Introduction and Objective: Maize (Zea mays L.) is the third most important cereal after wheat and rice, which accounts for 26% of total cereal cultivated area and 37% of their production. Maize is a valuable raw material for extracting oil, starch, alcohol, glucose, plastic, lactic acid, acetic acid, acetone and paint, and it is possible to made paper, cardboard, and nitrocellulose from its stalks. As an important nutrient, phosphorus plays important role in energy transfer, photosynthetic oxidation-reduction reactions, as well as in biochemical compounds including nucleic acids, structural proteins, enzymes, and signal transmission. Due to the predominance of calcareous soils with high pH in arid and semi-arid agricultural climates, the amount of available phosphorus is limited. In order to increase available phosphorus for plants, large amounts of phosphorus chemical fertilizers are needed regularly. However, a large amount of phosphorus in fertilizers may be converted to insoluble phosphate, immediately by reaction with calcium in the soil after its application. On the other hand, indiscriminate use of phosphorus chemical fertilizers leads to many environmental problems such as surface runoff of phosphorus, eutrophication of aquatic ecosystems, reduction of biodiversity and abnormal changes in soil salt concentration and pH. Breeding cultivars that absorb phosphate or phosphorus more efficiently is one of the solutions to deal with the stress of phosphorus deficiency as a trait with low heredity. The use of selection indices can be an effective method for the indirect selection of traits with complex genetics.  

  Materials and Methods: 93 maize genotypes prepared from different research centers were evaluated in terms of agro-morphological traits under normal and phosphorus deficit conditions using completely randomized design with three replications in the crop year 2016-2017 in an open area conditions. For this purpose, after analyzing the soil of different regions, soil with low phosphorus content (7.240 mg/kg) was selected and the pots (15 kg) were filled with soil and sand in a ratio of two to one. Usable soil phosphorus was determined by the Olsen method in the soil science laboratory of Urmia University. In both optimal and phosphorus deficit conditions, soil was strengthened with nitrogen fertilizers (in the amount of 9 g/pot (during three stages during the growth period)), potassium sulfate (in the amount of 13.5 g/pot), Sequestrin (1.5 g /pot), manganese sulfate (0.225 g/pot), zinc sulfate (0.99 g/pot), copper sulfate (0.3 g/pot) and boric acid (0.21 g/pot). In optimal conditions, phosphorus fertilizer in the form of triple superphosphate was added to each pot in the amount of 6 g/pot. Phosphorous fertilizer was not added to the soil in the phosphorus deficit conditions. With the beginning of the tasseling stage, various agro-morphological and chemical traits were measured. In the physiological maturity stage, the ears relating to each replication were separated from the plants. Seed yield per plant was determined by separating the seeds on the ears of each plant and weighing it. In order to select the desired genotypes, four selection indexes including Smith-Hazel, Pesek-Baker, Brim and Robinson indices were calculated. In this study, the same weight was considered for the attributes, which is done in this way in most studies. To select the best selection index, different criteria, including the genetic gain of traits (∆G), expected gain (∆H) and relative efficiency of selection index (RE) were calculated.

  Results: According to analysis of variance results, the effect of genotype and stress was significant on all studied traits at the probability level of 1%. Also, the interaction effect of genotype × stress was significant on all studied traits except for flag leaf length (FLL), flag leaf width (FLW), number of leaves (NL), plant height (PH), stem diameter (SD), ear length (EL), number of rows per ear (RpE), number of grain per row (GpR), ear diameter (ED) and 100 seeds weight (HGW). In the normal conditions (without stress) in the Smith-Hazel index (optimal index), the highest coefficient (7.21 and 3.98,) was observed in FLW and ear length (EL) traits, respectively, and the lowest ones (-3.03) was observed in RpE trait. In the phosphorus deficit conditions, the highest coefficient (3.91) was observed in EL and the lowest ones (-5.35) was observed in RpE trait. In the Pesak-Baker index, under normal conditions, the highest coefficient (5.64) was observed in GpR trait and the lowest ones (-9.28) was observed in EL. In phosphorus deficit conditions, the highest coefficient (8.49) was seen in FLW trait, followed by EL (4.53) and the lowest ones (-2.17) was observed in RpE. The highest coefficient in Robinson's index under normal conditions was 2.21 for LW, which was -5.91 for that under phosphorus deficit conditions. In this index, the lowest coefficient was seen in RpE (-0.92). In Robinson's index, under phosphorus deficit conditions, the highest coefficient (1.46) was seen in EL, and the lowest ones (-5.92) was seen in FLW, followed by RpE with a value of -2.13. The Smith-Hazel index with expected gain (∆H) value of 296.306 and 229.374, and relative efficiency of (RE) of 1.0011 and 1.0839 and the Brim with expected gain (∆H) value of 296.217 and 233.083, and relative efficiency of selection index (RE) of 0.9995 and 1.0836, in normal and phosphorus deficit conditions, respectively, were the best indices. Under both normal and phosphorus deficit conditions, biomass yield, seed yield, and plant height had the highest coefficients for these indices. Based on both indices, genotype with cod number 7 and the genotype with cod number 10 are introduced as superior genotypes under normal and phosphorus deficient conditions, respectively.

  Conclusion: In general, the results of present investigation showed that in both normal and phosphorus deficit conditions, selection based on the Smith-Hazel and Brim indices will increase the biomass yield, seed yield and plant height due to their highest relative efficiency and expected gain. The selected genotypes after validation at the molecular level with different technologies, such as studying the expression of genes involved in tolerance to phosphorus deficit conditions using Real time PCR, can be used in the production of hybrid varieties as a way to reduce the use of phosphorus fertilizers.