Reexamining the relationship between fall dormancy and winter hardiness in alfalfa

Although phenotypic correlations generally show winter hardiness and late summer and autumn growth (fall dormancy, FD) in alfalfa (Medicago sativa L.) to be strongly associated (r = 0.90), the genetic relationship between the traits has been poorly documented. This study was conducted to rigorously characterize this relationship in a segregating population. We developed an F-1 population of 200 plants derived from crossing an elite subsp. sativa clone, ABI408, and a semi-improved subsp. falcata clone, WLSFAL-6. We clonally propagated the parents and progeny and planted replicated trials at Ames and Nashua, IA in 1998. Plant height, a measure of fall dormancy, was measured in October 1998 at both locations. Winter injury was rated in April 1999. Transgressive segregation in both directions was noted in all cases, except for plant height at Ames, where no F-1 genotype was shorter than WISFAL-6. The heritability of fall growth was estimated at 0.69 +/- 0.044 on an entry mean basis and 0.29 +/- 0.035 on a plot basis, and of winter injury at 0.73 +/- 0.039 and 0.39 +/- 0.040, respectively. Phenotypic correlations between the traits based on entry means were not evident (P < 0.05) but were weakly negative based on plot data (-0.11 at both Ames and Nashua, P < 0.01). The genetic correlation between fall height and winter injury was -0.16 +/- 0.048. A weak association exists in this population-taller plants in the fall have less winter injury. These results provide compelling evidence that considerable improvement in both fall growth and winter hardiness can be achieved simultaneously. We suggest that the varied pathways controlling winter hardiness provide ample opportunity for selection of agronomically desirable genotypes.