839

http://journals.tubitak.gov.tr/agriculture/

Turkish Journal of Agriculture and Forestry Turk J Agric For(2016) 40: 839-854© TÜBİTAKdoi:10.3906/tar-1606-14

Cherry tree growth models for orchard management improvement

Mirjana LJUBOJEVIĆ1,*, Vladislav OGNJANOV1, Goran BARAĆ1, Jovana DULIĆ1,Maja MIODRAGOVIĆ1, Mirjana SEKULIĆ1, Nataša JOVANOVIĆ LJEŠKOVIĆ2

1Faculty of Agriculture, University of Novi Sad, Novi Sad, Serbia2Faculty of Pharmacy, University Business Academy in Novi Sad, Novi Sad, Serbia

* Correspondence: mirjana.ljubojevic@polj.uns.ac.rs

1. IntroductionIn cherries, introduction of dwarfing rootstocks (Lang, 2000; Robinson, 2005) and newer cultivars (Kappel, 2002) has provided new opportunities for the development of high-density cherry orchards. Since the 1970s, increased interest has been given to both tree size and architecture at the level of individual leaf, individual branch, and the canopy. Breeders seek to create the ‘ideotype’ through detailed studies on how various parts of the plant interact with each other in order to achieve the best performance in a specific environment (Borojević, 1990; Wu, 1998). Developmental aspects of the architectural models and selection criteria should include the distinction of rhythmic and continuous growth in relation to apex development, configuration of the branching points, plagiotropy and orthotropy of tree axes, and topological properties that describe the number and spatial arrangement of branch links (Prusinkiewicy and Remphrey, 2000).

Fruit tree size is predominantly reduced through: 1) conventional breeding for novel tree growth habits, 2) use of induced mutations, 3) dwarfing rootstock available in many species that restrict the tree’s growth to various sizes, 4) controlled pruning that produces fruit trees of a more

manageable size even on rootstocks of medium vigor, 5) root pruning or root restriction growth by pots, and 6) application of growth retardants.

Although the sweet and sour cherries are species rich in diversity for plant growth habits (Ljubojević et al., 2012; Ognjanov et al., 2012), variability in marketed cultivars represents only a tiny part of the total genetic diversity. Conventional breeding programs have focused mainly on fruit characteristics with little interest for the diversity of tree habit. Meristems and phytomers are development regions that shape the morphology of the plant architecture, and numerous studies have emphasized the role of plant hormones in these processes (Sussex and Kerk, 2001). Research in the past decade has led to the identification of genes that are important in regulating the plant architecture (Wu, 1998; Segura et al., 2009). Present studies emphasize tree architecture as genetic, ontogenetic, and environmental interaction (Segura et al., 2009). Thus, integration of molecular markers in breeding programs should be a powerful tool for hastening cultivar development (Kenis and Keulemans, 2007). Only a few genetic linkage maps are available for sweet or sour cherry and quantitative trait loci (QTLs) have been reported

Abstract: Twelve cultivars and selections of sweet and sour cherries were surveyed and characterized agromorphologically. A total of 33 characters, mainly defined by the International Plant Genetic Resources Institute and the International Union for the Protection of New Cultivars of Plants, were used to describe the tree, branches, leaves, blooming, and fruits. This allowed unequivocally clear separation between six distinct growth forms—compact, dwarf, upright (pillar), columnar, weeping, and standard form—as well as evaluation of their interaction with four rootstocks of different vigor. Natural tree habit diversity within distinct growth forms has the potential to reduce pruning and training requirements in cherry, particularly in medium- and high-density planting. The reported unique growth forms, within compact and columnar tree habit of sour and sweet cherries, combine specific genetic potential of scion fruiting habits and vigor. Genetic diversity was confirmed by simple sequence repeats. Research showed that multiple interactions exist between rootstock and genotypic determined components of the tree architecture, namely branching density, shoot growth dynamics, flower bud differentiation, and distribution. Developed ideotypes are new foundational concepts of modern production strategies bringing significant contribution to improvement of cherry orchard training systems, light interception, efficient fruit harvest, and economics.

Key words: Breeding, ideotype, orchard technologies, Prunus avium L., Prunus cerasus L., tree architecture

Received: 03.06.2016 Accepted/Published Online: 01.10.2016 Final Version: 14.12.2016

Research Article

840

LJUBOJEVIĆ et al. / Turk J Agric For

only for sour cherry. Until now, most of the efforts have concentrated on the use of molecular markers in order to identify the  S-alleles controlling gametophytic self-incompatibility (Hauck et al., 2006), characterize cultivars’ skin and flesh colors (Sooriyapathirana et al., 2010), and assess genetic diversity (Barać et al., 2013). While upright habit is predominant in cherry trees, instances of spreading and drooping forms exist (Ognjanov et al., 2012). In addition, genetic traits for tree size control can be classified as compact, semidwarf, and dwarf. Unconventional tree architectures include columnar and weeping habits (Scorza et al., 1999; Fideghelli et al., 2003).

Mutagenesis has already been used to induce many useful traits affecting plant size, blooming time and fruit ripening, fruit color, self-compatibility, self-thinning, and resistance to pathogens (Predieri, 2001). In Italy, an irradiation program of the main sweet cherry cultivars, aimed at inducing ‘compact’ mutations, began in 1967. Although the mutagenous treatment seems to produce more pronounced effects in irradiated cultivars, none resulted in any dwarf mutants. Compact mutants are characterized by an emphasized basal dominance and early fruiting. The main problem, before the commercial release of compact selections, was the genetic instability observed in several clones (Fideghelli et al., 2003).

Based on the fundamental studies of RG Hatton at East Malling (UK) between 1920 and 1938, tree size control has been obtained through the adoption of dwarfing rootstocks in combination with training systems and pruning techniques. Rootstocks have been used for propagating temperate fruit trees for more than 2000 years. Many of the rootstocks used, besides providing a simple method of propagation, also affect cropping, scion growth, degree of lateral branching, the individual length of shoots, branching angle, and adaptability to different environmental conditions (Tobutt, 1985a; Webster, 1995; Whiting et al., 2005). However, none of the rootstocks for cherries that have been released to date provide the same size control in different agroecological conditions (Ljubojević, 2012).

Cultural techniques directed at controlling tree growth—summer and winter pruning, root pruning, and root space restriction—were reviewed by Ferree et al. (1992). Even though results of tree size control through the application of the growth regulator paclobutrazol were encouraging (Webster and Quinlan, 1984; Jacyna at al., 1989), this agent does not conform to the current integrated fruit production practice.

The objective of this study was to survey, identify, and characterize sweet and sour cherry autochthonous germplasms and cultivars for vegetative tree habit, flowering, and fruit parameters and to compare the different vigor of rootstocks/crown architecture interactions. The research goal was conservation and

utilization of cherry genetic diversity in crown architecture parameters and reproductive traits. The aim of germplasm characterization was establishment of genetic relationships between standard cultivars and indigenous accessions. The ultimate aim was selection of initial material for breeding scion cultivars of reduced vigor with reliable cropping and good fruit quality suitable for intensive high-density orchard management grafted on moderately vigorous rootstocks adaptable to drier climate. The potential for breeding ornamental forms with dwarf, columnar, or weeping habit was also investigated.

2. Materials and methods2.1. Plant materialThe plant material represents the biodiversity of P. cerasus and P. avium germplasm with over 1000 in situ evaluated plants in 3 years within natural populations, in private collections, and in productive orchards. It consists of three selections and four cultivars of sour cherry and two selections and three cultivars of sweet cherry grafted onto four different size-controlling rootstocks. These selections are the outcome of a search for unique and rare traits significant for the realization of the set of breeding objectives using positive individual selection made from a sample comprising more than 200 accessions and cultivars ex situ. For accession mapping a GPS record-keeping program was used.2.2. Scion selections and cultivars2.2.1. Sour cherry (P. cerasus L.)Selection PC_0904 – low vigor upright form, local farmer selection from Central Serbia;

Selection PC_0905 – medium vigor pillar form, local farmer selection from South Serbia;

Selection PC_1001 – low vigor weeping form, local farmer selection from province of Vojvodina;

‘Maynard’ – spur-type cultivar, ‘Delbard nursery’, France;

‘Lara’ – (‘Kelleriis 14’ × ‘Rexelle’) low vigor cultivar bred at PKB ‘Agroekonomic’, Boleč-Belgrade, Serbia;

‘Kelleriis 14’ – (self-pollination of Ostheimer Weichsel) low vigor standard cultivar; and

‘Érdi  bőtermő’ – (‘Pandy Meggy 38’ × ‘Nagy Angol’) medium vigor standard cultivar. 2.2.2. Sweet cherry (P. avium L.)Selection PA_1001 – semidwarf selection from local nursery, province of Vojvodina, Serbia;

Selection PA_1002 – dwarf selection from natural population, central Serbia;

‘Victoria’ – columnar cultivar from ‘Ahrens+Sieberz’ nursery, Germany;

‘Sara’ – columnar cultivar from ‘Ahrens+Sieberz’ nursery, Germany; and

‘Summit’ – (‘Van’ × ‘Sam’), standard cultivar.

841

LJUBOJEVIĆ et al. / Turk J Agric For

2.3. RootstocksDwarf cherry rootstock ‘Gisela 5’ – GS (P. cerasus ‘Schattenmorelle’ × P. canescens), moderately vigorous rootstock ‘Oblačinska’ sour cherry – OB (P. cerasus), vigorous rootstock ‘Colt’ – CL (P. avium sel. F299/2 × P. pseudocerasus), and vigorous rootstock ‘Mahaleb’ – MH (P. mahaleb) were chosen as reference rootstocks of different vigor. All investigated rootstocks and selections were subsequently propagated in the nursery of the Faculty of Agriculture, Novi Sad. P. mahaleb was sexually propagated by sowing seeds taken from one individual tree with rootstock seedling progeny expressing a high level of uniformity in seedling and plant morphology. ‘Gisela 5’, ‘Colt’, and ‘Oblačinska’ sour cherries were propagated by softwood cuttings taken from 4-year-old virus-free mother trees. An experimental orchard was established in 2009 at the Faculty of Agriculture, Novi Sad, experimental station at Rimski Šančevi (43°57′N, 20°26′E, 80 m a.s.l.). Weather conditions are characterized by an average annual temperature of 11.3 °C and total annual rainfall of 609.3 mm. Soil at the orchard site is classified as vertisol clay/loam, low in organic matter, with pH in 0.01 M KCl of 6.67.

Planting distance was 4 m × 2 m. Ten representative trees within each rootstock/genotype combination were planted for data collection. The plants were cultivated until the stabilization of their phenological and reproductive cycle. Plant characterization was conducted in July, after completion of intensive growth, in three consecutive years, in the third, fourth, and fifth vegetations. Pruning measures were carried out in order to enhance the natural characteristics of growth and plant development. Annually, standard agricultural practices (thinning, fertilization, pest control, and irrigation) were performed.2.4. Agromorphological characterizationMorphological analysis included both qualitative and quantitative characteristics of trees, branches, leaves, and fruits (according to the International Union for the Protection of New Cultivars of Plants (UPOV) Cherry Descriptor, 2006). Investigated qualitative characteristics of trees and branches included plant vigor (PV), habit (H), anchorage intensity (AI), branching intensity (BI), and coloration of braches (BC), while quantitative characteristics comprised tree height (TH), crown diameter (CD), crown volume (CV), crown shape index (CSI), scion diameter (SD), rootstock diameter (RD), grafting point diameter (GPD), length of internodes (LI), and branching angle (BA). Tree crown volume calculations for different crown shapes were based on three variables: crown height, diameter, and crown shape index, as proposed by Changok (2007). According to the proposed methodology, shape indices were labeled as spheroid – S4 (whereby crown shape index of 0.667 was used in calculation), expanded parabolic – S5 (0.625), parabolic – S6 (0.500), fat cone – S7 (0.375), and cone – S8 (0.333).

CD2 × CH × Crown shape index = Crown volumeLeaf qualitative characters—leaf shape (LS), leaf blade

color of upper side (LC), leaf tip (LT), leaf base shape (BS), and incision of margins (MI)—as well as quantitative characteristics, such as leaf blade length (LL), leaf blade width (LW), and leaf petiole length (LP), were examined.

Fruit qualitative characteristics pertained to harvest maturity (FM), fruit skin color (FC), and fruit flesh color (FFC). Examined quantitative characteristics included fruit weight (FW), fruit height (FH), fruit width (FW), fruit thickness (FT), petiole length (FPL), fruit stone weight (FSW), and soluble solids content (SSC).

Vegetative measurements were performed on 30 replicates (three 1-year-old shoots, from ten plants per genotype), with the exception of leaf features, where five leaves were taken from each of 30 replicate shoots. Scion and rootstock diameters were measured at 5 cm in distance above and below the grafting point. Pomological measurements were conducted on 30 fruits per genotype/rootstock combination. Fruit skin and flesh color were evaluated descriptively, while fruit characteristics related to size and weight were measured. Soluble solid content was measured using a handheld refractometer.

Conceptual framework and vocabulary for increasingly complete and accurate characterization of tree architectures was accepted from Hallé and Oldeman (1970) and Prusinkiewicz and Remphrey (2000).2.5. Molecular characterizationIn molecular analysis young leaf samples were collected from indigenous landraces, cultivars, and species used in development of new high-density tree growth cherry ideotypes. Total genomic DNA was extracted using the protocol of Dellaporta et al. (1983). DNA quality and concentration was assessed by NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). A set of 10 simple sequence repeat (SSR) primer pairs was used in the diversity analysis, including UCD-CH12, UCD-CH17, UCD-CH19 (Struss et al., 2003), BPPCT005, BPPCT008, BPPCT027, BPPCT028, BPPCT032, BPPCT034, and BPPCT038 (Dirlewanger et al., 2002), and polymerase chain reaction (PCR) was performed using MultiGene OptiMax (Labnet International Inc., Edison, NJ, USA) according to published protocols. PCR products were separated by electrophoresis using 2% MetaPhor agarose (Cambrex Bio Science Rockland, Inc., Rockland, ME, USA) at 80 V for 12 h in 1X TBE, stained with ethidium bromide (0.8 mg/mL), and visualized using UV light on the FOTO/Analyst Investigator/FX Workstation (Fotodyne Incorporated, Hartland, WI, USA). Fragment sizes were estimated in relation to the GeneRuler Low Range DNA Ladder (Thermo Scientific).

Genetic similarity between accessions was calculated by the Nei and Li similarity coefficient (Nei and Li, 1979)

842

LJUBOJEVIĆ et al. / Turk J Agric For

using FreeTree (Pavlicek, 1999). Cluster analysis was performed using the unweighted pair group method with arithmetic average (UPGMA) and a dendrogram was constructed using FigTree v1.4.2 software (http://tree.bio.ed.ac.uk/software/figtree/).2.6. Statistical analysisStatistical analysis of quantitative characteristics included basic statistics and multivariate discriminant analysis (MDA) in order to provide information about differences between investigated rootstock/scion combinations. Results were processed using STATISTICA 12 software (StatSoft, DELL, Round Rock, TX, USA).

3. Results3.1. Germplasm preservation workThe Faculty of Agriculture, Novi Sad, benefits from an extensive network of collaborators across Serbia, who contribute greatly to the germplasm preservation work and serve as invaluable sources of breeding material and ideas. Through this initiative, more than 200 accessions were selected and conserved to date ex situ. Germplasm preservation work has resulted in new selections stemming from interactions of different vigor, habit, and branching within six distinct growth forms, namely compact, dwarf, upright (pillar), columnar, weeping, and standard. 3.2. Qualitative vegetative characteristicsTree vegetative nonmetric characterizations of sweet and sour cherry are summarized in Tables 1 and 2. Irrespective of the rootstock, very weak tree vigor was exhibited in PC_0904 and PA_1002, which could be declared as dwarf genotypes. Vigor scores were low for sour cherry PC_1001 as well as sweet cherry PA_1001 on all rootstocks exhibiting highly compact form, such as ‘Van’ in sweet cherry and ‘Schattenmorelle’ and ‘Kelleriis 14’ in sour cherry. Depending on rootstock, ‘Gisela 5’ and ‘Oblačinska’ sour cherry, medium vigor and semiupright crown was observed in ‘Victoria’ and ‘Sara’, whereas PC_0905 had dense upright crown. ‘Érdi  bőtermő’ and ‘Summit’ are standard medium vigor cultivars. Generally, the lowest scores for plant vigor were observed on ‘Gisela 5’ rootstock with scores of 1–5. ‘Oblačinska’ sour cherry used as rootstock significantly decreased tree vigor (scores 1–5) and facilitated precocity, while also positively affecting bearing.  Selection PA_1002 was completely incompatible with ‘Oblačinska’ sour cherry.

Plant habit varied from upright in PC_0904 (Figure 1A) and PC_0905, through semiupright in ‘Maynard’, ‘Lara’ (Figure 1B) and ‘Victoria’, to spreading in ‘Kelleriis 14’, ‘Sara’, PA_1001 (Figure 1C), PA_1002, and ‘Summit’, with occurrence of pendant growth habit in ‘Érdi bőtermő’ and especially in PC_1001. Natural branching intensity ranged from extremely weak in PC_1001 to strong in

‘Maynard’, whereby the majority of accessions had a medium branching habit. Only in a few accessions was branching intensity dependent on the rootstock selection, and it was most pronounced in ‘Oblačinska’ sour cherry rootstock.

Generally, leaf color was stronger and more glossy in sour (scores of 3 to 5) than in sweet cherries (score of 3). Sweet cherry genotypes had leaves with large foliar surface area. Narrower leaves, with elongated leaf tips and obtuse base shapes, were more common in sweet cherry genotypes, while sour cherries had elliptic leaves, with shorter leaf tips and both acute and obtuse base shapes. Owing to the narrow elliptic leaf shape, sour cherry PC_0904 was more similar to sweet cherries. Acute shape of leaf base (score 1) was determined only in PC_0904 and ‘Lara’ sour cherry, while all other sweet and sour genotypes had a higher value (score of 2) of the basal leaf angle, which was closer to obtuse. Leaf tip length was generally classified as medium (score of 5) with the exception of sour cherry PC_0904 and ‘Érdi bőtermő’ (score of 7). In sweet cherry, leaf blade margin incisions were serrate only, while both crenate and serrate forms were noted in sour cherry.

Anchorage intensity was dependent on genotype/rootstock combinations and was much weaker in ‘Oblačinska’ sour cherry and ‘Gisela 5’ rootstocks in combination with top-heavy central leader trees of medium vigor and vigorous scion genotypes. Sweet cherry trees grown on these rootstocks require support in the first 3–4 years. Anchorage seems to be adequate for ‘Colt’ and ‘Mahaleb’ without support. Anchoring was excellent for P. mahaleb due to deep, thick roots of significant length, as well as ‘Colt’, with a shallow but dense network of fine roots.3.3 Quantitative vegetative characteristics Quantitative vegetative parameters are shown in Tables 3 and 4. Average tree height of sour cherries was 144 cm, and it was 25% less than in sweet cherries (191 cm) due to the presence of pillar and columnar types. Crown diameter, volume, and shape index pertaining to sour cherries on average exceeded those associated with sweet cherries. Minimum values of crown diameter and crown volume were observed in ‘Victoria’/’Gisela 5’ (30 cm and 0.02 m3, respectively), while the maximum values were noted in ‘Summit’/’Mahaleb’ (3.64 m3) and ‘Érdi bőtermő’/’Mahaleb’ (1.72 m3). Parameters related to trunk thickness were the lowest for ‘Gisela 5’ and the greatest for ‘Mahaleb’. Narrow branching angle (25–45°) was noted in both species, for sour cherries - PC_0904, PC_0905, and ‘Maynard’ and sweet cherry PA_1002. These selections are characterized by almost completely pillar growth.

Leaf parameters proved to be distinctive and useful for agromorphological characterization of cherry germplasm. The smallest leaves were observed for ‘Maynard’, ‘Lara’, and

843

LJUBOJEVIĆ et al. / Turk J Agric For

Table 1. Qualitative vegetative characteristics of investigated sour cherry germplasm (average values for 3-year period).

Variety Selection Rootstock PV H BI BC LS LC LT BS MI AI

PC_0904

GS 1* 1 5 3 2 3 7 1 2 7

OB 3 1 7 5 2 3 7 1 2 7

CL 3 1 5 3 2 3 7 1 2 7

MH 5 1 5 3 2 5 7 1 2 7

PC_0905

GS 5 1 5 5 4 3 5 2 2 7

OB 5 1 5 3 4 3 5 2 2 5

CL 7 1 5 3 4 3 5 2 2 7

MH 9 1 5 3 4 5 5 2 2 7

PC_1001

GS 1 4 1 3 4 5 5 2 2 5

OB /† / / / / / / / / /

CL / / / / / / / / / /

MH 5 4 3 3 4 5 5 2 2 7

‘Maynard’

GS 3 2 5 3 4 5 5 2 2 5

OB 5 2 7 3 4 3 5 2 2 7

CL 5 2 7 3 4 3 5 2 2 7

MH 7 2 5 3 4 5 5 2 2 7

‘Lara

GS 1 1 5 5 4 3 5 1 2 5

OB 1 2 5 5 4 3 5 1 2 5

CL 3 2 5 5 4 3 5 1 2 5

MH 3 2 5 5 4 3 5 1 2 7

‘Kelleriis 14’

GS 1 3 5 3 4 3 5 2 2 5

OB 3 3 7 3 4 3 5 2 2 5

CL 5 3 5 3 4 3 5 2 2 7

MH 5 3 5 3 4 3 5 2 2 7

‘Érdi bőtermő’

GS 3 4 5 5 4 3 7 2 2 7

OB 5 4 7 5 4 5 7 2 2 5

CL 7 4 5 5 4 3 7 2 2 7

MH 7 4 5 5 4 5 7 2 2 7

* - Scores according to UPOV descriptor for sour cherry.Plant vigor (PV): 1 – very weak, 3 – weak, 5 – medium, 7 – strong, 9 – very strong.Habit (H): 1 – upright, 2 – semiupright, 3 – spreading, 4 – dropping. Branching intensity (BI): 3 – weak, 5 – medium, 7 – strong.Coloration of braches (BC): 3 – weak, 5 – medium, 7 – strong.Leaf shape (LS): 1 – narrow elliptic, 2 – elliptic, 3 – circular, 4 – ovate, 5 – obovate.Leaf color (LC): 3 – light, 5 – medium, 7 – dark.Length of leaf tip (LT): 3 – short, 5 – medium, 7 – long.Leaf base shape (LB): 1 – acute, 2 – obtuse, 3 – truncate.Incision of margins (MI): 1 – only crenate, 2 – both crenate and serrate, 3 – only serrate.Anchorage intensity (AI): 3 – weak, 5 – medium, 7 – strong.† - Incompatible combinations, no data available.

844

LJUBOJEVIĆ et al. / Turk J Agric For

‘Kelleriis 14’ sour cherries, with leaf length ranging from 6.4 to 11.6 cm, leaf width from 3.0 to 6.3 cm, and petiole length from 1.3 to 2.8 cm. Sour cherry ‘Maynard’ was also characterized by the lowest internodes length, irrespective of the rootstock type (1.0–1.5 cm). The largest leaves were measured in sweet cherries ‘Sara’, PA_1001, and ‘Summit’, with leaf length, leaf width, and internodes length ranging from 11.2 to 16.2 cm, 5.4 to 8.1 cm, and 2.5 to 4.5 cm,

respectively. As expected, all leaf characteristics values were lower for sour cherries relative to sweet cherry cultivars. Generally, accessions on ‘Gisela 5’ had the smallest leaves.

In order to observe similarities among investigated genotypes and genotype/rootstock interactions MDA was conducted for each rootstock separately (data not shown). MDA of the tree’s characteristics for the first three rootstocks (‘Gisela 5’, ‘Oblačinska’, and ‘Colt’) showed that

Table 2. Qualitative vegetative characteristics of investigated sweet cherry germplasm (average values for 3-year period).

Variety Selection Rootstock PV H BI BC LS LC LT BS MI AI

PA_1001

GS 2* 3 7 3 2 3 7 1 3 3

OB /† / / / / / / / / /

CL 3 3 5 3 2 3 7 2 3 7

MH 5 3 5 3 2 3 7 2 3 7

PA_1002

GS 1 3 3 3 2 3 7 2 3 7

OB / / / / / / / / / /

CL 1 3 3 3 2 3 7 2 3 7

MH 2 3 3 3 2 3 7 2 3 7

‘Victoria’

GS 3 2 3 3 2 3 5 2 3 5

OB 5 2 3 3 2 3 7 2 3 3

CL 7 2 3 3 2 3 5 2 3 7

MH 7 2 3 3 2 3 7 2 3 7

‘Sara’

GS 4 3 3 3 2 3 7 2 3 5

OB 6 3 5 3 2 3 7 2 3 3

CL 5 3 4 3 2 3 7 2 3 7

MH 7 3 4 3 2 3 7 3 3 7

‘Summit’

GS 4 3 3 5 2 3 7 2 3 7

OB 5 3 5 5 2 3 7 2 3 5

CL 7 3 5 3 2 3 7 2 3 7

MH 7 3 5 3 2 3 7 2 3 7

* - Scores according to UPOV descriptor for sweet cherry.† - Incompatible combinations, no data available. Plant vigor (PV): 1 – very weak, 3 – weak, 5 – medium, 7 – strong, 9 – very strong.Habit (H): 1 – upright, 2 – semiupright, 3 – spreading, 4 – dropping. Branching intensity (BI): 3 – weak, 5 – medium, 7 – strong.Coloration of braches (BC): 3 – weak, 5 – medium, 7 – strong.Leaf shape (LS): 1 – narrow elliptic, 2 – elliptic, 3 – circular, 4 – ovate, 5 – obovate.Leaf color (LC): 3 – light, 5 – medium, 7 – dark.Length of leaf tip (LT): 3 – short, 5 – medium, 7 – long.Leaf base shape (LB): 1 – acute, 2 – obtuse, 3 – truncate.Incision of margins (MI): 1 – only crenate, 2 – both crenate and serrate, 3 – only serrate.Anchorage intensity (AI): 3 – weak, 5 – medium, 7 – strong.

845

LJUBOJEVIĆ et al. / Turk J Agric For

significant canonical coefficients on the first axis were coefficients for tree height, crown volume, grafting point diameter, and rootstock diameter, while the canonical coefficients for leaf length and width, as well as crown shape index, had a higher discriminative value on the second axis. For rootstock ‘Mahaleb’, statistically significant on the first axis were coefficients for crown volume, leaf length, crown

shape index, scion diameter, and grafting point diameter, while on the second axis important coefficients were those for leaf width and rootstock diameter. Since similar characteristics were significant for all four rootstocks, a single MDA scatterplot was constructed to simplify the observation (Figure 2). In the plot of the first two MDA axes, there was a clear separation of the investigated

(A) (B)

(C)

Figure 1. A. Sour cherry selection PC_0904, B. Sour cherry variety ‘Lara’, C. Sweet cherry selection PA_1001.

846

LJUBOJEVIĆ et al. / Turk J Agric For

Table 3. Quantitative vegetative characteristics of the investigated sour cherry germplasm (average values for 3-year period).

TH*

(cm)CH(cm)

CD(cm)

CV(m3)

CSI SD(cm)

RD(cm)

GPD(cm)

BA(°)

LI(cm)

LL(cm)

LW(cm)

LP(cm)

PC_0904

GS 80 69 60 0.13 S4‡ 2.5 3.0 2.5 25 1.8 10.9 3.0 2.1

OB 110 100 90 0.42 S4 3.0 4.0 3.0 25 2.5 10.3 3.6 1.9

CL 80 70 60 0.12 S5 3.0 3.5 3.0 35 2.0 9.3 3.6 1.7

MH 130 123 100 0.64 S4 5.0 5.5 5.0 30 2.6 10.7 4.9 2.8

PC_0905

GS 130 90 50 0.07 S7 2.5 3.5 2.5 30 1.8 8.7 4.7 2.0

OB 180 150 120 0.57 S8 3.5 4.5 3.5 35 2.6 7.1 5.5 1.6

CL 150 134 70 0.17 S8 5.0 6.0 4.0 30 2.5 8.8 6.0 2.0

MH 200 187 130 0.83 S8 10.0 11.0 10.0 45 2.5 11.1 6.3 2.8

PC_1001

GS 110 83 55 0.13 S4 3.0 4.5 2.5 45 1.5 9.0 5.0 1.3

OB /† / / / / / / / / / / / /

CL / / / / / / / / / / / / /

MH 140 110 110 0.65 S5 5.5 6.0 5.0 45 3.0 10.4 5.6 1.8

‘Maynard’

GS 90 70 40 0.05 S5 2.5 3.0 3.0 30 1.0 6.7 3.6 1.0

OB 170 153 85 0.54 S5 4.0 4.5 4.5 30 1.2 8.5 4.4 1.0

CL 150 120 75 0.27 S6 4.5 5.0 4.5 40 1.5 7.6 4.9 1.4

MH 170 160 120 0.68 S7 7.0 8.0 7.0 30 1.4 8.1 4.2 1.4

‘Lara’

GS 80 55 70 0.13 S5 2.5 2.5 2.5 40 1.2 6.5 3.5 1.3

OB 95 70 90 0.28 S5 4.0 5.0 4.0 40 1.6 6.7 3.7 1.3

CL 142 112 110 0.71 S4 4.0 4.5 4.0 45 1.8 7.0 3.8 1.6

MH 168 128 110 0.81 S4 5.0 6.0 4.5 45 2.0 7.1 4.0 1.9

‘Kelleriis 14’

GS 95 40 60 0.08 S4 3.0 3.0 2.5 50 1.4 6.4 3.4 1.4

OB 160 140 100 0.69 S5 4.0 4.0 3.5 45 1.6 7.6 4.5 1.4

CL 155 125 110 0.74 S5 4.0 4.5 4.0 50 1.8 7.4 4.4 1.6

MH 183 143 120 1.01 S5 5.0 6.0 4.5 50 1.9 7.8 4.6 1.8

‘Érdi bőtermő’

GS 140 95 95 0.42 S5 4.5 5.0 3.5 70 1.6 9.2 4.4 1.8

OB 160 120 110 0.76 S4 4.0 4.5 3.0 60 2.5 10.1 5.1 2.4

CL 180 120 120 0.90 S4 4.5 6.5 4.5 50 2.3 8.9 4.2 2.1

MH 226 179 140 1.72 S5 5.5 6.0 5.0 55 2.5 11.6 5.7 2.4

* - Abbreviations used: tree height (TH), crown diameter (CD), crown volume (CV), crown shape index (CSI), scion diameter (SD), rootstock diameter (RD), grafting point diameter (GPD), length of internodes (LI), branching angle (BA) leaf blade length (LL), leaf blade width (LW), and leaf petiole length (LP). † - Incompatible combinations, no data available.‡ - Shape indices: spheroid – S4, expanded parabolic – S5, parabolic – S6, fat cone – S7, and cone – S8.

847

LJUBOJEVIĆ et al. / Turk J Agric For

genotypes into the four groups. The first two groups belong to sour cherries, while the next two comprised sweet cherries, observed from the left to the right side of the plot. Sour cherry cultivars ‘Lara’, ‘Maynard’, and ‘Kelleriis 14’ differed from selections PC_0904, PC_0905, PC_1001, and cultivar ‘Érdi bőtermő’, as indicated by the DA scatterplot. Among sweet cherries, there was a clear separation between cultivars ‘Victoria’ and ‘Sara’ at the bottom and selections PA_1001, PA_1002, and cultivar ‘Summit’ at the right top segment of the scatterplot. Selection PC_0904 and cultivar ‘Érdi bőtermő’, the closest to the coordinate system’s center, represent genotypes with the most stable response on all four rootstocks.

Rootstock affects tree scion growth, vigor, habit, and production of lateral branches. ‘Oblačinska’ sour cherry and ‘Gisela 5’ had favorable influence on scion precocity, abundance of flowering, propensity of flowers to set fruits, and ripening time. In size-controlling rootstocks, such as ‘Gisela 5’, tree architecture, yield, and fruit quality varied as a function of genetic factors, environmental conditions, and age-induced morphological changes owing to specific scion/rootstock interaction. 3.4. Pomological characteristics Pomological characteristics were investigated through assessment of flower and fruit characteristics. Distribution of flowers along 1- and 2-year-old branches revealed

Table 4. Quantitative vegetative characteristics of the investigated sweet cherry germplasm (average values for 3-year period).

TH*

(cm)CH(cm)

CD(cm)

CV(m3) CSI SD

(cm)RD(cm)

GPD(cm)

BA(°)

LI(cm)

LL(cm)

LW(cm)

LP(cm)

PA_1001

GS 125 70 80 0.23 S4‡ 2.5 2.0 2.5 45 2.2 12.3 5.2 2.8

OB /† / / / / / / / / / / / /

CL 175 120 125 0.98 S4 4.0 4.0 4.5 45 2.7 13.8 5.8 3.1

MH 180 135 140 1.30 S5 5.0 5.0 6.0 45 2.7 14.2 5.8 3.5

PA_1002

GS 95 60 50 0.06 S6 2.0 2.0 2.0 4.0 2.0 12.4 5.4 3.3

OB / / / / / / / / / / / / /

CL 100 70 70 0.17 S6 2.0 2.5 3.0 35 2.0 11.2 5.5 3.9

MH 160 146 70 0.27 S7 3.0 3.5 3.5 35 2.2 14.3 5.5 4.5

‘Victoria’

GS 80 42 30 0.02 S6 2.5 2.5 3.0 60 2.5 12.1 6.2 2.5

OB 190 160 100 0.63 S6 5.0 4.0 5.0 45 2.4 13.7 6.3 2.6

CL 300 246 90 0.59 S7 5.5 4.0 5.5 45 2.5 13.6 6.1 3.1

MH 320 290 110 1.38 S6 6.5 5.5 7.0 40 2.7 14.5 7.0 2.9

‘Sara’

GS 145 109 55 0.09 S8 4.5 4.5 5.0 40 2.5 12.5 6.3 3.3

OB 220 190 80 0.32 S8 6.0 6.0 7.0 45 2.9 15.5 7.3 3.1

CL 185 157 70 0.30 S6 4.0 4.5 5.5 40 2.6 14.1 5.7 3.1

MH 200 187 70 0.36 S6 7.0 6.0 7.5 45 2.9 15.0 7.0 3.0

‘Summit’

GS 150 93 90 0.22 S7 4.0 3.5 3.5 45 1.7 13.9 5.4 3.1

OB 180 120 140 0.92 S6 4.5 4.0 5.0 45 2.8 14.9 6.4 3.0

CL 200 145 150 1.71 S4 4.5 4.5 4.5 55 2.7 15.2 7.0 3.2

MH 240 214 170 3.64 S3 6.5 8.0 7.0 45 3.0 15.1 8.1 3.6

* - Abbreviations used: tree height (TH), crown diameter (CD), crown volume (CV), crown shape index (CSI), scion diameter (SD), rootstock diameter (RD), grafting point diameter (GPD), length of internodes (LI), branching angle (BA) leaf blade length (LL), leaf blade width (LW), and leaf petiole length (LP). † - Incompatible combinations, no data available.‡ - Shape indices: spheroid – S4, expanded parabolic – S5, parabolic – S6, fat cone – S7, and cone – S8.

848

LJUBOJEVIĆ et al. / Turk J Agric For

great genotype differences in flowering patterns. All rootstock combinations (sour cherry selection PC_0904, ‘Kelleriis 14’, and ‘Lara’) had a considerably larger number of flowers on 1- than 2-year-old branches (Figure 3). Selection PC_1001 flowered only on 1-year-old branches without any vegetative buds, resulting in bare wood in consecutive years. Sour cherries flowered more frequently on 1-year-old branches than sweet cherries, where flowers were observed on the base of 1-year-old branches only. Fertility of flowers on 1-year-old branches in PC_0904 and PC_1001 was very low, while it was medium in ‘Kelleriis 14’ and excellent in the cultivar ‘Lara’.

Fruit characterization included some of the most important qualitative and quantitative characteristics (Table 5). Fruit pomological characteristics showed no or little variation on different rootstocks; thus, results were presented as averages pertaining to all four rootstocks. Fruit weight was generally dependent on genotype. Among sour cherries, only PC_1001 stood out due to unique and valuable pomological properties. More specifically, it had the highest fruit weight (10.1 g), fruit height, width, and thickness (27.6, 27.0, and 23.6, respectively); however, its

yield efficiency was compromised by low flower fertility. Very small (2.1 g) and acidic but decorative fruit was determined in cultivar ‘Maynard’. In sweet cherries, values were more consistent, ranging from 5.5 g (‘Victoria’) to 8.0 g (‘Summit’). Stalk length ranged from 25.8 mm to 58.0 mm in sour cherry and from 33.5 mm to 45.0 mm in sweet cherry germplasm. Fruit stone weight varied from 0.2 to 0.5 g in sour cherry and from 0.2 to 0.4 g in sweet cherry. With respect to qualitative fruit characteristics, significant differences were noted among investigated germplasms. Some of the sweet and sour cherries—‘Maynard’, ‘Érdi  bőtermő’, ‘Lara’, PC_1001, and PA_1002—had fruit with dark red to black skin and a red pulp. Sour cherry selection PC_0905 had fruit with an orange-red skin and yellow-pink colored pulp. Fruit flesh color in sweet cherry was mainly medium red. At the other extreme was ‘Maynard’ with 19.2% soluble solid content, compared to ‘Kelleriis 14’ with only 7.4%. Marked contrasts were observed in yield efficiency, ranging from negligible productivity as in PC_0904 to high yield per unit area as in sour cherries ‘Kelleriis 14’ and ‘Lara’ and sweet cherries PA_1001 and ‘Summit’.

Figure 2. Scatterplot for discriminant analysis canonical coefficients based on sour and sweet cherry morphological characteristics.

849

LJUBOJEVIĆ et al. / Turk J Agric For

Number of flowers on one- (blue) and two-year-old shots (red) on rootstock ‘Gisela 5’.

Number of flowers on one- (blue) and two-year-old shots (red) on rootstock ‘Oblačinska’ sour cherry.

Number of flowers on one- (blue) and two-year-old shots (red) on rootstock ‘Colt’

Number of flowers on one- (blue) and two-year-old shots (red) on rootstock ‘Mahaleb’

0 50 100 150 200

PC_0904PC_0905MaynardPC_1001

Erdi BotermoVictoria

SaraSummit

0 50 100 150 200 250 300

PC_0904PC_0905Maynard

Erdi BotermoKelleris 14

LaraPA_1001

VictoriaSara

Summit

0 50 100 150 200 250

PC_0904PC_0905Maynard

Erdi BotermoLara

Kelleriis 14PA_1001PA_1002

VictoriaSara

Summit

0 50 100 150 200

PC_0904PC_0905MaynardPC_1001

Erdi BotermoLara

Kelleriis 14PA_1001PA_1002

VictoriaSara

Summit

Figure 3. Distribution of flowers along 1- and 2-year-old branches in compatible rootstock/scion combinations.

Table 5. Average pomological characteristics of investigated cherries on four rootstocks, during a 3-year period.

Har

vest

m

atur

ity

Frui

t ski

n co

lor

Frui

t fles

h co

lor

Frui

t wei

ght

(g)

Frui

t he

ight

(m

m)

Frui

t wid

th

(mm

)

Frui

t th

ickn

ess

(mm

)

Stal

k le

ngth

(m

m)

Frui

t sto

ne

wei

ght (

g)

Solu

ble

solid

co

nten

t (%

)

Sour cherry

PC_0904 12 June 5* 4* 3.5 16.3 16.5 16.7 44.2 0.3 10.1

PC_0905 12 June 3 3 4.6 16.9 20.1 17.2 50.3 0.2 14.6

PC_1001 16 June 7 4 10.1 27.6 27.0 23.6 58.0 0.5 13.5

‘Maynard’ 18 June 8 5 2.1 12.7 15.6 14.0 25.8 0.5 19.2

‘Érdi bőtermő’ 10 June 7 4 6.8 19.7 23.8 21.3 48.4 0.3 12.4

‘Kelleriis 14’ 11 June 5 4 4.6 17.1 18.8 17.2 45.9 0.3 7.4

‘Lara’ 11 June 7 4 4.8 17.8 18.6 17.3 38.7 0.4 9.5

Sweet cherry

PA_1001 12 June 5 3 6.6 19.4 23.1 19.4 45.0 0.2 13.2

PA_1002 13 June 7 4 6.8 20.5 22.7 20.5 41.5 0.3 12.9

‘Victoria’ 13 June 5 4 5.5 20.6 20.5 19.9 39.5 0.3 13.9

‘Sara’ 30 May 5 4 5.8 20.7 20.8 19.9 33.5 0.4 12.5

‘Summit’ 12 June 5 4 8.0 24.7 25.9 21.2 40.2 0.3 12.5

* - Scores according to UPOV Cherry Descriptor (2006).

850

LJUBOJEVIĆ et al. / Turk J Agric For

3.5. Molecular characterizationAnalysis using SSR markers revealed a high level of diversity within Prunus sp. accessions originating from Serbia. Molecular markers differentiated indigenous sweet and sour cherry germplasms, represented as advanced selections featuring specific traits for reported novel ideotypes concerning tree growth, as diverse compared to old and commercial cultivars, sour cherry progenitor species P. fruticosa Pall. and P. mahaleb L. as standards. Cluster analysis separated all accessions into four main clusters according to their taxonomy (Figure 4). Clusters representing sweet and sour cherry are nearest to each other, followed by clusters representing Prunus fruticosa and Prunus mahaleb as outgroups in this research. Low vigor sour cherry landraces from Serbia, PC_0904 and PC_1001, were clustered together and close to low and medium vigor standard cultivars ‘Keleris 14’ and ‘Érdi  bőtermő’, respectively. PC_0905 was in a separate subcluster and together with ‘Lara’ and ‘Maynard’. The relationships within Prunus avium support phenotypic observations where dwarf and semidwarf accessions PA_1001 and PA_1002 are distinct from standard cultivars and they represented a base for development of a new weak tree vigor sweet cherry ideotype.

4. DiscussionRootstock research and breeding programs have produced a wide series of dwarfing rootstocks for sweet cherry, but the cultivars grafted on to the most promising rootstock, ‘Gisela 5’, show water deficiency symptoms under hot

and dry summer climate. This is typical of the important European cherry-growing countries in Central and South Europe. In conditions of high soil temperatures and drought stress, when grown on sandy and clay loam soils, it is difficult to develop a management system that promotes new shoot growth to balance its early, heavy production. This situation negatively influences the crop and fruit quality in subsequent years and leads to many bare branches in the canopy, lack of an adequate number of leaves to ensure a good fruit size, and early senescence of trees (Bujdosó et al, 2004; Bujdosó and Hrotkó, 2005).

Sweet and sour cherry scion breeding has been focused mainly on fruit characteristics with little attention dedicated to the diversity of tree habit. These species are rich in diversity for plant growth habit with relatively little effort to genetically alter scion tree growth habit towards efficient high-density production systems. To improve cherry production efficiency, research needs to focus on developing novel production systems that increase fruit production, yield, and fruit quality while improving labor efficiency and safety (Laurens et al., 2000; Whiting et al., 2005). Plant growth and development vary as a function of deterministic and opportunistic factors and their interactions (Hallé et al., 1978). There is evident paucity of breeding initiatives aiming to genetically alter cherry tree growth habit towards optimizing the canopy and high photosynthetic activity in high-density production systems. A selection process resulting in an ideotype of cherry trees suitable for high-density orchards involves overcoming several morphological, physiological, and

Figure 4. UPGMA dendrogram based on 10 SSR markers showing high level of diversity between analyzed landraces and cultivar of different cherry species.

851

LJUBOJEVIĆ et al. / Turk J Agric For

cultural constrains (Laurens et al., 2000). Our work has demonstrated the performance and breeding potential of tree growth habits as a combination of specific variabilities in vigor, tree habit, and type of fruiting. This creates many challenges for scion breeding to suit intensive, efficient, easily managed cherry orchards on moderately vigorous rootstocks tolerant to warm, dry climate during the summer.

Vigor has a strong impact on many characteristics linked to production, including precocity, productivity, and fruit size. Global cherry production relies on the use of vigorous spreading orthotropic scion cultivars grafted onto different vigor rootstocks. It has been necessary to explore the possibilities of breeding less vigorous scion cultivars because small plagiotropic trees would allow more intensive planting and high quality yield on 1- and 2-year-old branches, faster picking, more effective spraying, and cheaper netting. Selection PA_1001 is such an architectural tree model, characterized by low to medium vigor with medium apical dominance. It is not only mechanically strong, but its aforementioned traits are highly consistent irrespective of the rootstock, and it can be trained to produce a desirable framework.

Tree habit characterizes the overall tree form, which in our germplasm varies from erect to weeping. Selection PA_1001, in addition to low vigor, has well-shaped tree habit for the purpose of high-density orchard management—the trunk is monopodial, branches are wide semiorthotropical with the position of flowers on the base of 1-year-old and evenly distributed along 2-year-old wood. Light summer pruning further shifts orthotropy to plagiotropy growth, increasing the interception of diffuse light inside the crown and leaf area, consequently enhancing the whole crown’s photosynthesis activity. At the same time, that selection combines yield, fruit size, fruit firmness, and cracking tolerance. These findings are in agreement with those reported by Goncalves et al. (2008), who documented the differences of biological behavior of sweet cherry cultivars with open and dense canopies. It has been shown that thick shoots, such as in PA_1001, could support heavier fruit load and provide better water and mineral supply to the fruit, especially during and immediately after anthesis, when the tree is without leaves and relies on branch reserves, thus assuring fruit of higher quality (Lang, 2011).

The columnar phenotype, as described in apple, is a very valuable compact growth form (Tobutt, 1985b; Ognjanov et al., 1998; Ognjanov, 2011); however, thus far, it has not been detected in sour and sweet cherry germplasm. Sweet cherry cultivars ‘Sara’ and ‘Victoria’, declared as columnar, have a vigorous upright cordon with reduced number of lateral shoots due to many latent buds, nearly normal internodes, and scarce unevenly distributed

fruiting spurs. The tree of PA_1002 selection naturally grows like a columnar sturdy cordon with scarce spreading short shoots of low density and shorter internodes. The flowers and fruits develop very densely on spurs and short shoots. A similar growth type has never been described in sweet cherry germplasm. Our columnar sour cherry model is a monopodial cordon with low apical dominance, a large number of sylleptic branches with slow elongation, as in the cultivar ‘Lara’. Very slow growth is a result of genetically low vigor of the scion with additional possibilities of rootstock/scion interaction, as in ‘Lara’/’Oblačinska’ sour cherry. The branches are plagiotropic with lateral flowering along whole 1-year-old wood and good fertility of flowers and fruit set, as in Roux and Cook’s model (Prusinkiewicy and Remphrey, 2000). Such a unique architectural tree model can be additionally trained by successive summer pruning after harvest, resulting in the reproductive stage until June and sylleptic growth from the lateral meristem in the axis of leaves without a period of dormancy. Described columnar tree ideotypes have a naturally narrow canopy that is ideal for high-density spindle and super spindle trees developed to suit the narrow “fruiting wall” canopy architecture conceptualized by Lang (2011). With respect to the production value, the suggested ideotype is valuable for planting high-density, pedestrian orchards that afford regular, high, and early yields. It also assists in easier management and cost reduction, while providing options for protecting the orchard against rain and bird damage, as well as achieving a safer and more productive working environment. Such trees insure adequate light penetration to all fruits and leaves, resulting in homogeneous development, ripening, and coloration of fruits. Light transmittance, as in the proposed canopy concept, improves photosynthetic rate, enhances leaf morphoanatomy, and elevates chlorophyll concentration due to increased palisade and parenchyma thickness (Goncalves, 2008). Furthermore, dense planting of such trees is possible, whereby a simple system of pruning is required. The shift in cherry orchard management toward dwarf trees trained as fruiting walls achieves perfect alignment of improved genetics and applied physiological advances that makes the use of protected orchard strategies much more feasible. Smaller trees facilitate erection and usage of shelters, such as various structures, multibay high tunnels or hoop houses, which can serve as protection from rain, hail, and birds, and also some insects and diseases. They can provide increased protection from spring frosts and wind, and can advance ripening dates to broaden the harvest season or obtain fruit for earlier, higher-value market windows (Lang, 2011).

A much less desirable approach to high density planting is the pillar growth habit, Scarrone and Stone’s model, distinguished by medium vigor and strong apical

852

LJUBOJEVIĆ et al. / Turk J Agric For

dominance, where the apex continues to grow a narrow monopodial orthotropic crown forming a shaded branched sympodium with a distinct main stem (Prusinkiewicy and Remphrey, 2000). This model is characteristic of the ‘Jachim’ and ‘Boas’ cherry cultivars particularly recommended for use in home gardening (Schuster et al., 2014a, 2014b). A similar growth type has previously been described for peaches as pillars (Scorza et al., 1989, 1999). The PC_0905 sour cherry selection has a pillar tree habit, as well. It is characterized by extremely erect branching, medium length internodes, long leaves, and a dense canopy. However, it is difficult to achieve a sufficiently open crown to allow good flower bud differentiation on 2-year-old branches inside the canopy. This selection has a good fruit set, with fruit quality characteristics for ‘Montmorency’, and is highly resistant to fungal diseases. Abundant flowering is produced only on 2-year-old branches.

Standard cherry trees are often difficult to grow as dwarfs or in pots. Dwarf fruit trees produce regular-sized fruit on smaller trees. In our research, the interaction of PC_0904 sour cherry and ‘Maynard’ with ‘Oblačinska’ sour cherry and ‘Gisela 5’ has unique ornamental value. ‘Maynard’ provides an opportunity for plant pruning to unusual shapes as espalier and topiary. Miniature trees are popularly used in backyards and balcony gardens. Shrub tree forms, such as P. eminens (P. cerasus × P. friticosa), as well as PC_0904, have an orthotropic module, forming a branched sympodium with upright branching angle, with no distinct main stem—Leeuwenberg’s model (Prusinkiewicy and Remphrey, 2000). Bors (2005) reported on only a few other dwarf sour cherries, such as P. fruticosa selections and P × kerrasis (75% P. cerasus and 25% P. fruiticosa).

Sour cherry PC_1001 is a result of pseudomonopodial axes, mainly built by the superposition of scarce lateral shoots formed on the curve of parent shoots with mixed orientation. It flowers only on 1-year-old shoots, where initially orthotropic shoots become pendulous under fruit weight, which creates a plagiotropic appearance as in Champagnat’s model (Prusinkiewicy and Remphrey, 2000). Based on our findings, PC_1001 can be characterized as an ornamental weeping sour cherry selection. Absence of vegetative buds on 1-year-old branches renders them bare wood in the second year. New 1-year-old wood is naturally a result of only terminal bud development. Summer pruning after harvest produces multiple 1-year-old shoots on which flower buds are evenly distributed. Additional breeding value of PC_1001 stems from it being the largest sour cherry fruit in the sour cherry germplasm in our collection.

Estimation of tree crown volume using the Changok (2007) classification improved the accuracy and reduced the analysis complexity without reliance on optical

imagery. Two-dimensional side views of idealized crown shapes minimized the subjectivity of age, cultivar, and rootstock influence estimation and enabled the direct calculation of crown volume using a geometric formula. Finally, shape is a result of the tree growth in height, width, and rigidity under the influence of gravity, tropism, and yield weight. Thus, achieving desired and controllable tree architecture requires modeled growth element regulation using functions based on botanical and pomological characterizations. Incorporation of biomechanics, structural  analysis, and growth dynamics, into architectural tree models created a multiyear simulation of tree growth under genetic, environmental, and topiary influences, obtaining a realistic tree shape at every stage of its development.

The entire indigenous reported germplasm differs from the germplasm prevalent in West Europe, East Europe, and the Near East. The Balkan Peninsula is a secondary center of domestication and cultivation of sweet and sour cherry. Several thousand years of germplasm introgression and on-farm selective breeding in cherries on the Balkan Peninsula can be traced back for several thousand years, since the ancient Silk Road as a trading route between Europe and Central Asia from the 2nd century BCE to the 15th century CE. As a result of various geographic regions, agroclimatic conditions, plant vigor and canopy shape suitable for small yards, and fruit preference for the Morello type, numerous landraces in cherry have been developed. Parallel to the usual morphological identifying methods, SSRs prove to be essential for precise identification of germplasm diversity and understanding of determinants of genetic diversity. Indigenous sweet and sour cherry advanced selections from Serbia used in this study represent novel germplasm, a wealth of morphological diversity, and original enhancement to cherry breeding, especially concerning tree growth habit.

Sweet and sour cherry are rich in diversity for natural plant growth habit, making the development of new cultivars with crown architecture for high-density training systems feasible. Our results provide valuable contributions to the theoretical foundation for the development of a plant ideotype that is a result of the search for efficient selection criteria for traits that secure optimum canopy for high-density cherry orchards and enhanced photosynthetic activity. The ideotype reported in PA_1001 and ‘Lara’ allows modeling of the relationships between intercorrelated developmental characteristics, such as vigor, tree habit, and type of fruiting suitable for intensive high-density cherry orchards. Our studies on fundamental growth characteristics and fruiting physiology define qualitative breeding criteria for model-based crown architecture genetic improvement aligned with modern cherry production strategies.

853

LJUBOJEVIĆ et al. / Turk J Agric For

This means that small trees could be obtained as an interaction of medium vigor and more drought-resistant rootstocks and scion with greater precision in tree architecture, with columnar or monopodial plagiotropic growth, which “deconstructs” the leaf and fruit populations into spatial and developmental relationships that can be managed to optimize tree size, crown architecture, yields, and uniform fruit quality in cherry high-density planting systems.

The reported germplasm is a valuable gene pool for further breeding. Development of new cultivars with different growth habits is feasible through further

intercrossing to obtain “mixed” growth types suitable for high-density production systems, amateur gardens, and potted trees.

AcknowledgmentsThis research was supported and funded by the Ministry of Education, Science and Technological Development as one of the research topics in the project “Selection of sweet and sour cherry dwarfing rootstocks and development of intensive cultivation technology based on sustainable agriculture principles,” evidence number TR 31038, for the period 2011–2016.

References

Barać G, Ognjanov V, Obreht D, Ljubojević M, Bošnjaković D, Gašić K (2013). Genotypic and phenotypic diversity of cherry species collected in Serbia. Plant Mol Biol Rep 32: 92-108.

Borojević S (1990). Principles and Methods of Plant Breeding. Amsterdam, the Netherlands: Elsevier. Bors HR (2005). Dwarf sour cherry breeding at the University of Saskatchewan. Acta Hortic 667: 135-140.

Bujdosó G, Hrotkó K (2005). Rootstock-scion interactions on dwarfing cherry rootstocks in Hungary. Hortic Sci (Prague) 32: 129-137.

Bujdosó G, Hrotkó K, Stehr R (2004). Evaluation of sweet and sour cherry cultivars on German dwarfing rootstocks in Hungary. Journal of Fruit and Ornamental Plant Research 12: 233-244.

Changok L  (2007). Estimation of urban tree crown volume based on object-oriented approach and LIDAR data. MSc, ITC Enschede, Enschede, the Netherlands.

Dellaporta SL, Wood J, Hicks JB (1983). A plant DNA minipreparation: version II. Plant Mol Biol Rep 1: 19-21.

Dirlewanger E, Cosson P, Tavaud M, Aranzana MJ, Poizat C, Zanetto A, Arus P, Laigret F (2002). Development of microsatellite markers in peach [Prunus persica (L.) Batsch] and their use in genetic diversity analysis in peach and sweet cherry (Prunus avium L.). Theor Appl Genet 105: 127-138.

Ferree DC, Myers SC, Schupp JR (1992). Root pruning and root restriction of fruit trees - current review. Acta Hortic 322: 153-166.

Fideghelli C, Sartori A, Grassi F (2003). Fruit tree size and architecture. Acta Hort 622: 279-293.

Goncalves B, Correia MC, Silva PA, Bacelar AE, Santos A, Moutinho-Pereira MJ (2008). Leaf structure and function of sweet cherry tree (Prunus avium L.) cultivars with open and dense canopies. Sci Hortic 116: 381-387.

Hallé F, Oldeman RAA (1970). Essai sur l’architecture et dynamique de la croissance des arbres tropicaux. Paris, France: Masson and Co. (in French).

Hallé F, Oldeman RAA, Tomlinson PB (1978). Tropical Trees and Forests: An Architectural Analysis. Berlin, Germany: Springer-Verlag.

Hauck NR, Yamane H, Tao R, Iezzoni  A (2006).  Accumulation of nonfunctional S-haplotypes results in the breakdown of gametophytic self-incompatibility in tetraploid Prunus. Genetics 172: 1191-1198.

Jacyna T, Wood DSE, Trappitt MS (1989). Application of paclobutrazol and Promalin (GA4 + 7 + BAP) in the training of ‘Bing’ sweet cherry trees. New Zeal J Crop Hort 17: 41-47.

Kappel F (2002). Managing the self-fertile Lapins and Sweetheart sweet cherries in high density systems. Compact Fruit Tree 35: 75-77.

Kenis K, Keulemans J (2007). Study of the tree architecture of apple (Malus × domestica Borkh.) by QTL analysis of growth traits. Mol Breeding 19: 193-208.

Lang GA (2000). Precocious, dwarfing, and productive – how will new cherry rootstocks impact the sweet cherry industry? Horttechnology 10: 719-725.

Lang GA (2011). Producing first-class sweet cherries: integrating new technologies, germplasm, and physiology into innovative orchard management strategies. In: Proceeding of the 3rd Conference on “Innovations in Fruit Growing”, Belgrade, pp. 59-74.

Laurens F, Audergon J, Claverle J, Duval H, Germain E, Kervella J, Le Leyec M, Lauri P, Lespinasse J (2000). Integration of architectural types in Franch programmes of ligneous fruit species genetic improvement. Fruit 55: 141-152.

Ljubojević M (2012). Cherry genetic diversity in low vigorous vegetative rootstock breeding. PhD, University of Novi Sad, Novi Sad, Serbia.

Ljubojević M, Ognjanov V, Bošnjaković D, Barać G, Ognjanov M, Mladenović E, Čukanović J (2012). Sweet and sour cherry decorative forms. Genetika 44: 367-375.

Nei M, Li WH (1979). Mathematical model for studying genetic variation in terms of restriction endonucleases. P Natl Acad Sci USA 76: 5269-5273.

854

LJUBOJEVIĆ et al. / Turk J Agric For

Ognjanov V (2011). ‘Smaragd’ apple. HortScience 46: 952-954.

Ognjanov V, Ljubojević M, Ninić-Todorović J, Bošnjaković D, Barać G, Čukanović J, Mladenović E (2012). Morphometric diversity of dwarf sour cherry germplasm in Serbia. J Hortic Sci Biotech 87: 117-122.

Ognjanov V, Vujanic-Varga D, Gašic K (1998). Breeding columnar apples in Novi Sad. Acta Hortic 484: 207-210.

Pavlicek A, Hrda S, Flegr J (1999). Free-Tree freeware program for construction of phylogenic trees on the basis of distance data and bootstrap/jackknife analysis of the tree robustness. Application in the RAPD analysis of the genus Frankelia. Folia Biol 45: 97-99.

Predieri S (2001). Mutation induction and tissue culture in improving fruits. Plant Cell Tiss Org 64: 185-200.

Prusinkiewicy P, Remphrey W (2000). Characterization of architectural tree models using L-systems and Petri nets. In: L’arbre – The Tree 2000: Papers Presented at the 4th International Symposium on the Tree, pp. 177-186.

Robinson TL (2005). Developments in high density sweet cherry pruning and training systems around the world. Acta Hortic 667: 269-272.

Schuster M, Grafe C, Wolfram B (2014a). New results of sour cherry breeding in Germany. Acta Hortic 1020: 71-73.

Schuster M, Grafe C, Wolfram B, Schmidt H, (2014b). Cultivars resulting from cherry breeding in Germany. Erwerbs-Obstbau 56: 67-72.

Scorza R, Bassi D, Dima A, Rizzo M (1999). Developing new peach tree growth habits for higher density plantings. In: 42nd Annual IDFTA Conference, Hamilton, Canada, pp. 18–20.

Scorza R, Lightner GW, Liverani A (1989). The pillar peach tree and growth habit analysis of compact × pillar progeny. J Amer Soc Hort Sci 114: 991-995.

Segura V, Durel C, Costes E, (2009). Dissecting apple tree architecture into genetic, ontogenetic and environmental effects: QTL mapping. Tree Genet Genomes 5: 165-179.

Sooriyapathirana SS, Khan A, Sebolt AM, Wang D, Bushakra JM, Wang KL, Allan AC, Gardiner SE, Chagné H, Iezzoni AF (2010). QTL analysis and candidate gene mapping for skin and flesh color in sweet cherry fruit (Prunus avium L.). Tree Genet Genomes 6: 821-832.

Struss D, Ahmad R, Southwick SM, Boritzki M (2003). Analysis of sweet cherry (Prunus avium L.) cultivars using SSR and AFLP marker. J Am Soc Hortic Sci 128: 904-909.

Sussex ML, Kerk MN (2001). The evolution of plant architecture. Curr Opin Plant Biol 4: 33-37.

Tobutt RK (1985a). New approaches to breeding sweet cherry scion cultivars at East Malling, with particular reference to small tree size. Acta Hortic 169: 43-50.

Tobutt RK (1985b). Breeding columnar apples at East Malling. Acta Hortic 159: 63-68.

UPOV (2006). Sweet Cherry Guidelines for the Conduct of Tests for Distinctness, Uniformity and Stability. Geneva, Switzerland: International Union for the Protection of New Cultivars of Plants.

Webster AD (1995). Rootstock and interstock effects on deciduous fruit tree vigour, precocity, and yield productivity. New Zeal J Crop Hort 23: 373-382.

Webster AD, Quinlan JD (1984). Chemical Control of Shoot Growth. Report of East Malling Research Station for 1983. East Malling, UK: East Malling Research Station.

Whiting MD, Lang G, Ophardt, D (2005). Rootstock and training system affect sweet cherry growth, yield, and fruit quality. HortScience 40: 582-586.

Wu LR (1998). Genetic mapping of QTLs affecting tree growth and architecture in Populus: implication for ideotype breeding. Theor Appl Genet 96: 447-457.