Abstract Sink-source ratio in tomato was manipulated, in six glasshouse experiments, by fruit pruning (trusses pruned to two to seven fruits immediately after fruit set of each truss), truss pruning (removal of every other truss at anthesis) and truss pruning in plants with two shoots. Periodic destructive harvest were conducted for about 100 d after flowering of the first truss. Dry matter production was not influenced by sink-source ratio, whereas dry matter distribution between fruits and vegetative parts was greatly affected. The fraction of dry matter distributed to the fruits at the end of the fruit pruning experiments (F fruits ) could be described accurately as a saturation-type function of number of fruits retained per truss (N f ): F fruits = 0.660 (l-e -0.341N f ). Specific leaf area and internode length decreased and plant leaf area increased when sink-source ratio was reduced. Removal of every other truss at anthesis did reduce dry matter partitioning into the fruits, but it did not influence internode length. Plant development (number of visible leaves at the end of the experiments) was not influenced by sink-source ratio. In four experiments some plants were pruned to one fruit per truss. Final dry matter production was 8-24% lower for these plants, compared with plants with more than one fruit per truss. This was, at least party, the result of less light interception by these plants, which had strongly curled leaves pointing downwards. Results indicate that effects of sink demand on dry matter production per unit of intercepted radiation and probably on leaf photosynthetic rate in commercial tomato production can be ignored.