The performance of elite experimental maize (Zea mays L.) hybrids test
ed in 154 regional trials conducted in the Argentine corn belt during
the 1979-1991 period was considered to estimate the genetic gain for g
rain yield in the region. The relation between the relative performanc
e of elite experimental hybrids nested in years in comparison with a c
ommon check and the number of years since 1979 was assessed using regr
ession models. Models containing only a lineal coefficient suggest tha
t the rate of genetic gain was approximately 1.05 q ha(-1) year(-1). S
imilar rates were obtained when fitting models to the same series of e
nvironments (trials) classified as favorable or unfavorable when the y
ield of the check hybrid was greater or smaller than its general mean
across location within years, respectively. When compared to first deg
ree models, second degree polynomial regressions exhibited greater coe
fficients of determination in most instances. Although the full expres
sion of the genetic improvement is restricted by the environment, the
results suggest that maize breeding programs succeeded in developing e
lite hybrids that express higher yields in better environments as well
as in suboptimal environments.