The idea that restricting the growth of roots provides a means to control plant growth has been known for thousands of years. The Japanese and Chinese have practised the art of "bonsai" probably for at least 1,000 years. We are all also well aware of the benefits of root restriction as a horticultural practice, where not only the unwanted vegetative shoot growth (the bonsai effect) is reduced, but where there is also a promotion of the diversion of photosynthetic carbohydrates into fruit dry matter and yield.
Scientists at East Malling Research (EMR), over many years, have developed their understanding of the processes that control this diversion of carbohydrates into fruit rather than unwanted woody plant parts. Much of this understanding has enabled a modern, worldwide fruit industry to evolve and be highly economic.
Earlier, more practical, approaches developed the grafting of clonal apple-root systems onto favoured fruit scion varieties, that impose a "dwarfing" response on the scion. This not only enhances the tree's precocity (time to fruiting - maturity), but also its yield efficiency (fruit weight per tree size). Both these factors are key foundation stones in modern, high-density fruit orchards, pioneered at EMR, but now utilised throughout the world.
A few years ago work undertaken at East Malling involved imposing physical root restrictions on the growth of apple trees, in a manner similar to that achieved within a container, but in an orchard. This involved planting young Cox apples in orchard soil contained within flexible synthetic polymer membranes. Initial experiments involved membranes that were of different weaves and pore sizes (spaces between the weave).
We found that selection of the correct membrane was critical as some roots were able to grow through due to the gaps in the weave. In these cases the impact on controlling tree size was not as great as when the roots remained contained by the membrane. One treatment actually girdled the roots (restricted the amount of living tissue in the root). The effect of this was not as great as complete root restriction, but much better than no restriction. This "partial restriction" approach has now been effectively exploited commercially by amenity tree nurseries.
Where the roots did not penetrate the membrane, shoot extension was reduced and cropping efficiency increased. Over a number of growing seasons the trees, which were subject to root-restriction, showed less annual growth (shoot and stem girth increments) resulting in smaller trees.
This was an important piece of work because it informed us about how naturally occurring growth processes in the plant reduced shoot growth. Chemicals were synthesised in the root system and transported to the shoot where they reduced growth. This, at least in part, has also helped us understand the control of growth in grafted combinations of dwarfing rootstocks and scions of apple trees. It also made us think about whether such an approach was universal in the sense that by restricting tree roots we could control the growth and development of any species of tree.
Due to the removal of water below foundations, urban trees are considered by some people as the primary candidates for causing shrinkage in clay soils that leads to subsidence damage to property. This situation has been made worse by the impacts of climate change as another cause of reducing the water content of clay soils in the summer.
Trees in the urban environment are a precious resource and add considerably to the quality of life. Scientists at EMR were engaged in a Horticulture LINK project to examine what the impact of trees was on soil-water-induced shrinkage and to determine whether there were other ways to control tree growth to avoid potential subsidence damage. Nobody wanted to see extensive tree felling as the only solution.
As a small part of this project, EMR scientists started an experiment in 2000, based on EMR's earlier experiments with root-restricted apples. It involved the field planting of amenity trees of lime (Tilia cordata Award of Garden Merit (AGM)) and Norway maple (Acer platanoides AGM) in a range of tree pits with different soil volumes contained within a geotextile woven membrane, Terram 3000, believed to be impermeable to roots.
The membranes were carefully folded to minimise root penetration. Four different pit volumes (no restriction, 2,000, 1,000 and 250 litres) were used. The growth of the trees was measured over several years and the experiment lasted seven years, but in 2007 EMR had to finally curtail the experiment due to financial constraints.
The results were very interesting as during the experiment we were able to show that the root-restricting membranes were very effective at controlling tree growth. Measurements of shoot extension or incremental stem girths of the trees in the smallest pit size (250 litres) showed reductions compared to the unrestricted trees at the end of the second growing season for both the lime and Norway maple.
But after the third year, differences in growth could be attributed directly to the differences in the volumes of soil available to the roots. Again, both lime and Norway maple showed similar responses, but lime had the greater overall growth. These reductions in tree growth were apparent with both the 250-litre and 1,000-litre pits, while the 2,000-litre pits were no different from the unrestricted control trees. By the time the experiment was stopped there were marked differences in tree size as well as in the trunk cross-sectional area.
Inspection after the trees were extracted revealed that some roots had penetrated the membranes. This was particularly evident with the smallest pits. With the larger sizes there was an occasional root escaping - slightly more penetrated with Norway maple than lime. Although differences between species may not be unexpected, this subject warrants further evaluation. Particularly important is the speed with which root restriction could effectively control shoot extension growth and tree size with more commercially relevant pit sizes, ie 1,000 litres. The possibility remains to investigate other membrane products of higher specification than we were able to use, which could control growth even more effectively.