Abstract:
Sorghum is a food security crop in tropical marginal areas. Improvement strategies for yield
under such conditions are important. Genetic improvement for yield is done thr0ll:gh improvement
of its components. Due to yield component compensation, this improvement strategy
is made less rewarding. It is therefore important to know compe1J.satory mechanisms existing
for better improvement strategies. Studies on the nature of component compensations in
sorghum are limited and virtually lacking under Tanzanian conditions. A field experiment
was conducted to investigate the nature of interrelationships among components of yield in
sorghum at the experimental plots of Hombolo Research Station in the semi-arid zone of
Dodoma Region, Tanzania. Six treatment combinations, of rain water harvesting techniques
and fertilizer were laid out in a randomized comple,te block design (ReED) with four replications
during the growing season of 1996/97. Number of grains was an important component
which was significantly correlated (r=O. 982***) with grain yield and had a high positive direct
effect (0.979) on yield. Average grain weight was not import~nt in influencing yield of
sorghum. Plant biomass had a negative direct effect (-1.2997) on average grain weight but
was not important in influencing number of grains. Plant height and percent light intercepted
directly influenced number of grains and average grain weight negatively. The negative influence
(- 0.8712) of plant height on number of grains was compensated to a low relationship
(r= -0.337) mainly by its positive indirect influence (0.3780) through light interception. Improvement
strategies should aim at shorter plants with more grains, of lesser canopy development
and biomass in these semi arid areas.
