Dwarf pea plants bearing two cotyledonary shoots were obtained by removing the epicotyl shortly after germination, and the patterns of distribution of (14)C in these plants was investigated following the application of [(14)C]IAA to the apex of one shoot. Basipetal transport to the root system occurred, but in none of the experiments was (14)C ever detected in the unlabelled shoot even after transport periods of up to 48 h. This was true both of plants with two equal growing shoots and of plants in which one shoot had become correlatively inhibited by the other, and in the latter case applied whether the dominant or subordinate shoot was labelled. In contrast, when [(14)C]IAA was applied to a mature foliage leaf of one shoot transfer of (14)C to the other shoot took place, although the amount transported was always low. Transport of (14)C from the apex of a subordinate shoot on plants bearing one growing and one inhibited shoot was severely restricted compared with the transport from the dominant shoot apex, and in some individual plants no transport at all was detected. Removal of the dominant shoot apex rapidly restored the capacity of the subordinate shoot to transport apically-applied [(14)C]IAA, and at the same time led to rapid cambial development and secondary vascular differentiation in the previously inhibited shoot. Applications of 1% unlabelled IAA in lanolin to the decapitated dominant shoot maintained the inhibition of cambial development in the subordinate shoot and its reduced capacity for auxin transport. These results are discussed in relation to the polarity of auxin transport in intact plants and the mechanism of correlative inhibition.