One of the interesting phenomena of intercrosses is in regards to continuous, quantitative traits such as fruit size. Here's F2 #28, with an avg. weight of about 15 grams - well below that of the
smaller parent, Indigo Rose.
So - what we can conclude from this F2 plant is that some contribution of genes from the much larger parent (Cherokee Purple) must have contributed genetic material that further reduced fruit size in the context of the smaller parent's genes. How is this possible?
There are a few known tomato genes with large effects on size, and it is likely that Cherokee Purple is homozygous for the "big" alleles at these genes. However, there are undoubtedly many more genes that have smaller effects on size too. Cherokee Purple probably has some alleles that act to reduce fruit size slightly, but in the context of the Cherokee Purple genome these are overwhelmed by the action of the other loci that have much stronger effect to promote bigness.
In F2 28, I speculate that this plant has inherited "small" alleles from CP for loci that in IR are represented by "big" alleles. It is easier to understand this if one remembers that the ancestral fruit size was very small - just a few grams. In this view, IR certainly has some alleles that promote size way beyond the ancestral state. In fact almost no tomatoes eaten routinely today are as small as the ancestral species/subspecies. Exceptions are "currant" varieties, such as one of our family favorites, Hawaiian Currant. Here's an example: