In order to screen the suitable genotypes of doubled haploid winter wheat from the view of effective agronomic traits on yield for planting in cold semiarid regions of country, a trial was planed with 39 lines of doubled haploid wheat and 2 lines of no irrigated wheat (Sabalan and zar) and the irrigated line Zarrin. The design of the trial was Rectangular Lattice and contained 3 replications that was conducted in the crop year 2001-2002. The lines and varieties were studied according to the following traits: Time to initial stem, flowering and ripening, length of flag leaf, number of tillers, number of spikes, plant height, length of spike, number of seed per spike, 1000 kernel weight, harvest index and economic yield. Analysis of variance and statistical analysis of yield and dependent traits showed that there is a significant difference among different ecotypes. Differences medicate that there is a wide genetic diversity among the mentioned traits. Grouping of genotypes for the above traits showed the genotypes that initial stem and flowers more rapidly and need less time for physiological maturity nave more yield. Correlation between necessary time for physiological maturity and economic yield was negative and significant and correlation between number of tillers and yield was positive and significant. From the view of spike length, 1000 kernel weight and number of seeds there was a significant difference amongst the genotypes. Correlation between harvest index and yield was positive and much significant (r= 0.456) and correlations between yield and 1000 kernel weight , yield and height, rooo kernel weight and height were positive and much significant (r= 0.288, r= 0.364 ana r= 0.4 respectively) that these Show the Importance of selecting genotypes with the mentioned traits. It sounds that the genotypes selected for semiarid regions should, as well as suitable yield, have the following traits: suitable height, high 1000 kernel weight, Precocious, less tillers, more fertile tillers and suitable harvest index. In tills study the genotypes 5 and 28 from the view of some traits were similar to no irrigated improved and commercial-lines that were also selected for the complementary trial. Correlation between necessary time for physiological maturity and economic yield was negative and significant and correlation between number of tillers and yield was positive and significant. From the view of spike length, 1000 kernel weight and number of seeds there was a significant difference amongst the genotypes. Correlation between harvest index and yield was positive and much significant (r= 0.456) and correlations between yield and 1000 kernel weight, yield and height, 1000 kernel weight and height were positive and much significant (r= 0.288, r= 0.364 and r= 0.4 respectively) that these snow the importance of selecting genotypes with the mentioned traits. It sounds that the genotypes selected for semiarid regions should, as well as suitable yield, have the following traits: suitable height, high 1000 kernel weight, precocious, less tillers, more fertile tillers and suitable harvest index.