SELECTIOi^ WITHOUT ARTIFICIAL CROSSING. 



41 



Table C. — Method of determining relative frequency of the various types in the progeny 

 of generation F.,, fith cross-fertilization, and selection to the dominant type, when two 

 pairs of characters are involved. 



Matings. 



Relative Ireqaency of each type in the progeny of the various matings. 



AABB 



AABb 



AAbb 



AaBB 



AaBb 



Aabb 



aaBB 



aaBb 



aabb' 



Fi, AaBbX AaBb .... 1 X 1= 1 .... 10a 



F2, AABBXAABB...AXI= 1... 

 AABBXAABb 1X2= 2... 



1 



2 



1 



2 



4 



2 



1 



2 



1 



1 

 1 

 1 

 1 

 1 

 1 

 1 

 1 

 1 



1 

 1 

 1 

 1 

 1 



















1 

















AABBxAaBB.. 1X2= 2... 





1 

 1 













AABBxAaBb 1X4= 4... 



1 

 1 

 2 

 1 

 2 





1 











A ABbxA ABB.... 2X'i= 2... 

 AABbXAABh 2X2= 4... 













1 















AABbXAaBB 2X2= 4... 



1 

 1 

 1 

 1 



2 

 2 

 1 

 1 

 2 

 2 



1 

 2 











AABbX iaBb 2X4= 8 



1 



1 









AaBBxAABB....2Xl= 2... 

 AaBBXAABb 2X2= 4... 









1 





1 













AaBBxAaBB 2X2= 4... 







1 

 1 







AaBBxAaBb . 2X4= 8... 



1 

 1 

 2 

 1 

 2 





2 

 1 

 2 

 2 

 4 





1 





AaBbXAABB....AXl= 4... 

 AaBbxAABb 4X2= 8... 









1 



1 







J.a£5X^a5jB . 4X2= 8... 



1 

 1 



1 

 1 



1 



2 





AaBbXAaBb.... 4X4=16... 

 Frequency of types in F3 



1 



2 



1 



10 



16 



4 



16 



16 



4 



4 



4 



1 



Relative frequency of se- 

 lected tjrpes in F3 



1 



1 







1 























a 16 progeny assumed to avoid fractions in the table. 



Since we do not know what percentage of cross-fertilization occurs 

 in corn or any other open-fertihzed crop, it has been assumed in what 

 follows that corn is completely cross-fertilized. The actual results in 

 practice would be intermediate between figures 2 and 4 when there is no 

 selection to type and between figures 3 and 5 when there is such selection. 



Referring again to figure 2, and supposing that we desire to select 

 from this mixed population and perpetuate the type YYSS — that is, 

 pure yellow starch-forming corn — we would at once discard the fol- 

 lowing forms: YYss, Yyss, ijySS, yySs, and yijss. All of these would 

 either be white or of the sweet type, or both. Discarding all these 

 types that are not yellow starch-forming types, we would still have 

 left the second, fourth, and fifth types shown in figure 2, all of which 

 would be yeUow and would have starchy grains, because the presence 

 of these two characters is dominant over their absence, and these types 

 heterozygote for one or both of these characters can not be distinguished 

 by inspection from type 1, which is the type we wish to perpetuate. 



Selecting for seed those plants which do have j^ellow, starchy grains — 

 that is, types 1,2,4, and 5 of figure 2 — and planting them where they 

 can cross with each other but not with other corn,^ the next year the pro- 

 portion of each of the nine t}^es that would appear would be as shown 

 in Table YI, column 3. If we make the same kind of selection again 

 the next year the proportions of the nine types would be as in column 

 4. Table YI shows the results that would be obtained at the end of the 

 sixth generation by tins kind of selection in a cross-fertilized crop. 



« Complete cross-fertilization is here assumed. The actual results are intermediate 

 between those shown in figures 2 and 4. 

 165 



