Introduction Safflower is a plant adaptable to areas with limited rainfall during winter and spring and dry air at flowering, seed filling and maturity stages, and tolerant to water deficit. The positive correlation coefficient observed between safflower seed yield with precipitation and low air temperature during seed germination to flowering and flowering to seed maturity. Furthermore, precipitation and low temperature during flowering to seed maturity significantly increased seed oil percentage. Therefore, it seems that, by adjusting the planting time can be adapted phenological stages of plant with appropriate weather conditions. The aims of this research were study the possibility reducing the growth period of safflower spring varieties with maintaining performance through delay in planting time, evaluate seed and oil yields at different planting times, and identify the best varieties for cold and semi-arid areas. Materials and Methods The experiment was conducted at the East Azarbaijan Agriculture and Natural Resources Research Center (46°2¢E, 37°58¢N, 1347 m a.s.l.) during 2014-2015 growing season. According to Koppen climatic classification system, the area climate is semi-arid and cold. This study was carried out as factorial experiment based on randomized complete block design with three replications. Treatments were four planting dates (30 March, 9, 19 and 29 April) and three safflower spring cultivars (Sina, Soffeh and Goldasht). Plant spacing between rows was 24 cm and final plant density was 70 plant per m2. Each plot consisted of 6 rows in 5 meters. Plants were harvested on the 11th and 14th of August in the first and second years of experiment, respectively. At the harvest time, in order to control boarder effects, plants from the sides of each plot were removed. Measured traits were plant height, capitulum diameter, seed yield, capitula number per plant, seed number per capitulum and 1000-seed weight. Ten plants in each plot were used to determine plant height and seed yield components. Also, seed oil content was determined by nuclear magnetic resonance (NMR) method. Results and Discussion The results indicated that interaction effect of year, planting date and cultivar was significant on the plant height, paniculum number per plant, seed number per paniculum, 1000 seed weight , seed yield and seed oil percent. A delay in planting date from 30 March, significantly reduced yield components, seed and oil yields. Soffeh cultivar had the greater plant height and Goldasht and Sina were located in the next rank. Goldasht cultivar had higher seed yield with the largest capitulum diameter and the most 1000 seed weight. The highest oil yield was belonged to the Soffeh (1161 kg ha-1) and Goldasht cultivar (754 kg per ha1) in the first and second year, respectively. There were positive significant correlations between capitulum diameter, 1000 seed weight and seed yield. The Goldasht and Soffeh cultivars provided acceptable performance that were seen appropriate for cultivation in Tabriz cold and semi-arid plain. Delayed planting led to significant decrease in seed and oil yields and therefore is not be recommended. Conclusions The results of this study indicated that the most appropriate date to cultivate spring safflower in Tabriz cold and semi-arid plain is 30 March and planting is not recommended at the late season. Soffeh and Goldasht cultivars are suitable for cultivation in this region.