The use of true potato seed (TPS) as a propagule for potato tuber (Solanum tuberosum L.) production is a viable alternative to the use of seed tubers. For this technology to be successful, efficient production of high- quality botanical seed is crucial. The objectives of the research reported in this thesis were to define production practices that maximise quantity and quality of botanical seed produced, and to contribute to the understanding of above-ground development in potato.The first part of the study describes experiments on seed production under field conditions in three contrasting agro-ecological zones in Peru. Seed production is described as a function of number of flowers produced, berry set and number of seeds produced per berry. Hundred-seed weight is used as seed-quality parameter. In the second part of the thesis, flower production is analysed as a function of inflorescence production, number of flower primordia initiated per inflorescence and flower primordia survival.The results of the field experiments indicated that generally, later-produced inflorescences on a shoot or in a field, and later-produced flowers in an inflorescence have a lower berry set and produce fewer and smaller seeds per flower. The effect of the position of the flower in the inflorescence does not affect seed size in all cultivars. Hundred-seed weight of production from primary inflorescences was increased when later-produced inflorescences were not used for seed production, but this could not compensate for the reduction of seed yield. Increasing plant density reduced the number of inflorescences per shoot and the number of flowers per inflorescence. Flower production per m2 increased with plant density in two of the three cultivars used. Berry set, number of seeds per berry and 100-seed weight were reduced when comparing flowers at similar positions on the shoot. However, because increasing plant density shifted the flower production from later- to earlier- formed flowers, the effect on average berry set, number of seeds per berry and seed size of the total seed production was relatively small. The effect of plant density on seed production was largely determined by the effect on flower production. Artificial extension of the photoperiod and interruption of the night with incandescent light increased the flower production under warm tropical conditions. This effect was principally a result of an increased inflorescence production. Photoperiod treatments did not affect the seed production per flower.Experiments in controlled conditions showed that increasing photoperiod and temperature increased the production of inflorescence positions per shoot, number of flower primordia per inflorescence and flower primordia survival in the temperature range of 15-25 °C (24-h average). Shoot development and flowering in potato were quantified as functions of rates and durations of leaf and flower primordia initiation, and of stem production. Effects of increasing temperature and photoperiods on shoot development and flowering were a result of increasing thermal durations of stem production, and leaf and flower primordia production of individual stems. The effects on individual stems were, however, small compared to the effects on stem and inflorescence production.Conclusions from the study for practical TPS production are that the last produced flowers in a field have a strongly reduced potential for seed production and that seed production can be best increased by increasing flower production through longer photoperiods and higher temperatures.