This paper tested the hypothesis that the final number of fruits (nº/m³) depends on the present number of floral buds, flowers and/or fruitlets. This hypothesis is based on the idea that the plant completes itself with the edafoclimatic conditions available for its growth and development. For this, the hypothesis was tested in orange "Valencia" (Citrus sinensis (L.) Osbeck.) grafted on mandarin "Cleopatra" (VCL) (Citrus reshni Hort. ex Tan.) and lime "Cravo" (VCR) (Citrus limonia Osbeck). The following relations between reproductive structures were tested: Fruits= f(Fruitlets), Fruits= f(Flowers), Fruits= f(floral buds), as well as Flowers=f(floral buds) and Fruitlets= f(Flowers), besides Fruits= f(Flowers of the north side) for VCL and Fruits= f(buds of the north side) for VCR suggested by cluster analysis. The evaluation of the number of reproductive structures was made in four cardinal points of the canopy (1m³/sector). Due to the high R², agreement index (d) and relative low mean percentage absolute error (EAPM) resulted from exponential regression in relation to Flowers = f(floral buds) (R²=0,85 and R²=0,82; d= 0,95 and d= 0,95; EAPM= 15,2% and EAPM= 19,1%, for VCL and VCR respectively) and mainly for fruits number (for example: Fruits = f(floral buds); (R²=0,62 and R²=0,71; d= 0,89 and d= 0,88; EAPM= 5,0% and EAPM= 2,1%, for VCL and VCR respectively), the hypothesis could be accepted, which means that, it is possible to estimate the future number of fruits from the present number of reproductive structures (floral buds and flowers ).
Citrus sinensis; nonlinear model; floral bud; fruitlets; fruit
