Scion - rootstock relationships and root behaviour in glasshouse roses

In the Netherlands, the cultivation of cut roses in the glasshouse is commonly carried out yearround. Most cultivars are grown on a rootstock. The first part of this study investigates various rootstocks for their ability to influence production and quality throughout the year. The mutual influence between scion and root system during normal practical growing management in soil without artificial light was followed and its physiological background studied. In the second part attention was mainly focussed on shoot-root ratio, root carbohydrates and mortality and regeneration of roots.In Part I, in one long term experiment, the effect of 17 root systems on plant behaviour of the scion <em>cultivar Rosa</em> 'Varlon' grown in soil without supplementary lighting was studied for three years under conditions of relatively low greenhouse temperatures. It became clear that the root system influenced the building up of the rose bush, and that this particularly took place during the first year after planting.Considering only the plants propagated by stenting, the number of bottom-breaks for <em>R.</em> 'Varlon' plants varied between 1.5 for <em>R.</em> 'Motrea' and 2.3 per plant for <em>R. indica</em> 'Major'. About 90 % of all the bottom-breaks were formed in the initial months after planting. The quality of these bottom- breaks, e.g. the diameter at the time they were harvested, was 15 percent higher for the best rootstock, <em>R</em> . 'Mme Alfred Carrière', compared with the least, <em>R.</em> 'Paul's <em></em> Lemmon Pillar'. The percentage of secondary thickening of bottom-breaks after three years varied somewhat per rootstock and was on average about 20%. Eventually, the difference between rootstocks with respect to bottom-break diameter was about the same as obtained during their harvest.After three years it became apparent that plants on <em>R.</em> 'Moonlight' produced the highest number of branches per bottom-break, i.e. 2.4, whereas the bottom-breaks of <em>R. zigrii</em> had only 1.4 laterals. The diameter of these laterals varied from 10.2 mm for plants on <em>R.</em> 'Fredica' to 8.6 mm. for plants on <em>R.</em> 'Paul's Lemmon Pillar'.The number of shoots at 60 cm above ground level and exceeding 0.5 cm, diameter, the socalled structural shoots, was 6.6 per plant for <em>R.</em> 'Moonlight', which means that it exceeded the least <em>R. zigrii</em> by 60%. This means that the same tendency was found here as was noticed for the number of branches per bottom-break. It became clear that competition existed between the number of structural shoots per plant; the more shoots produced, the more eventually died off. The diameter of the structural shoots reached a maximum when grown on <em>R.</em> 'Mine Alfred Carrière' and a minimum when grown on <em>R.</em> 'Paul's Lemmon Pillar'. This result is comparable with that found for the diameter of the bottom-breaks.The method of propagation proved to be an important factor with respect to the building up of the plant; the plants bench grafted on <em>R. canina</em> 'Inermis' produced the highest total number of bottom-breaks but their diameter, secondary thickening and weight remained behind all the other rootstocks including the root-grafted <em>R. canina</em> 'Inermis' plants.In the same <em>R.</em> 'Varlon' experiment differences of more than 100% between rootstocks, irrespective of the method of propagation, were found for the period with the lowest production, i.e. winter. The differences between rootstocks in average production per year was less (about 60%). The highest production was found for <em>R.</em> 'Varlon' on its own roots and <em>R. indica</em> 'Major', the lowest for both <em>R. canina</em> types, irrespective of the method of propagation, and <em>R. zigrii.</em> Further, it proved that some of the rootstocks exceeded others during the whole year, whereas some diverged only in some periods. The high production of <em>R.</em> 'Varlon' might be related to the abscence of the graft union.The number of axillary buds per plant and the number of released buds, but also the readiness to break out and, thereafter, the time required to develop a harvestable flower, are important factors in determining the total amount of flowers produced. In this study it was shown that the differences found in bud break between rootstocks during the year explained, at least partly, the differences found in production between rootstocks and between the seasons of the year. Also, flower stem development after axillary bud break proved to be dependent on rootstock. The importance of the presence of the subtending leaf and the diameter of the parent shoot depends on the chosen rootstock. This in turn depends on the method of carbohydrate supply (from actual photosynthesis or from storage) for the release of the bud and the development of the young shoot.The effect of rootstock on production (number of flowers) as well as on the quality of the harvested flower (length and weight) was more pronounced in winter than it was in summer. The quality parameters used were well correlated. However, the flower stem length was less influenced by season than flower weight.Finally, using the data from the long term <em>R.</em> 'Varlon' experiment, production and quality were quantified by linear regression models, using the plant development parameters, irrespective of the rootstock, as regressors. Production as well as quality was (partly) explained, for each season, by the transverse sections of the branches of the bottom-breaks. The number of bottombreaks was less important for production, even less than the number of their laterals; the former parameter is only of importance for production in spring and summer. The importance of the transverse sections is probably due to their role in the storage of reserves and transport capacity. Therefore, easiness of bud break and the time required for flower development is better when emerging from thick parent shoots. This is especially emphasized by the more pronounced role of the transverse section of the shoot for production in winter and for production by older bushes.It was suggested, when comparing the data concerning bush development parameters at the start and after three years, that the development of the plant in the first six months after planting was important for the building up of the rose bush and therefore for production and quality throughout its whole life.In Part II, examination of the plant's root system revealed that root fresh weight for the root systems used as rootstock for <em>R.</em> 'Varlon' differed markedly after three years. The root system of <em>R.</em> 'Mme Alfred Carrière' had 40% less weight compared with <em>R.</em> 'Moonlight'. Further, some evidence was found to suggest that propagation technique may play a role in root characteristics, such as root number and average root length.In this study, the shoot-root ratio of young rose plants (younger than one year), varied between three and six and proved to depend on growing technique, environmental factors and the combination of rootstock and scion cultivar used. With progressing plant age shoot-root ratio increased to 10 -15 for three years old plants. It can be stated that the mutual influence of shoot and root, with respect to partitioning of assimilates between shoot and root, depends on the combination and can be influenced by climatic conditions, plant age and growing technique.For <em>R.</em> 'Varlon' grown on different root systems, a close relationship was found to exist between the plant's shoot-root ratio after three years and the ratio between total flower weight produced over three years and final root weight. It was suggested that a fixed share of assimilates of the upper parts is used for flower formation irrespective of the root genotype used.From several experiments it became clear that disturbance of the balance between shoot and root, either by shoot or root pruning, was followed by a re-establishment of the original shootroot ratio. This is achieved independently of the scion-rootstock combination used, or the age of the plant. When frequently repeated intervention was carried out by removing upper parts, the shoot- root ratio reached a new stable balance but at a lower level. It was suggested that in this case storage of reserves was taken over by the roots.In experiments with rooted leaf cuttings it was found that a surplus of assimilates can be reflected in a higher root mass. On the other hand, it was also found that a higher root mass can indicate that the sink activity and/or storage capacity of the root is higher.The amount of carbohydrates stored in the roots of a rose plant during the year is related to the development of the shoot and root and the photosynthetic activity of the leaves. It was shown that apart from the time of year the combination of scion and rootstock used also had a prominent effect. With respect to the storage of carbohydrates in rose roots, it became clear that rootstocks originating from 'winter active' plants, such as <em>R. multiflora</em> 'Multic' and <em>R. indica</em> 'Major' showed a less pronounced variation during winter, compared with the originally hardier ones such as the <em>R.</em> canina types. Further, aspects such as the difference in storage capacity, sink activity, redistribution and climatic factors may play a role in the storage of carbohydrates in roots.Root growth was related to the presence of leaves. Consequently the removal of leaves by harvesting flower stems or other methods of shoot pruning, resulted in inhibited root growth within one or two weeks. Additionally, root mortality may occur depending on the severity of pruning. The extent of root mortality proved to be dependent to some degree on the rootstockused. Their susceptibility to root mortality depended on the amount of carbohydrates stored, internal competition between roots and shoot-root competition.After root mortality the best root regeneration was observed when more carbohydrates were available. This means that a root system with a higher storage capacity (i.e. mostly thicker roots), plants with more leaves, plants grown under a higher light intensity, or plants treated with supplementary sucrose, showed better root regeneration. Stimulating the root's sink activity by application of auxin improved root regeneration too. The optimal concentration of auxin (IBA) applied for root regeneration was dependent on the temperature used for plant growth. With respect to temperature, root regeneration was optimal around 17 °C.Root growth of <em>R.</em><em>canina</em> 'Inermis' plants grown in peat was not limited between pH values 4 to 8. Root growth for plants grown on hydroculture was optimal when pH was about 6. Dormant plants only remained inactive at a pH of 3.5 or lower, but young active roots disintegrated within a few hours at this pH value. Bud release was inhibited at pH levels <em></em> that hampered root growth, probably because production of cytokinin remained behind.Reduced root growth during a short period or to a relatively low extent was not necessarily damaging for the plant. In fact, shoot growth was enhanced when root mass was decreased temporarily by removing root ends. The emergence after a while of a higher number of root apices, suggested that the imroved shoot growth was due to a higher production of cytokinin. A more pronounced reduction of root growth led to less shoot development, more water stress and a higher susceptibility to disease.In conclusion, it may be said that the whole life of a rose plant and therefore also flower production and quality, is influenced by competition between upper parts and roots. Good development requires not only an adequate assimilating system but also good storage and redistribution capabilities. Storage capacity is of especial importance during unfavourable conditions. For optimal production throughout the year, the roots should be active and after mortality, a quick recovery is required to ensure the correct balance between shoot and root. In this respect the behaviour of the rootstock plays an important role.Therefore, new scion cultivars should be assessed when growing on rootstocks which have adapted properly to the grower's method of plant management and his growing conditions. Cultivars for year round production for instance should never be tested on <em>R.</em><em>canina</em> selections, due to their poor winter activity.For the correct balance between shoot and root, plant management should he carried out in dependence of climatic conditions, rooting medium, scion and rootstock combination and plant condition itself.The requirements for the ideal rootstock as well as for the ideal scion cultivar are tremendous . Therefore, it is more practical to select rootstock and scion cultivars in separate programmes. Finally, the most suitable combination for a certain situation could then be combined by grafting.