dc.description.abstract | Eight early maturing inbred lines of maize with prolific
potential were crossed in a diallel system (excluding
reciprocals) to determine the inheritance and relationship
among yield and yield components. The inbred lines also were
crossed to four testers in a top cross system. The parents,F1
hybrids,top cross progenies and testers were evaluated during
the long rain season at the University of Nairobi, Kabete Field
Station and at National Horticultural Research Centre, Thika,
in 1991. F heterosis was exhibited by some hybrids for all 1
traits studied. Yield heterosis over the better parent averaged
38.4% at Kabete and 41.5% at Thika. High yielding parents did
not necessarily produce high yielding hybrids. However, hybrids
yielding better than the high parent had at least one of the
yield components higher than the high parent values. Kernel
weight appeared to contribute significantly to yield heterosis.
Three hybrids; DC22xDC26, DC15xDC22 and DC3xDC5 gave
outstanding yields and consistently ranked first, second and
third, respectively, at both locations. General and specific
combining ability determined by Griffing's method 2, model I
analysis were both important for days to flowering, plant
height, duration to maturity, kernel row number and kernel
weight. Specific combining ability alone was important for ear
number, ear length and grain yield. Parent DC5, DC15 and DC37
transmitted earliness; DC22 increased kernel row number;
(x)
DC15, DC22 and DC37 increased ear length; DC5 and DC22
.increased kernel weight; DC15, DC22 and DC26 higher yields.
Grain yield was positively correlated with ear number, kernel
row number, 100 kernel weight, ear length and days to
flowering. Correlations among yield components were either
negative or nonsignificant, except for plant height with ear
length and days to flowering, and for 100 kernel weight with
ear length at Kabete. These associations ruled out the
possibility of indirect selection by use of yield components
for improvement of yield. Direct selection for yield is
therefore suggested. Top crosses did not perform better than
Fl hybrids for important traits such as, ear number, 100 kernel
weight and grain yield. However, they underscored the
importance of genetic diversity in the expression of heterosis.
They particularly excelled in heterotic values for earliness
and reduced plant height. Top crosses; 844BxDC26, 573AxDC3,
Pop.42xDC3 and Pop.42xDC37 gave significant positive combining
ability effects for grain yield. The four testers ranked inbred
lines differently at both locations. Single cross and top cross
tests identified good/poor general combiners differently and
also ranked inbred lines differently at the two locations | en |