Quince Rootstocks Affect Some Vegetative and Generative Traits

ABSTRACT

A study was carried out in order to select the most suitable vegetative rootstock for ‘Isfahan’ quince commercial cultivar at the Isfahan Agricultural Research Center, Iran. The effect of four quince rootstocks including BA29, A, B, and C, along with quince and hawthorn seedlings on some qualitative and quantitative traits of ‘Isfahan’ quince cultivar was investigated during 2013–2017. According to the results, the highest and lowest diameter of rootstock, grafting union, and scion were observed in the quince and hawthorn seedlings, respectively. The highest length of the annual branch and the height of the tree was observed in quince seedlings and BA29 rootstocks. Quince seedling had the lowest yield/trunk cross-sectional area (0.22 kg/cm²) and hawthorn (0.35 kg/cm²), QA (0.29 kg/cm²) and BA29 (0.29 kg/cm²) rootstocks had the highest amount of this trait. Characteristics of the fruit were affected by rootstock, so that the hawthorn, BA29, and QA rootstocks induced higher fruit weight, total soluble solid/total acidity, and fruit firmness to the scion. The most leaf iron and magnesium were found in the trees grafted on the hawthorn (69.33 and 71.83 mg/kg) and BA29 (75.57 and 61.07 mg/kg) rootstocks. Nitrogen of leaf in the trees grafted on the BA29 was higher than other rootstocks (2.22%). The highest leaf phosphorus (0.15%) and potassium (1.44%) were observed in the grafted trees on the hawthorn rootstock. According to the results, for the establishment of ‘Isfahan’ quince orchard, three rootstocks including the hawthorn (low growth), QA (medium growth), and BA29 (relatively vigorous) is recommended.

KEYWORDS:

• Seedling rootstock
• quince vegetative rootstocks
• hawthorn
• nutrient elements
• yield

Introduction

Fruit species include diverse cultivars, genotypes, accessions and have been recognized for their human health benefits. They have important non-nutritive, nutritive, and bioactive substances such as flavonoids, phenolics, anthocyanins, and phenolic acids, as well as nutritive compounds such as sugars, essential oils, carotenoids, vitamins, and minerals (Fazenda et al., 2019; Halasz et al., 2010; Senica et al., 2019).

Quince belongs to the Rosaceae family and Pomoideae subfamily. The cultivars of this species are typically grafted on the seedling or clonal rootstocks. The East Malling Research Center introduced selected clonal rootstocks of the Cydonia oblonga Mill., namely, QA, QB, and QC as dwarfing rootstocks. In the Southern Fruit Research Institute, the BA29 clonal rootstock was introduced after several years of research. These rootstocks are dwarfing for quince and pear cultivars, but dwarfing of them is not so much that it can be established very dwarfing trees such as apple on the Malling rootstocks (Abdollahi et al., 2012). Another rootstock that used for quince cultivars is hawthorn. Crataegus sp. is used as a rootstock for quince and pear due to its compatibility with calcareous soils and relative tolerance to drought (Mohammadi et al., 2015).

The rootstocks have a significant effect on the growth characteristics of the fruit trees, and in some species of fruit trees, their effect is up to 50% of the economic outcome (Mezey and Lesko, 2014). The rootstocks can be effective on tree growth, precocity, production, nutrition elements of the tree (Milosevic and Milosevic, 2015), alternate bearing (Reig et al., 2018), chilling and frost tolerance (Robinson, 2004), drought tolerance (Tworkoski et al., 2016), root system depth, pest, and disease tolerance (Beers et al., 2006), germination time, and physical and chemical composition of fruit (Kviklys et al., 2017).

A two-year comparison of some vegetative growth characteristics of quince genotypes showed that the use of the quince rootstock compared to hawthorn rootstock increased the number of nodes, the growth rate, and the internode length (Abdollahi et al., 2012).

The concentration of leaf nutrition elements varies with the rootstock (Fallahi et al., 2002), and the dwarfing rootstocks absorb less nutrition element than vigorous rootstocks (Ikinci et al., 2014). It has been reported that dwarfing rootstocks reduce the concentration of elements such as nitrogen and potassium in the leaf of the scion compared to the vigorous rootstocks (Sotiropoulos, 2008).

Iron deficiency is a nutritional problem all over the world that affects the quality and quantity of the product. The use of iron deficiency tolerant rootstocks is one of the most effective ways to solve this problem (Donnini et al., 2009). In the research conducted by Mohammadi et al. (2015), the leaves of the seedling rootstocks had a lower iron content than the hawthorn and pear rootstocks. Iglesias and Asin (2005) were studied the growth of seven pear rootstocks from hybridization, along with BA29, A, and C quinces and reported that the pear hybrids and BA29 rootstocks were superior in production and chlorosis resistance compared to other rootstocks.

‘Isfahan’ cultivar has high quality and great fruit with golden yellow color and special flavor that has been cultivated in many orchards of Iran. Considering the undeniable effects of rootstock on the characteristics of cultivar, this study was conducted to investigate the graft compatibility and some of the vegetative and reproductive characteristics of ‘Isfahan’ cultivar on four quince clonal rootstocks including BA29, A, B, and C along with two hawthorn and quince seedling rootstocks in the climatic conditions of the center of Iran that is the main zone of quince productions.

Material and Methods

Orchard Establishment

Four quince clonal rootstocks, including BA29, A, B, C, which were previously propagated with mound layering method (also called stooling), along with hawthorn and quince seedling rootstocks were cultivated in the nursery at intervals of 10 cm. In September 2007, ‘Isfahan’ cultivar was grafted on each of these rootstocks. Grafted seedlings were transferred to the orchard in Isfahan Agricultural Research Station in March 2008. Seedlings were planted at a distance of 4 × 3 m and cut at a height of 80 cm. In order to train in Espalier form, metal foundations up to 2.5-m high were placed at intervals of 4 m in March. These foundations were connected to each other with three rows of galvanized wire at intervals of 70, 120, and 180 cm from the ground. Each year, with the growth of the trees, the branches, and the main arms of the trees were guided on these wires. Irrigation, pest, disease, and weed management were carried out in the orchard, according to commercial cultural practices.

Evaluation of Traits

Some grafting incompatibilities in fruit trees appear in the long term, so in the present study, the grafting compatibility was visually observed in the nursery and orchard. The vegetative traits of the trees consisted of the diameter of rootstock at a distance of 5 cm lower than the grafting union, the diameter of the grafting union, the diameter of the cultivar at a distance of 10 cm above the grafting union, the height of the tree and length of annual branches were recorded in the winter. The length and diameter of trees were measured in meters and caliber, respectively. Incompatibility symptoms in the grafting union of trees were also studied. Chlorophyll index were measured by chlorophyll meter (Spad). After bearing in 2013, some fruit characteristics, including, yield of each tree, the ratio of yield to the trunk cross-sectional area (TCSA), fruit weight, TSS, TA, taste index (TSS/TA), and firmness of fruit tissue were investigated.

The weight of the fruit was measured using a scale. The total fruit weight of each tree was considered as the yield. The yield divided by trunk cross-sectional area (π× (radius)2), and obtained yield/TCSA.

Fruit firmness was measured by a Penetrometer (model EFFEGI, Italy, plunger diameter 11.1 mm, depth 7.9 mm), at opposite peeled sides and expressed as lbf/sq.in. Total soluble solids (TSS) were determined by extracting and mixing two drops of juice from the two cut ends of each fruit into a digital Refractometer (ATAGO N-1α, Japan) at 22°C. Titrable acids (TA) were determined in 10 g of pulp samples by titration of extracted juice with sodium hydroxide (0.1 N) up to pH 8.1 and expressed as a percent of malic acid. Taste index was calculated from TSS to TA ratio.

Qualitative traits, including, bearing time, tree, and fruit size of ‘Isfahan’ on different rootstocks were recorded using the IBPGR descriptor. Nitrogen, phosphorus, potassium, iron, and magnesium contents were investigated by sampling from the leaf of trees that were taken in June.

The Kjeldahl method is used to determine the nitrogen content in leaf samples (Jones, 2001). Briefly, 0.3 g of fine dry powder was digested in concentrated H2SO4 and distilled with NaOH (40%), and ammonium nitrogen was fixed in H3BO3 (2%) and titrated with 0.1 N H2SO4. The P content of the samples was determined by the vanadate-molybdate colorimetric method (Chapman and Pratt, 1982). The absorbance of samples was measured at 470 nm in a UV/visible spectrophotometer (model PG Instrument+80, Leicester, UK). Potassium (K) was determined by the flame photometric method as described by Jones (2001). The digested extract was diluted by calcium chloride (CaCl2) at 1:9 ratios (v/v) and the absorbance was measured at 766.5 nm. Magnesium (Mg) was measured using atomic absorption spectroscopy. Briefly, digested extracts were diluted with distilled water (1:9 v/v), and then 4.75 ml of lanthanum nitrate [La (NO3) 3] was added to 250 ml of the diluted extract. Finally, the absorbance was measured at 285.2 nm by atomic absorption (Jones, 2001). To measure iron, 2 g of leaf samples was dried in an oven for 72 hours at 72°C. Then, one gram of dried leaves was ashed in the Chinese cruise for 6 h. The ash was digested with 10 ml of chloride acid in boiling water for 45 min. After filtering the samples, the extract volume increased to 50 ml using distilled water. The iron content of the samples was measured using the atomic absorption spectrometer (Varian-Spect AA 200, USA) (Reuter and Robinson, 1997).

Analysis of variance

The experiment was conducted as a randomized complete block design with three replicates and six trees per replicate for 5 years. Recorded data were analyzed using SAS software (version 9.1) and combined variance analysis was performed by Duncan’s test at a probability level of 5%.

Results

Interaction of rootstock and year on measured traits was not significant, but the effect of each of the rootstock and year factors alone was significant on most measured traits (data not shown).

The highest and the lowest diameter of rootstock, scion, and grafting union were observed in the grafted trees on quince and hawthorn seedlings, respectively. In all the rootstocks, the diameter of grafting union was more than the diameter of rootstock and scion. Although the genus and species of clonal rootstocks (QA, QB, QC, and BA29) and ‘Isfahan’ cultivar were the same, the least difference between the diameter of the grafting union and the diameter of rootstock and scion was observed in quince seedling rootstocks. The diameter of the scion to the diameter of the rootstock ratio was varied from a minimum of 0.969 in the combination of ‘Isfahan’ cultivar on QC rootstock to 1.02 in the combination of this cultivar on the hawthorn rootstock (Figure 1). During the years of experiment, the growth of the seedlings was normal and there was no sign of incompatibility at the grafting union.

Figure 1. Diameter of rootstock, scion, and grafting union in grafting of ‘Isfahan’ cultivar on different rootstocks

According to Table 1, the highest of the tree height was observed on the quince seedling and then BA29 rootstocks. The quince seedling rootstock produced the highest length of the annual branch. The length of annual branches in BA29 rootstock had no significant differences with the quince seedling rootstock. QC induced the lowest length of annual branches in ‘Isfahan’ cultivar. The hawthorn seedling rootstock produced trees with the lowest height.

Table 1. The effect of different rootstocks on the average of some traits of ‘Isfahan’ quince cultivar from 2013 to 2017

Rootstock	Length of annual branch (cm)	Height of the tree (cm)	Yield (kg/trees)	Yield/TCSA* (kg/cm^2)	Chlorophyll index
Q.A	24.3b	182.9d	23.1a	0.29ab	67.57a
Q.B	24.8b	195.8c	20.5b	0.26c	59.05b
Q.C	23.9c	185.1d	22.1a	0.3b	60.15b
PQBA29	25.3ab	218.5b	23.8a	0.29ab	60.21b
Quince seedling	26.5a	230.4a	18.4c	0.22d	57.39b
Hawthorn seedling	24.4b	173.8e	22.3a	0.35a	57.72b

* TCSA: Trunk Cross-Sectional Area

Similar letters in each column indicate no significant difference at the 5% level of LSD.

QA, QC, BA29, and hawthorn seedling rootstocks had more yield than QB and quince seedling rootstocks. The lowest yield belonged to quince seedling rootstock. The yield of this rootstock was 20.44% lower than the yield of trees on the QA rootstock. The hawthorn seedling rootstock showed the highest yield/TCSA. Yield/TCSA of the BA29 and QA rootstocks did not differ significantly from the hawthorn rootstock. The lowest yield/TCSA was observed in quince seedlings. The highest of the chlorophyll index were found on the QA rootstock. There were no significant differences between other rootstocks (Table 1).

The diameter of rootstock, scion, and grafting union increased from 2013 to 2017, and the diameter difference between them reduced further over time (Figure 2).

Figure 2. Diameter of rootstock, scion, and grafting union in different years

According to Table 2, the highest length of the annual branch was observed in the second year (2014), and the amount of this trait did not show significant differences in the last two years. The height of the tree also increased over time. Productivity in ‘Isfahan’ cultivar started from the third year and yield and yield/TCSA increased in the following years.

Table 2. The average of some traits of ‘Isfahan’ quince cultivar in different years

Year	Length of annual branches (cm)	Height of the tree (cm)	Yield (kg/trees)	Yield/TCSA (kg/cm^2)
2013	27.4b	163.1d	–	–
2014	28a	185.6c	–	–
2015	23.8d	199.8c	5.37	0.05c
2016	24.5c	215.1b	24.5	0.22b
2017	24.2c	222.9a	45.9	0.35a

Similar letters in each column indicate no significant difference at the 5% level of LSD.

Rootstock also affects the quantitative and qualitative characteristics of the fruit. The highest fruit weight of ‘Isfahan’ was obtained on QA and BA29 rootstocks. These two rootstocks along with the hawthorn rootstock had the most TSS for ‘Isfahan’ cultivar. Quince seedling rootstock induced the lowest fruit weight and TSS as well as the highest TA in ‘Isfahan’ cultivar. Fruit weight and TSS in this rootstock were 23.61% and 15.78% lower than fruit weight and TSS in QA rootstock, respectively. TA in quince seedling rootstock was 10.81% higher than QB and BA29 rootstocks. Fruit taste index on BA29 rootstock was more than other rootstocks. After that, Hawthorn and QA rootstocks were placed. The taste index in quince seedling rootstock was lower than the other rootstocks. The highest fruit firmness was induced on hawthorn and BA29 rootstocks. Fruit firmness on these two rootstocks was 7.14% higher than fruit firmness on QC and quince seedlings (Table 3).

Table 3. The effect of rootstock on the average of fruit traits of ‘Isfahan’ quince cultivar from 2013 to 2017

Rootstock	Fruit weight (g)	Fruit TSS (%)	Fruit TA (%)	TSS/TA	Fruit firmness (lbf/sq.in)
Q.A	288a	19a	3.5c	5.42b	6.8b
Q.B	230d	17c	3.3d	5.15c	6.6c
Q.C	263c	18b	3.6b	5c	6.5d
PQBA29	287a	19a	3.3d	5.75a	7a
Quince seedling	220e	16d	3.7a	4.32d	6.5d
Hawthorn seedling	270b	19a	3.5c	5.42b	7a

Similar letters in each column indicate no significant difference at the 5% level of LSD.

The rootstock affected qualitative traits such as fruiting time, fruit size, and tree size. Hawthorn, BA29, and QA rootstocks produced fruits larger than the fruits of other rootstocks. Tree size in quince seedling and BA29 was large and in QC and hawthorn was small. The QA, QC, and hawthorn rootstocks were early bearing and quince seedling rootstock were late bearing (Table 4).

Table 4. Qualitative traits of ‘Isfahan’ quince cultivar on different rootstocks from 2013 to 2017

Rootstock	Bearing time	Fruit size	Tree size
Q.A	Early	Large	Medium
Q.B	Middle	Medium	Medium
Q.C	Early	Medium	Small
PQBA29	Middle	Large	Large
Quince seedling	Late	Medium	Large
Hawthorn seedling	Early	Large	Small

Leaf iron and magnesium were also affected by the rootstock. The highest leaf iron was observed in BA29 and hawthorn rootstocks. QA, QC, and quince seedling had the least amount of iron in the leaves. The highest amount of magnesium was found in the leaves of grafted ‘Isfahan’ cultivar on hawthorn rootstock. In general, BA29 and hawthorn rootstocks had more iron and magnesium. In the other rootstocks, the rates of these two elements were about the same level (Figure 3).

Figure 3. The effect of different rootstocks on the average of leaf iron and magnesium levels in ‘Isfahan’ quince cultivar from 2013 to 2017

The rootstocks had a significant effect on the amount of nitrogen and potassium in the leaves but had no significant effect on the phosphorus content of the grafted cultivar. BA29 absorbed the most nitrogen. The most potassium uptake occurred at hawthorn rootstock (Figure 4).

Figure 4. The effect of different rootstocks on the average of leaf nitrogen, phosphorus, and potassium percentage in ‘Isfahan’ quince cultivar from 2013 to 2017

Discussion

In this study, diameter of rootstock, scion, and grafting union, in grafted trees on hawthorn seedlings were lower than other rootstocks (Figure 1). A comparison of the scion to rootstock diameter ratio also showed a relative stability in grafted compounds (between 0.969 and 1.02). This ratio was reported from 0.938 in ‘Bartlett’ pear cultivar on the hawthorn rootstock to 1.342 in ‘Haru Sweet’ cultivar on the same rootstock by Abdollahi et al. (2012). It was reported that ‘Williams’ pear scion on the pear seedling and QC rootstocks had the highest and the lowest scion diameter, respectively (Francescatto et al., 2014). Abdollahi et al. (2012) unexpectedly reported that the hawthorn rootstock diameter in the three grafted combinations of ‘Spadona,’ ‘Bartlett,’ and ‘Haru Sweet’ scions were higher than the diameter of other rootstocks (pear seedling and QA). In their opinion, the low growth of the grafted scion on hawthorn rootstock caused the accumulation of nutrients in below the grafting union and significant growth of rootstock.

The incompatibility of the scion and rootstock reveals some symptoms such as significant differences in growth rate of the rootstock or scion, not complete connection of grafting union, fracture of the grafting union, or early abscission of the scion leaves (Tukey, 1964). In this study, no significant differences were observed in the diameter of cultivar and rootstocks or other symptoms of incompatibility. In rootstocks with higher diameter, the diameter of the cultivar was also higher, indicating that growth rate in all the grafting compounds followed a linear pattern and increased with increasing rootstock diameter. The decreasing trend of difference in the diameter of rootstock, cultivar, and grafting union over time (Figure 2) also showed the grafting compatibility between the studied rootstocks with ‘Isfahan’ cultivar.

Quince seedling and then BA29 rootstocks produced trees with the highest height and length of the annual branch. QC and hawthorn seedling rootstocks had the lowest length of the annual branch and tree height, respectively (Table 1). The effect of the dwarfing rootstock on tree height caused by the restriction of annual growth in comparison to vigorous rootstocks such as seedling rootstocks. Abdollahi et al. (2012) also reported the lowest height of pear cultivars on hawthorn seedling rootstock. They found that different cultivars of grafted pear on hawthorn and QA rootstocks were very dwarfing and semi-dwarfing, respectively. Also, Gizala 6 rootstocks in sweet cherry (Morandi et al., 2019) and M9 rootstock in apple (Milosevic et al., 2018) had more dwarfing. The rootstock influence on tree growth is also confirmed by Tworkoski and Miller (2007).

BA29, QA, hawthorn, and QC had higher yields than quince seedling and QB rootstocks (Table 1). According to Du Plooy et al. (2002), vegetative tissues compete with fruit for carbohydrate uptake, leading to a negative correlation between production and vegetative growth. In this study, no significant relationship was found between growth vigor and the amount of yield, so that BA29 and quince seedling vigorous rootstocks had the highest and lowest yields, respectively. Different results have been reported by researchers about the relationship between rootstock vegetative growth and the amount of yield. So that, Dodangeh et al. (2012) showed that the MM106 apple rootstock, which was dwarfer than the M2, M4, M7, and MM111 rootstocks, produced more yield. More production in dwarfing rootstock is also reported by Morandi et al. (2019). Contrary to these results, Milosevic et al. (2018) reported that ‘Red Chief’ apple cultivar on M9 dwarfing rootstock had limited vegetative growth with fewer leaves and fruiting branches, which confirmed by Tareen et al. (2003). It seems that the interaction between the rootstock and the scion is more important than the growth vigor of the rootstock to achieve higher yields.

Yield efficiency is the ratio of produced fruit to wood in the tree and is obtained by dividing the amount of produced fruit (kg) by the trunk cross-sectional area (cm²) (Hassani et al., 2012). Hawthorn seedling, BA29, and QA rootstocks had the highest yield/TCSA and as a result, the highest yield efficiency (Table 1). In most reports, the yield efficiency in dwarfing rootstock was higher than that of the vigorous rootstocks. For example, Sotiropoulos (2008) reported that apple production in dwarf trees was more than in vigorous trees. It has also been reported that lower growth of Gisela 6 shoots in sweet cherry leads to more resource allocation to fruits and increase yield efficiency (Hrotko and Rozpara, 2017). In this study, BA29 vigorous rootstock had a high yielding, after hawthorn.

‘Isfahan’ cultivar on QA rootstock had the highest chlorophyll index. The other rootstocks did not differ significantly with each other (Table 1). The effect of rootstock on spad index and chlorophyll content has also been reported by other researchers (Hrotko and Rozpara, 2017; Pirmoradian et al., 2017).

The highest fruit weight of ‘Isfahan’ was obtained on QA, BA29, and hawthorn rootstocks (Table 3). There are conflicting reports about the effect of rootstock on fruit weight. In a research, the highest fruit weight was observed in grafted apple cultivars on the low-growth rootstocks such as M7 and MM106, and the lowest fruit weight was found in grafted cultivars on seedling vigorous rootstocks (Sotiropoulos, 2008), while Milosevic et al. (2018) reported the weight of the ‘Red Chief’ fruit on the M4 rootstock was more than the M9 dwarfing rootstock. Normally, fruit size and weight are influenced by the number of fruits (De Salvador et al., 2006), but the grafted ‘Isfahan’ cultivar on BA29, QA, and hawthorn rootstocks produced more yield with larger fruit weight and size. On the opposite side was the quince seedling rootstock.

TSS, TA, and taste index are affected by the rootstock (Milosevic et al., 2018). In the current research, the fruit TSS ranges varied from 16% (quince seedling) to 19% (QA, BA29, and hawthorn) (Table 3). Rodriguez-Guisado et al. (2009) studied five quince genotypes in Spain and reported that TSS was between 11.5 and 14.7 which was lower than the TSS values in ‘Isfahan’ cultivar on all studied rootstocks. ‘Isfahan’ has high TSS and a sweet taste, which is also consumed as fresh fruit. In this study, quince seedling rootstocks induced the lowest TSS and the highest TA in ‘Isfahan’ cultivar. Similarly, in a study by Biranvand et al. (2011), TA content in apple seedling rootstock was higher than MM106 vegetative rootstock, but TSS content in MM106 was higher than apple seedling rootstock. Morandi et al. (2019) also reported TSS of the grafted fruits on the Gisela 6 was more than the vigorous rootstocks. Rootstock vigor causes a change in the allocation of resources between vegetative and reproductive organs, which can affect fruit production and quality (Morandi et al., 2019), but in this study, hawthorn, QA and BA29 showed low, medium, and high growth, respectively. These three rootstocks induced the highest TSS and taste index in the grafted cultivar and no significant relationship was found between tree growth vigor and fruit quality.

Fruit firmness on hawthorn and BA29 rootstocks was higher than fruit firmness on QC and quince seedling rootstocks (Table 3). In a study, MM106 induced more fruit firmness than M4 and M9 to the scion, and M4 showed the lowest fruit firmness (Milosevic et al., 2018). In contrast, Tareen et al. (2003) reported that the rootstock had no effect on the fruit firmness of ‘Starking Delicious’ apple cultivar. It seems that these differences affected by factors pre- and post-harvest and the amount of the yield. Greater yield leads to the production of smaller and firmer fruits (Milosevic et al., 2018). In the present study, the rootstocks that induced a higher yield to grafted cultivar also produced firmer fruits, but these fruits were not smaller. The fruit firmness can be changed by absorbing elements such as calcium from different rootstocks.

The rootstock had an effect on the precocity of the tree. QC, QA, and hawthorn were earlier bearing than other rootstocks (Table 4). Galli et al. (2011) also reported that the ‘Abate Fetel’ pear cultivar on QC and Adams rootstocks was earlier bearing and smaller than when it grafted on BA29. Similarly, Webster et al. (1997) mentioned that dwarfing rootstocks such as QC propagate vegetatively and induce early bearing to the tree. Other researchers reported that pear trees on the quince rootstocks were early bearing and had higher yields than seedling rootstocks (Francescatto et al., 2014; Lepsis et al., 2013). It has been reported that the BA29 is not early bearing and the fruits on this rootstock have a larger size. The bearing on this rootstock was in the fifth year after planting (Tatari et al., 2016).

Hawthorn, BA29, and QA rootstocks produced fruits larger than the others (Table 4). The effect of the rootstock on fruit size has also been reported previously, so that ‘Abate Fetel’ and ‘Conference’ pear cultivars that grafted on the BA29 produced larger fruit (Galli et al., 2011; Iglesias and Asin, 2005). According to research results, the average fruit size on the BA29 rootstock was slightly larger than the other rootstocks (Alonso et al., 2011). Similarly, North and Cook (2008) reported that in ‘Forelle’ pear cultivar, the most yield with suitable size and optimum quality were obtained from grafted pear trees on QA and BA29 rootstocks.

The quince seedling and BA29 rootstocks produced the larger size of the tree (Table 4). The results of the present study are in agreement with the results of Iglesias and Asin (2005) who reported the BA29 vegetative rootstock induced more growth vigor. The rootstock can affect on the transfer of nutrients and carbohydrates between the roots and branches. Also, the rootstock influence the distribution of carbohydrates, elements, and growth regulators and affect the vegetative growth of the plant (Rufato et al., 2014). Tree size was not directly related to fruit weight and size. This is also confirmed by Morandi et al. (2019). Carried out research on apple and pear rootstocks has also shown that the rootstock was effective in controlling tree size and dwarfing rootstocks reduced tree size. The low root volume of dwarfing rootstocks results in reduced shoot growth (Jackson, 2003).

The highest iron was observed in leaves of grafted ‘Isfahan’ on BA29 and hawthorn rootstocks (Figure 3). Similarly, Prado and Alcantara (2011) reported that grafted pear on BA29 had more leaf iron. Ghasemi et al. (2010) also found that the use of hawthorn rootstock caused higher resistance of tree to soil stress conditions such as drought, salinity, and iron deficiency due to the alkalinity of the root environment. The highest amount of iron was reported in grafted ‘Isfahan’ on quince and hawthorn seedlings by Abdollahi et al. (2012). Comparison of two rootstocks of hawthorn and quince seedlings showed relative equality of both rootstocks in iron uptake. In the present study, the amount of iron in these two rootstocks was not equal. In another study, significant differences were observed in soil iron uptake by different rootstocks of stone fruit trees, and the positive effect of both rootstock and cultivar on the amount of iron uptake was reported in grafted peach cultivars on these rootstocks (Tsipouridis et al., 2005). It was reported that the leaves of grafted pear trees on dwarfing rootstocks had a lower concentration of micronutrients (Ikinci et al., 2014), which confirmed in QA and QC but did not confirm in hawthorn. It should be noted that iron measurement of leaves is not normally a good criterion for assessing plant iron status, because leaf iron may precipitate in the leaf Apoplast and cause the error in evaluation (Prado and Alcantara, 2011). The Mg element in the leaves of hawthorn rootstock was also more than others (Figure 3).

BA29 absorbed the most nitrogen. The most potassium uptake was found in hawthorn rootstock (Figure 4). Hadavand et al. (2016) reported that trees with higher yields accumulated more nitrogen and less iron in their leaves. Sotiropoulos (2008) also showed that dwarfing rootstocks had lower concentrations of potassium and phosphorus than seedling rootstocks, which none of them is consistent with the findings of this study. Due to the decrease in vegetative growth induced by the hawthorn rootstock, it has been reported that this rootstock can reduce the vegetative growth of the tree, cause dwarfing, and thus reduce the nutritional needs of the grafted tree (Abdollahi et al., 2012).

Conclusion

The quince cultivation industry using modern horticultural systems requires high-density orchards and early bearing trees. For high-density orchards, rootstock growth control is very important (Elkins et al., 2012). Based on the obtained results, BA29, QA, and hawthorn rootstocks are preferred because of their yield, yield efficiency, taste index, fruit firmness, and early bearing. The choice of these rootstocks depends on the orchard conditions so that hawthorn is used to have a higher density orchard and QA and BA29 are used to have an orchard with less density. Also, according to the higher uptake of iron by BA29 and hawthorn rootstocks and its higher tolerance to alkaline soils in such soils these two rootstocks are recommended. Due to the susceptibility of the quince species to the severity of winter damage, the hawthorn rootstocks are also more suitable than the quince rootstocks in zones with the cold winter.