In a popcorn breeding program, the procedures and techniques used for breeding popcorn are similar to those used for breeding dent corn with some minor adjustments needed. The adjustments in popcorn breeding are due to the poorer agronomic traits performance of popcorn germplasm compared with dent corn germplasm. The reasons for the poorer agronomic traits of popcorn are: the limited effort focused on improvement of popcorn, agronomic traits of popcorn are not the most important traits emphasized in selection, and the base germplasm available for popcorn breeding program does not perform agronomically as well as the base dent corn germplasm.

The most important quality trait evaluated by a popcorn breeder is popping expansion volume; defined as the volume of popped corn per gram of unpopped corn; therefore, improvement in this trait is considered the most important objective in popcorn breeding. High expansion volume is associated with increased kernel tenderness. Also, from a commercial standpoint, popcorn genotypes with high expansion volumes will produce more popped corn than genotypes with low expansion volumes

Heritability estimates for the same trait in the same population vary depending on the type of heritability computed. However, heritability estimates must be defined relative to the trait of interest, the population, and the environment unit because these factors affect the estimates.

Parent-offspring regression is a measure of narrow-sense heritability and is not inflated by non-additive components of variance, such as dominance. Assumptions must be met to correctly interpret the results from parent-offspring regression; however, no environmental correlation was met for this study. The discrepancy did arise in the heritability estimates for popping expansion per kernel. The parents (F2) were grown and measured in at the AEERC in 2002 and the offspring (F3) were also grown and measured in 2003 and 2004 at AEERC, which gave upwardly biased heritability estimates for AEERC. This can be seen in the comparison of the estimate of popping expansion was 0.47 (0.42, 0.51) at AAERC, 2003 and 0.47 (0.43, 0.50 at Curtiss Farm, 2003, and 0.60 (0.57, 0.64) at AAERC, 2004. The components of variance heritability were higher than those observed from parent-offspring regression because these did not take midparent data into account and therefore were not biased by environmental correlations and remained constant across environments. An explanation is the fact that parent-offspring regression measures narrow-sense heritability while F3 generation estimates are in the broad sense. Another possible explanation is the fact that the heritability obtained from the components of variance are a ratio of the covariance of full-sibs in the F3 generation to the phenotypic variance of full-sib families while parent-offspring regression is a ratio of the parent-offspring covariance to the parent-phenotypic variance. Overall, popping expansion traits were correlated with each other at individual environments and across environments for parent-offspring regression and components of variance.

Since heritability is referred as the effectiveness of selection for a trait depending on the relative importance of genetic and non genetic factors of phenotypic differences among genotypes in a population, and based on the results from parent-offspring regression and components of variance; selection in the F2 generation for popping expansion traits should be practiced with moderate selection to maintain improved agronomic traits, previous results agree in accordance with the results of this research that selection in the F2 generation for popping expansion traits should be successful and small improvement is observed in subsequent generations of inbreeding. Broad-sense heritability of popping expansion will facilitate popcorn breeders to develop long-term genetic improvement combining desirable attributes of popcorn and dent corn in popcorn hybrids.

Because phenotypic selection for popping expansion and other agronomic traits related is a laborious and time consuming, phenotypic correlations among traits in a popcorn population let breeders know how traits are related and allow them to develop breeding methods according to the desirable characters of a good cultivar including high popping expansion traits. The correlation coefficients obtained in this research for all the popping expansion traits suggest that it would be feasible to develop germplasm which combines both high popping expansion of popcorn and high grain yield of dent corn as result of as short-term breeding term effort.