yield/plant and harvest index and number of pods/plant are
comparable to those previously reported for lentil.
Associations between plant characters are of value in breeding
programs helping in their planning and evaluation. The above results
of F4 generation show that seed yield/plant was correlated positively
and significantly with number pods/plant and number seeds/plant in
all crosses. The results suggest that selection for high number of
pods/plant would be efficient for improving seed yield. In this regard,
Hamdi (1987) compare of the efficiency of indirect selection through
selection for 100 seed weight or for number of pods/plant. The ratio
was 0.55 if indirect selection was applied to 100-seed weight, whereas
it was 0.69 if selection was performed on number of pods/plant. These
results indicate that improving seed yield by selection for pods per
plant should be more efficient than selection for seed weight. Manara
and Manara (1988) also reported that maximum weight should be
given to pod number during selection for high yielding lines in lentil.
It should be mentioned, however, that counting the number of pods on
a plant is a more laborious than measuring its seed yield.
The relationship between the two main yield components, number
of pods/plant and number of seeds/plant, showed strong association,
indicating that selection for one of them is enough to improve yield.
Moreover, in most cases there was no significant correlation between
number of pods/plant and number of seeds/plant with number of
seeds/pod. These results indicate that it would be difficult to achieve a
response to selection for high levels of pods/plant and /or seeds/plant
and number of seeds/pod.
Number of branches/plant (primary and secondary branches) was
correlated negatively or not correlated, in most cases with harvest
index. Tejinder Singh (1977) found that number of primary and
secondary branches had negative direct effects on seed yield. She
observed that a bushy plant type results in relatively more plant dry
matter than seed yield, and hence lower harvest index. She suggested
it would thus be desirable to develop varieties with less branching to
allow better light penetration through the plant canopy for good pod
formation. In the present investigation, the results indicate that it is
possible to select for high seed yield/plant with a reduced number of
2.2. F5 generation
2.2. 1. Nean performance of F5 families
The analyses of variance were made for each F5 cross
separately and the results are given in Table (26). The data reveal
significant differences among F5-families for several characters in
each cross. For example, the F5–families varied significantly in plant
height in cross 4 and cross 5 only, seed yield/plant in cross 3 and 4
and days to flowering in all crosses.
Regarding the data of cross 1, the performances of the families
(12 families) for plant height, number of branches/plant, growth vigor,
days to flowering and days to maturity are given in Tables (27).
Table 26. Mean square values of studied characters of the F5-families of the
five lentil crosses (Giza 9 and Sinai 1 excluded).
Character Cross number
1 2 3 4 5
Plant height 79.36 35.42 32.23 44.61** 49.54**
Branch./plant 1773.4* 752.45** 765.66 885.69** 737.90
No. of pods/plant 16811.9** 4459.1** 3210.5 2446.6* 4997.95*
No. of seeds/plant 9837.1 2165.8* 2412.0 1589.8* 2971*
No. of seeds/pod 0.041* 0.012 0.022** 0.030* 0.030**
Biological y/plant 220.79 78.963 20.57 29.21* 42.18
Seed yield/plant 0.407 0.366 0.988** 0.474** 0.782
Growth vigor 0.566* 0.410 0.533 0.437 0.364
Days to flowering 1716.2** 182.56* 212.72* 213.28** 463.46**
Days to maturity 29.52 32.27* 25.26 33.69** 28.59
Harvest index 623.83 227.57 4.18 275.73* 154.08
**,* Significant at 0.01 and 0.05 levels of probability, respectively.