Introduction. The production of `Tahiti' limes in southern Brazil has increased in the last 20 years and has become a good alternative for citrus growers in regions where Citrus Variegated Chlorosis has affected sweet orange groves. One of the main goals to insure success in growing `Tahiti' limes is to apply the right techniques to induce out-of-season flowering and winter fruit harvest when the fruit prices are the highest. This experiment evaluated the effects of gibberellic acid application on fruit set of `Tahiti' lime trees related to out-of-season flowering and accumulated degree days. Materials and methods. Eight-year-old plants were treated with five concentrations of gibberellic acid [ (0, 10, 25, 50 and 100) mg$\cdot$L-1] applied during anthesis of out-of-season bloom. Accumulation of degree days for fruit set and fruit development in the conditions of São Paulo State, Brazil, were calculated. The percentage of fruit set determined by weekly counts of the number of flowers from anthesis to the maturation of the fruits were determined on each previously marked branch. Data were submitted to regression analysis. Results and discussion. In regard to fruit set, it was observed that 10 mg gibberellic acid$\cdot$L-1 was enough to retain more than 10% of fruits in relation to the number of formed flowers. Starting from anthesis, (837.7 and 1 115.2) accumulated degree days were necessary for the stabilization of the fruit set and fruit maturation, respectively.

Introduction. Tahiti lime trees grafted on Rangpur lime grown in São Paulo State, Brazil, in sub-tropical climatic conditions, have a high level of vigor, poor tolerance to Phytophthora spp. and low productivity. Using Poncirus trifoliata var. monstrosa Flying Dragon rootstock which allows one to obtain dwarfed trees and high yields could solve this problem. Materials and methods. In an experiment set up in November 1994 in Brazil, the tree size, fruit production and quality of Tahiti lime grafted on trifoliate orange Flying Dragon were evaluated at four planting spaces: 4 m × 1 m (2 500 trees$\cdot$ha-1); 4 m × 1.5 m (1 666 trees$\cdot$ha-1); 4 m × 2 m (1 250 trees$\cdot$ha-1) and 4 m × 2.5 m (1 000 trees$\cdot$ha-1), in a rand- omized block design, with four trees per plot. The cultural practices did not include supple- mentary irrigation. Results. Trees planted in the 4 m × 1 m planting space had greater can- opy diameters (2.75 m) than trees planted with the other planting spaces. Whatever the densities considered, tree height did not differ. Average fruit yield (1998 to 2000) was the greatest on the 4 m × 1 m planting space with 21.6 t$\cdot$ha-1, significantly differing from the 4 m × 2.5 m planting space with 13.1 t$\cdot$ha-1. Fruit quality was affected by the plant planting space, but all the production was commercially acceptable. Discussion. The use of a high density planting of Tahiti lime grafted on P. trifoliata var. Flying Dragon would be of com- mercial interest, due to greater yields than trees in traditional cropping systems.