A study was conducted to quantify the effects of rootstock and training system on C allocation in apple. Dry-matter distribution was determined at harvest in 5-year-old `Golden Delicious' apple (Malus domestica Borkh.) trees on four rootstocks (MM.111 EMLA, M.7a, M.26 EMLA, and M.9 EMLA) and in three training systems (three-wire palmette, free-standing central leader, and nonpruned). Mobilizable carbohydrate content was determined at harvest and leaf fall in trees from the same planting on MM.111 EMLA and M.9 EMLA in all three training systems. Training system effects interacted with rootstock effects in dry weights of branches and of fruit. Nonpruned system shoot and fruit dry weights reflected known rootstock vigor; whereas, pruned system (three-wire and central leader) shoot dry weights were greatest and fruit dry weights were lowest in trees on M.7a. Rootstock affected the partitioning of dry matter between above- and below-ground tree components, with MM.111 EMLA accumulating significantly more dry matter in the root system than trees on the other rootstocks. Trees in the central leader and the three-wire palmette systems partitioned more dry weight into nonbearing 1-year shoots than trees in the nonpruned system. Root starch content at harvest was greater in trees on MM.111 EMLA than on M.9 EMLA, and root sucrose and sorbitol were less in trees on MM.111 EMLA compared to M.9 EMLA. At leaf fall, starch in young roots was equal in trees on both rootstocks, and sorbitol again was lower in trees on MM.111 EMLA. Harvest starch content of roots, shoots, and branches was lower in nonpruned than in pruned trees. At leaf fall, root, shoot, and branch starch content increased in nonpruned and central leader-trained trees but did not increase in three-wire palmette-trained trees.
If the inline PDF is not rendering correctly, you can download the PDF file here.