Tomato is one of the most popular vegetables grown worldwide with production of 126.24 million tonnes. In India, it occupies an area of 0.54 million ha with production of 8.85 million tonnes and average yield of 16.39 t/ha (FAO, 2007). Plant growth characteristics range from indeterminate to highly determinate types. Branches in the indeterminate types keep growing and producing fruits until adverse temperatures kill the plant. Earlier, tomato cultivation was season-bound. Production scenario has changed in the last few years. Nowadays, tomatoes are grown round the year. However, in the North-Western parts of hot, arid regions of India, tomato production and productivity is limited by various biotic and abiotic stresses. In the recent past, despite extreme temperature regimes prevalent and environmental stresses, tomato cultivation has been gaining localized popularity. The main reasons are remunerative prices, availability of hybrid seeds from the private sector to growers, and exploitation of water sources by resourceful farmers. Yield potential is very low due to extremes of environmental in the hot, arid region of Western Rajasthan, thereby upsetting economics of the crop both ways, i.e., low market-price of the produce due to glut, and non- realization of full potential of hybrids owing to environmental susceptibility. In tomato, when the ambient temperature exceeds 35°C, seed germination, seedling and vegetative Short communication J. Hortl. Sci. Vol. 7(2):194-198, 2012 Variability and association studies in tomato germplasm under high-temperature arid region Hanif Khan and D.K. Samadia Central Institute for Arid Horticulture, Bikaner-334 006, India E-mail: hanif.gene@gmail.com ABSTRACT Genetic variability and correlation studies were carried out for 12 traits in 23 genotypes of tomato grown during spring-summer of 2009 under hot, arid agro-climatic conditions. Genetic variability, heritability and expected genetic advance revealed significant differences for all the traits studied. PCV and GCV were high for fruit weight, number of fruits per plant, plant height and fruit yield per plant. High heritability with high genetic advance as percentage of mean was observed for yield per plant (93.2%) as also for average fruit weight (92.8%), number of fruits per plant (73.4%) and plant height (50.1%) indicating the role of additive gene effects and for effectiveness of selecting for these traits. Correlation studies revealed that fruit yield had significant positive correlation with fruit weight, fruit length, fruit diameter and number of fruits per plant, both at the genotypic and phenotypic levels, indicating mutual association of these traits. Negative correlation of days to flowering and days to first harvest on yield per plant suggested indirect selection for earliness for yield improvement. Key words: Tomato, arid environment, germplasm, variability growth, flowering, fruit set and fruit ripening are adversely affected (Miller et al, 2001). In this part of the country, tomato does not set fruit at high temperature and ripening is greatly affected by extremes of high and low temperatures. This warrants development of genotypes for environmentally stressed arid areas, for extended crop-harvest period (Samadia, 2006). Heat tolerance is generally defined as the ability of a plant to grow and produce economic yield under high temperatures. Saeed et al (2007) suggested that a genotype, that produces better yields under high temperature conditions, is heat tolerant. Germplasm is considered as a reservoir of variability for various traits (Vavilov, 1951). Genetic variability is the first step in plant breeding for crop improvement readily available in the germplasm. Large diversity for various characters exists is in tomato germplasm but potential variability for marketable fruit production under hot, arid environment is very limited. Fruit set and development of colour (lycopene production) under high temperatures with maximum 48oC and minimum 30oC and mean temperature >37oC during fruit harvest period (April and May) in the summer season crop, are major breeding objectives for hot agro-climatic conditions. The experimental material comprised 23 genotypes of tomato including 14 germplasm lines collected from arid and semi-arid regions. Five progeny lines of selections too 195 Variability in tomato under high temperature were included. The experimental material was grown in randomized block design with three replications, under field conditions in the summer of 2009 at Central Institute for Arid Horticulture, Bikaner. The climate of this eco sub- region is very hot, with dry summers and chilling winters. The material was sown in January in a nursery, and transplanted after 35 days in furrows 20-25cm deep and 10m long, in rows at 60 x 45cm plant spacing and accommodating 22 plants of each genotype in every replication. The crop stood in the field for four months (February 22 to June 20) as a spring-summer crop for evaluation and tomato lines were characterized under hot arid conditions. Observations were recorded on five competitive plants in respect of twelve horticultural traits, namely, plant height, number of branches, days to flowering, days to first harvest, number of fruits per plant, fruit weight, fruit length, fruit diameter, number of locules per fruit, pericarp thickness, Total Soluble Solids (TSS) and fruit yield per plant. Observations were also recorded on growth habit (determinate/indeterminate), number of flowers per cluster, per cent fruit-set under varying temperature conditions, and fruit colour development (lycopene). During April – June, maximum day temperature was > 40oC, except for six days. To assess yield potential of the genotypes, fruits harvested per plant during the 60 days of harvest period (15th April- 15th June) were accounted for. Data were analyzed using standard procedures of Gomez and Gomez (1984) and statistical package INDOSTAT Version 8.5. Genotypic and phenotypic co-efficient of variation, heritability (in a broad sense) and genetic advance were calculated following Burton and de Vane (1953). Data pertaining to growth and yield attributes of 23 tomato genotypes are presented in Table 1. Mean, range and coefficient of variation of 12 characters of the 23 genotypes is presented in Table 2. Analysis of Variance revealed a wide range of variability and highly significant differences among the tomato lines for all the characters studied (Tables 1 and 2). However, absolute variability among various characters cannot form the criteria for assigning a specific character as showing the highest degree of variability. Genotypic and phenotypic coefficients of variation (GCV and PCV), expressed as percentage, were high for fruit weight, plant height, number of fruits per plant, and fruit yield per plant; these were moderate for days to flowering, fruit length, fruit diameter, number of branches per plant, number of locules per fruit, pericarp thickness and TSS. GCV which indicates the extent of genetic variability in a population, ranged from 3.81% in days to first harvest, to 45.61% in fruit weight results on yield and component characters are in accordance with Sharma et al (2009) and Satesh et al (2007). GCV estimates were considerably high for yield-component characters such as fruit weight, number of fruits per plant and yield per plant. These, thus, offer a better scope for improvement through selection. Differences in magnitude of GCV and PCV were found to be very low (Table 2) indicating little influence of environmental factors on expression of the traits observed. In such a situation, selection can be effective on the basis of phenotype alone. Similar projections have been made by Samadia et al (2006) in tomato. Two lines, namely, SM2 L1 P9 and SM2 L2 P3, were found to have the best potential for yield per plant (5.15 kg and 4.46 kg, respectively). These lines were also superior in terms of red ripened quality fruits. Of the 23 lines evaluated, 12 were indeterminate in growth habit, four lines semi-determinate and seven determinate. Only five lines, viz., KSB-54, KSB-76, SM2 L1 P9, SM2 L2 P3 and SM2 L8 P5, developed fully-red fruit under high temperature (maximum day temperature > 42oC). Highest fruit weight was observed in SM2 L1 P9 (95.83g), followed by SM2 L8 P5 (80.40g). Maximum pericarp thickness was observed in SM2 L1 P9. Pooled data on average number of fruit/plant showed maximum number fruits per plant in KSB- 54 (102.9), followed by KSB-57 (93.8); however both these lines were indeterminate type, bearing small-sized fruits and failed to develop red fruits in of May and June. All determinate type genotypes, except SM2 L1 P9, stopped fruit-set by 10th May (when maximum day temperature exceeded 45oC). Three lines, KSB-54, KSB-57 and SM2 L1 P9, set fruit at high temperatures in the month of May. In the second fortnight of June, indeterminate lines too stopped growing; fruit development was inhibited, followed by their entry into the senescence phase. Estimates for heritability are a predictive instrument in expressing reliability of the phenotypic value. It, therefore, helps the breeder select for a particular trait when its heritability is high. In the present study, all the traits exhibited high heritability. Highest broad-sense heritability (e” 98%) was recorded for fruit weight, fruit length, fruit diameter, number of fruits/plant, plant height, number of locules in the trait and TSS (oBrix). In the present study, fruit weight and yield per plant were seen to be highly variable. Based on variability and heritability estimates, it is concluded that improvement by direct or indirect selection in tomato is possible for traits like fruit weight, number of fruits per plant/ per cluster, and yield per plant. Genetic advance also showed a pattern similar to that of heritability in broad-sense. Genetic J. Hortl. Sci. Vol. 7(2):194-198, 2012 196 advance, as per cent mean, was higher for a trait of much interest to the breeder, that is, yield per plant (93.2%) as also for average fruit weight (92.8%), number of fruits per plant (73.4%) and plant height (50.1%). Genetic advance (as per cent of mean) was the least for days to first harvest (7.5%), followed by days to flowering (17.6). In general, traits that show high heritability, with high genetic advance as per cent of mean, are genetically controlled by additive gene action, indicating easy selection and improvement in the breeding lines; Whereas, traits showing high heritability, with moderate or low genetic advance as per cent of mean, can be improved by inter-mating superior genotypes, and subsequent selection in the segregating population. These results corroborate the view of Ara et al (2009). High genotypic and phenotypic coefficients of variation, heritability and genetic advance as percentage of mean for number of fruits/plant, yield per plant and average fruit weight suggests their importance in selection for tomato improvement. Phookan et al (1998) and Mariame et al (2003) also observed similar trends in their study on 29 genotypes of tomato under summer conditions. Figure 1 shows average linkage dendrogram of standardized Euclidean distance. The more distant genotypes, falling in different clusters are expected to be better parents in a hybridization programme. Data on genotypic and phenotypic co-efficient of variation in traits is presented in Table 2. Phenotypic co- efficients of correlation, in general, were higher in magnitude than corresponding genotypic ones thereby, suggesting, an inherent association among the various characters studied. Similar observations were made by Singh (2005) and Samadia et al (2006). Days to flowering showed significant negative-correlation with fruit weight, fruit length, fruit Table 1. Performance of tomato lines per-se under hot, arid agro-climate Line DF DH FW FL FD F/P PH BR/P NL/F PCTH TSS Y/P KSB-4 26.3 59.7 45.10 2.91 3.60 31.7 78.1 6.74 3.8 3.8 5.87 1.428 KSB-5 21.0 58.3 25.05 3.21 4.23 44.1 107.7 8.31 3.1 3.0 7.83 1.099 KSB-11 21.7 57.0 23.53 2.84 3.79 79.8 149.8 10.89 3.6 2.8 6.17 1.879 KSB-13 21.3 54.7 35.20 3.21 3.92 74.3 147.9 9.62 3.2 3.2 8.00 2.616 KSB-29 22.0 59.0 32.83 3.09 4.31 55.0 132.5 8.59 3.4 3.0 6.73 1.805 KSB-30 22.7 57.7 35.63 3.20 4.25 86.7 110.2 9.67 3.0 3.1 7.36 3.093 KSB-41 20.3 60.6 41.57 3.13 4.21 76.6 123.1 8.26 3.2 3.7 6.83 3.186 KSB-54 20.0 55.3 31.30 3.14 3.93 102.9 132.7 8.62 4.0 3.4 6.47 3.220 KSB-57 20.3 53.3 32.13 2.54 3.19 93.8 123.9 7.40 4.1 3.0 6.17 3.015 KSB-58 22.7 57.7 36.04 3.22 3.58 40.4 132.8 7.65 3.5 3.1 5.90 1.453 KSB-60 20.3 57.3 47.67 3.07 3.97 70.9 89.3 8.00 3.9 3.8 7.00 3.358 KSB-72 21.0 55.3 38.57 3.34 4.22 55.6 144.8 6.56 3.3 3.3 7.33 2.140 KSB-74 21.7 58.3 20.44 2.98 3.67 41.8 122.3 7.28 3.0 2.9 7.10 0.854 KSB-76 21.7 60.0 26.97 3.26 3.80 40.8 121.5 9.31 3.2 3.4 7.53 1.100 PKM-1 20.3 53.3 67.60 3.71 4.82 38.6 89.8 6.14 4.2 3.8 6.20 2.610 KSB-29-1 21.7 57.7 34.50 3.14 4.29 66.7 117.0 8.17 3.4 3.5 7.27 2.303 CIAH-SEL 1 21.3 58.0 32.17 3.20 4.13 82.8 139.7 9.17 3.6 3.6 5.87 2.662 CIAH-SEL 2 22.3 58.7 21.10 2.89 3.42 69.4 107.1 6.44 3.7 3.0 6.03 1.464 SM2 L1 P9 19.0 59.0 95.83 4.94 6.08 53.7 80.8 5.61 5.2 5.3 7.50 5.152 SM2 L1 P16 18.3 59.6 66.93 4.32 5.45 39.6 61.2 5.56 4.4 5.0 7.37 2.649 SM2 L2 P3 18.3 59.3 53.37 4.38 5.38 83.7 71.2 5.60 4.5 4.9 8.07 4.462 SM2 L2 P4 18.6 60.3 45.64 3.65 4.81 38.5 81.3 5.61 4.3 4.8 7.83 3.363 SM2 L8 P5 17.3 61.7 80.40 4.74 5.79 40.7 73.7 6.31 4.3 5.1 7.37 3.269 CD at (P=0.05) 0.89 1.15 2.51 0.14 0.18 3.56 5.03 0.51 0.15 0.19 0.16 0.195 DF= days to flowering, DHR= days to harvest, FW= fruit weight (g), FL= fruit length (cm), FD= fruit diameter (cm), F/P= number of fruits per plant, PH= plant height (cm), BR/P= number of branches per plant, NL/F= number of locules per fruit, PCTH= pericarp thickness (mm), TSS= Total Soluble Solids (oBrix), Y/P= yield per plant (kg) Fig 1. Average linkage dendrogram of 23 tomato genotypes. Vertical distances are arbitrary, while, horizontal distances indicate linkage between genotypes Hanif Khan and Samadia J. Hortl. Sci. Vol. 7(2):194-198, 2012 197 diameter, TSS, and fruit yield per plant. This indicates that under high temperature conditions, early-flowering lines gave superior performance with respect to fruit yield and quality. Similarly, days-to-first-harvest was also negatively associated with fruit length, fruit diameter and yield per plant suggesting, that, early harvest is preferred for indirect selection for higher yield and larger fruit size under high temperature conditions. Fruit weight showed significant positive-correlation with fruit length, fruit diameter, TSS and yield per plant, whereas it showed negative correlation with number of primary branches. In this study, indeterminate- type lines having higher number of branches had smaller fruits. Fruit length showed significant positive-correlation with fruit diameter and yield per plant, whereas, it showed significant negative-correlation with number of fruits and Table 3. Correlation coefficient for 12 traits in tomato lines DF DH FW FL FD F/P PH BR/P NL/F PCTH TSS Y/P DF -0.115 -0.523** -0.267* -0.277* -0.095 0.366** 0.374** 0.152 0.121 -0.502** -0.522** DH 0.202 -0.280* -0.372** -0.402** -0.146 -0.211 0.128 0.282* 0.163 -0.263* FW 0.878** 0.863** 0.201 -0.187 -0.615** 0.223* 0.624** 0.237* 0.686** FL 0.958** -0.250* -0.149 -.0291* 0.168 0.524** 0.214 0.627** FD -0.178 -0.206 -0.332** 0.147 0.554** 0.234* 0.628** F/P 0.307* 0.285* 0.115 -0.283* -0.424** 0.519 PH 0.738** -0.173 -0.153 -0.181 -0.123 BR/P -0.182 -0.193 -0.166 -0.203 NL/F -.0201 -0.208 0.154 PCTH 0.283* 0.546** TSS 0.156 Y/P *, ** significant correlation at 5% and 1%, respectively DF= days to flowering, DHR= days to harvest, FW= fruit weight (g), FL= fruit length (cm), FD= fruit diameter (cm), F/P= number of fruits per plant, PH= plant height (cm), BR/P= number of branches per plant, NL/F= number of locules per fruit, PCTH= pericarp thickness (mm), TSS= Total Soluble Solids (oBrix), Y/P= yield per plant (kg) Table 2. Analysis of Variance (ANOVA) for yield and yield-contributing traits in tomato Trait Grand mean Range SE Co-efficient of variation Heritability % (H) Genetic advance GCV PCV GA GA % of mean DF 20.84 17-27 0.689 8.89 9.27 92 3.66 17.6 DH 57.93 53-62 0.681 3.81 3.99 91 4.34 7.5 FW 42.14 19.2 - 97.6 4.964 45.61 45.93 99 39.53 92.8 FL 3.40 2.45 - 5.02 0.165 18.25 18.42 98 1.26 37.2 FD 4.30 3.12 - 6.2 0.198 17.65 17.82 99 1.55 36.1 F/ P 59.61 30.5 -107.3 4.827 35.83 36.05 98 44.00 73.4 PH 110.3 59.8 -154.3 5.929 24.47 24.63 98 55.60 50.1 BR/P 7.63 5.2 - 11.3 0.518 18.86 20.26 86 3.05 40.1 NL/F 3.72 3.40 - 5.20 0.432 15.82 17.50 98 0.55 9.2 PCTH 3.63 2.8 5.3 15.64 16.84 93 1.2 28.3 TSS 6.93 5.8 - 8.2 0.154 10.47 10.56 98 1.48 21.4 Y/P 2.438 0.790 - 5.17 0.276 45.48 47.74 91 2.26 93.2 DF= days to flowering, FW= fruit weight (g), FL= fruit length (cm), FD= fruit diameter (cm), F/P= number of fruits per plant, PH= plant height (cm), BR/P= number of branches per plant, NL/F= number of locules per fruit, PCTH= pericarp thickness (mm), TSS= Total Soluble Solids (oBrix), Y/P= yield per plant (kg) branches per plant. Fruit diameter also showed positive correlation with TSS and yield per plant while showing negative correlation with number of branches/plant. Number of fruits/plant showed positive correlation with plant height, number of branches per plant and yield, whereas it showed negative correlation TSS and pericarp thickness. Pericarp thickness showed significant positive-correlation with fruit weight, fruit length, fruit width and yield per plant. A strong association of number of branches with days to flowering and plant height was noticed; however, it had significant negative-association with fruit weight. Hence, selection for early flowering and higher fruit weight and number of fruits per plant can indirectly help improve fruit yield per plant. Similar results were noticed by Mohanty (2002) and Singh et al (2002). Variability in tomato under high temperature J. Hortl. Sci. Vol. 7(2):194-198, 2012 198 REFERENCES Ara, A., Raj, N., Nazeer, A. and Khan, S.H. 2009. Genetic variability and and selection parameters for yield and quality attributes in tomato. Indian J. Hort., 66: 30-35 Burton, G.V. and de Vane, E.H. 1953. Estimating heritability in tall fescue (Festuca arundinacea L.) from replicated clonal material. Agron. J., 45:478-481 FAO. 2007. FAOSTAT Database. Food and Agriculture Organization, Rome, Italy Gomez, K.A. and Gomez, A.A. 1984. Statistical procedures for agricultural research. 2nd Edn., John Wiley and Sons, New York, USA Miller, P., Lanier, W. and Brandt, S. 2001. Using Growing Degree Days to Predict Plant Stages. Ag/Extension Communications Coordinator, Communications Services, Montana State University-Bozeman, Bozeman, MO, USA Mariame, F., Ravishankar, H. and Dessalegne, L. 2003. Study on variability in tomato germplasm under conditions of Central Ethiopia. Veg. Crops Res. Bulletin, 58:41-50 Mohanty, B.K. 2002. Studies on variability, heritability, interrelationship and path analysis in tomato. Ann. Agril. Res., 2:65-69 Phookan, D.B., Talukdar, P., Shadeque, A. and Chakravarty, B.K. 1998. Genetic variability and heritability in tomato (Lycopersicon esculentum) genotypes during summer season under plastic-house condition. Indian J. Agril. Sci., 68:304-306 Singh, P., Singh, S., Cheema, D.S. and Dhaliwal, M.S. 2002. Genetic variability and correlation study of some heat tolerant tomato genotypes. Veg. Sci., 29:68-70 Saeed, A., Hayat, K., Khan, A.A. and Iqbal, S. 2007. Heat tolerance studies in tomato (Lycopersicon esculantum Maill.). Intl. J. Agril. Biol., 9:649- 652 Samadia, D.K., Aswani, R.C. and Dhandar, D.G. 2006. Genetic analysis for yield components in t o m a t o landraces. Haryana J. Hortl. Sci., 35:116-119 Satesh, K., Sharma, J.P., Singh, A.K., Tiwari, S.P. and Neerja, S. 2007. Harvest index, quality and morphometrical evaluation for identification of superior types in tomato. Environ. Ecol., 25:399- 402 Sharma, J.P., Singh, A.K., Satesh, K. and Sanjeev, K. 2009. Identification of traits for ideotypeselection in tomato. Mysore J. Agril. Sci., 43:222-226 Singh, A.K. 2005. Genetic variability, correlation and path co-efficient studies in tomato under cold arid region of Ladakh. Prog. Hort., 37:437-443 Vavilov, N.I. 1951. The origin, variation, immunity and breeding of cultivated plants. Soil Sci., 72:483-482 (MS Received 17 February 2012, Revised 16 July 2012) Hanif Khan and Samadia J. Hortl. Sci. Vol. 7(2):194-198, 2012