Genetic variability in many modern crops is very limited because of bottlenecks during domestication and past selection pressures. This narrow genetic base has resulted in a lack o f genetic variability in some crops, and increasing the susceptibility to biotic and abiotic stresses, and may limit responsiveness to market needs. Tomato (Lycopersicon esculentum M ill.) is one of many autogamous crop species in which the exploited germplasm has been severely reduced as a result o f the process o f domestication, and particularly because the initial germ plasm used to generate much of the material exploited in current varieties, represented a very small fraction of the initial variability available. The concept of genetic base broadening has been suggested as a means of mitigating this lack o f diversity in modern crops, with the aim to utilise the rich genetic resources available in wild relatives, vintage varieties, and land races. Genetic base broadening programmes involve the systematic utilisation of an arrangement o f genetic variability in such a way as to generate a mass of newly adapted gene stocks available as parents in breeding programmes.
This research examines options available within a genetic base broadening program me, limited by space and time. Different populations were created by hybridisation in order to examine options and feasibility within a base broadening programme. These included a study of the genetic diversity of the genus Lycopersicon, using 43 accessions o f different taxa to examine the level o f genetic variability in tomato, and the richness o f diversity available in wild relatives and vintage/landrace tomato cultivars. Hybridisation was conducted as part of genetic base broadening program me to create inter-taxon and intra-taxon crosses between selected tomato cultivars and wild relatives. As part o f possible strategies, double crosses between inter-taxon populations were tested and analysed. The created populations were selfed and examined using morphological and molecular markers for polymorphism , genetic distances and heterozygosity indices from genetic population analysis computational program packages Popgene and NTSYS.
Results are presented for these populations over a number of generations and reviewed against possible strategies for conservation and utilisation of this sample of populations for future breeding programmes. Results showed that there is large genetic diversity at morphological and molecular level between and within Lycopersicon taxa. L. esculentum presented limited genetic diversity within the accessions examined, and a narrow genetic base. However, substantial sources of genetic diversity are available to incorporate into the cultivated tomato from both wild relatives and old varieties and landraces of the cultivated species.
After hybridisation, the created populations did not follow the expectation of autogamous crops, and revealed only a tendency toward decreasing genetic variability in further generations. The F t generation behaved as expected, for both morphological and molecular markers, but in F2 and F3 generations, the results fluctuated from increasing to decreasing values for all indices examined. However, from the data obtained it was possible to theorise about the number of parents to be involved, and the created population size that should be used in genetic base broadening programmes, along with strategies for the conservation of the created genetic variability.
The methods utilised in this project, morphological and molecular markers, gave valuable information about the genetic diversity in self-pollinating generations. However, morphological characters were more limited than molecular markers in respect to information accuracy, because of the number and type o f traits selected. The sample size affected both type of markers. From the genetic indices utilised, average gene diversity (Hs), total gene diversity (//,), and effective number of alleles (Ae) were more informative than the arbitrary mean proportion o f polymorphic loci (P) and number of polymorphic alleles (A). However, all indices had some m erit and usefulness in analysing the data obtained in this research.
For the future, it is hoped to use the experience gained with Lycopersicon spp utilising morphological and molecular markers in order to answer some more of the questions that will arise in any genetic base broadening program me.
