Unbranched rooted layers of the rootstock Mailing 9 (M.9 clone Nie 29) ≈600 mm long were treated twice in autumn (4 and 13 October 1995) in the stoolbed as follows: a) non-treated control; b) the distal 300 mm of the shoot was painted with an aqueous solution of TIBA [250 mg 1−12,3,5-triiodo-benzoic acid with 2 ml l−1Tensospray wetting agent (100 g l−1alkyl-aryl-polyethyleneglycol)]; c) the proximal ®300 mm of the shoot was painted with an aqueous solution of Promalin® (3000 mg 1−1N-(phenylmethyl)-1H-purine 6-amine and 3000 mg 1−1gibberellins A4+A7without a wetting agent); d) the above TIBA and Promalin treatments combined on the same shoot. Twenty rooted stoolshoots per treatment were forced on 20 October 1995 (autumn), 5 December 1995 (mid-winter), and 21 February 1996 (late winter). The growth rates of the terminal bud, an upper lateral bud (≈200 mm from the apex) and a lower lateral bud (≈400 mm from the apex) were observed. Twenty additional rooted shoots per treatment were allowed to growin situ.Budburst relative to bud position was regularly recorded until no further change was observed. Two months later, length and position of the new shoots was recorded. The application of TIBA to the distal shoot half reduced the ability of the distal buds to establish dominance. The application of Promalin to the proximal lateral buds increased the growth rate sufficiently to overcome the distal inhibitions associated with acrotony, and burst simultaneously with the terminal bud. Promalin increased the basitonic branching from proximal buds. The data suggest the hormonal control of the correlative phenomena that determine tree architecture.