While this subject has occupied tomato geneticists for several decades progress has been very slow. In the main this is due to the lack of satisfactory methods for dealing with multiple-gene-determined characters. However, in part, the slow progress has been due to the haphazard choice of material for hybridization studies. It would seem logical to begin a study of size and shape inheritance by surveying the distribution of tomato varieties throughout the world.
Only a beginning has been made in such a survey (Jenkins, 1948). Nevertheless a number of importent facts have emerged. Probably the most important of these is that in areas of ancient cultivation (Mexico, Peru and the intervening countries), there are two main trends in the evolution of fruit size and shape. In the first of these, beginning with a small two-loculed berry of one gram or less in weight, there has been a gradual increase in locule size resulting in berries weighing up to 100 grams. Together with this increase in locule size, there has been an increase in the size of the seeds, the size of the placenta and most conspicuously an increase in thickness of the outer wall of the locules. Accompanying the increase in locule size there has been a gradual change in shape from, spherical to oval (i.e. with a greater polar diameter). Present evidence indicates that the increase in locule size has been due to the accumulation of many gene mutations, the bulk of which are recessive. The basic oval shape of the larger two-loculed fruits may be further modified by additional mutant genes resulting in pear-shaped and nipple-tipped tomatoes.
The second main trend in the evolution of the tomato fruit has been an increase in the number of locules. This multiplication of locules has resulted in an increase in the equatorial diameter until ultimately the fruit becomes kidney-shaped. In contrast to the increase in size of locules, the increase in number of locules seems to have a simpler genetic basis. In some crosses the increase in locule number seems to be due largely to a single recessive mutation, in other crosses there are evidences of numerous modifying genes. The fact that there are multi-locular types with all sizes of locules indicates that there may have been several independent mutations to the multilocular condition.
In contrast to the areas of primative cultivation, the Multi-locular types have their locules rediating from a fibrous, central placenta there has been a third evolutionary trend, which is largely confined to the tomatoes of the United States. In this third type the locules are irregular in shape and scattered throughout a soft central placental region.
Crosses involving a representative samle of the different types are now under study and will be reported in greater detail at a later date.