A distinctive feature of the tomato species is their sympodial index (SPI)-the number of leaf nodes per sympodium (i.e., between successive inflorescences). All of the colored- fruited species plus L. hirsutum have a mean SPI of 3. whilst the remaining spp., all green-fruited, average 2. In all of the accessions I have examined, the SPI is remarkably consistent, deviations being infrequent enough to class as teratisms. A departure from normal in one sympodium is usually compensated by a departure in the opposite direction in one of the adjoining sympodia. Thus, in indeterminate L. esculentum with a normal SPI of 3, a sympodium with 4 leaf nodes will almost always be contiguous to one with 2 nodes. In progenies derived from crossing L. esculentum with other spp. that we have studied, it has been difficult to analyze inheritance because the former parent was usually sp, and +/sp segregation and its radical departure from normal sympodial sequence tend to obscure SPI segregation. The impression obtained from such segregations suggested a complex quantitative determination of SPI.
During the past season we had appropriate material for analyzing this character in a large hybrid population derived from intercrossing three spp. The project was planned for other purposes, but in the process of scoring many characters, we decided to include SPI. The pedigree of the cross was: L. esculentum ms-31 x F1. (L. cheesmanii f. minor LA1508 x L. pennellii (LA 716). Despite the radical differences among these three spp., fertility of all plants in the population was sufficient for fruit production, mortality rate was extremely low, and vigor was excellent. Even though the ms-31 parent was sp, determinate habit did not segregate because the male parent was homozygous for the dominant allele. Therefore, in respect to SPI, the pedigree was essentially a backcross of 3 x F1(3x 2). The character was scored by counting the number of leaf nodes in 10 sympodia and deriving the mean per plant. Sympodia were examined from different branches of each plant, avoiding the lower, less stable regions. The resultant distribution, shown in the graph to the left is strikingly bimodal with peaks at 2 and 3. Clearly, a major gene plus an undetermined number of quantitative genes of minor effect are at work. This experience suggests that it might be relatively easy to convert an SPI3 species to SPI2 or vice versa, the high dominance of the 2 parent greatly simplifying conversion of L. esculentum to SPI. All linkage tests of the SPI locus completed to date have proved negative.