Genetic variability and selection in natural populations of vineyard peach (Prunus persica ssp. vulgaris Mill.) in the Krusevac region (Central Serbia)

Variabilidad genética y selección en poblaciones naturales de durazno nativo de los balcanes (Prunus persica ssp. vulgaris Mill.) en la región Krusevac (Serbia Central)

Tomo Milosevic¹*, Nebojsa Milosevic²

¹ Faculty of Agronomy, Department of Fruit Growing & Viticulture, Cara Dusana 34, 32000 Cacak, Serbia, *Autor responsable: (tomom@tfc.kg.ac.rs).

² Fruit Research Institute, Department of Pomology and Fruit Breeding, 32000 Cacak, Kralja Petra 1/9, Serbia, Email: (mnebojsa@tfc.kg.ac.rs).

Received: December, 2008.
Approved: March, 2010.

ABSTRACT

In situ selection of vineyard peach (Prunus pérsica spp. vulgaris Mill.) genotypes was conducted in 2006 and 2007 in the region of Kfusevac (Central Serbia). Identification, observation and recording of the phenological and pomological properties of the highest–quality genotypes were carried out. Twenty–eight genotypes of the highest quality were selected out of 2093 genotypes, based on biological and pomological properties, and compared to those of cv. Redhaven (control). The onset of flowering varied (April 7–21), being early in two genotypes and late in one genotype, as compared to Redhaven, while 25 genotypes showed an intermediate flowering date, as Redhaven did. Campanula flowers were found in six genotypes, as in Redhaven, and the rosaceous type in 22 genotypes. All genotypes showed late harvest time in relation to Redhaven. Fruit weight and stone weight ranged between 10.23 ±0.5 to 78.03±3.6 g, and 1.91 ±0.1 to 5.40±0.9 g. Fruits were predominantly rounded, while the stones were ovoid. The selected vineyard peach genotypes showed favorable phenological and pomological properties that can be used in further breeding programs.

Key words: breeding, flowering, fruit and stone, genetic diversity, genotype.

RESUMEN

Se realizó la recolección in situ de genotipos de durazno nativo de los Balcanes (Prunus pérsica spp. vulgaris Mili.) en 2006 y 2007, en la región Krusevac (Serbia Central). Se efectuó la identificación, la observación y el registro de las propiedades fenológicas y pomológicas de los genotipos de más alta calidad. De 2093 genotipos se seleccionaron 28 de la más alta calidad con base en las propiedades biológicas y pomológicas, para compararlos con la la cv. Redhaven (testigo). El inicio de la floración varió (7–21 de abril), siendo temprana en dos genotipos y tardía en un genotipo, comparado con Redhaven, mientras que 25 genotipos mostraron una fecha de floración intermedia, como Redhaven. Se encontraron flores campanuladas en seis genotipos, como en Redhaven, y de tipo rosáceo en 22 genotipos. Todos los genotipos tuvieron un periodo de cosecha tardío respecto a Redhaven. El peso de los frutos y de los huesos osciló entre 10.23±0.5 y 78.03±3.6 g, y 1.91±0.1 y 5.40±0.9 g. Los frutos fueron predominantemente redondos, mientras que los huesos fueron ovoides. Los genotipos de durazno nativo seleccionados mostraron propiedades fenológicas y pomológicas favorables, y se pueden usar en programas de reproducción.

INTRODUCTION

In Serbia, vineyard peaches (Prunus pérsica spp. vulgaris Mill.) are primarily cultivated in old vineyards which have steadily become less numerous. Intensive management in new vineyards have led to a significant reduction in the vineyard peach populations. Thus, there is a need for protection and genetic conservation of the remaining vineyard peach as a source of germplasm by establishing collection plantings as gene banks in order to develop new cultivars. Research programs are conducted in a few stages to achieve preservation of genotypes to develop new cultivars and rootstocks, increasing disease and pest resistance and using fruits either for fresh consumption or for processing (Xu et al., 2006; Moreno and Gogorcena, 2007; Pandey et al., 2008). Native peach populations in the Balkan Peninsula are a valuable source of germplasm made up of seed–propagated genotypes. The Balkan Peninsula is a secondary center of peach gene divergence due to the genetic variability of germplasm, environmental diversity and human activity (Vujanic–Varga and Ognjanov, 1992).

Selection of vineyard peach genotypes in Serbia for biodiversity conservation has been investigated by Paunovic et al. (1998), Gasic and Ognjanov (1999) and Gasic et al. (2001). The directional selection of vineyard peach genotypes has also been investigated. Milutinovic et al. (1994) performed selection for pomological properties, and Vujanic–Varga et al. (1996) for breeding genotypes, suitable for juice production. Fungi and virus resistance of vineyard peach genotypes has been studied by Paunovic et al. (1992) and Ognjanov et al. (2000); others have focused in the selection of genotypes suitable to be used as rootstocks for peach cultivars (Misic et al., 1990; Paunovic et al., 1992; Ognjanov et al., 1996a). Selection of superior vineyard peach genotypes having a variety of uses, primarily as rootstocks of peach and nectarine cultivars was performed by Paunovic et al. (1992). The objective of the present study was to identify vineyard peach genotypes characterized by positive biological and pomological traits in situ for further selection, collection and orchard establishment for germplasm conservation purposes.

MATERIALS AND METHODS

Study area and plant material

In 2006 and 2007 all accessions were collected between the towns of Trstenik (43° 37' N, 20° 59' E, altitude 180–230 m, municipality area 448 km²) and Krusevac (43° 30' N; 21° 25' E, altitude 290–330 m, municipality area 854 km²), in the Krusevac vineyard region. Also, in situ identification and examination of vineyard peach genotypes were conducted. Out of the 2093 genotypes identified in the vineyards, 493 trees were selected, singling out 28 genotypes for further studies and they were compared to cv. Redhaven (control). A single–plant selection method was used. When selecting trees for seed utilization, their health, fruit–bearing potential, maturation dates, fruit size and stone weight were emphasized. The trees had to be sound and fruit–bearing, and fruit maturation as late as possible from mid–August to the end of September (small fruits, small stones and good stone–flesh separation). When selecting genotypes for fruit consumption, their sanitary status, large fruit size, high flesh/ stone ratio and specific organoleptic traits (skin ground color, red over color, flesh color) were emphasized. The date of maturity and stone–flesh separation was not relevant.

Experimental procedure and analysis of phenological, pomological and organoleptic traits

Skin ground color, red over color and flesh color were measured using a Minolta chroma meter (CR–300, Minolta, Ramsey, NJ) tristimulus color analyzer calibrated to a white porcelain reference plate. In addition, a trained panel of five experts classified the peaches visually, according to the perception of the skin, red over and flesh colors.

The described fruit, stone and flower properties were categorized according to IBPGR descriptors (Bellini et al. 1984): 1) time of flowering (TF) with nine categories: 1=extremely early, through 9=extremely late; 2) flower type (FT): 1=rosaceous, 2=campanulate; 3) date of maturity (DM) with nine categories: 1=extremely early, through 9=extremely late; 4) fruit size (FS): 1=extremely small, 3=small, 5=intermediate, 7=large, 9=extremely large; 5) fruit shape (FSh): 1=very flat, 2=slightly flat, 3=rounded, 4=ovate, 5=oblong, 6=elongated; 6) ground color (GC) of the skin of fully mature fruit: 1=green, 2=greenish–cream, 3=cream, 4=cream–yellow, 5=yellow, 6=orange–yellow; 7) red over color (ROC): 0=no red over color, 1=none to red–trace, 2=red–trace, 3=red striped, 4=red–mottled, 5=partly–red, 6=medium–red, 7=mostly–red, 8=full–red, 9=red–wine; 8) flesh color (FC): 1 = white–greenish, 2=white, 3=white–cream, 4=yellow–greenish, 5=yellow, 6=yellow–orange, 7=yellow–red, 8=red; 9) stone size (SS): 1=extremely small, 3=small, 5=medium, 7= large, 9=extremely large; 10) stone shape (SSh): 1=flat, 2=rounded, 3=ovoid, 4=elongated, 5=very elongated; 11) stone adherence (SA) to flesh of fully ripe fruit: 1=freestone, 2=semi–freestone, 3 = clingstone.

Statistical analysis

Means of fruit and stone weight were compared by LSD tests (p < 0.05) for all the variable using the program Statistica (StatSoft Inc., Tulsa, Oklahoma, USA), involving the performance of oneway analyses of variance.

The relationship between some phenological, pomological and organoleptic traits was evaluated by Pearson's correlation (p<0.05). In addition, a principal component analysis (PCA) was performed to evaluate relationships among genotypes and among variables using the PRINCOMP procedure of SAS (SAS Institute Inc., North Carolina, USA).

RESULTS AND DISCUSSION

Evaluation of phenological traits

Results on the onset of flowering and fruit maturity in the selected vineyard peach genotypes are shown in Table 1. The TF was from April 7^th to 25th; the earliest flowering was recorded with the FA G 1 (April 7–11) and FA G 10 genotypes (April 8–12), and the latest flowering with the FA G 24 genotype (April 21–25). According to Bellini et al. (1984), Redhaven is a cultivar with intermediate TF. As compared to Redhaven, FA G 1 and FA G 10 exhibited an early TF, and FA G 24 a late one; the remaining 25 genotypes were medium–flowering, as Redhaven (Milosevic, 1996). Genotypes FA G 2, FA G 11, FA G 16, FA G 20, FA G 24 and FA G 25 produced campanula flowers, as Redhaven did, whereas the remaining ones gave rosaceous flowers. In 12 vineyard peach genotypes selected in the former Yugoslavia for juice production, the TF was between April 24^th and April 30^th, similar to the Redhaven, as they belong to intermediate–flowering genotypes (Vujanic–Varga et al., 1996). The same authors reported that only two genotypes produced campanula flowers, the remaining 12 giving rosaceous ones. Similar flowering and flower–type tendencies were observed in our study.

Evaluation of fruit physical attributes

Results on fruit and stone traits of vineyard peach are shown in Table 2. There were significant differences (p<0.05) among accessions regarding FW, which ranged from 10.23 ±0.5 (FA G 16) to 78.03±3.6 g (FA G 20). Fruit weight was less than 30.0 g for 15 genotypes, between 30.0 and 60.0 g for nine genotypes, and higher than 60.0 g for four genotypes. The largest fruits were found with genotypes FA G 24 (73.13±3.3 g), FA G 11 (76.90±3.1 g) and FA G 20 (78.03±3.6 g), while the smallest ones with the genotype FA G 16 (10.23±0.5 g), followed by FA G 17 (15.84±0.8 g) and FA G 19 (16.18±0.9 g). The genotypes in our study were categorized as having extremely small fruits, the genotypes FA G 24, FA G 11 and FA G 20 being the only ones having small fruits (Bellini etai, 1984; Zanetto et al., 2002). The FW of peach genotypes from Mt. Fruska Gora intended for consumption and of those for seed utilization was between 66.0 to 93.0 g and 22.0 to 34.0 g (Vujanic–Varga et al., 1988). Besides, the FW ranged from 51.2 to 230.0 g in 12 genotypes for juice production (Vujanic–Varga et al., 1996), and 32.0 to 114.5 g in 24 genotypes intended for generative rootstock production (Ognjanov et al., 1996a). According to Paunovic et al. (1992), the FW of 15.84 to 68.70 g was determined in 35 genotypes in the Krusevacka Zupa region, whereas Paunovic (1963) observed that the largest–fruit genotype produced a FW of 154.0 g. The analysis of the results in our study suggests that the small–fruit vineyard peach genotypes and large–fruit ones showed characters for seed exploitation aimed at rootstock development and for fresh consumption or processing, which is in accordance with studies carried out on peach (Misic et al., 1990; Paunovic, 1963; Paunovic et al., 1992).

In the genotypes, rounded FSh was dominant (14 genotypes) over slightly flat (one genotype), ovate (six genotypes) and oblong (seven genotypes). The dominant FSh in the vineyard peach genotypes selected to be used as rootstocks was rounded (Ognjanov et al., 1996a), and that in the genotypes selected for juice production ranged from slightly flat, rounded and ovate to oblong (Vujanic–Varga et al., 1996). The dominant GC, ROC and FC was yellow, without red over color and white (Table 2), which is in agreement with studies on peach (Paunovic et al., 1992, 1998).

There were significant differences among accessions concerning the SW which ranged from 3.01 ±0.5 to 3.92±0.5 g in 10 genotypes, 2.20±0.2 to 2.94±0.4 g in seven genotypes and 4.00±0.7 to 4.85±0.7 g in six genotypes (Table 2). The smallest stones were found with FA G 1 (1.91 ±0.1 g) and FA G 3 (1.96±0.2 g); while the largest were FA G 20 (5.12±0.8 g), FA G 24 (5.21 ±0.9 g) and FA G 11 (5.40±0.9 g). The most predominant SSh was ovoid in 23 genotypes (Table 2); genotypes FA G 11, FA G 20, FA G 24, FA G 26 and FA G 28 displayed an elongated stone shape. Regarding the stone–flesh separation, it was only in genotypes FA G 7 and FA Gil that the stone was adhered to the flesh; in the remaining 26 genotypes the separation occurred. In 24 vineyard peach genotypes intended to be used as rootstocks, the average SW was 2.9 to 8.8 g (Ognjanov et al., 1996a). These authors further reported that the SSh varied: rounded in four genotypes, ovoid in 14, elongated in two, very elongated in one. All genotypes were characterized by stone–flesh separation, and there was higher seed germination in small–stone genotypes. The SW in 14 peach genotypes coming from Mt. Fruska Gora (Vojvodina) and selected to be used as rootstocks varied between 2.4 and 4.0 g (Vujanic–Varga et al., 1988), and the stone weight in 35 genotypes was 1.84 to 5.92 g (Paunovic et al., 1992). In 12 genotypes from the former Yugoslavia and selected for juice production, the average SW was 3–9 to 9–6 g, the most common SSh being ovoid and the stone separating from the flesh in all genotypes (Vujanic–Varga et al., 1996). For rootstock production, vineyard peach genotypes to be selected should produce fruits from 20.0 to 50.0 g average weight, medium late to late maturity, good separation of the stone from the flesh, and white–fleshed (Misic, 1984).

Evaluation of organoleptic attributes

The GC was yellow in 18 genotypes, creamy yellow in six, creamy in two and greenish–creamy in two (Table 2). Twenty selected genotypes displayed no red color over the skin, four none to red trace, one a trace of red, and three a partly red trace. The FC was white in 22 genotypes, white–cream in one, yellow–greenish in two and yellow in three. Out of the 35 selected genotypes, nine were white flesh, and seven yellow (Paunovic et al., 1992). Of the 12 genotypes selected for juice production, 11 were yellow–fleshed and only one red–fleshed (Vujanic–Varga et al., 1996). In our study, the largest–fruit genotypes were yellow (FA G 24), yellow–greenish (FA G 11) and white–fleshed (FAG 20).

Correlations and principal component analysis

Significant correlations were found between the set of nine phenological and pomological traits of genotypes (Table 3). Results show a high correlation between FW and FC (0.732), which can be used for early selection of large fruit genotypes. A moderate significant correlation was detected between FW and FC (0.463), FW and TF (0.488), FC and SA (0.409), SW and SA (0.462) and SW and TF (0.440). Time of flowering and DM were not significantly correlated, which implies that the number of days from blossom to maturity is highly variable in these genotypes. Low correlations between fruit traits and date of maturity were found (Table 3): FS and DM (–0.017), GC and DM (0.361) and FC and DM (0.147). Opposite to our results, Badenes et al. (1998) reported that ripening time correlated with fruit weight, which may be explained by differences in the plant material and in the size of the group of genotypes studied. Ground color was not correlated with FC (–0.039) in our study, which was in agreement with Lewallen and Marini (2003) who reported that for peach GC does not seem to be a good indicator of FC and fruit firmness, because fruit with the same hue angle showed greatly–differing firmnesses.

Principal components analysis (PC) is used to establish genetic relationships among genotypes and to study correlations among fruit and organoleptic attributes and phenological characteristics within sets of peach genotypes (Iezzoni and Pritts, 1991; Badenes et al., 1998; Wu et al., 2003). The first two principal components (PC) of the accessions accounted for 48.99 % of the total variance among genotypes (Table 4).

Fruit weight, flesh color, stone weight, stone adherence and time of flowering (on PC1), ground color and data of maturity (on PC2) and fruit size and red over color (on PC3) explained the largest portion of the variance (Table 5). Higher positive values for PC1 represent genotypes with larger fruit and stone and later time of flowering (Figure 1). This group included genotypes FA G 11, FA G 15, FA G 20, FA G 22, FA G 23, FA G 24 and FA G 25. Genotypes with higher positive PC2 values were characterized by later time of flowering and yellow ground color. Genotypes such as FA G 1, FA G 2, FA G 3, FA G 6, FA G 8, FA G 9, FA G 17, FA G 18, GA G 19, FA G 21 and FA G 27, integrate a different group as shown in Figure 1. Higher positive PC3 values represent genotypes with red over color. This group comprises genotypes FA G 4, FA G 5, FA G 10, FA G 12, FA G 14 and FA G 28 (Figure 1).

Genetic variability of peach is due to the evolutionary and geographic origin of the germplasm, the physiographic variation of the collection sites, and seed movements due to diverse biotic and abiotic factors (Morales–Nieto et al., 2006). Greater genotypic variation in our study suggests the existence of genetic potential readily available to develop peaches with larger fruit, smaller stone, different ground color and presence of red over color (Wu et al., 2003). Also, there is genetic potential to develop peaches with later time of flowering (Badenes et al., 1998). These genotypes showing valuable biological and pomological attributes can be immediately shared with the farmers and the breeders community.

CONCLUSIONS

The peach genotypes selected in the region of Krusevac (Central Serbia) showed substantial variability in terms of the tested attributes. Most of the evaluated genotypes show a high potential to solve some of the important problems in peach production for fresh consumption and fruit juice.

Many genotypes produced extremely small fruits and stone. The most dominant fruit shape, skin ground color, red over color and flesh color were rounded, yellow, without red color and white.

Significant correlation was found among some peach physical and organoleptic attributes and phonological characteristics, which could reduce the number of pomological traits to be studied in peach germplasm. The PCA was successfully used for grouping genotypes according to similar phenological and pomological traits.

The authors are grateful to Ms. Jelena Krstic, Faculty of Agronomy, Cacak, for her translation of the paper into English.

LITERATURE CITED

Badenes, M.L., J. Martínez–Calvo, and G. Llácer. 1998. Estudio comparativo de la calidad de los frutos de 26 cultivares de melocotonero de origen norteamericano y dos variedades población de origen español. Invest. Agr. Prod. Prot. Veg. 13: 56–70.        [ Links ]

Bellini, E., R. Watkins, and E. Pomarici. 1984. Descriptor list for peach (Prunus persica). International Board for Plant Genetic Resources, Rome (Italy); Commission of the European Communities, Brussels (Belgium); Committee on Disease Resistance Breeding and Use of Genebanks, IBPGR Secretariat. Rome, Italy. 34 p.        [ Links ]

Gasic, K., and V. Ognjanov. 1999. Genetic polymorphism in vineyard peach germplasma. Plant Genet. Res. Newsl. 5: 25–27.        [ Links ]

Gasic, K., V. Ognjanov, R. Boskovic, K.R. Tobutt, and C. James. 2001. Characterisation of vineyard peach biodiversity. Acta Hort. 546: 119–125.        [ Links ]

Graselly, C. 1985. Selection of peach seedling rootstock. Acta Hort. 173: 245–249.        [ Links ]

Iezzoni, A.F., and M.P. Pritts. 1991. Applications of principal component analysis to horticultural research. HortSci. 26: 334–338.        [ Links ]

Lewallen, K.S., and R.P. Marini. 2003. Relationship between flesh firmness and ground color in peach as influenced by light and canopy position. J. Am. Soc. Hort. Sci. 128: 163–170.        [ Links ]

Milosevic, T. 1996. Flowering and ripening of peach in the ecological conditions of Cacak. Jug. Vocarstvo 30: 291–296 (in Serbian).        [ Links ]

Milutinovic, M., G. Surlan–Momirovic, D. Nikolic, M. Milutinovic, and V. Rakonjac. 1994. Investigation of pomological characteristics of vineyard peach. In: X Yugoslav Conference of Peach Improvement and Processing, Grocka. Proceedings of Scientific Papers. 10: 23–28 (in Serbian).        [ Links ]

Misic, D.P. 1984. Fruit Rootstocks. Nolit, Belgrade. 208 p (in Serbian).        [ Links ]

Misic, D.P. 1987. General Fruit Breeding. Nolit, Belgrade. 270 p (in Serbian).        [ Links ]

Misic, D.P., Z.V. Pavlovic, R.R. Todorovic, and A.M. Mirkovic 1990. Evaluation of vineyard peach as a peach rootstock. Fruit Varieties J. 44: 99–101.        [ Links ]

Morales–Nieto, C, A. Quero–Carrillo, O. Le–Blanc, A. Hernández–Garay, J. Pérez–Pérez, and S. González–Muñoz. 2006. Caracterización de la diversidad del pasto nativo Bouteloua curtipendula Michx. Torr. mediante marcadores de AFLP. Agrociencia 40: 711–720.        [ Links ]

Moreno, A.M., and Y. Gogorcena. 2007. Breeding, selection and conservation of peach and Prunus genetic resources. J. Amer. Soc. Hort. Sci. 132: 670–674.        [ Links ]

Ognjanov, V., M. Vojnovic, D. Vujanic–Varga, K. Gasic, M. Krstic, and V. Jankovic–Dozet. 1996a. Selection of vineyard peach genotypes suitable as seedling rootstock. Jug. Vocarstvo 30: 123–128 (in Serbian).        [ Links ]

Ognjanov, V, D. Vujanic–Varga, and K. Gasic. 1996b. Vineyard peach suitable for processing. Plant Genet. Res. Newsl. 2: 28–29.        [ Links ]

Ognjanov, V, D. Vujanic–Varga, K. Gasic, and B. Nadj. 2000. Disease resistance in apple, pear and peach germplasm originating from the Balkan Peninsula. Acta Hort. 513: 63–68.        [ Links ]

Pandey, A., E. Roshini–Nayar, K. Venkateswaran, and CD. Bhandari. 2008. Genetic resources of Prunus (Rosaceae) in India. Genet. Res. Crop Evol. 55: 91–104.        [ Links ]

Paunovic, A.S. 1963. Seed germination of deciduous fruit trees grown in Yugoslavia. Acta Hort. 3: 5–10.        [ Links ]

Paunovic, A.S., A.S. Paunovic, T.M. Milosevic, M.T. Tisma, and A. Obradovic. 1992. Selection of native vineyard peach germplasm. Acta Hort. 315: 133–140.        [ Links ]

Paunovic, A.S., K. Gasic, E. Mratinic, M. Nikolic, D. Ogasanovic, V. Ognjanov, M. Stanisavljevic, L. Rados, and M. Radulovic 1998. Fruit gene bank in Yugoslavia: Genetic resources and possibility fruit germplasm conservation. Jug. Vocarstvo 30: 39–50 (in Serbian).        [ Links ]

Vujanic–Varga, D., V Ognjanov, D. Lalic, and A. Horvat. 1988. Investigation of vineyard peach population in Fruska Gora. Jug. Vocarstvo 22: 137–142 (in Serbian).        [ Links ]

Vujanic–Varga, D., and V. Ognjanov. 1992. Conservation of vineyard peach populations in Yugoslavia. Plant Genet. Res. Newsl. 90: 28–30.        [ Links ]

Vujanic–Varga, D., M. Krstic, M. Vojnovic, K. Gasic, and V. Jankovic–Dozet. 1996. Selection of vineyard peach suitable for juice production. Jug. Vocarstvo 30: 129–136 (in Serbian).        [ Links ]

Wu, B., B. Quilot, J. Kervella, M. Génard, and S. Li. 2003. Analysis of genotypic variation of sugar and acid contents in peaches and nectarines through the principle component analysis. Euphytica 132: 375–384.        [ Links ]

Xu, H.D., S. Wahyuni, Y. Sato, M. Yamaguchi, H. Tsunematsu, and T. Ban. 2006. Genetic diversity and relationships of Japanese peach (Prunuspérsica L.) cultivars revealed by AFLP and pedigree tracing. Genet. Res. Crop Evol. 53: 883–889.        [ Links ]

Zanetto, A., L. Maggioni, K.R. Tobutt, and F. Dosba. 2002. Prunus genetic resources in Europe: achievement and perspectives of a networking activity. Genet. Res. Crop. Evol. 49: 331–337.        [ Links ] ]]> 1998 13 56-70 1984 34 1999 5 25-27 2001 546 119-125 1985 173 245-249 1991 26 334-338 2003 128 163-170 1996 30 291-296 1994 10 Grocka 23-28 1984 208 1987 270 1990 44 99-101 2006 40 711-720 2007 132 670-674 1996 a 30 123-128 1996 b 2 28-29 2000 513 63-68 2008 55 91-104 1963 3 5-10 1992 315 133-140 1998 30 39-50 1988 22 137-142 1992 90 28-30 1996 30 129-136 2003 132 375-384 2006 53 883-889 2002 49 331-337