HETEROSIS IN A NEW POPULATION 435 



concentrated further by another round of selfs and outcrosses to the tester. 

 The objective would be to concentrate the factors which complement those 

 of the tester to bring about the ideal hybrid. If successful, the final hybrid 

 could consist of a cross between two pools of germplasm, rather than indi- 

 vidual inbred lines, with the tester used as the female parent if so desired. 



The race Conico (Figs. 27.1 and 27.2) is almost universally distributed in 

 Zone 1 where it is used for June plantings at the beginning of the rainy 

 season. With respect to general plant characters it is probably the poorest 

 corn in Mexico. The plants are sparsely leafed, have a poor tassel, and have 

 an extremely weak root system. As a normal practice, dirt is hilled around 

 each individual stalk to keep the plants upright until harvest. Its one out- 

 standing feature is its ability to grow and develop grain of high test weight 

 under relatively low temperature conditions. 



To improve this race, incorporation of germplasm from another race with 

 a strong root system seemed highly desirable. The materials most likely to 

 be of use were the earliest varieties of the race Conico Norteno. For the pur- 

 pose of improving the corn varieties for the rainy season of Zone 1, therefore, 

 two varieties were selected. One, designated as Mex. 39, was one of the best 

 yielding varieties of the race Conico at 2200 meters elevation. The other, 

 designated as Leon I, was a good yielding early variety of the race Conico 

 Norteno commonly grown in Zone 2 at 1800 to 2000 meters elevation. This 

 second variety was a little late in maturity in Zone 1 and was not as well 

 adapted to the cool growing season as was Conico. Selected first generation 

 selfed lines from Mex. 39 were topcrossed on Leon I and selected Si lines 

 from Leon I were tested with Mex. 39. The best combiners with the respec- 

 tive testers under conditions best suited to Conico were then crossed in all 

 possible combinations, and tested for yield. From these results a double 

 cross that might also make a good synthetic was predicted, made, and tested. 

 The pedigree of this double cross, its yield relative to Mex. 39, and its prob- 

 able yield in F2 are given in Table 27.6. 



The female parent of this double cross consisted of a single cross between 

 two Si lines from Leon I (Race Conico Norteno). The male parent consisted 

 of a cross between an Si line and a composite of four Si lines all from Mex. 39 

 (Conico). This pollinator also is being propagated and used as a synthetic 

 variety with good results. 



Although based on only one year's results, the data in Table 27.6 indicate 

 substantial differences in yield capacity between the hybrid and the open- 

 pollinated variety Mex. 39. Perhaps what is more important at the moment 

 is that the advanced generation progeny of this hybrid shows promise of 

 being substantially superior in yield capacity to Mex. 39 which is one of the 

 better varieties of the race Conico. The actual yields of the double cross and 

 the six possible single crosses between the four parental lines, together with 

 the open-pollinated variety Mex. 39 as check in Table 27.6, were all deter- 



