Impaginato 245 Adv. Hort. Sci., 2019 33(2): 245-255 DOI: 10.13128/ahs-23991 Selection of open pollination progenies in some pear species in order to achieve dwarf and drought tolerant rootstocks M. Tatari 1 (*) H. Abdollahi 2, M. Henareh 3, M. Dehqani 4 1 Horticulture Crops Research Department, Isfahan Agricultural and Natural Resources Research and Education Centre, AREEO, Isfahan, Iran. 2 Temperate Fruits Research Center, Horticultural Scinces Research Institue, AREEO, Karaj, Iran. 3 Horticulture Crops Research Department, West Azarbaijan Agricultural and Natural Resources Research and Education Center, AREEO, Urmia, Iran 2 Soil and Water Research Department, Isfahan Agricultural and Natural Resources Research and Education Center, AREEO, Isfahan, Iran. Key words: drought stress, genotypes, growth vigor, Pyrus spp., seedlings. Abstract: One of the important products in Iran and Isfahan province is the pear that its cultivation has been limited by drought stress and global warming in recent years. The use of drought-tolerant rootstocks is one of the available solu- tions for pear orchards in semi-arid regions. In addition, the lack of dwarf or semi-dwarf rootstocks, which are appropriate and compatible with Iran climatic conditions, limited high density pear orchards. In order to obtain drought toler- ant rootstocks, in this research fruit of Pyrus glabra, Pyrus syriaca, and Pyrus salicifolia, along with P. communis cv. Spadona, Dargazi, as well as Khoj n. 1 and n. 2 species were collected from different regions of Iran in August and September of 2016. The seeds were separated from the flesh and dried at room temperature. The seeds were cultivated in uniform and light texture of soil in November in the field condition to break the seed dormancy. Seedlings were irrigated regularly for three months in order to establish in the soil; drought stress began in July. In order to apply drought stress, irrigation time was consid- ered based on 80% of allowed water depletion. Morphological traits of seedlings were recorded before stress and at the end of the stress period (late September). The viability percentage of seedlings after drought stress was between 14.28% (P. communis cv. Dargazi progenies) to 82.55% for P. salicifolia. Comparison of the means and cluster analysis, among populations showed that the three populations of P. salicifola, P. glabra and P. communis cv. Khoj n. 2 had the lowest height and were placed in the same group. After studying single genotypes in these three populations, genotypes no. 31, 32, 41, 57 and 12 from P. salicifolia, genotypes 10, 11, 7, 3 and 9 of P. glabra and genotype 4 of P. com- munis cv. Khoj n. 2 populations were selected as drought tolerant and dwarf genotypes and were taken to the propagation phase for future evaluation. 1. Introduction Pear is from the Rosaceae family and the Pyrus genus. This species has been cultivated in Iran since ancient times, and Iran is one of the earliest (*) Corresponding author: mtatari1@gmail.com Citation: TATARI M., 2019 - Selection of open pollination progenies in some pear species in order to achie- ve dwarf and drought tolerant rootstocks. - Adv. Hort. Sci., 33(2): 245-255 Copyright: © 2019 Tatari M. 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 September 2018 Accepted for publication 25 February 2019 AHS Advances in Horticultural Science http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ Adv. Hort. Sci., 2019 33(2): 245-255 246 areas of pear distribution and variation in the world (Abdollahi, 2011). Drought stress is the most impor- tant environmental stress that occurs annually with extreme damage to crops, especially in arid and semi-arid regions (Xoconostle-Cazares et al., 2010). In dry and semi-arid regions of Iran, the amount of annual evaporation is higher than precipitation. In recent years, the cultivation of fruit trees, such as pears, has been limited due to climate change and reduce rainfall. Considering that the drought toler- ance of the rootstocks is transmitted to the scion (Landsberg and Jones, 1981), the choice of drought tolerant rootstocks with a low water requirement is one of the solutions to drought problem of pear orchards in arid and semi-arid regions such as Iran (Cheruth et al., 2009). Another problem in the cultivation of pears in Iran is the lack of suitable dwarf or semi-dwarf rootstocks that are compatible with cultivars and climatic condi- tions of the country for high density orchards. The pear and quince are used as rootstocks for pear trees. The quince rootstocks are very dwarf (EM-C) and semi-dwarf (BA29), but these rootstocks could not overcome the problem of graft incompatibility with some pear cultivars. These rootstocks are not also tolerant to winter frosts and chlorosis caused by iron deficiency in the calcareous soils (Harotko, 2007). In recent years, a number of clonal rootstocks have been introduced into Iran such as Pyrodwarf, Fox-11 and some of the American rootstocks of the Old home × Farmingdale series (Abdollahi, 2011). The use of native pear species as a seed or clonal root- stocks can be a solution for access to proper root- stocks with good adaptability to the climatic condi- tions of the country. Some of these species are dwarf and have a high tolerance to unfavorable environ- mental conditions that can be used to produce clonal rootstocks (Abdollahi et al., 2012). In the pear genus, 22 species have been identified in the world that are mainly native to Europe, Africa and Asia. Of the 22 identified species, there are 12 species in Iran (Sabeti, 1995). In some parts of the world, these species are used as the rootstock for pear commercial cultivars. For example, a large num- ber of pear cultivars in Turkey are grafted on P. elaeagrifolia, in Syria and Lebanon on P. syriaca, in ancient Yugoslavia, Turkey and Greece on P. amyg- daliformis, in the south of Russia on P. salicifolia and in Algeria and Morocco are grafted on P. longipes (Henareh, 2015). W i l d g e r m p l a s m h a s e v o l v e d i n n a t u r a l d r y ecosystems for tolerance of stress conditions such as h i g h t e m p e r a t u r e s , d r o u g h t a n d s a l i n i t y . Identification of wild pear germplasm is very impor- tant to use as a rootstock in semi-arid regions (Zarafshar et al., 2014). According to this, three wild pear genotypes from P. syriaca were exposed to four irrigation treatments. The results showed that the Coile wild genotype (P. syriaca) was more tolerant to drought conditions compared with other genotypes due to its high relative water content during drought stress and non-decreasing dry weight. Drought stress reduced leaf photosynthesis, stomatal conductance and transpiration in a number of pear species (Javadi and Bahramnejad, 2011). In another study, drought tolerance was evaluat- ed in three populations of wild pear germplasm (P. boisseriana) in greenhouse conditions. Among the three populations studied, the collected population from semi-arid regions showed higher drought toler- ance than the other two populations that were col- lected from semi-humid areas, and recovered more rapidly after irrigation. These seedlings were intro- duced as promising sources for use as rootstock for commercial pear cultivars in drought conditions (Zarafshar et al., 2014). In a research carried out by Ghasemi et al. (2014), chlorophyll index was signifi- cantly different between studied pistachio root- stocks. In these rootstocks, chlorophyll index in stress conditions was lower than drought conditions. So, the reduction of chlorophyll content can be caused by chlorophyll degradation under drought stress con- ditions, which leads to reduced pure photosynthesis. The rootstocks of European pear species are from very vigor to relatively dwarf. These rootstocks are compatible with all the pear tree cultivars and have considerable tolerance to the adverse conditions of soil and fire blight disease. Other species of Pyrus genus, such as P. syriaca Boiss is compatible with commercial pear cultivars and is currently used in some countries (Radnia, 1996). Seed cultivation after cross or open pollination is one of the breeding methods that is frequently used in breeding programs. In this way, it is possible to achieve a wide variation for choice of new rootstocks and cultivars. For example, the Manon cultivar was obtained from open pollination of Beurre Bosc culti- var. The Pib-BU3 dwarf pear rootstock was obtained from open pollination of P. longipes, while Pi-BU4 and Pi-BU7 rootstocks were obtained from open pol- lination of the P. pyrifolia species (Mohan Jain and Priyadarshan, 2009). Currently, most pear cultivars in Iran are grafted on different seedling rootstocks of the P. communis Tatari - Selection of drought tolerant pear rootstocks 247 species. Considering that wild pear genotypes grow on rocks and dry or low moisture soils, as well as some of them have a small growth in the form of shrubs, so some of them can be used as drought tol- erant and dwarf rootstocks for commercial pear culti- vars (Ashraf and Karimi, 1991; Henareh, 2015). The aim of this research was an evaluation of drought stress tolerance in some wild pear species, and then selection of the dwarf genotypes among the drought tolerant genotypes. 2. Materials and Methods Plant materials In this research, seeds from open pollination of Pyrus syriaca Boiss., Pyrus salicifolia Pall. and Pyrus. glabra Boiss. along with P. communis L. cv. Spadona and Dargazi as well Khoj including large fruit (Khoj no. 1) and small fruit (Khoj no. 2) were evaluated. P. syriaca Boiss. is distributed from west Azarbaijan to Fars in the Zagros Mountains and northwest of Iran (Abdollahi, 2011). This species has shown compatibil- ity with Spadona and Kochia pear cultivars and have improved the growth of the scion in calcareous soils. It is currently used as a rootstock in some countries (Fallouh et al., 2008). P. glabra species is known as Anchuchek in Iran, and has spread mainly in the Zagros Mountains. The seeds of this species are large a n d a r e c o n s u m e d a s s n a c k s i n F a r s p r o v i n c e (Abdollahi, 2011). P. communis is more commonly known as Khoj and is scattered in the forests of the north, West Azarbaijan, Sardasht and Baneh in Kordestan province. The fruit of this species is very diverse. Two types of native Khoj including large fruit (Khoj n. 1) and small fruit (Khoj n. 2) were studied in this research (Moazedi et al., 2014). P. communis cv. Dargazi is native commercial cultivar of Iran that has been introduced as tolerant rootstock (Mansouryar et al., 2017). P. communis cv. Spadona is high yielding and has high resistance to chlorosis due to iron defi- ciency. It is also relatively tolerant to psylla and fire blight. P. salicifolia species spread in the northwest of Iran, including west and east Azerbaijan provinces (Abdollahi, 2011). Cultivation of seeds Fruits of studied species were collected in August and September of 2016 from different regions of West Azarbaijan and Isfahan provinces as well as northern regions of Iran and transferred to the labo- ratory. The seeds were separated from the flesh and kept in a cool and dry place in paper bags after wash- ing and drying, In order to eliminate the chilling requirement, seeds were cultivated in separate rows in the nursery soil with a sandy loam texture at Isfahan Agricultural and Natural Resources Research Center, in December. From each of these species, 500 seeds were cultivated at intervals of 2-3 cm in the nursery and each produced seedlings were stud- ied as a genotype. During chilling period, adequate moisture was provided and prevented from drying of the culture bed. Irrigation of seedlings After emergence of seeds, the seedlings were irri- gated for three months in order to establish in the nursery soil. Determination of irrigation time based on allowed water depletion of pear trees is after a 5 0 % d e c r e a s e i n h u m i d i t y , b u t b e c a u s e o f t h e seedling roots had not yet been sufficiently devel- oped, and needed enough moisture for better growth, so allowed water depletion was considered 35%. In the irrigation intervals, soil moisture at vari- ous depths of the soil was measured up to a depth of 100 cm by the Time Domain Reflectometry device (TDR, Trase6050X1) (Doorenbos and Pruitt, 1977; Alizadeh, 2006). To obtain the amount of irrigation (irrigation vol- ume), first, net irrigation depth was calculated according to formula 1. In= ∑ [(θFCi - θBLi)×Di] (1) In this formula, In is the net irrigation depth (mm), θFCi is the moisture content of the field capacity for each layer, θBLi is the soil moisture before irrigation for each layer, Di is the root development depth (mm), and i is the number of each soil layer. Then, according to formula 2, gross irrigation depth was calculated. Ig = In/(1-Lr)×Ei (2) In this formula, Ig is the gross irrigation depth (mm), In is the net irrigation depth (mm), (1-Lr) is the amount of leaching and Ei is the irrigation efficiency (usually 80-90% for drip irrigation). Irrigation volume was obtained from the multipli- cation of gross irrigation depth in the irrigation plot area. This volume was controlled by the installed counter on the pipe before irrigation plot. For irriga- tion of seedlings leaked tubes with drippler at 10 cm intervals were used. The outlet flow of each drippler was measured in one liter per hour at an appropriate pressure. In this study, the efficiency of the drip sys- tem and the leaching requirement was considered 90% and 10% respectively. Adv. Hort. Sci., 2019 33(2): 245-255 248 Drought stress Drought stress began in July, which coincided with the beginning of the stress period in Isfahan region. Irrigation frequency was changed in order to apply drought stress, and irrigation time was considered based on 80% of allowed water depletion. Table 1 shows the volume of pure and impure irrigation water and number of irrigation per month. Evaluated traits The morphological traits of seedlings were record- ed separately for each genotype at the end of the stress period (late September). These traits included height and diameter of seedlings, number and length of internode, crown width, number of suckers and branches, chlorophyll index and leaf dimensions. Before applying stress, seedling height and diameter at 5 cm above the soil surface were recorded. The difference in both seedling height and diameter b e f o r e a n d a f t e r s t r e s s w e r e a l s o c a l c u l a t e d . Qualitative traits, including leaf chlorosis and tri- chome, as well as seedling growth vigor were record- ed after stress using national guideline for distinct- ness, uniformity and stability in pear (DUS guileline) in pears. The abbreviation and measurement method of evaluated traits are given in Table 2. Data analysis Mean comparison was calculated with SAS soft- ware (version 9.1). Descriptive statistics including mean, minimum, maximum and coefficient of varia- tion and also cluster analysis were performed by Ward method based on Squared Euclidean Distance with SPSS software (version 15). 3. Results Viability percent Viability percent of seedlings of the studied pear species after the stress was shown in Table 3. According to the results, P. salicifolia showed the highest survival in drought stress conditions. After that, P. communis cv. Khoj n. 1 and 2 were placed in the next rank. The lowest percentage of seedling via- Table 2 - Symbol and measurement method for recorded traits (based on DUS guideline) Table 1 - Number of irrigation, net and gross volume of irrigation water Irrigation February March April May June July August September October November Total Number of irrigation 3 3 5 7 8 1 1 1 1 1 32 Net volume of irrigation water (m3.ha-1) 540 540 900 1260 1440 180 180 180 180 180 5580 Gross volume of irrigation water (m3.ha-1) 666.6 666.6 1111.1 1555.5 1777.7 222.2 222.2 222.2 222.2 222.2 6888.2 Characteristic Symbol Unit Measurement method Leaf chlorosis LC Code No chlorosis (1), low chlorosis (3), medium chlorosis (5), high chlorosis (7), very high chlorosis (9) Leaf trichome LT Code No trichome (1), low trichome (3), medium trichome (5), high trichome (7), very high trichome (9) Seedling growth vigor GV Code Very low (1), low (3), medium (5), high (7), very high (9) Seedling height SE cm Ruler Seedling diameter SD mm Caliper Height difference HD cm Calculation Diameter difference DD cm Calculation Internode number IN Number Counting Internode length IL cm Ruler (average of internodes in branches) Crown width CW cm Meter Number of secondary branches NSB Number Counting Number of suckers NS Number Counting Chlorophyll index CI - Chlorophyll meter (Spad) Leaf length LL cm Ruler (average of 10 leaves) Leaf width LW cm Ruler (average of 10 leaves) Table 3 - Seedlings viability percent of pear cultivars and spe- cies after drought stress Species Viability percent Pyrus communis cv. Spadona 14.77 Pyrus communis cv. Dargazi 14.28 Pyrus communis cv. Khoj n. 1 55.17 Pyrus communis cv. Khoj n. 2 43.79 Pyrus syriaca 20.33 Pyrus salicifolia 82.55 Pyrus glabra 17.74 Tatari - Selection of drought tolerant pear rootstocks 249 were lower than other species. Similarly, P. commu- nis cv. Spadona showed the highest height difference before and after applying stress. T h e s e e d l i n g d i a m e t e r o f P . c o m m u n i s c v . Spadona, Khoj n. 1, Dargazi and P. syriaca was more than the other seedling diameter, So that P. glabra and P. communis cv. Khoj n. 2 had the lowest seedling diameter. No significant difference was observed among the seedling diameter of the species before and after drought stress. P. communis cv. Spadona had the highest number of internode after the end of drought stress. The low- est number of internode belonged to P. glabra and P. communis cv. Khoj n. 2. P. salicifolia produced the crown with the widest (24.71 cm) width, compared to other species. It should be noted that P. communis cv. Spadona, Khoj n. 1 and P. syriaca did not show any significant difference with Pyrus salicifolia. P. communis cv. Khoj n. 2 had the lowest crown width (4.75 cm). P. salicifolia did not produce a secondary branch. There were no significant differences in the number of secondary branches among the other species. P. salicifolia showed the highest chlorophyll index, but did not show significant differences with P. communis cv. Spadona, Khoj n. 2, P. syriaca and P. glabra. The lowest chlorophyll index belonged to P. communis cv. Dargazi (Table 5). P. communis cv. Spadona , with a length of 6.06 cm and a width of 2.7 cm had the highest leaf length and width compared with other species. After P. communis cv. Spadona, P. communis cv. Khoj n. 1 had the highest leaf length and width. The lowest leaf length was related to P. communis cv. Khoj n. 2 with an average of 2.01 cm. P. syriaca also produced the leaves with the lowest width (1.37 cm). P. communis cv. Spadona and Dargazi had the longest internode, but did not show significant differ- ences with P. communis cv. Khoj n. 1 and P. syriaca. bility during drought stress belonged to P. communis cv. Spadona and Dargazi. Traits before applying stress The mean comparison of recorded traits among populations before applying drought stress is pre- sented in Table 4. According to the results, P. com- munis cv. Spadona had the highest seedling height. The seedling height in the wild species of P. glabra, P. salicifolia and P. communis cv. Khoj n. 2 was the low- est. P. communis cv. Khoj n. 2 and P. glabra produced seedlings with the lowest diameter. The largest seedling diameter belonged to P. communis cv. Spadona and P. syriaca (Table 4). Generally, without drought stress conditions, growth of P. glabra, P. sali- cifolia and P. communis cv. Khoj n. 2 were lower than P. communis cv. Spadona, Khoj n. 1, Dargazi and P. syriaca. Traits after applying stress The mean comparison of recorded traits among populations after stress is presented in Table 5. P. communis cv. Spadona had the highest seedling height (36.73 cm) after stress. The seedling height of P. glabra, P. salicifolia and P. communis cv. Khoj n. 2 Table 4 - The mean comparison of the seedling height and dia- meter among populations of pear species before applying stress Similar letters in each column indicate no significant difference (LSD). Table 5 - Mean comparison of some traits in the seedling populations of pear species after applying stress Similar letters in each column indicate no significant difference (LSD). Species Seedling height (cm) Seedling diameter (mm) Pyrus communis cv. Spadona 33.52 a 3.44 a Pyrus communis cv. Khoj n. 1 23.38 b 3.15 ab Pyrus communis cv. Dargazi 24.23 b 3.15 ab Pyrus syriaca 20.92 b 3.31 a Pyrus salicifolia 12.80 c 2.62 bc Pyrus communis cv. Khoj n. 2 11.83 c 1.74 d Pyrus glabra 7.54 c 2.25 cd Species Seedling height (cm) Height difference (cm) Diameter (mm) Diameter difference (mm) Internode number Crown width (cm) Number of secondary branches Chlorophyll index Leaf length (cm) Leaf width (cm) Internode length (cm) Pyrus communis cv. Spadona 36.73 a 3.20 a 4.34 a 0.89 a 20.23 a 10.03 ab 0.23 ab 44.99 ab 6.06 a 2.70 a 1.91 a Pyrus communis cv. Khoj n. 1 26.65 b 2.80 ab 4.00 a 0.84 a 14.56 b 10.62 ab 0.68 a 42.48 b 5.45ab 2.23 b 1.21 ab Pyrus communis cv. Dargazi 25.83 b 1.60 ab 3.81 a 0.66 a 14.66 b 7.83 b 0.66 a 42.02 b 4.73 bc 2.10 bc 1.93 a Pyrus syriaca 21.60 b 1.41 ab 4.02 a 0.70 a 12.91 bc 9.16 ab 0.25 ab 49.77 ab 4.50 c 1.37 d 1.34 ab Pyrus salicifolia 14.21 c 0.67 b 3.56 ab 0.94 a 10.05 cd 11.24 a 0.00 b 53.83 a 4.30 c 1.73 c 0.94 b Pyrus communis cv. Khoj n. 2 13.67 c 0.84 b 2.63 c 0.88 a 8.31 de 4.75 c 0.21 ab 46.53 ab 2.01 d 1.97 bc 0.71 b Pyrus glabra 10.22 c 0.68 b 2.92 bc 0.66 a 5.27 e 8.31 b 0.36 ab 48.85 ab 3.98 c 2.10 bc 0.73 b 250 Adv. Hort. Sci., 2019 33(2): 245-255 The lowest internode length belonged to P. salicifo- lia, P. glabra, and P. communis cv. Khoj n. 2 (Table 5). Cluster analysis of populations According to the results of cluster analysis, species were classified into three groups at five s q u a r e d E u c l i d e a n d i s t a n c e b a s i s o f t h e W a r d method (Fig. 1). P. communis cv. Spadona and Dargazi were placed in the first group. P. syriaca and P. communis cv. Khoj n. 1 in the second group, while P. communis cv. Khoj n. 2, P. salicifolia and P. glabra species in the third group. Study of single genotypes The mean and the range of traits for each of the examined genotypes, as well as the coefficient of variation in each trait are presented in Table 6. Each genotype was identified by a number and genotypes with minimum and maximum values were presented. Internode length (66.12%), seedling height before stress (60.45%) and seedling height after stress (56.85%) had the highest and chlorophyll index (14.34%) had the lowest coefficient of variation. In order to achieve dwarfing genotypes, the trait of seedling height was considered. The three pear populations, including P. salicifolia, P. glabra and P. communis cv. Khoj n. 2 had a lower mean seedling height than other populations, so the height of single genotypes of them after stress was shown in figures 2, 3 and 4, respectively. In P. salicifolia, genotypes no. 31 (4.5 cm), 32 (5 cm), 41 (5.5 cm), 57 (6 cm) and 12 (6 cm) had the lowest seedling height. Genotypes Fig. 2 - Height of single genotypes in P. salicifolia after stress. Table 6 - Mean, Range and coefficient of variation in studied traits Fig. 1 - Grouping of pear species based on measured characteri- stics by Ward method. Characteristics CV (%) Standard deviation Mean Minimum Maximum Genotype (number of genotypes) Rate Genotype (number of genotypes) Rate Seedling height before stress 60.45 9.83 16.26 P. glabra (42) 1.4 P. communis cv. Spadona (23) 53.9 Seedling diameter before stress 28.35 0.76 2.68 Py. communis cv. Khoj n. 2 (33) 0.75 P. communis cv. Spadona (64) 6.3 Seedling height after stress 56.85 10.28 18.08 P. glabra (10 & 11) 3 P. communis cv. Spadona (2) 55 Seedling diameter after stress 26.47 0.94 3.55 P. communis cv. Khoj n. 2 (8) 1.6 P. salicifolia (8) 7.28 Height difference 55.32 2.19 1.82 P. salicifolia (24 & 28) 0 P. communis cv. Khoj n. 1 (9) & P. syriaca (3, 4,5,6 & 7) 19 Diameter difference 55.17 0.48 0.87 P. salicifolia (47) 0.04 P. salicifolia (66) 3.06 Internode number 44.65 5.00 11.20 P. glabra (7,10 & 11) 3 P. communis cv. Khoj n. 1 (14) 26 Crown width 30.67 3.11 10.14 P. communis cv. Khoj n. 1 (5) & P. glabra (10) 4 P. communis cv. Khoj n. 1 (8) 30 Number of secondary branches 37.22 0.67 0.18 Many genotypes 0 P. communis cv. Khoj n. 2 (4) 4 Chlorophyll index 14.34 7.81 54.46 P. communis cv. Khoj n. 1 (8) 23.2 P. salicifolia (36) 69.8 Leaf length 31.93 1.37 4.29 P. communis cv. Khoj n. 2 (15) 1 P. communis cv. Spadona (13) 7.2 Leaf width 26.70 0.51 1.91 P. syriaca (5) 1 P. communis cv. Spadona (2) 3.3 Internode length 66.12 0.82 1.24 P.salicifolia (30) 0.3 P. communis cv. Spadona (1) 10 Tatari - Selection of drought tolerant pear rootstocks 251 no. 10 (3 cm), 11 (3 cm), 3 (6 cm), 9 (6 cm) and 7 (6 cm) showed the lowest seedling height in the P. glabra species. Genotype no. 4 had the lowest height from P. communis cv. Khoj n. 2 population. Cluster analysis of single genotypes based on all stud- ied traits Cluster analysis among the single genotypes of the P. salicifolia, P. glabra and P. communis cv. Khoj n. 2 populations was shown in figures 5, 6 and 7. Genotypes no. 12, 31, 32, 41 and 57 of P. salicifolia (Fig. 5) and genotypes no. 10, 11, 3, 9 and 7 of P. glabra (Fig. 6) were placed in the groups close to each other. Qualitative traits for selected genotypes were pre- sented in Table 7. All genotypes had a small trichome and had low or very low growth potentials. Leaf chlorosis was not observed in genotypes except for the genotype no. 31 from P. salicifolia and genotype n o . 3 f r o m P . g l a b r a t h a t h a d l o w c h l o r o s i s . Genotypes no. 12 and 32 of P. salicifolia were green and very green, respectively. Genotypes no. 9, 7, 10 and 11 of P. glabra and genotype no. 4 of the P. com- munis cv. Khoj n. 2 preserved their green color after applying drought stress. 4. Discussion and Conclusions Viability percent After the drought stress period, 35.51% of the genotypes survived, and the rest of them were dried. Most surviving genotypes belonged to P. salicifolia species. P. communis cv. Spadona and Dargazi Fig. 3 - Height of single genotypes in P. glabra after stress. Fig. 4 - Height of single genotypes in P. communis cv. Khoj n. 2 after stress. Fig. 5 - Grouping of drought tolerant genotypes in P. salicifolia species based on measured traits by Ward method. Fig. 6 - Grouping of drought tolerant genotypes in P. glabra spe- cies based on traits measured by Ward method. Adv. Hort. Sci., 2019 33(2): 245-255 252 showed the lowest percentage of survival (Table 3). For a long time, wild pear genotypes have been con- sidered in Iran’s plateau due to the tolerance to biot- ic and abiotic stresses (Javadi et al., 2005). The adap- tation of wild pears with rocky areas and dry or low moisture soils can lead to more tolerance of them under drought stress conditions compared with domestic and commercial rootstocks (Henareh, 2015). Traits before and after applying stress The analysis of variance showed that the investi- gated species had a significant difference in most of the studied traits, which is due to the diversity among populations, so it is possible to select species for different values of a trait. According to table 4, in normal conditions, seedling height and diameter of P. glabra, P. salicifolia and P. communis cv. Khoj n. 2 were less than P. communis cv. Spadona, Khoj n. 1, Dargazi and P. syriaca. Similarly, after applying drought stress, the seedling height and diameter of P. glabra, P. salicifolia and P. communis cv. Khoj n. 2 species were lower than other populations (Table 5). Morphological adaptations in plants can be one of the adaptive mechanisms under drought stress (Pire et al., 2007). The first reaction of plants against drought stress is a reduction in their vegetative growth. Drought stress affects the vegetative charac- teristics of trees, including their height (Higgs and Jones, 1990). Due to height difference among popu- lations before and after stress, it seems that the effect of drought stress on the seedling height trend of these populations before and after applying the stress is almost same. Before drought stress, the seedling diameter of P. communis cv. Spadona, Khoj n. 1, Dargazi and P. syri- aca was higher than the seedling diameter of P. sali- cifolia, P. glabra and P. communis cv. Khoj n. 2, while the diameter difference before and after drought stress among compared species did not show a sig- nificant difference, therefore, it can be concluded that the effect of drought on the seedling diameter among populations was not the same. Other results also showed that the negative effect of drought stress on seedling diameter was less than its effect on seedling height (Haghighatnia et al., 2013). Growth of branch and internode length is an appropriate index for detecting the effect of drought stress on the plants, so that the occurrence of drought stress can be observed even before the change in the water potential of the leaves (Grimplet et al., 2007). Among the remaining genotypes after the stress, P. communis cv. Spadona had the highest and P. glabra and P. communis cv. Khoj n. 2 had the lowest number and length of internode. The chlorophyll meter indicates the relative Fig. 7 - Grouping of drought tolerant genotypes in P. communis cv. Khoj n. 2 based on the measured traits by the Ward method. Table 7 - Growth vigor, trichome and chlorosis of leaf in selected genotypes Species No. genotype Leaf chlorosis Leaf trichome Seedling growth vigor Pyrus salicifolia 12 No chlorosis Low trichome Low 31 Low chlorosis Low trichome Low 32 No chlorosis Low trichome Low 41 No chlorosis Low trichome Low 57 No chlorosis Low trichome Low Pyrus glabra 3 Low chlorosis Low trichome Low 9 No chlorosis Low trichome Low 7 No chlorosis Low trichome Low 10 No chlorosis Low trichome Very low 11 No chlorosis Low trichome Very low Pyrus communis cv. Khoj n. 2 4 No chlorosis Low trichome Low Tatari - Selection of drought tolerant pear rootstocks 253 chlorophyll concentration, based on the difference between the light transmittance in two red and infrared wavelengths, which correlates with the chlorophyll content of the leaves (Hoel and Solhaug, 1998). In the present study, P. salicifolia and P. com- munis cv. Dargazi showed the highest and the lowest chlorophyll index, respectively. Preservation and not decomposition of chlorophyll in P. salicifolia during drought stress indicates the tolerance of that species to this stress (Tarahomi et al., 2010). Cluster analysis of populations Drought tolerance in plants has a direct or indirect relationship with a complex of traits, therefore, all the traits should be considered for selection of toler- ant plants. For this reason, in this research cluster analysis was used to facilitate the selection of drought tolerant species. Cluster analysis classified species into three groups at five squared Euclidean distance (Fig. 1). P. communis cv. Spadona and Dargazi were placed in the first group that had the lowest survival rate after drought stress. The highest seedling diameter belonged to the plants of this group. In the second group, P. syriaca and P. commu- nis cv. Khoj n. 1 were placed, which showed low to moderate survival rate after drought stress. In total, the most crown width, primary and secondary seedling height and diameter, as well as the highest number of internode were observed in the species of the first and second groups. P. communis cv. Khoj n. 2, P. salicifolia and P. glabra species formed the third group. These populations had the lowest seedling diameter and height. The lowest seedling growth vigor was also found in these species; therefore, selection of dwarf and drought tolerant genotypes in this group was more possible than other groups. It seemed that traits such as seedling height and diam- eter were the most effective grouping traits. In the research carried out by Aran et al. (2012), seedling h e i g h t w a s a l s o o n e o f t h e i m p o r t a n t t r a i t s i n seedling grouping. Of course, different values of some traits were seen in this group. For example, the crown width and viability percentage were observed at low, medium and high levels in this group. Study of single genotypes The mean, the range of changes, and the coeffi- cient of variation of each trait were shown in Table 6. Coefficient of variation shows the extent of variability in relation to the mean of the population. In the traits with a high coefficient of variation has provided a higher selection range. Genetic variation helps the plant to overcome environmental changes and also provides more chance for selection of new cultivars (Liu, 2006). In this research, high variation was observed for internode length and seedling height before and after stress. The variability of some traits in 15 cultivars of Vitis vinifera L. was previously inves- tigated by Mousazadeh et al. (2014). They reported that leaf traits had the highest diversity among the studied traits. Doulati Baneh et al. (2013) and Tahzibihagh et al. (2012) also observed a high varia- tion in the morphological characteristics of the leaves of grape and pear cultivars. In the current research, the coefficient of variation in leaf dimensions was not high compared to other traits, because of less scatter in relation to the mean of the population. The high diversity coefficient for internode length and seedling height before and after stress indicates the high range of variation in these traits among the studied seedlings, so it is possible to select genotypes based on these two traits. Survived genotypes after drought stress were selected in order to dwarfing. For this purpose, seedling height of each genotype was investigated separately. Considering that the average height of three populations including P. salicifolia, P. glabra and P. communis cv. Khoj n. 2 were lower than other populations, therefore selection was carried out among genotypes of these populations. The highest number of genotypes was selected from P. salicifolia. The least seedlings height was belonged to P. glabra. Only a dwarf genotype was selected from the P. com- munis cv. Khoj n. 2 population (Figs. 2, 3 and 4). Finally, 11 dwarf genotypes were selected from these species. Cluster analysis of single genotypes Cluster analysis was performed for three popula- tions with a lower seedling height. According to this, five selected genotypes of P. salicifolia (Fig. 5), five genotypes of P. glabra (Fig. 6) and a genotype of the P. communis cv. Khoj n. 2 population (Fig. 7) were classified in the same subgroups. These genotypes had lower seedling height and internode number. Leaf width was also lower in these genotypes. Q u a l i t a t i v e t r a i t s o f 1 1 s e l e c t e d g e n o t y p e s showed that all genotypes had low trichome. The survival and tolerance of these genotypes to drought stress was not along with the increase of leaf tri- chome. The growth vigor of genotypes was low, and two genotypes 10 and 11 from P. glabra had very low growth vigor. These genotypes also had the lowest seedling height. In fact, their growth vigor was corre- lated with the seedling height. This positive correla- tion was observed between seedling height and Adv. Hort. Sci., 2019 33(2): 245-255 254 growth vigor in bitter cherry seedlings (Mahlab) ( G a n j i - M o g h a d a m a n d K h a l i g h i , 2 0 0 6 ) . U n d e r drought stress, studied genotypes did not show chlorosis, only genotypes 3 of P. salicifolia and 31 of P. glabra were greenish-yellow after drought stress. Selected genotypes were transferred to a propaga- tion phase through cutting and layering. After studying single genotypes in these three populations, genotypes no. 31, 32, 41, 57 and 12 from P. salicifolia, genotypes 10, 11, 7, 3 and 9 of P. glabra and genotype no. 4 of P. communis cv. Khoj n. 2 populations were selected as drought tolerant and dwarf genotypes and were taken to the propagation phase for future evaluation. References ABDOLLAHI H., 2011 - Pear, butany, cultivars and root- stocks. - Publication of Agricultural Education, Iranian Ministry of Agriculture, Tehran, Iran, pp. 196 (In Persian). A B D O L L A H I H . , A T A S H K A R D . , A L I Z A D E H A . , 2 0 1 2 - Comparison of dwarfing effects in two hawthorn and quince rootstocks on some commercial pear cultivars. - Iran J. Hort. Sci., 43: 53-63 (In Persian). ALIZADEH A., 2006 - Principles of irrigation systems design. - Imam Reza University Press, pp. 452 (In Persian). A R A N M . , F A T T A H I M O G H A D A M M . R . , Z A M A N I Z . , JODAKHANLOO A., 2012 - Growth characteristics of some plum seedlings in Karaj climate conditions. - Seed Plant Improv. J., 27: 149-165 (In Persian). A S H R A F M . , K A R I M I F . , 1 9 9 1 - S c r e e n i n g f o r s o m e cultivar/line of black gram for resistance to water stress. - J. Trop. Agric., 68: 57-62. CHERUTH A.J., MANIVANNAN P., WAHID A., FAROOQ M., Al-JUBURI SOMASUNDARAM H., PANNEERSELVAM R., 2009 - Drought stress in plants: A review on morpho- logical characteristics and pigments composition. - Int. J. Agric. Biol., 11: 1560-8530. DOORENBOS J., PRUITT W.H., 1977 - Guidelines for predict- ing crop water requirements. - FAO Irrigation and Drainage, Rome, Italy, Paper No. 24, pp. 144. DOULATI BANEH H., ABDOLLAHI R., ASLANPOUR M., 2013 - Morphological study of some wild grape genotypes of Sardasht and Piranshahr regions, Iran. - Seed Plant Improv. J., 29: 519-533 (In Persian). FALLOUH I., Al-MAARRI K., HADDAD S., 2008 - Study of the grafting compatibility between some clones of Syrian wild pear Pyrus Syriaca Boiss. with four pear commer- cial cultivars. - Damascus J. Agron. Sci., 24: 237-250. GANJI-MOGHADAM E., KHALIGHI A., 2006 - Genetic varia- tion of mahaleb (Prunus mahaleb L.) on some Iranian populations using morphological characters. - J. Appl. Sci., 6: 651-653. GHASEMI M., ARZANI K., YADOLLAHI A., HOKMABADI H., 2014 - Effect of drought stress on fluorescence and chlorophyll index on four pistachio seedlings. - Water Res. Agric., 27: 475-485. GRIMPLET J., DELUC L.G., GRAMER G.R., CUSHMAN J.C., 2007 - Integrating functional genomics with salinity and water deficit stress respones in wine grape-Vitis v i n i f e r a , p p . 6 4 3 - 6 6 8 . - I n : J E N C K S M . A . , P . M . HASEGAWA, and S.M. JAIN (eds.) Advances in molecu- lar breeding towards drought and salt tolerant crops. Springer-Verlag, Dordrecht, The Netherlands, pp. 817, HAGHIGHATNIA H., NADIAN H., REJALI F., TAVAKOLI A.R., 2013 - The effect of two arbuscular mycorrhizal fungi on vegetative growth and phosphorus absorption on Citrus aurantifolia under drought stress conditions. - Seed Plant Prod. J., 28: 403-417 (In Persian). HAROTKO K., 2007 - Advances and challenges in fruit root- stock research. - Acta Horticulturae, 732: 33-42. HENAREH M., 2015 - Investigation of Iranian pear species capability in order to achieve a dowarfing and fire blight tolerant rootstocks. - Final Research Report. Agricultural and Natural Resources Research Center of Azarbaijan - Gharbi, pp. 95 (In Persian). HIGGS K.H., JONES H.G., 1990 - Response of apple root- stocks to irrigation in south east England. - J. Hortic. Sci., 65(2): 129-141. HOEL B.O., SOLHAUG K.A., 1998 - Effect of irradiance on chlorophyll estimation with the Minolta SPAD-502 Leaf chlorophyll meter. - Ann. Bot., 82: 389-392. J A V A D I T . , A R Z A N I K . , E B R A H I M Z A D E H H . , 2 0 0 5 - Evaluation of soluble carbohydrates and proline in nine Asian pear cultivars (Pyrus serotinia) under drought stress. - Iran J. Biol., 17: 12-24 (In Persian). JAVADI T., BAHRAMNEJAD B., 2011 - Relative water con- tent and gas exchanges of three wild pear genotypes under water stress conditions. - J. Hortic. Sci., 24(2): 223-233. LANDSBERG J.J., JONES H.G., 1981 - Apple orchards, pp. 419-469. - In: KOZLOWSKI T.T. (ed.) Water deficit and plant growth. Volume 6. Woody plant communities. Academic Press, London, UK, pp. 598. LIU M.J., 2006 - Chinese jujube: Botany and horticulture. - Hortic. Rev., 32: 229-299. MANSOURYAR M., ABDOLLAHI H., ERFANI MOGHADAM J., SALAMI S.A., 2017 - Study of antioxidant enzymes activity and morphological changes in some vigorous pears inoculated with cause of fire blight disease (Erwinia amylovora). In vitro conditions. - Iran J. Hort. Sci. Technol., 18: 81-88. MOAZEDI R., ZAARE NAHANDI F., MAHDAVI Y., KAMRANI M., EBRAHIMI M.A., 2014 - Assessment of genetic rela- tionships of some cultivars of Asian pears (Pyrus pyrifo- lia Nakai) with some native pears of northern Iran using SSR markers. - Int. J. Farm. Allied Sci., 3: 923-929. MOHAN JAIN S., PRIYADARSHAN P.M., 2009 - Breeding plantation tree crops: Temperate species. - Springer Science Business Media, New York, USA, pp. 290. Tatari - Selection of drought tolerant pear rootstocks 255 MOUSAZADEH R., SHOUR M., TEHRANIFAR A., DAVARINA- JAD GH.H., MOKHTARIAN A., 2014 - Evaluation of genetic variation of some grape cultivars based on morphological traits. - J. Plant Prod. Res., 21: 179-192. PIRE R., PEREIRA A., DIEZ J., FERERES E., 2007 - Drought tolerance assessment of a Venezuelan grape rootstock and possible conditions mechanism. - Agrociencia, 41: 435-446. RADNIA H., 1996 - Fruit tree rootstocks. - Publication of Agricultural Education, Karaj, Iran, pp. 637 (In Persian). SABETI H., 1995 - Forests, trees and shrubs in Iran. - Publication of Yazd University, Yazd, Iran, pp. 81 (In Persian). TAHZIBIHAGH F., ABDOLLAHI H., GHASEMI A., FATHI D., 2012 - Vegetative and reproductive traits of some Iranian native pear (Pyrus communis L.) cultivars in cli- matical conditions of Karaj. - Seed Plant Improv. J., 27(1): 37-55 (In Persian). TARAHOMI G., LAHOTI M., ABASI F., 2010 - Effect of drought stress on variations of soluble sugar chloro- phyll and pottasium in Salvia leriifolia benth. - Q. J. Biol. Sci., 2(9): 1-7. XOCONOSTLE-CAZARES B., RAMIREZ-ORTEGA F.A., FLO- RES-ELENES L., RUIZ-MEDRANO R., 2010 - Drought tol- erance in crop plants. - Am. J. Plant Physiol., 5: 241- 256. ZARAFSHAR M., AKBARINIA M., ASKARI H., HOSSEINI S.M., R A H A I E M . , S T R U V E D . , S T R I K E R G . G . , 2 0 1 4 - Morphological, physiological and biochemical respons- es to soil water deficit in seedlings of three populations of wild pear tree (Pyrus boisseriana). - Biotechnol. Agron. Soc. Environ., 18: 353-366.