STUDIES ON GENETIC VARIABILITY IN AGRONOMIC AND FRUIT QUALITY TRAITS AMONG SOME TOMATO (SOLANUM LYCOPERSICUM L.) GENOTYPES BY OCHAR KINGSLEY (10442177) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL CROP SCIENCE DEGREE JULY, 2015 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I, Ochar Kingsley, hereby declare that except for the references to other people’s work which have been duly cited, this thesis is the result of my original findings and that this thesis has neither in part nor in whole been presented for a degree in Ghana or elsewhere. OCHAR KINGSLEY SIGNATURE……………………………….. (STUDENT) DATE………………………………………... PROF. E. T. BLAY SIGNATURE……………………………….. (SUPERVISOR) DATE……………………………………….. PROF. G. O. NKANSAH SIGNATURE……………………………….. (SUPERVISOR) DATE……………………………………….. PROF. I. K. ASANTE SIGNATURE…………………………….. (SUPERVISOR) DATE……………………………………... University of Ghana http://ugspace.ug.edu.gh ii ABSTRACT In the present study, a total of twenty (20) tomato genotypes were evaluated simultaneously under greenhouse and field conditions in order to determine genetic variability present in the agronomic and fruit quality traits. The experiments were laid out in a Randomized Complete Block Design (RCBD) with three (3) replications. Both individual and combined analyses of variance were employed. The individual analysis of variance showed a significant (P < 0.01) variability among the genotypes for almost all characters studied. The components of variance estimated on individual location basis indicated a moderate to high GCV, high broad sense heritability as well as high genetic gain for almost all traits. However, stem diameter, pH and gallic acid recorded low estimated GCV under greenhouse conditions while plant height, stem diameter, chlorophyll content, number of days to 50 % flowering, fruit set percentage, number of days to fruit maturity and pH recorded lower estimates for GCV under field conditions. The combined analysis of variance (ANOVA) showed significant (P < 0.01) difference among the genotypes, locations and their interactions. The component of variance based on a two-factor analysis of variance revealed considerable effect of location and Genotype x Location interaction on the expression of traits especially, fruit phytochemical composition traits. Relative performance of the genotypes differed with location. Among the genotypes evaluated, MONGAL F1, PLATINUM F1, NKANSAH HT, WOSOWOSO, ROMA and SUMO F1 displayed superior performance for fruit yield across all locations. Large variation was observed among the genotypes for all fruit quality traits studied. Genotypes produced under greenhouse conditions recorded higher amounts of flavonoid composition while phenolic acids content were higher among field-produced tomato genotypes. Number of trusses per plant, number of fruits per plant as well as total fruit weight per plant showed a positive significant association with fruit yield under both greenhouse and field conditions. Consistency of correlation across the two locations was observed. However the correlation coefficients differed due to the effects of G x L interaction. University of Ghana http://ugspace.ug.edu.gh iii DEDICATION I dedicate this research work to the Almighty God and the entire members of my family University of Ghana http://ugspace.ug.edu.gh iv ACKNOWLEDGEMENT I thank the Almighty God for His unfailing love and unique favour upon me throughout the period of pursuing this programme. I am sincerely grateful to Prof. E. T. Blay, Prof. G. O. Nkansah and Prof. I. K. Asante for their supervisory roles, guidance and constructive suggestions that culminated in the successful completion of this work. I am grateful to Mr. Kwabena Kensah, Mr. Musah Bukari and workers of EnviroDome Greenhouse unit at the University of Ghana Forest and Horticultural Crops Research Centre (FOHCREC), Okumani- Kade for their support throughout the greenhouse and field experimental activities. My sincere gratitude goes to all workers of FOHCREC for their assistance in various ways. I thank Mr. W. A. Asante (Department of Crop Science) and Mr. Kojo Obeng (Department of Botany) for their immense support in diverse ways throughout the period of this work. University of Ghana http://ugspace.ug.edu.gh v TABLE OF CONTENT DECLARATION ................................................................................................................... i ABSTRACT ......................................................................................................................... ii DEDICATION ..................................................................................................................... iii ACKNOWLEDGEMENT .................................................................................................... iv TABLE OF CONTENT ........................................................................................................ v LIST OF ABBREVIATIONS............................................................................................... xi LIST OF TABLES ............................................................................................................. xvi CHAPTER ONE ................................................................................................................... 1 1.0 INTRODUCTION ...................................................................................................... 1 CHAPTER TWO .................................................................................................................. 4 2.0 LITERATURE REVIEW ................................................................................................ 4 2.1 Origin, domestication and distribution of tomato .......................................................... 4 2.2 Botany of tomato ......................................................................................................... 4 2.2.1 Vegetative description ........................................................................................... 4 2.2.2 Reproductive description ....................................................................................... 5 2.3 Nutritional and health benefits of tomato ..................................................................... 5 2.4 Agronomic and morphological evaluation of tomato .................................................... 7 2.5 Quality characteristics of tomato .................................................................................. 8 2.6 Antioxidants activity on free radicals ........................................................................... 9 2.7 Some antioxidant compounds in tomato ..................................................................... 10 2.7.1 Ascorbic acid ...................................................................................................... 10 University of Ghana http://ugspace.ug.edu.gh vi 2.7.2 Carotenoids ......................................................................................................... 10 2.8 Phenolic compounds .................................................................................................. 12 2.9 Tomato improvement ................................................................................................. 14 2.10 Evaluation of tomato characters for genetic variability ............................................. 15 CHAPTER THREE ............................................................................................................. 25 3.0 MATERIALS AND METHODS ................................................................................... 25 3.1 Experimental Location ............................................................................................... 25 3.2 Genetic Materials ..................................................................................................... 25 3.3 Nursery and Nursery Management Practices .............................................................. 25 3.4 Experimental Design and Field Layout ...................................................................... 27 3.5 Preparation of Experimental Sites .............................................................................. 28 3.5.1 Field experiment.................................................................................................. 28 3.5.2. Greenhouse experiment ...................................................................................... 29 3.6 Transplanting ............................................................................................................. 29 3.7 Agronomic Practice ................................................................................................... 30 3.8 Data Collection .......................................................................................................... 31 3.8.1 Agronomic traits ..................................................................................................... 31 3.8.2 Fruit physical and quality traits ............................................................................ 33 3.9 Statistical Analysis of Data ........................................................................................ 40 3.9.1 Estimation of variance components ..................................................................... 40 3.9.4 Correlation coefficient estimation ........................................................................ 44 CHAPTER FOUR ............................................................................................................... 45 University of Ghana http://ugspace.ug.edu.gh vii 4.0 RESULTS ..................................................................................................................... 45 4.1 Mean performance of tomato genotypes for agronomic traits across locations ............ 45 4.1.1 Mean plant height (cm) (PHT) ............................................................................. 45 4.1.2 Mean stem diameter per plant (cm) (SD) ............................................................. 45 4.1.3 Mean number of leaves per plant (NL) ................................................................ 46 4.1.4 Mean chlorophyll content (nm) (CC) ................................................................... 46 4.1.5 Mean number of primary branches per plant (NPB) ............................................. 48 4.1.6 Mean root length per plant (cm) (RTL) ................................................................ 48 4.1.7 Mean number of days to 1st flowering (NFF) ....................................................... 49 4.1.8 Mean number of days to 50 % flowering (FPF) ................................................... 50 4.1.9 Mean number of days to 1st fruit set (NFS) .......................................................... 50 4.1.10 Mean number of days to 50 % fruit set (FPFS) .................................................. 51 4.1.11 Mean percentage fruit set (FSP) ......................................................................... 51 4.1.12 Mean number of days to fruit maturity (DFM) ................................................... 52 4.1.13 Mean number of trusses per plant (TPP) ............................................................ 53 4.1.14 Mean number of flowers per truss (FPT) ........................................................... 53 4.1.15 Mean number of fruits per truss (NFPT) ............................................................ 55 4.1.16 Mean number of fruits per plant (FPP) ............................................................... 55 4.1.17 Mean single fruit weight per plant (g) (SFW) .................................................... 57 4.1.18 Mean total fruit weight (g) per plant (FWP) ....................................................... 57 4.1.19 Mean fruit yield per plant (t/ha) (YPP) .............................................................. 58 4.2 Mean performance of tomato for fruit physical and quality traits across locations ...... 59 University of Ghana http://ugspace.ug.edu.gh viii 4.2.1 Mean tomato fruit length per plant (cm) (FL) ...................................................... 59 4.2.2 Mean fruit diameter (cm) (FD) ............................................................................ 61 4.2.3 Mean fruit shape index of tomato (SI) ................................................................. 61 4.2.4 Mean number of locules per fruit (LOC) .............................................................. 62 4.2.5 Mean fruit pericarp thickness (mm) (PTK) .......................................................... 63 4.2.6 Mean fruit firmness (kg/cm2) (FF) ....................................................................... 63 4.3.5 Mean fruit pH value of tomato (pH) .................................................................... 64 4.3.6 Mean fruit dry matter content of tomato (g) (FDM) ............................................. 64 4.3.7 Mean total soluble solids content (0Brix) (TSS) ................................................... 66 4.3.8 Mean acidity content (mg/100 ml) (TA) .............................................................. 66 4.3.9 Mean flavour index (TSS/TA) ............................................................................. 68 4.3.10 Mean malic acid content (mg/100 ml) (MA) ...................................................... 68 4.3 Mean performance of tomato for phytochemical composition across locations ........... 70 4.4.1 Mean β-Carotene content in tomato fruit (mg/100 ml) (β -CARO) ....................... 70 4.4.2 Mean lycopene content in tomato fruit (mg/100 ml) (LYC) ................................. 70 4.4.3 Mean ascorbic acid content in tomato fruit (mg/100 ml) (ASC) ........................... 71 4.4.4 Mean gallic acid content in tomato fruit (mg/100 ml) (GA) ................................. 71 4.4.5 Mean rosmarinic acid content in tomato fruit (mg/100 ml) (RA).......................... 73 4.4.6 Mean vanillic acid content in tomato ((mg/100 ml) (VA) ..................................... 73 4.4.7 Mean catechin content in tomato fruit (mg/100 ml) (CAT) .................................. 75 4.4.8 Mean genestein content in tomato (mg/100 ml) (GEN) ........................................ 75 4.4.9 Mean hesperetin content in tomato fruit (mg/100 ml) (HES) ................................ 76 University of Ghana http://ugspace.ug.edu.gh ix 4.4.10 Mean kaempferol content in tomato fruit (mg/100 ml) (KAE) ........................... 76 4.4.11 Mean myricetin content in tomato fruits (mg/100 ml) (MYR) ............................ 77 4.4.12 Mean rutin content in tomato fruit (mg/100 ml) (RUT) ...................................... 78 4.4.13 Mean quercetin content in tomato fruit (mg/100 ml) (QUE) ............................... 78 4.4.14 Mean antioxidant activity (mg/100 ml) (AA) ..................................................... 80 4.4 Variance components, heritability and genetic gain of tomato characters ................... 81 4.4.1Variance components of tomato traits studied under greenhouse conditions.......... 81 4.4.2 Heritability of tomato traits studied under greenhouse conditions ...................... 82 4.4.3 Genetic gain (GAM) of tomato traits studied under greenhouse conditions .......... 82 4.4.4 Variance components of tomato traits studied under field conditions ................... 82 4.4.5 Heritability of tomato traits studied under field conditions ................................... 87 4.4.6 Genetic gain (GAM) of tomato traits studied under field conditions .................... 87 4.4.7 Variance components of tomato traits estimated over combined locations ........... 91 4.4.8 Heritability of tomato traits estimated over combined locations ........................... 92 4.4.9 Genetic gain (GAM) of tomato traits estimated over combined locations ............. 92 4.5 Character association among tomato genotypes.......................................................... 97 4.5.1 Correlation of tomato traits studied under greenhouse conditions ........................ 97 4.5.2 Correlation of tomato traits studied under field conditions ................................... 98 4.5.3 Correlation among tomato traits over combined locations .................................. 101 CHAPTER FIVE .............................................................................................................. 105 5.0 DISCUSSION ............................................................................................................. 105 5.1 Variability in Agronomic Performance of Tomato Genotypes .................................. 105 University of Ghana http://ugspace.ug.edu.gh x 5.1.1 Vegetative characteristics of tomato genotypes .................................................. 105 5.1.2 Reproductive and phenological characteristics of tomato genotypes .................. 106 4.1.3 Yield and yield components characteristics of tomato genotypes ....................... 108 5.4 Fruit physical and quality characteristics of tomato genotypes ................................. 110 5.3 Antioxidant activity of tomato genotypes ................................................................. 116 5.1.4 Genotype x Location interaction effects on tomato fruit yield ............................ 116 5.3 Genetic Variance, heritability and genetic gain of tomato characters ........................ 118 5.3.1 Individual location basis .................................................................................... 118 5.3.2 Combined location basis .................................................................................... 120 5.4 Character association of traits among tomato genotypes ........................................... 122 CHAPTER SIX ................................................................................................................. 126 6.0 CONCLUSIONS AND RECOMMENDATIONS ....................................................... 126 6.1 Conclusions ........................................................................................................ 126 6.2 Recommendations ............................................................................................... 127 REFERENCES ................................................................................................................. 129 APPENDIX ...................................................................................................................... 148 University of Ghana http://ugspace.ug.edu.gh xi LIST OF ABBREVIATIONS AA Antioxidant activity ANOVA Analysis of Variance ASC Ascorbic acid content CARO Carotene content CAT Catechin content CC Chlorophyll Content CV (%) Coefficient of variation expressed as percentage DAT Days after Transplanting df Degree of freedom DFM Number of days to fruit maturity DNA Deoxyribonucleic acid DPPH diphenyl-2-picrylhydrazyl ECV Environmental coefficient of variability EGA Expected genetic advance EMS Expectation of means FAO Food and Agricultural Organization FAOSTAT Food and Agriculture Organization Statistical Databases University of Ghana http://ugspace.ug.edu.gh xii FD Fruit diameter FF Fruit firmness FL Fruit length FOHCREC Forest and Horticultural Crops Research Centre FPF Number of days to 50 % flowering FPFS Number of days to 50 % fruiting FPP Number of fruits per plant FPT Number of flowers per truss FSI Fruit shape index FSP Percentage fruit set FW Fresh Weight FWP Total fruit weight per plant GA Gallic acid GAM Genetic advance as percentage of mean GCV Genotypic coefficient of variability G X L Genotype-by-location interaction GEN Genestein content HES Hesperetin content h2b Broad Sense Heritability University of Ghana http://ugspace.ug.edu.gh xiii IC50 Inhibition Coefficient at 50 % KAE Kaempferol content k Selection differential LOC Number of locules LSD Least Significant Difference LYC Lycopene MA Malic acid content MS Mean square MYR Myricetin content n.d. No date N.P.K Nitrogen, Phosphorus and Potassium NaOH Sodium hydroxide NFF Number of days to first flowering NFPT Number of fruits per truss NFS Number of days to fruit set NL Number of leaves NPB Number of primary branches per plant PCV Phenotypic coefficient of variability PHT Plant height University of Ghana http://ugspace.ug.edu.gh xiv PTK Pericarp thickness QUE Quercetin content RCBD Randomized Complete Block Design RA Rosmarinic acid content ROS Reactive Oxygen Species RUT Rutin content SFW Single fruit weight SD Stem diameter SSD Single Seed Decent t/ha tonnes per hectare TA Titrable acidity TPP Number of truss per plant TSS Total Soluble Solids content TSS/TA Flavour indicator (ratio of TSS to TA) UNESCO United Nations Educational Scientific and Cultural Organization VA Vanillic acid content WHO World Health Organization YPP Fruit Yield per plant University of Ghana http://ugspace.ug.edu.gh xv λ Wave length σ2E Environmental variance σ2G Genotypic variance σ2GL Genotype x environment interaction variance σ2L Location variance σ2P phenotypic variance University of Ghana http://ugspace.ug.edu.gh xvi LIST OF TABLES Table 3. 1 Tomato genotypes used in the study .................................................................... 26 Table 3. 2 Climatic data in the greenhouse environment ...................................................... 27 Table 3. 3 Climatic data at the field experimental location ................................................... 28 Table 3. 4 Physicochemical properties of soil at the field experimental location .................. 29 Table 3. 5 Physicochemical properties of soil in the greenhouse environment .................... 30 Table 3.6 Phenolic compounds and their standard curves…………………………………...38 Table 3.7 Flavonoid compounds their standard curves………………………………………39 Table 3. 8 Format for individual analysis of variance .......................................................... 41 Table 3. 9 Estimation of variances and broad sense heritability on individual location basis 41 Table 3. 10 Format for a two-factor (combined) analysis of variance ................................... 42 Table 3. 11 Estimation of variance and broad sense heritability on combined location basis 43 Table 3. 12 Methods for computing coefficient of variability and genetic gain (EGA & GAM) ............................................................................................................................ 43 Table 4. 1 Mean Plant height, Stem diameter and Number of leaves per plant ..................... 47 Table 4. 2 Mean Chlorophyll content, Number of primary branches per plant and Root length ............................................................................................................................ 49 Table 4. 3 Mean Number of days to 1st flowering, 50 % flowering and Number of days to 1st fruit set ................................................................................................................ 52 Table 4. 4 Mean Number of days to 50 % fruit set, Percentage fruit set and Number of days to fruit maturity ....................................................................................................... 54 Table 4. 5 Mean Number of trusses per plant, Number of flowers per truss and Number of fruits per truss ..................................................................................................... 56 Table 4. 6 Mean Number of fruits per plant (FPP) ............................................................... 58 University of Ghana http://ugspace.ug.edu.gh xvii Table 4. 7 Mean Single fruit weight per plant, Total fruit weight per plant, and Fruit yield per plant .................................................................................................................... 60 Table 4. 8 Mean Fruit length, Fruit diameter and Fruit shape index ..................................... 62 Table 4. 9 Mean Number of locules, Fruit pericarp thickness and Fruit firmness ................. 65 Table 4. 10 Mean pH value, Fruit dry matter content and Total soluble solids content ......... 67 Table 4. 11 Mean Titrable acidity, TSS/TA and Malic acid content of tomato fruits ............ 69 Table 4. 12 Mean β-Carotene, Lycopene and Ascorbic acid contents of tomato fruits .......... 72 Table 4. 13 Mean Gallic, Rosmarinic and Vanilic acids content of tomato fruits ................. 74 Table 4. 14 Mean Catechin, Genestein and Hesperetin content of tomato genotypes ............ 77 Table 4. 15 Mean Kaempferol, Myricetin and Rutin contents of tomato genotypes .............. 79 Table 4. 16 Mean Quercetin content and Antioxidant activity .............................................. 81 Table 4. 17 Estimate of variances, heritability and genetic advance for 16 vegetative and reproductive characters under greenhouse conditions........................................ 84 Table 4. 18 Estimate of variance, heritability and genetic advance for 15 fruit physical and quality characters of tomato under greenhouse conditions ................................ 85 Table 4. 19 Estimate of variance, heritability and genetic advance for 14 antioxidant characters of tomato under greenhouse conditions ............................................ 86 Table 4. 20 Estimate of variance, heritability and genetic advance for 16 vegetative and reproductive characters of tomato under field conditions .................................. 88 Table 4. 21 Estimate variance, heritability and genetic advance for 15 fruit physical and quality characters of tomato under field conditions ........................................... 89 Table 4. 22 Estimate of variance, heritability and genetic advance for antioxidant characters of tomato under field conditions ....................................................................... 90 Table 4. 23 Estimate of variance, heritability and genetic advance for 16 vegetative and reproductive characters of tomato over combined locations .............................. 94 University of Ghana http://ugspace.ug.edu.gh xviii Table 4. 24 Estimate of variance, heritability and genetic advance for 16 fruit physical and quality characters of tomato over combined locations ....................................... 95 Table 4. 25 Estimate of variance, heritability and genetic advance for antioxidant characters of tomato over combined locations ................................................................... 96 Table 4. 26 The Pearson’s correlation matrix of selected agronomic and quality traits of tomato under greenhouse conditions ................................................................. 99 Table 4. 27 The Pearson’s correlation matrix of selected agronomic and fruit quality traits of tomato under field conditions ......................................................................... 102 Table 4. 28 The Pearson’s correlation matrix of selected agronomic and quality traits of tomato over combined greenhouse and field locations .................................... 104 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE 1.0 INTRODUCTION Tomato (Solanum lycopersicum L., 2n = 24) is one of the most widely produced, consumed and processed vegetable crops (Tekeoka et al., 2001, Rukshar et al., 2012). Epidemiological and medical research reports reveal an association between the consumption of tomato and its products with a reduction in certain chronic and degenerative diseases such as those related to the cardiovascular system, certain types of cancer as well as ageing conditions (Harrigan et al., 2007; Tambo and Gbemu, 2010). The medicinal and health benefits associated with tomato are mainly attributed to the antioxidant function of important compounds present in the fruits. It is therefore crucial to monitor the quality components of tomato since consumers and processors currently show preference for specific quality traits of the fruits (Radzevicius et al., 2013). Development of tomato cultivars that combine high yielding potential with improved nutritional and quality traits could be essential in the quest to meet the needs of tomato growers, fresh tomato consumers as well as the processing industry (Radzevicius et al., 2013). Several studies have indicated that the tomato crop is adaptable to different climatic conditions and diverse cropping systems (Van der Hoeven et al., 2002; Shankara et al., 2005; Hossain et al., 2010). This characteristic of the crop enables it to be transferred or introduced from one geographical location to another (Kader and Rolle, 2004). In Ghana, wide range of tomato genetic resources are under cultivation on both small and large scale commercial production systems (Blay et al., 1999). This represents very rich source of genetic materials that can be evaluated, exploited and utilized in tomato improvement programmes (Lecomte et a.l, 2003). Yet knowledge is limited in terms of tomato genetic resources in Ghana that are endowed with better fruit quality traits particularly their antioxidant content. This makes genetic variability study in the crop very essential. University of Ghana http://ugspace.ug.edu.gh 2 Agro-morphological evaluation is an essential technique that has been extensively used by many researchers to determine variability in crop genetic resources. The method is simple, fast and cost effective (Hoogendijk and Williams, 2001). However, the expressions of quantitative traits are reported to be associated with the masking effects of several interrelated environmental factors (Panthee et al., 2012; Nour et al., 2013). Consequently, a particular genotype of a crop is subjected to performing differently across different environments or locations irrespective of its inherent genetic potential. It is thus very significant that the genetic assessment of a crop is carried out over different locations or environments (Nwosu et al., 2014). Over the years, open field tomato production systems have been the main system of production practiced especially in the less developed regions of the world. Currently, tomato fruits are produced year round to satisfy the rising consumer and processors’ demand in greenhouse production systems. Enhancing increase in tomato farmers’ incomes, reduction in poverty level as well as ensuring food security necessitates development of tomato cultivars with better performance for yield trait under different growing conditions. Considering the nutritional significance of tomato, and the need to satisfy the rising consumers’ and processors’ demand in terms of quantity and quality, it is important to develop cultivars with desirable quality traits as well as high yields (Willcox et al., 2003). Knowledge about genetic variability in tomato in terms of its agronomic and fruit quality components is indispensable for achieving the important breeding objective of improvement in yield and fruit quality traits of tomato (Emami, et al., 2013; Osekita and Ademiluyi (2014). This study was therefore carried out to evaluate twenty tomato genotypes for genetic variability in their agronomic and fruit quality traits under greenhouse and field conditions so as to identify better performing ones for incorporation into crop improvement programmes. University of Ghana http://ugspace.ug.edu.gh 3 OBJECTIVE The main purpose of the study was to identify genetic variability in tomato genotypes that will form the basis for development of cultivars with high yielding ability and desirable quality traits. The specific objectives of the study were to; 1. Determine genetic variability among the genotypes based on selected agronomic and fruit quality traits. 2. Select promising tomato genotypes for incorporation into future breeding programmes. 3. Determine the genetic and environmental effects on the studied traits 4. Estimate the heritability and genetic gain of the studied traits 5. Determine the association among the selected agronomic and fruit quality traits University of Ghana http://ugspace.ug.edu.gh 4 CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 Origin, domestication and distribution of tomato Tomato (Solanum lycopersicum L.) is reported to have originated from the Andean regions of Western and Southern America (Peralta and Spooner, 2007) though other hypotheses suggest Mexico instead (Bai and Lindhaut, 2007). While the early history of tomato domestication is reported to be unclear, Peru is suggested as the most probable centre of diversity of tomato wild relatives whereas Mexico is the original centre of domestication (Larry and Joanne, 2007). Currently, wild tomato genotypes are still found growing in their centres of origin and domestication. Genetic variability studies meant to develop desirable traits in the cultivated tomato’s wild relative, Solanum lycopersicum var cerasiforme to be incorporated into tomato breeding programmes have been carried out (Larry and Joanne, 2007). New tomato species are continuously being identified and preserved in various gene banks (Peralta et al., (2005). Domestication of tomato in its centre of origin occurred and remained there until its introduction to Europe around the 15th C where domestication continued over the 18th and 19th centuries (Sims, 1980). Tomato is adaptable to a wide range of environments and agro- ecological zones; hence it is found cultivated throughout the world as one of the most important vegetable crops. The introduction of tomato in West Africa is reported to have occurred around the 16th and 17th centuries (Norman, 1992). It is extensively cultivated in Eastern, Central and West Africa especially in Ghana and Nigeria (De Lannoy, 2001). 2.2 Botany of tomato 2.2.1 Vegetative description Tomato (Solanum lycopersicum L.), a member of the Solanaceae family (Peralter and Spooner, 2007) is a short-term duration perennial herb but usually grown as an annual vegetable crop. The crop is dicotyledonous with a tap root system which can grow to a length University of Ghana http://ugspace.ug.edu.gh 5 of about 50 cm with characteristic dense lateral and adventitious roots (Akinfasoye, 2011). The glandular stem of the plant can grow to about 2 to 4 m high (Shankara et al., 2005). Its compound leaves consist of ovate to oblong leaflets spirally arranged along the hairy weak wooden stem by the petiole (Yeboah, 2011). Tomato plants are usually classified as determinate, semi-determinate or indeterminate based on their growth habits. The determinate types discontinue growing further at about 1.5 m high following the formation of flower cluster at the terminal growing point. Unlike the indeterminate types, determinate tomatoes are generally self-supporting and require no staking. Their cultivation is less labour-intensive, making them highly popular for commercial cultivation. The indeterminate types in contrast continue to grow after flowering though growth may cease when cultivated under tropical conditions or due to diseases and pests attacks (Shankara et al., 2005). 2.2.2 Reproductive description The cultivated tomato (Solanum lycopersicum L.) plant is an autogamous species (flowers are self- pollinated) though cross-pollination may occur (Agong et al., 2001, Shankara et al., 2005). The plant produces a cluster of inflorescence with an average of 6 to 12 bisexual flowers whose petals are coloured yellow. The 6 stamens and anthers are bright yellow and surround the style with an elongated sterile tip. The ovary is superior with 2-9 locules. The fruit is fleshy and variable in shape, length and diameter. The immature fruits change from green to yellow, orange or red at ripe stage. The brightly coloured edible fruits are larger among the cultivated types compared to the wild species. Fruits of the determinate types are known to be fast ripening compared to the indeterminate ones. The slower rate in ripening together with the high leaf to fruit ratio improves the taste of indeterminate types. 2.3 Nutritional and health benefits of tomato Tomato fruits are valued all over the world by consumers and processors due to the presence of many important nutritional components like vitamins, minerals and antioxidant University of Ghana http://ugspace.ug.edu.gh 6 constituents reported to play essential role in human health (Reddy et al., 2013). It is a rich source of vitamin C, lycopene, beta-carotene, folate, potassium, flavonoids and vitamins (Willcox et al., 2003). The chemical constituents such as β-carotene, potassium and vitamins (A, and C) are significant determinants of nutritional quality of tomato. The fruits are also major sources of micronutrients, minerals, and oxalic acids (ascorbic, citric, malic and fumeric). Report by Borguini and Da Silva Torres (2009) revealed a number of vital nutritional components in tomato fruits such as minerals, vitamins (vitamin C, E, B6, folic acid, niacin), potassium and trace elements. Several epidemiological and medical reports have indicated a high rate of the occurrence of chronic and degenerative diseases throughout the world and that phytochemical compositions in fruits and vegetables are associated with reduced risk of many of such chronic and degenerative diseases. Report by WHO ( 2003) indicated that diet containing important phytochemical components could help reduce or avoid about 90% Type II diabetes, 80% cardiovascular-related conditions as well as 1/3 of certain types of cancer. Consumption of fruits and vegetables and their associated nutritional and several other metabolite compositions in reducing the risk associated with those disease conditions have currently received research attention. The tomato fruit is an excellent source of organic acids and other important metabolites such as carotenoids, lycopene, phenolics acids and flavonoids which possess antioxidant properties (Mertenz-Talcott et al., 2003). Abdul-Hammed et al., (2009) suggested a need for consumption of fruits and vegetables due to the rising rate of chronic diseases including obesity, diabetes, different kinds of cancer and heart-related diseases. Flavonoids for instance are important components of the tomato fruit reported to be associated with reduction in cancer-related diseases by inhibiting the excretion of certain molecules from cancerous cells (Yu and Sacco, 2005). Also, Hounsome et al. (2008) indicated the role of antioxidant composition of tomato in protecting the human cell from University of Ghana http://ugspace.ug.edu.gh 7 oxidants responsible for several human cancerous conditions known to occur in the stomach, oesophagus, lung, pharynx, endometrium, pancreas and colon. 2.4 Agronomic and morphological evaluation of tomato Agronomic and morphological evaluation involves identification and recording of visible characters of plants including height, leaf type, leaf colour, hairiness and floral features (Godia, 2014; Nwosu et al., 2014). The use of morphological data is one of the most important methods in estimating genetic variability in a crop. The method is simple, less costly and concise and thus used as diagnostic tool in evaluating the performance of different genotypes under specific growing environments (Hoogendijk and Williams, 2001). Osei et al. (2014) reported that morphological approach to genetic variability studies involves growing sample plants in an environment and taking visual score of their agro-morphological characters. Such characters are good markers that can be used to assess genetic variability existing in the crop and to further facilitate systematic crop improvement programmes aimed at understanding their adaptability to diverse environments. Agro-morphological evaluation in a crop gives direction to the choice of parents in hybridization breeding. Descriptors developed by the IBPGR are used as guide during such data recordings (Ibitoye et al., 2009). Characterization and evaluation of plants in the most precise, quick and reliable manner is highly essential in crop improvement programmes. Genetic variability and genotype performance evaluation are easy to be conducted in the required growing conditions of a crop through the use of morphological data (Nwosu et al., (2014). Blay et al. (1999) carried out an experiment on morphological and agronomic characterization of some tomato (Lycopersicum esculentum) germplasm in Ghana. The results of the study showed an existence of variability among several traits including plant height, fruit set, number of fruits per plant, fruit weight, number of locules per fruit and fruit yield. Fruit set percentage was low due to the sensitivity of genotypes to heat stress. University of Ghana http://ugspace.ug.edu.gh 8 2.5 Quality characteristics of tomato Generally, consumers and processors of tomato show preference for specific tomato fruit quality traits and hence tomato growers are obliged to produce specific varieties in order to successfully compete in the markets. While such traits like appearance, firmness and flavour influence fresh market demands, total soluble solids content, pH and firmness determine the quality of tomato in the processing industry. This phenomenon also calls for the evaluation of available cultivars and the development of new tomato varieties which are consumer focused (Lecomte et al., 2003). Quality parameters of tomato are determined by a number of physical and chemical traits of the fruits (Aoun et al., 2013). Fruit flavour, sweetness, taste, texture, firmness and nutritional properties are vital determinants of fruit quality that influence consumers of fresh tomato (Carli et al., 2009; Rocha et al., 2012). Most of these traits are in turn dictated by several other components which are themselves quality determinant traits. These quality components include acidity, sugars, total soluble solids content, ascorbic acid, lycopene, β-carotene, flavonoids, and phenolic acids. The level of production and thus the concentration of these compounds in fruits are not only influenced by genetic factors but also the crop’s surrounding environment as well as genotype-by-environment interaction effects (Chattopadhyay et al., 2013). These factors therefore bring about significant variability among different tomato genotypes and could be exploited for further genetic improvement of the crop (Rocha et al., 2012). Currently, tomato research attention has in addition to improvement in yield and resistance to biotic and abiotic stresses, incorporated fruit quality component traits. Efforts are therefore directed at developing tomato varieties with improved antioxidant composition in order to fulfill their medicinal and epidemiological significance (Harrigan et al., 2007). University of Ghana http://ugspace.ug.edu.gh 9 2.6 Antioxidants activity on free radicals Antioxidants represent a valuable group of compounds which when in human body at lower concentrations are capable of inhibiting and counteracting the oxidative species and their associated damaging effects on the body cells and tissues (Shiow, 2003). Since oxidative processes of free radicals result in various chronic and degenerative diseases in humans, it can be inferred that by their counteracting activity, antioxidant compounds are involved in the prevention of chronic and degenerative diseases. Such chronic and degenerative diseases according to medical research reports include cardiovascular diseases, certain type of cancers, arthritis, neurological diseases, fibrosis, ageing and several others (Bachir et al., 2014). It is reported that several enzymes within the human body are involved in the free radical scavenging function. However, plant antioxidant (phytochemical or chemo-protective compounds) commonly obtained in human diet represent an important and more abundant source. Some of the most frequently reported phytochemical compounds with antioxidant roles include vitamins (ascorbic acid), carotenoids (β-carotene and lycopene), phenolic compounds (phenolic acids and flavonoids) as well as some mineral elements especially trace elements (Nour et al., 2013). Contrast to the animal cell where antioxidant production is limited, the plant cell produces a broad range of these antioxidant compounds whose concentration and antioxidant capacity vary among different fruits and edible plant parts. In a study by Bachir et al. (2014), fruits are endowed with a wide range of antioxidant compounds capable of neutralizing the oxidative damage arising from free radical accumulation and consequently lower the risks of chronic and degenerative diseases. The total composition of antioxidant compounds in tomato is reported to be highly influenced by genetic factors and to some extent by environmental factors (Nour et al., 2013). University of Ghana http://ugspace.ug.edu.gh 10 2.7 Some antioxidant compounds in tomato 2.7.1 Ascorbic acid Tomato (Solanum lycopersicum L.) is an important source of ascorbic acid with a property of scavenging the free radicals that may potentially cause several chronic and degenerative diseases in humans (Sablani et al., 2006). Clinical and epidemiological research therefore suggest a 100–120 mg/L daily intake of ascorbic acid so as to help avoid the risk associated with cardiovascular, stroke and cancer among the growing human population (Abdul-Fatawu, 2013). The relative concentration of ascorbic acid like many other antioxidant compounds is reported to vary among different genotypes due to genetic and environmental influence (Peng et al., 2008; Adalid et al., 2010). Singh et al. (2014) evaluated tomato hybrids for growth, yield and quality performance and recorded an ascorbic acid content range of 12.65–15.63 mg/100 g. According to this author, high ascorbic acid content in tomato fruits improves upon the quality of the fruits and thus required as an important determinant of the suitability of a genotype for processing. Variability in ascorbic acid content has been reported in earlier studies by Radzevicius (2013) who found ascorbic acid range of 8.20 to 16.20 mg/100 g and Aoun et al. (2013) who recorded ascorbic acid a range of 6.01 to 12.94 mg/100 g. Ascorbic acid content has been reported to be higher in fruits produced under field conditions due to high light intensity as compared with greenhouse cultivation environment characterized by being shady. Lower light intensity reduces the synthesis of sugar which is a substrate for ascorbic acid biosynthesis, hence reduced ascorbic acid content (Lee and Kader, 2000). 2.7.2 Carotenoids Tomato is an important source of many carotenoid compounds chiefly among which are β – carotene and lycopene. The concentration of carotenoid compounds in tomato is greatly determined by the genotype or cultivar of the plant (Kuti and Konuru, 2005). Other factors reported to have influence on the amount of carotenoids in plants include the growing University of Ghana http://ugspace.ug.edu.gh 11 environment of the plant such as climatic, fertilizer and soil conditions. Different tomato cultivars have been reported to be variable in lycopene content from 13, 48, 49 and 45 mg/100 g (Ilic et al., 2014). Lycopene is the most abundant micronutrient among the entire carotenoid component in tomato fruits and it is formed during the ripening stage of the tomato fruit. Ibitoye et al. (2009) indicated that lycopene concentration in fresh tomato is responsible for the deep red colour variation among different tomato genotypes and it is known to be directly correlated with fruit ripeness and increasing pH. This compound accounts for almost 80 to 90 % of the overall carotenoid content of the tomato fruit higher than beta-carotene content (5-10%) Both β-carotene and lycopene play essential health promoting roles due to their antioxidant capacity. They reduce the occurrence of many chronic and degenerative diseases (like cardiovascular, diabetes and certain cancer related conditions by neutralizing the effects of oxidative damage to cells (Tepic et al., 2006). A study conducted by Capanoglu et al. (2010) revealed that almost 85 % of lycopene requirement in the human body is sourced from tomato and besides being the most abundant of all carotenoids compounds, it is the most effective antioxidant in neutralizing free radicals and by far, the most stable to changes during processing. Humans and other animals are unable to synthesize lycopene and hence depend on diets that contain lycopene (Tapier et al., 2004). A study by Kuti and Konuru (2005) who evaluated the effects of genotype and cultivation environment on lycopene content in red-ripe tomatoes, a significant variability among the studied tomato genotypes was reported and that both environment and cultivar factors influence lycopene content in tomato fruits. The concentration of lycopene in tomato fruits has been reported to differ as a result of genotype factor and growing environmental conditions (light and temperature). Also University of Ghana http://ugspace.ug.edu.gh 12 lycopene concentration is reported to be higher in fruit skin (approximately 37 %). This has been indicated to be 3-6 times higher than the entire pulp content. Whereas the outer pericarp is known to contain high amount of lycopene and carotenoids, the proportion of carotenoids is higher in the locules (Toor and Savage, 2005) 2.8 Phenolic compounds Phenolic compounds are essential for plant growth, reproduction and development. They also play a role in plant pigmentation, as protective agents against ultraviolet light, act as anti-feedants and anti-pathogen, pesticides and as structural material for plant stability (Msaada et al. (2014). Their essential function as antioxidant compounds as well as their inhibitory effects on oxidative stress and some harmful enzyme systems make phenolic compounds an attractive area by most researchers. Many important phenolic compounds are known to exist and are diverse in chemical structure irrespective of the common presence of hydroxyl group. Phenolic acids and flavonoids represent different categories of phenolic compounds which form essential component of plants by virtue of their antioxidant activity (Jaffel et al., 2011). Phenolic acids are regarded as strong antioxidant compounds capable of scavenging all oxidative molecules (free radicals) through their unique hydroxyl groups (Sroka and Cicowski, 2003). Important compounds of phenolic acid group include gallic acid, vanillic acid, rosmarinic acid, salicylic, etc. They are extensively distributed in plant species and play important function in human body as antibacterial, antiviral, anti- inflammatory and as antioxidant compounds (Klem et al., 2000) Flavonoids are a group of poly-phenolic compounds formed as secondary plant metabolites and are reported to be the largest and most studied category of plant phenols that occur by nature. These compounds possess anti carcinogenic, anti-inflammatory, anti-allergic and anti- mutagenic properties (Tokusoglu et al., 2003). Flavonols are a category of flavonoids predominant in almost all plants and thus form a component of human diet. The composition University of Ghana http://ugspace.ug.edu.gh 13 of flavonols in plants is influenced by both environment (temperature, soil, light nutrition and pathogen) and genotype specific factors (Tokusoglu et al., 2003). Some of the most common flavonols include quercetin, kaemferol, myricetin, catechin, hesperitin, rutin, isorhamatin and genestein. Among the flavonols, quercetin and kaemferol are the most abundant in fruits with quercetin being highly potent in defending the human body against reactive oxygen species (Martinez- Valverde et al., 2002). Lee and Kader (2000) mentioned that ascorbic acid and quercetin concentrations in tomato are known to increase with exposure to sun light. Riadh et al. (2009) evaluated lycopene content of advanced breeding lines of tomato for their bioactive compounds and antioxidant activity. Significant differences were observed among the genotypes for all studied characters (agronomic, total carotenoids, lycopene, total phenolics, flavonoids, ascorbic acid as well as their antioxidant activities). Bachir et al. (2014) conducted a study on antioxidant activity of eight tomato (Solanum lycopersicum L.) varieties and recorded significant difference among the genotypes for lycopene (which varied between 3.90 to 7.70 mg/100 g), phenolic compounds (with a range of 20.6 to 49.55 mg/100 g). Generally, the concentration of phytochemical compounds has been found to be higher in fruits produced under field conditions than that of the greenhouse. Tomato fruits produced under field condition tend to receive higher amount of light and thus associated with high phenolic content. Light increases phenolic compounds biosynthesis by increasing enzymatic reactions (Caliman et al., 2010). High environmental conditions including temperature, relative humidity, light intensity and rainfall increases the expression of rutin and thus increase its content in fruits. Higher amounts of rutin in fruits has also been reported to have no negative effects on certain fruit quality traits like carotenoids, vitamin C, pH, fruit taste, colour or TSS. University of Ghana http://ugspace.ug.edu.gh 14 2.9 Tomato improvement Since the early period of tomato domestication and its subsequent introduction across various regions of the world, several attempts have been made to enhance improvement in the available genotypes of the crop (Deery, 2012). According to Shende et al. (2012), the use of exotic tomato genetic materials and incorporation of economically essential novel genes into existing tomato genetic resource has resulted to success in tomato improvement programmes. The existence of over 83 300 tomato accessions in the gene banks globally is an indication of a great variability in the crop and this provides an essential tool with which breeders work (FAO, 2010). Deery (2012) assumed that tomato improvement programmes have been so much focused on morphological traits to enhance increased yield. Similarly, Maul et al. (2000) reported that breeding work has over the years centered on developing varieties which show disease resistance, response to fertilization and higher nutrient content. Today, tomato breeding objective has been geared towards developing varieties which are adaptable to growth constraints, disease and pest resistance, fruit productivity and more importantly, quality attributes (Albrecht et al., 2010). Several reports confirm the current rising global population and its parallel consumption rate of vegetable crops which necessitates monitoring of nutritional constituents of horticultural crops including tomato. Carli, et al. (2009) therefore suggest the need to develop tomato genotypes that meet consumers’ preferred fruit quality traits by considering sensory attributes of tomato as a key breeding objective irrespective of the difficulties reported to be involved in their assessment. Improvement of quality traits of vegetable crops like tomato with known health promoting implications of their micro-phytonutrients has therefore gained research focus in recent years. It has been hypothesized that tomatoes, both cultivated and wild species alike possess important agronomic and physicochemical (quality) characters that can be exploited in tomato improvement programmes (Firas et al. 2012; Panthee, 2012). The success in tomato University of Ghana http://ugspace.ug.edu.gh 15 improvement programmes, either conventional, molecular or tissue culture techniques very much depend on knowledge about the nature and magnitude of genetic variability in agronomic and consumer desired quality attributes of the crop. Equally significant is an understanding of the individual and collective roles of environment and genotype factors in causing variability in crop performance (Mutumpike, 2013). According to Shende et al. (2012), improvement programmes in tomato has over the years relied on such breeding methods as pedigree, hybridization and backcrossing among parents of desirable traits. Such breeding programmes have led to the development of improved hybrid varieties with better yield and resistance to certain pathogens. There have also been value addition to important agronomic characters including higher fruit setting, earliness, uniformity in fruit ripening, adaptation, firmness and long shelf life (Kalloo and Banerjee, 2000; Hazra and Chattopadhyay, 2009; Shende et al., 2012). In recent times important tomato breeding systems combine conventional and nonconventional methods (molecular or biotechnological tools).Whereas the conventional breeding methods involving hybridization are postulated to be a prolonged breeding strategy with limited success rate (Pessarakli and Dris, 2004), the molecular-based techniques are fast with high rate of success. Breeding work in tomato started with selection and progressed through hybridization to molecular techniques. As a result of the growing consumption of fresh tomato fruits and its related products all over the world, efforts are now being made to develop tomato varieties with improved yield and quality traits including high antioxidant composition by employing both conventional and biotechnological means (Romer et al., 2000). 2.10 Evaluation of tomato characters for genetic variability Genetic variability in a given trait is an important requirement for successful breeding programmes to be conducted (Bello et al., 2012). Genetic constitution of any organism or University of Ghana http://ugspace.ug.edu.gh 16 species is composed of the heritable or additive variance, the surrounding environment made up of non-heritable components and the interaction between these two factors. These factors further determine the phenotypic character of an organism. Variability in the phenotypic value is therefore the result of variations due to the genotypic values and deviations arising from the environment whose relative proportions control the genetic distinctiveness of a population (Pradip, 2013). Partitioning the phenotypic value into its components becomes significant in variability studies as it aids in quantifying genotypic parameters including heritability, co-heritability, genetic gain and genotypic correlation. This further gives an idea as to whether selection for any specific trait in cultivar development will be transmitted to the successive generation or not (Jagatpati et al., 2013; Pradip, 2013). Estimation of genotypic coefficient of variation (GCV) gives a true suggestion of the magnitude of genetic variation in a studied population (Mohamed et al., 2012; Vinod et al., 2013). The values of phenotypic and genotypic coefficient of variability can be categorized as high (>20 %), moderate (10 -20) or low (< 10 %) (Reddy et al., 2013). Higher values of phenotypic variance and phenotypic coefficient of variability than the corresponding genotypic component is an indication of the relative role of environment on the expression of characters (Ullah et al., 2011). A study conducted by Nwosu et al. (2014) revealed very small differences in values between PCV and GCV in several traits including days to 1st flowering, days to 50 % flowering, days to fruit ripening, fruit length, fruit per inflorescence, fruit diameter, fruit weight and number of days to fruit maturity. The study thus indicated a lower environmental influence on the expression of the traits, hence the possibility of selecting those traits for further improvement programmes. On the other hand, wider differences in values were recorded between PCV and GCV for some traits including plant height, number of branches per plant, number of fruits per plant and fruit yield. This suggested a greater imposition of environment than genotypic University of Ghana http://ugspace.ug.edu.gh 17 factors on the expression of such traits. Similarly, Osekita and Ademiluyi (2014) recorded a wide difference between phenotypic and genotypic coefficient of variability (PCV and GCV) for some traits including number of fruits per plant, number of clusters per plant, and number of locules per plant. Mohamed et al. (2012) studied heritability and genetic variability for different plants and fruit characters of 30 tomato genotypes and reported significant variability in the studied genotypes for all characters that were scored. High genetic variance as well as phenotypic and genotypic coefficient of variation was observed in fruit weight. The results implied a higher magnitude of variability among the studied traits. However, lowest values were recorded for number of fruits per plant (0.17 and 0.39) and days to 50 % flowering (0.0552 and 0.0885). Genotypic coefficient of variation was high for most characters suggesting a higher contribution of genetic component to total variation. In order to measure genetic variability in tomato for yield and resistance to bacterial wilt disease, Pradeepkumar et al. (2001) estimated genetic variability, heritability and genetic advance for some agronomic and fruit morphological characters. Higher values of genotypic coefficient of variation were recorded for all the characters studied (plant height, number of days to fruit maturity, number of fruits per plant, pericarp thickness, number of locules, total soluble solids, average fruit weight and fruit yield) which suggests an existence of genetic variation among the genotypes. Reddy et al. (2013) observed a wide range of variability for all studied characters. Higher values of phenotypic coefficient of variation (PCV) and genotypic coefficient of variation (GCV) were recorded for most characters including number of clusters per plant, number of fruits per plant, fruit weight and yield, as well as fruit acidity. Number of primary branches per plant, ascorbic acid and TSS were higher (>20 %) for PCV but moderate values were recorded for GCV. Also, moderate PCV and GCV were recorded for plant height, number of University of Ghana http://ugspace.ug.edu.gh 18 flowers per cluster, number of fruits per cluster and fruit length. However, characters such as number of days to 50 % flowering, days to 50 % fruit set and days to fruit maturity had both lower PCV and GCV. Although, values for PCV were higher than their corresponding GCV, differences observed between the PCV and that of the GCV were smaller indicating high genetic control in the expression of the studied traits. Vinod et al. (2013) studied genetic parameters and correlation for yield and quality traits in tomato (Solanum lycopersicum L.) and observed moderate to low values of the genetic parameters estimated in most traits. The phenotypic variance and coefficients of variability were greater than their genotypic counterparts for all traits studied. Higher values of phenotypic and genotypic variances were recorded for plant height, average fruit weight and number of fruits per plant establishing a greater contribution of genetic constitution to total variation. Shushay et al. (2013) evaluated 36 tomato genotypes for genetic variability and association of characters in tomato (Solanum lycopersicum L.) and reported highly significant variation among the genotypes for all traits studied. Generally the phenotypic variances and coefficient of variations were higher than the corresponding genotypic values. Genotypic and phenotypic coefficient of variation estimated were higher (>20 %) for most traits including number of matured fruits per plant, fruit set percentage, total yield per hectare, number of fruit clusters per plant, average weight of fruits per plant and single fruit weight per plant. This indicated the potential for improvement of the traits via selection. Number of primary branches per plant, fruit polar diameter, fruit equatorial diameter, number of days to flowering, days to 50 % flowering and total soluble solids recorded medium values (10–20 %) for phenotypic and genotypic coefficient of variation. Also, the difference in value between PCV and GCV was found to be high for number of primary branches per plant, number of fruits per cluster, average weight of fruits per plant and single fruit weight per plant. This suggests that the University of Ghana http://ugspace.ug.edu.gh 19 expression of the traits were affected by environmental factors. Nonetheless, differences in value between PCV and GCV were observed to be narrow for total soluble solids, number of matured fruits per plant, fruit set percentage, and fruit equatorial diameter suggesting a negligible effect of environment on the expression of the traits. 2.11 The concept of heritability and genetic advance Heritability indicates the proportion of the total variability that exists among species as a result of genetic factors whereas genetic advance measures the degree of genetic gain. Heritability estimated along with genetic gain gives a good indication of the gene action involved in the inheritance of a character. Jagatpati et al. (2013) stated that the extent of heritability estimated for different parameters is dictated by the genotypic make up of the genetic materials and this is required for choosing breeding strategies. Heritability of a trait has been categorized as high, medium (moderate) and low when the values estimated are > 60 %, 30 – 60 % and < 30 % respectively ( Reddy et al., 2013). Similarly, values of genetic gain (expressed as genetic advance as percentage of mean) has been classified as high (>20 %), moderate (10 – 20 %) and low (<10 %) (Reddy et al., 2013). A trait with very high heritability value (≥80 %) suggests a close correspondence between the genotype and the phenotype. This permits easy selection for the trait as the effects of environment on the expression of the traits is minimal (Shushay et al., 2013) Variation in a particular trait resulting from genotypic features is best determined by estimating heritability. Estimate of heritability enables a breeder to determine the phenotypic variability of a trait coming from the contribution of inherent genetic factors. Heritability has a predictive role and expresses the reliability of the phenotypic value. Whereas high values of heritability and genetic advance for a trait gives an idea about the possibility of selection in University of Ghana http://ugspace.ug.edu.gh 20 early generations of the character, low values of heritability indicate limited scope of selection of characters for genetic improvement (Ullah, 2010). Higher heritability value along with higher genetic advance gives a better indication of additive gene action for a trait and consequently allows effective selection in early generations (Atnafua, 2013; Farshadfar and Esterghari, 2014) using simple methods like pure line, mass selection, bulk or SSD (Reddy et al., 2013). Similar reports have been made by Manju and Sreelathakumary (2002) that higher values of heritability suggest existence of fixable additive factors which can be improved upon by effective selection. For selection to be effectively done, heritability must accompany genetic advance. High heritability value accompanied by low genetic advance suggests the presence of non-additive gene action. On the other hand, higher values of GCV, heritability and genetic advance are better indication for selection of traits for improvement programmes. Riaz et al. (2013) reported that higher estimate of genetic advance is an indication that additive genes control the character hence the character can be improved upon through selection. However, a high value of heritability along with low genetic gain suggests that non additive gene action controls the expression of the trait and such a phenomenon does not permit simple selection methods. Rather, improvement of such traits could be amenable to hybrid development or heterosis breeding procedures involving the development of transgressive segregants (Reddy et al., 2013). Anshuman et al. (2013) stated that information obtained about heritability along with genetic advance provides a better scope of selection. Genotypic coefficient of variability which indicates all the genotypic variability transmitted from parents to successive generation is manifested in heritability estimated. An environmental effect on a character is known to be higher when the estimated broad sense heritability is lower. Estimation of heritability alone does not give any indication regarding the amount of gain or improvement that will result University of Ghana http://ugspace.ug.edu.gh 21 from selection of individual genotypes. Understanding heritability in broad sense along with genetic advance is therefore valuable to arrive at a better conclusion (Ajmal et al., 2009). Shushay et al. (2013) recorded very high heritability values (> 80 %) in tomato genotypes along with high genetic gain (> 20 %) for most traits studied suggesting a minimal influence of environment on their expression. For instance, number of mature fruits per plant, fruit set percentage, number of days to fruit maturity, total fruit yield, number of days to 50 % fruiting, number of locules per fruit, plant height, number of flowers per plant and total soluble solids recorded very high heritability estimates. Mehta and Asati (2008) observed high broad sense heritability estimates for plant height, number of fruits per cluster, fruit weight per plant, total fruit yield, number of locules as well as total soluble solids content. Nwosu et al. (2014) evaluated 19 tomato genotypes for genetic variability and correlation and recorded higher values for heritability and genetic advance for most characters studied including days to 50 % flowering, fruits per plant, fruit length and diameter, as well as fruit weight. It could be inferred therefore that the inherent genetic effects on the phenotypic expression of such traits are basically additive and consequently, a greater and effective selection response for those characters is possible (Farshadfar and Estehghari, 2014). High broad sense heritability values were estimated for most characters (plant height, average fruit weight, number of branches per plant and number of days to 50 % flowering) in a study conducted by Mohamed et al. (2012) giving an indication that the characters were influenced by additive gene action valuable for making selection decisions. In the study, Plant height recorded the highest heritability value (97%). Osekita et al. (2014) recorded different magnitudes of heritability ranging from low, moderate and high. Heritability was lowest for number of cluster per plant (2.60) and highest for number of days to 50 % flowering. Fruit shape index, plant height, average fruit weight and pericarp thickness also recorded higher heritability values (> 60 %). University of Ghana http://ugspace.ug.edu.gh 22 Pradeepkumar et al. (2001) observed higher values of heritability along with high genetic advance for all parameters studied in an experiment that evaluated some tomato genotypes for variability in yield and bacterial wilt resistance. The results indicated that environmental factors played a lesser significant role in the expression of the traits. The higher value estimated for genetic advance also implied that additive gene action was higher. Effective selection of the traits for further crop improvement programmes is therefore possible. In a study that evaluated 19 tomato genotypes for variability, Reddy et al. (2013) observed high values of heritability (> 60 %) along with high genetic advance (> 20 %) for most of the characters including plant height, cluster number per plant, number of flowers per cluster, number of fruits per cluster, number of fruits per plant, fruit length, yield per plant, ascorbic acid, TA and TSS. Number of primary branches per plant, days to fruit maturity, and fruit width recorded moderate values of heritability (30-60 %). The results suggest higher contribution of additive gene action than environmental factors to the expression of the traits, hence a potential for selection of the traits for further improvement. Whereas number of days to 50 % flowering, days to first fruit set and fruit width recorded moderate (10-20 %) values of genetic gain, days to fruit maturity (first and last harvests) had low genetic gain (< 10 %). A study by Hadhayatullah et al. (2008) revealed high heritability estimated for plant height, number of fruits per plant, fruit weight per plant, fruit length and diameter, single fruit weight pericarp thickness, number of locules and total soluble solids in tomatoes. 2.12 Genotype x Environment Interaction in Variability Studies Environmental factors coupled with the genotype of a plant are responsible for the existence of variability in the performance of a crop. The consequences associated with interaction between genotype and environment may pose difficulties in deciding on better performing genotypes in terms of desirable characters for any given environment (Mutumpike, 2013). For a homogeneous environment where genotype-by-environment interaction is absent, a top University of Ghana http://ugspace.ug.edu.gh 23 performing genotype will definitely perform well irrespective of location. On the other hand, as a result of the existence of genotype-by-environment interaction, the performance of a given genotype will vary across different environments. Understanding the magnitude of genotype, environment and their interactive effects is essential requirement for establishing the possibility of developing desirable cultivars with high environmental stability or for specific environments. Apart from this, Genotype -by- environment interaction studies to discover variation among genotypes is required in selecting new and best cultivars with respect to their desirable quality traits (Rosello et al., 2011). The growing interest of consumers and processors for quality tomato fruits has been recognized and thus tomato breeders are making efforts to develop varieties with both improved yield and quality traits such as improvement in the content of carotenoids, lycopene, ascorbic acid, TSS, TA and many other antioxidant components in recent times (Causse et al., 2002). Knowledge about differences in performance of different varieties under different growing environments is valuable to any crop improvement programme by facilitating the identification of optimal location for future breeding. Similarly, Panthee et al. (2012) reported that in addition to inherent genetic factors which bring about variability among crop genotypes, the growing environment or location (including climatic, soil and cultural management conditions) have significant impact on a crop’s performance. As a result, the performance of a genotype may not necessarily be the same across diverse environments. This phenomenon is reported to be due to the interaction between the genotype and the environment and requires adequate understanding in order to plan developing cultivars that are adapted to or perform well across multi environments. This is valuable in selection that targets variables including lycopene, ascorbic acid, soluble solids and TA (Panthee et al., 2012). University of Ghana http://ugspace.ug.edu.gh 24 Panthee et al. (2012) determined the magnitude of genotype x environment interactions affecting tomato fruit quality in three locations and reported significant variability among the tomato genotypes involved as well as the location and their interaction. The study further revealed a higher environmental effect on lycopene contents but least affected titrable acidity. The study also indicated that no location-specific pattern of performance was noticed among the genotypes thus proofing the existence of genotype x environment interaction. Rosello et al. (2011) reported a significant influence of environment on the expression of phenotypic traits of tomato and a higher genotypic along with a considerable amount of GXE interaction variation in a study that evaluated genotype, environment and their interaction on carotenoids and ascorbic acid accumulation in tomato germplasm. Results also indicated that phenotypic expression of traits was significantly influenced by environmental impacts. University of Ghana http://ugspace.ug.edu.gh 25 CHAPTER THREE 3.0 MATERIALS AND METHODS 3.1 Experimental Location Two experiments were conducted simultaneously under field and greenhouse growing conditions at the University of Ghana Forest and Horticultural Crops Research Centre (FOHCREC), Okumaning-Kade in the Eastern Region of Ghana between November, 2014 and March, 2015. FOHCREC-Kade is in the Kwaebibrim District and forms part of the forest agro-ecological zone of Ghana (Ofosu-Budu, 2003; Nkansah et al., 2007). Geographically, the study area is located between latitude 60, 0854’N and longitude 00, 5400’W at about 114 m above sea level. The soils of the area are predominantly Haplic Acrisol (FAO/UNESCO, 1990). The mean annual rainfall ranges between 1200 mm-1300 mm and has a bimodal rainfall pattern of two peaks that occurs around June to July and September to October. It has a mean annual temperature of 25-38 0C (Ofosu-Budu, 2003). 3.2 Genetic Materials A total of 20 tomato genotypes were evaluated for their genetic variability in agronomic and fruit quality traits. The source of the genetic materials, type and growth habits are shown in Table 3.1. 3.3 Nursery and Nursery Management Practices Seedlings of all the twenty (20) tomato genotypes were first produced under greenhouse conditions. Plastic seed trays consisting of 98 cells filled with carbonated rice husk (biochar) as growth medium was used to raise seedlings. Prior to seed sowing, the seed trays and the growth medium were sterilized using sodium hypochlorite solution in order to control microbial and fungal infection. Daily watering of seedlings (early morning and late afternoon) was carried out following germination using watering can. University of Ghana http://ugspace.ug.edu.gh 26 Table 3. 1 Tomato genotypes used in the study GENETIC MATERIAL SOURCE GROWTH HABIT WOSOWOSO UNIVERSITY OF GHANA INDETERMINATE NKANSAH HT FOHCREC DETERMINATE PECTOMECH AGRISEED COMPANY LTD DETERMINATE ROMA AGRISEED COMPANY LTD DETERMINATE ROMA VF AGRISEED COMPANY LTD DETERMINATE BUFFALO AGRISEED COMPANY LTD DETERMINATE 11-172 FOHCREC INDETERMINATE L11 FOHCREC INDETERMINATE NS 504 FOHCREC DETERMINATE #20880 FOHCREC INDETERMINATE SHAKTIMAN FOHCREC INDETERMINATE HEINZ-1370 AGRISEED COMPANY LTD DETERMINAT CHERRY AGRISEED COMPANY LTD DETERMINATE MONGAL F1 AGRISEED COMPANY LTD DETERMINATE NIRVANA F1 DIZENGOFF GHANA LTD SEMI-DETERMINATE INLAY F1 AGRISEED COMPANY LTD INDETERMINATE PLATINUM F1 AGRISEED COMPANY LTD DETERMINATE THORGAL F1 AGRISEED COMPANY LTD DETERMINATE COBBRA F1 AGRISEED COMPANY LTD DETERMINATE SUMO F1 AGRISEED COMPANY LTD DETERMINATE When seedlings were 7 days old, the fungicide, Mancozep 80 WP (Mancozeb dithiocarbonate) at 10 g per litre of water was sprayed on seedlings to control fungal infection especially damping off. This was repeated at days 12 and 17 before healthy and uniform University of Ghana http://ugspace.ug.edu.gh 27 seedlings were transferred to the experimental sites at day 21. The chemical fertilizer, N.P.K. (19:19:19) at the rate of 70-90 g/15 L was applied at weekly intervals to provide nutrients to the seedlings. 3.4 Experimental Design and Field Layout The 20 tomato genotypes were arranged in a Randomized Complete Block Design (RCBD) with three replications. The seedlings were transplanted in a spacing of 30 cm x 40 cm in both field and greenhouse experimental locations. In the field experiment, each genotype was transplanted in four rows of five plants while under greenhouse conditions, seedlings were transplanted in two drip rows of five plants each. Data were collected from 6 tagged plants in the middle rows. The climatic conditions (mean monthly temperature, relative humidity, and rainfall (for field experiment only) prevalent in the two experimental sites were recorded (Tables 3.2 and 3.3). Table 3. 2 Climatic data in the greenhouse environment Temperature (0C) Relative humidity (%) Month Maximum Minimum Maximum Minimum November, 2014 40.70 16.70 79.89 31.23 December, 2014 39.26 16.70 81.20 34.23 January, 2015 40.65 16.68 88.40 33.25 February, 2015 40.34 16.71 82.30 34.24 SOURCE: FOHCREC University of Ghana http://ugspace.ug.edu.gh 28 Table 3. 3 Climatic data at the field experimental location Temperature ( 0C) Relative humidity (%) Rainfall Month Maximum Minimum Maximum Minimum (mm) November, 2014 34.40 23.51 94.98 39.52 0.47 December, 2024 34.57 22.79 99.11 45.15 1.90 January, 2015 34.57 22.54 98.64 46.19 2.11 February, 2015 34.80 22.60 99.43 47.75 0.70 SOURCE: FOHCREC 3.5 Preparation of Experimental Sites 3.5.1 Field experiment A site previously cultivated to cucumber was cleared and burnt. A composite soil sample taken from 15 locations in the field at 0-20 cm depth was analyzed for physical and chemical properties using standard laboratory procedures. Soil analysis was carried out in the laboratory of the Department of Soil Science, University of Ghana, Legon. The samples were adequately mixed, air dried, ground and sieved on a 2.0 mm mesh before analysis. Table 3.4 shows the physical and chemical properties of the soil at the field experimental site. The experimental field was tilled thoroughly before lining and pegging of the field was carried out. University of Ghana http://ugspace.ug.edu.gh 29 Table 3. 4 Physicochemical properties of soil at the field experimental location S/N Property Value 1 Texture Sandy clay loam 2 Sand 9%) 52.1 3 Silt (%) 25.9 4 Clay (%) 22.0 5 pH 6.3 6 Total nitrogen (%) 0.6 7 Available phosphorus (mg/kg) 7.74 8 Available potassium (cmol/kg) 0.6 9 Organic carbon 2.26 10 Electrical conductivity (ds/m) 0.73 3.5.2. Greenhouse experiment Beds which have previously been used to cultivate tomato was removed and replaced with fresh soil medium. A composite soil sample was analysed for physical and chemical composition (Table 3.5). Crops were planted on raised beds of approximately 21.5 m length, 0.3 m width and 0.2 m high. The greenhouse area was 24 m x 10 m and consisted of a total of 7 beds each of which was made up of 2 rows. The entire greenhouse environment was sterilized by hot water treatment and sprayed with insecticide (cydim super) both within and outside the surroundings. 3.6 Transplanting Three weeks after seed sowing, uniform and healthy seedlings of each genotype were transplanted in the experimental sites. Transplanted seedlings had an average height of 10-13 cm with 5-6 leaves. Prior to transplanting, seedlings were well watered before removal from the nursery to the field. In each case, watering was carried out immediately after transplanting in order to avoid stress. University of Ghana http://ugspace.ug.edu.gh 30 Table 3. 5 Physicochemical properties of soil in the greenhouse environment S/N Property Value 1 Texture Sandy clay loam 2 Sand 9%) 41.45 3 Silt (%) 29.92 4 Clay (%) 27.10 5 pH 5.10 6 Total nitrogen (%) 0.13 7 Available phosphorus (mg/kg) 10.18 8 Available potassium (cmol/kg) 0.54 9 Organic carbon 1.53 10 Electrical conductivity (ds/m) 0.61 3.7 Agronomic Practice Daily watering of transplanted seedlings was carried out throughout the experiment early morning and late afternoon. Immediately after transplanting, crop starter solution was applied to seedlings at the rate of 60-150 ml/15 L to enable seedlings overcome transplanting shock. The N.P.K. fertilizer (19:19:19) was applied in solution to each transplanted seedling at two weeks after transplanting at the rate of 70-90 g/15 L of water and repeated every 2 weeks as a vegetative booster. Sulphate of ammonia was also applied at the rate of 125 kg /ha in split doses during the reproductive growth stage of the crop. Also, multi K (80 %) was applied to crops at the rate of 150-200 g/15 L of water once every two weeks as fertigation at flowering and fruiting stages of the crops according to the manufacturers instruction. A fruit set enhancer was sprayed on the plants at the rate of 100-150 ml/15 L of water during flowering and fruit formation stages of the crop following the manufacturer’s instruction. Cydim super (an insecticide) was sprayed on crops at the rate of 35 ml/15 L of water at weekly interval to control various kinds of insects in both the field and the greenhouse University of Ghana http://ugspace.ug.edu.gh 31 experimental locations. Cuprofix 30 (fungicide and bactericide) at the rate of 40-60 g/15 L of water was also sprayed on crops at pre-harvest interval of 10-14 days to control fungal and bacterial infection. Weeds were regularly removed with hoe as and when necessary and lower yellowish leaves pruned just after data on number of leaves at the reproductive stage had been scored. Tomato plants were provided with support in the form of trellis in both experiments. Fruits were harvested by hand picking when they had reached their physiological maturity (fully ripe) stage. Harvesting was done 3 times at an interval of 5 days until fruits of all varieties had been picked. 3.8 Data Collection 3.8.1 Agronomic traits Mean plant height (PHT): This was determined as the mean perpendicular height (cm) of six tagged plants measured from the soil level to the tip of the shoot at reproductive stage (30 DAT) of the crop using a meter rule. Mean stem diameter (SD): This was determined as the mean diameter (cm) of six tagged plants measured at about 10 cm from the base of the plant using Vernier caliper at reproductive stage (30 DAT) of the crop. Mean number of leaves per plant (NL): This was determined as the mean number of leaves of six tagged plants counted at reproductive (30 DAT) growth stage of the crop. Mean chlorophyll content (CC): Chlorophyll content (nm) was determined on six tagged plants at reproductive stage of the plant growth (30 DAT) using the chlorophyll meter and the mean value determined University of Ghana http://ugspace.ug.edu.gh 32 Mean number of primary branches per plant (NPB): The number of primary branches of six tagged plants was counted at the end of the growing season and the mean number of branches per plant determined. Mean number of days to first flowering (NFF): This was determined by counting the number of days from transplanting to first flower emergence of each genotype per replication Mean number of days to 50 % flowering (FPF): This was determined by counting the number of days from transplanting until 50 % of the tagged plants per genotype had flowered. Mean number of days to fruit set (NFS): This was determined by counting the number of days from transplanting to first fruit appearance. Mean number of days to 50 % fruit set (FPFS): This was determined by counting the number of days from transplanting until 50 % of tagged plants per plot had fruited. Mean number of days to fruit maturity (DFM): This was determined by counting the number of days from transplanting until 50 % of the tagged plants had at least one fruit ripened (at breaker stage) Mean number of trusses per plant (TPP): The number of trusses per plant was counted on tagged plants and the mean number of trusses per plant determined Mean number of flowers per truss (FPT): The number of flowers per truss was counted for 10 trusses on each of the tagged plants and the mean number determined Mean number of fruits per truss (NFPT): The number of fruits per truss was counted for each of the six tagged plants and the mean number per truss determined University of Ghana http://ugspace.ug.edu.gh 33 Mean fruit set percentage (FSP): Fruit set percentage was determined by dividing the number of fruits by the number of flowers per cluster and the mean computed and expressed in percentage. Mean number of fruits per plant (FPP): Number of fruits per plant was determined by counting the number of harvested fruits from each of the six recorded plants and the mean number determined. Mean single fruit weight per plant (SFW): Single fruit weight was determined as total weight (g) of fruits harvested from tagged plants divided by the total number of fruits harvested from tagged plants Mean fruit weight per plant (FWP): Mean fruit weight (g) per plant was determined by weighing the harvested fruits from the six tagged plants using an electronic balance and the mean determined Mean fruit yield per plant (YPP): The mean fruit weight for each of the six record plants was used to calculate yield per hectare and converted to yield in tonnes per hectare (t/ha) 3.8.2 Fruit physical and quality traits Fruit samples per genotype were collected separately from the experiments conducted under each of the greenhouse and field conditions during the peak period of harvest when the fruits were fully ripe and were in their mature stage. Composite samples of 10 fruits were taken from the selected tagged plants from all the three (3) replications to determine fruit physical and quality characters. Fruit physical characters were scored shortly after harvest while composite samples of fruits per genotype from replications were carried to the laboratory for chemical trait analysis. University of Ghana http://ugspace.ug.edu.gh 34 Tomato fruit physical characteristics Mean fruit length (FL): Fruit length (cm) was recorded at harvest from ten fruits per genotype, from the stem end (pedicel attachment) to the blossom end (apex) using Vernier caliper and the mean determined. Mean fruit diameter (FD): Fruit diameter (cm) was recorded at harvest from ten fruits at the largest diameter of cross-sectional fruits to one decimal place using Vernier caliper and the mean determined. Mean fruit shape index (FSI): This was determined by dividing the mean polar diameter (fruit length) value by the mean equatorial diameter (fruit diameter) value of the fruit. Mean pericarp thickness (PTK): This was determined by making a transverse cut through 10 randomly selected fruits per each genotype and the thickness of the pericarp measured using Vernier caliper at four cardinal points and their mean determined. Mean number of locules (LOC): This was determined by counting the number of locules from 10 randomly dissected fruits of each genotype per replication and their mean determined. Mean fruit firmness (FF): This was determined shortly after harvest using a hand held penetrometer (HANNA Instruments; model GY-3 of higher precision). Readings (kg/cm2) were recorded on 3 fruits per genotype per replication and the mean value determined. Tomato fruit quality characteristics Mean fruit dry matter content (total solids): Dry matter content of fruits (g) was determined by drying 5 g of fresh fruits in an oven set at 70 0C to a constant weight. University of Ghana http://ugspace.ug.edu.gh 35 Determination of pH and total soluble solids (TSS) Sample preparation Ten fruits per genotype per replication were selected separately from each experimental location. The fruits were thoroughly washed with distilled water; cut open and macerated using a blender (DOUBLE-M German Superior quality multifunctional blender DM-106 A). Macerated samples were then used to determine each of the fruit quality traits. Determination of pH Sample juice extracted from blended fruits per each genotype per location was poured into separate beakers. A digital pH meter (JENWAY 3520) was then used for the pH readings. The readings were taken in triplicates and the average for each replication determined. Determination of total soluble solids content (TSS) The total soluble solids content was determined by placing a thin film of blended tomato sample on a DIGITAL BENCH MODEL refractometer (HANNA Instruments HI 96801 0- 85) and the total soluble solids content scored in % 0Brix units (percentage solids). In each case, measurements were taken in triplicates for each sample by recording the readings on the prism scale. Before each sample test was carried out, the prism plate was cleaned with distilled water and wiped dry with soft tissue. The activity was done in triplicates per genotype per replication and the mean value of the total soluble solids content computed. Determination of acidity and malic acid content Acidity of tomato samples was determined by diluting 3 ml of the extracted fruit juice with 10 ml distilled water. The resulting mixture was then titrated against 0.1 M NaOH using phenolphthalein indicator until the content of the conical flask changed from colourless to a University of Ghana http://ugspace.ug.edu.gh 36 stable pink. Titration was carried out in triplicates for each genotype and the mean of the titre values determined. Total acidity content (expressed as citric acid concentration) and malic acid content were calculated according to the procedure of Hawkins (n.d.) as indicated below: 𝑇𝑖𝑡𝑟𝑎𝑏𝑙𝑒 𝑎𝑐𝑖𝑑𝑖𝑡𝑦 (𝑚𝑔/𝑚𝑙 𝑐𝑖𝑡𝑟𝑖𝑐 𝑎𝑐𝑖𝑑) = 𝑇𝑖𝑡𝑟𝑒 𝑥 𝑎𝑐𝑖𝑑 𝑓𝑎𝑐𝑡𝑜𝑟 0.0064 𝑥 100 𝑥 10 10 (𝑚𝑙 𝑗𝑢𝑖𝑐𝑒 ) source: Hawkins (n.d.) Malic acid content (mg/ml) = 𝑇𝑖𝑡𝑟𝑒 𝑥 𝑎𝑐𝑖𝑑 𝑓𝑎𝑐𝑡𝑜𝑟 0.0067 𝑥 100 𝑥 10 10 (𝑚𝑙 𝑗𝑢𝑖𝑐𝑒 ) source: Hawkins (n.d.) Determination of flavour indicator (TSS/TA) Flavour indicator (TSS/TA) was determined by dividing sugar concentration (% 0Brix) value by the value of the citric acid concentration (g/L) according to the equation provided below: 𝑇𝑆𝑆/𝑇𝐴 𝑟𝑎𝑡𝑖𝑜 = 𝐵𝑟𝑖𝑥 𝑣𝑎𝑙𝑢𝑒 𝐶𝑖𝑡𝑟𝑖𝑐 𝑎𝑐𝑖𝑑 (𝑚𝑙 /𝐿) Determination of ascorbic acid content Standardization of the iodine solution Samples of solid ascorbic acid (0.05 g) were weighed in triplicate and placed in 3 separately labeled conical flasks. A volume of 30 ml distilled water and 5 drops of starch solution were added to the content of each of the flasks. A clean burette was filled with 50 ml iodine solution and titrated against ascorbic acid solution. The initial and the final volume readings from the burette were recorded. The procedure was repeated for the other two ascorbic acid solutions. University of Ghana http://ugspace.ug.edu.gh 37 Preparation of sample Fruit samples of 100 g were cut into pieces topped up with 50 ml of distilled water and blended. A volume of 10 ml of distilled water was added to the sample and then strained through Whatman filter paper. The filtrate was then collected in a beaker and later poured into a graduated measuring cylinder and made up to 100 ml with distilled water. Titration of fruit juice Twenty milliliters (20 ml) of each sample solution was measured into a conical flask and topped up with 25 ml of distilled water followed by 1 ml of starch indicator solution. The sample was then titrated against the standardized iodine solution until a permanent dark blue- black colour was formed due to the starch-iodine complex. Determination of amount of ascorbic acid in the sample Determination of ascorbic acid content in the sample was carried out according to the procedure of Kartz (2013). Step 1 Concentration (M) of Iodine solution (MIodine) 𝑀 𝑖𝑜𝑑𝑖𝑛𝑒 = 𝑚𝑎𝑠𝑠 𝑎𝑠𝑐𝑜𝑟𝑏𝑖𝑐 𝑎𝑐𝑖𝑑 𝑥 1 𝑚𝑜𝑙𝑒 𝑎𝑠𝑐𝑜𝑟𝑏𝑖𝑐 𝑎𝑐𝑖𝑑 176.12 𝑔 (𝑎𝑠𝑐𝑜𝑟𝑏𝑖𝑐 𝑎𝑐𝑖𝑑) 𝑥 1000 𝑚𝑙/𝐿 𝑣𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑖𝑜𝑑𝑖𝑛𝑒(𝑚𝑙) Source: Kartz, 2013 Step 2 Concentration of ascorbic acid in the tomato fruit sample (mg) (Mgascorbic acid) 𝑀𝑔 𝑎𝑠𝑐𝑜𝑟𝑏𝑖𝑐 𝑎𝑐𝑖𝑑 = 𝑀 (𝑖𝑜𝑑𝑖𝑛𝑒 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛) 𝑥 𝑚𝑙 (𝑖𝑜𝑑𝑖𝑛𝑒 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 𝑥 176.12 𝑔/𝑚𝑜𝑙𝑒] Source: Kartz, 2013 Determination of carotene and lycopene content The β-carotene and lycopene contents were determined separately according to the method used by Kipandula et al. (2014). An amount of 5 ml of 70 % methanol was added to 5 g of University of Ghana http://ugspace.ug.edu.gh 38 sample tomato extract and thoroughly shaken for a minute. The content of the resulting mixture was filtered through Whatman No. 4 filter paper. The absorbance of the filtrates were then measured at wave length (λ) = 453, 505 and 663 nm using a spectrophotometer. β- carotene and lycopene contents were then calculated according to the following equations: β-Carotene (mg/100 ml) = 0.216 A663 - 0.304 A505 + 0.452 A453 Lycopene (mg/100 ml) = -0.0458 A663 + 0.372 A505 - 0.0806 A453 Determination of phenolic acid content The composition of phenolic compounds in the methanolic extracts of tomato samples was determined according to the Folin-Ciocalteau spectrophotometric process (Harbone, 1973). Distilled water (1200 µl) and aqueous sodium carbonate solution (450 µl) were added to each sample (150 µl, 10 mg/ml). The Folin-Ciocalteau reagent (100 µl) was then added to the mixture and vortexed. The content was made to stand for 1 hr 30 min after which absorbance were read at the respective wave lengths for each phenolic acid compound using UV/visible spectrophotometer (SpectraMax plus 384, United States). The concentration of the individual phenolic acid compounds was determined according to their standard curves of linear equations (Table 3.6). Table 3. 6 Phenolic acid compounds and their standard curves Phenolic aid compounds Standard curve Gallic acid (mg/100 ml) Y = 0.0871x - 0.102 Vanillic acid (mg/100 ml) Y = 0.053x + 0.012 Rosmarinic acid (mg/100 ml) Y = 0.069x + 0.022 University of Ghana http://ugspace.ug.edu.gh 39 Determination of flavonoid content The modified aluminium chloride colorimetric procedure of Barros et al. (2007) was used for the determination of flavonoid content in the tomato samples. Sample extract (100 µl, 100 mg/ml) was added to 500 µl of distilled water and sodium nitrite, NaNO2 (5%, 30 µl). The resulting mixture was made to stand for 5 minutes after which a solution of aluminium chloride AlCl3. H2O (10 %, 30 µl) was added to the mixture. Again the mixture was allowed to stand for 6 minutes after which sodium hydroxide, NaOH (1 M, 200 µl) and 110 µl distilled water were added to the solution and vortexed. Measurements of absorbance of solution was made at various wave length (SpectraMax plus 384, Unitd States). The concentration of individual flavonoid compounds was calculated according to their respective standard curves and the results expressed as mg/ml of the extract (Table 3.7). Table 3.7 Flavonoid compounds and their standard curves Flavonoid compound Wave length (nm) Standard curves Rutin (mg/100 ml) 425 Y = 0.010x + 0.223 Quercetin (mg/100 ml) 415 Y = 0.015x + 0.246 Catechin (mg/100 ml) 276 Y = 0.011x + 0.0154 Hesperitin (mg/100 ml) 350 Y = 0.0067x + 0.466 Kaemferrol (mg/100 ml) 368 Y = 0.028x + 0.435 Myricetin (mg/100 ml) 266 Y = 0.025x + 0.226 Genestein (mg/100 ml) 260 Y = 0.021x + 0.190 Determination of antioxidant activities An amount of 5 µl of DPPH solution was added to 100 µl of methanolic sample extracts. An amount of 1 ml of methanolic extract was added to 0.002 % DPPH prepared solution The same amount of the sample extract was added to the standard solution to be tested separately. The resulting mixtures were allowed to stand in the dark for 20 min after which optical University of Ghana http://ugspace.ug.edu.gh 40 density was measured at 517 nm using spectrophotometer against methanol. Percentage inhibition was calculated from the optical density record values according to the formula as given below: Percentage inhibition of DPPH activity = (A – B/A) x 100 where A = optical density of the blank and B = optical density of sample. Analysis was carried out in triplicates and the results expressed in mean values. 3.9 Statistical Analysis of Data Data collected under field and greenhouse growing conditions were analyzed for variances. Both individual and combined analyses of variance were performed. The Fisher’s Least Significant Difference, LSD was used to separate means that differed significantly based on parameters gathered following the procedure described by Gomez and Gomez 1984). GENSTAT statistical software (12th edition) was used for the data analysis. The variance component format was employed in the computation of variances, coefficient of variability, heritability and genetic advance. 3.9.1 Estimation of variance components 3.9.1.1 Individual analysis of variance Variances of each trait were estimated separately for field and greenhouse data based on the RCBD analysis of variance following the format in Table 3.8 (Usman, 2013). The genotypic, environmental and phenotypic variances were computed from mean squares as per the methods suggested by Farshadfar and Estehghari (2014) (Table 3.9). Broad sense heritability (h2b) was estimated for all the key traits following the method adopted by (Farshadfar and Estehghari, 2014) (Table 3.9). University of Ghana http://ugspace.ug.edu.gh 41 Table 3. 8 Format for individual analysis of variance Source of variation df MS EMS F-test Replication r-1 MSR σ 2 E + gσ 2 R Genotype g-1 MSG σ 2 E + rσ 2 G MSG/MSE Error (r-1) (g-1) MSE σ 2 E Total gr-1 Where df = degree of freedom; MS = mean square; EMS = expected mean square; MSR = mean square due to replication; MSG = mean square due to genotypes; MSE = mean square of error; σ2G, σ2R and σ 2 E are variances due to genotype, replication and Error respectively; r = Number of replications; g = number of genotypes. Source: Usman (2013) Table 3. 9 Estimation of variances and broad sense heritability on individual location basis Genotypic parameter Symbol Determination method Environmental variance σ2E MSE Genotypic variance σ2G (MSG - MSE)/r Phenotypic variance σ2P σ 2 p = σ 2 G + σE 2 Broad sense heritability h2b σ 2 G/ σ 2 P Source: Farshadfar and Estehghari (2014) 3.9.1.2 Combined analysis of variance A two-factor analysis of variance (ANOVA) was performed for the combined data scored on all traits from the two experiments (greenhouse and open field) following the procedure described by Ntawuruhunga and Dixon (2010) (Table 3.10) for the randomized complete block design (RCBD). Fisher’s Least Significant Difference (LSD) was used for the mean comparison. Based on the model used, Replication (R) and Location (L) were treated as random effect whereas genotype (G) was treated as fixed effect. The linear additive model used was; University of Ghana http://ugspace.ug.edu.gh 42 Yijk = µ + Gi + Lj + (G x L)ij + Rj(k) + Eijk where, Yijk is the observation on the ith genotype in the jth location in the kth replication, µ is the general mean, Gi is the fixed effect of the jth genotype, Lj is the effect of the jth location, (G x L)ij is the interaction of the jth genotype with jth location, Rk(j) is the effect of kth randomized block within the jth location and Eijk is the experimental error associated within the ijkth observation. Table 3. 10 Format for a two-factor (combined) analysis of variance Source of variation df MS EMS F-test Replication (R) r-1 MSR Replication (Location) r(l-1) MSR(L) Location (L) l-1 MSL σ 2 E + rσ 2 GL + rg σ 2 L MSL/MSGE Genotype (G) g-1 MSG σ 2 E + rσ 2 GL + rl σ 2 G MSG/MSE Genotype x Location (G x L) (g-1) (l-1) MSGE σ 2 E + r σ 2 GL MSGE/MSE Residual (gl-1) (r-1) MSE σ 2 E Total glr-1 Where df = degree of freedom; MS = Mean square; EMS = expected mean square; MSR = Mean square due to replication; MSL = Mean square due to location; MSG = Mean square due to Genotype; MSGL= Mean square due to genotype x location; MSE = Error mean square; σ 2 L = Location variance, σ2GL = Genotype by location variance, σ 2 E = Error variance, r = Number of replications, l = Number of locations and g = Number of genotypes. Source: Ntawuruhunga and Dixon (2010) 3.9.1.3 Estimation of variances and heritability for combined data The estimation of variance components for genotype, phenotype, environment (locations), and genotype x location interaction as well as heritability (h2b) were carried out following the two-factor analysis of variance method as adopted by (Usman, 2013) (Table 3.11). University of Ghana http://ugspace.ug.edu.gh 43 Table 3. 11 Estimation of variance and broad sense heritability on combined location basis Genotypic parameter Symbol Determination method Environmental variance σ2E MSE Location variance σ2L (MSL – MSGL)/ rg Genotypic variance σ2G (MSG – MSGL)/rl G x L interaction variance σ2GL (MSGL- MSE)/r Phenotypic variance σ2P σ 2 G + (σ 2 GE/l) + (σ 2 E/rl) Broad sense heritability h2b σ 2 G/ σ 2 P Source: Usman (2013) 3.9.1.4 Estimation of coefficient of variability Genotypic (GCV), Environmental (ECV) and phenotypic coefficient of variation (PCV) were computed according to the formulae adopted by Jalal and Ahmad (2011) and Farshadfar and Estehghari (2014) and categorized as high (> 20 %), moderate (10 – 20 % ) and low < 10 % ) following the procedure adopted by Reddy et al. (2013) (Table 3.12) Estimation of expected genetic advance and genetic advance as percentage of mean The expected genetic advance (EGA) for selection intensity (k) at 5% (2.06) and the genetic advance as percentage of mean (GAM) were estimated according to the procedure of Jalal and Ahmad (2011) adopted from Johnson et al. (1955) (Table 3.12). Table 3. 12 Methods for computing coefficient of variability and genetic gain (EGA & GAM) Genotypic parameter Symbol Determination method Genotypic coefficient of variability GCV % (√σ2G/GM)*100 Phenotypic coefficient of variability PCV % (√σ2P/GM)*100 Environmental coefficient of variability ECV % (√σ2E/GM)*100 Expected genetic advance EGA k * √σ2P * h 2 b Genetic advance as percentage of mean GAM (EGA/GM)* 100 University of Ghana http://ugspace.ug.edu.gh 44 3.9.4 Correlation coefficient estimation The Pearson’s correlation coefficient was used to establish the association between selected agronomic and fruit quality characters of the tomato genotypes. University of Ghana http://ugspace.ug.edu.gh 45 CHAPTER FOUR 4.0 RESULTS 4.1 Mean performance of tomato genotypes for agronomic traits across locations 4.1.1 Mean plant height (cm) (PHT) The mean values for plant height at flowering for the 20 tomato genotypes are presented in Table 4.1. The lowest mean plant height of 98.22 cm under greenhouse conditions was recorded by genotype HEINZ-1370 while the highest mean value of 158.83 cm was recorded by genotype L11. Under field conditions, the lowest mean value of 41.06 cm was recorded by genotype NKANSAH HT while the highest mean value of 61.33 cm was recorded by genotype MONGAL F1. The genotype HEINZ-1370 recorded the lowest mean plant height of 71.34 cm across the two locations while the highest mean value of 105.39 cm was recorded by genotype L11. Based on the combined analysis of variance, the effects of genotype, location and genotype x location interaction were significantly (P< 0.01) different (Appendix 11). 4.1.2 Mean stem diameter per plant (cm) (SD) The mean values for stem diameter are presented in Table 4.1. The lowest mean stem diameter value of 6.58 cm under greenhouse conditions was recorded by NIRVANA F1 while the highest mean value of 8.70 cm was recorded by SUMO F1. Under field conditions, the lowest mean value of 6.86cm was recorded by genotype 11 – 172 while the highest mean value of 10.25 cm was recorded by SUMO F1. Across the two locations, the lowest mean stem diameter value of 7.09 cm was recorded by genotype 11-172 while the highest mean value of 9.48 cm was recorded by genotype SUMO F1. Significant (P< 0.01) differences were observed among the 20 genotypes for stem diameter based on the combined analysis of University of Ghana http://ugspace.ug.edu.gh 46 variance. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 11). 4.1.3 Mean number of leaves per plant (NL) The mean values for number of leaves per tomato plant are presented in Table 4.1. The lowest mean value of 23 under greenhouse conditions was recorded by genotype HEINZ- 1370 while the highest mean value of 61 was recorded by the genotype WOSOWOSO. Under field conditions, the lowest mean value of 30 was recorded by genotype THORGAL F1 while the highest mean value of 67 was recorded by genotype NKANSAH HT. Across the two locations, the lowest mean value of 28 was recorded by genotype THORGAL F1 while the highest mean value of 57 was recorded by genotype NKANSAH HT. From the combined analysis of variance, significant differences were observed among the 20 tomato genotypes for the trait. The effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 11). 4.1.4 Mean chlorophyll content (nm) (CC) The mean values for chlorophyll content per tomato plant among the genotypes evaluated are presented in Table 4.2. The lowest mean chlorophyll content of 22.97 nm under greenhouse conditions was recorded by genotype ROMA VF while the highest mean value of 44.74 nm was recorded by genotype #20880. Under field conditions, the lowest mean value of 30.97 nm was recorded by genotype CHERRY while the highest mean value of 48.91 nm was recorded by the genotype NKANSAH HT. Across the two locations, the lowest mean value of 27.96 nm was recorded by genotype CHERRY while the highest mean value of 41.44 nm was recorded by genotype NIRVANA F1. Significant (P< 0.01) differences were observed among the 20 genotypes for chlorophyll content. The combined analysis of variance showed that effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 11). University of Ghana http://ugspace.ug.edu.gh 47 Table 4. 1 Mean Plant height, Stem diameter and Number of leaves per plant Plant height (cm) Stem diameter (cm) Number of leaves Genotype Green house Open field pooled mean Green house Open field pooled mean Green house Open field pooled mean PECTOMECH 115.86 48.38 82.12 6.80 7.89 7.35 25.00 37.00 31.00 PLATINUMF1 144.50 47.11 95.81 7.19 7.64 7.42 34.00 46.00 40.00 NS 504 143.50 54.39 98.95 7.86 8.33 8.10 31.00 33.00 32.00 NKANSAH HT 102.94 41.06 72.00 7.08 8.08 7.58 47.00 67.00 57.00 CHERRY 134.55 49.44 92.00 7.66 7.89 7.78 54.00 54.00 54.00 11-172 139.89 57.22 98.56 7.31 6.86 7.09 33.00 39.00 36.00 ROMA 124.83 48.00 86.42 8.42 8.20 8.31 33.00 47.00 40.00 L11 158.83 51.95 105.39 6.86 7.58 7.22 32.00 40.00 36.00 HEINZ-1370 98.22 44.46 71.34 7.75 10.14 8.95 23.00 43.00 33.00 NIRVANA F1 128.72 52.33 90.53 6.58 8.36 7.47 30.00 42.00 36.00 BUFFALO 141.70 51.41 96.56 7.28 8.05 7.67 36.00 48.00 42.00 INLAY F1 139.22 51.60 95.41 7.40 8.68 8.04 33.00 39.00 36.00 MONGAL F1 130.74 61.33 96.04 7.34 9.28 8.31 43.00 65.00 54.00 #20880 111.87 49.39 80.63 7.18 9.03 8.11 28.00 44.00 36.00 SUMO F1 128.33 45.89 87.11 8.70 10.25 9.48 38.00 42.00 40.00 ROMA VF 109.11 42.48 75.80 8.41 8.42 8.42 35.00 43.00 39.00 SHAKTIMAN 131.78 49.20 90.49 7.32 7.47 7.40 28.00 38.00 33.00 THORGAL F1 124.11 51.78 87.95 7.39 8.42 7.91 26.00 30.00 28.00 COBBRA F1 127.83 50.00 88.92 7.50 7.86 7.68 33.00 33.00 33.00 WOSOWOSO 133.50 49.50 91.50 7.47 8.58 8.03 61.00 45.00 53.00 MEAN 128.24 49.85 89.17 7.48 8.35 7.91 35.00 44.00 40.00 LSD(0.05) 1.95 3.51 2.01 0.40 1.36 0.73 1.26 5.81 2.99 University of Ghana http://ugspace.ug.edu.gh 48 4.1.5 Mean number of primary branches per plant (NPB) The mean values for number of primary branches per plant among the tomato genotypes evaluated are presented in Table 4.2. The lowest mean value of 2 under greenhouse conditions was recorded by genotype SHAKTIMAN while the highest mean value of 6 was recorded by genotype CHERRY. Under field conditions, the lowest mean value was 3 and this was recorded by the 8 genotype while the highest mean value of 5was recorded by genotype CHERRY. Across the two locations, the highest mean value of 6 was recorded by genotype CHERRY while the lowest value of 3 was recorded by 5 genotype. Significant differences were observed among the genotypes based on the combined analysis of variance. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 11). 4.1.6 Mean root length per plant (cm) (RTL) The mean values for root length per tomato plant among the genotypes evaluated are presented (Table 4.2). The lowest mean root length of 15.00 cm under greenhouse conditions was recorded by genotype THORGAL F1 while the highest mean root length of 32.00 cm was recorded by genotype COBBRA F1. Under field conditions, the lowest mean root length was 12.00 and was recorded by the genotype CHERRY while the highest mean value of 27.00 was recorded by genotype SUMO F1. The genotype CHERRY recorded the lowest mean root length of 16.00 cm across the two locations while the highest mean value of 25 cm was recorded by genotype ROMA VF. Combined analysis of variance showed significant (P< 0.01) genotype, location and genotype x location effects (Appendix 11). University of Ghana http://ugspace.ug.edu.gh 49 Table 4. 2 Mean Chlorophyll content, Number of primary branches per plant and Root length Chlorophyll content (nm) Number of primary branches Root length (cm) Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Mean PECTOMECH 25.33 40.02 32.68 3.00 3.00 3.00 20.00 19.00 19.50 PLATINUMF1 32.12 44.12 38.12 3.00 4.00 4.00 16.67 22.00 19.33 NS 504 26.28 39.89 33.09 3.00 4.00 4.00 24.00 16.00 20.00 NKANSAH HT 23.68 48.91 36.30 3.00 4.00 4.00 24.67 15.00 19.83 CHERRY 24.95 30.97 27.96 6.00 5.00 6.00 20.00 12.00 16.00 11-172 32.22 42.53 37.38 3.00 3.00 3.00 22.67 15.00 18.83 ROMA 37.61 38.16 37.89 3.00 4.00 4.00 21.33 20.00 20.67 L11 34.20 40.20 37.20 4.00 4.00 4.00 17.00 25.67 21.33 HEINZ-1370 40.01 37.16 38.59 4.00 3.00 4.00 22.00 15.00 18.50 NIRVANA F1 37.87 45.00 41.44 4.00 3.00 4.00 16.33 22.00 19.17 BUFFALO 26.71 38.15 32.43 3.00 3.00 3.00 17.33 25.00 21.17 INLAY F1 28.46 41.28 34.87 3.00 4.00 4.00 21.00 21.00 21.00 MONGAL F1 32.26 40.04 36.15 4.00 4.00 4.00 20.67 18.00 19.33 #20880 44.74 34.77 39.76 3.00 4.00 4.00 21.53 26.00 23.77 SUMO F1 32.74 39.85 36.30 3.00 4.00 4.00 21.00 27.00 24.00 ROMA VF 22.97 36.44 29.71 3.00 4.00 4.00 31.00 19.00 25.00 SHAKTIMAN 32.26 43.03 37.65 2.00 3.00 3.00 17.33 19.00 18.17 THORGAL F1 28.48 41.39 34.94 3.00 3.00 3.00 15.00 18.00 16.50 COBBRA F1 31.01 37.26 34.14 4.00 3.00 4.00 32.00 15.00 23.50 WOSOWOSO 28.41 41.39 34.90 4.00 5.00 5.00 24.67 20.33 22.50 MEAN 31.12 40.03 35.57 3.00 4.00 4.00 21.31 19.50 20.40 LSD(0.05) 1.72 3.26 1.89 0.40 0.65 0.38 1.48 1.70 1.10 4.1.7 Mean number of days to 1st flowering (NFF) The mean values for number of days to 1st flowering among the tomato genotypes are presented in Table 4.3. The lowest mean value of 15 days under greenhouse conditions was recorded by genotype NKANSAH HT, CHERRY, MONGAL F1, ROMA VF, THORGAL F1 and WOSOWOSO while the highest mean value of 20 was recorded by genotypes PECTOMECH, NIRVANA F1 and #20880. Under field conditions, the lowest mean value of 12 days was recorded by NKANSAH HT while the highest mean value of 24 days was University of Ghana http://ugspace.ug.edu.gh 50 recorded by the genotype 11- 172. Tomato genotypes NKANSAH HT, CHERRY, MONGAL F1 and WOSOWOSO recorded the lowest mean value of 14 days across the two locations while the highest mean value of 22 days was recorded by the genotype 11-172. From the combined analysis of variance, the effects of genotype, location and genotype x location interaction were significantly (P< 0.01) different (Appendix 11). 4.1.8 Mean number of days to 50 % flowering (FPF) The mean values for number of days to 50 % flowering for all the 20 tomato genotypes are presented in Table 4.3. The lowest mean value of 18 days under greenhouse conditions was recorded by genotype NKANSAH HT, CHERRY, MONGAL F1 and COBBRA F1 while the highest mean value of 29 days was recorded by genotype PECTOMECH. Under field conditions, the lowest mean value of 21 days was recorded by genotypes NKANSAH HT and CHERRY while the highest mean value of 29 days was recorded by the genotype ROMA VF. The genotype NKANSAH HT and CHERRY recorded the lowest mean value of 20 days across the two locations while the highest mean value of 27 was recorded by the genotype PECTOMECH. The effects of genotype, location and genotype x location interaction were significantly (P < 0.01) different based on the combined analysis of variance (Appendix 11). 4.1.9 Mean number of days to 1st fruit set (NFS) The mean values for number of days to 1st fruit set among tomato genotypes evaluated are presented in Table 4.3. The lowest mean number of days to 1st fruit set was 26 days under the greenhouse conditions and this was recorded by genotype NKANSAH HT while the highest mean value of 48 days was recorded by genotype INLAY F1. Under field conditions, the lowest mean value of 26 days was recorded by genotype NKANSAH HT while the longest mean number of days to 1st fruit set was 42 days and was recorded by genotype #20880. Across the two locations, the lowest mean value of 26 days was recorded by genotype NKANSAH HT while the highest mean value of 44 days was recorded by genotype #20880. University of Ghana http://ugspace.ug.edu.gh 51 The effects of genotype, location and genotype x location interaction were significantly (P < 0.01) different based on the combined analysis of variance (Appendix 11). 4.1.10 Mean number of days to 50 % fruit set (FPFS) The mean values for number of days to 50 % fruit set are presented in Table 4.4. The lowest mean value of 28 days under greenhouse conditions was recorded by genotype NKANSAH HT while the highest mean value of 56 days was recorded by genotype SUMO F1. Under field conditions, the lowest mean value of 36 days was recorded by genotype NKANSAH HT while the highest mean value of 52 days was recorded by genotype HEINZ-1370. Across the two locations, the lowest mean value of 29 days was recorded by genotype WOSOWOSO while the highest mean value of 52 days was recorded by the genotype L11. Combined analysis of variance showed significant (P < 0.01) genotype, location and genotype x location effects. 4.1.11 Mean percentage fruit set (FSP) The mean values for percentage fruit set among the tomato genotypes evaluated are presented in Table 4.4. The lowest mean percentage fruit set of 38.56 % under greenhouse conditions was recorded by genotype #20880 while the highest mean percentage value of 85 % was recorded by genotype PLATINUM F1. Under field conditions, the lowest mean percentage fruit set of 47.58 % was recorded by genotype NIRVANA F1 while the highest mean percentage value of 65.39 % was recorded by genotype CHERRY. Across the two locations, the lowest mean value of 47.29 % was recorded by genotype NS 504 while the highest mean value of 72.93 % was recorded by genotype PLATINUM F1. The combined analysis of variance showed significant (P < 0.01) genotype and genotype x location effects (Appendix 12). Also, the effect of location was significantly (P < 0.05) different. University of Ghana http://ugspace.ug.edu.gh 52 Table 4. 3 Mean Number of days to 1st flowering, 50 % flowering and Number of days to 1st fruit set Number of days to 1st Flowering Number of days to 50 % flowering Number of days to 1st fruit set Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 20.00 14.00 17.00 29.00 25.00 27.00 36.00 40.00 38.00 PLATINUMF1 16.00 17.00 17.00 19.00 25.00 22.00 35.00 35.00 35.00 NS 504 19.00 22.00 21.00 23.00 27.00 25.00 41.00 37.00 39.00 NKANSAH HT 15.00 12.00 14.00 18.00 21.00 20.00 26.00 26.00 26.00 CHERRY 15.00 13.00 14.00 18.00 21.00 20.00 36.00 36.00 36.00 11-172 19.00 24.00 22.00 23.00 27.00 25.00 45.00 35.00 40.00 ROMA 17.00 13.00 15.00 20.00 23.00 22.00 41.00 39.00 40.00 L11 16.00 17.00 17.00 20.00 23.00 22.00 41.00 39.00 40.00 HEINZ-1370 18.00 17.00 18.00 19.00 27.00 23.00 41.00 41.00 41.00 NIRVANA F1 20.00 18.00 19.00 23.00 27.00 25.00 46.00 40.00 43.00 BUFFALO 19.00 14.00 17.00 23.00 27.00 25.00 45.00 33.00 39.00 INLAY F1 17.00 14.00 16.00 19.00 23.00 21.00 48.00 32.00 40.00 MONGAL F1 15.00 13.00 14.00 18.00 26.00 22.00 30.00 34.00 32.00 #20880 20.00 13.00 17.00 23.00 27.00 25.00 46.00 42.00 44.00 SUMO F1 18.00 16.00 17.00 22.00 26.00 24.00 40.00 40.00 40.00 ROMA VF 15.00 17.00 16.00 25.00 29.00 27.00 37.00 41.00 39.00 SHAKTIMAN 16.00 16.00 16.00 20.00 26.00 23.00 42.00 40.00 41.00 THORGAL F1 15.00 14.00 15.00 19.00 27.00 23.00 37.00 37.00 37.00 COBBRA F1 17.00 17.00 17.00 18.00 24.00 21.00 29.00 35.00 32.00 WOSOWOSO 15.00 13.00 14.00 22.00 22.00 22.00 28.00 27.00 27.00 MEAN 17.00 16.00 17.00 21.00 25.00 23.00 39.00 37.00 38.00 LSD(0.05) 0.68 0.94 0.62 0.92 1.26 0.78 1.00 2.31 1.22 4.1.12 Mean number of days to fruit maturity (DFM) The mean number of days to fruit maturity among the tomato genotypes evaluated has been presented in Table 4.4. The tomato genotype NKANSAH HT recorded the least number of University of Ghana http://ugspace.ug.edu.gh 53 days to fruit maturity under each of the greenhouse and field conditions with mean values of 55 and 56 days respectively. The highest mean value of 77 days to fruit maturity under greenhouse conditions was recorded by the genotype 11–172 whereas genotype 11-172 recorded the maximum value of 78 days under field conditions. Across the two locations, the least mean value of 56 days was recorded by the genotype NKANSAH HT while the highest mean value of 75 days was recorded by the genotypes L11 and NIRVANA F1. Combined analysis of variance showed significant (P < 0.05) difference between locations. The effects of genotype and genotype x location interaction were also significantly (P < 0.01) different (Appendix 12). 4.1.13 Mean number of trusses per plant (TPP) The mean values for number of trusses per plant among the tomato genotypes are shown in Table 4.5. The lowest mean number of trusses recorded was 6.00 under greenhouse conditions and this was recorded by genotype #20880 while the highest mean value of 21.00 was recorded by genotypes NKANSAH HT and MONGAL F1. The lowest mean value of 10.00 under field conditions was recorded by #20880 while the highest mean value of 33 was recorded by genotype NKANSAH HT. Across the two locations, the lowest mean value of 8.00 was recorded by genotype #20880 while the highest mean value of 27.00 was recorded by genotype NKANSAH HT. Significant (P< 0.01) differences were observed among all genotypes evaluated for the trait. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 12). 4.1.14 Mean number of flowers per truss (FPT) The mean values for number of flowers per trusses per tomato plant are presented in Table 4.5. The lowest mean value of 4 under greenhouse conditions was recorded by genotype L11, INLAY F1 and SHAKTIMAN while the highest mean value of 7 was recorded by genotype NKANSAH HT, CHERRY, #20880, COBBRA F1 and WOSOWOSO. University of Ghana http://ugspace.ug.edu.gh 54 Table 4. 4 Mean Number of days to 50 % fruit set, Percentage fruit set and Number of days to fruit maturity Number of days to 50 % fruit set Fruit set percentage (%) Number of days to fruit maturity Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 54.00 46.00 50.00 50.93 55.11 53.02 63.00 63.00 63.00 PLATINUMF1 44.00 42.00 43.00 85.46 60.39 72.93 62.00 65.00 63.50 NS 504 49.00 51.00 50.00 43.04 51.54 47.29 62.00 66.00 64.00 NKANSAH HT 28.00 36.00 32.00 69.43 54.76 62.10 55.00 56.00 56.00 CHERRY 39.00 43.00 41.00 73.86 65.39 69.63 62.00 64.00 63.00 11-172 54.00 44.00 49.00 52.47 53.32 52.90 77.00 71.00 74.00 ROMA 42.00 48.00 45.00 48.61 54.44 51.53 60.00 66.00 63.00 L11 54.00 50.00 52.00 68.75 53.94 60.85 72.00 78.00 75.00 HEINZ-1370 48.00 52.00 50.00 49.55 51.10 50.33 70.00 71.00 71.00 NIRVANA F1 50.00 48.00 49.00 47.93 47.58 47.76 75.00 74.00 75.00 BUFFALO 48.00 44.00 46.00 54.55 54.87 54.71 68.00 67.00 68.00 INLAY F1 54.00 46.00 50.00 60.95 56.55 58.75 71.00 63.00 67.00 MONGAL F1 35.00 39.00 37.00 64.45 54.65 59.55 57.00 63.00 60.00 #20880 54.00 44.00 49.00 38.56 62.59 50.58 73.00 74.00 74.00 SUMO F1 56.00 46.00 51.00 67.52 57.87 62.70 73.00 69.00 71.00 ROMA VF 47.00 47.00 47.00 50.47 60.21 55.39 64.00 65.00 65.00 SHAKTIMAN 44.00 50.00 47.00 69.94 51.61 60.78 63.00 70.00 67.00 THORGAL F1 42.00 44.00 43.00 43.65 54.97 49.31 60.00 58.00 59.00 COBBRA F1 41.00 41.00 41.00 60.16 51.79 55.98 58.00 58.00 58.00 WOSOWOSO 29.00 29.00 29.00 42.85 56.16 49.51 57.00 57.00 57.00 MEAN 46.00 45.00 45.00 57.16 55.39 56.28 65.00 66.00 65.00 LSD(0.05) 1.16 3.50 1.79 12.38 5.89 6.67 0.95 2.23 1.18 University of Ghana http://ugspace.ug.edu.gh 55 Under field conditions, the lowest mean value of 5 was recorded by the genotype #20880, MONGAL F1 and WOSOWOSO while the highest mean value of 9 was recorded by genotype NKANSAH HT. Across the two locations, the lowest mean value of 5 was recorded by genotype L11, INLAY F1, SUMO F1 and SHAKTIMAN while the highest mean value of 8±0.27 was recorded by genotype NKANSAH HT. The effects of genotype, location and genotype x location interaction were significantly (P< 0.01) different (Appendix 12). 4.1.15 Mean number of fruits per truss (NFPT) The mean values for number of fruits per truss among the tomato genotypes evaluated have been presented in Table 4.5. The lowest mean value of 3.00 under greenhouse conditions was recorded by 13 genotypes while the highest mean value of 5.00 was recorded by genotype NKANSAH HT, CHERRY and COBBRA F1. Under field conditions, the lowest mean value of 2.00 was recorded by genotype ROMA while the highest mean value of 5.00 was recorded by NKANSAH HT and CHERRY Across the two locations, the lowest mean value of 3.00 was recorded by 14 genotypes while the highest mean value of 5.00 was recorded by genotype NKANSAH HT and CHERRY. Based on the combined analysis of variance, significant (P < 0.05) difference was observed among the genotypes, locations and genotype x location interaction effects (Appendix 12). 4.1.16 Mean number of fruits per plant (FPP) The mean values for number of fruits per tomato plant are shown in Table 4.6. The lowest mean value of 3.00 was recorded by genotype L11 under greenhouse conditions while the highest mean value of 31.00 was recorded by genotype CHERRY. Under field conditions, the lowest mean value of 5.00 was recorded by genotype L11 and THORGAL F1 while the highest mean value of 35 was recorded by genotype CHERRY. University of Ghana http://ugspace.ug.edu.gh 56 Table 4. 5 Mean Number of trusses per plant, Number of flowers per truss and Number of fruits per truss Number of trusses per plant Number of flowers per truss Number of fruits per truss Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 8.00 12.00 10.00 6.00 6.00 6.00 3.00 3.00 3.00 PLATINUMF1 14.00 16.00 15.00 5.00 7.00 6.00 4.00 4.00 4.00 NS 504 13.00 15.00 14.00 5.00 7.00 6.00 3.00 3.00 3.00 NKANSAH HT 21.00 33.00 27.00 7.00 9.00 8.00 5.00 5.00 5.00 CHERRY 18.00 32.00 25.00 7.00 7.00 7.00 5.00 5.00 5.00 11-172 9.00 17.00 13.00 6.00 8.00 7.00 4.00 4.00 4.00 ROMA 11.00 17.00 14.00 5.00 7.00 6.00 4.00 2.00 3.00 L11 13.00 11.00 12.00 4.00 6.00 5.00 3.00 3.00 3.00 HEINZ-1370 13.00 13.00 13.00 6.00 6.00 6.00 3.00 3.00 3.00 NIRVANA F1 16.00 12.00 14.00 5.00 7.00 6.00 3.00 3.00 3.00 BUFFALO 12.00 12.00 12.00 6.00 8.00 7.00 3.00 3.00 3.00 INLAY F1 14.00 12.00 13.00 4.00 6.00 5.00 3.00 3.00 3.00 MONGAL F1 21.00 29.00 25.00 5.00 7.00 6.00 4.00 4.00 4.00 #20880 6.00 10.00 8.00 7.00 5.00 6.00 3.00 3.00 3.00 SUMO F1 14.00 14.00 14.00 5.00 5.00 5.00 3.00 3.00 3.00 ROMA VF 11.00 19.00 15.00 5.00 7.00 6.00 3.00 3.00 3.00 SHAKTIMAN 7.00 11.00 9.00 4.00 6.00 5.00 3.00 3.00 3.00 THORGAL F1 15.00 13.00 14.00 6.00 6.00 6.00 3.00 3.00 3.00 COBBRA F1 13.00 15.00 14.00 7.00 7.00 7.00 5.00 3.00 4.00 WOSOWOSO 15.00 15.00 15.00 7.00 5.00 6.00 3.00 3.00 3.00 MEAN 13.00 16.00 15.00 6.00 7.00 6.00 4.00 3.00 3.00 LSD(0.05) 1.55 1.72 1.13 0.70 0.70 0.58 0.52 0.79 0.46 University of Ghana http://ugspace.ug.edu.gh 57 Across the two locations, the lowest mean value of 4.00 was recorded by genotype L11 while the highest mean value of 33.00 was recorded by genotype CHERRY. Significant (P< 0.01) differences were observed among the genotypes evaluated based on the combined analysis of variance. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 12). 4.1.17 Mean single fruit weight per plant (g) (SFW) The mean values for single fruit weight per tomato plant are presented in Table 4. 7. The lowest mean value for single fruit weight per plant was recorded by genotype CHERRY under each of the greenhouse and field conditions with values 20.70 g and 17.70 g respectively. The highest mean value for single fruit weight per plant under greenhouse and field conditions was recorded by genotype SUMO F1 with values 127.74 and 96.89 respectively. Across the two locations, the lowest mean value of 19.20 was recorded by genotype CHERRY while the highest mean value of 112.32 was recorded by the genotype SUMO F1. Combined analysis of variance showed significant (P< 0.01) genotype, location and genotype x location interaction effects (Appendix 12). 4.1.18 Mean total fruit weight (g) per plant (FWP) The mean values of total fruit weight per plant are presented in Table 4.7. The lowest mean value of 150.33 g under the greenhouse conditions was recorded by genotype L11 while the highest mean value of 1492.72 was recorded by genotype PLATINUM F1. Under field conditions, the lowest mean value of 242.08 g was recorded by genotype NS 504 while the highest mean value of 1472 g was recorded by genotype MONGAL F1. Across the two locations, the lowest mean value of 217.99 g was recorded by the genotype L11 while the highest mean value of 1475.12 was recorded by the genotype PLATINUM F1. Significant (P< 0.01) differences were observed among the 20 genotypes. The effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 12). University of Ghana http://ugspace.ug.edu.gh 58 Table 4. 6 Mean Number of fruits per plant (FPP) Number of fruits per plant Genotype Greenhouse Open field Pooled Mean PECTOMECH 5.00 17.00 11.00 PLATINUMF1 19.00 21.00 20.00 NS 504 9.00 11.00 10.00 NKANSAH HT 27.00 33.00 30.00 CHERRY 31.00 35.00 33.00 11-172 7.00 9.00 8.00 ROMA 8.00 12.00 10.00 L11 3.00 5.00 4.00 HEINZ-1370 5.00 7.00 6.00 NIRVANA F1 4.00 8.00 6.00 BUFFALO 12.00 16.00 14.00 INLAY F1 9.00 11.00 10.00 MONGAL F1 19.00 21.00 20.00 #20880 4.00 6.00 5.00 SUMO F1 7.00 7.00 7.00 ROMA VF 6.00 12.00 9.00 SHAKTIMAN 4.00 10.00 7.00 THORGAL F1 5.00 5.00 5.00 COBBRA F1 20.00 10.00 15.00 WOSOWOSO 7.00 9.00 8.00 MEAN 11.00 13.00 12.00 LSD(0.05) 0.69 1.39 0.76 4.1.19 Mean fruit yield per plant (t/ha) (YPP) The mean values for fruit yield among tomato genotypes are presented in Table 4.7. The lowest mean fruit yield of 12.54 t/ha under greenhouse conditions was recorded by genotype L11 while the highest mean value of 124.68 t/ha was recorded by genotype MONGAL F1. Under the field conditions, the lowest mean fruit yield of 23.80 t/ha was recorded by University of Ghana http://ugspace.ug.edu.gh 59 genotype L11 while the highest mean fruit yield of 122.69 t/ha was recorded by genotype MONGAL F1. Across the two locations, the lowest mean value of 18.17 t/ha was recorded by genotype L11 while the highest mean value of 123.69 t/ha was recorded by genotype MONGAL F1. Significant (P< 0.01) differences were observed among the 20 tomato genotypes evaluated. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 13). 4.2 Mean performance of tomato for fruit physical and quality traits across locations 4.2.1 Mean tomato fruit length per plant (cm) (FL) The mean value for fruit length of the tomato genotypes evaluated has been presented in Table 4.8. The lowest mean value of 2.96 cm under greenhouse conditions was recorded by genotype CHERRY while the highest mean value of 5.64 cm was recorded by genotype 11- 172. Under field conditions, the lowest mean tomato fruit length of 2.91 cm was recorded by genotype CHERRY while the highest mean fruit length of 7.05 cm was recorded by genotype ROMA VF. Across the two locations, the lowest mean fruit length of 2.94 cm was recorded by genotype CHERRY while the highest mean value of 6.24 cm was recorded by genotype ROMA VF. Combined analysis of variance showed significant (P< 0.01) genotype, location and Genotype x Location interaction effects (Appendix 13). University of Ghana http://ugspace.ug.edu.gh 60 Table 4. 7 Mean Single fruit weight per plant, Total fruit weight per plant, and Fruit yield per plant Single fruit weight per plant (g) Total fruit weight per plant (g) Fruit yield per plant (t/ha) Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 58.72 37.53 48.13 276.67 607.11 441.89 23.05 50.70 36.88 PLATINUMF1 81.59 71.82 76.71 1492.72 1457.51 1475.12 124.40 121.45 122.93 NS 504 90.54 39.63 65.09 759.95 242.08 592.02 63.34 35.34 49.34 NKANSAH HT 42.75 34.95 38.85 1156.65 1160.05 1158.35 96.38 96.67 96.53 CHERRY 20.70 17.7 19.20 603.16 601.78 602.47 50.28 50.14 50.21 11-172 56.06 55.56 55.81 356.98 503.95 430.47 29.74 41.99 35.87 ROMA 73.04 64.50 68.77 600.05 773.32 686.69 50.01 64.44 57.23 L11 59.80 58.24 59.02 150.33 285.65 217.99 12.54 23.80 18.17 HEINZ-1370 104.4 84.73 94.57 524.77 506.75 515.76 43.73 42.24 42.99 NIRVANA 57.85 57.4 57.63 212.36 483.22 347.79 17.69 40.30 29.00 BUFFALO 51.93 45.62 48.78 583.99 709.32 646.66 48.63 59.12 53.88 INLAY F1 59.33 42.11 50.72 502.23 406.75 454.49 41.85 33.89 37.87 MONGAL F1 79.73 76.06 77.90 999.42 1472.36 1235.89 124.68 122.69 123.69 #20880 73.69 52.51 63.10 187.53 299.45 243.49 15.62 24.95 20.29 SUMO F1 127.74 96.89 112.32 922.49 650.15 786.32 76.87 54.18 65.53 ROMA VF 56.71 62.22 59.47 358.30 721.67 539.99 29.85 60.14 45.00 SHAKTIMAN 50.81 39.29 45.05 178.78 386.02 282.40 14.90 32.17 23.54 THORGAL F1 57.57 65.27 61.42 265.78 342.51 304.15 22.10 28.54 25.32 COBBRA F1 53.36 57.69 55.53 1062.56 597.34 829.95 88.55 49.78 69.17 WOSOWOSO 122.01 96.60 109.31 932.72 824.50 878.61 77.73 68.71 73.22 MEAN 68.90 57.80 63.37 606.40 660.70 633.50 52.60 55.10 53.80 LSD(0.05) 5.48 3.36 3.14 49.50 61.70 162.10 3.56 5.15 3.05 University of Ghana http://ugspace.ug.edu.gh 61 4.2.2 Mean fruit diameter (cm) (FD) The mean values for tomato fruit diameter are presented in Table 4.8. The lowest mean fruit diameter of 3.05 cm under greenhouse conditions was recorded by genotype CHERRY while the highest mean value of 7.18 cm was recorded by genotype WOSOWOSO. Under field conditions, the lowest mean value of 2.80 cm was recorded by genotype CHERRY while the highest mean fruit diameter of 6.79 cm was recorded by genotype WOSOWOSO. Across the two locations, the lowest mean fruit diameter of 2.93 cm was recorded by genotype CHERRY while the highest mean value of 6.99 cm was recorded by the genotype WOSOWOSO. Significant (P< 0.01) differences were observed among the 20 genotypes based on the combined analysis of variance. Also, the effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 13). 4.2.3 Mean fruit shape index of tomato (SI) The mean values for fruit shape index of all the tomato genotypes evaluated are presented in Table 4.8. The smallest tomato fruit shape index of 0.57 under greenhouse conditions was recorded by the genotype WOSOWOSO while the largest mean fruit shape index of 1.15 was recorded by genotype PLATINUM F1. Under field conditions, the smallest mean fruit shape index of 0.57 was recorded by genotype WOSOWOSO while the highest mean value of 1.66 was recorded by genotype ROMA VF. The genotype WOSOWOSO recorded the lowest mean value of 0.57 across the two locations while the highest mean value of 1.29 cm was recorded by genotype 11-172. Based on the combined analysis of variance, significant (P < 0.01) differences were observed among the tomato genotypes. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 13). University of Ghana http://ugspace.ug.edu.gh 62 Table 4. 8 Mean Fruit length, Fruit diameter and Fruit shape index Fruit length (cm) Fruit diameter (cm) Fruit shape index Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 4.37 5.56 4.97 4.57 4.94 4.76 0.94 1.12 1.03 PLATINUMF1 4.49 4.76 4.63 3.94 4.74 4.34 1.15 1.00 1.08 NS 504 4.28 4.67 4.48 4.90 5.18 5.04 0.88 0.90 0.89 NKANSAH HT 3.95 3.98 3.97 3.72 3.46 3.59 1.07 1.16 1.12 CHERRY 2.96 2.91 2.94 3.05 2.80 2.93 0.98 1.07 1.03 11-172 5.64 5.28 5.46 4.32 4.16 4.24 1.3 1.28 1.29 ROMA 5.57 5.04 5.31 5.56 4.39 4.98 1.00 1.17 1.09 L11 5.03 4.94 4.99 5.36 5.66 5.51 0.99 0.86 0.93 HEINZ-1370 5.16 4.79 4.98 5.88 6.11 6.00 0.89 0.79 0.84 NIRVANA 3.92 5.04 4.48 3.74 4.10 3.92 1.06 1.25 1.16 BUFFALO 5.50 6.08 5.79 4.81 4.47 4.64 1.14 1.38 1.26 INLAY F1 5.09 5.5 5.30 6.03 5.06 5.55 0.85 1.04 0.95 MONGAL F1 4.3 4.49 4.40 6.82 6.07 6.45 0.64 0.74 0.69 #20880 5.16 5.04 5.10 6.03 5.77 5.90 0.86 0.91 0.89 SUMO F1 4.67 4.70 4.69 6.74 6.24 6.50 0.70 0.76 0.73 ROMA VF 5.43 7.05 6.24 6.08 4.21 5.15 0.89 1.66 1.28 SHAKTIMAN 3.92 5.21 4.57 3.68 4.43 4.06 1.01 1.2 1.11 THORGAL F1 4.48 4.55 4.52 4.23 5.31 4.77 1.07 0.86 0.97 COBBRA F1 5.01 3.64 4.33 4.72 3.97 4.35 1.06 0.93 1.00 WOSOWOSO 4.09 4.04 4.07 7.18 6.79 6.99 0.57 0.57 0.57 MEAN 4.65 4.86 4.76 5.07 4.89 4.98 0.95 1.03 0.99 LSD(0.05) 0.25 0.52 0.29 0.37 0.52 0.32 0.12 0.10 0.07 4.2.4 Mean number of locules per fruit (LOC) The mean values recorded for number of locules per fruit have been presented in Table 4.9. The lowest mean number of locules was 2.00 under greenhouse conditions and this was University of Ghana http://ugspace.ug.edu.gh 63 recorded by 7 tomato genotypes while the highest mean value of 8.00 locules was recorded by genotype WOSOWOSO. Under field conditions, the smallest mean number of locules recorded was 2 and it was recorded by 8 genotypes while the highest mean value of 8.00 was recorded by genotype WOSOWOSO. Across the two locations, the lowest mean value of 2.00 was recorded by 7 genotypes while the highest mean value of 8.00 was recorded by genotype WOSOWOSO. Based on the combined analysis of variance, differences observed among the 20 tomato genotypes were significantly (P < 0.01) different. The effects of location and genotype x location interaction were not significant (P < 0.05) (Appendix 13). 4.2.5 Mean fruit pericarp thickness (mm) (PTK) The mean values for fruit pericarp thickness recorded in the study are presented in Table 4.9. The smallest mean value of 3.00 mm under greenhouse conditions was recorded by the genotype CHERRY while the maximum mean value of 6.19 mm was recorded by the genotype COBBRA F1. Under field conditions, the lowest mean pericarp thickness of 2.71 mm was recorded by genotype CHERRY while the highest mean pericarp thickness of 6.01 mm was recorded by the genotype ROMA VF. The genotype CHERRY recorded the lowest mean pericarp thickness of 2.86 mm across the two locations while the highest mean value of 5.51 mm was recorded by genotype PLATINUM F1 .The combined analysis of variance showed significant (P < 0.01) genotype, location and genotype x location interaction effects (Appendix 13). 4.2.6 Mean fruit firmness (kg/cm2) (FF) The mean values for fruit firmness among the tomato genotypes evaluated are presented in Table 4.9. The lowest mean value of 2.87 kg/cm2 under greenhouse conditions was recorded by genotype MONGAL F1 while the highest mean value of 5.13 kg/ cm2 was recorded genotype 11-172. Under field conditions, the lowest mean value of 2.40 kg/cm2 was recorded by genotype MONGAL F1 while the highest mean value of 3.73 kg/cm2 was recorded by University of Ghana http://ugspace.ug.edu.gh 64 ROMA VF. Tomato genotype MONGAL F1 recorded the lowest mean value of 2.64 kg/cm2 across the two locations while the highest mean value of 4.27 kg/cm2 was recorded by the genotype 11-172. Significant (P< 0.01) difference was observed among the 20 genotypes based on the combined analysis of variance (Appendix 13). The effects of location and genotype x location interaction were also significantly (P< 0.01) different. 4.3.5 Mean fruit pH value of tomato (pH) The mean pH values of the tomato genotypes are presented in Table 4.10. The lowest mean pH value of 3.90 under greenhouse conditions was recorded by genotype WOSOWOSO while the highest mean value of 5.11 was recorded by genotype #20880. Under field conditions, the lowest mean value of 4.22 was recorded by the tomato genotype 11 - 172 while the highest mean value of 4.52 was recorded by seven tomato genotypes (PLATINUM F1, CHERRY, L11, BUFFALO, #20880, SHAKTIMAN, and WOSOWOSO). Across the two locations, the lowest mean value of 4.09 was recorded by NKANSAH HT while the highest mean value of 4.81 was recorded by the genotype #20880. Significant (P< 0.01) differences were observed among the 20 genotypes. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 14). 4.3.6 Mean fruit dry matter content of tomato (g) (FDM) The mean values showing fruit dry matter content of tomato are presented in Table 4.10. The lowest mean value for dry matter content of 0.22 g under greenhouse conditions was recorded by genotype ROMA while the highest mean dry matter content of 0.65 g was recorded by the genotype ROMA VF. Under field conditions, the lowest mean dry matter content of tomato was 0.27 g and was recorded by MONGAL F1, SHAKTIMAN and WOSOWOSO while the highest mean value of 0.49 g was recorded by ROMA VF. University of Ghana http://ugspace.ug.edu.gh 65 Table 4. 9 Mean Number of locules, Fruit pericarp thickness and Fruit firmness Number of locules Fruit pericarp thickness (mm) Fruit firmness (kg/cm2) Genotype Green house Open field Poole Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 2.00 2.00 2.00 5.72 3.42 4.57 3.40 3.37 3.39 PLATINUMF1 2.00 2.00 2.00 5.80 5.22 5.51 3.50 2.83 3.17 NS 504 3.00 3.00 3.00 4.00 5.16 4.58 3.90 2.80 3.35 NKANSAH HT 3.00 3.00 3.00 4.58 3.42 4.00 4.63 3.50 4.07 CHERRY 2.00 2.00 2.00 3.00 2.71 2.86 3.63 3.23 3.43 11-172 2.00 2.00 2.00 3.75 3.80 3.78 5.13 3.40 4.27 ROMA 3.00 5.00 4.00 5.80 3.75 4.78 5.07 3.27 4.17 L11 3.00 3.00 3.00 3.99 4.25 4.12 3.85 3.13 3.49 HEINZ-1370 4.00 4.00 4.00 5.16 5.10 5.13 3.63 3.67 3.65 NIRVANA 2.00 2.00 2.00 4.21 4.00 4.11 3.90 3.57 3.74 BUFFALO 4.00 4.00 4.00 4.72 5.69 5.21 3.67 3.57 3.62 INLAY F1 3.00 3.00 3.00 4.45 4.53 4.59 4.10 3.47 3.79 MONGAL F1 5.00 5.00 5.00 5.04 4.99 5.02 2.87 2.40 2.64 #20880 5.00 5.00 5.00 5.10 4.89 5.00 3.77 3.43 3.60 SUMO F1 7.00 7.00 7.00 4.56 4.44 4.50 4.50 3.33 3.92 ROMA VF 2.00 2.00 2.00 4.96 6.01 5.49 3.33 3.73 3.53 SHAKTIMAN 4.00 2.00 3.00 4.08 3.47 3.78 4.43 2.43 3.43 THORGAL F1 5.00 5.00 5.00 4.00 3.94 3.97 3,23 2.50 2.87 COBBRA F1 2.00 2.00 2.00 6.19 3.00 4.60 3.70 3.37 3.54 WOSOWOSO 8.00 8.00 8.00 4.03 4.08 4.06 3.70 2.50 3.10 MEAN 3.55 3.55 3.55 4.66 4.29 4.48 3.90 3.18 3.54 LSD(0.05) 0.02 1.51 1.13 0.52 1.01 0.57 0.61 0.51 0.39 University of Ghana http://ugspace.ug.edu.gh 66 The genotype ROMA recorded the lowest mean value of 0.29 g across the two locations while the highest mean value of 0.57 g was recorded by ROMA VF. From the combined analysis of variance, Significant (P< 0.01) differences were observed among the 20 genotypes. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 13). 4.3.7 Mean total soluble solids content (0Brix) (TSS) The mean total soluble solids content of tomato fruits are presented in Table 4.10. The lowest mean total soluble solids content of 2.26 0Brix under greenhouse conditions was recorded by ROMA while the highest mean value of 6.58 0Brix was recorded by ROMA VF. Under field conditions, the lowest mean total soluble solids content of 2.58 0Brix was recorded by genotype SHAKTIMAN while the highest mean value of 6.98 0Brix was recorded by genotype WOSOWOSO. Across the two locations, the lowest mean value of 2.66 0Brix was recorded by genotype ROMA while the highest mean value of 6.26 0Brix was recorded by genotype WOSOWOSO. Significant (P< 0.01) differences were observed among the tomato genotypes evaluated. The effects of genotype and genotype x location interaction were significantly (P< 0.01) different (Appendix 14). 4.3.8 Mean acidity content (mg/100 ml) (TA) The mean value for acidity content of tomato fruits are presented in Table 4.11. The lowest mean acidity content of 0.19 mg/100 ml under greenhouse conditions was recorded by ROMA and MONGAL F1 while the highest mean value of 0.44 mg/100 ml was recorded by HEINZ-1370 and ROMA VF. Under field conditions, the lowest mean value of 0.32 mg/100 ml was recorded by genotype SHAKTIMAN while the highest acidity content of 0.66 mg/100 ml was recorded by genotype WOSOWOSO. Across the two locations, the lowest mean value of 0.27 mg/100 ml was recorded by genotype MONGAL F1 while the highest mean value of 0.54 was recorded by WOSOWOSO. Significant (P< 0.01) differences were University of Ghana http://ugspace.ug.edu.gh 67 observed among the 20 genotypes. Also, the effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 14). Table 4. 10 Mean pH value, Fruit dry matter content and Total soluble solids content pH Fruit dry matter content (g) Total soluble solids (0Brix) Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 4.04 4.31 4.18 0.38 0.38 0.38 3.85 3.91 3.88 PLATINUMF1 4.24 4.52 4.38 0.35 0.35 0.35 3.64 3.45 3.55 NS 504 4.13 4.23 4.18 0.36 0.32 0.34 3.62 2.63 3.13 NKANSAH HT 3.92 4.27 4.09 0.41 0.39 0.40 3.95 4.33 4.14 CHERRY 4.57 4.52 4.54 0.43 0.36 0.40 4.41 4.50 4.46 11-172 4.67 4.22 4.44 0.51 0.38 0.45 5.28 6.64 5.86 ROMA 4.23 4.30 4.26 0.22 0.36 0.29 2.26 3.05 2.66 L11 4.87 4.52 4.70 0.44 0.35 0.40 4.33 5.17 4.75 HEINZ-1370 4.34 4.23 4.29 0.50 0.41 0.46 5.18 3.90 4.54 NIRVANA 4.08 4.30 4.19 0.52 0.36 0.44 5.30 2.74 4.02 BUFFALO 4.20 4.52 4.36 0.43 0.39 0.41 4.39 3.15 3.77 INLAY F1 4.13 4.24 4.19 0.51 0.38 0.45 5.22 3.79 4.51 MONGAL F1 4.66 4.30 4.48 0.36 0.27 0.32 3.78 4.12 3.95 #20880 5.11 4.52 4.81 0.32 0.38 0.35 3.16 4.23 3.70 SUMO F1 4.32 4.23 4.28 0.33 0.37 0.35 3.40 3.44 3.42 ROMA VF 4.29 4.30 4.29 0.65 0.49 0.57 6.58 4.22 5.40 SHAKTIMAN 4.12 4.52 4.32 0.55 0.27 0.41 5.55 2.58 4.07 THORGAL F1 4.20 4.23 4.22 0.63 0.28 0.46 6.44 3.93 5.19 COBBRA F1 4.07 4.30 4.18 0.29 0.38 0.34 2.93 3.23 3.08 WOSOWOSO 3.90 4.52 4.21 0.55 0.27 0.41 5.54 6.98 6.26 MEAN 4.30 4.36 4.33 0.44 0.36 0.40 4.44 4.00 4.22 LSD(0.05) 0.03 0.03 0.02 0.04 0.07 0.04 0.29 0.2 0.19 University of Ghana http://ugspace.ug.edu.gh 68 4.3.9 Mean flavour index (TSS/TA) The mean values for flavour index (TSS/TA) of tomato fruits are presented in Table 4.11. The lowest flavour index (TSS/TA) of 9.61 under greenhouse conditions was recorded by PECTOMECH while the highest mean value of 19.49 mg/100ml was recorded by genotype L11. Under field conditions, the lowest mean flavour index (TSS/TA) of 6.76 was recorded by genotype NS 504 while the highest flavour index (TSS/TA) of 14.40 was recorded by genotype 11 - 172. The genotype PLATINUM F1 recorded the lowest mean value of 9.39 across the two locations while the highest mean value of 15.67 as recorded by genotype MONGAL F1. The combined analysis of variance showed significant (P< 0.01) differences among the 20 genotypes. Also, the effects of genotype and genotype x location interaction were significantly (P< 0.01) different (Appendix 14). 4.3.10 Mean malic acid content (mg/100 ml) (MA) The mean value for malic acid contents of the tomato fruits are presented in Table 4.11. Under greenhouse conditions, the lowest mean value of 0.22 mg/100 ml was recorded by MONGAL F1 while the highest mean value of 0.48 mg/100 ml was recorded by HEINZ- 1370. Under field conditions, the lowest mean value of 0.34 mg/100 ml was recorded by SHAKTIMAN while the highest mean value of 0.69 mg/100 ml was recorded by the genotype WOSOWOSO. Across the two locations, the lowest mean value of 0.29 mg/100 ml was recorded by MONGAL F1 while the highest mean value of 0.57 mg/100ml was recorded by the genotype WOSOWOSO. Significant (P < 0.01) differences among the genotypes and locations for malic acid content were observed based on the combined analysis of variance (Appendix 13). Also, Genotype x Location interaction effects was significantly (P < 0.01) different. University of Ghana http://ugspace.ug.edu.gh 69 Table 4. 11 Mean Titrable acidity, TSS/TA and Malic acid content of tomato fruits Titrable acidity (mg/100 ml) Flavour index (TSS/TA) Malic acid content (mg/100 ml) Genotype Green house Open field Pooled Mean Green house Open field Pooled mean Green house Open field Pooled mean PECTOMECH 0.42 0.36 0.39 9.61 10.89 10.25 0.45 0.37 0.41 PLATINUMF1 0.36 0.45 0.41 10.31 8.46 9.39 0.35 0.44 0.40 NS 504 0.28 0.43 0.36 12.12 6.76 9.44 0.33 0.46 0.40 NKANSAH HT 0.28 0.44 0.36 13.24 9.55 11.40 0.36 0.48 0.42 CHERRY 0.27 0.47 0.37 16.15 9.76 12.96 0.28 0.48 0.38 11-172 0.34 0.47 0.41 15.78 14.40 15.09 0.36 0.47 0.42 ROMA 0.19 0.36 0.28 11.49 8.89 10.19 0.23 0.36 0.30 L11 0.26 0.46 0.36 19.49 10.81 15.15 0.23 0.52 0.38 HEINZ-1370 0.44 0.36 0.40 11.66 10.64 11.15 0.48 0.38 0.43 NIRVANA 0.39 0.34 0.37 13.00 8.24 10.62 0.41 0.36 0.39 BUFFALO 0.27 0.36 0.32 15.59 8.75 12.17 0.30 0.35 0.33 INLAY F1 0.38 0.33 0.36 13.91 11.38 12.65 0.37 0.39 0.38 MONGAL F1 0.19 0.34 0.27 19.27 12.06 15.67 0.22 0.35 0.29 #20880 0.24 0.36 0.30 12.75 11.79 12.27 0.26 0.38 0.32 SUMO F1 0.28 0.43 0.36 12.07 8.58 10.33 0.30 0.42 0.36 ROMA VF 0.44 0.39 0.42 16.07 10.74 13.41 0.43 0.41 0.42 SHAKTIMAN 0.36 0.32 0.34 15.48 7.97 11.73 0.38 0.34 0.36 THORGAL F1 0.37 0.42 0.40 17.55 9.41 13.48 0.38 0.44 0.41 COBBRA F1 0.23 0.34 0.29 12.73 9.47 11.10 0.24 0.36 0.30 WOSOWOSO 0.42 0.66 0.54 13.26 10.62 11.94 0.44 0.69 0.57 MEAN 0.32 0.40 0.36 14.08 9.96 12.02 0.34 0.42 0.38 LSD(0.05) 0.03 0.03 0.02 0.80 0.62 0.54 0.03 0.04 0.03 University of Ghana http://ugspace.ug.edu.gh 70 4.3 Mean performance of tomato for phytochemical composition across locations 4.4.1 Mean β-Carotene content in tomato fruit (mg/100 ml) (β -CARO) The mean value for β-Carotene content of tomato fruits are presented in Table 4.12. The tomato genotype ROMA VF recorded the lowest value for β-Carotene content under greenhouse conditions with a mean value of 7.31 mg/100 ml while the highest mean value of 34.39 mg/100 ml was recorded by NIRVANA F1. Under field conditions, the lowest mean value for β-Carotene content of 13.31 was recorded by PECTOMECH while the highest mean β-carotene content of 33.79 mg/100ml was recorded by genotype 11–172. Across the two locations, the lowest mean value of 13.60 was recorded by ROMA VF while the highest mean value of 30.38 mg/100 ml was recorded by the genotype 11-172. The combined analysis of variance showed significant (P < 0.01) genotype, location and genotype x location interaction effects (Appendix 14). 4.4.2 Mean lycopene content in tomato fruit (mg/100 ml) (LYC) The mean value for lycopene content in tomato fruit are shown in Table 4.12. The lowest mean value of 7.50 mg/100 ml under greenhouse conditions was recorded by THORGAL F1 while the highest mean lycopene content of 30.74 mg/100 ml was recorded by NIRNANA F1. Under field conditions, the lowest mean lycopene content of 6.56 mg/100 ml was recorded by genotype PECTOMECH while the highest mean value of 25.79 mg/100 ml was recorded by genotype 11–172. The genotype THORGAL F1 recorded the lowest mean value of 8.83 mg/100 ml across the two locations while the highest mean value of 26.50 was recorded by the genotype NIRVANA F1. Significant (P < 0.01) differences were observed among the genotypes and locations based on the combined analysis of variance (Appendix 14). Also, Genotype x Location interaction effects was significantly (P < 0.01) different. University of Ghana http://ugspace.ug.edu.gh 71 4.4.3 Mean ascorbic acid content in tomato fruit (mg/100 ml) (ASC) The mean value for ascorbic acid content in tomato fruits are presented in Table 4.12. The tomato genotype INLAY F1 recorded the lowest ascorbic acid content of 6.11 mg/100 ml under greenhouse conditions while the highest mean value of 17.39 mg/100 ml was recorded by CHERRY. Under field conditions, the lowest mean value of 5.33 mg/100 ml was recorded by L11/while the highest mean value of 12.94 mg/100 ml was recorded by genotype ROMA. The genotype INLAY F1 recorded the lowest mean value of 7.11 mg/100 ml across the two locations while the highest mean value of 13.98 mg/100 ml was recorded by genotype CHERRY. Significant (P < 0.01) differences were observed among the genotypes and locations based on the combined analysis of variance (Appendix 14). Also, Genotype x Location interaction effects was significantly (P < 0.01) different. 4.4.4 Mean gallic acid content in tomato fruit (mg/100 ml) (GA) The mean value for gallic acid content in tomato fruit has been presented in Table 4.13. The lowest mean gallic acid content of 1.50 mg/100 ml under greenhouse conditions was recorded by COBBRA F1 while the highest mean value of 1.77 mg/100 ml was recorded by genotype NIRVANA F1. Under field conditions, the lowest mean value of 1.49 mg/100 ml was recorded by genotype L11 while the highest mean gallic acid content of 2.72 mg/100 ml was recorded by PLATINUM F1. Across the two locations, the lowest mean value of 1.54 mg/100 ml was recorded by the genotype NKANSAH HT while the highest mean value of 2.18 mg/100 ml was recorded by PLATINUM F1. Significant (P< 0.01) differences were observed among the 20 genotypes. Also, the effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 14). University of Ghana http://ugspace.ug.edu.gh 72 Table 4. 12 Mean β-Carotene, Lycopene and Ascorbic acid contents of tomato fruits β-Carotene content (mg/100 ml) Lycopene content (mg/100 ml) Ascorbic acid content (mg/100 ml) Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 29.19 13.31 21.25 22.12 6.56 14.34 12.89 8.92 10.91 PLATINUMF1 14.75 24.70 19.73 11.45 19.23 15.34 13.72 10.72 12.22 NS 504 8.13 27.02 17.58 8.36 10.03 13.70 11.72 8.50 10.11 NKANSAH HT 21.19 25.16 23.18 12.84 13.12 12.98 13.89 12.17 13.03 CHERRY 16.72 26.46 21.59 14.76 19.28 17.02 17.39 10.56 13.98 11-172 26.97 33.79 30.38 22.31 25.79 24.05 14.28 11.06 12.67 ROMA 23.07 25.92 24.50 13.55 17.02 15.29 8.33 12.94 10.64 L11 18.95 21.50 20.23 19.50 12.48 15.99 9.72 5.33 7.53 HEINZ-1370 13.29 31.03 22.16 11.04 24.99 18.02 8.42 11.11 9.77 NIRVANA F1 34.39 26.18 30.29 30.74 22.25 26.50 12.56 9.28 10.92 BUFFALO 13.23 19.32 16.28 10.75 13.63 12.19 10.11 9.82 9.97 INLAY F1 10.66 28.28 19.47 19.50 12.48 15.99 6.11 8.11 7.11 MONGAL F1 19.68 15.33 17.51 12.55 9.86 11.21 15.39 5.83 10.61 #20880 17.74 18.39 18.07 15.29 13.46 14.38 12.67 12.05 12.36 SUMO F1 13.50 29.85 21.68 12.15 20.01 16.08 14.00 9.94 11.97 ROMA VF 7.31 19.89 13.60 11.29 13.53 12.41 8.33 8.00 8.17 SHAKTIMAN 33.71 22.28 28.00 28.52 18.74 23.63 8.39 12.06 10.23 THORGAL F1 17.13 21.96 19.55 7.50 10.15 8.83 13.11 7.89 10.50 COBBRA F1 27.59 20.69 21.14 23.86 7.65 15.76 11.11 6.83 8.97 WOSOWOSO 30.50 15.91 23.21 16.40 8.99 12.70 12.50 7.11 9.81 MEAN 19.89 23.35 21.62 16.22 15.41 15.82 11.73 9.41 10.57 LSD(0.05) 0.82 4.95 2.44 0.70 1.23 0.69 4.58 0.76 2.28 University of Ghana http://ugspace.ug.edu.gh 73 4.4.5 Mean rosmarinic acid content in tomato fruit (mg/100 ml) (RA) The mean values for rosmarinic acid content in tomato fruit are present in Table 4.13. The lowest mean value of 0.66 mg/100 ml under greenhouse conditions was recorded by PECTOMECH while the highest mean value of 1.08 mg/100 ml was recorded by genotype NIRVANA F1. Under field conditions, the lowest mean value of 0.72 mg/100 ml was recorded by genotype L11 while the highest mean value of 2.28 mg/100 ml was recorded by PLATINUM F1. Across the two locations, the lowest mean value of 0.70 mg/100ml was recorded by the PECTOMECH while the highest mean value of 1.59 mg/100ml was recorded by PLATINUM F1. Significant (P< 0.01) differences were observed among the 20 genotypes based on the combined analysis of variance. The effects of location and genotype x location interaction were also significantly (P< 0.01) different (Appendix 14). 4.4.6 Mean vanillic acid content in tomato ((mg/100 ml) (VA) The mean values for vanillic acid content in tomato fruits have been presented in Table 4.13. The lowest mean value of 0.57 mg/100ml under greenhouse conditions was recorded by genotype NKANSAH HT while the highest mean vanillic acid content of 1.22 mg/100ml was recorded by the genotype NIRVANA F1. Under field conditions, the lowest mean value of 0.75 mg/100 ml was recorded by genotype L11 while the highest mean value of 2.78 mg/100 ml was recorded by PLATINUM F1. Across the two locations, the lowest mean value of 0.83 mg/100 ml was recorded by NKANAH HT while the highest mean value of 1.88 mg/100 ml was recorded by PLATINUM F1. Significant (P< 0.01) differences were observed among the 20 genotypes. The effects of genotype and genotype x location interaction were significantly (P< 0.01) different (Appendix 14). University of Ghana http://ugspace.ug.edu.gh 74 Table 4. 13 Mean Gallic, Rosmarinic and Vanilic acids content of tomato fruits Gallic acid (mg/100 ml) Rosmarinic acid (mg/100 ml) Vanillic acid (mg/100 ml) Genotype Green house Open field Pooled mean Green house Open field Pooled mean Green hose Open field Pooled mean PECTOMECH 1.61 1.66 1.64 0.66 0.73 0.70 0.79 0.88 0.84 PLATINUMF1 1.63 2.72 2.18 0.90 2.28 1.59 0.98 2.78 1.88 NS 504 1.53 1.64 1.59 0.78 0.92 0.85 0.82 1.00 0.91 NKANSAH HT 1.38 1.70 1.54 0.58 0.98 0.78 0.57 1.09 0.83 CHERRY 1.53 1.66 1.60 0.78 0.94 0.86 0.82 1.04 0.93 11-172 1.60 1.79 1.70 0.86 1.11 0.99 0.93 1.25 1.09 ROMA 1.55 1.71 1.63 0.80 1.00 0.90 0.86 1.12 0.99 L11 1.64 1.49 1.57 0.91 0.72 0.82 1.00 0.75 0.88 HEINZ-1370 1.53 1.67 1.60 0.78 0.95 0.87 0.82 1.05 0.94 NIRVANA F1 1.77 1.59 1.68 1.08 0.84 0.96 1.22 0.91 1.07 BUFFALO 1.72 1.72 1.72 1.01 1.01 1.01 1.13 1.13 1.13 INLAY F1 1.57 1.65 1.61 0.82 0.92 0.87 0.88 1.01 0.95 MONGAL F1 1.54 1.65 1.60 0.79 0.92 0.86 0.84 1.01 0.93 #20880 1.56 1.55 1.56 0.81 0.80 0.81 0.87 0.86 0.87 SUMO F1 1.57 2.33 1.95 0.82 1.78 1.30 0.87 2.13 1.50 ROMA VF 1.53 1.71 1.62 0.77 1.00 0.89 0.82 1.11 0.97 SHAKTIMAN 1.66 1.75 1.71 0.93 1.05 0.99 1.03 1.17 1.10 THORGAL F1 1.65 1.66 1.66 0.93 0.93 0.93 1.02 1.03 1.03 COBBRA F1 1.50 1.65 1.58 0.73 0.93 0.83 0.76 1.02 0.89 WOSOWOSO 1.54 1.60 1.57 0.79 0.86 0.83 0.84 0.93 0.89 MEAN 1.58 1.75 1.66 0.83 1.03 0.93 0.89 1.16 1.03 LSD(0.05) 0.04 0.45 0.11 0.15 0.59 0.15 0.12 0.75 0.19 University of Ghana http://ugspace.ug.edu.gh 75 4.4.7 Mean catechin content in tomato fruit (mg/100 ml) (CAT) The mean values for catechin content in tomato fruits have been presented in Table 4.14. The lowest mean value of 54.66 mg/100 ml under greenhouse conditions was recorded by INLAY F1 while the highest mean catechin content of 190.22 mg/100 ml was recorded by THORGAL F1. Under field conditions, the lowest mean value of 42.03 mg/100 ml was recorded by genotype MONGAL F1 while the highest mean value of 311.02 mg/100 ml was recorded by WOSOWOSO. Across the two locations, the lowest mean value of 53.04 mg/100 ml was recorded by genotype INLAY F1 while the highest mean value of 192.84 mg/100 ml was recorded by WOSOWOSO. Significant (P< 0.01) differences were observed among the 20 genotypes. Also, the effects of genotype and genotype x location interaction were significantly (P< 0.01) different (Appendix 15). 4.4.8 Mean genestein content in tomato (mg/100 ml) (GEN) The mean value for genestein content in tomato fruits has been presented in Table 4.14. The lowest mean genestein content of 35.40 mg/100 ml under greenhouse conditions was recorded by ROMA VF while the highest mean value of 112.77 mg/100 ml was recorded by genotype NIRVANA F. Under field conditions the lowest mean value of 23.70 mg/100 ml was recorded by ROMA VF while the highest mean genestein content of 164.56 mg/100 ml was recorded by WOSOWOSO. Across the two locations, the lowest mean value of 29.55 mg/100ml was recorded by genotype ROMA VF while the highest mean value of 102.66 mg/100 ml was recorded by genotype WOSOWOSO. Combined analysis of variance showed significant (P< 0.01) differences among the 20 genotypes evaluated. Also, the effects of genotype and genotype x location interaction were significantly (P< 0.01) different (Appendix 15). University of Ghana http://ugspace.ug.edu.gh 76 4.4.9 Mean hesperetin content in tomato fruit (mg/100 ml) (HES) The mean hesperetin content in tomato fruit among the genotypes are presented in Table 4.14. The genotype INLAY F1 recorded the lowest mean value of 26.78 mg/100 ml under greenhouse conditions while the highest mean value of 273.63 mg/100 ml was recorded by THORGAL F1. Under field conditions, the lowest mean hesperetin content of 18.44 mg/100 ml was recorded by NIRVANA F1 while the highest mean value of 495.11 mg/100 ml was recorded by WOSOWOSO. The genotype INLAY F1 recorded the lowest mean hesperetin content of 22.98 mg/100 ml across the two locations while the highest mean value of 278.45 mg/100 ml was recorded by WOAOWOAO. Based on the combined analysis of variance, significant (P< 0.01) genotype, location and genotype x location interaction effects was observed (Appendix 15). 4.4.10 Mean kaempferol content in tomato fruit (mg/100 ml) (KAE) The mean value for kaempferol content in tomato fruits are presented in Table 4.15. The lowest mean kaempferol content of 9.46 mg/100 ml under greenhouse conditions was recorded by INLAY F1 while the highest mean value of 68.48 mg/100 ml was recorded by the genotype NIRVANA F1. Under field conditions, the lowest mean value of 1.75 mg/100 ml was recorded by the genotype ROMA VF while the highest mean Kaempferol content of 120.07 mg/100 ml was recorded by WOSOWOSO. Across the two locations, the lowest mean value of 6.04 mg/100ml was recorded by genotype ROMA VF while the highest mean value of 67.21 mg/100 ml was recorded by WOSOWOSO. Significant (P < 0.01) differences among the genotypes and locations for kaempferol content were recorded based on the combined analysis of variance. The effects of location and Genotype x Location interaction were also significantly (P < 0.01) different (Appendix 15). University of Ghana http://ugspace.ug.edu.gh 77 Table 4. 14 Mean Catechin, Genestein and Hesperetin content of tomato genotypes Catechin (mg/100 ml) Genestein (mg/100 ml) Hesperetin (mg/100 ml) Genotype Green house Open field Pooled mean Green house Open field Pooled mean Green house Open field Pooled mean PECTOMECH 104.44 63.68 84.06 56.35 34.98 45.67 116.39 41.59 78.99 PLATINUM F1 74.65 70.53 72.59 40.81 38.57 39.69 61.96 54.18 58.07 NS 504 161.93 43.49 102.71 86.46 24.49 55.48 221.77 44.37 113.07 NKANSAH HT 189.50 111.53 150.52 101 60.04 80.52 272.33 129.37 200.85 CHERRY 154.15 113.84 134.00 82.39 61.27 71.83 207.51 133.63 170.57 11-172 110.62 185.36 147.99 59.58 98.75 79.17 127.70 264.73 196.22 ROMA 113.54 75.83 94.69 61.11 41.37 51.24 133.07 64.00 98.54 L11 92.43 60.11 76.27 50.05 33.03 41.54 90.70 35.11 62.91 HEINZ-1370 160.72 64.18 112.45 85.83 35.21 60.52 219.50 42.33 130.92 NIRVANA F1 65.78 51.02 58.40 112.77 28.36 70.57 44.38 18.44 31.41 BUFFALO 65.77 75.36 70.57 48.36 55.04 51.70 45.11 65.29 55.20 INLAY F1 54.66 51.42 53.04 56.41 28.57 42.49 26.78 19.18 22.98 MONGAL F1 104.66 42.03 73.35 56.46 30.64 43.55 116.77 30.08 59.93 #20880 69.74 62.84 66.29 50.39 34.56 42.48 85.53 40.11 62.82 SUMO F1 80.43 260.11 170.27 57.74 137.9 97.82 74.59 401.77 238.18 ROMA VF 64.45 42.13 53.29 35.4 23.70 29.55 43.07 33.48 23.28 SHAKTIMAN 112.58 81.73 97.16 60.61 39.37 49.99 131.3 63.63 97.47 THORGAL F1 190.22 54.16 122.19 101.28 57.94 79.61 273.63 21.96 147.80 COBBRA F1 69.21 78.30 73.76 37.89 42.65 40.27 51.78 68.44 60.11 WOSOWOSO 74.66 311.02 192.84 40.75 164.56 102.66 61.78 495.11 278.45 MEAN 105.71 94.93 100.32 64.08 53.55 58.82 120.28 103.34 109.39 LSD(0.05) 3.70 6.64 3.75 5.12 2.36 2.78 23.83 7.08 12.11 4.4.11 Mean myricetin content in tomato fruits (mg/100 ml) (MYR) The mean value for myricetin content in tomato fruits have been presented in Table 4.15. The lowest mean value of 19.93 mg/100 ml under greenhouse condition was recorded by ROMA University of Ghana http://ugspace.ug.edu.gh 78 VF while the highest mean value of 85.07 mg/100 ml was recorded by NIRVANA F1. Under field conditions, the lowest mean myricetin content of 10.11 mg/100ml was recorded by ROMA VF while the highest mean value of 128.43 mg/100 ml was recorded by genotype WOSOWOSO. The genotype ROMA VF recorded the lowest mean value of 15.02 mg/100 ml across the two locations while the highest mean value of 76.43 mg/100 ml was recorded by WOSOWOSO. Significant (P< 0.01) differences were observed among the 20 genotypes. Also, the effects of location and Genotype x Location interaction were significantly (P< 0.01) different (Appendix 15). 4.4.12 Mean rutin content in tomato fruit (mg/100 ml) (RUT) The mean value for rutin content in tomato fruit are presented in Table 4.15. The lowest mean value of 44.37 mg/100 ml under greenhouse conditions was recorded by INLAY F1 while the highest mean value of 188.48 mg/100 ml was recorded by THORGAL F1. Under field conditions, the lowest mean value of 23.47 mg/100 ml was recorded by genotype MONGAL F1 while the highest rutin content of 321.37 mg/100 ml was recorded by the tomato genotype WOSOWOSO. Across the two locations, the lowest mean value of 37.86 mg/100 ml was recorded by genotype ROMA VF while the highest mean value of 191.37 mg/100 ml was recorded by WOSOWOSO. Significant (P< 0.01) differences were observed among the 20 genotypes. Also, the effects of location and genotype x location interaction were significantly (P< 0.01) different (Appendix 15). 4.4.13 Mean quercetin content in tomato fruit (mg/100 ml) (QUE) The mean quercetin content in tomato fruit among the genotypes evaluated have been presented in Table 4.16. The lowest mean value of 29.04 mg/100 ml under the greenhouse condition was recorded by INLAY F1 while the highest mean value of 124.12 mg/100 ml was recorded by the genotype THORGALF1. University of Ghana http://ugspace.ug.edu.gh 79 Table 4. 15 Mean Kaempferol, Myricetin and Rutin contents of tomato genotypes Kaempferol (mg/100 ml) Myricetin (mg/100 ml) Rutin (mg/100 ml) Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 26.04 11.22 18.63 37.53 19.58 28.555 94.13 49.25 71.69 PLATINUM F1 14.45 14.18 14.32 24.45 22.59 23.52 61.48 56.91 59.20 NS 504 48.63 2.29 25.46 62.8 10.61 36.705 157.36 27.05 92.21 NKANSAH HT 59.46 32.29 45.88 74.96 40.65 57.805 187.70 101.92 144.81 CHERRY 45.56 33.31 39.44 59.38 41.66 50.52 148.81 104.48 126.65 11-172 28.75 64.78 46.77 40.23 73.17 56.7 100.92 183.14 142.03 ROMA 29.62 16.60 23.11 41.54 24.96 33.25 104.14 62.70 83.42 L11 22.64 9.66 16.15 32.24 18.04 25.14 87.59 45.37 66.48 HEINZ-1370 48.15 11.40 29.78 62.29 19.77 41.03 156.03 49.70 102.87 NIRVANA F1 68.48 5.66 37.07 85.07 14.02 49.545 79.60 35.37 57.49 BUFFALO 13.33 16.11 14.72 30.83 36.47 33.65 54.03 64.87 59.45 INLAY F1 9.46 5.84 7.65 37.62 14.2 25.91 44.37 35.81 40.09 MONGAL F1 26.13 4.22 15.18 37.59 15.93 26.76 94.36 23.47 58.92 #20880 11.57 10.86 11.22 32.5 19.22 25.86 54.95 48.37 51.66 SUMO F1 17.09 97.66 57.38 38.7 106.02 72.36 67.06 265.36 166.21 ROMA VF 10.33 1.75 6.04 19.93 10.11 15.02 50.14 25.58 37.86 SHAKTIMAN 29.24 15.18 22.21 41.11 23.54 32.325 103.08 59.14 81.11 THORGAL F1 59.74 7.18 33.46 75.27 15.54 45.405 188.48 37.48 112.98 COBBRA F1 12.20 17.67 14.94 22.03 26.03 24.03 55.37 65.37 60.37 WOSOWOSO 14.35 120.07 67.21 24.43 128.43 76.43 61.37 321.37 191.37 MEAN 29.76 24.90 27.33 44.025 34.03 39.0 97.55 83.14 90.34 LSD(0.05) 13.32 2.36 6.71 3.32 1.86 1.87 8.14 4.17 4.45 Under field conditions, the lowest mean value of 15.36 mg/100 ml was recorded by genotype MONGAL F1 while the highest quercetin content of 212.71 mg/100 ml was recorded by the genotype WOSOWOSO. Across the two locations, the lowest mean value of 23.71 mg/100 ml was recorded by ROMA VF while the highest mean value of 126.05 was recorded by University of Ghana http://ugspace.ug.edu.gh 80 genotype WOSOWOSO. Significant (P< 0.01) differences were observed among the 20 tomato genotypes. Also, the effects of location and Genotype x Location interaction were significantly (P< 0.01) different (Appendix 15). 4.4.14 Mean antioxidant activity (mg/100 ml) (AA) The mean values for antioxidant activity in tomato among all the genotypes evaluated are presented in Table 4.16. The lowest mean IC50 value of 1.96 mg/100 ml under greenhouse conditions was recorded by SHAKTIMAN while the highest mean value of 5.61 mg/100 ml was recorded by PECTOMECH. Under field conditions, the lowest mean IC50 value of 0.92 mg/100 ml was recorded by genotype WOSOWOSO while the highest value of 10.28 mg/100 ml was recorded by #20880. The genotype WOSOWOSO recorded the lowest mean value of 1.58 mg/100 ml across the two locations while the highest mean value of 6.53 mg/100 ml was recorded by genotype #20880. Combined analysis of variance showed significant (P< 0.01) genotype, location and Genotype x Location interaction effects (Appendix 15). University of Ghana http://ugspace.ug.edu.gh 81 Table 4. 16 Mean Quercetin content and Antioxidant activity Quercetin content (mg/100 ml) Antioxidant activity (mg/100 ml) Genotype Green house Open field Pooled Mean Green house Open field Pooled Mean PECTOMECH 61.22 31.30 46.26 5.61 3.39 4.50 PLATINUM F1 39.45 36.34 37.90 5.31 2.87 4.09 NS 504 103.50 16.40 59.95 3.17 1.73 2.45 NKANSAH HT 123.60 66.41 95.01 3.10 2.19 2.65 CHERRY 97.67 68.12 82.90 2.67 4.21 3.44 11-172 65.74 120.56 93.15 3.64 6.72 5.18 ROMA 67.89 40.26 54.08 3.50 2.83 3.17 L11 52.41 28.70 40.56 4.91 3.25 4.08 HEINZ-1370 102.48 31.60 67.04 3.68 7.86 5.77 NIRVANA F1 73.75 22.04 47.90 2.11 9.29 5.70 BUFFALO 36.61 42.78 39.70 2.01 2.71 2.36 INLAY F1 28.04 22.34 25.19 2.60 3.69 3.15 MONGAL F1 61.38 15.36 38.37 3.47 4.99 4.23 #20880 36.21 30.71 33.46 2.78 10.28 6.53 SUMO F1 43.17 175.38 109.28 4.10 2.48 3.29 ROMA VF 31.89 15.53 23.71 2.90 5.57 4.24 SHAKTIMAN 67.18 36.79 51.99 1.96 2.98 2.47 THORGAL F1 124.12 23.45 73.79 3.61 2.63 3.12 COBBRA F1 35.38 42.04 38.71 3.91 4.24 4.08 WOSOWOSO 39.38 212.71 126.05 2.24 0.92 1.58 Mean 64.55 53.94 59.25 3.36 4.24 3.80 LSD(0.05) 5.48 3.05 3.14 0.31 0.42 0.26 4.4 Variance components, heritability and genetic gain of tomato characters 4.4.1Variance components of tomato traits studied under greenhouse conditions The components of variance were estimated for all traits scored under greenhouse conditions (Tables 4.17-4.19). With the exception of stem girth, pH and gallic acid content which recorded low values (<10 %) for genotypic and phenotypic coefficient of variation (GCV and PCV), all the other studied traits had moderate (10–20 %) to high (>20%) estimates. Number of fruits per plant recorded the highest estimated value of GCV and PCV (74.24 and 74.34 respectively) whereas gallic acid had the least estimated values of the same genotypic parameter (5.04 and 5.27 respectively). Generally, the PCV estimated were higher than the University of Ghana http://ugspace.ug.edu.gh 82 corresponding GCV for all the agronomic and quality characters evaluated indicating the relative role of environment on the expression of the traits. However, the ratio of the GCV to PCV was near unity for almost all the characters studied. 4.4.2 Heritability of tomato traits studied under greenhouse conditions The estimate of broad sense heritability for tomato characters studied under greenhouse conditions are presented in Tables 4.17–4.19. High estimate of broad sense heritability (> 60 %) was recorded for most of the studied traits. The lowest broad sense heritability value of 42 % was recorded by the trait ascorbic acid. Traits including number of fruits per plant, number of locules per fruit, pH, β-carotene, lycopene and catechin recorded the highest estimated broad sense heritability value of 100 %. 4.4.3 Genetic gain (GAM) of tomato traits studied under greenhouse conditions The estimated genetic gain expressed as genetic advance as percentage of mean (GAM) of all tomato characters scored under greenhouse conditions is presented in Table 4.17–4.19. Genetic gain was high (> 20 %) for almost all the studied traits. However, gallic acid recorded low (< 10 %) estimated value while stem diameter, pH and rosmarinic acids recorded moderate (10 – 20 %) values. Number of fruits per plant recorded the highest value (152.73) for estimated GAM. 4.4.4 Variance components of tomato traits studied under field conditions The results of the estimated genotypic parameters for tomato traits gathered from the open field experimental location are presented in Tables 4.20 – Table 4.22. The magnitude of variability for most characters ranged from moderate to high as showed by the moderate (10 – 220 %) and high (> 20 %) estimated values for both GCV and PCV. However, chlorophyll content, number of days to 50 % flowering, fruit set percentage, number of days to fruit maturity and pH recorded low estimates (<10%) of both GCV and PCV. However, number of University of Ghana http://ugspace.ug.edu.gh 83 days to 50 % flowering, percentage fruit set, number of days to fruit maturity and pH recorded low estimates (<10%) of both GCV and PCV. Also, plant height and stem girth recorded low (< 10 %) values of GCV. Higher values of GCV and PCV were observed in most of the quality parameters. The highest estimated values of GCV and PCV (100.01and 100.07 respectively) were recorded by quercetin (an antioxidant trait) while the lowest values (2.952.and 98 respectively) was recorded by the trait pH. Generally, the estimated PCV values were higher than their corresponding GCV values. The ratio of GCV to PCV was close to unity for almost all traits suggesting the dominant role of genetic factors in the phenotypic expression of the traits. University of Ghana http://ugspace.ug.edu.gh 84 Table 4. 17 Estimate of variances, heritability and genetic advance for 16 vegetative and reproductive characters under greenhouse conditions Variance component Heritability & genetic gain Character GM σ2G σ 2 P σ 2 E GCV PCV ECV H 2 b (%) EGA GAM Pant height (cm) 128.24 225.81 227.19 1.38 11.72 11.75 0.92 99 30.86 24.07 Stem girth (cm) 7.48 0.28 0.34 0.06 7.03 7.76 3.27 82. 0.98 13.13 Number of leaves 35.00 89.84 90.42 0.58 27.08 27.17 2.18 99 19.46 55.61 Chlorophyll content 31.12 30.63 32.37 1.74 17.79 18.28 4.24 95 11.25 35.64 Number of primary branches 3.00 0.56 0.62 0.06 24.94 26.25 8.16 90 1.47 48.84 Root length per plant 21.31 19.76 20.58 0.82 20.86 21.29 4.25 96 8.97 42.11 Number of days to 1st flowering 17.00 3.64 3.81 0.17 11.23 11.49 2.43 96 3.84 22.61 Number of days to 50 % flowering 21.00 9.11 9.42 0.31 14.38 14.62 2.65 97 6.12 29.12 Number of days to first fruit set 38.00 43.21 43.57 0.36 17.37 17.37 1.58 99 13.49 35.49 Number of days to 50 % fruit set 46.00 69.50 69.99 0.49 18.12 18.19 1.52 99 17.11 37.20 Percentage fruit set 57.16 137.70 193.77 56.07 20.53 24.35 13.10 71 20.38 35.65 Number of days to fruit maturity 65.00 46.64 46.97 0.33 10.51 10.54 0.88 99 14.02 21.57 Number of truss per plant 13.00 14.72 15.60 0.88 29.51 30.38 7.22 94 7.68 59.06 Number of flowers per truss 6.00 0.94 1.12 0.18 16.19 17.66 7.07 84 1.83 30.56 Number of fruits per truss 4.00 0.65 0.75 0.10 20.10 21.60 7.91 87 1.54 38.54 Number of fruits per plant 11.00 66.68 66.86 0.18 74.24 74.34 3.86 100 16.80 152.72 GM = Grand mean, σ2G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 85 Table 4. 18 Estimate of variance, heritability and genetic advance for 15 fruit physical and quality characters of tomato under greenhouse conditions Variance component Heritability &genetic advance Character GM σ2G σ 2 P σ 2 E GCV PCV ECV H 2 b EGA GAM Single fruit weight 68.90 685.70 696.69 10.99 38.01 38.31 4.81 98 53.52 77.67 Total fruit weight per plant 606.40 133003.00 171274.00 38271.0 60.14 68.25 32.26 78 662.04 109.18 Fruit yield per plant 52.60 1239.23 1243.87 4.65 66.93 67.05 4.10 99 71.93 136.75 Number of locules per fruit 4.00 3.09 3.09 0.00 43.97 43.97 0.00 100 3.62 90.58 Fruit length (cm) 4.65 0.48 0.50 0.02 14.90 15.20 3.04 96 1.40 30.07 Fruit diameter (cm) 5.07 1.39 1.44 0.05 23.29 23.70 4.41 97 2.39 47.13 Fruit shape index 0.95 0.03 0.03 0.00 17.78 18.99 6.64 88 0.33 34.32 Pericarp thickness 4.66 0.63 0.73 0.10 17.09 18.39 6.79 86 1.52 32.71 Fruit firmness 3.90 0.30 0.44 0.14 14.13 17.09 9.60 68 0.94 24.08 pH 4.30 0.10 0.10 0.0002 7.35 7.36 0.33 100 0.65 15.13 Total soluble solids content 4.44 1.35 1.38 0.03 26.13 26.42 3.90 98 2.36 53.24 Titrable acidity 0.32 0.01 0.01 0.0003 25.32 25.90 5.41 96 0.16 51.01 TSS/TA 14.08 7.44 7.74 0.30 19.38 19.76 3.89 96 5.51 39.14 Malic acid 0.34 0.01 0.01 0.0003 23.38 24.37 5.09 96 0.16 48.01 Fruit dry matter content 0.44 0.01 0.01 0.001 26.09 27.08 7.24 93 0.23 51.79 GM = Grand mean, σ2G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of meas. University of Ghana http://ugspace.ug.edu.gh 86 Table 4. 19 Estimate of variance, heritability and genetic advance for 14 antioxidant characters of tomato under greenhouse conditions Variance component Heritability & genetic advance Character GM σ2G σ 2 P σ 2 E GCV PCV ECV H 2b EGA GAM β-Carotene (mg/100ml) 19.89 67.37 67.62 0.25 41.27 41.34 2.51 100 16.88 84.85 Lycopene (mg/100ml) 16.22 41.83 42.01 0.18 39.87 39.96 2.62 100 13.29 81.96 Ascorbic acid (mg/100ml) 11.73 5.59 13.26 7.67 20.16 31.05 23.61 42 3.16 26.97 Gallic acid (mg/100ml) 1.58 0.01 0.01 0.001 5.04 5.42 2.00 86 0.15 9.64 Vanillic acid (mg/100ml) 0.89 0.02 0.03 0.01 14.51 18.35 11.24 63 0.21 25.62 Rosmarinicacid (mg/100ml) 0.83 0.01 0.02 0.01 12.05 17.04 12.05 50 0.15 17.55 Rutin (mg/100ml) 97.55 2131.56 2155.80 24.24 47.33 47.60 5.05 99 94.57 96.95 Quercetin( mg/100ml) 64.55 939.26 950.25 10.99 47.48 47.75 5.14 99 62.77 97.23 Catechin (mg/100ml) 105.71 1877.26 1882.27 5.01 40.99 41.04 2.12 100 89.14 84.32 Hesperetin (mg/100ml) 120.28 6065.47 6273.37 207.90 64.75 65.85 11.99 97 157.75 131.15 Kaempferol (mg/100ml) 29.76 327.52 392.49 60.81 60.81 66.57 27.08 83 34.06 114.43 Myricetin (mg/100ml) 44.02 367.02 371.05 4.03 43.52 43.76 4.56 99 39.25 89.16 Genestein (mg/100ml) 64.08 518.38 527.96 9.58 35.53 35.86 4.83 98 46.47 72.52 Antioxidant activity (IC50) 3.36 1.08 1.11 0.03 30.89 31.32 5.15 97 2.11 62.77 GM = Grand mean, σ2G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 87 4.4.5 Heritability of tomato traits studied under field conditions Broad sense heritability estimated for all characters studied under the open field condition is presented in Table 4.20 – Table 4.22. Estimated broad sense heritability was high (> 60 %) for most of the studied traits. However, traits including stem girth, fruit set percentage, number of fruits per truss, fruit dry matter content, gallic acid, vanillic acid and rosmarinic acid recorded moderate (30-60%) heritability values. The highest broad sense heritability estimate of 100 % was recorded by flavonoid compounds, rutin, quercetin, catechin, hesperetin, Kaempferol, myricetin and genestein while the lowest value of 40 % was recorded by gallic acid. 4.4.6 Genetic gain (GAM) of tomato traits studied under field conditions The estimated genetic gain expressed as genetic advance as percentage of mean of all tomato traits studied is presented in Table 4.20 – 4.22. Moderate (10 – 20 %) to high (> 20 %) genetic gain (GAM) was recorded by most of the tomato traits (Table 4.4 – 4.6). However, the traits percentage fruit set and pH recorded low (< 10 %) values of GAM while moderate (10 – 20 %) estimates were recorded by plant height, stem girth, chlorophyll content, number of days to 50 % flowering and gallic acid. The lowest estimated GAM was observed for percentage fruit set (9.97) and pH (5.88). Higher values were particularly observed among the quality traits compared to that of the agronomic traits. The highest estimated value of 281.59 % was recorded by Hesperetin while the lowest value of 6.03 % was recorded by fruit pH. University of Ghana http://ugspace.ug.edu.gh 88 Table 4. 20 Estimate of variance, heritability and genetic advance for 16 vegetative and reproductive characters of tomato under field conditions Variance component Heritability & genetic advance Character GM σ2G σ 2 P σ 2 E GCV PCV ECV H2b EGA GAM Pant height (cm) 49.85 20.34 24.85 4.51 9.05 10.00 4.26 82 8.41 16.86 Stem diameter (cm) 8.35 0.47 1.14 0.67 8.24 12.81 9.80 41 0.91 10.92 Number of leaves 43.75 89.12 101.45 12.33 21.58 23.02 8.03 88 18.23 41.66 Chlorophyll content 40.03 13.78 17.66 3.88 9.27 10.50 4.92 78 6.76 16.88 Number of primary branches 4.00 0.39 0.54 0.15 15.61 18.37 9.68 72 1.09 27.33 Root length per plant 19.50 17.40 18.45 1.05 21.39 22.03 5.25 94 8.34 42.79 Number of days to 1st flowering 16.00 9.12 9.44 0.32 18.87 19.20 3.54 97 6.11 38.22 Number of days to 50 % flowering 25.00 4.95 5.53 0.58 8.90 9.40 3.05 90 4.33 17.34 Number of days to first fruit set 37.00 18.72 20.68 1.96 11.69 12.29 3.78 91 8.48 22.92 Number of days to 50 % fruit set 46.00 25.36 29.84 4.48 10.95 11.88 4.60 85 9.56 20.79 Percentage fruit set 55.39 13.79 26.48 12.69 6.71 9.29 6.43 52 5.52 9.97 Number of days to fruit maturity 66.00 28.15 29.96 1.81 8.04 8.29 2.04 94 10.59 16.05 Number of truss per plant 16.00 50.94 52.02 1.08 44.61 45.08 6.50 98 14.55 90.93 Number of flowers per truss 7.00 0.78 1.12 0.34 12.59 15.10 8.33 70 1.51 21.63 Number of fruits per truss 3.00 0.27 0.50 0.23 17.32 23.57 15.99 54 0.79 26.22 Number of fruits per plant 13.00 71.68 72.38 0.70 65.12 65.44 6.44 99 17.36 133.51 GM = Grand mean, σ2L = Variance due to location, σ 2 G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 GL = Genotype x location variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 89 Table 4. 21 Estimate variance, heritability and genetic advance for 15 fruit physical and quality characters of tomato under field conditions Variance component Heritability & genetic advance Character GM σ2G σ 2 P σ 2 E GCV PCV ECV H 2 b EGA GAM Single fruit weight 57.82 424.99 429.11 4.12 35.65 35.83 3.51 99 42.26 73.09 Total fruit weight per plant 660.70 117805.71 119198.71 1393.00 51.95 52.26 5.65 99 702.91 106.39 Fruit yield per plant 55.10 817.96 827.67 9.71 51.91 52.21 5.66 99 58.57 106.30 Number of locules per fruit 4.00 2.34 3.18 0.84 38.22 44.56 22.91 74 2.70 67.52 Fruit length (cm) 4.86 0.74 0.84 0.10 17.66 18.82 6.51 88 1.66 34.14 Fruit diameter (cm) 4.89 0.99 1.09 0.10 20.38 21.38 6.47 91 1.96 40.02 Fruit shape index 1.03 0.06 0.07 0.004 24.17 24.94 6.14 94 0.50 48.27 Pericarp thickness 4.29 0.68 1.05 0.37 19.22 23.89 14.18 65 1.37 31.87 Fruit firmness 3.18 0.15 0.24 0.09 12.31 15.51 9.43 63 0.64 20.14 pH 4.36 0.02 0.02 0.0003 2.95 2.98 0.40 98 0.26 6.03 Total soluble solids content 4.00 1.36 1.38 0.02 29.19 29.40 3.54 99 2.39 59.70 Titrable acidity 0.40 0.01 0.01 0.0004 20.21 20.82 5.00 94 0.16 40.41 TSS/TA 9.96 2.97 3.11 0.14 17.30 17.71 3.76 95 3.47 34.83 Malic acid 0.42 0.01 0.01 0.001 18.95 20.39 7.53 86 0.15 36.27 Fruit dry matter content (g) 0.36 0.003 0.005 0.002 15.21 19.64 12.42 60 0.09 24.28 GM = Grand mean, σ2L = Variance due to location, σ 2 G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 GL = Genotype x location variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 90 Table 4. 22 Estimate of variance, heritability and genetic advance for antioxidant characters of tomato under field conditions Variance component Heritability & genetic advance Character GM σ2G σ 2 P σ 2 E GCV PCV ECV H2b EGA GAM β-Carotene (mg/100 ml) 23.35 26.71 35.66 8.95 22.13 25.58 12.81 75 9.21 39.46 Lycopene (mg/100 ml) 15.41 31.24 31.79 0.55 36.27 36.59 4.82 98 11.41 74.06 Ascorbic acid (mg/100 ml) 9.41 4.75 4.96 0.21 23.16 23.67 4.87 96 4.39 46.69 Gallic acid (mg/100 ml) 1.75 0.05 0.13 0.08 13.20 20.87 16.16 40 0.30 17.19 Vanillic acid (mg/100 ml) 1.16 0.15 0.36 0.21 33.39 51.72 39.50 42 0.52 44.40 Rosmarinicacid (mg/100 ml) 1.03 0.09 0.22 0.13 29.13 45.54 35.01 41 0.40 38.38 Rutin (mg/100 ml) 83.14 6557.17 6563.52 6.35 97.40 97.44 3.03 100 166.73 200.54 Quercetin( mg/100 ml) 53.94 2910.14 2913.55 3.41 100.01 100.07 3.42 100 111.06 205.90 Catechin (mg/100 ml) 94.94 5387.53 5403.67 16.14 77.31 77.43 4.23 100 150.98 159.02 Hesperetin (mg/100 ml) 98.49 18143.76 18162.13 18.37 136.76 136.83 4.35 100 277.34 281.59 Kaempferol (mg/100 ml) 24.90 1041.21 1043.24 2.03 129.59 129.72 5.72 100 66.41 266.69 Myricetin (mg/100 ml) 34.03 1030.72 1031.98 1.26 94.34 94.40 3.30 100 66.10 194.23 Genestein (mg/100 ml) 53.55 1435.03 1437.07 2.04 70.74 70.79 2.67 100 77.98 145.62 Antioxidant activity (mg/100 ml) 4.24 6.32 6.39 0.07 59.31 59.63 6.24 99 5.15 121.50 GM = Grand mean, σ2L = Variance due to location, σ 2 G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 GL = Genotype x location variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 91 4.4.7 Variance components of tomato traits estimated over combined locations Components of variance estimated over the combined greenhouse and open field locations are presented in Tables 4.23 – 4.25. The estimated variance due to location (σ2L) was higher than the corresponding genotypic variance (σ2G) for some vegetative and reproductive traits. Such traits included plant height, stem girth, chlorophyll content, root length and number of days to first flowering. Similarly, quality traits including fruit firmness, acidity, TSS/TA, malic acid, fruit dry matter content, β-carotene, gallic acid, vanillic acid, rosmarinic acid, myricetin, genestein as well as antioxidant activity recorded higher values of variance due to location (σ2L) than corresponding genotypic variance (σ2G). Generally, estimated values of phenotypic (σ2P) and interaction (σ 2 GL) variance were higher than the corresponding genotypic variance (σ2G) for most of the traits. The variances due to G x L were much larger than the environmental variance components for most of the studied characters indicating that the traits differed considerably with environment. The estimated environmental variances for stem girth, number of fruits per truss, number of locules per fruit, vanillic acid and rosmarinic acid were higher than the variance component due to G x L interaction. This indicated that those traits varied strongly with the environment as a result of unpredictable environmental factors. With the exception of characters number of days to 1st fruit set and number of flowers per truss, the ratio of the genotypic variance to genotype by location interaction variance (σ2G: σ 2 G x L) observed for almost all the studied traits were far from unity. Among all the studied characters, high (> 20 %) estimated values of GCV and PCV were recorded by number of truss per plant, number of fruits per truss, number of fruits per plant, single fruit weight per plant, fruit weight per plant, total fruit yield per plant and number of University of Ghana http://ugspace.ug.edu.gh 92 locules per fruit. The differences between the GCV and PCV were large for most of the studied traits indicating a significant role of the environment on the expression of those traits. High (> 20 %) estimated GCV and PCV was recorded by number of truss per plant, number of fruits per truss, number of fruits per plant, single fruit weight per plant, fruit weight per plant, total fruit yield per plant and number of locules. The differences between the GCV and PCV were large for most of the studied traits indicating a significant role of environment on the expression of those traits. 4.4.8 Heritability of tomato traits estimated over combined locations Heritability estimated for agronomic traits over the combined locations are presented in Tables 4.23 – 4.25. Among the vegetative and reproductive traits studied, number of leaves, number of days to first fruit set, number of days to 50 % fruit set, number of days to fruit maturity, number of truss per plant, number of flowers per truss, number of fruits per truss and number of fruits per plant recorded high ( > 60 %) broad sense heritability estimates. Moderate (30 – 60 %) heritability values were observed by plant height, stem girth, number of primary branches and number of days to 1st flowering while number of days to 50 % flowering and percentage fruit set recorded low (<30 %) heritability estimates. Vanillic acid recorded the lowest heritability estimate (11 %). The highest estimated value of 96 % was recorded by number of locules per fruit. 4.4.9 Genetic gain (GAM) of tomato traits estimated over combined locations The estimated values of genetic gain expressed as genetic advance as percentage of mean are presented in Tables 4.23 – 4.25. Number of leaves per plant, number of days to 50% fruit set, number of truss per plant, number of fruits per truss and number of fruits per plant recorded University of Ghana http://ugspace.ug.edu.gh 93 higher values of genetic gain (> 20 %). With the exception of characters including plant height, number of primary branches per plant, number of days to 1st flowering, number of days to first fruit set, and number of flowers per truss where moderate (10 – 20 %) GAM was recorded, all other vegetative and reproductive traits had low (< 10 %) estimated values. High (> 20 %) GAM was observed for almost all the quality traits except Pericarp thickness, fruit firmness, malic acid and fruit dry matter where low estimate of GAM was recorded. Titrable acidity and TSS/TA recorded moderate genetic gain (10 – 20 %). Among the antioxidant composition traits, very low GAM was observed for lycopene content, ascorbic acid, gallic acid, vanillic acid and rosmarinic acid contents. Non-significant estimated genetic gain values were recorded for all the other polyphenol traits. University of Ghana http://ugspace.ug.edu.gh 94 Table 4. 23 Estimate of variance, heritability and genetic advance for 16 vegetative and reproductive characters of tomato over combined locations Variance components Heritability and genetic advance Character GM σ2L σ 2 G σ 2 P σ 2 GL σ 2 E GCV PCV ECV H 2 b EGA GAM Pant height (cm) 89.17 3086.69 41.03 82.55 82.03 3.06 7.18 10.19 1.96 50 9.30 10.43 Stem girth (cm) 7.91 0.37 0.20 0.35 0.16 0.40 5.65 7.46 8.00 57 0.70 8.83 Number of leaves 40.00 32.44 53.60 72.62 35.77 6.78 18.30 21.30 6.51 74 12.96 32.39 Chlorophyll content 35.57 38.44 00 12.68 24.45 2.70 - 10.01 4.62 00 00 00 Number of primary 4.00 0.00 0.20 0.36 0.27 0.11 11.18 14.90 8.29 56 0.69 17.29 Root length per plant 20.40 0.32 00 13.03 25.75 0.91 - 17.69 4.68 00 00 00 Number of days to 1st flowering 17.00 0.85 2.35 4.41 4.02 0.29 9.01 12.35 3.17 53 2.30 13.54 Number of days to 50 % flowering 23.00 10.12 1.12 4.15 5.91 0.46 4.60 8.86 2.95 27 1.13 4.92 Number of days to first fruit set 38.00 1.32 15.17 23.04 15.35 1.13 10.25 12.63 2.80 66 6.51 17.14 Number of days to 50 % fruit set 45.00 0.26 27.78 38.02 19.67 2.43 11.71 13.70 3.46 73 9.28 20.63 Percentage fruit set 56.28 00 13.02 50.11 62.96 33.69 6.41 12.58 10.31 26 3.79 6.73 Number of days to fruit maturity 66.00 00 31.46 34.61 5.93 1.06 8.50 8.91 1.56 91 11.02 16.69 Number of truss per plant 15.00 2.32 18.67 25.91 14.17 0.96 28.80 33.94 6.53 72 7.55 50.36 Number of flowers per truss 6.00 0.11 0.44 0.70 0.42 0.25 11.08 13.89 8.33 64 1.09 18.19 Number of fruits per truss 3.00 0.01 0.38 0.44 0.08 0.16 20.41 22.19 13.33 85 1.16 38.67 Number of fruits per plant 12.00 3.23 61.62 65.47 7.56 0.44 65.42 67.43 5.43 94 15.69 130.73 GM = Grand mean, σ2L = Variance due to location, σ 2 G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 GL = Genotype x location variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 95 Table 4. 24 Estimate of variance, heritability and genetic advance for 16 fruit physical and quality characters of tomato over combined locations Variance component Heritability & genetic advance as % of mean Character GM σ2L σ 2 G σ 2 P σ 2 GL σ 2 E GCV PCV ECV H 2 b (%) EGA GAM Single fruit weight 63.37 56.57 456.76 507.31 98.62 7.46 33.73 35.54 4.31 90 41.78 65.92 Fruit weight per plant 633.50 92.40 104328.17 118171.67 21057.33 19889.0 50.99 54.26 22.26 88 625.19 98.69 Fruit yield per plant 53.80 00 871.66 951.32 156.98 7.05 54.88 57.33 4.94 92 58.22 108.21 No. of locules per fruit 4.00 0.02 2.65 2.75 0.06 0.44 40.70 41.48 16.58 96 3.29 82.25 Fruit length 4.76 0.01 0.39 0.51 0.22 0.06 13.04 14.95 5.15 76 1.11 23.41 Fruit diameter 4.98 0.003 0.96 1.09 0.23 0.08 19.66 20.95 5.68 88 1.89 38.00 Fruit shape index 0.99 0.002 0.03 0.04 0.02 0.004 15.97 19.34 6.39 68 0.27 27.17 Pericarp thickness 4.48 0.04 0.14 0.44 0.52 0.24 8.20 14.72 10.94 31 0.42 9.41 Fruit firmness 3.54 0.27 0.06 0.17 0.17 0.11 7.11 11.47 9.37 38 0.32 9.07 pH 4.33 00 0.16 0.32 0.33 0.02 9.09 13.13 3.27 48 0.56 12.97 Total soluble solids 4.22 0.05 0.41 0.89 0.95 0.03 15.11 22.29 4.10 46 0.89 21.10 Titrable acidity 0.36 0.003 0.001 0.004 0.005 0.0003 8.78 17.57 4.81 25 0.03 9.05 TSS/TA 12.02 8.37 1.46 3.37 3.73 0.22 10.06 15.26 3.90 43 1.64 13.67 Malic acid 0.38 0.003 0.001 0.006 0.01 0.0005 8.32 20.38 5.88 17 0.03 7.00 Fruit dry matter 0.40 0.003 0.002 0.01 0.01 0.001 11.18 18.26 7.91 38 0.06 14.10 GM = Grand mean, σ2L = Variance due to location, σ 2 G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 GL = Genotype x location variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 96 Table 4. 25 Estimate of variance, heritability and genetic advance for antioxidant characters of tomato over combined locations Variance component Heritability and genetic advance as %of mean Character GM σ2L σ 2 G σ 2 P σ 2 GL σ 2 E GCV PCV ECV H 2 b (%) EGA GAM β-Carotene (mg/100 ml) 21.62 3.07 00 29.33 57.15 4.51 - 25.05 9.82 00 00 00 Lycopene (mg/100 ml) 15.19 00 2.74 19.70 33.79 0.36 11.04 29.59 4.00 14 1.27 8.49 Ascorbic acid (mg/100 ml) 10.57 2.37 00 3.26 5.21 3.93 - 17.08 18.76 00 00 00 Gallic acid (mg/100 ml) 1.66 0.01 00 0.04 0.06 0.04 - 11.79 12.05 00 00 00 Vanillic acid (mg/100 ml) 1.03 0.03 0.01 0.06 0.08 0.10 7.93 24.43 30.70 11 0.05 5.30 Rosmarinicacid (mg/100 ml) 0.93 0.02 0.01 0.04 0.04 0.07 7.60 21.05 28.45 13 0.05 5.66 Rutin (mg/100 ml) 90.34 00 00 2438.54 4872.10 14.95 - 54.66 4.28 00 00 00 Quercetin( mg/100 ml) 59.25 00 00 1099.00 2195.50 7.48 - 55.95 4.62 00 00 00 Catechin (mg/100 ml) 100.32 00 00 1986.18 3968.82 10.64 - 44.42 3.25 00 00 00 Hesperitin (mg/100 ml) 109.39 00 00 6685.07 13333.13 111.0 - 74.74 9.63 00 00 00 Kaempferol (mg/100 ml) 27.33 00 00 407.24 803.11 34.12 - 73.84 21.37 00 00 00 Myricetin (mg/100 ml) 39.03 8.50 00 414.77 828.65 2.65 - 52.18 4.17 00 00 00 Genestein (mg/100 ml) 58.82 0.06 00 553.80 1105.66 5.83 - 40.01 4.10 00 00 00 Antioxidant activity (mg/100 ml) 3.80 0.18 00 2.06 4.10 0.05 - 37.75 5.88 00 00 00 GM = Grand mean, σ2L = Variance due to location, σ 2 G = Genotypic variance, σ 2 P = Phenotypic variance, σ 2 GL = Genotype x location variance, σ 2 E = Environmental variance, GCV = Genotypic coefficient of variability, PCV = Phenotypic coefficient of variability, ECV =Environmental coefficient of variability, h2b = Broad sense heritability, EGA = Expected genetic advance and GAM = Genetic advance as percentage of mean. University of Ghana http://ugspace.ug.edu.gh 97 4.5 Character association among tomato genotypes 4.5.1 Correlation of tomato traits studied under greenhouse conditions The Pearson’s correlation coefficients among selected agronomic and fruit quality traits studied under greenhouse conditions revealed an association between trait pairs (Table 4.26). A significant and positive association was observed between number of trusses per plant (TPP) and number of primary branches per plant (NPB) (r = 0.588**). Number of fruits per plant (FPP) correlated with number of primary branches per plant (NPB) (r = 0.517*) and number of trusses per plant (TPP) (r = 0.643**). Number of fruits per plant (FPP) associated with number of primary branches per plant (NPB) (r = 0.517*) and number of trusses per plant (TPP) (r = 0.643**). Ascorbic acid content of fruit (ASC) correlated with number of primary branches per plant (NPB) (r = 0.474*), number of truss per plant (TPP) (r = 0.535**) and number of fruits per plant (FPP) (0.518*) while total fruit weight per plant (FWP) correlated with number of trusses per plant (TPP) (r = 0.559**) and number of fruits per plant (FPP) (0.687**). Tomato fruit yield (YPP) correlated positively and significantly with number of trusses per plant (TPP) (r= 0.562**) fruit weight per plant (FWP) (r =1.00**) and number of fruits per plant (FPP) (r = 0.686**). A positive correlation was observed between each of the following trait pair: number of days to fruit maturity (DFM) and number of days to 50 % fruit set (FPFS) (r = 0.804**), fruit length (FL) and number of days to fifty percent fruit set (FPFS) (r = 0.459*) as well as titrable acidity (TA) with total soluble solids content (TSS) (0.744**). Gallic acid (GA) also showed a positive and significant association with plant height (PHT) (r = 0.457*) and number of days to fruit maturity (DFM) (0.554**). On the other hand, a significant negative association was observed between certain trait pairs (Table 4.26). Fruit yield per plant (YPP) showed a negative significant association with University of Ghana http://ugspace.ug.edu.gh 98 number of days to 50 % fruit set (FPFS) (r = -0.573**), number of days to fruit maturity (DFM) (-0.627**) and gallic acid content (GA) (-469**). A significant negative correlation was observed for fruit length (FL) and number of primary branches per plant (NPB) (r = - 0.561**) and number of fruits per plant (FPP) (r = -0.460*), ascorbic acid content (ASC) (r = -0.557**) and quercetin content in fruit (QUE) (r = -0.447*). The association of number of fruits per plant (FPP) with number of days to 50 % flowering (FPF) (r = -0.454*), number of days to 50 % fruit set (FPFS) (r = -0.624**) and gallic acid (GA) (-460*) were found to be significantly negative. Fruit weight per plant (FWP) correlated negatively with number of days to 50 % fruit set (FPFS) (r = -0.572**), number of days to fruit maturity (DFM) (r = -6.27**) and gallic acid content (GA) (r = -0.462*). Moreover, a negative and significant association was recorded for trait pairs including number of trusses per plant (TPP) and number of days to 50 % fruit set (FPFS) (r = -0.534**). Quercetin content in fruit (QUE) correlated negatively and significantly with fruit diameter (FD) (r = -0.481*). 4.5.2 Correlation of tomato traits studied under field conditions The results of the correlation among selected agronomic and fruit quality traits of field- produced tomato genotypes were significant for a number of traits (Table 4.27). A significant positive association was observed between trait pairs number of trusses per plant (TPP) and number of primary branches per plant (NPB) (r = 0.605**), number of fruits per plant (FPP) and number of days to fruit maturity (DFM) (r = 0.740**). Fruit length (FL) correlated with number of days to 50 % flowering (FPF) (r = 0.630**) while fruit diameter (FD) associated with single fruit weight per plant (SFW) (r = 0.744**), titrable acidity (TA) (r = 0.744**) and total soluble solids content (TSS) (r = 0.751**). Antioxidant activity (AA) correlated with number of days to fruit maturity (DFM) (r = 0.674**) University of Ghana http://ugspace.ug.edu.gh 99 Table 4. 26 The Pearson’s correlation matrix of selected agronomic and quality traits of tomato under greenhouse conditions Variables PHT NPB FPF FPFS DFM TPP FPP FWP YPP SFW FL FD FF TSS TA LYC ASC GA QUE PHT - NPB 0.185 - NFF 0.116 -0.250 - FPFS 0.207 -0.384 0.436 - DFM 0.192 -0.251 0.311 0.804** - TPP 0.089 0.588** -0.380 -0.534** -0.386 - FPP -0.019 0.517* -0.454* -0.624** -0.607** 0.643** - FWP 0.049 0.230 -0.323 -0.572** -0.627** 0.559** 0.687** - YPP 0.050 0.231 -0.324 -0.573** -0.627** 0.562** 0.686** 1.000** - SFW -0.051 -0.066 0.218 0.061 0.020 -0.077 -0.360 0.309 0.308 - FL -0.031 -0.561** 0.001 0.459* 0.307 -0.368 -0.460* -0.193 -0.194 0.170 - FD -0.099 -0.069 -0.015 0.072 -0.023 -0.107 -0.344 0.169 0.174 0.727** 0.426 - FF 0.056 -0.228 0.144 0.140 0.335 -0.064 -0.088 -0.154 -0.160 -0.011 0.229 -0.158 - TSS -0.076 -0.005 -0.203 -0.015 0.225 0.101 -0.315 -0.421 -0.419 -0.134 -0.081 -0.068 -0.223 - TA -0.269 -0.131 0.136 0.136 0.204 -0.149 -0.384 -0.312 -0.316 0.138 -0.063 -0.033 -0.193 0.744** - LYC 0.105 -0.022 0.301 0.182 0.418 -0.190 -0.193 -0.345 -0.344 -0.297 -0.163 -0.305 0.270 0.091 0.175 - ASC 0.050 0.474* 0.141 -0.277 -0.258 0.535** 0.518* 0.380 0.382 -0.056 -0.557** -0.265 -0.163 -0.192 -0.242 -0.110 - GA 0.457* -0.211 0.394 0.460* 0.554** -0.291 -0.518* -0.462* -0.460* -0.091 0.065 -0.224 -0.132 0.290 0.214 0.343 -0.119 - QUE -0.300 0.198 -0.062 -0.332 -0.260 0.399 0.219 -0.050 -0.050 -0.200 -0.447* -0.481* 0.077 0.137 0.014 -0.242 0.304 -0.209 - AA 0.079 0.025 0.139 0.252 -0.132 -0.043 0.009 0.205 0.204 0.151 0.159 -0.005 -0.130 -0.364 -0.017 -0.172 0.184 -0.102 -0.025 * = significant at P < 0.05 (r > 0.413), ** = significant at P < 0.01 (r > 0.526), PHT = Plant height, NPB = Number of primary branches per plant, FPF = Number of days to 50 % flowering, FPFS = Number of days to 1st fruit set, DFM = Number of days to fruit maturity, = TPP = Number of truss per plant, FPP = Number of fruits per plant, FWP = Fruit weight per plant, YPP = Yield per plant, FL = Fruit length, FD = Fruit diameter, FF = Fruit firmness, TSS = Total soluble solids content, TA = Titrable acidity content, LYC = Lycopene content, ACS = Ascorbic acid content, GA = Gallic acid content, QUE = Quercetin content, AA = Antioxidant activity. University of Ghana http://ugspace.ug.edu.gh 100 (r = 0.674). Number of days to fruit maturity (DFM) correlated positively and significantly with number of days to 50 % flowering (FPF) (r = 0.555**) and number of days to 50 % fruit set (FPFS) (r = 0.617**). A significantly positive association of fruit weight per plant with number of trusses per plant (TPP) (r = 0.629**) and number of fruits per plant (FPP) (r = 0.566**) was recorded. Tomato fruit yield per plant (YPP) showed a high significant positive association with number of trusses per plant (TPP) (r = 0.628**), total fruit weight plant (FWP) (r = 1.00**) and number of fruits per plant (FPP) (r = 0.566**). The correlations between fruit firmness (FF) with number of days to 50 % flowering (FPF) (r = 0.573**) and fruit length (FL) (0.641**) were positive and significantly high. There was also significant positive association between lycopene content of fruits (LYC) and total soluble solids content of fruit (TSS) (r = 0.731**). Gallic acid content (GA) associated significantly (P < 0.5) with fruit weight per plant (FWP) (r = 0.502*) as well as yield per plant (YPP) (r=0.502*). The flavonoid compound quercetin (QUE) correlated positively and significantly with single fruit weight per plant (SFW) (r = 0.476*), total soluble solids content (TSS) (0.596**) and titrable acidity (TA) (r = 0.736**). A significantly negative correlation was observed for some agronomic and fruit quality traits. The number of trusses per plant (TPP) showed a significant negative correlation with number of days to 50 % flowering (FPF) (r = -0.555**) and number of days to fruit maturity (DFM) (r = -0.490*). A significant negative association was observed to exist in some trait pairs including fruit weight per plant (FWP) and number of days to 50 % fruit set (FPFS)(r = - 0.469*), fruit yield per plant(YPP) and number of days to 50 percent fruit set (FPFS) (r = - 0.469*). Fruit length (FL) correlated with number of trusses per plant (TPP) (r = 0.495*), fruit diameter (FD) (and number of fruits per plant (FPP) (r = -0.632**), total soluble solids (TSS) and number of days to 50 % fruit set (FPFS) (r = -0.494*). Titrable acidity content of University of Ghana http://ugspace.ug.edu.gh 101 tomato fruits (TA) correlated with number of days to 50 % fruit set (-0.669**). Quercetin content of fruit (QUE) showed a negative significant correlation with number of days to 50 percent flowering (NFPFS) (-596**). 4.5.3 Correlation among tomato traits over combined locations The correlation among selected agronomic and fruit quality characters o f tomato determined over the combined greenhouse and open field locations are presented in Table 4.29. Positive and significant association was observed between trait pairs number of days to 50 % fruit set (NFPFS) and number of days to 1st flower observed (NFF) (r = 0.518*), number of trusses per plant (TPP) and number of primary branches per plant (NPB) (r = 0.689**), fruit length (FL) and number of days to 50 % fruit set (FPFS) (r = 0.553**), fruit diameter (FD) and single fruit weight per plant (SFW) (r = 0.826**), acidity (TA) and total soluble solids content of fruit (TSS) (r = 0.757**). Number of days to fruit maturity (DFM) correlated positively and significantly with number of days to 50 % flowering (FPF) (r = 0.514*), number of days to 50 % fruit set (FPFS)) (r = 0.791**) and lycopene content (LYC) (r = 0.702**). Number of fruits per plant associated positively and significantly with number of primary branches per plant (NPB) (r = 0.507*), number of truss per plant (TPP) (r = 0.847**) well as ascorbic acid (ASC) (r =0.516*). There was significantly positive character association between fruit weight per plant (FWP) and number of truss per plant (TPP) (r = 0.647**) as well as number of fruits per plant (FPP) (r = 0.622**). The association of fruit yield per plant (YPP) with number of truss per plant (TPP) (r = 0.648**), fruit weight per plant (FWP) (r = 1.00**) and number of fruits per plant (FPP) (r = 0.621**) was found to be highly significant. Quercetin (QUE) showed a significantly positive correlation with titrable acidity (TA) (r = 0.536*) and ascorbic acid (ASC) (r = 0.474*). University of Ghana http://ugspace.ug.edu.gh 102 Table 4. 27 The Pearson’s correlation matrix of selected agronomic and fruit quality traits of tomato under field conditions Variables PHT NPB FPF FPFS DFM TPP FPP FWP YPP SFW FL FD FF TSS TA LYC ASC GA QUE PHT - NPB 0.032 - NFF 0.166 -0.425 - FPFS -0.141 -0.271 0.431 - DFM 0.142 -0.103 0.519* 0.617** - TPP -0.154 0.605** -0.555** -0.382 -0.490* - FPP -0.266 0.320 -0.529** -0.350 -0.334 0.740** - FWP -0.037 0.379 -0.244 -0.469* -0.386 0.629** 0.566** - YPP -0.037 0.378 -0.244 -0.469* -0.386 0.628** 0.566** 1.000** - SFW -0.028 0.076 0.250 -0.123 0.012 -0.226 -0.402 0.248 0.248 - FL -0.090 -0.256 0.630** 0.439 0.382 -0.495* -0.291 -0.152 -0.152 0.048 - FD 0.185 0.020 0.198 -0.102 0.081 -0.379 -0.632** -0.027 -0.027 0.744** 0.097 - FF 0.002 0.055 0.573** 0.106 0.284 -0.208 -0.238 0.179 0.178 0.357 0.641* 0.423 - TSS 0.127 0.148 -0.286 -0.494* -0.227 0.043 0.209 0.039 0.039 0.245 -0.142 0.233 -0.122 - TA -0.071 0.292 -0.369 -0.669** -0.379 0.147 0.194 0.145 0.145 0.296 -0.357 0.215 -0.108 0.751** - LYC 0.287 -0.020 -0.020 -0.086 0.050 -0.053 0.150 -0.006 -0.006 0.098 0.139 0.159 -0.009 0.731** 0.117 - ASC -0.429 -0.046 -0.046 0.207 0.345 0.030 0.406 -0.002 -0.002 -0.232 0.017 -0.360 -0.142 -0.220 -0.160 -0.151 - GA -0.241 0.043 0.136 -0.010 0.049 0.015 0.135 0.502* 0.502* 0.329 0.008 0.031 0.212 -0.174 0.108 -0.260 0.265 - QUE -0.107 0.178 -0.260 -0.531** -0.180 -0.008 -0.008 0.086 0.085 0.476* -0.293 0.299 -0.218 0.596** 0.736* 0.144 0.045 0.233 - AA 0.102 -0.160 0.419 0.364 0.674** -0.219 -0.060 -0.255 -0.255 0.000 0.186 -0.040 0.147 -0.005 -0.395 0.408 0.201 -0.233 -0.301 * = significant at P < 0.05 (r > 0.413), ** = significant at P < 0.01 (r > 0.526), PHT = Plant height, NPB = Number of primary branches per plant, FPF = Number of days to 50 % flowering, FPFS = Number of days to 1st fruit set, DFM = Number of days to fruit maturity, = TPP = Number of truss per plant, FPP = Number of fruits per plant, FWP = Fruit weight per plant, YPP = Yield per plant, FL = Fruit length, FD = Fruit diameter, FF = Fruit firmness, TSS = Total soluble solids content, TA = Titrable acidity content, LYC = Lycopene content, ACS = Ascorbic acid content, GA = Gallic acid content, QUE = Quercetin content, AA = Antioxidant activity. University of Ghana http://ugspace.ug.edu.gh 103 In addition, the antioxidant activity (AA) of tomato across the locations associated positively with number of days to 50 % fruit set (FPFS) (r = 0.505) and number of days to fruit maturity (DFM) (r = 0.555). However, a significantly negative correlation was observed between certain traits determined over the combined locations (Table 4.28). Trait pair association was found negative and significant between number of days to 50 % flowering (FPF) and number of primary branches per plant (NPB) (r = -0.630), ascorbic acid content (ASC) and fruit length (FL) (r = -4.87). Quercetin content (QUE) associated negatively with fruit length (FL) (-0.476). Negative and significant correlation was observed for number of truss per plant (TPP) with number of days to 50 % flowering (FPF) (r = -0.673), number of days to 50 % fruit set (FPFS) (r = -0667) and number of days to fruit maturity (DFM) (r = -0.585). Similarly, number of fruits per plant (FPP) negatively associated with number of days to 50 % flowering (FPF) (r = -0.609), number of days to 50 % fruit set (FPFS) (r = -0.580) and number of days to fruit maturity (DFM) (-0.551). Both fruit weight per plant (FWP) and fruit yield per plant (YPP) showed separate negative significant correlation with number of days to 50 % fruit set (FPFS) ( r = -0.637, r = -0.636) and number of days to fruit maturity (DFM) (r = -0.576, r = -0.575). A significantly negative association was also observed for fruit length (FL) with number of primary branches per plant (NPB) (r = -0.634), number of truss per plant (TPP) (r = -0.596) and number of fruits per plant (FPP) (-0.584). University of Ghana http://ugspace.ug.edu.gh 104 Table 4. 28 The Pearson’s correlation matrix of selected agronomic and quality traits of tomato over combined greenhouse and field locations Variables PHT NPB NFF FPFS DFM TPP FPP FWP YPP SFW FL FD FF TSS TA LYC ASC GA QUE PHT - NPB 0.184 - NFF 0.093 -0.630** - FPFS 0.085 -0.437 0.518* - DFM 0.156 -0.284 0.514* 0.791** - TPP -0.109 0.689** -0.673** -0.667** -0.585** - FPP -0.090 0.507* -0.609** -0.580** -0.551** 0.847** - FWP -0.004 0.326 -0.401 -0.637** -0.576** 0.647** 0.622** - YPP -0.002 0.328 -0.400 -0.636** -0.575** 0.648** 0.621** 1.000** - SFW -0.066 0.041 0.205 -0.063 0.028 -0.187 -0.453* 0.274 0.274 - FL -0.064 -0.634** 0.395 0.553** 0.402 -0.596** -0.584** -0.286 -0.286 0.131 - FD 0.000 0.058 0.172 0.021 0.033 -0.249 -0.526* 0.093 0.096 0.826** 0.282 - FF -0.226 -0.158 0.030 0.325 0.369 -0.151 -0.028 -0.250 -0.254 -0.146 0.315 -0.241 - TSS 0.021 0.126 -0.079 -0.191 -0.002 0.052 -0.171 -0.218 -0.218 0.080 0.068 0.149 -0.119 - TA -0.087 0.092 0.065 -0.251 -0.129 -0.026 -0.124 -0.050 -0.053 0.305 -0.090 0.141 -0.126 0.757** - LYC 0.176 -0.083 0.237 0.405 0.702** -0.264 -0.214 -0.372 -0.373 -0.158 -0.023 -0.354 0.422 0.044 0.015 - ASC -0.184 0.195 0.041 -0.252 -0.005 0.380 0.516* 0.270 0.268 -0.145 -0.487* -0.404 0.154 -0.121 -0.006 0.119 - GA 0.202 -0.157 0.156 0.174 0.187 -0.122 0.032 0.384 0.381 0.307 0.081 -0.056 -0.028 -0.216 0.108 0.121 0.278 - QUE -0.098 0.334 -0.053 -0.443 -0.204 0.296 0.140 0.124 0.122 0.359 -0.476* 0.100 0.169 0.377 0.536** 0.027 0.474* 0.019 - AA -0.252 -0.165 0.318 0.505* 0.555** -0.239 -0.255 -0.239 -0.237 0.001 0.236 0.014 0.120 -0.027 -0.190 0.370 0.145 -0.013 -0.357 * = significant at P < 0.05 (r > 0.413), ** = significant at P < 0.01 (r > 0.526), PHT = Plant height, NPB = Number of primary branches per plant, NFF = Number of days to 1st flowering, FPFS = Number of days to 1st fruit set, DFM = Number of days to fruit maturity, = TPP = Number of truss per plant, FPP = Number of fruits per plant, FWP = Fruit weight per plant, YPP = Yield per plant, FL = Fruit length, FD = Fruit diameter, FF = Fruit firmness, TSS = Total soluble solids content, TA = Titrable acidity content, LYC = Lycopene content, ACS = Ascorbic acid content, GA = Gallic acid content, QUE = Quercetin content, AA = Antioxidant activity. University of Ghana http://ugspace.ug.edu.gh 105 CHAPTER FIVE 5.0 DISCUSSION 5.1 Variability in Agronomic Performance of Tomato Genotypes 5.1.1 Vegetative characteristics of tomato genotypes Vegetative traits play significant role in the overall development of the tomato crop. In the present study, significant variation was observed among the tomato genotypes for vegetative traits in all locations as well as across locations. Plant height and root length were significantly higher under greenhouse conditions as compared with the genotypes produced under field conditions. Similar observation was made by Rajasekar et al. (2013) who reported 150.25 cm and 52.41 cm of plant height under greenhouse and field conditions respectively. According to this author relative humidity was higher in the greenhouse environment which consequently resulted to higher vegetative growth and improved productivity of tomato fruit yield. Estimated values recorded for stem diameter, number of leaves per plant, chlorophyll content and number of primary branches per plant in this work was higher under field conditions than those of the greenhouse. Contrast to the present findings, Rajasekar et al. (2013) found that tomato leaves were higher in number under greenhouse conditions compared with that of the field conditions. Similarly, Atnafua and Rao (2014) recorded higher values for number of primary branches per plant under greenhouse conditions than that of the field conditions. Differences in observations made could be attributed to differences in the genotypes evaluated, surrounding environment and seasons of the study. Vicente et al. (2011) indicated that chlorophyll content is higher at the vegetative growth stage than at the reproductive growth stage. In this study, measurement of chlorophyll content was carried out at the reproductive growth stage of the plants and thus probably resulted to lower values recorded. University of Ghana http://ugspace.ug.edu.gh 106 5.1.2 Reproductive and phenological characteristics of tomato genotypes The growth and productivity of the tomato crop is significantly influenced by its reproductive or phenological characters. Understanding these economically important characters is essential in planning a crop improvement programme because genotypes with short reproductive or phenological cycle along with high fruit yield are normally selected for commercial cultivation (Singh et al., 2014). The results of the present study indicated that greenhouse-produced tomato genotypes were earlier in attaining 50 % flowering as well as number of days to fruit maturity. Higher values for percentage fruit set and number of fruits per truss were also recorded under greenhouse conditions. However, the field-produced genotypes were early in attaining 1st flower opening and fruit set. Also, number of truss per plant as well as number of flowers per truss was higher under field conditions. The significance of genotype, location and Genotype x Location interaction for all the reproductive and phenological characters implied that variability existed among the genotypes evaluated in each location. According to Singh et al. (2014) both genotypic and environmental factors influence tomato plants to flower early or delay in flowering. Generally, the performances of the genotypes for reproductive and phenological traits were inconsistent across the experimental locations due to the interaction between genotype and the environment. Number of days to 1st flowering varied from 15 to 20 days and 12 to 24 days after transplanting under greenhouse and field conditions respectively. The presence of variability among tomato genotypes for number of days to 1st flowering has been reported in earlier studies. Earlier report by Parvej et al. (2010) indicated that phenological traits such as number of days to 1st flowering, 1st fruit set and number of days to fruit maturity were earlier under greenhouse conditions as compared with results obtained under field conditions. University of Ghana http://ugspace.ug.edu.gh 107 Amarananjundeshwara et al. (2008) recorded variation in number of days to 1st flowering which ranged from 25 to 30 days after transplanting. Percentage fruit set is an important character for increasing the productivity of tomato in terms of fruit yield. It serves as a good indicator of the resistance or tolerance of a genotype to the surrounding environmental conditions particularly temperature (Singh et al., 2014). According to Singh et al. (2014), higher percentage fruit set is required to achieve high fruit yield. High percentage fruit set was recorded under greenhouse conditions than that of the field and varied greatly among the experimental locations. Genotypes such as PLATINUM F1 (72.93 %) and NIRVANA F1 (47.76) recorded maximum and minimum values for fruit set percentage across locations. Earlier investigations reported minimum and maximum values of 50.33% to 84.09 % (Singh, 2014), 72.84 % to 86.21 % (Sharma et al., 2011) and 83.1 to 93.9 % (Pandey et al., 2006). Differences recorded by different authors may be due to differences in the genetic materials evaluated. Genotype PLATINUM F1, CHERRY, SUMO F1 and NKANSAH HT showed a better mean performance for this trait. They are therefore considered as superior heat resistant or tolerant genotypes. Number of days to fruit maturity was higher under greenhouse conditions as compared with that of the field. The genotypes showed high variability for the trait in each location and across locations. Similar observations were previously reported by Nagalakshani et al. (2001) and Cheema et al. (2004). Genotypes NKANSAH HT, WOSOWOSO, COBBRA F1 and THORGAL F1 showed early fruit maturity across locations and represent good genetic materials for increasing the productivity of tomato. Reproductive and phenological characters are influenced by both cultivar and environmental conditions. Tomato genotypes evaluated in this experiment showed variation in their reproductive development characters within and across locations as a result of genotypic differences. The results of the present study indicated that number of truss per plant and University of Ghana http://ugspace.ug.edu.gh 108 number of flowers per truss were higher under field conditions while number of fruits per truss was higher under greenhouse conditions. A study by Parvej et al. (2010) revealed that reproductive development traits including number of trusses per plant, number of fruits per truss and number of flowers per truss were higher under greenhouse conditions than that of the field. Such differences may result from differences in surrounding environment as well as differential response of different genotypes to different environments. Singh et al. (2014) observed that higher number of flower clusters contributed to increased fruit yield in tomato and attributed this observation to inherent genetic factors. Variability in reproductive development characters has been reported in previous studies and ranged from 6.9 to 8.6 for number of truss per plant (Osekita and Ademiluyi, 2014), 2.9 to 7.8 for number of flowers per truss (Emami et al., 2013) and 1.72 to 5.11 for number of fruits per truss. 4.1.3 Yield and yield components characteristics of tomato genotypes Fruit yield is an important trait in genetic improvement of tomato. It is a complex trait and thus influenced by several yield components. Considering the average performances of all genotypes under greenhouse conditions, a total of 7 genotypes (MONGALF1, PLATINUM F1, NKANSAH HT, COBBRA F1, WOSOWOSO, SUMO F1, and NS 504) showed superior performance for fruit yield. These genotypes are good materials for tomato production under greenhouse experimental conditions. Similarly, 7 tomato genotypes namely MONGAL F1, PLATINUM F1, NKANSAH HT, WOSOWOSO, ROMA, ROMA VF and BUFFALO exhibited better fruit yield performance among all the genotypes evaluated under open field conditions for fruit yield. The mean fruit yield of these genotypes exceeded their respective location averages. Across the two locations, four tomato genotypes showed superior performance. These genotypes were MONGAL F1, PLATINUM F1, NKANSAH HT and WOSOWOSO. They represent good materials for tomato production irrespective of differences in locations considered in this experiment. Common among most of the University of Ghana http://ugspace.ug.edu.gh 109 genotypes with superior performance for fruit yield was high scores for one or more yield component traits including number of fruits per plant, number of trusses per plant, single fruit weight and total fruit weight per plant. The tomato genotypes generally showed diverse yield performances which agreed with findings by other workers (Hussain et al., 2001; Mansour et al., 2009; Singh et al., 2009). This was as a result of differences in number and type of genetic materials used, location as well as growing environment. A narrow difference in fruit yield was recorded between the genotypes produced under greenhouse and that produced under field conditions. Some genotypes including PECTOMECH, L11, NIRVANNA F1, #20880 and SHAKTIMAN performed poorly under greenhouse conditions. These genotypes showed poor adaptation to greenhouse conditions. Thus a lower average value of 52.60 t/ha under greenhouse conditions than that of the field (55.10 t/ha) was recorded. Average tomato fruit yield as reported by Atnafua (2014) was higher under polyhouse condition (80.6 t/ha) than that of the open field (57.2 t/ha). Parvej (2010) compared phenological development and productive ability of tomato and obtained high fruit yield under polyhouse condition (81 t/ha) than that of the field (57 t/ha). Also, Blay et al. (1999) evaluated 8 tomato genotypes under field conditions and reported low fruit yield (6 117.30 – 11 359.2 kg/ha). This was attributed to poor rainfall, high day and night temperatures in the experimental period, as well as attack by root knot nematodes. Other reports on tomato fruit yield differences among cultivars include that of Firas et al. (2012) who reported minimum and maximum yield of 15 907 kg/ha and 42 908 kg/ha respectively. Singh et al. (2014) recorded average fruit yield of tomato in a range of 135.10 – 1046.80 q/ha. According to this author, genotypes with medium and large numbers of fruits per plant produced more fruit yield as compared with those with large fruit sizes but smaller number of fruits per plant. This observation agrees with the present study where University of Ghana http://ugspace.ug.edu.gh 110 genotype NKANSAH HT with medium sized fruits and higher number of fruits per plant (27, 33) produced high fruit yield per plant. 5.4 Fruit physical and quality characteristics of tomato genotypes The quality of fruits and vegetables including tomato is determined by their chemical compositions. While inherent genetic factors play major role in determining fruit quality components, environment could alter the quantitative composition of the fruit (Mpofu et al., 2006). For the fresh consumption of tomato as well as in the tomato processing industries, desirable tomato fruit physical and quality features include fruit pericarp thickness, fruit firmness, total soluble solids contents, acidity, flavour, pH, phenolic acids and flavonoid compounds (Caliman et al., 2010). Fruit firmness Tomato fruit firmness has been reported as one of the most pertinent qualities required for processing in the canning industry. It determines fruit shelf life as well as susceptibility of tomatoes to damages associated with harvest, sorting, processing, packaging and transportability (Radzevicius et al., 2013). High fruit pericarp thickness corresponds with better fruit firmness and consequently, its suitability for processing (Saimbhi et al., 2001). Genotype factor is known to be the major determinant of tomato fruit firmness. In the present study, fruits obtained from greenhouse-produced genotypes were found to be relatively firmer (3.90 kg/cm2) than fruits harvested from field-produced genotypes (3.18 kg/cm2). Comparatively, genotypes 11 – 172, ROMA, NKANSAH HT, PECTOMECH, SUMO F1, and INLAY F1 were superior in fruit firmness in the present study. Total soluble solids content Between fifty and sixty-five percent (50 - 65 %) of total soluble solids content (TSS) of tomato is made up of mainly sugars (glucose and fructose) which are essential quality parameters of tomato in the processing industry. Higher amounts of TSS are correlated with University of Ghana http://ugspace.ug.edu.gh 111 higher recovery product yield (Manashi, 2011; Emami et al., 2013; Singh et al., 2014). In terms of processing, a total soluble solids content range of 4 to 8 0Brix has been reported to be desirable. The present study showed a wide variability among the genotypes for TSS content and ranged from 2.26 0Brix (Roma) to 6.58 0Brix (ROMA VF) under greenhouse and 2.58 0Brix (SHAKTIMAN) to 6.98 (WOSOWOSO) under field conditions. A related range of TSS content have been reported in previous studies by Blay et al. (1999), Caliman et al. (2010), Emami et al. (2013) and Singh et al. (2014). In contrast to findings by Caliman et al. (2010), greenhouse-produced genotypes in this study recorded higher TSS values than the field-produced genotypes. Out of the 20 tomato genotypes evaluated, 11 genotypes under green house conditions as well as 8 genotypes under field conditions recorded TSS values in the range of 4 – 8 0Brix indicating their suitability for the processing industry. Fruit acidity content Acidity content of tomato fruits represents another important quality determinant trait of tomato for processing. In the processing industry, acidity reduces pH of the pulp and prevents the growth of microbes. Besides, high acidity gives better tomato flavour, improves its palatability and nutritive value of the fruit as well as reduction in processing time and temperature of the product (Manashi, 2011). Wide variations among genotypes for acidity content have been reported in tomato by previous authors (Manashi, 2011; Aoun et al. 2013; Singh et al., 2014) and this have been mainly attributed to genetic factors. Caliman et al (2010) reported that tomato genotype with acidity value greater than 0.32 % is suitable for processing. The results of the current study revealed 10 and 11 tomato genotypes with acidity content greater than 0.30 mg/ml (0.32 %) under greenhouse and field conditions respectively. Values obtained in this study were in agreement with findings by Caliman et al. (2010) who recorded percentage acidity values of 0.26 – 0.37 under protected conditions and 0.28 – 0.40 under field conditions . From the results of the present work, tomato genotypes produced University of Ghana http://ugspace.ug.edu.gh 112 under field conditions were more acidic (0.40 mg/100 ml) compared with the corresponding genotypes evaluated under greenhouse (0.32 mg/100 ml) conditions. The low acidity content under greenhouse condition could be attributed to low photosynthetic activity of the genotypes produced under greenhouse conditions as compared with that produced under field conditions (Caliman et al., 2010). This is not unexpected as field-produced genotypes recorded higher mean chlorophyll content (40.30 mg/ml) than the greenhouse genotypes (31.12 mg/ml). Fruit flavour index (TSS/TA) Tomato fruit flavour influences consumers or processors’ preference for specific tomato cultivars. It is determined by TSS/TA ratio which is also a taste determinant quality trait. Variation in tomato flavour composition among different genotypes has been attributed mainly to varietal differences and environmental factors such as plant nutrition (Caliman et al, 2010; Manashi, 2011; Aoun et al 2013). Tomato genotype with a TSS/TA value > 10 is reported to have good flavour required for fresh fruit consumption and processing (Caliman et al. (2010). Among the 20 tomato genotypes evaluated, superior flavour content (>10) was recorded by almost all the tomato genotypes under greenhouse conditions except PECTOMECH (9.61). More than half of the tomato genotypes evaluated under field conditions recorded poor flavour composition of < 10 TSS/TA value. This finding suggests that better fruit flavour composition could be obtained under greenhouse environment compared with that of field conditions. The present results contradicted findings by Caliman et al. (2010) who recorded an average TSS/TA value of 12.07 and 16.82 for greenhouse and field conditions respectively. Differences in flavour indexes observed by different authors could be due to differences in the experimental materials used as well as the surrounding environment. Fruit pH Fruit pH along with TSS is desirable indexes for the assessment of related quality traits of tomato in the processing industry (Agong et al., 2001). Previous study by Aoun et al. (2013) revealed that fruit acidity affects flavour only when pH value is lower. At pH value less than 4.1, growth of pathogens like Bacillus coagulans in tomato products are reduced (Majid, University of Ghana http://ugspace.ug.edu.gh 113 2007), while values less than 4. 49 are suitable in terms of fruit sourness (Aoun et al., 2013). In the processing industry, a suitable tomato genotype for processing should have pH value lower than 4.5 (Caliman et al., 2010). Blay et al. (1999) and Singh (2014) reported relatively smaller variability for pH among different genotypes. Findings from the present study showed that pH value differed among the genotypes with appreciably higher values recorded by field-produced genotypes (4.36) than those produced under greenhouse conditions (4.30). This observation was in accordance with findings by Caliman et al. (2010) who recorded a relatively greater pH value under field conditions (4.49) than that obtained under protected conditions (4.48). Fruit pH values obtained in this study revealed that most of the tomato genotypes recorded values that make them suitable for processing. Carotenoids content of fruits In the plant kingdom, beta carotene and lycopene are essential phytochemical compounds. Manashi (2011) mentioned bright colours of fruits (either yellow or red) are good quality characteristic that influence consumers’ choice for tomato. Such colours are affected by the composition of β-carotene and lycopene in the fruit. Composition of carotenoids in fruits varies greatly among different genotypes as well as environments (Frusciante, 2000). Various reports have indicated the effects of light intensity and high temperature in stimulating carotenoids biosynthesis. Variation was observed among the genotypes for lycopene concentration of fruits and ranged from 7.31 mg/100 ml (ROMA VF) to 33.71 mg/100 ml (SHAKTIMAN) among the greenhouse-produced genotypes and 13.31 mg/100 ml (PECTOMECH) to 33.79 mg/100 ml (11-172) among the field-produced tomato genotypes. Generally, lycopene content under field conditions (23.35 mg/100 ml) was appreciably higher than the amount recorded by the corresponding greenhouse genotypes (19.89 mg/100 ml). A different observation was made by Caliman et al. (2010) who obtained higher average lycopene content under protected conditions (7.81 mg/100 ml) than field produced genotypes (7.47 mg/100 ml). Manashi et al. (2011) recorded a narrow difference in carotene content University of Ghana http://ugspace.ug.edu.gh 114 when 5 tomato genotypes were evaluated under field conditions. Previous findings have also revealed an appreciable level of variation among different genotypes for lycopene content under different growing conditions. For instance, Frusciante (2000) reported lycopene content in the range for 5.5 – 7.5 mg/100 g whereas Radzevicius (2013) found 2.02 – 4.95 range of lycopene. These differences in observed values could be attributed to differences in genotypic and prevalent environmental factors such as temperature and light intensity. Ascorbic acid content of fruit A wide variability in ascorbic acid composition was recorded among tomato genotypes as a result of both genetic and surrounding environmental factors. Tomato genotypes that contain high ascorbic acid content are preferred for processing and determine the nutritional quality of the processed product (Singh et al., 2014). The results of the present study indicated that greenhouse-produced tomato genotypes are richer in ascorbic acid content than the field- produced ones. The composition of ascorbic acid was found to be variable within each location as well as across locations. Tomato varieties that contain high ascorbic acid content are suitable for consumption and thus establish the nutritional worth of the variety (Singh et al., 2014). Variability in ascorbic acid composition has also been recorded in previous studies by Singh et al. (2014) (12.6 – 15.63 mg/100 g), Sharma et al. (2011) (17.29 – 26.21 mg.100 g) and Vinod et al. (2013) (21.63 – 26.70 mg/100 g). The small fruited CHERRY variety recorded the highest ascorbic acid composition under greenhouse condition. This is in agreement with findings by Vinod et al. (2013) who indicated that tomato genotypes with small fruits contain high ascorbic acid content. Comparatively, the highest ascorbic acid composition across the two locations was recorded by CHERRY, NKANSAH HT, 11-172, #20880 and PLATINUM F1. University of Ghana http://ugspace.ug.edu.gh 115 Phenolic acid compounds: Phenolic acid compounds including gallic, vanillic and rosmarinic acids have chemo- protective or antioxidant property and thus are able to scavenge effects associated with reactive oxygen species in the human body. In this way, several human-related chronic and degenerative diseases are prevented. The composition of these compounds in the tomato fruit is not only influenced by cultivar factors but also the cultivation practice, surrounding environment, extraction procedure, temperature, processing as well as storage conditions (Supathra et al., 2013). For all the phenolic acid compounds, field-produced genotypes recorded maximum amount as compared with those produced under greenhouse conditions. Brezeanu et al. (2013) observed a significant amount of rosmarinic acid content in tomato fruits among tomato genotypes evaluated under conventional system. The results of the present study revealed five (5) tomato genotypes as comparatively better source of phenolic acid compounds. These genotypes included PLATINUM F1, 11-172, BUFFALO, SUMO F1 and SHAKTIMAN. They contain a relatively higher amounts of gallic, vanillic and rosmarinic acids composition across locations. Flavonoid compounds Quercetin, kaempferol and myricetin are reported to be the main flavonoid compounds present in tomato and its related products (Tokusoglu et al., 2003). In addition, tomato fruits also contain rutin, catechin, hesperetin and genestein as important flavonoid compounds. The concentration of these compounds in fruits varies among genotypes as a result of differences in genetic makeup and cultivation environment. In the present experiment greenhouse- produced tomato genotypes recorded high composition for all the flavonoid compounds studied. A wide variability within each location and across locations was observed among the genotypes for all the flavonoid compounds studied and this could be attributed to both genotype and environment factors. Generally, genotypes WOSOWSO, SUMO F1, NKANSAH HT, 11-172 and CHERRY displayed superior performance for all flavonoid University of Ghana http://ugspace.ug.edu.gh 116 compounds studied across locations. Koh et al. (2009) recorded quercetin and Kaempferol content in commercial broccoli with a range of 0.03 to 10. 85 and 0.24 to 13.20 mg/100 g FW respectively. In this work, quercetin and kaempferol content in tomato fruits ranged from 23.71 to 126 mg/100 ml FW and 6.04 to 57 mg/100 ml respectively across locations. Differences in results could be attributed to differences in plant materials evaluated. 5.3 Antioxidant activity of tomato genotypes Generally, the antioxidant activity of vegetables and fruits differ among different cultivars due to differences in their genotypic makeup and environmental conditions (Marsic et al., 2011). The results of the present study revealed that antioxidant activity of the field-produced genotypes was higher than the same genotypes produced under greenhouse conditions. In each of the two experimental locations, variability for antioxidant activity in terms of their IC50 values was observed. In a related study, Ferreira et al. (2012) recorded the most excellent antioxidant activity in a local tomato variety whose IC50 value was 1.63 mg/100 ml. The results of the present work showed that the local tomato genotype WOSOWOSO proved to be the highest in antioxidant activity under field conditions as well as across locations. In a related study in organic Baby-Leaf salads, Aires et al. (2013) reported antioxidant activity (expressed as IC50) ranged from 0.23 to 3.03 mg/ml. Generally, WOSOWOSO, BUFFALO and NS 504 recorded the lowest IC50 values across the two locations suggesting that their antioxidant activities were comparatively superior to the rest of the genotypes. 5.1.4 Genotype x Location interaction effects on tomato fruit yield Though, the tomato crop is recognized to be adapted to different agro-climatic conditions and cropping systems (Sunil et al., 2013), several studies have indicated a significant G x E interaction effects on the expression of important traits of tomato genotypes. Such interaction effects have been reported in tomato fruit yield as well as several quality characters (Kuti and Konuru, 2005; Rosello et al., 2011; Cebolla-Cornejo et al., 2011 and Panthee, et al., 2012). University of Ghana http://ugspace.ug.edu.gh 117 In the present study, differences in genotype, location and G x L interaction effects were significant (P < 0.01) for almost all the studied characters indicating the role of varietal difference, location and G x L interaction effects on the expression of individual traits. The highly significant value recorded for location suggested that the two locations were diverse. These findings are not unexpected because yield as a complex trait is influenced by a number of yield components which are quantitative in character. Such quantitative characters exhibit continuous variation in their expression and are subject to genotype-by-environment interaction effects (Causse et al, 2002). For instance, the genotype COBBRA F1 performed better with a total fruit yield of 88.55 t/ha relative to its overall mean performance across the two locations and also, it ranked among the top 5 genotypes with superior yield (above location mean) under greenhouse conditions. However it produced 49.78 t/ha (and ranked 15th) under the open field conditions relative to its overall mean across the locations and thus categorized outside the first 5 superior genotypes under field locations. This pattern of performance is inconsistent and a manifestation of reversibility in rank as well as differences in levels of environmental sensitivity across the two locations. This makes COBBRA F1 an unstable and environmentally sensitive for fruit yield. Apart from MONGAL F1, PLATINUM F1 and NKANSAH HT all the other genotypes evaluated exhibited irregular patterns of performance as a result of genotype-by-environment interaction effects. Similar pattern of performance was observed among the genotypes for most fruit quality traits. The present results was in conformity with earlier findings by Panthee et al. (2012) who reported significant effects of genotype-by-environment interaction on all studied traits except ascorbic acid content of fruits. University of Ghana http://ugspace.ug.edu.gh 118 5.3 Genetic Variance, heritability and genetic gain of tomato characters 5.3.1 Individual location basis Genetic variability studies in a crop are fundamental to crop improvement programmes aimed at yield and quality traits. The significant (P < 0.01 and P < 0.05) genotypic difference obtained from the results of the analysis of variance (ANOVA) and the estimate of genetic variance indicated an existence of an amount of variability in the 20 tomato genotypes evaluated. Similar findings have been reported earlier by Sharma et al. (2009) and Dar and Sharma (2011). Estimation of genotypic coefficient of variation (GCV) gives a true suggestion of the magnitude of genetic variation in a studied population. In the present study, moderate (10 – 20 %) to high (> 20 %) estimate of GCV and PCV for most of the agronomic traits under both greenhouse and field conditions showed appreciable amounts of phenotypic and genotypic variability. The relatively higher estimate of PCV than the corresponding GCV indicated a relative effects of environment (to some degree) on the expression of the traits. However, the narrow difference between PCV and GCV and the closeness to unity in values observed in the ratio of GCV to PCV suggest that genetic control in the expression of the traits was predominant. This suggests that simple phenotypic selection could be made among the genotypes. Also, selection could be effective for most characters at the early stage of a breeding program since response to selection is directly proportional to variability present in the experimental materials evaluated (Falconer and Mackay, 1996). Generally, the magnitude of variability observed among the genotypes for agronomic traits was higher under the greenhouse conditions than the open field conditions. Among the quality traits studied, moderate to high estimate of PCV and GCV was observed for most traits under greenhouse and field conditions. This indicated the existence of variability in the genotypes for such traits. However, the low estimate of GCV for pH, gallic acid, rosmarinic acid and vanilic acid suggests that the expression of those traits were University of Ghana http://ugspace.ug.edu.gh 119 influenced by environment. For crop improvement purpose, selection for such characters will need to be carried out in large replicates across multiple location and years (Hallauer, 2007). Besides, the fruit quality traits were more variable compared to the agronomic traits under both locations. The results of the present investigation were in agreement with findings by Nwosu et al. (2014) who recorded low difference between PCV and GCV for most of the characters studied including days to flower, days to 50 % flower, days to fruit ripening, fruit length, fruit per inflorescence, fruit diameter, fruit weight and number of days to fruit maturity. Report by Reddy et al. (2013) indicated an estimate of moderate to high PCV and GCV as well as smaller differences between them for most of the traits they studied. Similar findings have also been reported by other authors (Pradeepkumar et al., 2001; Jiregna et al., 2012; Shankar et al., 2013; Shushay et al., 2013) The estimate of GCV along with heritability provides a true indication of the magnitude of heritable component of variation. Estimate of heritability enables a breeder to determine the extent to which genetic variability contributes to phenotypic variability of a trait. In the present study, broad sense heritability estimated for all traits under greenhouse and field conditions were very high with the exception of ascorbic acid and rosmarinic acid contents which recorded moderate values. High GCV as well as high heritability are useful to breeders to make effective selection and also indicate that selection could be made on phenotypic basis. However, this does not necessarily suggest high genetic gain for a particular character unless it is associated with high genetic advance. In this study, very high heritability estimates accompanied with high genetic gain was recorded by most of the traits evaluated under each of the locations. These findings suggest that those traits could easily be passed on to the next generation. High estimate of heritability may be the result of the diverse nature of the genotypes included in the study (Shushay et al., 2013). Phenotypic selection of such traits in the early generation University of Ghana http://ugspace.ug.edu.gh 120 for further improvement could therefore be accomplished via simple selection methods like pure line, mass selection, bulk or single seed decent. High estimate of broad sense heritability (61 % to 100 %) along with high genetic advance was recorded for almost all agronomic and physicochemical traits of tomato varieties studied by Shankar et al. (2013). However, in the present study high heritability and moderate genetic gain was recorded for days to 50 % flowering, number of flowers per cluster and TSS. High and medium heritability estimate along with low genetic gain recorded for percentage fruit set and pH respectively under open field conditions suggested that non-additive gene action controlled their expression; hence selection cannot be achieved through simple selection methods. Instead such traits could be improved through development of hybrids or could possibly be used as transgressive segregants in heterosis breeding programme (Reddy et al., 2013). The results of the present study is in accordance with findings made by Shushay et al. (2013), Hadhayatullah et al. (2008) and Mehta and Asati (2008) who recorded very high (P > 80 %) estimates of heritability along with high genetic gain for most of the traits studied. A related study by Vinod et al. (2010) indicated a very high estimate of heritability for all traits with a range from 80 % to 99 %. Several related findings have been reported in tomato genetic variability and heritability studies (Kaushik et al., 2011; Reddy et al., 2013). 5.3.2 Combined location basis On the basis of the combined analysis of variance, the observed variation recorded for almost all characters studied was significant (P < 0.01 and P < 0.05) among the genotypes. This observation could be attributed to environment or location or genotype-specific differences as well as G x L interaction effects. The significant (P < 0.01) mean squares of location and G x L interaction suggest that the locations were diverse. This resulted in differences in response of genotypes to location changes as manifested in their varied performances across different locations. The challenge then is the need to develop separate varieties tailored to each University of Ghana http://ugspace.ug.edu.gh 121 specific location. Therefore, one of the most important rationales behind genetic variability studies is the identification of suitable environments where a trait which is intended to be improved upon is most likely to show a consistent expression. Environmental factors may influence the genotype and consequently, the expression of a trait in a cultivar evident by the magnitude of variability due to genotype-by-environment interaction (G x E). Genotype-by-environment interaction (G x E) causes a reduction in the overall genetic gain of desired traits. This phenomenon may create difficulties in selection of desirable traits for crop improvement programmes. Knowledge about G x E interaction is therefore required in the selection of ideal location, traits and the genotype for inclusion in a breeding programme (Gruneberg et al, 2005). The results in the present investigation based on the combined analysis of variance indicated that the estimated variance due to location (σ2L) was higher than the corresponding genotypic variance (σ2G) for some agronomic traits including plant height, stem diameter, chlorophyll content, root length per plant and number of days to 50 % flowering confirming the significance of location effects on those traits. Among the fruit quality parameters, larger estimates of genotypic variance (σ2G) than variance due to location (σ2L) was found for most fruit physical traits including single fruit weight, fruit weight per plant, fruit yield per plant, number of locules, fruit length and diameter as well as shape index. As a result high estimate of heritability was recorded for such parameters indicating that they were predominantly affected by inherent genetic factors. It should be noted that very large variance estimates due to phenotypic, environmental and G x L as well as phenotypic coefficient of variation than the corresponding genotypic parameter were observed for almost all the fruit chemical traits. This resulted in negative GCV and heritability estimates. Among the traits studied, the value obtained for the ratio of σ2G to σ 2 G x L (σ 2 G:σ 2 G x L) ratio was close to unity only for number of primary branches per plant, number of days to first fruit University of Ghana http://ugspace.ug.edu.gh 122 set, number of flowers per truss and fruit shape index. This implied that the effects of location or environment as well as G x L interaction effects accounted for variability in most of the traits. This observation gives an indication that differential response of genotypes to location difference was very possible. Highly significant genotype-by-location interaction implied that independent analysis of data for each location is required (Khan et al., 2013) rather than analysis carried out on combined location basis. The present observation agrees with earlier findings by Causse et al. (2003) who recorded a significant G x E interaction for lycopene, TSS and TA. Information regarding the nature of a trait, its expression as well as interaction with environment could best be understood from the estimate of broad sense heritability. The results of the present study revealed that almost all fruit quality traits recorded very low heritability estimates. This implied that most of the quality traits were environmentally sensitive and the ability to control and predict variability based on genetic information alone becomes negligible (Panthee, 2012). This also implied that selection of superior genotypes for most of the traits would not be effective at early generation. In contrast to the present findings, high heritability estimate has been reported by Premah et al. (2011) for TSS (77.95), ascorbic acid (74.7 %), and TA (99.94 %). Similarly, Dar and Sharma, (2011) reported high heritability estimates for lycopene (92 %) and ascorbic acid (94 %). These contrasting results could be due to differences in the experimental materials, growing conditions as well as cultural practices. 5.4 Character association of traits among tomato genotypes In selection programs, yield, yield components and quality traits are among the most economically essential traits usually targeted by plant breeders. According to Falconer and Mackay (1996), many important characters in crops are positively or negatively correlated because they are influenced by the same gene or because they are developmentally or University of Ghana http://ugspace.ug.edu.gh 123 structurally related. Knowledge of correlated traits is essential in determining whether or not selection for a particular trait will influence another (Girdthai et al., 2012). In the present study, number of trusses per plant, number of fruits per plant as well as total fruit weight per plant showed a significant and positive association with fruit yield irrespective of the crops growing condition considered in this experiment. This shows a consistency of correlation across the two locations. However the correlation coefficients differed due to G x L interaction effect. Indirect selection gives a correlated response in the target trait so far as the targeted trait and the secondary traits are associated (Falconer and Mackey, 1996). Those traits with significantly positive association with fruit yield could therefore be selected for maximizing fruit yield in tomato in any of the two growing conditions. This result was in agreement with findings by previous authors including Hadhayatullah et al. (2008), Gosh et al. (2010), Jiregna et al. (2012) and Shushay et al. (2013). These authors independently reported a positive and significant correlation of fruit yield with number of fruit clusters per plant and number of fruits per plant at both the genotypic and /or phenotypic levels. Singh (1993) indicated that a high fruit yield response is achieved when the secondary trait also has a high estimate of heritability as well as high correlation coefficient. The present study revealed high heritability estimates as well as high correlation coefficient for the traits associated with fruit yield. Negative correlation of fruit yield with number of days to 50 % fruit set (under both greenhouse and open field condition) and number of days to fruit maturity (greenhouse condition only) indicated that genotypes that attain early fruit set and mature early tend to have higher fruit yield. This finding agrees with results reported by Jiregna et al. (2012) and Shushay et al. (2013). A positive and significant correlation was observed between fruit yield and gallic acid (a phenolic and antioxidant phytochemical compound) under the field experimental location. University of Ghana http://ugspace.ug.edu.gh 124 This indicated that indirect selection of one trait for the other could be feasible under field conditions. Moreover, the significantly positive association between most traits suggests that each of those pairs of traits is controlled by the same or similar genes or is developmentally or structurally related (Falconer and Mackay, 1996). Indirect selection of one trait for the other should then be appropriate. This is in agreement with earlier report where average fruit weight correlated positively and significantly with number of primary branches per plant at both genotypic and phenotypic levels (Vinod et al., 2013), number of days to 50 % fruit set and number of days to fruit maturity (Shushay et al., 2013). On the other hand, negative association observed for trait pairs point out that indirect selection for any one of those traits may possibly not be advantageous. Antioxidant compounds considered in this discussion included lycopene, ascorbic acid, gallic acid and quercetin. Under greenhouse growing environment, ascorbic acid correlated positively and significantly with number of primary branches per plant, number of trusses per plant and number of fruits per plant suggesting that those pairs of traits are controlled by the same gene. Similar observations were made earlier by Vinod et al. (2013). Indirect selection for any one of those traits is therefore possible. Genotypes with superior performance for traits positively correlated with ascorbic acid could be used for further development of materials endowed with high ascorbic acid content. A negative association between ascorbic acid content and fruit length also suggests that genotypes with shorter fruit length contain higher amount of ascorbic acid content and hence could be used as indirect selection under green house conditions. Moreover, the antioxidant compound quercetin showed a significant and positive correlation with titrable acidity and ascorbic acid. This observation indicated that indirect selection of the trait titrable acidity or ascorbic acid could contribute to enhancing quercetin content in tomato fruit over combined locations. Under the combined locations genotypes containing University of Ghana http://ugspace.ug.edu.gh 125 high ascorbic acid content and titrable acidity content could be used to develop high quercetin content cultivars. Quercetin showed a negative significant correlation with fruit length and fruit diameter under greenhouse conditions suggesting that indirect selection for these traits to increase gallic acid content would be less useful. Thus genotypes with reduced fruit length or reduced diameter are possibly better in terms of their gallic acid composition. Also, under field growing conditions a positive significant association of quercetin with single fruit weight per plant, total soluble solids content and titrable acidity was observed. These results revealed that the traits are under the control of the same or similar genes. Indirect selection of those traits could therefore maximize quercetin content in tomato fruits under field conditions. However a negative and significant association of quercetin with number of days to 50 % fruit set suggests that it is undesirable to select the later trait for maximizing the amount of quercetin in the fruit. Antioxidant scavenging activity, expressed as IC50 (inhibition coefficient ) associated positively and significantly only with number of days to 50 % fruit set implying that the two traits are being controlled by the same or similar genes or they are developmentally or structurally correlated. The lower the IC50 value scored for a genotype the better its antioxidant potential to deactivate the harmful effects associated with reactive oxygen species (ROS) in humans (Kipandula et al., 2014). This also implied that a reduction in number of days to 50 % fruit set should show a corresponding reduction in the IC50 value (increased antioxidant activity). Genetic and environmental variation study in bread and Durum wheat by Sukkalovic et al. (2013) indicated a highly significant and positive association of antioxidant capacity with total phenolic composition. The association of antioxidant activity with phenolic compound (gallic acid) studied in the present experiment was not significant. University of Ghana http://ugspace.ug.edu.gh 126 CHAPTER SIX 6.0 CONCLUSIONS AND RECOMMENDATIONS 6.1 Conclusions 1. Systematic breeding programmes aimed at improving tomato fruit yield and fruit quality traits require information on the nature and magnitude of genetic variability, heritability and character association in agronomic and fruit quality traits in the crop. Results of the present study revealed an existence of useful genetic variability in the genotypes. The estimate of moderate to high GCV, high broad sense heritability as well as high genetic gain for almost all traits suggest the potential for selection and utilization of suitable genotypes based on different parameters to obtain high yield and fruit quality traits. 2. Following the present findings, promising tomato genotypes based on their fruit yield (t/ha) under greenhouse conditions were MONGAL F1, PLATINUM F1, NKANSAH HT, COBBRA F1, WOSOWOSO, ROMA, SUMO F1 and NS 504. 3. Tomato genotypes MONGAL F1, PLATINUM F1, NKANSAH HT, COBBRA F1, WOSOWOSO, ROMA, SUMO F1, ROMA VF and BUFFALO showed superior yield performance under field conditions. 4. Across the two locations, MONGAL F1, PLATINUM F1, NKANSAH HT, COBBRA F1, WOSOWOSO, ROMA and SUMO F1were superior genotypes for fruit yield. Most of the top performing genotypes for yield recorded high scores for one or more yield component traits including number of fruits per plant, trusses per plant, single fruit weight and total fruit weight per plant. 5. Highest amount of flavonoid compounds across the two locations were recorded by tomato genotypes WOSOWSO, SUMO F1, NKANSAH HT, 11-172, CHERRY and THORGAL F1. University of Ghana http://ugspace.ug.edu.gh 127 6. The amount of phenolic acids composition was found to be highest in genotypes PLATINUM F1, BUFFALO, 11-172, SHAKTIMAN, THORGAL F1 and NIRVANA F1. 7. Genotypes produced under greenhouse conditions recorded higher amount of flavonoid composition while phenolic acids content was higher under field conditions. 8. Highest antioxidant scavenging activity among the genotypes across locations was recorded by WOSOWSO, BUFFALO and NS 504. 9. Generally, the performance of the genotypes evaluated differed across locations due to the interaction between the genotypes and environment. The expressions of most traits were influenced by genetic and environmental factors as well as their interaction. 10. Number of trusses per plant, number of fruits per plant and total fruit weight per plant showed a positive significant association with fruit yield and are potentially useful traits as indirect selection indexes for yield improvement under both greenhouse and field conditions 6.2 Recommendations 1. The results of the current study showed that tomato genotypes MONGAL F1, PLATINUM F1, NKANSAH HT, COBBRA F1, WOSOWOSO, ROMA and SUMO F1are recommended for commercial production under both greenhouse and field conditions by virtue of their superior fruit yield per plant across the two growing environments. 2. The present study was carried out under a single growing season; hence further evaluation of the genotypes could be conducted over multiple seasons or years. Evaluation of the genotypes across multiple locations and years along with appropriate stability analysis procedure will further be required to identify stable University of Ghana http://ugspace.ug.edu.gh 128 tomato genotypes for fruit yield and other important chemical composition traits. This will also provide an in depth information regarding G x E interaction and tomato fruit yields. University of Ghana http://ugspace.ug.edu.gh 129 REFERENCES Abdul-Fatawu, A. (2013). Effects of stage of ripening and month of harvest on the properties of Citrus sinensis var late Valencia stored at ambient conditions. Mphil. Thesis, KNUST. Abdul-Hammed, M., Bello, L. A., and Olajire, A. A. (2009). Comparison of the biochemical and physiological properties of Nigerian tomato fruits ripened under different conditions. African Journal of Food, Agriculture, Nutrition and Development, 9 (9): 1859-1877. Albrecht, E., Escobar, M. and Roger T. C. (2010). “Genetic diversity and population structure in the tomato-like nightshades solanum lycopersicoides and S. sitiens.” Annals of Botany 105 (4): 535-554. Adalid, A. M., Rosello, S. and Nuez, F. (2010). Evaluation and selection of tomato accessions (Solanum lycopersicum) for content of lycopene, β-carotene and ascorbic acid. J Food Compos Anal 23(6): 613-618. Agong, S. G., Scittenhelm and S. Friedt (2001). Genotypic variation of Kenyan tomato (Lycopersicon esculentum L.) Germplasm. The Journal of Food Technology in Africa, vol. 6, No. 1, 13-17. Aires, A., Marques, E., Carvalho, R., Rosa, E. A. S. and Saavedra M. J. (2013). Evaluation of biological value and appraisal of polyphenols and glucosinolates from organic baby-leaf salads as antioxidants and antimicrobials against important human pathogenic bacteria. Molecules: a Journal of Synthetic Chemistry and Natural Product Chemistry, 18(4): 4651-4668 University of Ghana http://ugspace.ug.edu.gh 130 Ajmal, S. U., Zakir, N. and Mujahid, M. Y. (2009). Estimation of genetic parameters and character association in wheat. J. Agric. Biol. Sci., 1: 15-18. Akinfasoye, J. A., Dotun, J. O., and Ajayi, E. O (2011). Phenotypic relationship among Agronomic Characters of Commercial Tomato (Lycopersicum esculentum) Hybrids. American Eurasian Journal of Agronomy, 4 (1): 17-22. Amarananjundeshwara, H., Shyyamalamma, S., Chikkasubbanna, V., and Ajayakumar, M. Y., (2008). Performance of tomato hybrid under greenhouse conditions. Bangladesh J. Agric. Res., 35: 367 – 373. Anshuman, V., Dixit, N. N., Dipika, Sharma, S. K. and Marker, S (2013). Studies on heritability and genetic advance estimates in maize genotypes. Bioscience Discovery, 4(2): 165-168. Aoun, A. B., Lechiheb, B., Benyahya, L. and Ferchichi, A. (2013). Evaluation of fruit quality traits of traditional varieties of tomato (Solanum lycopersicum) grown in Tunisia. African Journal of Food Science, vol. 7(10), pp. 350-354. Atnafua, B. and Rao, T. N. (2014). Estimate of heritability, genetic advance and correlation study for yield and its attributes in maiz (Zea mays L.). Journal of Plant Sciences, 2014; 2 (1): 1-4. Bachir, B.M., Louaileche, H. and Mouhoubi, Z. (2014). Antioxidant activity of eight tomato (Solanum lycopersicum L.) varieties grown in Algeria. Journal of Food Technology Research, 1(3): 133-145. Bai, Y. and Lindhout, P. (2007). Domestication and breeding of tomatoes: What have we gained and what can we gain in the future? Ann. Bot., 100: 1085-1094. University of Ghana http://ugspace.ug.edu.gh 131 Barros, L., Duenas, M., Pinelam, J., Carvalho, A. M., Buelga, C. S. and Ferreira, I. C. F. R. (2012). characterization and quantification of phenolic compounds in four tomatoes (Lycopersicon esculentum l.) farmers’ varieties in northeastern portugal home gardens. Plant Food for Human Nutrition, Vol. 67 (3), 229-234. Bello, O., Ige, S. A., Afolabi, M. S., Abdulmaliq, S. Y. and Mahmood, J. (2012). Heritability and Genetic Advance for Grain Yield and its Component Characters in maize (Zea mays L.). International Journal of Plant Research, 2(5): 138 – 145. Blay, E. T., Danquah, E. Y., Offei, S. K., and Kudadjie (1999). Morphological and Agronomic Characterization of Some Tomato (Solanum lycopersicum L.) Germplasm in Ghana. Ghana Jounal of Agric. Science. Vol 32, No. 2, 169-175. Borguinni, R. G. and Da Silva Torres, E. A. F. (2009). Evaluation and selection of tomato products as dietary sources of antioxidants. Food Rev Int 25 (4): 313-125 Brezeanu, P. M., Munteanu, N., Brezeanu, C., Ambarus, S., Draghici, E., Calin, M. and Cristea, T. O. (2013). Antioxidant activity in selected tomato genotypes cultivated in conventional and organic culture systems. African Journal of Biotechnology, vol. 12(20), pp. 2884 – 2899. Caliman, F. R.B., Silva, D. J., Stringheta, P.C., Fontes, P. C. R., Moreira, G. R.., and Momtavani, E. C. (2010). Quality of tomatoes grown under protected Environment and field conditions. ADESIA (Chile) Mayoo-Agosto, Volume 28, No. 2, Paginas 75-82. Capanoglu, E., Beekwilder, J., Boyacioglu, D., De Vos, R. C.H., Hall, R.D. (2010). The effects of industrial food processing on potentially health-beneficial tomato antioxidants. Crit Rev Food Sci Nutr 50:919-930. University of Ghana http://ugspace.ug.edu.gh 132 Carli, P., Arima, S., Fogliano, V., Tardella, L., Frusciante, L., and Ercolano, M. (2009). Use of network analysis to capture key traits affecting tomato organoleptic quality. Journal of Experimental Botany, vol. 60, NO. 12, pp. 3379-3386. Causse, M., Saliba-Colombani, V., Lecomte, L., Duffe, P., Rousselle,P. and Buret, M. (2002). QTL analysis of fruit quality in fresh market tomato: a few chromosome regions control the variation of sensory and instrumental traits. Journal of experimental Botany, Vol. 53, No. 377, Fruit Development and Ripening Special Issue, pp. 2089-2098. Causse, M., Buret, M., Robini, K and Verschave, P. (2003). Inheritance of Nutritional and Sensory Quality Traits in Fresh Market Tomato and Relation to Consumer Preference. Journal of Food Science, Vol. 68, Nr. 7. Cebolla-Cornejo, J., Rosello, S., Valcarcel, M., Serrano, E., Beltran, J. and Nuez, F. (2011). Evaluation of genotype and environment effects on taste and aroma flavor components of Spanish fresh tomato varieties. J. Agr. Food Chem. 59:2440-2450. Chattopadhyay, A. Chakraborty, I. V. I. and Siddique, W. (2013). Characterization Determinate Tomato Hybrids: Search for Better Processing Qualities. Journal of Food ProcessingTechnol. http//dx.doi.org/10.4172/2157-7110.1000222. Cheema, D. S., Kaur, P. and Kaur, S. (2004). Off-season cultivation of tomato under net house conditions. Acta Horticulture (ISHS), 659, PP: 177-181. Dar, R. A., and Sharma, J. P. (2011). Genetic variability studies of yield and quality traits in tomato (Solanum lycopersicum L.). Int. J. Plant Breed. Genet. 5(2): 168 – 174 Deery, S. J. (2012). Monitoring rhizosphere microbial community of tomato. PhD Thesis. University of Nottingham. University of Ghana http://ugspace.ug.edu.gh 133 De Lannoy, G. (2001). Vegetables: In crop production in tropical Africa. R. H. Raemaekers (ed), DGIC, Brusses. 475pp. Emami , A., Homauni-Far, M. Razavi, R. and Eivazi, A. R. (2013). Introduction of superior tomato cultivars (Solanum lycopersicum L.). Peak Journal of Food Science and Technology vol. 1(2), pp. 19-26. Falconer, D. S. and Mackay, T. F. C. (1996). Introduction to quantitative genetics, 4th Edition. Pearson Education Limited, Edinburgh Gate, Harlow, England. Pp. 464. FAO (2010). Plant genetic resource for food and agriculture. Rome, Food and Agriculture Organization of the United Nations. FAO/UNESCO (1990). FAO/UNESCO Soil map of the world Generalized from the FAO/UNESCO Soil map of the world (FA), 1971-1981. Proceedings of the 14th International Congress of Soil Science, July 22-27, Kyoto, Japan. Farshadfar, E. and Estehghari, M. R. (2014). Estimation of genetic architecture for agro morphological characters in common wheat. International Journal of Bioscience (IJB), Vol. 5, no. 6, p 140-147. Ferreira I. C. F. R., Pinela, J., Barros, L. and Carvalho, A.M. (2012). Nutritional composition and antioxidant activity of four tomato (Lycopersicon esculentum L.) farmers’ vatieties in Northern Portugal home gardens. Food Chemistry Toxicol. 50 (3- 4):829-34. Firas, A., Hussein, K., Mahmood, S., Abdulla, A and Mohannad, A. A. (2012). Genetic Analysis and Correlation of Yield and Fruit Quality Traits in Tomato (Solanum lycopersicum L.). NY Sci J. 5(10):142-145]. ISSN: 1554-0200). University of Ghana http://ugspace.ug.edu.gh 134 Frusciante, L., Barone, B. Carputo, D. Ercolano, M. R., Della Roca, F. and Esposito, S. (2000). Evaluation and use of plant biodiversity for food and pharmaceutical. Fitoterapia 71, 66 – 72. Girdthai, T. Jogloy, S. Vorasoot, N., Akkasaeng, C., Wongkaew, S., Patanothai, A. and Holbrook, C. C. (2012). Inheritance of the physiological traits for drought resistance under terminal drought conditions and genotypic correlations with agronomic traits in peanut. SABRAO Journal of Breeding and Genetics, 44 (2) 240 – 262. Godia, A. K. G. (2014). Evaluation of Some Introduced Fresh tomato (Solanum lycopersicum L.) for genetic variability and adaptability in Ghana using morphological and molecular markers. M.Sc Thesis, KNUST. Gomez, K. V. and Gomez, A. A. (1984).Statistical procedure for Agricultural Research. 2nd Edn., John Wiley and sons Inc., New York.Pp.357-427. Gosh, K. P., Islam, A. K. M. A., Mian, A. K. and Hussain, M. (2010). Variability and character association in F2 segregating population of different traits in tomato (Solanum lycopersicum L.). J. Appl. Sci. Environ. Manage. 14 (2): 91 -95. Gruneberg, W. J., Manrique, K, Zhang, D. and Hermann, M. (2005). Genotype x environment interactions for a diverse set of sweet potato clones evaluated across varying ecogeographic conditions in Peru. Journal of Crop Science, FAO, Vol. 45, no. 6, pp. 2160-217. Hadhayatullah, S. A., Jatoi, A., Ghafoor, A. and Mahmood, T. (2008). Path coefficient analysis of yield components in tomato (Lycopersicum esculentum).Pak. J. Bot., 40: 627-635. Hallauer, A. R. (2007). History, contribution and future of quantitative genetics in plant breeding: Lessons from maize. Crop Science 47: S5 – S19. University of Ghana http://ugspace.ug.edu.gh 135 Harbone, J. B. (1973). Phytochemical methods. A guide to modern techniques in plant analysis. London, Chapman and Hall, New York. Pp. 49 - 188 Harrigan, G. G., Stork, L. G., Riordan, S., Reynolds, T. L., Ridley, W. P., Masucci, J. D., MacIsaac, S., Halls, S. C., Orth, R., Smith, R. G., Ween L., Brown, W. E., Wesch, M., Riley, R,. MacFarland, D., Pandravada, A. Glenn, K. C. (2007). Impact of genetics and environment on nutritional and metabolic components of maize grain, J. Agric. Food Chem. 55, pp. 6177-6185. Hawkins, W. (n.d.). Determination of fruit acids by titration and calculation of the sugar/acid ratio. www.hawkinswatts.com/documents/OECD Hazra, P., Chattopadhyay, S. (2009). Tomato BCTH 4 for fresh consumption. Indian Horticulture 54:22-23 Hoogendijk, M. and Williams, D. (2001). Characterizing genetic diversity of home garden crop species: some examples from the Americas. In proceedings of the second Int. home Gardens Worship. 17-19. July, 2001. Witzenhansen, Federal Republic of Germany Home gardens and in site conservation of plant genetic resources in farming systems. Eds. J. W. Watson and P. B. Eyzaguirre. Pp 34-40. Hossain, M. E., Alam, M. J., Hakim, M. A., Amanullah, A. S. M., and Ahsanullan, A. S. M. (2010). An assessment of physicochemical properties of some tomato genotypes and Varieties grown at Ragpur. Bangladesh Res. Pub. J. 4(3): 135-243. Hounsome, N., Hounsome, B., Tomos, D., and Edwards-Jones, G. (2008). Plant metabolites and nutritional quality of vegetables. JFS R; Concise Reviews/Hypotheses in Food Science 1 – 18. Hussain, S. I., Khokhar, K. M., Mahmood, T., Laghari, M. H. and Mahmud, M. M. (2001). Yield potential of some exotic and local tomato cultivars grown for summer production. Pak. J. Sci., 4: 1215 – 1216. University of Ghana http://ugspace.ug.edu.gh 136 Ibitoye, D. O., Akin-Idowu, P. E. and Ademoyegu, O. Y. (2009). Agronomic and lycopene evaluation in tomato (Lycopersicon lycopersicum Mill.) as a function of genotype. World Jour. of Agricultural Sciences 5(5): 892-895. Ilic, S. Z., Kapoulas, N. and Sunic, L. (2014). Tomato fruit quality from organic and conventional production. In: Agricultural and Biological Sciences, “organic Agriculture towards sustainability, edited by Vytautas Pilipavivius, Lithunia, InTech Publisher Jaffel, K., Sai, S., Bouraoui, N. K., Ammar, R. B., Legendre, L., Lachaal, M. and Marzouk, B. (2011). Influence of salt stress on growth, lipid peroxidation and antioxidative enzyme activity in borage (Borago officialis L). Plant Biosyst. 145, 362-369. Jagatpati, T., Narayan, C.A., and Mobina, P. (2013). Biometrical evaluation of genotypic variability and genetic advance in amaranth cultivars. The ScieTech, Journal of Science and Technology., vol.2, Issue2, p-26-30. Jallal, A. A., and Ahmad, H. A (2011). Genetic Variability, Heritability, Phenotypic and Genotypic correlation Studies for Yield and Yield Components in Promising Barley Genotypes. Journal of Agricultural Science, Vol. 4, No. 3, p. 193. Jiregna, T., Derbew, B. and Kassahun, B. (2012). Genetic Associations Analysis among Some Traits of Tomato (Lycopersicum esculenttum Mill.) Genotypes in West Showa, Ethiopia. International Journal of Plant Breeding and Genetics 6 (3): 129- 139. Johnson, H. W., Robinson, H. F. and Comstock, R. E. (1955). Estimates of genetic and Environmental variability in soybeans. Agronomy Journal 47:314-318. Kader, A. A. and Rolle, R. S. (2004). The role of post-harvest management in assuring the quality and safety of horticultural produce.FAO Agricultural services bulletin 152. University of Ghana http://ugspace.ug.edu.gh 137 Kaloo, G. and Barnerjee, M.K. (2000). H-24: Moderately leaf curl resistant variety of tomato (Lycopersicon esculentum Mill.). Vegetable Science 27:117-120. Kartz, D. A. (2013). Determination of vitamin C in foods, (unpublished). Kaushik, S. K. Tomar, D. S. and Dixit, A. K. (2011). Genetics of fruit yield and its contributing characters in tomato (Solanum lycopersicum L.). Journal of Agriculture and Sustainable Development. Vol. 3(10), pp. 209-213 Khan, A. A., Alam, M.A., Alam, M. K. and Alam, M. J. (2013). Genotypic and phenotypic correlation and path analysis in durum wheat (Triticum turgidum, L. var. durum). Bangladesh J. Agril. Res. 38(2): 219-225. Kipandula, W. L., Mwanza, B., Nguu, E. and Ogoyi, D. (2014). Antioxidant activities of selected indigenous vegetables from Kenya and Malawi. African Journal of Biotechnology, vol. 13(17), pp.1824-1834. Klem, M. A., Nair, M. G., Strasburg, G. M. and Dewitt, D. L. (2000). Antixidant and cyclooxygenate inhibitory phenolics compounds from Ocimum sanctum Linn. Phytomedicine 7:7-13. Koh, E., Wimalasiri, K. M. S., Chassy, A. W. and Mitchell, A. E. (2009). Content of ascorbic acid, quercetin, kaempferol and total phenolics. Journal of Food Composition and analysis (Impact factor: 1.99), 22(7-8): 637 – 643. Kuti, J. O. and Koruru, H. B. (2005). Effects of genotype and cultivation environment on lycopene content in red-ripe tomatoes. J Sci Food Agric 85(12): 2021-2026. Larry, R. and Joanne, L. (2007). Genetic resources of tomato. In: Rasdan M. K., Mattoo, A. K., editors. Genetic improvement of solanaceous crops.vol. 2, Enfield, N H: Science Publishers. pp 25 – 75. University of Ghana http://ugspace.ug.edu.gh 138 Lecomte, L., Hospital, F., Buret, M., Duffe, P., Servin, B and Causse, M. (2003). Efficiency of marker assisted backcross for mapping tomato fruit quality, a multi- trait characteristic, Molecular breeding, (in press). Lee, S. K. and Kader, A. A. (2000). Preharvest and postharvest factors influencing vitamin C content of horticultural crops. Postharvest Biol. Technol. 37, 817-826. Majid, R. (2007). Genome Mapping and Molecular Breeding of tomato. Int. J. Plant Genomics p. 52. Marsic, N. K., Gasperi, L., Abram, V., Budic, M. and Vidrih, R. (2011). Quality parameters and total phenolic content in tomato fruits regarding cultivar and microclimatic conditions. Turkish Journal of Agriculture and Forestry, 35(2), 185– 194. Martinez-Valverde, I., Periago, M. J., Provan, G. and Chesson, A. (2002). Phenolic compounds, lycopene and antioxidant activity in commercial varieties of tomato (Lycopersicum esculentum). J. Sci. Food Agric. 82, 323-330. Manashi, D. P. (2011). Physicochemical properties of five different tomato cultivars of Meghalaya, India and their suitability in food processing. As. J. Food Ag-Ind. 4903), 187-203. Manju, P. R. and Sreelathakumary, I. (2002). Genetic variability, heritability and genetic advance in hot chilli (Capsicum chinense Jacq.). Journal of Tropical Agriculture 40: 4-6. Mansour, A., Ismail, H. M., Ramadan, M. F. and Gyulai, G. (2009). Variation in tomato (Lycopersicon esculentum) cultivars grown under heat stress. J. verbr. Lebensm. 4: University of Ghana http://ugspace.ug.edu.gh 139 118 – 127. NHB, 2010. Horticulture data 2009 – 2010. National Horticulture Board, India. Maul, F., Sargent, S. A., Sims, C. A., Baldwin, E.A., Balaban, M. O. and Hubr, D. J. (2000). Tomato flavour and aroma quality as affected by storage temperature. Journal of Food Science, 65, 1228-1237. Mehta, N. and Asati, B. S. (2008).Genetic Relationship of Growth and Development Traits with Yield in Tomato (Lycopersicum esculentum Mill.). Karnataka Journal of Agricultural Science, 21, 92-96 Mertenz-Talcott, S. U., Talcott, S. T., Percival, S. S. (2003). Low concentrations of Quercetin and ellagic acid synergistically influence proliferation, cytotoxicity and apoptosis in MOLT-4 Human leukemia cells. J Nutr 133:2669-2674. Mohamed, S. M., Ali, E.E and Mohamed, T. Y.(2012). Study of heritability and genetic variability among different plant and fruit characters of tomato (Solanum lycopersicum L.). International Journal of Scientific and Technology Research, 1 (2):55-58. Mpofu, A., Harry, D., Sapirstein, H. D., Beta, T. (2006). Genotype and environment variation in phenolics content, phenolics acid composition and antioxidant activity of hard spring wheat. J Agric. Food Chem., 54: 1265-1270. Msaada, K., Jemia, M. B., Salem,,N., Bachrouch, O., Sriti, S. T., Battaieb, I. Jabri, I., Kefi, S., Limam, F. and Marzouk, B. (2014). Antioxidant activity of methanolic extracts from three coriander (Coriandrum sativum L.) fruit varieties. Arabian Journal of Chemistry. htt://dx.doi.org/10.1016/j.arabjc.2013.12.011 University of Ghana http://ugspace.ug.edu.gh 140 Mutumpike, M. (2013). Growth response of tomato (Lycopersicum esculentum Mill) to Different Growing Media under Greenhouse and Field Conditions. MSc Thesis, University of Zambia. Nagalakshmi, S. N., Nandakumar, N., Palanisamy, D. and Sreearayann, V. V. (2001). Natural ventilated polyhouse for vegetable cultivation. South Indian Horticulture, 49 (Special), pp: 345 – 346. Nkansah, G. O., Okyere, P., Coffie, R. and Voisard, J. M. (2007). Evaluation of four exportable hybrid okra varieties in three different ecological zones of Ghana. Ghana J. of Horticulture. 6:25-31. Norman, J.C. (1992). Tropical vegetable crops. Elms Court: Pub, Arthur H. Stockwell Ltd. Pp 15-53. Nour, V., Trandafir, I. and Ionica, M. E.(2013). Antioxidant Compounds, Minerals and Antioxidant Activity of several Tomato Cultivars Grown in Southwestern Romania. Not. Bot Horti Agrobo, 41 (1): 136-142. Ntawuruhunga, P. and Dixon, A. G. O. (2010). Quantitative variation and interrelationship between factors influencing cassava yield. Journal of Applied Biosciences 26: 1594 – 1602. Nwosu, D. J., Onakoya, O. A., Okere, A. U., Babatunde, A. O. and Popoola, A. F. (2014). Genetic Variability and Correlations in Rainfed Tomato (Solanum spps.) Accessions in Ibadan, Nigeria. Greener Journal of Agricultural Sciences ISSN:2276-7770, ICV (2012):6.15,Vol.4(5), PP.211-219. Ofosu-Budu, K. G. (2003). Performance of Citrus Root Stocks in the Forest Zone of Ghana. J. Hort, 3:1-9. University of Ghana http://ugspace.ug.edu.gh 141 Osei, M. K., Bonsu, K. O., Agyeman, A. and Choi,H. S. (2014). Genetic Diversity of Tomato Germplasm in Ghana using Morphological Characters. International Journal of Plant and Soil Science, 3 (3): 220-231, Article no. IJPSS. 2014. 001. 12334 Osekita O. S. and Ademiluyi, A. T. (2014). Genetic advance, heritability, and character association of components of yield in some genotypes of tomato (Lycopersicon esculentum Mill.) Wettsd. Academia Journal of Biotechnology 2(1): 006-010 Pandey, Y. R., Pun, A. B. and Upadhyay, K. P. (2006). Participatory varietal evaluation of rainy season tomato under plastic house condition. Nepal Agric. J., 7: 11 – 15. Panthee, D., (2012). Advances in Breeding Vegetables and Fruits for Enhanced Nutritional Content. 2012 Annual Conference of the American Society for Horticultural Science Parvej, M. R., Khan, M. A. H. and Awal, M. A. (2010). Phenological development and production potentials of tomato under polyhouse climate. The Journal of Agricultural Sciences, Vol. 5(1) pp. 19-31 Peng, Y., Zhang, Y. and Ye, J. (2008). Determination of phenolics compounds and ascorbic acid in different fractions of tomato by capillary electrophoresis with electrical detection. Journal of Agricultural and Food Chemistry. vol. 56, p. 1838- 1844 Peralta, I. E. and Spooner, D. M. (2005). Morphological characterization and relationships of wild tomatoes (Solanum L. Section Lycopersicum). Afestschrift for William G. D’Arcy. T. B. Croat, V. C. Hollowell and R. C. Keating, Missouri Botanical Garden Press. 104:227-157. Peralta, I. E. and Spooner, D. (2007). History, origin and early cultivation of tomato (solanaceae). Genetic improvement of solanaceous crops. M. K. Razdan and A. K. Mattoo. Enfield (NH), Science Publisher. 2: 1 – 24. University of Ghana http://ugspace.ug.edu.gh 142 Pessarakli, M. M. and Dris, R. (2004). Pollination and breeding of eggplant. Journal of Food, Agriculture and Environment 2 (11): 218-219. Pradeepkumar, T. Bastian, D., Joy, M. Radhakrishman, N. V. and Aipe, K. C. (2001). Genetic variation in tomato for yield and resistance to bacterial wilt. J. Trop. Agric., 39: 157-158. Pradip, K. S. (2013). Research Methodology: A Guide for Researchers in Agricultural Science Social Science and Other Related Fields. Springer, West Bengal, India. Premah, G., Indiresh, K. M. and Santhosha, H. M. (2011). Studies on genetic variability in cherry tomato (Solanum lycopersicum var. cerasiforme). Asian J. Hort. 6: 207- 209. Radzevicius, A., Viskelis, P., Viskelis, J., Bobinaite, R. Karkleleine, R. and Juskeviciene, D. (2013). Tomato fruit quality of different cultivars grown in Lithuania. International Journal of Biological, Veterinary, Agricultural and Food Engineering Vol. 7, No. 7. Rajasekar, M., Arumugam, T. and Kumar, S. R. 92013). Influence of weather and growing environment on vegetable growth and yield. Journal of Horticulture and Forestry, vol. 5(10), pp. 160-167. Reddy, B.R., Reddy, D. S., Reddaiah, K. and Sunil, N. (2013). Studies on genetic variability, heritability and genetic advance for yield and quality traits in tomato (Solanum lycopersicum L.). Int. J. Curr. Microbiol. App. Sci 2(9): 238-244. Riadh, I., Chafik, H. and Imen, T. (2009). Bioactive compounds and Antioxidant activity of tomato high active lycopene content advanced breeding lines. In: Daami-Remadi M University of Ghana http://ugspace.ug.edu.gh 143 (Ed) Tunisian Plant Science and Biotechnology I. The Africa Journal of Plant Science and Biotechnology 3 (Special Issue 1), pp 1-6 Riaz, M., Farooq, J., Mahmood, A., Sadiq, M. A. and Yaseen, M. (2013). Genotypic variability for reoot/shoot parameters under water stress in some advanced lines of cotton (Gossypium hirsuntum L.). Genetic and Molecular Research Online Journal. Rocha, M.M. de C., Delizarosire, C.B.M., Carmo, M. G. F., Abboud, A. C. S. (2012). A study to guide breeding of new cultivars of organic cherry tomato following a consumer-driven approach. Journal of food Research International, vol. 51, pp. 265- 273 Romer, S., Fraser, P. D., Kiano, J. W. Shipton, C. A., Misawa, N., Schuch, W. and Bramley, P. M. (2000). Elevation of the provitamin A content of transgenic tomato plants. Nat Biotechnol 18, pp. 666-669 Rosello, S., Adalid, A.M., Cebolla-Cornejo, J. and Nuez, F. (2011.). Evaluation of genotype, environment and their interaction on carotenoid and ascorbic acid accumulation in tomato germplasm. J. Sci. Food Agr. 91:1014-1021. Rukhsar, A.D., Jag, P. S., Ambreen, N. and Chopra, S. (2012). Germplasm evaluation for yield and fruit quality traits in tomato (Solanum lycopersicum L.). African Journal of Agricultural Research Vol. 7(46), pp. 6143-6149. Sablani, S.S., Luopara, A. and Al-Balushi, K. (2006). Influence of bruising and storage temperature on vitamin C content of tomato fruit. Journal of Food, Agriculture and Environment, 4: 54-56. Saimbhi, M. S., Singh, S. and Cheema D. S. (2001). Physicochemical characters of exotic varieties of tomato. Haryana J. Hort. Sci. 30 (3 &4): 279 – 280. University of Ghana http://ugspace.ug.edu.gh 144 Sharma, J. P., Singh, A. K., Kumar, S. and Kumar, S. (2009). Identification of traits for ideotype selection in tomato (Lycopercicon esculentum Mill.). Mysore J. Trop. Agric., 39: 157-158. Sharma, M. K., Kumar, R. and Kumari, S. (2011). Identifying superior quality F1 tomato hybrids for year round production under low cost plastic greenhouses in North-West Himalayas. Vegetable Science, 38(1): 30-34 Shankar, A., Reddy, R. V. S. K., Sujatha, M. and Pratap, M. (2013). Genetic variability studies in F1 generation of tomato (Solanum lycopersicon L.). IOSR Journal of Agriculture and Verterinary Science (IOSR-JAVS). Volume 4, Issue 5, PP 31-34. Shankara, N., Joep V. L. D. J., Marja, D. G., Hilmi, M, Barbara and Van dam (2005). Arodok 17 (2005), © Agromisa Foundation and CTA, Wageningen, 2005. Digigrafi, Wageningen, Netherlands ISBN Agromisa: 90-80573-039-2, ISBN CTA: 92-9081- 299-0, pp. 6-54. (www.prota.org) Shende, V. D., Tania, S., Subhra, M. and Arup, C. (2012). Breeding Tomato (Solanum lycopersicum L.) for higher productivity and better processing qualities. SABRAO Journal of Breeding and Genetics. 44(2) 302-321. Shiow, Y. W. (2003). Antioxidant Capacity of Berry Crops, Culinary Herbs and Medicinal Herbs. Acta Hort. 620, ISHS 2003. Shushay, C., Deerbew, B. and Fetien, A. (2013). Genetic Variability and Association of Character in Tomato (Solanum lycopersicum L.) Genotype in Northern Ethiopia. International Journal of Agricultural Research 8 (2): 67 – 78. Sims, W. L. (1980). History of tomato production for industry around the world. Acta Horticulturae: 100: 25-26. University of Ghana http://ugspace.ug.edu.gh 145 Singh, B. D. (1993). Plant Breeding Principles and Methods. Kalayani Publishers, New Delhi, pp. 677 Singh, S., Pandey, V. B., Singh, D. R. and Srivastava, R. C. (2009). Evaluation of tomato cultivars under protected conditions in Bay island conditions. Proceedings of the national Seminar on Production System Management in Adverse Condition for Higher Productivity in A and N Islands, December 22-24, India. Singh Tribhuwan, Narendra Singh, Abhishek Bahuguna, Mayank Nautiyal and V.K. Sharma, (2014). Performance of Tomato (Solanum lycopersicum L.) Hybrids for Growth, Yield and Quality inside Polyhouse under Mid Hill Condition of Uttarakhand. American Journal of Drug Discovery and Development, 4: 202-209. Sroka, Z. and Cicowski, W. (2003). Hydrogen peroxide scavenging, antioxidant and anti- radical activity of some phenolics acids. Food Chem. Toxicol (6). 41, 753-758. Sukalovic, V. H. T., Dodig, D., Basic, Z. , Kandic, V., Zilic, S., Delic, N. and Miritescu, M. (2013). Genotypic and environment variation of bread and durum wheat proteins and antioxidant compounds. In: Romanian Agricultural Research, No. 30(30), 125- 134. Sunil, M. K. Akhilesh, K. P., Kumar, A. P., Kumar, A. S. Sati, K. and Kumar, D. (2013). Studies on genetic parameters to improve the genetic architecture of tomato (Solanum lycopersicum, L.). International Journal of Applied Biology and Pharmaceutical Technology, Volume 4, Issue 4. Supathra, L., Nattaya, O., Kornkanok, A., Akkarach, B. and Kanit K. (2013). Phenolic acids content and antioxidant capacity of fruit extracts from Thailand. Chiang Mai J. Sci. 40(4): 636-642 University of Ghana http://ugspace.ug.edu.gh 146 Tambo, J. A. and Gbemu, T. (2010). Resource-use Efficiency in Tomato Production in the Dangme West District, Ghana. Conference on International Research on Food Security, Natural Resource Management and Rural Development. Tropentag, ETH Zurich, Swzld Tapier, H., Townsend, D. M. and Tew, K.. D.(2004). The Role of Carotenoids in the Prevention of Human pathologies and biomed pharmacotherapy, 58: 100-110. Tepic, A. N., Vejicic, B. L., Takac, A. J., Kristic, B. D. and Calic, L. J. (2006). Chemical heterogeneity of tomato inbred lines. APTEFE, 37:1-192. Tekeoka, G. R., Dao, L., Flessa, S., Gillespie, D. M., Jewell and Huebner, B. (2001). Processing effects on lycopene content and antioxidant activity of tomatoes. Journal of Agricultural and Food Chemistry, 49: 3713-3717. Toor, R. K.. and Savage , G. P. (2005). Antioxidant activity in different fractions of tomatoes. Food Research International. Vol. 38, pp. 487 – 494. Tokusoglu, O., Unal, M. K. and Yildirim, Z. (2003). HPLC-UV and GC-MS characterization of the flavonol aglycons Quercetin, kaemferol and myricetion in tomato paste and other tomato-based products. ACTA Chromatographica, No. 13. Ullah, M. Z., Hassan, M. J., Rahman, A. H. M. A. and Saki, A. I. (2011). Genetic Variability, Character Association and Path Coefficient Analysis in Radish (Raphanus sativus L.). A Scientific Journal of Krishi Foundation. The Agriculturalists 8(2): 22-27. Usman, A. (2013). Genetic analysis of resistance to rosette disease of groundnut (Arachis hypogaea L.). Mphil. Thesis, University of Ghana, Legon. University of Ghana http://ugspace.ug.edu.gh 147 Van der Hoeven, R. S., Ronning, C., Giovannomi, J. J., Martin, G. and Tanksley, S. D. (2002). Deductions about the number, organization and evolution of genes in the tomato genome based on analysis of a large expressed sequence tag collection and selective genomic sequencing. The Plant Cell 14, pp. 1441-1456. Vicente, R., Morcuende R. and Babiano, J. (2011). Differences in Rubisco and chlorophyll content among Tissies and growth stages in tomato (Lycopersicon esculentum, Mill) Varieties. Agronomy Research 9 (Special Isssues II), 501 – 507. Vinod, K., Nandan, R., Srivastava, K., Sharma, S. K., Kumar, K. and Kumar, A. (2013). Genetic parameters and correlation for yield and quality traits in tomato (Solanum lycopersicum L.). Plant Archives, Vol. 13, no. 1, pp 463-467, India. WHO (2003). Diet, Nutrition and the prevention of Chronic Diseases, Report of a Joint WHO/FAO Expert Consultation on, WHO Technical report series vol. 916. Geneva, World Health Organization. Willcox, J. K., George, L. and Lazarus, S. (2003). Tomatoes and cardiovascular health. Food Sci. Nutr. 43: 1 - 18 Yeboah, A. K. (2011). A survey on post harvest handling, preservation and processing methods of tomato (Solanum lycopersicum, L.) in the Dormaah and Tano South Districts of the Brong Ahafo Region of Ghana. MSc thesis, KNUST. Yu, S. A. and Sacco, V. (2005). Study of antimicrobial activity of lemon peel extract. Brit J Pharm Toxicol., 2 (3): 119-122. University of Ghana http://ugspace.ug.edu.gh 148 APPENDIX Appendix 1. Means squares of PHT, SG, NL, CC, NPB, RTL, NFF, FPF and NFS measured under greenhouse conditions MEAN SQUARE Source of variation df PHT SG NL CC NPB RTL NFF FPF NFS Replication 2 4.40 0.05 1.17 3.85 0.14 1.09 0.47 0.42 1.52 Genotype 19 678.90** 0.89** 270.11** 93.63** 1.74** 60.10** 11.10** 27.65** 130.00** Residual 38 1.38 0.06 0.58 1.08 0.06 0.82 0.17 0.31 0.36 Total 59 219.60 0.32 87.40 30.98 0.61 20.25 2.91 9.12 42.45 CV (%) 11.75 7.27 27.26 18.46 24.14 21.01 11.05 13.65 16.70 *Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; PHT= Plant height; SG = Stem girth; NL= Number of leaves CC = Chlorophyll content; NPB = Number of primary branches per plant, RTL = Root length; NFF = Number of days to 1st flowering; FPF = Number of days to 50 % flowering and NFS = Number of days to 1st fruit set. University of Ghana http://ugspace.ug.edu.gh 149 Appendix 2. Mean squares of FPFS, FSP, DFM, TPP, FPT, NFPT, FPP, SFW and FWP measured under greenhouse conditions MEAN SQUARE Source of variation df FPFS FSP DFM TPP FPT NFPT FPP SFW FWP Replication 2 1.72 6.55 0.47 1.32 0.02 0.06 0.14 13.18 25359.00 Genotype 19 209.00** 469.18** 140.24** 45.04** 3.01** 2.04** 200.23** 2068.08** 437280.00** Residual 78 0.49 56.07 0.33 0.88 0.18 0.10 0.18 10.99 38271.00 Total 59 67.68 187.43 45.39 15.12 1.09 0.72 64.60 673.52 166327.63 CV (%) 18.13 21.88 10.30 30.46 18.68 21.74 77.52 38.01 62.96 **Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; FPFS = Number of days to 50 % fruit set; FSP = Fruit set percentage; DFM = Number of days to fruit maturity; TPP = Number of truss per plant; FPT = Number of fruits per truss, NFPT = Number of fruits per truss; FPP = Number of fruits per plant; SFW =Single fruit weight and FWP = Fruit weight per plant University of Ghana http://ugspace.ug.edu.gh 150 Appendix 3. Means squares of YPP, LOC, PTK, FF, FL, FD, SI, FDM, and MA measured under greenhouse conditions MEAN SQUARE Source of variation df YPP LOC PTK FF FL FD SI FDM MA Replication 2 7.66 0.00 0.11 1.03 0.04 0.03 0.00 0.00 0.00 Genotype 19 3722.33** 9.28 2.00** 1.05** 1.46** 4.23** 0.09** 0.04** 0.02** Residual 78 4.65 0.00 0.10 0.14 0.02 0.05 0.00 0.00 0.00 Total 59 1201.97 2.99 0.71 0.46 0.49 1.39 0.03 0.01 0.01 CV (%) 66.97 48.77 17.52 15.17 15.01 23.43 18.55 26.09 23.53 *Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; YPP = Yield per plant; LOC = Number of locules per fruit; PTK = Pericarp thickness; FF = Fruit firmness; FL = Fruit length; FD = Fruit diameter; SI = Fruit shape index; FDM = Fruit dry matter content and MA= Malic acid content of fruit. University of Ghana http://ugspace.ug.edu.gh 151 Appendix 4. Mean squares of PH, TSS, TA, TSS/TA, β-CARO, LYC, ASC, GA and VA measured greenhouse conditions MEAN SQUARE Source of variation df pH TSS TA TSS/TA β-CARO LYC ASC GA VA Replication 2 0.00 0.03 0.00 0.44 0.32 0.09 8.05 0.00 0.00 Genotype 19 0.30* 4.07** 0.02** 22.63** 202.37** 125.67** 24.45** 0.02** 0.06** Residual 78 0.00 0.03 0.00 0.30 0.25 0.18 7.67 0.001 0.01 Total 59 0.10 1.33 0.01 7.49 65.34 40.59 13.09 0.01 0.02 CV (%) 5.48 26.21 25.32 19.59 41.30 40.17 24.34 5.34 15.83 *Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; pH = fruit pH; TSS = Total soluble solid content; TA = Titrable acidity content of fruit; TSS/TA = TSS: TA ratio or ripeness index; TI = Fruit taste index; β-CARO = Beta carotene content of fruit; LYC = Lycopene content of fruit and ASC = Ascorbic acid content of fruit University of Ghana http://ugspace.ug.edu.gh 152 Appendix 5. Mean squares of RA, QUE, RUT, CAT, KAE, HES, MYR, GEN and AA measured under greenhouse conditions . Mean square Source of variation df RA QUE RUT CAT KAE HES MYR GEN AA Replication 2 0.00 9.46 6.18 0.87 82.94 83.0 0.43 5.83 0.09 Genotype 19 0.04** 2828.77** 6418.92** 5636.78** 1047.25** 18404.30** 1105.10** 1564.71** 3.27** Residual 78 0.01 10.99 24.24 5.01 64.97 207.90 4.03 9.58 0.03 Total 59 0.02 918.36 2082.93 1818.49 382.00 6063.48 358.49 510.26 1.08 CV (%) 13.71 6.87 47.42 41.01 62.79 65.12 43.60 35.64 31.03 **Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; RA = Rosmarinic acid; QUE = Quercetin; RUT = Rutin; CAT= Catechin; KAE = Kaempferol; HES = Hesperetin acid ; MYR = Myricetin; GEN = Genestein and AA = Antioxidant activity. University of Ghana http://ugspace.ug.edu.gh 153 Appendix 6. Means squares of PHT, SG, NL, CC, NPB, RTL, FFF, FPF, and NFS measured under field conditions MEAN SQUARE Source of variation df PHT SG NL CC NPB RTL NFF FPF NFS Replication 2 21.89 2.84 26.47 14.64 0.28 0.65 5.52 3.35 2.12 Genotype 19 65.54** 2.09** 279.69** 45.23** 1.32** 53.25** 27.68** 15.42** 58.12** Residual 78 4.51 0.67 12.33 3.88 0.15 1.05 0.32 0.58 1.96 Total 59 24.75 1.20 98.91 17.56 0.53 17.85 9.31 5.45 20.05 CV (%) 9.40 10.00 21.61 9.70 17.76 21.60 19.83 9.15 12.25 *Significant at P < 0.01; CV = Coefficient 0f variability; PHT= Plant height; SG = Stem girth; NL= Number of leaves and RTL = Root length NFF = Number of days to 1st flowering; FPF = Number of days to 50 % flowering and NFS = Number of days to 1st fruit set. University of Ghana http://ugspace.ug.edu.gh 154 Appendix 7. Mean squares of FPFS, FSP, DFM, TPP, FPT, NFPT, FPP, SFW and FWP measured under field conditions MEAN SQUARE Source of variation df FPFS FSP DFM TPP FPT NFPT FPP SFW FWP Replication 2 2.83 15.32 0.22 1.11 0.02 0.14 0.97 0.76 2854.00 Genotype 19 80.56** 54.07** 86.26** 153.89** 2.67** 1.04** 215.73** 1279.09** 354810.14** Residual 78 4.48 12.69 1.81 1.08 0.34 0.23 0.70 4.12 1393.00 Total 59 28.93 26.10 28.95 50.29 1.08 0.49 69.96 414.59 115255.24 CV (%) 12.21 7.66 9.14 41.89 15.85 22.20 64.08 35.60 52.05 *Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; FPFS = Number of days to 50 % fruit set; FSP = Fruit set percentage; DFM = Number of days to fruit maturity; TPP = Number of truss per plant; FPT = Number of fruits per truss, NFPT = Number of fruits per truss; FPP = Number of fruits per plant; SFW =Single fruit weight and FWP = Fruit weight per plant University of Ghana http://ugspace.ug.edu.gh 155 Appendix 8. Mean squares of measured traits YPP, LOC, PTK, FF, FL, FD, DI, FDM and MA measured under field conditions MEAN SQUARE Source of variation df YPP LOC PTK FF FL FD SI FDM MA Replication 2 19.72 2.19 0.94 0.31 0.28 0.44 0.00 0.01 0.00 Genotype 19 2463.58** 7.85** 2.41** 0.55** 2.31** 3.08** 0.19** 0.01** 0.02** Residual 78 9.71 0.84 0.37 0.09 0.10 0.10 0.00 0.00 0.00 Total 59 800.28 3.14 1.05 0.25 0.82 1.07 0.00 0.00 0.01 CV (%) 52.04 50.45 20.87 13.74 18.04 20.71 24.66 15.16 19.47 *Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; YPP = Yield per plant; LOC = Number of locules per fruit; PTK = Pericarp thickness; FF = Fruit firmness; FL = Fruit length; FD = Fruit diameter; SI = Fruit shape index; FDM = Fruit dry matter content and MA= Malic acid content of fruit University of Ghana http://ugspace.ug.edu.gh 156 Appendix 9. Mean squares of measured PH, TSS, TA, TSS/TA, B-CARO, LYC, ASC, GA and VA measured under field conditions MEAN SQUARE Source of variation df pH TSS TA TSS/TA β-CARO LYC ASC GA VA Replication 2 0.00 0.00 0.00 0.04 0.86 0.20 0.30 0.02 0.05 Genotype 19 0.05** 4.11** 0.02** 9.05** 89.09** 94.26** 14.46** 0.24** 0.66** Residual 78 0.00 0.02 0.00 0.14 8.95 0.55 0.21 0.08 0.21 Total 119 0.02 1.34 0.01 3.01 34.49 30.72 4.80 0.13 0.35 CV (%) 2.93 29.29 19.41 17.43 23.34 36.31 23.34 16.17 40.26 *Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; pH = fruit pH; TSS = Total soluble solid content; TA = Titrable acidity content of fruit; TSS/TA = TSS: TA ratio or ripeness index; TI = Fruit taste index; β-CARO = Beta carotene content of fruit; LYC = Lycopene content of fruit, ASC = Ascorbic acid content of fruit, GA = Gallic acid and VA = Vanillic acid University of Ghana http://ugspace.ug.edu.gh 157 Appendix 10. Mean squares of RA, QUE, RUT, CAT, KAE, HES, MYR, GEN, AA measured under field conditions MEAN SQUARE Source of variation df RA QUE RUT CAT KAE HES MYR GEN AA Replication 2 0.03 8.89 12.02 32.64 4.06 55.43 4.24 8.35 0.05 Genotype 19 0.408** 8733.82** 19677.85** 16178.73** 3125.67** 54449.66** 3093.43** 4307.14** 19.04** Residual 78 0.13 3.41 6.35 16.14 2.03 18.37 1.26 2.04 0.07 Total 59 0.21 2815.08 6341.44 5221.60 1008.02 17548.35 997.15 1388.64 CV (%) 35.18 100.03 97.42 77.36 129.65 136.79 94.37 70.76 59.39 **Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; RA = Rosmarinic acid; QUE = Quercetin; RUT = Rutin; CAT= Catechin; KAE = Kaempferol; HES = Hesperetin acid ; MYR = Myricetin; GEN = Genestein and AA= Antioxidant activity University of Ghana http://ugspace.ug.edu.gh 158 Appendix 11. Mean squares of PHT, SG, NL, CC, NPB, RTL, NFF, FPF AND NFS measured across greenhouse and field locations MEAN SQUARE Source of variation df PHT SG NL CC NPB RTL NFF FPF NFS Replication 2 16.59 1.27 8.73 7.43 0.29 1.68 4.01 2.76 3.61 Location 1 185450.69** 22.97** 2060.65** 2382.35** 0.80* 97.20** 63.08** 625.63** 126.08** Genotype 19 495.30** 2.09** 435.70** 62.80** 2.13** 35.18** 26.43** 24.89** 138.22** Genotype x Location 19 249.14** 0.89* 114.10** 76.06** 0.93** 78.17** 12.36** 18.18** 47.18** Residual 78 3.06 0.40 6.78 2.70 0.11 0.91 0.29 0.46 1.13 Total 119 1679.56 0.95 109.69 44.09 0.57 19.70 6.98 12.48 31.46 CV (%) 10.21 7.46 21.48 9.23 17.64 11.90 13.11 9.02 12.79 *Significant at P < 0.05; **Significant at P < 0.01; CV = Coefficient of variability; PHT= Plant height; SG = Stem girth; NL= Number of leaves; RTL = Root length; NFF = Number of days to 1st flowering; FPF = Number of days to 50 % flowering and NFS = Number of days to 1st fruit set. University of Ghana http://ugspace.ug.edu.gh 159 Appendix 12: Mean squares of FPFS, FPS, DFM, TPP, FPT, NFPT, FPP, SFW and FWP measured across greenhouse and field locations MEAN SQUARE Source of variation df FPFS FSP DFM TPP FPT NFPT FPP SFW FWP Replication 2 4.23 14.17 0.10 2.21 0.01 0.17 0.85 10.12 6144.00 Location 1 76.80** 94.10* 5.04* 182.43** 7.86** 0.97* 217.16** 3697.22** 88605.00* Genotype 19 228.13** 300.68** 207.64** 155.47** 4.17** 2.66** 392.84** 3043.86** 709030.00** Genotype x Location 19 61.43** 222.57** 18.86** 43.47** 1.52** 0.41** 23.12** 303.31** 83061.00** Residual 78 2.43 33.69 1.06 0.96 0.25 0.16 0.44 7.46 19889.00 Total 119 48. 106.66 36.90 33.96 1.14 0.61 68.54 570.55 140352.88 CV (%) 14.14 12.58 9.25 34.29 12.92 20.02 67.99 35.44 54.26 *Significant at P < 0.05; **Significant at P < 0.01; ns = Not significant at 0.05; CV = Coefficient of variability; FPFS = Number of days to 50 percent fruit set; DFM = Number of days to fruit maturity; TPP = Number of truss per plant; FPT = Number of flowers per truss; FPP = Number of fruits per plant; Single fruit weight and Fruit weight per plant. University of Ghana http://ugspace.ug.edu.gh 160 Appendix 13: Mean squares of YPP, LOC, PTK, FF, FL, FD, SI, FDM and MA measured across greenhouse and field locations. MEAN SQUARE Source of variation df YPP LOC PTK FF FL FD SI FDM MA Replication 2 25.10 1.10 0.63 1.24 0.19 0.26 0.00 0.01 0.00 Location 1 182.18** 1.67ns 3.94** 16.53** 1.36** 0.93** 0.19** 0.20** 0.19** Genotype 19 5707.92** 16.52** 2.61** 0.99** 3.04** 6.53** 0.22** 0.02** 0.02** Genotype x Location 19 477.98** 0.62ns 1.80** 0.61** 0.73** 0.78** 0.07** 0.02** 0.02** Residual 78 7.05 0.44 0.24 0.11 0.06 0.08 0.00 0.00 0.00 Total 119 994.24 3.06 0.91 0.49 0.66 1.23 0.05 0.01 0.01 CV (%) 57.30 48.77 14.74 11.52 14.96 20.94 19.39 15.91 16.31 **Significant at P < 0.01; ns = Not significant at P < 0.05; CV = Coefficient of variability; df = degree of freedom; YPP = Yield per plant; LOC = Number of locules per fruit; PTK = Pericarp thickness; FF = Fruit firmness; FL = Fruit length; FD = Fruit diameter; SI = Fruit shape index; FDM = Fruit dry matter content and MA= Malic acid content of fruit. University of Ghana http://ugspace.ug.edu.gh 161 Appendix 14. Mean squares of PH, TSS, TA, TSS/TA, β-CARO, LYC, ASC, GA and VA measured across locations MEAN SQUARE Source of variation df pH TSS TA TSS/TA β-CARO LYC ASC GA VA Replication 2 0.00 0.02 0.00 0.26 0.23 0.03 4.85 0.01 0.03 Location 1 0.04** 5.84** 0.21** 513.72** 360.02** 3.21** 161.58** 0.80** 2.17** Genotype 19 1.94** 5.31** 0.02** 20.20** 115.51** 118.20** 19.46** 0.14** 0.38** Genotype x Location 19 1.01** 2.87** 0.02** 11.42** 175.95** 101.74** 19.55** 0.23** 0.34** Residual 78 0.02 0.03 0.00 0.22 4.51 0.36 3.93 0.04 0.10 Total 119 0.05 1.37 0.01 9.52 52.52 35.38 10.23 0.07 0.20 CV (%) 4.28 22.29 16.88 15.38 20.30 27.94 17.05 9.04 15.83 *Significant at P < 0.05; **Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; pH = fruit pH; TSS = Total soluble solid content; TA = Titrable acidity content of fruit; TSS/TA = TSS: TA ratio or flavour indicator; TI = Fruit taste index; β- CARO = Beta carotene content of fruit; LYC = Lycopene content of fruit; ASC = Ascorbic acid content of fruit; GA= Gallic acid and VA = Vanilic acid University of Ghana http://ugspace.ug.edu.gh 162 Appendix 15. Mean squares of RA, RUT, CAT, KAE, HES, MYR, GEN and AA measured across greenhouse and field locations MEAN SQUARE Source of variation df RA QUE RUT CAT KAE HES MYR GEN AA Replication 2 0.02 0.18 16.24 20.38 29.39 106.60 2.01 7.49 0.10 Location 1 0.23** 3379.34** 6231.98** 3481.66** 710.07** 14247.40** 2998.80** 3326.68** 23.09** Genotype 19 1.28** 4968.60** 11465.53** 9898.41** 1729.73** 32743.60** 1709.91** 2549.03** 9.96** Genotype x Location 19 0.20** 6593.99** 1463.10** 11917.09** 2443.45** 40110.40** 2488.61** 3322.81** 12.35** Residual 78 0.07 7.48 14.95 10.64 34.12 111.00 2.65 5.83 0.05 Total 119 0.12 1879.43 4229.16 3519.72 695.13 11826 697.32 969.43 3.79 CV (%) 21.24 48.57 48.39 40.49 62.13 67.53 43.26 35.05 33.89 *Significant at P < 0.05; **Significant at P < 0.01; CV = Coefficient of variability; df = degree of freedom; RA = Rosmarinic acid; QUE = Quercetin; RUT = Rutin; CAT= Catechin; KAE = Kaempferol; HES = Hesperetin acid; MYR = Myricetin; GEN = Genestein and AA = Antioxidant activity University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh