It has been proposed that fw2.2 encodes a negative fruit-growth regulator that underlies natural fruit-size variation in tomato (Lycopersicon spp.) via heterochronic allelic variation of fw2.2 expression, rather than by variation in the structural protein itself. To further test the negative regulator and the transcriptional control hypotheses, a gene dosage series was constructed, which produced a wider range of fw2.2 transcript accumulation than can be found in natural tomato populations. Fruit developmental analyses revealed that fw2.2 transcript levels were highly correlated (negatively) with fruit mass, supporting the negative regulator and transcriptional regulation hypotheses. Further, the effect of fw2.2 on fruit mass was mediated by repressing three-and two-dimensional cell division in placental and pericarp tissues, respectively. Finally, fw2.2 had little effect on fertility and seed size/number, indicating that fruit size effects of fw2.2 are due largely to expression in the maternal tissues of developing fruit and not mediated through fertility or seed-setting-related processes.Crop domestication began in several regions around the world about 7,000 to 10,000 years ago-a very recent event in the entire evolutionary history of plants (White and Doebley, 1998). In the short span of crop domestication, genetic changes followed by repeated cycles of human selection have fundamentally altered the morphology, physiology, and overall environmental adaptations of a handful of wild species, leading to the formation of modern crops (Diamond, 2002). However, when and how these events took place remains unclear.Genetic studies have demonstrated that most traits that distinguish modern crops from their related wild species are due to quantitative trait loci (QTLs) with distinct effects (White and Doebley, 1998;Grandillo et al., 1999;Mackay, 2001;Barton and Keightley, 2002). With the advent of molecular markers in combination with statistical methodology, dissecting the genetic and molecular bases of these traits is no longer an impossible mission. During the last decade, we have witnessed the cloning of several major QTLs related to crop domestication in maize (Zea mays; Doebley et al., 1997), tomatoes (Lycopersicon esculentem; Frary et al., 2000;Fridman et al., 2000;, and rice (Oryza sativa; Yano et al., 2000).A key morphological change that has accompanied the domestication of many fruit and vegetable crops has been the dramatic expansion of fruit and explosion of shape variation. Tomato is a classic example. The wild forms of tomato bear small (approximately 1-2 g), round, seed dense berries-ideal for reproduction and dispersal. In contrast, cultivated tomatoes typically produce fruit that weigh anywhere from 50 to 1,000 g, come in a wide variety of shapes (e.g. round, oblate, pear-shaped, torpedo-shaped), and are not well adapted for seed dispersal in the wild. Genetic studies involving crosses of wild and cultivated tomatoes have shown that most of the variation in size and shape can be attributed to fewer t...