In recent years, growers of temperate tree crops have successfully adopted closer planting systems and smaller trees to achieve earlier cropping, high regular yields and improved management but in tropical fruit production such systems remain largely unexploited. It is widely accepted that tree vigour is strongly dependent on root activity although details of the controlling mechanisms are not clear. One of the most promising hypotheses is that shoot and root growth are controlled by a feedback system based on nutrients, carbohydrates, or hormones either individually or in combination. If this is correct, the key to manipulating tree size will involve a better understanding of root-shoot relationships in the control of vigour in tropical fruit trees.
This study was concerned mainly with establishing the relationship between root and shoot growth in selected tropical fruit trees: mango (Mangifera indica 'Kensington Pride'), grapefruit (Citrus x parodist 'Wheeney'), and macadamia (Macadamia integrifolia 'H2'). A wide range of physiological and morphological characteristics were measured in trees of varying degrees of vigour with the aim of identifying any process causally related to control of vigour and to determine whether the control of vigour could be attributed to hormonal or to stress (water or nutrient) effects.
A number of the techniques investigated for controlling tree vigour were successful in achieving this objective. In all cases the reduction in tree vigour was expressed by a reduction in tree height and/or dry weight, but stem thickness, leaf area, lateral growth and dry matter distribution were also affected in some cases. Although the effects of chemical restriction (by the ethylene generator CEPA or the gibberellin antagonist paclobutrazol) on shoot elongation in mango were superficially similar to those of physical root restriction, different mechanisms appear to have been involved. Root restriction and paclobutrazol increased the diversion of metabolites to the roots, so lowering the S/R (shoot/root) ratio, whereas CEPA (at non-toxic levels) reduced shoot elongation without any effect on dry matter distribution. The lowering of the S/R ratio by paclobutrazol was due to a smaller proportion of dry matter in the stems while trees with restricted root systems had a lower proportion of dry matter in both stems and leaves. Root and chemical restriction also differed in their effects on trunk development. It is therefore suggested that the reduction in shoot elongation, although one of the main effects of root restriction, is only part of the mechanism which controls vigour.
Apple (Malus x domestica 'Granny Smith'), though it is not a tropical fruit tree species, was included in this study because no dwarfing rootstocks of tropical trees were available. Using apple trees grafted onto either EM 26 or M 779, respectively dwarfing - inducing or non - dwarfing rootstocks, this experiment attempted to determine if the mechanism controlling tree vigour in root restricted trees was the same as that in trees on dwarfing rootstocks. The similarity in the responses suggests that this may be the case.
The analysis of xylem sap in these experiments provides no support for the hypothesis that shoot vigour is under the control of a specific, hormonal, root signal such as cytokinin or gibberellins. Similarly the reduction in tree vigour did not appear to be due to nutrient deficiencies, lack of assimilates or stomatal behaviour although it may have been affected in some cases by water stress.
It is concluded that the most satisfactory hypothesis to account for the control of vigour by root restriction is that of Brouwer, which depends on competition between shoot and root for an essential metabolite. No direct evidence has been obtained for the identity of this substance, but it is proposed that it may be a specific nitrogen compound, produced in the roots and essential for shoot growth.