The epistatic interaction of alleles at the VRN-H1 and VRN-H2 loci determines vernalization sensitivity in barley. To validate the current molecular model for the two-locus epistasis, we crossed homozygous vernalization-insensitive plants harboring a predicted "winter type" allele at either VRN-H1 (Dicktoo) or VRN-H2 (Oregon Wolfe Barley Dominant), or at both VRN-H (Calicuchima-sib) loci and measured the flowering time of unvernalized F(2) progeny under long-day photoperiod. We assessed whether the spring growth habit of Calicuchima-sib is an exception to the two-locus epistatic model or contains novel "spring" alleles at VRN-H1 (HvBM5A) and/or VRN-H2 (ZCCT-H) by determining allele sequence variants at these loci and their effects relative to growth habit. We found that (a) progeny with predicted "winter type" alleles at both VRN-H1 and VRN-H2 alleles exhibited an extremely delayed flowering (i.e. vernalization-sensitive) phenotype in two out of the three F(2) populations, (b) sequence flanking the vernalization critical region of HvBM5A intron 1 likely influences degree of vernalization sensitivity, (c) a winter habit is retained when ZCCT-Ha has been deleted, and (d) the ZCCT-H genes have higher levels of allelic polymorphism than other winterhardiness regulatory genes. Our results validate the model explaining the epistatic interaction of VRN-H2 and VRN-H1 under long-day conditions, demonstrate recovery of vernalization-sensitive progeny from crosses of vernalization-insensitive genotypes, show that intron length variation in VRN-H1 may account for a continuum of vernalization sensitivity, and provide molecular markers that are accurate predictors of "winter vs spring type" alleles at the VRN-H loci.