Variation among 41 genotypes of tomato (Lycopersicon esculentum Mill.) for crossability to L-peruvianum (L.) Mill.

Even with the aid of tissue culture, crosses between Lycopersicon esculentum (E) and L. peruvianum (P) typically yield few progeny. To determine whether some E genotypes produce more progeny per fruit that others when crossed with P, 41 E genotypes were crossed with pollen bulked from five P accessions. This first experiment (expt 1) was replicated over 2 years. In a second experiment (expt 2), differences among three genotypes each of E and P, and among individual plants within E genotypes were investigated. The E genotypes for expt 2 were chosen for relatively high and low crossability based on results of expt 1. The P genotypes for expt 2 were from different accessions than those used in expt 1. For both experiments, the 15 largest ovules from each ripe fruit were cultured aseptically for 1 month. Out of 1228 fruit, 753 hybrids were obtained. For expt 1, significant genotype by year interactions were observed. Within each year, there were significant differences among E genotypes for crossability. In expt 2, significant effects were found for E genotypes, but not for interactions between E and P genotypes, P genotypes, nor plants within E genotypes. Moreover, general crossability for E genotypes using bulked pollen (expt 1) was indicative of general crossability with three P accessions not present in the bulk (expt 2). Thus, selecting E genotypes of high crossability to P is the key to obtaining progeny for gene introgression. Rare production of E x P seed which was large and had brown seed coals typical of E seed indicated strong selection pressure to maintain separate species, but gene exchange in nature may be possible albeit at a low rate over long periods of time. (C) 1997 Annals of Botany Company.