Agro-Morphological Characterization and Assessment of Variability in Local Germplasm of Cucumis melo L. In Iran

Agricultural Engineering Research Journal 7(3): 30-35, 207 ISSN 228-3906 IDOSI Publications, 207 DOI: 0.5829/idosi.aerj.207.30.35 Agro-Morphological Characterization and Assessment of Variability in Local Germplasm of Cucumis melo L. In Iran 2 Majid Rashidi, Karim Arabsalmani and Behnam Zand Greenhouse Cultivation Research Department, Tehran Agricultural and Natural Resources Research and Education Center, AREEO, Varamin, Iran 2 Seed and Plant Improvement Research Department, Tehran Agricultural and Natural Resources Research and Education Center, AREEO, Varamin, Iran Abstract: The agro-morphological characterization is fundamental in order to provide information for genetic resources conservation and breeding programs. The present study was carried out to characterize ten local varieties of melon (Cucumis melo L.) selected by Tehran Agricultural and Natural Resources Research and Education Center, AREEO, Varamin, Iran. The study was carried out in a in a randomized block design with three replications. A descriptor list with 4 characters (nine quantitative and five qualitative) related to fruit and seed was adopted. Quantitative and qualitative data underwent analysis of variance and principal component analysis (PCA). With the traits retained in the PCA, a cluster analysis was performed using the Unweighted Pair Group Method with Arithmetic Mean (UPGMA). Significant differences were noted for the whole of the quantitative traits and high degree of polymorphism was observed for almost all of the qualitative characters. Cluster analysis separated the varieties in different groups. Key words: Cucumis melo L. Diversity Morphology Multivariate Analysis INTRODUCTION characters [7-9] and molecular markers such as isozymes [0, ], amplified fragment length polymorphism (AFLP), Melon (Cucumis melo L.) is the one of polymorphic random amplified polymorphic DNA (RAPD) [2, 3] and species among the Cucurbitaceae family []. The species simple sequence repeat (SSR) [4]. is generally known as melon, it is also called sweet melon, In Iran, melon is among the main vegetable crops muskmelon, casaba and cantaloupe [2]. Munger and grown and consumed, therefore of its economic Robinson [3] defined six groups of melon namely; importance. It ranks third after cucumber and tomato. In Cantaloupe, Inodorus, Flexuosus, Conomon, Chito, 20, 0447 ha were dedicated to this crop and its Dudaim and Momordica. A recent taxonomy of melon production amounted to 04482 tones [5]. However local identified 6 groups [4], five in subspecies agrestis melon genetic resources are currently being lost due to (Acidulous, Chinensis, Conomon, Makuwa and severe genetic erosion caused by the replacement of local Momordica) and in subspecies melo (Adana, Ameri, varieties by modern varieties and improper management Cantaloupensis, Chandalak, Chate, Chito, Dudaim, and inadequate regeneration procedures of germplasm Inodorus, Flexuosus, Reticulates and Tibish). The genus collections. Therefore, a survey of the genetic diversity is melon considered originated from Africa [5]. Recently, necessary to encourage rational management and phylogenetic data demonstrated that Cucumis originated selection programs involving the local Cucumis melo from Asia. It is located in tropical and subtropical regions germplasm. and is grown in temperate climate [6]. The aim of the present study was to determine the Cucumis melo is thought to contain the most diverse agro-morphological variation in ten local varieties of varieties in the genus Cucumis. Genetic diversity in this melon selected by selected by Tehran Agricultural and species has been analyzed using several morphological Natural Resources Research and Education Center, Corresponding Author: Dr. Majid Rashidi, Ph.D., Greenhouse Cultivation Research Department, Tehran Agricultural and Natural Resources Research and Education Center, AREEO, Varamin, Iran. E-mail: majidrashidi8@yahoo.com & m-rashidi@areeo.ac.ir. 30

Agric. Engineering Res. J., 7(3): 30-35, 207 AREEO, Varamin, Iran and to provide useful information to facilitate the choice of genitors for melon breeding program. MATERIALS AND METHODS Research Site: This study was carried out at the research site of Tehran Agricultural and Natural Resources Research and Education Center, Varamin, Iran. The research site is located at at latitude of 35 9 N and longitude of 5 39 E and is 000 m above mean sea level, in arid climate (50 mm rainfall annually) in the center of Iran, where the summers are dry and hot while the winters are cool. The soil of the research site is classified as an Aridisol (fine, mixed, active, thermic, typic haplocambids). Weather Parameters: The mean monthly rainfall and temperature of the research site during the year of study is given in Fig.. Soil Sampling and Analysis: To determine soil physical and chemical properties of the research site, a composite soil sample (from 2 points) was collected from 0-30 cm depth 30 days before planting during the study years. Soil samples were analysed in the laboratory for N, P, K, Fe, Zn, Cu, Mn, B, EC, ph, organic carbon (OC), particle size distribution and dry bulk density. Details of soil physical and chemical properties of the research site are given in Table. Plant Materials: Ten local varieties of melon (Cucumis melo L.) selected by Tehran Agricultural and Natural Resources Research and Education Center, AREEO, Varamin, Iran were used in this study. They consisted of Semsoori (SEM), Green Teel (GRT), Torogh Teel (TOT), Magasi (MAG), Atashi (ATA), Shahabadi (SHA), Dastjah (DASJ), Esfahan (ESF), Dastgerd (DASTG) and Abhine (ABH). Field Methods: The study was carried out in a randomized block design with three replications from April to September 203 at the research site of Tehran Agricultural and Natural Resources Research and Education Center, Varamin, Iran. The size of each plot was 0.0 m long and 7.50 m wide. A buffer zone of 2.50 m spacing was provided between plots. There were three furrows in each plot. The furrows had 0.0 m long, 75 cm wide and 50 cm depth and seeds were sown manually on the both sides of each furrow by keeping row to row and plant to plant distance 250 cm and 00 cm, respectively. Recommended levels of N (85 kg ha ), P (70 kg ha ) and K (30 kg ha ) were Table : Soil physical and chemical characteristics of the experimental site Soil characteristics Values Texture Clay-loam Sand (%) 24.0 Silt (%) 39.0 Clay (%) 37.0 EC (ds m ).70 ph 7.50 Organic carbon (%) 0.95 Available P (mg kg ) 40.4 Available K (mg kg ) 460 Available Fe (mg kg ) 2.84 Available Mn (mg kg ) 2.9 Available Zn (mg kg ).50 Available Cu (mg kg ).3 used as ammonitrate, phosphoric acid and potassium nitrate, respectively. Pest and weed controls were performed according to general local practices and recommendations. All other necessary operations except those under study were kept normal and uniform for all the treatments. Observation and Data Collection: Data was collected on 4 agro-morphological (nine quantitative and five qualitative) parameters related to fruit and seed according to the combined standards of descriptor lists of IPGRI (International Plant Genetic Resources Institute) [6] and UPOV (International Union for the Protection of New Varieties of Plants) for melon (Table 2). Observations were recorded on three randomly selected plants of each variety per replication. A digital caliper was used to measure length, diameter and thickness. Also, fruit and seed weights were measured by using a digital balance. Seed weight was determined for 00 seeds in three replications for each variety. Statistical Analysis: The data collected were analyzed using the computer software SAS 6.. Analysis of variance (ANOVA) was used to determine differences between varieties. Duncan s Multiple Range test at 5% probability was performed to compare the means of different treatments. Also, correlation between pairs of morphological characters was evaluated using Pearson s correlation coefficient. Multivariate relationships among varieties were revealed through a Principal Component Analysis (PCA) for quantitative characters and Factorial Correspondence Analysis (FCA) for qualitative characters. With the traits retained in the PCA and the dimensions obtained in the FCA, a cluster analysis was performed using the Unweighted Pair Group Method with Arithmetic Mean (UPGMA). This analysis was used to 3

Agric. Engineering Res. J., 7(3): 30-35, 207 Table 2: Agro-morphological traits used for local melon varieties characterization Descriptor / Trait Acronym Type Source State/Unit Fruit Fruit shape FS QL IPGRI Globular, 2 Flattened, 3 Oblate, 4 Elliptical, 8 Elongate Secondary fruit skin color SFC QL IPGRI 2 Light-yellow, 4 Pale green, 5 green, 6 Dark green, 7 Orange, 8 Brown Skin netted SN QL UPOV low, 2 medium, 4 High Fruit length FL QN IPGRI cm Fruit number FN QN UPOV number Fruit diameter FD QL UPOV cm Fruit weight FW QN IPGRI kg Flesh color FC QL IPGRI Number of female flower NFF QN UPOV number Ovary length OL QN IPGRI cm Flesh diameter FD QN UPOV cm Fruit sugar TSS QL UPOV % Seed Seed cavity diameter SCD QN UPOV cm Seed weight SW QN IPGRI G QL: qualitative QN: quantitative Rainfall Temperature 80.0 35.0 70.0 30.0 60.0 50.0 40.0 30.0 20.0 0.0 0.0 Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Rainfall (mm) 25.0 20.0 5.0 0.0 Temperature ( C) 5.0 0.0 Fig. : Mean monthly rainfall and temperature of the research site during the year of study study patterns of variance and relationships among accessions, where accessions with close genetic distances were placed in close proximity in the dendrogram. RESULTS AND DISCUSSION The collected data was extracted on the basis of 4 descriptors starting from ten melon varieties at a rate of three replicates per variety. Based on the measurements and morphological observations, the examined melon varieties showed a wide range of variability for almost all of the traits studied. Analysis of variance applied on quantitative characters (Table 3) showed that differences among cultivars for the whole studied characters were significant (P<0.0). The difference between landrace of cantaloupe was also significant. It is evident that diversity in population. There was non-significant difference between the ten cantaloupe landrace for number of fruit. This show that the Gene function related to hormonal changes at fruit set time of female flower in total populations was dominant and after the formation of 3 to 4 fruit, fruit set is prevented. The significant differences in number of female flower confirmed the gene function because despite the increasing number of female flowers on the different population, not increase the number of fruit. The Duncan s test at % revealed two groups of means depending on the descriptors (Table 4). Fruit weight (FW), Fruit length (FL), Skin netted (SN) and Flesh color (FC) were the most discriminating character. The genotype Semsoori consistently recorded significant 32

Agric. Engineering Res. J., 7(3): 30-35, 207 Table 3: Analysis of variance for quantitative and qualitative characters Characters MS C.V. (%) Fruit shape (FS).899 43.22 Secondary fruit skin color (SFC).64 47.40 Skin netted (SN) 2.068 3.5 Fruit length (FL) 6.344 6.47 Fruit diameter (FD) 7.064 2.72 Seed cavity diameter (SCD) 40.478 53.89 Total soluble solids (TSS ) 6.338 32.70 Flesh color (FC) 0.384 2.63 Seed weight (SW) 6.583 26.58 Flesh diameter (FD) 0.854 8.57 Fruit weight (FW) 0.82 33.6 Fruit number (FN) ns.927 33.98 Ovary length (OL) 3.274 9.97 Number of female flower (NFF) 263.456 23.3 ns = Non significant, = Significant at 5%, = Significant at % Table 4: Means comparison for quantitative traits in ten local melon varieties Population FS SFC SN FL FN FW NFF OL FD TSS SCD SW SEM 4 a 2.4 abc 4 a 5.63 ab 3.53 a.27 c 3.34 ab 3.34 b 3.0 e 5.73 a 0.59 abc 6.7 c GRT 2.4 b 2.66 ab.2 c 5.33 ab.76 c.43 c 3.57 ab 3.57 b 4.23 de 4.4 ab.02 abc 235.6 ab SHA.93 bc.5 c 3 ab 3.26 b 3 b 2.95 a 0.48 b 0.48 c 20.6 a 3. b 7.75 c 22.5 ab DASJ. c 2.23 abc.93 bc 7.64 a 3 b 2.29 ab 2.59 b 2.59 b 7.5 bc 4.9 ab 2.77 a 245.4 ab ABH.3 bc.5 c.5 bc 6.59 a 3.06 ab 2.69 ab 2.9 b 2.9 b 8.7 ab 4.2 ab 6.59 a 207.2 ab ESF.96 bc.43 c.43 ab 3.4 b 3 b 2.43 ab 9.87 b 9.87 c 9 ab 2.9 b 3.4 c 267. a MAG.73 bc 2.93 a 2.63 b 8.9 a 3. ab.98 bc 5.49 a 5.49 a 5.34 cd 4. ab 8.9 ab 24.8 ab DASTG.56 bc.36 c 2 bc 8.33 a 3.06 ab 2.84 a 3.39 ab 3.39 b 8.68 ab 3.66 ab 8.33 ab 245 ab TOT.53 bc 2.3 abc.83 bc 5.32 ab 2.93 b.57 c 3.22 ab 3.22 b 4.76 de 4.43 ab 5.32 abc 99. ab ATA c.76 bc 2 bc 2.85 b.56 c.36 c 0.4 b 0.4 c 4.2 de 5.33 a 2.85 bc 77.9 b Table 5: Pearson s correlation coefficients between pairs of quantitative characters SCD FD FL SW NF TSS NFF OL FW FD SCD 0.0 FD -0.04 0.0 FL 0.898-0.9 0.0 SW 0.34 0.89 0.30 0.0 NF 0.49 0.32 0.254-0.94 0.0 TSS 0.97 0.594 0.044-0.748-0.073 0.0 NFF 0.658 0.534 0.65-0.328 0.585 0.334 0.0 OL 0.84 0.970 0.080-0.756 0.42 0.578 0.680 0.0 FW 0.75-0.400 0.479 0.362-0.592 0.550-0.734-0.537 0.0 FD 0.834-0.42 0.780 0.329 0.257-0.054 0.430 0.222-0.480 0.0 = Significant at 5%, = Significant at % (P<0.0) differences from the others varieties in many Shahabadi presented the heaviest fruits (2.98 kg), the quantitative characters. It showed the highest value for biggest (20.6 cm) and the heaviest seeds (245.2 g); number of fruit (3.53) and fruit shape (4), but the lowest whereas Magasi had the longest ovary length (5.49 values for fruit weight (.27 kg), fruit diameter (3.0 cm), mm). Significant differences between varieties were also seed weight (.04 kg) and seed weight (6.7 g). However, reported in previous data [7]. Generally, the highest rates Dastjah, Dastgerd, Esfahan and Shahabadi were obtained for Semsoori and Magasi genotypes. presented the highest average values for skin color (2.9), Relations between quantitative traits were expressed (2.6), (2.4) and (2.2), respectively and the lowest average by correlation matrix (Table 5). According to this table, values for TSS (3 to 3.5). Whereas Semsoori had the six morphological features were significantly correlated lowest value for fruit diameter (3.0 cm) and the highest at 5% or % probability level. The main positive value for fruit number (3.53) and TSS (5.73 %). correlation appeared as follows: fruit weight with fruit 33

Agric. Engineering Res. J., 7(3): 30-35, 207 Fig. 2: Dendrogram obtained from cluster analysis of local melon varieties diameter (r=0.89), seed cavity diameter with fruit length couple and Magasi. According to Pitrat et al. [9], (r=0.898) and fruit diameter (r=0.834); ovary length with Cucumis melo var. flexuosus or Snake melon has a low fruit diameter (r=0.970); number of fruit with number of similarity with each other varieties of melon and female flower (r=0.585). On the other hand, strong represented other group of melon. negative correlation was detected between number of female flower with fruit weight (r=-0.95), fruit diameter CONCLUSION (r=-0.87), seed weight (r=-0.83) and seed size (r=-0.70); peduncle length with fruit weight (r=-0.66) and fruit Agro-morphological traits considered in this study diameter (r=-0.6). Negative correlation between leaves showed a large variability in ten local melon varieties. development and fruits development was also detected in Results obtained could be used to establish a catalogue tomato [8] suggesting a competition for assimilates of local melon varieties. Further studies involving between vegetative parts and reproductive ones. molecular markers could be very promising. A cluster dendrogram (Fig. 2) combining quantitative and qualitative characters was carried out in order to ACKNOWLEDGEMENT study the general pattern of variance and to establish relationship among the ten melon varieties. At an average The financial support provided by the Agricultural distance of.0, hierarchical clustering process leads to Research, Education and Extension Organization of Iran two major groups. The first group (G I) included under research award number 07-2-79-204 is gratefully Semsoori, Green Teel, Torogh Teel and Atashi acknowledged. whereas the second group (G II) contained the six varieties Shahabadi, Dastjah, Esfahan, Dastgerd, Magasi REFERENCES and Abhine. The nearest varieties of first group in the dendrogram are Green Teel and Torogh Teel with. Decker-Walters, D.S., J.E. Straub, S.M. Chung, Semsoori being close to them and followed by Atashi E. Nakata and H.D. Quemada, 2002. Diversity in in a larger distance. The nearest varieties of second group free-living populations of Cucurbita pepo are Shahabadi with Dastgerd and also Dastjah with (Cucurbitaceae) as assessed by random amplified Esfahan but first couple are larger distance with second polymorphic DNA. Systematic Botany, 27(): 9-28. 34

Agric. Engineering Res. J., 7(3): 30-35, 207 2. Nayar, N.M. and R. Singh, 998. Taxonomy, 2. Lopez-Sese, A.I., J.E. Staub and M.L. Gomezdistribution, ethnobotanical uses in Cucurbits Guillamon, 2003. Genetic analysis of Spanish melon (Nayar, N.M. and T.A. More, eds.). Science (Cucumis melo L.) germplasm using a standardized Publishers Inc., U.S.A., pp: -8. molecular marker array and reference accessions. 3. Munger, H.M. and R.W. Robinson, 99. Theoretical and Applied Genetics, 08 (): 4-52. Nomenclature of Cucumis melo L. Cucurbit Genetics 3. Nhi, P.T.P., Y. Akashi, T.T.M. Hang, K. Tanaka, Cooperative Reports, 4: 43-44. Y. Aierken, T. Yamamoto, H. Nishida, C. Long and 4. Pitrat, M., P. Hanelt and K. Hammer, 2000. Some K. Kato, 200. Genetic diversity in Vietnamese melon comments on infraspecific classification of cultivars landraces revealed by the analyses of morphological of melon. Acta Horticulturae, 50: 29-36. traits and nuclear and cytoplasmic molecular markers. 5. Kerge, T. and M. Grum, 2000. The origin of melon, Breeding Science, 60: 255-266. Cucumis melo: A review of the literature. In: the 7th 4. Monforte, A.J., J. Garcia-Mas and P. Arus, 2003. EUCARPIA Meeting on Cucurbit Genetics & Genetic variability in melon based on microsatellite Breeding, pp: 37-44. variation. Plant Breeding, 22: 53-57. 6. Pech, J.C., A. Bernadac, M. Bouzayen, A. Latch, 5. FAOSTAT, 203. FAO Statistical Yearbook. Food C. Dogimont and M. Pitrat, 2007. Biotechnology in and Agriculture Organization of the United Nations. agriculture and forestry, Vol. 60. Transgenic crops V. Available online at http://www.fao.org. In: Pua E.C. and M.R. Davey (eds.) Melon, Springer- 6. IPGRI, 2006. Descriptors for mango (Mangifera Verlag, Berlin, Heidelberg, pp: 209-240. indica L.). International Plant Genetic Resources 7. Laghetti, G., R. Accogli and K. Hammer, 2008. Institute, Rome, Italy. Different cucumber melon (Cucumis melo L.) races 7. Henan, I., I. Tlili, R. Ilahy, T. Rhim and H. Jebari, 203. cultivated in Salento (Italy). Genetic Resources and Evaluation of qualitative parameters and Crop Evolution, 55(4): 69-623. physicochemical properties of local varieties of 8. Escribano, S. and A. Lazaro, 2009. Agro- muskmelon (Cucumis melo L.) grown in Tunisia. morphological diversity of Spanish traditional melons Global Science Books, 7: 7-2. (Cucumis melo L.) of the Madrid provenance. Genetic 8. Gautier, H., S. Guichard and M. Tchamitchian, 200. Resources and Crop Evolution, 56: 48-497. Modulation of competition between fruits and leaves 9. Nasrabadi, H.N., H. Nemati, A. Sobhani and by flower pruning and water fogging and M. Sharifi, 202. Study on morphologic variation of consequences on tomato leaf and fruit growth. different Iranian melon cultivars (Cucumis melo L.). Annals of Botany, 88 (4): 645-652. African Journal of Agricultural Re., 7(8): 2764-2769. 9. Pitrat, M., M. Chauvet and C. Foury, 2000. Diversity, 0. Akashi, Y., N. Fukuda, T. Wako, M. Masuda and history and production of cultivated cucurbits. Acta K. Kato, 2002. Genetic variation and phylogenetic Horticulturae, 492: 24-250. relationships in East and South Asian melons, Cucumis melo L., based on the analysis of five isozymes. Euphytica,.25: 385-396.. McCreight, J.D., J.E. Staub, A.I. Lopez-Sese and S. Chung, 2004. Isozyme variation in Indian and Chinese melon (Cucumis melo L.) germplasm collection. Journal of the American Society for Horticultural Science, 29(6): 8-88. 35