concept

Long-branch attraction

A systematic error of maximum parsimony: when two lineages change fast and long - two "long branches" - they accumulate so many novelties that some coincidentally match. The method counts those matches, mistakes them for evidence of kinship, and pulls the two long branches together near the base of the tree - even though they are not close relatives. The danger is that it fails *confidently*, producing a clean but wrong answer. Molecular data can usually break the trap; morphology alone cannot.