GENETIC DIVERGENCE IN COMBINING ABILITY AND MAIZE HYBRIDS

To estimate the genetic parameters for a local maize population (Zea mays L.J.
Halt diallel cross mating design was used. The data showed the best crosses
(M1 OxM 11) and (M11 xM 18) for grain yield /plot: and (M10xM12) tor plant height and
days to tasseling; (M11xM16) for too-kernet weight; (M15xM18) for number of
rows/ear, number of kernels/row and 100·kernel weight; (M12xM16), (M13xM15) and
(M 14xM 18) for plant height (towards shortness) and ear height (towards low ear
placement); and the cross (M16xM18) for days to tasseling towards earliness and
grain yield/plot. Generally, most studied traits showed significant differences for G.C.A
mean squares except, number of rows/ear and plant height at the two years and
combined data. Highly significant specific combining ability (S.C.A) mean squares,
were found for most studied traits. For crosses x years interaction mean squares
show highly significant differences for most studied traits wer,::; found. while for number
ot rows/ear. plant height. and days to tasseling non- significant differences were
observed For G.C.A x years mean squares the dat.l showed divided between the
seven traits under study. For S.C.A x years -nean squares, highly sig,!1ificant
differences for most studied traits were foun.j ";'re ratio between K²G.C.A / K'S.C.A.
was found to be less than unity at the comr-r.ed data for most studied traits except for
days to tasseling indicating that. the uommance genetic variance controlled the
.behavior of most studied traits rathr-: .ian the additive ones. Generally the best inbred
line was M 18 for ear diameter a: the two growing years and the combined data. for
lCO-kernel.weight at the second year and combined data, and ptant height (toward
shortness) at the first year and combined data. The best crosses were (M11xM16) for
100-kernRI·weight (M13xM1S) for ear height (towards low ear placement) at the two
years and combined data. The percentages of economic heterosis of F I hybrids
relative to the check variety (S.c155) were ranged from (1.60% to 33.66%) positive
desirable values and from (·5.78%) to (-6.40%) negative desirable values at combined
data. Heterosis could not be considered as a function of genetic divergence.
Moreover. genetic distance based on morpho-agronomical markers was not
significantly correlated with specific combining ability, heterosis and the mean
performance. Hence, it is impossible to predict the hybrid performance from genetiC
distance itself.