Impaginato 441 Adv. Hort. Sci., 2020 34(4): 441­448 DOI: 10.13128/ahsc­7725 Variations in tree and fruit characteris­ tics revealed potential dwarfing geno­ types within Iran’s pomegranate germplasm A.A. Ghasemi Soloklui 1, A. Gharaghani 1, 2 (*), A. Sarkhosh 3 1 Department of Horticultural Sciences, Faculty of Agricultural Sciences, Shiraz University, Shiraz, Iran. 2 Drought Research Centre, Faculty of Agricultural Sciences, Shiraz University, Shiraz, Iran. 3 Horticultural Sciences Department, University of Florida, Gainesville, FL 32611, Florida, USA. Key words: internodes length, rootstocks, stomatal density, sucker. Abstract: This study aimed to explore diversity in dwarfing tendencies, deter­ mine the correlation of measured traits with dwarfing, and identify and select promising dwarf candidates as potential scions or rootstock cultivars. Growth habit, vegetative attributes, fruit physicochemical characteristics,and leaf stom­ atal densityof 19 Iranian pomegranate cultivars, which have been collected across the country and established in the Yazd pomegranate germplasm, were assessed. Results showed that the cultivars differed in almost all measuredtraits. The tree height and canopy width, current year’s shoot, and internode length were within the range of 1.97­4.6 m, 1.53­4 m, 15­41.5 cm 1.96­3.39 cm, respec­ tively. Moreover, a positive correlation was observed between tree height and internode length (r= 0.55), whereas a negative correlation was obtained between stomatal density and tree height (r= ­0.44). Based on characteristics measured for the vegetative growth, ‘Malas No. 1 Saravan’ and ‘Torosh Nar Riz Zirab’ proved dwarfing habit. ‘Rabab Poost Ghermez Neyriz’,a commercial culti­ var, showed semi­dwarfing growth and ‘Khajei Ghasrodasht Fars’, ‘Shahsavar Seydan Marvdasht’, ‘Bihaste Ravar’, ‘Bihaste Sangan Khash’, ‘Torosh Goli Naz Behshahr’ and ‘Anar Siah’ resulted in vigorous trees. This preliminary study found promising dwarf and semi­dwarf genotypes at Iran’s pomegranate germplasm. 1. Introduction According to historical documents, pomegranates originated in central Asia, especially in parts of Iran, and believed to have spread to nearby areas due to traveling and incursion (Harlan, 1975; Levin, 1994; Verma et al., 2010). The main Iranian collection of pomegranate in Yazd contains 762 accessions, including wild, semi­wild, and commercial types (Behzadi Shahrbabaki, 1998; Zamani et al., 2007). These diverse and valuable (*) Corresponding author: agharghani@shirazu.ac.ir Citation: GHASEMI SOLOKLUI A.A., GHARAGHANI A., SARKHOSH A., 2020 ­ Variations in tree and fruit characteristics revealed potential dwarfing genotypes within Iran’s pomegranate germplasm. Adv. Hort. Sci., 34(4): 441­448 Copyright: © 2020 Ghasemi Soloklui A.A., Gharaghani A., Sarkhosh A. 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 24 December 2019 Accepted for publication 30 November 2020 AHS Advances in Horticultural Science http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ Adv. Hort. Sci., 2020 34(4): 441­448 442 genetic resources would benefit from a further genetic improvement to develop dwarfing scion and rootstock genotypes to establish modern high­density pomegranate orchards. Ingels et al. (2002) suggested that the term ‘dwarf tree’ applies to a tree that appears smaller than usual owing to selection of dwarf genotypes, specific training or pruning meth­ ods, or grafting on dwarfing rootstocks. Besides, in classifying trees according to size, dwarf trees are approximately 2.5 m or less when they mature (Castle, 1992). Dwarf trees have many benefits com­ pared to larger vigorous ones, such as being able to be spaced closer together without suffering from excessive crowding or the need for frequent, severe pruning. Moreover, dwarf trees allow for ease of pruning, pest control, fruit thinning, spraying, harvest­ ing, and increased production of high­grade fruit, higher fruit quality and decreased production costs (Tukey, 1964). The advantages of dwarf and semi­ dwarf genotype trees have been demonstrated in the fruit industry resulting in the widespread use of dwarfing rootstocks in tree crops such as apple (Looney and Lane, 1983), cashew (Moura, 2001), and peach (DeJong et al., 2005). Lately, breeding efforts have resulted in the selection of dwarfing rootstock or scion cultivars in almost all temperate and tropical fruit tree crops (Busov et al., 2003). The primary iden­ tification of dwarf genotypes should be based on field evaluations of fruit tree cultivars and genotypes. Genetic dwarfism often manifests itself in distinctive morphological characteristics, easily identified, and often appears to a casual observer (Castle, 1992). Analysis of morphological diversity as tree height, canopy shape, internode, and branching pattern is a useful method for detecting dwarfing phenotypes in a population of many genotypes. Although some pome­ granate cultivars, such as Nana, are considered dwarf­ ing (Terakami et al., 2007), these are ornamentals, and there are no published reports on new dwarfing pomegranates. The objectives of this study were: 1) evaluate tree growth habits, vegetative and fruit char­ acteristics of 19 Iranian pomegranate cultivars, 2) to examine diversity in dwarfing potential, 3) determine inter­correlations among measured traits with dwarfism, and 4) identify promising candidates with dwarfing potential as pomegranate scions and root­ stocks. 2. Materials and Methods Plant materials The experiment was conducted in 2015, at the Agriculture and Natural Resources Research Centre Table 1 ­ Name, origin, fruit and tree characteristics of pomegranate cultivars used in this study Cultivars Provinces Cities Skin color Aril color Taste Uses Anar Siah Esfahan Esfahan Black Dark red Sweet Medicinal Bihaste Ravar Kerman Ravar Yellow­pink White Sweet Local Bihaste Sangan Khash Sistan Baluchistan Khash White White Sweet Local Jangali Poost Ghermez Roodbar Gilan Roodbar Red Red Sweet­sour Local Khajei Ghasrodasht Fars Fars Shiraz Pink White Sweet Local Malas Pishva Varamin Tehran Varamin yellow White Sweet Local MalasYazdi Yazd Yazd Red Red Sweet­sour Commercial Makhmal Malas Shahreza Esfahan Shahreza Red Red Sweet­sour Local Malas No. 1 Saravan Sistan Baluchistan Saravan Yellow White Sweet­sour Local Poost Nazok Torosh Abarkuh Yazd Abarkuh Red Red Sweet­sour Local Poost Sefid Dezfoul Khuzestan Dezfoul Yellow­white White Sweet­sour Local Rabab Poost Ghermez Neyriz Fars Neyriz Red Red Sweet­sour Commercial Rabab Poost Ghermez Kazeroon Fars Kazeroon Red Pink Sweet­sour Local Sefid Biardal Borujen Chahar Mahal­e Bakhtiar Borujen Yellwo Pink Sour Local Shirin Jangal Sisangan Mazandaran Sisangan Red­yellow Pink Sweet­sour Local Shirin Semnan Semnan Semnan Green­yellow White Sweet Local Shahsavar Seydan Marvdasht Fars Marvdasht White­yellow White Sweet­ Local Torosh Goli Naz Behshahr Mazandaran Behshahr White­yellow White Sour Local Torosh Nar Riz Zirab Fars Darab Green­yellow White Very sour Wild Ghasemi Soloklui et al. ‐ Potential dwarfing within Iran’s pomegranate germplasm 443 (ANRRC), Yazd Province, Iran. Nineteen pomegranate cultivars were used in the study (Table 1). Two of the cultivars, ‘Rabab Poost Ghermez Neyriz’ and ‘Malas Yazdi’, are commercial cultivars widely cultivated in the country. The rest are of local importance in dif­ ferent provinces, except ‘Torosh Nar Riz Zirab’, a non­commercial (semi­wild type) from the Darab region. The pomegranate cultivars have been planted in a randomized complete block design (RCBD) with three replications per cultivar. Trees were 26­year­ old at the time of the experiment and managed fol­ lowing the region’s recommended orchard practices. Measurement of vegetative characteristics A total of 6 trees per cultivar (3 trees × 3 repli­ cates) was used for vegetative traits measurements. Since the pomegranate trees are trained to the multi­ ple trunk (3­4 trunks) system (the common practice in Iran’s commercial orchards), individual trunks diameters were measured at 30 cm above the soil surface and then averaged to get the value for the single trunk diameter. Bark thickness was measured in a small­detached section of bark at 30 cm above the soil surface. Tree height (from the ground level up to the tree peak) and canopy width (in the widest point). Numbers of suckers were simply counted on trees and for shoot angle (o), the insertion of shoots that came directly from the scaffold were measured. In addition, the current year’s shoot length (cm) was evaluated after shoot growth stopped (in November) on three scaffolds of trees. Moreover, internode length (cm) was calculated by dividing the current year’s shoot length by its corresponded node num­ ber. Measurement of fruit quality attributes Physical properties. At harvest (which was varied for each cultivar), measurements of fruit physical properties were done on 90 randomly selected fruits per cultivar (10 fruits per tree × 3 trees per replicate × 3 replicate = 90 fruits for each cultivar). Fruits were weighed using a digital balance. Peel and arils were carefully separated manually from the fruit to mea­ sure the edible portion. The extracted arils were col­ lected in a tray and mixed thoroughly to assure uni formity.The edible portion of the fruit was deter­ mined using the following formula (Ghasemi Soloklui et al., 2019): Edible portion of fruit (%)= fresh weight ­ peel weight ­ capillary membranes x 100 fruit weight Chemicals properties. At harvest, total soluble solids content (TSS) and total acidity (TA) were mea­ sured in juice extracted from 90 fruits per cultivar (10 fruits per tree × 3 trees per replicate × 3 replicates = 90 fruits for each cultivar). TSS (in °Brix) was deter­ mined using a digital refractometer (model PR­1, Atago, Japan) and TA by titration to pH end­point 8.2 with 0.1 N NaOH and expressed as citric acid equiva­ lent (g CAE100 mL­1) (Horwitz, 1980). Measurement of stomatal density In late summer, fully expanded leaves (15 leaves per replication) were collected randomly from the midpoint of the current season’s shoots. Stomata numbers were determined using the replica method (Soleimani et al., 2002). The stellate hairs were removed from the lower surface of each leaf using an adhesive tape. A thin film of cellulose acetate was painted directly onto the lower epidermis of the leaf. The cellulose acetate was allowed to dry at room temperature before being peeled from the leaf. Sections were taken from the middle of each leaf. The slides were coded, and a binocular microscope was used for stomatal counts at x 40 magnification. Stomatal density was counted in a field area of one mm2. Statistical analysis Analysis of variances (ANOVA) was performed using SAS version 9.1 (SAS, 2003.). The means were carried out at P<0.05 using Duncan’s multiple range tests. Correlation between pairs of traits was deter­ mined using Pearson’s correlation coefficient. 3. Results Vegetative characteristics Measured vegetative traits are presented in Table 2 and figure 1. Tree height varied between 1.97 to 4.6 m (Fig. 1). The highest tree height value was observed in ‘Khajei Ghasrodasht Fars’ (4.6 m), fol­ lowed by ‘Bihaste Ravar’ (4.3 m), ‘Bihaste Sangan Khash’ (4.26 m), and ‘Shahsavar Seydan Marvdasht’ (4.26 m). ‘Malas No. 1 Saravan’ showed the smallest tree height (1.97 m) (Fig. 2), some cultivars such as ‘Rabab Poost Ghermez Neyriz’ and ‘Shirin Semnan’ categorized as medium height (Fig. 1). Among the 19 cultivars, the greatest canopy width (4 m) was observed in ‘Shahsavar Seydan Marvdasht’, while ‘Makhmal Malas Shahreza’showed the smallest canopy width (1.50 m) (Fig. 1). ‘Khajei Adv. Hort. Sci., 2020 34(4): 441­448 444 Ghasrodasht Fars’ had the greatest single trunk diam­ eter (121.33 mm), whereas ‘Malas No.1 Saravan’, had the lowest single trunk diameter (61 mm) (Table 2). The longest current year’s shoot length (41.50 cm) was recorded in ‘Poost Nazok Torosh Abarkuh’, while the shortest (15 cm) obtained in ‘Poost Sefid Dezfoul’ (Table 2). Internode length varied from 1.96 cm (‘Malas No. 1 Saravan’) to 3.39 cm (‘Khajei Ghasrodasht Fars’) (Table 2). The correlation between internode length and tree height was also statistically significant (r=0.55; p=0.0001) (Fig. 3). Shoot angle was within the range of 46.6 to 90 (Table 2). Significant differences were observed in the num­ ber of suckers among the pomegranate cultivars. The highest number of suckers per tree was recorded in ‘Sefid Biardal Borujen’ (215), while the lowest num­ ber was counted in ‘Rabab Poost Ghermez Kazeroon’. In addition, the highest (3.24 mm) and lowest (1.55 mm) bark thickness were recorded in ‘Poost Nazok Torosh Abarkuh’ and ‘Sefid Biardal Borujen’, respec­ tively (Table 2). Considering all vegetative characteristics, in par­ ticular, tree height, trunk diameter and internode length; cultivars could be classified into four groups, Fig. 1 ­ Tree height and canopy width in 19 Iranian pomegranate cultivars. Similar letters indicate non­significant differen­ ces among cultivars (P≤0.05). Fig. 2 ­ Comparison of tree height between a vigorous pomegra­ nate cultivar (A) and ‘Malas No. 1 Saravan’ (B), the most dwarf cultivar in the pomegranate genotypes studied. Table 2 ­ Vegetative tree characteristics of the pomegranate cultivars used in the experiment Similar letters in each column indicate non­significant differences among cultivars at P≤0.05. Cultivars Trunk diameter (mm) Current year shoot length (cm) Internode length (cm) Shoot angle (o) Number of suckers Bark thickness (mm) Anar Siah 114 ab 21 bcdefg 3.11 ab 57 defg 18 gh 2.03 b Bihaste Ravar 98 ab 24.16 b 3.17 ab 60.16 defg 55 bc 2.60 b Bihaste Sangan Khash 99 ab 17 defg 2.83 abc 53.33 fg 90 cd 2.22 b Jangali Poost Ghermez Roodbar 78 ab 15.66 fg 2.57 abc 63.33 defg 191 bcd 2.83 b Khajei Ghasrodasht Fars 121.33 a 22 bcde 3.39 a 65 def 8.33 h 2.56 b Malas Pishva Varamin 97.67 ab 16.33 efg 3.17 ab 65 def 161 abc 1.82 b Malas Yazdi 74.67 ab 23 bcd 2.14 c 46.66 g 70 efg 2.46 b Makhmal Malas Shahreza 77 ab 26.50 b 2.05 c 86.66 ab 180 abc 1.77 b Malas No. 1 Saravan 61.0 b 17.6 cdefg 1.96 c 60 defg 22 gh 1.70 b Poost Nazok Torosh Abarkuh 93.83 ab 41.50 a 3.39 a 63.33 defg 70 efg 3.24 a Poost Sefid Dezfoul 89 ab 15 g 2.39 bc 73.33 abcde 93 def 1.64 g Rabab Poost Ghermez Neyriz 110 ab 21 bcdefg 2.52 abc 56.66 efg 143 bcd 2.23 bcdefg Rabab Poost Ghermez Kazeroon 107.17 ab 16.33 efg 2.33 bc 83.33 abc 1 h 1.62 g Sefid Biardal Borujen 111.33 ab 23.50 bc 2.61 abc 86.66 ab 215 a 1.55 g Shirin Jangal Sisangan 73.76 ab 17.6 cdefg 2.01 c 68.33 cdef 89 def 2.10 cdefg Shirin Semnan 99 ab 16.83 defg 2.80 abc 75 abcd 126 def 2.92 ab Shahsavar Seydan Marvdasht 114 ab 21.3 bcdef 3.17 ab 70 bcdef 35 fgh 2.64 abcd Torosh Goli Naz Behshahr 67.75 ab 17.6 cdefg 2.33 bc 90 a 35 fgh 2.40 bcdef Torosh Nar Riz Zirab 62.17 b 21.3 bcdef 2.52 abc 66.66 cdef 69 efg 2.42 bcdef Ghasemi Soloklui et al. ‐ Potential dwarfing within Iran’s pomegranate germplasm 445 1) vigorous cultivars: ‘Khajei Ghasrodasht Fars’, ‘Shahsavar Seydan Marvdasht’, ‘Bihaste Ravar’, ‘Bihaste Sangan Khash’, ‘Torosh Goli Naz Behshahr’ and ‘Anar Siah’, 2) Semi­vigorous cultivars: ‘Malas Pishva Varamin’, ‘Rabab Poost Ghermez Kazeroon’, ‘Poost Nazok Torosh Abarkuh’, ‘Jangali Poost Ghermez Roodbar’, ‘Sefid Biardal Borujen’, ‘Malas Yazdi’ and ‘Shirin Jangal Sisangan’, 3) Semi­dwarf cul­ tivars: ‘Rabab Poost Ghermez Neyriz’, ‘Shirin Semnan’, ‘Makhmal Malas Shahreza’ and ‘Poost Sefid Dezfoul’, and 4) Dwarf cultivars: ‘Torosh Nar Riz Zirab’ and ‘Malas No. 1 Saravan’. Stomata density As shown in figure 4, a large variation in stomatal density (from 46.91 to 108.91 stomata per mm2) was observed among studied cultivars. ‘Shirin Semnan’, showed the highest stomatal density, while ‘Shahsavar Seydan Marvdasht’ had the lowest stom­ atal density. Results of Pearson correlation analysis provide significant negative correlations between stomatal density and tree height (r = ­0.44; P =0.0005) (Fig. 5). Fig. 3 ­ Pearson’s correlation coefficients between internode length and tree height in 19 Iranian pomegranate culti­ vars. Fruit quality traits The highest fruit weight was perceived in ‘Shahsavar Seydan Marvdasht’ (378.17 g), followed by ‘Malas Yazdi’ (230.83 g) and ‘Jangali Poost Ghermez Roodbar’ (214.67 g), while the smallest fruit (62.17 g) was observed in ‘Torosh Nar Riz Zirab’ (a semi­wild cultivar) (Table 3). The percentage of the edible portion of the fruit ranged from 49.80 (in ‘Torosh Nar Riz Zirab’) to 71.98% (in ‘Biardal Borujen’) (Table 3). The highest (18.7°Brix) and low­ est (13.0°Brix) TSS was measured in ‘Sefid Biardal Borujen’ and ‘Shahsavar Seydan Marvdasht’, respec­ tively (Table 3). Fruit juice pH varied from 3.13 to 4.43 among the studied pomegranate cultivars, with the minimum and maximum pH measured respec­ tively in ‘Torosh Nar RizZirab’, and ‘Anar Siah’. Moreover, the highest and lowest TA were observed in ‘Torosh Nar Riz Zirab’ (8.47 g CAE 100 mL­1) and ‘Bihaste Sangan Khash’ (0.50 g CAE 100 mL­1), respec­ tively. 4. Discussion and Conclusions The results of this research present a wide range of vigor and dwarfing potential in Iranian pomegran­ ates. In high­density orchards, controlling tree vigor and canopy size is important for enhancing the orchard efficiency and productivity (Umar and Sharma, 2008). Vegetative growth can be defined by several parameters such as; total shoot length, internode length, number of terminals and lateral shoot, and trunk cross­sectional area (Weibel et al., 2003). In classifying trees according to size, dwarf trees are approximately 2.5 m or less in height when mature (Castle, 1992). Considering the above men­Fig. 4 ­ Frequency distribution of stomatal density in 19 Iranian pomegranate cultivars. Bars indicate SE (n=45). Fig. 5 ­ Pearson’s correlation coefficients between stomatal den­ sity and tree height in 19 Iranian pomegranate cultivars. ous rootstock. The results of the current study also showed a wide variation in sucker production among cultivars. In this regard, cultivars such as ‘Rabab Poost Ghermez Kazeroon’ and ‘Khajei Ghasrodasht Fars’ with the lowest number of suckers have the advantage of easy management and also may be suitable for preferred single trunk training system in modern fruit orchards. Stomata are directly responsible for the trade­off between water loss and carbon acquisition (Raven, 2002). Stomatal density as a quantitative attribute is genetically determined (Gailing et al., 2008). Some plant species have been reported as possessing gen­ erally high heritability (i.e., less dependence on envi­ ronmental conditions) in their stomatal traits (Sharma and Dunn, 1969; Orlovic et al., 1998). Drogoudi et al. (2012) reported stomatal density among four pomegranate cultivars ranging from 68 to 149.9 stomata mm­2. Also, Meena et al. (2011) reported astomatal density of 130.67 stomata mm­2 for pomegranate. These results are following the findings of the current study, and minor differences in the results could be due to cultivar or climate dif­ ferences. Interestingly, some of the vigorous cultivars such as ‘Bihaste Ravar’, ‘Shahsavar Seydan Marvdasht’ and ‘Bihaste Sangan Khash’ had low 446 Adv. Hort. Sci., 2020 34(4): 441­448 tioned vegetative attributes and keeping in mind the Castle (1992) scale, ‘Malas No. 1 Saravan’ and ‘Torosh Nar Riz Zirab’ were the most dwarf size, about half that of vigorous cultivars. Thus, this culti­ var has potential to be used directly as dwarfing pomegranate rootstocks, although a more detailed study on propagation, graft compatibility, and toler­ ance to biotic and abiotic stress will shed more light on the suitability of these cultivars as rootstock. As a dwarfing source, these cultivars could be utilized as parents in breeding programs to develop superior dwarf scion and rootstock cultivars. This study demonstrated that internode length was associated with tree size. In general, some culti­ vars such as ‘Makhmal Malas Shahreza’, ‘Poost Sefid Dezfoul’ and ‘Malas No. 1 Saravan’ have the smallest tree sizes and shorter internodes lengths than other cultivars. The average internode length depended on the number of nodes per extension unit (Costes and Garcia­Villanueva, 2007). Dwarf trees usually pro­ duce very short internodes length, resulting in branches more compact than vigorous trees (Ingels et al., 2002). Obtained results for internode length are in agreement with those of Murase et al. (1990), on peach trees grafted on dwarfing rootstocks that had shorter internodes than trees grafted on vigor­ Table 3 ­ Fruit characteristics of the pomegranate cultivars used in the experiment Similar letters in each column indicate non­significant differences among cultivars at P≤0.05. Cultivars Fruit weight (g) Edible portion (%) TSS (°Brix) pH TA (%) Anar Siah 124.0 ef 53.31 cd 13.8 hi 4.43 a 0.70 hi Bihaste Ravar 140.32 e 54.30 cd 14.0 fghi 4.06 cd 0.55 i Bihaste Sangan Khash 156.0 de 60.39 abcd 13.83 ghi 4.38 ab 0.50 i Jangali Poost Ghermez Roodbar 214.67 b 61.88 abcd 18.33 ab 3.49 hij 2.17 efg Khajei Ghasrodasht Fars 207.50 bc 59.53 abcd 14.66 efghi 3.65 gh 2.51 def Malas Pishva Varamin 131.0 ef 67.81 ab 15.5 defgh 3.89 def 1.41 ghi MalasYazdi 230.83 b 49.80 d 16.0 cdef 3.70 fgh 1.85 fg Makhmal Malas Shahreza 133.50 ef 60.32 ab 17.66 abc 4.2 bc 0.6 i Malas No. 1 Saravan 125.83 ef 61.52 abcd 15.33 defgh 3.34 ijk 3.30 cd Poost Nazok Torosh Abarkuh 129.50 ef 62.66 abcd 16.0 cdef 3.49 hij 2.8 de Poost Sefid Dezfoul 88.0 fg 57.94 bcd 16.66 bcde 4.16 bc 0.88 hi Rabab Poost Ghermez Neyriz 190.0 bcd 56.61 bcd 14.83 efghi 3.53 hi 2.57 def Rabab Poost Ghermez Kazeroon 161.67 cde 54.16 cd 15.8 cdefg 3.84 efg 1.55 k Sefid Biardal Borujen 159.33 de 71.98 a 18.66 a 3.28 jk 3.97 c Shirin Jangal Sisangan 192.67 bcd 55.10 bcd 15.5 defgh 4.04 cde 1.35 ghi Shirin Semnan 158.0 de 62.01 abcd 14.0 fghi 4.16 bc 0.70 hi Shahsavar Seydan Marvdasht 378.17 a 56.52 bcd 13 i 4.16 sm 0.61 i Torosh Goli Naz Behshahr 126.83 ef 64.67 abc 17.16 abcd 3.14 k 5.47 b Torosh Nar Riz Zirab 62.17 g 60.77 abcd 16.66 bcde 3.13 k 8.47 a Ghasemi Soloklui et al. ‐ Potential dwarfing within Iran’s pomegranate germplasm 447 stomatal density (between 46 to 52 stomata mm­2), whereas dwarf and semi­dwarf cultivars, including ‘Shirin Semnan’, ‘Torosh Nar Riz Zirab’ and ‘Poost Sefid Dezfoul’ possess very high stomatal density (108, 96 and 90 stomata mm­2, respectively). These results are in line with the findings of Barrientos­ Pérez and Sanchez­Colín (1982), who reported that stomatal density could be a good method to classify the growth habit in avocado trees (Barrientos­Pérez and Sánchez­Colín, 1982). Thus, the data on fruit attributes would provide useful information for selecting the best dwarf culti­ vars to be used directly as scion cultivars on their root or as parent materials in scion cultivars breeding programs. The evaluation of pomegranate fruit quali­ ty (physical and chemical) in the local material has previously been carried out in Iran (Akbarpour et al., 2009), Turkey (Özkan, 2001), Italy (Barone et al., 2001), and Greece (Drogoudi et al., 2005). Tehranifar et al. (2010) described important fruit traits of 20 pomegranate cultivars from different regions in Iran. They found that the fruit weight, peel percentage, aril percentage, and juice percentage were within the range of 196.89­315.28 g, 32.28­59.82%, 37.59­65% and 26.95­46.55%, respectively, which are in line with the results of the current study. Moreover, Yıldız et al. (2003) reported that promising pomegranate genotypes, selected from Hizan (Bitlis) in Turkey, had 192.3­388.3 g fruit weight, 28­55% juice percentage, 0.33­4.03% juice acidity and 10.0­17.0% juice soluble solids content. On the other hand, Mars and Marrakchi (1999) defined fruit characteristics of 30 pomegranate genotypes from Tunisia. They reported fruit weights ranging from 196.1 to 673.6 g, pH from 2.9 to 4.6, soluble solid contents from 13.3 to 16.9°Brix, and acidity from 0.2 to 3.1 g CAE 100 mL­1. Consequently, the pomegranate studied herein had many similarities to those described in other studies concerning fruit traits such as fruit weight, soluble solids content, pH and acidity. Minor differences in these traits across the studies could arise from differ­ ent plant materials and varied climatic conditions. In this study edible portion was between 49.80 to 71.98%; whereas, Al­Maiman and Ahmad (2002) reported an edible portion of about 55­60% of the total fruit weight. This study showed that most culti­ vars except ‘Torosh Nar Riz Zirab’ and ‘Poost Sefid Dezfoul’ have big and medium sized fruits. In general, considering vegetative characteristics and fruit quality attributes, ‘Rabab Poost Ghermez Neyriz’ a commercial cultivar with semi­dwarfing growth habit and good fruit quality (have big fruits, with high TSS contents and low acidity) is a promising candidate for establishing high­density orchards on its roots. Moreover, some cultivars such as ‘Shirin Semnan’, ‘Makhmal Malas Shahreza’ and ‘Malas No. 1 Saravan’, which categorized as dwarfing or semi­ dwarfing cultivars and possessed quite good fruit quality, have the potential to be used as a parent in breeding programs to develop dwarf pomegranate cultivars or dwarfing rootstocks. ‘Torosh Nar Riz Zirab’ is a semi­wild cultivar with small tree size (dwarf cultivar) but represents poor fruit quality attributes. Thus, this cultivar can be considered as a dwarfing rootstock in pomegranate production. However, a more detailed study on propagation, graft compatibility, and tolerance to biotic and abiot­ ic stress will shed more light on these cultivars’ potentials as rootstock. References AKBARPOUR V., HEMMATI K., SHARIFANI M., 2009 ­ Physical and chemical properties of pomegranate (Punica granatum L.) fruit in maturation stage. ­ Am. Eurasian. J. Agric. Environ. Sci., 6(4): 411­416. AL­MAIMAN S.A., AHMAD D., 2002 ­ Changes in physical and chemical properties during pomegranate (Punica granatum L.) fruit maturation. ­ Food. Chem., 76(4): 437­441. BARONE E., CARUSO T., MARRA F., SOTTILE F., 2001 ­ Preliminary observations on some sicilian pomegranate (Punica granatum L.) varieties. ‐ J. Am. Pomol. Soc., 55(1): 4­7. BARRIENTOS­PÉREZ F., SÁNCHEZ­COLÍN S., 1982 ­ Height variability obtained from a new dwarf avocado tree population. ­ Fruit. Breeding., XXI IHC, 140: 163­168. BEHZADI SHAHRBABAKI H., 1998 ­ Genetic diversity of pomegranate genotypes in iran: Nashr Amoozesh Keshavarzi. BUSOV V.B., MEILAN R., PEARCE D.W., MA C., ROOD S.B., STRAUSS S.H., 2003 ­ Activation tagging of a dominant gibberellin catabolism gene (ga 2‐oxidase) from poplar that regulates tree stature. ­ Plant. Physiol., 132(3): 1283­1291. CASTLE W S., 1992 ­ Tree size control and dwarfing root‐ stocks. ­ Fact Sheet HS­146. COSTES E., GARCIA­VILLANUEVA E., 2007 ­ Clarifying the effects of dwarfing rootstock on vegetative and repro‐ ductive growth during tree development: A study on apple trees. ­ Ann. Bot., 100(2): 347­357. DEJONG T., JOHNSON R., DOYLE J., RAMMING D., 2005 ­ Labor costs may be reduced research yields size‐con‐ euramericana dode (guinier)) and eastern cottonwood (Populus deltoides bartr.) clones. ­ Silvae. Gene., 47(4): 183­189. ÖZKAN Y., 2001 ­ Determination of pomological character‐ istics of Niksar district pomegranates (Punica granatum L.) of the Tokat province. ­ International Symposium on Sustainable Use of Plant Biodiversity to Promote New Opportunities for Horticultural Production, 598 RAVEN J.A., 2002 ­ Selection pressures on stomatal evolu‐ tion. ­ New Phytol., 153(3): 371­386. SAS, 2003 ­ SAS. Statistical analysis system. Sas release 9.1. ­ SAS Institute, Cary, NC, USA. SHARMA G.K., DUNN D.B., 1969 ­ Environmental modifica‐ tions of leaf surface traits in datura stramonium. ­ Can. J. Bot., 47(8): 1211­1216. SOLEIMANI A., LESSANI H., TALAIE A., 2002 ­ Relationship between stomatal density and ionic leakage as indica‐ tors of cold hardiness in olive (Olea europaea L.).­ Acta Horticulturae, 618: 521­525. TEHRANIFAR A., ZAREI M., NEMATI Z., ESFANDIYARI B., VAZIFESHENAS M.R., 2010 ­ Investigation of physico‐ chemical properties and antioxidant activity of twenty iranian pomegranate (Punica granatum L.) cultivars. ­ Sci. Hortic., 126(2): 180­185. TERAKAMI S., MATSUTA N., YAMAMOTO T., SUGAYA S., GEMMA H., SOEJIMA J., 2007 ­ Agrobacterium‐mediat‐ ed transformation of the dwarf pomegranate (Punica granatum L. var. Nana). ­ Plant. Cell. Rep., 26(8): 1243­ 1251. TUKEY H.B., 1964 ­ Tree structure, physiology and dwarf‐ ing: Dwarf fruit trees. ­ Cornell University Press, UK, pp. 562. UMAR I., SHARMA A., 2008 ­ Control of height through growth retardants in fruit trees. ­ Asia. J. Hort., 3: 473­ 478. VERMA N., MOHANTY A., LAL A., 2010 ­ Pomegranate genetic resources and germplasm conservation: A review. ­ Fruit.Veg. Cereal. Sci. Biotech., 4: 120­125. WEIBEL A., JOHNSON R.S., DEJONG T.M., 2003 ­ Comparative vegetative growth responses of two peach cultivars grown on size‐controlling versus standard rootstocks. ­ J. Am. Soc. Hortic. Sci., 128: 463­471. YILDIZ K.F., MURADOĞLU H.İ., OGUZ H. YILMAZ, 2003 ­ Pomological characteristics of pomegranate varieties rown in Hizan town of Bitlis. ­ Congress, Antalya, Turkey, pp. 238­240. ZAMANI Z., SARKHOSH A., FATAHI R., EBADI A., 2007 ­ Genetic relationships among pomegranate genotypes studied by fruit characteristics and rapd markers. ­ J. Hort. Sci. Biotech., 82:11­18. Adv. Hort. Sci., 2020 34(4): 441­448 448 trolling rootstocks for peach production. ­ Calif. Agric., 59(2): 80­83. DROGOUDI P., PANTELIDIS G., MANGANARIS A., 2012 ­ Morphological and physiological characteristics in pomegranate cultivars with different yields. ­ Cah. Options. Mediterr., 103: 67­69. DROGOUDI P.D., TSIPOURIDIS C., MICHAILIDIS Z., 2005 ­ Physical and chemical characteristics of pomegranates. ­ HortSci., 40(5): 1200­1203. GAILING O., LANGENFELD­HEYSER R., POLLE A., FINKELDEY R., 2008 ­ Quantitative trait loci affecting stomatal den‐ sity and growth in a quercus robur progeny: Implications for the adaptation to changing environ‐ ments. ­ Global. Change. Biol., 14(8): 1934­1946. GHASEMI SOLOKLUI A.A., GHARAGHANI A., ORAGUZIE N., RAMEZANIAN A., 2019 ­ Shelf life and changes in bio‐ chemical composition of ready‐to‐eat arils of nineteen Iranian pomegranate (Punica granatum L.) cultivars during storage. ­ Food Sci. Technol., 56(3): 1416­1426. HARLAN J.R., 1975 ­ Crops and man. ­ American Society of Agronomy, Madison, WI, USA, pp. 284. HORWITZ W., 1980 ­ Official methods of analysis ­ AOAC, Arlington, VA, Washington DC, Vol. 534, pp. 1038. INGELS C., GEISEL P.M., UNRUH C.L., 2002 ­ Fruit trees: Training and pruning deciduous trees ­ University of California, ANR Publications, 8057, pp. 8. LEVIN G.M., 1994 ­ Pomegranate (Punica granatum) plant genetic resources in Turkmenistan. ­ Plant Genetic Resources Newsletter, IPGRI, 106: 47­49. LOONEY N., LANE W., 1983 ­ Spur‐type growth mutants of Mcintosh apple: A review of their genetics, physiology and field performance. ­ Acta Horticulturae, 146: 31­46. MARS M., MARRAKCHI M., 1999 ­ Diversity of pomegran‐ ate (Punica granatum L.) germplasm in Tunisia. ­Genet. Resour. ­ Crop. Evol., 46(5): 461­467. MEENA K., SINGH R., PAREEK S., KASHYAP P., SHEIKH M., MOKASHI A., ROKHADE A., 2011 ­ Evaluation of pome‐ granate (Punica granatum L.) genotypes for morpho‐ logical and flowering characteristics under semi‐arid climate. ­ Acta Horticulturae, 890: 233­237. MOURA C.F.H., ALVES R.E., INNECCO R., FILGUEIRAS H.A.C., MOSCA J.L., PINTO S.A.A., 2001 ­ Physical char‐ acteritics of cashew apples for fresh fruit market. ­ Revista Brasileira de Fruticultura, 23: 537­540. MURASE S., YAMAZAKI T., INOMATA Y., SUZUKI K., 1990 ­ Dwarfing rootstock for peach. ­ Jpn. Agric. Res. Q., 23: 294­300. ORLOVIC S., GUZINA V., KRSTIC B., MERKULOV L., 1998 ­ Genetic variability in anatomical, physiological and growth characteristics of hybrid poplar (Populus x