Impaginato 367 Adv. Hort. Sci., 2023 37(4): 367­376 DOI: 10.36253/ahsc­13749 Yield related traits in some Persian wal­ nut cultivars: Analysis of genetic and genetic by environment interaction A. Soleimani 1 ,2 (*), V. Rabiei 1 (*), D. Hassani 2 (*), M.R. Mozaffari 3, R. Dastjerdi 2 1 Department of Horticulture, Faculty of Agriculture, University of Zanjan, Zanjan, Iran. 2 Temperate Fruits Research Center (TFRC), Horticultural Science Research Institute (HSRI), Agricultural Research, Education, and Extension Organization (AREEO), Karaj, Iran. 3 Agricultural and National Resources Research Centre of Kerman, Kerman, Iran. Key words: Climatic conditions, cultivar, heritability, Juglans regia L., stability. Abstract: The most important trait in tree species, including walnut, is the yield. In this study, the effect of genotype and their interaction with year on Nut weight, Kernel weight, Kernel percentage, Fruit set, Nuts number on Scaffold (Canopy) Cross Area (SCA), Nut weight on SCA and Kernel weight on SCA were evaluated on Caspian, Persia, Alvand, and Chaldoran walnut cultivars. The results showed that the effects of year, genotype, and year × genotype interac­ tion on all traits were significant. The results showed that Alvand had the high­ est number of nuts (41.8 per m2) and nut weight (472.1 g/m2) on (SCA). Heritability (H2 b) for kernel weight and kernel percentage, were estimated 0.75 and 0.80, respectively. The lowest value of H2 b (0.36) was belong to fruit set. The analyses of genetic and phenotypic correlations between traits showed that, the nut weight had (rg = 0.31, rp = 0.27) a moderate correlation with SCA same as kernel weight (rg = 0.34, rp = 0.29). The GGE biplot analysis explained most of the existing variations (>90%). The genetic effect (PC1) for all traits were higher respect to the genetic × environment interaction (PC2), especially for the kernel percentage (94.4%) and number and weight of nut and kernel on SCA (>90%). The lowest value of the PC1 was related to the fruit set (65.6%), which indicates the trait was more affected by genetic × environment interac­ tions (21.8%). So, this result showed that the yield­related traits in walnut is highly relevant to environment(year in this study) and evaluation of the new cultivars needs careful attention in this case. 1. Introduction The accurate identification of genotypes is a basic requirement for appropriate utilization of germplasm in practical breeding programs. The diverse climatic conditions, environment, and their interactions with (*) Corresponding author: rabiei@znu.ac.ir d.hassani@areeo.ac.ir Citation: SOLEIMANI A., RABIEI V., HASSANI D., MOZAFFARI M.R., DASTJERDI R., 2023 ­ Yield rela‐ ted traits in some Persian walnut cultivars: Analysis of genetic and genetic by environment interaction. ­ Adv. Hort. Sci., 37(4): 367­376. Copyright: © 2023 Soleimani A., Rabiei V., Hassani D., Mozaffari M.R., Dastjerdi R. This is an open access, peer reviewed article published by Firenze University Press (http://www.fupress.net/index.php/ahs/) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Competing Interests: The authors declare no competing interests. Received for publication 10 September 2022 Accepted for publication 1 August 2023 AHS Advances in Horticultural Science https://doi.org/10.36253/ahsc-13749 http://www.fupress.net/index.php/ahs/ http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ Adv. Hort. Sci., 2023 37(4): 367­376 368 genetic are the most important factor determining the performance of the cultivars (Fehr, 1987). Therefore, the genetic and environment implies the differential performance of genotypes that rises from the variations in the genotype’s sensitivities to the environmental conditions (Rawandoozi et al., 2021). Change of climate not only affect the phenology of tree species, but also affects its production. So, with considering the climate changes and abiotic stresses, walnut production in the world has encountered challenges more than ever before. On the other hand, selection for fruit quality traits is complex; because the most of these traits are often controlled by several loci that are also influenced by the envi­ ronment (Bliss, 2009). Nut and kernel weight as well as fruit set percentage could be considered as walnut yield components, while the yield efficiency could include nut and kernel weights produced on trunk cross area (TCA) or scaffold cross area (SCA) (Mahmoodi et al., 2015; Dogra et al., 2018; Hassani et al., 2020 b). These traits can be affected by envi­ ronmental conditions in several ways. For example, the climatic factors affect the receptivity period of walnut pistillate flowers and therefore affect the fruit set percentage and yield of walnut trees (Mariana and Sina Niculina, 2017). Dogra et al. (2018) calculated phenotypic and genetic broad sense heritability of walnut yield relat­ ed traits. Based on their study the pistillate flower density, fruit set percentage, circumference and cross section of tree trunk showed the highest corre­ lation with the yield. Some walnut trees somewhat show different alternate bearing habits, so the yield is affected by the crop load of the previous year (Mahmoodi et al., 2015). Marrano et al. (2019) reported that lateral bearing habit have a significant influence on yield of walnuts. Besides the leafing date had high heritability (88%) and was therefore recommended as a reliable character for improve­ ment of new cultivars. Combining analysis of variance and stability analy­ sis could determine the contribution of genetic, envi­ ronment and their interactions in traits. In spite of cli­ mate change is becoming a bigger challenge every day, determining the genetic and environmental effects can led to understand the response of the cul­ tivars to different environments and select the appropriate cultivars for specific environments and eventually to deal better with changing climate (Bliss, 2009; Rawandoozi et al., 2021). Research with number of genotypes evaluated in different locations and years, makes the genetic × environment analysis a major contest. The GGE biplot analysis is a beneficial tool for data analyzing in multi environment trials (Yan and Tinker, 2006). Rawandoozi et al. (2021) estimated the variance components, genetic × environment interaction and heritability of fruit quality related traits in nine peach and nectarine low to medium chill F1 full­sib families together with their parents in two locations. Based on their research the ripe date and fruit development period had high narrow sense heritability. Fruit weight and shape showed the lower heritability. Scariotto et al. (2013) based on budburst percent­ age and fruit­bearing shoot formation, evaluated the compatibility and stability of peach genotypes in four years. Arji (2018) investigated the stability of yield components of olive cultivars for three years. Despite the high priority for data availability regarding the climatic adaptability of walnut cultivars, there is need for a continuous basis research with the newly released cultivars. Therefore, this study is con­ ducted to evaluate the adaptability of some new Persian walnut cultivars to determine the variance components and cultivars adaptability affecting the yield components and yield efficiency traits, especially with increasing the climate change challenge. 2. Materials and Methods Plant materials and location Walnut yield component together with the yield efficiency traits in four newly released cultivars (i.e., Caspian, Persia, Alvand, and Chaldoran) (Hassani et al., 2020 b) with Chandler and Jamal as reference cul­ tivars, were evaluated in three consecutive years (2015­2017). The cultivars, grafted on Persian walnut seedlings rootstocks, were planted in Karaj in 2006 (35.76031 N, 50.96833 E; elevation: 1240 m a.s.l.; mean annual temperature: 15.8°C; and mean annual precipitation; 247 mm). Evaluated traits The data were recorded on yield component traits including nut and kernel weight and fruit set percent­ age together with the yield efficiency traits such as: nut and kernel weights produced on scaffold cross area (SCA). To estimate the number of pistillate flow­ ers and fruits on experimental trees, pistillate flowers and fruits were counted in sample branches and then were used to predict the whole trees using regression. To measure Scaffold Cross Area (SCA), the tree’s canopy diameter was measured. The SCA was then Soleimani et al. ‐ Genetic and environmental interactions on walnut yield 369 estimated using the canopy and the circle approxima­ tion. For nut and kernel traits, 30 samples in each treatment were evaluated. The tree nut and kernels’ yield were obtained from the number of nuts per tree multiplied per average nut and kernel weights. Next, the nut and kernel yield of trees were divided by the corresponding SCA’s, for estimating yield effi­ ciencies based on nut and kernel (Hassani et al., 2014). To calculate the fruit set percentage, the fruit number in sample branches were divided by the cor­ responding number of pistillate flowers. Statistical analysis The combined analysis of variances was carried out using general linear model (GLM) procedure. Means were separated by Duncan’s Multiple Range Test (DMRT) and Least Significant Difference (LSD). Phenotypic (σ2 p), Genetic (σ2g) and genetic × year interaction (σ2 gy) variances were obtained from their corresponding expected mean square in ANOVA table. Heritability in the broad sense (H2 b) was esti­ mated using the genetic and phenotypic variances (Visscher et al., 2008). The phenotypic and genetic correlations were estimated using the variances and variance­covariance matrices of traits (Dogra et al., 2018; Marrano et al., 2019). The GGE biplot analysis was employed to determine the year and genotype interaction, besides the combining analysis of vari­ ances (Yan and Tinker, 2006). 3. Results Yield‐related traits variability and analysis The descriptive statistics of the traits were report­ ed in Table 1. The nut weight varied from 8­15.3 g with the average of 11.2 g, while the kernel weight average was 5.8 g varying from 3.9­8.3 g. Though the variation in kernel percentage range were 39.4­ 66.7%. The fruit set average was 48.7%, with a wide range variation (14­82%) in different cultivars. Mean number of nuts on SCA were 26.8 with a range of 2.7­ 64.7. Moreover, the average of nut and kernel weight on SCA were 295.1 and 157.4 g/m2, respectively. Nut weight on SCA ranged 28.8­782.9 g/m2, while the ker­ nel weight on SCA ranged 12.1­409.7 g/m2. In gener­ al, a high variation was observed for the evaluated traits in different cultivars. The three years combined analysis of variance and genetic variance components for the studied traits are shown in Table 2. The effect of the year was sig­ nificant on fruit set percent, the nut number on SCA Table 1 ­ Descriptive statistics of the traits evaluated in walnut cultivars Evaluated traits Min Max Range Mean Variance Nut weight (g) 8 15.3 7.3 11.2 2.9 Kernel weight (g) 3.9 8.3 4.4 5.8 1.3 Kernel percentage 39.4 66.7 27.3 52.1 54.5 Fruit set percentage (%) 14 82 68 48.7 264.1 Nut number on scaffold cross area (no./m2) 2.7 67.4 64.7 26.8 271.6 Nut weight on scaffold cross area (g/m2) 28.8 782.9 754.1 295.1 29655.1 Kernel weight on SCA (g/m2) 12.1 409.7 397.6 157.4 9448.5 **, * and NS show statistical significance at the probability level of 1%, 5% and not significant, respectively. SCA = Scaffold cross area; Cv = coefficient of variance; H2 b = broad­sense heritability and SE = Standard error of H2 b. Table 2 ­ Combined analysis of variance, genetic variance components and broad­sense heritability (H2 b) of the traits in six walnut culti­ vars (2015­2017) Variance component DF Nut weight mean squares Kernel weight Kernel percentage Fruit set percentage Nuts number on SCA Nut weight on SCA Kernel weight on SCA Year 2 10.3 NS 0.99 NS 16.8 NS 995.4 * 1149.8 * 159004 * 47838 * Replication (year) 6 1.3 0.18 17.1 142.8 70.1 5671.1 2319.3 Genotype 5 16.2 ** 10.9 ** 430.9 ** 1197.8 * 1056.4 * 141510 * 46238 * Year x genotype 10 3.17 ** 0.71 ** 21.8 ** 330.3 ** 239.9 ** 30441 ** 10044 ** Error 30 0.47 0.19 6.1 95.3 76.7 7912.8 2300.1 Cv (%) 6.1 7.5 4.7 20.1 22.9 30.1 30.5 H2 b ­ 0.41 0.75 0.80 0.36 0.42 0.44 0.45 SE ­ 0.1 0.025 0.015 0.13 0.11 0.104 0.11 Adv. Hort. Sci., 2023 37(4): 367­376 370 and the nut and kernel weight on SCA (P≤0.05). However, it was not significant on nut weight, kernel weight and kernel percent. The effects of genotype and year × genotype interaction was statistically sig­ nificant in all traits (Table 2). A high broad­sense heritability (H2 b) obtained for kernel weight (0.75) and kernel percentage (0.80). The lowest value of H2 b (0.36) was belong to fruit set (Table 2). Based on the analysis of results for determining the effect of different years the highest fruit set per­ centage was observed in 2016 and 2015 with the average of 52.4% and 51%, respectively, although, the lowest fruit set percentage was 42.7% in 2017 (Fig. 1). The highest number of nuts on SCA; and nut and kernel weight on SCA were attained in 2016 with the corresponding averages of 34.3 nuts per m2; and 317.4 and 210.8 g per m2. Evaluation of traits in different genotypes during three experimental years showed that the average of nut weight varied from 9.2 g in Caspian to 13.2 g in Chaldoran. Kernel percentage varied from 42.2% in Chandler to 60.1% in Persia. Fruit set ranged from 33.8% in Persia to 62.7% in Jamal. Furthermore, fruit set was significantly lower in late leafing cultivars and genotypes such as Persia, Chandler, and Caspian (Marrano et al., 2019, Hassani et al., 2020 a) com­ pared with early to medium leafing ones (33.8­43.5% and 55.6­62.7%, respectively) (Fig. 2). Based on the results a wide range of differences was observed in yield efficiency traits (number of nuts per m2 SCA and weight of nut and kernel per m2 SCA). The highest number of nuts per m2 SCA was observed in Alvand with an average of 41.8 nuts/m2, while the lowest amount was recorded in Jamal with 9.1 nuts/m2. Moreover, the highest nut weight on SCA were observed in Alvand with 472.1 g/m2. Alvand and Chaldoran had the highest kernel weight on SCA with 239.7 and 218.6 g/m2, correspondingly. The lowest nut and kernel weight on SCA with 108.2 g/m2 and 50.4 g/m2 belonged to Jamal (Fig. 2). Genetic and genetic per environment (GGE) analysis According to statistically significant interactions between years and genotypes, the studied cultivars showed different responses to years. Analyzing the effect of genotypes and genotype × year interaction on the studied traits have been shown in GGE biplot diagrams in figure 3. In GGE biplot diagrams, the hor­ izontal axis (PC1) shows the effects of genotypes and the vertical axis (PC2) shows the interaction of geno­ type per year (environment). According to the results for nut weight (Fig. 3 a), the PC1 and PC2 explained respectively 79.5% and 19.2% of variability, with the Fig. 1 ­ Mean comparisons for the effect of years on fruit set per­ centage (a), nut number on scaffold cross area (SCA) (b), nut weight on SCA (c), and kernel weight on SCA (d). Soleimani et al. ‐ Genetic and environmental interactions on walnut yield 371 98.7% of total variability. The biplot 2a was divided into four sectors with a principal sector grouping the years 2015 and 2017, together with Chaldoran and Alvand with higher nut weight. While in the second sector, the year 2016 was grouped with Persia. For the percentage of kernel (Fig. 3 b), the PC1 and PC2 explained respectively 94.4% and 4.7% of variability, with 99.1% of total variability in five sectors. The principal sector grouped the years 2015 and 2017. Persia with high kernel percentage was included in this sector. The second sector were grouped the year 2016 together with Chaldoran and Caspian. These cultivars had greater kernel percentage than general average. In figure 3 c the PC1 and PC2 explained respectively 65.6% and 21.8% of variability, with 87.4% of total variability about fruit set percentage. The biplot for fruit set percent was divided into five sectors, too. The principal sector grouped the years 2015 and 2017, and Jamal with higher fruit set. The second sector grouped the year 2016 and Chaldoran. This cultivar had fruit set greater than average. For the fruit number on SCA (Fig. 3 d), the PC1 and PC2 explained 94.2% and 5.5% (99.7% of total) of variabil­ ity correspondingly. The GGE biplot was divided into four sectors. In the principal sector the years 2015 and 2016, and the cultivars Alvand, Caspian and Chaldoran were classified together with higher fruit number on SCA. In the second sector, the year 2017 and Persia were grouped together. Similar results were obtained for nut weight on SCA (Fig. 3 e). For the kernel weight on SCA (Fig. 3 f), the PC1 and PC2 explained respectively 90% and 9.4% of variability Fig. 2 ­ Mean comparisons of the effect of the cultivar and year × cultivar on fruit weight (a), kernel percent (b), fruit set (c), number of fruits on SCA (d), fruit weight on scaffold cross area (SCA) (e), and kernel weight on SCA (f) using Duncan multiple range test for genotypes and Least Significant Difference (LSD) test for interaction of year × cultivar. 372 Adv. Hort. Sci., 2023 37(4): 367­376 and 99.4% of total variability. The GGE biplot for ker­ nel weight on SCA was also divided into five sectors. The principal sector grouped the years 2015 and 2016, and the cultivars Alvand and Chaldoran with higher kernel weight on SCA. The second sector grouped the year 2017 and Persia that had greater kernel weight on SCA. The figure 4 shows scattering of walnut cultivars based on the yield efficiency traits in a biplot. In fig­ ure 4b Chandler, Persia, Chaldoran and Alvand had the highest nut weight on SCA and also nut weight. The same results on figure 4 c with Chaldoran and Alvand which had the highest kernel weight on SCA too. According to figure 4 d the highest fruit set per­ cent and nut weight on SCA also was belong to Alvand and Chaldoran. Genetic and phenotypic correlations The genetic and phenotypic correlations between the traits are reported in Table 3. These results showed that the nut weight did not considerably cor­ relate with kernel percentage and number of nuts on SCA. Nut weight had a moderate impact on fruit weight on SCA (rg = 0.31, rp = 0.27) and kernel weight on SCA (rg = 0.34, rp = 0.29). As expected, a high genetic and phenotypic correlation was observed between the number of nuts on SCA and nut weight on SCA (rg = 0.95, rp = 0.95) as well as kernel weight on SCA (rg = 0.90, rp = 0.91) (Table 3). Kernel weight on SCA was significantly correlated with most of the traits, but the highest genetic and phenotypic corre­ lations were observed between this trait and fruit weight on SCA (rg = 0.97, rp = 0.97). Fig. 3 ­ Genotype and Genetic × Environment (GGE) biplot of six walnut cultivars over three years (2015­2017) for nut weight (a), kernel percentage (b), fruit set percentage (c), nut number on scaffold cross area (SCA) (d), nut weight on SCA (e) and kernel weight on SCA (f). Soleimani et al. ‐ Genetic and environmental interactions on walnut yield 373 4. Discussion and Conclusions The significant effects of year, genetic and genetic × year interaction showed that cultivar and its inter­ action with environmental conditions as the main determinants of cultivar’s adaptability. Therefore, stable and compatible cultivars should be found and introduced for appropriate climate(s) (Rawandoozi et al., 2021). The low fruit set in 2017 caused the fruit production to be significantly lower compared to other two years, while there were no significant dif­ ferences in pistillate flowers (data not shown). Understanding the genotype and genetic × envi­ ronment interaction also is important for increasing Fig. 4 ­ Biplot of regression coefficients against average yields of walnut cultivars (The horizontal solid line represents the mean coeffi­ cient of regression and the vertical solid line denotes the average fruit weight on scaffold cross area (SCA). The standard error (±1SE) was included and represented by the dotted lines for both yield efficiency and regression coefficients) (a), biplot of nut weight with nut weight on SCA (b), kernel weight with kernel weight on CA (c), and percentage of fruit set with nut weight on SCA (d). Solid lines in graphs show the average of each trait. Table 3 ­ Genetic and phonotypic correlations of different traits G and P are the genetic and phenotypic correlations, respectively. SCA = Scaffold cross area. Kernel (%) Fruit set (%) Number of nuts on SCA2 Nut weight on SCA Kernel weight on SCA Nut weight G1 0.19 0.72 ­0.02 0.31 0.34 P1 0.12 0.53 ­0.03 0.27 0.29 Kernel (%) G 1 ­0.34 0.25 0.31 0.55 P ­0.26 0.26 0.30 0.52 Fruit set (%) G 1 ­0.29 ­0.09 ­0.14 P ­0.12 0.03 0.01 Number of nuts on SCA G 1 0.95 0.90 P 0.95 0.91 Nut weight on SCA G 1 0.97 P 0.97 Kernel weight on SCA G 1 P Adv. Hort. Sci., 2023 37(4): 367­376 374 the gain in cultivar improvement programs. The genotype’s main effect nd especially it’s G x E interac­ tion implies the different performance of genotypes across environments that arises from the various sensitivities to the different environments (Rawandoozi et al., 2021). For all of the yield related traits, GGE biplot have described most of the existing variations (more than 90 %), which indicates the rela­ tive validity of the biplot in explaining the variations of genotypes and genetic × environment interaction (Yan and Tinker, 2006). The effect of genotype in determining the walnut yield efficiency has been demonstrated by Dogra et al. (2018). Based on our results high amounts of PC1 has been recorded for kernel percentage (94.4%) as well as number and weight of nut and kernel on SCA (more than 90%). So, a high and stable production will be expected by selecting cultivars with higher yield efficiencies and more compatible with environmental conditions. The results on nut number and nut weight in cultivars are consistent with the findings of Mahmoodi et al. (2016). Fruit set is another important trait affecting the number of nuts in tree (McGranahan and Leslie, 2009; Sarikhani Khorami et al., 2014; Khadivi­Khub et al., 2015). It is clear that, number of pistillate flowers is relatively lower in cultivars with terminal bearing habit compared to lateral bearing ones (McGranahan and Leslie, 2009; Hassani et al., 2020 b). In terminal bearing cultivars like Jamal, the higher fruit set usual­ ly could compensate the production to some extent. Among the yield related traits, the lowest value of the PC1 was belong to fruit set (65.6%), which indi­ cates this trait is more affected by genetic × environ­ ment interaction (21.8%). The results of this study present the clear effect of genotype on fruit set (Fig. 2 c) which is consistent with Kumar et al. (2005). Due to the fact that the amount of pistillate flowers pro­ duced in walnut is lower than pome and stone fruit trees, higher fruit set (50­90%) is necessary in order to produce an adequate yield. In addition to genetic, genetic × environment interaction plays a very impor­ tant role in pollination and fruit set of cultivars (Cosmulescu et al., 2010; Mariana and Sina Niculina, 2017). According to the results, there was a signifi­ cant negative correlation between bud break and fruit set (R = ­0.54), so that with delayed leafing, the fruit set decreased. It is clear that the late leafing cul­ tivars deal better with late spring frosts (McGranahan and Leslie, 2006; Hassani et al., 2013; Hassani et al., 2020 b), but the pollination and fruit set were not the same in late and early leafing walnut cultivars. To obtain a sufficient fruit set, care must be taken regarding producing a sufficient pollen volume with an adequate overlap of pollen­shedding for the receptivity period of pistillate flowers. The response of cultivars could be affected by different climatic conditions in different years especially at leafing time and time of pollination (Cosmulescu et al., 2010; Sarikhani Khorami and Vahdati, 2019; Cao et al., 2020). Temperature is one of environmental factors influences the percentage of fruit set by influencing pollination factors, such as pistillate flowers receptiv­ ity and pollen­shedding period. In late leafing culti­ vars the environmental factors such as high tempera­ tures at the pollination time, could lead to lower effective pollination period and pistillate flower receptivity period, that could reduce the fruit set (Ramos, 1997). High broad­sense heritability relative to kernel weight and kernel percentage also indicated that these traits are less affected especially by genetic × environment interaction. Conversely, low­moderate heritability and high ratio of genetic × environment interaction for fruit set indicated substantial environ­ mental effects on this trait. The heritability values in the present study were somewhat lower than what reported by Eskandari et al. (2006), Dogra et al. (2018), and Marrano et al. (2019). In terms of yield stability, it seems that the genet­ ics, environment and their interaction could con­ tribute to various characteristics such as: fruit­bear­ ing habit, growth vigor, nut weight, kernel weight, kernel percentage, previous year crop load, pollina­ tion, fruit set and late spring frosts (Cosmulescu et al., 2010; Asma, 2012; Sarikhani Khorami et al., 2014; Dogra et al., 2018; Cao et al., 2020; Hassani et al., 2020 a). Generally, in low­yielding cultivars such as Jamal, year­by­year variations of traits were low. However, they were higher in cultivars with more production such as Chaldoran and Alvand (Hassani et al., 2020 a). Some studies have reported significant alternate bearing in walnut cultivars (Asma, 2012; Hassani et al., 2014; Mahmoodi et al., 2016). So, alternate bearing, opposed to genetic stability, is affecting the fruit production trends of walnut culti­ vars in different years (Amiri et al., 2010). Mahmoodi et al. (2015), reported the presence of 2­15% of alter­ nate bearing among different walnut cultivars. In majority of high­yielding cultivars, a heavy crop load is followed by a low fruit production in the subse­ quent year. Asma (2012) found that the yield is influ­ enced mostly by leafing time, fruit­bearing habit, tree Soleimani et al. ‐ Genetic and environmental interactions on walnut yield 375 size, nut and kernel weights, and kernel percentage. Moreover, Dogra et al. (2018) reported that yield is controlled polygenically and is influenced by environ­ mental conditions. They stated that pistillate flower density, fruit set, trunk section area, trunk circumfer­ ence, tree height, shoot length, pollen­shedding peri­ od, fruit weight, kernel percentage, and shell thick­ ness had affected the yield of walnut trees, which were in part consistent with the findings of the pre­ sent study. High variation was observed in yield components and yield efficiency traits with different environments and cultivars, and it was found that genetic by envi­ ronment interaction are the most important factors determining yield variations. Understanding the con­ tribution of genetics and genetic × environment interaction is very important. The genetic × environ­ ment interaction in fruit set was more than other yield­related traits, while the broad sense heritability (H2 b = 0.36) was the lowest value. Therefore, fruit set is most affected by variation of environmental condi­ tions, so that under undesirable climatic conditions it will be yield determining factor especially in late leaf­ ing walnut cultivars. Regarding the nut weight, the effect of genetic by environment interaction was strong while heritability was greatly affected by the environmental conditions. The contribution of genet­ ic × environment interaction on other traits related to yield efficiency was estimated to be less than 10%. Genetic and phenotypic correlation also indicated that nut weight, kernel weight and kernel percentage had a low­moderate correlation with nut and kernel weight on SCA. On the contrary, the nut number on SCA had the highest genetic and phenotypic correla­ tion with nut and kernel produced on SCA. The results showed that in walnuts, that is a nut tree species well adapted to temperate climate, the vari­ ability of yield related traits over environments (years), was highly significant. So, the change in cli­ mate in one hand and the scarcity of the resources (water and land) on the other hand emphasizes on more accurate evaluation on the new cultivars espe­ cially in yield related traits. Acknowledgements The authors sincerely thank the Temperate Fruits Research Center of Horticultural Science Research Institute (HSRI) for supporting the research. References AMIRI R., VAHDATI K., MOHSENIPOOR S., MOZAFFARI M.R., LESLIE C., 2010 ­ Correlations between some hor‐ ticultural traits in walnut. ­ HortScience, 45: 1690­1694. ARJI I., 2018 ­ Stability analysis of fruit yield of some olive cultivars in semi‐arid environmental condition. ­ Adv. Hort. Sci., 32(4): 517­524. ASMA B.M., 2012 ­ Pomological and phenological charac‐ terization of promising walnut (Juglans regia L.) geno‐ types from Malatya, Turkey. ­ Acta Scientiarum Polonorum Hortorum Cultus, 11: 169­178. BLISS F., 2009 ­ Marker‐assisted breeding in horticultural crops. ­ Inter. Symposium on Molecular Markers in Horticulture, 859: 339­350. CAO G.­X., LI R.­T., LI L., ZENG H., WANG J.­Y., 2020 ­ Gender specialization and factors affecting fruit set of the wind‐pollinated heterodichogamous Juglans regia. ­ Plant Species Biol., 35: 138­146. COSMULESCU S., BACIU A., BOTU M., ACHIM G., 2010 ­ Environmental factors’ influence on walnut flowering. ­ Acta Horticulturae, 861: 83­88. DOGRA R., SHARMA S., SHARMA D., 2018 ­ Heritability estimates, correlation and path coefficient analysis for fruit yield in walnut (Juglans regia L.). ­ J. Pharm. Phytochem., 7: 3707­3714. ESKANDARI S., HASSANI D., ABDI A., 2006 ­ Investigation on genetic diversity of Persian walnut and evaluation of promising genotypes. ­ Acta Horticulturae, 705: 159­ 166. FEHR W.R., 1987 ­ Principles of cultivar development. Volume 1. Theory and technique. ­ Macmillan Publishing Company, New York, NY, USA. HASSANI D., DASTJERDI R., SOLEIMANI A., JAFFARAGHAEI M., REZAEE R., VAHDATI K., DEHGHANI A., HADADNE­ JHAD H., ASEFNOKHOSTIN M., MOZAFFARI M., 2014 ­ A model for estimation of the potential yield of walnut trees. ­ Acta Horticulturae, 1050: 407­412. HASSANI D., HAGHJOOYAN R., SOLEIMANI A., ATEFI J., LONI A., 2013 ­ Evaluation of some walnut cultivars and selections in Iran. ­ Acta Horticulturae, 981: 59­64. HASSANI D., MOZAFFARI M.R., SOLEIMANI A., DASTJERDI R., REZAEE R., KESHAVARZI M., VAHDATI K., FAHADAN A., ATEFI J., 2020 a ­ Four new persian walnut cultivars of Iran: Persia, Caspian, Chaldoran, and Alvand. ­ HortSci., 55: 1162­1163. HASSANI D., SARIKHANI S., DASTJERDI R., MAHMOUDI R., SOLEIMANI A., VAHDATI K., 2020 b ­ Situation and recent trends on cultivation and breeding of Persian walnut in Iran. ­ Scientia Hortic., 270: 109369. KHADIVI­KHUB A., EBRAHIMI A., SHEIBANI F., ESMAEILI A., 2015 ­ Phenological and pomological characterization of Persian walnut to select promising trees. ­ Euphytica, 205: 557­567. KUMAR A., KUMAR K., SHARMA S.D., 2005 ­ Extent of fruit Adv. Hort. Sci., 2023 37(4): 367­376 376 set and retention under different modes of pollination in Persian walnut (Juglans regia L.) . ­ Acta Horticulturae, 696: 327­330. MAHMOODI R., HASSANI D., AMIRI M., AGHAEI M., VAH­ DATI K., 2015 ­ Relationship between some traits and nut production in walnut cultivars and genotypes. ­ J. Crop Production and Processing, 4: 63­74. (In Arabic). MAHMOODI R., HASSANI D., AMIRI M.E., JAFFARAGHAEI M., 2016 ­ Phenological and pomological characteris‐ tics of five promised walnut genotypes in Karaj, Iran. ­ J. Nuts, 7: 1­8. MARIANA B.I., SINA NICULINA C., 2017 ­ Effect of climatic conditions on flowering of walnut genotypes in Romania. ­ Journal Nuts, 8: 161­167. MARRANO A., SIDELI G.M., LESLIE C.A., CHENG H., NEALE D.B., 2019 ­ Deciphering of the genetic control of phe‐ nology, yield, and pellicle color in Persian walnut (Juglans regia L.). ­ Frontiers Plant Sci., 10: 1140. MCGRANAHAN G., LESLIE C., 2009 ­ Breeding walnuts (Juglans regia). Breeding plantation tree crops: Temperate species. ­ Springer, New York, NY, USA, pp. 249­273. MCGRANAHAN G.H., LESLIE C.A., 2006 ­ Advances in genet‐ ic improvement of walnut at the University of California, Davis. ­ Acta Horticulturae, 705: 117­122. RAMOS D.E., 1997 ­ Walnut production manual. ­ UCANR publications. RAWANDOOZI Z., HARTMANN T., BYRNE D., CARPENEDO S., 2021 ­ Heritability, Correlation, and Genotype by Environment Interaction of Phenological and Fruit Quality Traits in Peach. ­ J. Amer. Soc. Hortic. Sci., 146(1): 56­67. SARIKHANI KHORAMI S., ARZANI K., ROOZBAN M.R., 2014 ­ Correlations of certain high heritability horticultural traits in Persian walnut (Juglans regia L.). ­ Acta Horticulturae, 1050: 61­68. SARIKHANI KHORAMI S., VAHDATI K., 2019 ­ Determination of Persian walnut yield components and its correlation with phenological, morphological and biochemical traits. ­ Iranian J. Hortic. Sci., 50: 549­560. SCARIOTTO S., CITADIN I., RASEIRA M.D.C.B., SACHET M.R., PENSO G.A., 2013 ­ Adaptability and stability of 34 peach genotypes for leafing under Brazilian subtropical conditions. ­ Scientia Hortic., 155: 111­117. VISSCHER P.M., HILL W.G., WRAY N.R., 2008 ­ Heritability in the genomics era‐concepts and misconceptions. ­ Nature Reviews Genetics, 9: 255­266. YAN W., TINKER N.A., 2006 ­ Biplot analysis of multi‐envi‐ ronment trial data: Principles and applications. ­ Canadian J. Plant Sci., 86: 623­645.