Introduction
Dragon's head (Lallemantia iberica) is well-suited for cultivation in arid and semi-arid regions due to its low water requirements and short growth cycle (approximately 80-95 days). These characteristics make it particularly valuable for the climatic conditions of Iran, especially in Azerbaijan. Given the diversity in growth and yield characteristics among different ecotypes, evaluating various quantitative and qualitative traits using statistical methods can help identify and introduce the most suitable and adaptable ecotypes for the region. This study aimed to assess the seed oil percentage and yield of 49 ecotypes of Dragon’s head to determine the most promising ecotypes for potential use in breeding programs.
 Materials and Methods
To evaluate oil-related traits in 49 ecotypes of Dragon's head collected from different regions of Iran, a randomized complete block design (RCBD) with three replications was implemented in 2015 and 2016 at the research farm of the Faculty of Agriculture, University of Tabriz. The study measured key agronomic and qualitative traits, including fertile branch length, seed yield per unit area, seed count in the main and lateral branches, oil percentage, and oil yield. Data normality was assessed using SPSS software, and statistical analyses were conducted based on RCBD using MSTAT-C and SPSS. Mean comparisons were performed using Tukey's test at a 1% probability level. To classify the ecotypes, cluster analysis was conducted using Ward's method, and the optimal cut-off point for the resulting dendrogram was determined using discriminant function analysis. Figures were generated using Excel software.
 Results and Discussion
The results revealed significant differences among ecotypes in studied traits. Regression analysis showed that seed yield per unit area had the highest regression coefficient (1.098) for oil yield. Based on mean comparisons, ecotype No. 37 (Alwar village, Bostanabad) had the highest grain yield (116.3 gr m^-2), followed by ecotype No. 23 (Tabriz 4, 107.7 gr m^-2), and ecotype No. 24 (Kalvanaq, 14, 101.7 gr m^-2)  . The variance analysis of ecotype effect on seed oil percentage showed significant differences. Ecotype No. 8 (Kalvanaq 7) had the highest oil percentage (44.3 %), followed by ecotype No. 27 (Param 1, Harris village). Ecotype No. 37 (Alwar village, Bostanabad) had the highest oil yield per unit area with an average of 44.4 gr m^-2.
 Conclusion
Based on the findings, ecotypes No. 37 (from Alwar village, Bostanabad), No. 23 (known as Tabriz 4), and No. 24 (designated Kolvanagh 14) have been identified as the most suitable and compatible ecotypes for the Tabriz region, particularly in terms of seed and oil yield. These ecotypes demonstrated superior performance in various agronomic traits, suggesting their adaptability to the local environmental conditions and agricultural practices. Given their promising results, these ecotypes are strongly recommended for further cultivation efforts, as well as for inclusion in potential breeding programs aimed at enhancing productivity in this region. By focusing on these ecotypes, farmers and agricultural researchers can work towards improving not only the quantity but also the quality of the agricultural outputs, thereby contributing to the economic sustainability of the area. Additionally, the adoption of these ecotypes could lead to increased resilience against local pests and diseases, further supporting the agricultural ecosystem in Tabriz.