-

Selection index is one of the effective methods in indirect selection for a
higher yield. In this study a field experiment was conducted in 1997-98 at two locations (Kooshkak and Badjgah). Twenty five wheat (Triticum aestivum L.)
cultivars were used in a randomized complete block design in each location. Agronomic and physiological traits were measured in five developmental
stages (tillering (stage 1, 51), Stem elongation (52), ear emergence (53), dough
development (S4) and ripening (55)]. The significant traits were distinguished
and the genotypic correlation coefficients with yield, heritability, genotypic coefficient of variation, phenotypic coefficient of variation and genetic
advance were determined. Pathanalysis was done for 23 traits and 14 following traits wre selected:nurnber of kernels per main spike, number of kernels per
tiller spike, number of spikelets per main spike, number of spikelets per tiller spike, number of spike per square meter, percentage of infertile tillers, spike
length spike to stem dry weight ratio in tillers at (55), harvest index, Plant height, NAR21, NAR32, LAD3 and LAD4. The first method used was the
method of Robinson. The best indices resulted from this method had these

spike, number of spikelet per tiller spike. It could be recommended that since the tirst type of selection index is simple and precise, may be suitable for use
combination of the' following traits was used: seed yield, number of kernels per main spike, number of kernels per tiller spike, number of spike let per main
indices were optimum selection indices. In third method economical value was heritibility of traits and in the foruth, economical value was the product of
heritability in advances of desired traits. In the two last methods the
weight ratio of tillers at (55), harvest index, NAR2 and NAR3.This index was similar to the best of the first type of selection index. Third and fourth types of

traits: seed yield, number of kernels per main spike, number of kernels per tiller spike, spike to stem dry weight ratio in tillers at (55), harvest index,
NAR2 and NAR3. The second method was desired gain selection index. The best indicf;s had these traits: seed yield, number of kernels per main spike,
number of kernels per tiller spike, spike to stem dry weight ratio of tillers at (55), harvest index, NAR2 and NAR3. The second methos was desired gain
selection index. The best indices had these traits: seed yield, number of kernels per main spike, number of kernels per tiller spike, spike to stem dry
in applied experiments.