excellent genetic research which the western corn breeders were carrying on 

 at this time — such geneticists as W. A. Kellerman, W. T. Swingle, and 

 Willet M. Hays. They anticipated many of Mendel's findings and described 

 dominance, the reappearance of recessives (atavisms), and even Mendelian 

 ratios such as 1 to 1 and 3 to 1. They were all concerned with practical 

 results. Hays (1889), in particular, tried to synthesize superior breeds of 

 corn by hybridizing controlled varieties. 



Sanborn (1890) confirmed Beal's results and reported that his own 

 hybrid corn yielded in the ratio of 131 to 100 for his inbred. He also fol- 

 lowed Real's method of planting his parental stocks in alternate rows and of 

 detasseling one of them. He made an additional observation which we know 

 now is important: 



It is this outcrossed seed which will give the great crops for the next year. It will be 

 noted that I gained twelve bushels i)er acre by using crossed seed. The operation is simple 

 and almost costless and will pay one hundred fold for the cost involved. The cross must be 

 made every year using nen' seed, the product of the outcross of two pure seed. (Italics C. Z.) 



If our farmers had known of this discovery reported in 1890 they might 

 not have tried to use their own hybrid corn as seed. 



Singleton (1941) also called attention to a pre-Mendelian interpretation 

 of hybrid vigor by Johnson (1891) which, in the light of our present knowl- 

 edge, deserves more than passing notice. We can state it in Johnson's own 



words : 



That crossing commonly gives better offspring than in-and-in breeding is due to the 

 fact that in the latter both parents are likely to possess by inheritance the same imperfec- 

 tions which are thus intensified in the progeny, while in cross breeding the parents more 

 usually have different imperfections, which often, more or less, compensate each other in 

 the immediate descendants. 



We come next to a j)ublication of G. W. McClure (1892). This paper is 

 deservedly famous, and its many contributions are incorporated into our 

 modern genetics literature. Here we shall cite only the observations which 

 pertain to heterosis. McClure noted (1) that sterility and deformity often 



