Hypotheses that neuromotor systems are conserved during evolution are examined. Focus is on the fundamental assumption underlying such hypotheses, that neuromotor patterns are homologous. The criteria for testing hypotheses of homology are briefly reviewed and applied to several cases in which neuromotor conservatism has been proposed. It is concluded that few studies of neuromotor conservatism are complete enough to convincingly corroborate a hypothesis of homology. Particular problems include an absence of specific definitions of the parameters designating the conserved neuromotor pattern and the lack of sufficiently broad and detailed phylogenetic tests. The hypothesis that terrestrially feeding vertebrates exhibit a conservative feeding program, which has acted as a constraint in evolution, receives particular attention and it is concluded that existing data do not support this hypothesis.