Abstract:
The literature includes conflicting reports of effect of cane length (i.e. number of nodes per cane after pruning) on mean fruit number per cane and total yield at harvest with optimal number of buds per cane varying from 12 to 21. The aim of the present paper is to simulate the effect of cane length using the model of Agostini and Habib, published elsewhere in this volume.
This model uses a stochastic process to simulate both intensity and delay of flowering in the kiwifruit.
Six experimental treatments are simulated.
In each treatment, 8 canes (i.e. 8 replicates = one experimental treatment) are pruned to a given number of buds from 5 to 30. Then, 100 runs (i.e. 100 replicates of each experimental treatment) of the model are made to calculate the simulated distribution of mean fruit number per treatment.
It appears that the bud fertility decreases when the number of buds per cane increases.
From this study, it appears that the discrepancy between published results is mainly related to the between- and within-cane variability.
|