Five F3 selected families of bread wheat were evaluated at Ras Sudr Research
Station of (ORC), South-Sinai Governorate in 2000/2001 growing season under
salinity conditions of soil and irrigation water ( about 9875 and 10000 ppm,
respectively ). At 90 days after sowing, 10 guarded plants were selected randomly to
measure chemical contents in a composite sample of each F3 family leaves i.e proline
(lin. freslil· samples), KINa ratio, Mg, CI and S04 (in dry samples). At harvest, fifteen
competitive piants frorr» eacJn' pJbi weJ1e; selectee, to rec.ma' observations (ill, plant
tTeight, number. of spikes- / plant; spjke;leliTg)vl', number-of'qrairrs zspike lODCl'-grain
weigMt< and grain yield / plant. Simple correlations, phenotypic & genotypic..c:o.efficients
of vailiabilily, her.itability and path analysis were calcuiated; The obtained results can
be summarized as follbws: Inter-relationships among the studied characters under such saline conditions
varied in magnitude according to the F3 population genotypes. Correlations between grain yield / plant and each of the chemical composition of
leaves as well as yield components were positive and significant in most cases
except for CI content which was negatively correlated to grain yield / plant. For
improving grain yield under such saline conditions, the following traits: heavy
grains, high number of grains/spike, high content of proline in fresh leaves and low
chloride content in dry leaves considered to be a suitable selection criteria. Heritability values varied among traits and between populations studied therefore,
selection for the yield attributes may be effective for improving grain yield which
had low to moderate (GCV) values among F3 selected families. Hence, the direct
selection for improving yielding ability is not effective and path analysis must be
done to detect the joint effects. The total contribution of each chemical component in leaves showed that proline
content was the most powerful determinant of grain yield in S8" 18 F3 family as it
contributed by 20.33%. In general, chemical contents could be arranged over all
studied F3 populations as follows: S04 , proline, Mg, K / Na and CI. The main source of grain yield variation was 1000- grain weight followed by number
of grains / spike and number of spikes / plant in three or more of F3 families. Such
highly contributed traits easily measured and gave a valuable idea for selection to
yield improvement under saline environments.