In population improvement for combining ability, the use of selfed progenie
s increases genetic advance per cycle, but it can unduly increase the cycle
length. Haplodiploidization (HD) can be very efficient because it induces
complete homozygosity in a short period. We compared, from a theoretical ap
proach, the potential of recurrent selection with a tester using doubled ha
ploids (SDHT), with selection with testcrosses of S-0, S-1, or S-2 plants,
with or without the use of off-season nurseries, for an annual plant like m
aize (Zea mays L.). With the same selection intensity, and without off-seas
on nurseries, SDHT with a 4-yr cycle is the most efficient method. Efficien
cy increases with lower heritability, with an advantage in comparison to th
e test of So plants (S0T) of 40 to 50% at low heritability (h(2) < 0.15) an
d 12% at high heritability (h(2) = 0.8). Use of off-season nurseries reduce
s the advantage of SDHT. With a 3-yr cycle, SDHT remains the best at low he
ritability with a gain of 27% (for h(2) = 0.1) in comparison to SDHT with a
4-yr cycle. When using constant effective size, the advantage of SDHT is f
urther reduced or suppressed at the benefit of S0T in 2 yr, The use of HD i
n recurrent selection for combining ability has its biggest advantage when
heritability is low. Consideration of variety development will give more ad
vantage to HD.