[1] The axillary wood (or xylem) formed in this location is denser than in surrounding tissues of the tree's stem or branch, the wood grain pattern formed is tortuous and in these tissues there is typically a reduction in vessel length, diameter and frequency of occurrence (i.e. more of the xylem consists of fibre cells).
This is not the case in tree forks, where the growth rate of both branches is approximately equal, and no occluded knot is formed.
[4] A clear anatomical difference can be found on dissection between these branch attachments, in that the former has an initial knot that originates near to the stem's pith, those developing from latent buds will have a bud trace that originates near the stem's pith, and adventitious epicormic branches will exhibit neither of these internal features.
[8] In 1985, the researcher Dr. Alex Shigo published an anatomical model of branch attachment[9] based on analysis of extensive tree cross-sections and which came to be widely utilised in the arboricultural industry.
Such a wood grain arrangement would result in sap travelling from the foliage at the end of the branch to other foliage in the tree's crown, which is contrary to the 'source-to-sink' model widely accepted for sap distribution in all woody plants,[12] and can clearly be seen not to be the case by dissections of junctions formed in trees.