Gene duplication followed by nucleotide differentiation is one of the simplest mechanisms to develop new functions for genes. However, the evolutionary processes underlying the divergence of multigene families remain controversial. We used multigene families found within the diversity of toxic proteins in centipede venom to test two hypotheses related to venom evolution: the two-speed mode of venom evolution and the rapid accumulation of variation in exposed residues (RAVER) model. The two-speed mode of venom evolution proposes that different types of selection impact ancient and younger venomous lineages with negative selection being the predominant form in ancient lineages and positive selection being the dominant form in younger lineages. The RAVER hypothesis proposes that, instead of different types of selection acting on different ages of venomous lineages, the different types of selection will selectively contribute to amino acid variation based on whether the residue is exposed to the solvent where it can potentially interact directly with toxin targets. This hypothesis parallels the longstanding understanding of protein evolution that suggests that residues found within the structural or active regions of the protein will be under negative or purifying selection, and residues that do not form part of these areas will be more prone to positive selection. To test these two hypotheses, we compared the venom of 26 centipedes from the order Scolopendromorpha from six currently recognized species from across North America using both transcriptomics and proteomics. We first estimated their phylogenetic relationships and uncovered paraphyly among the genus Scolopendra and evidence for cryptic diversity among currently recognized species. Using our phylogeny, we then characterized the diverse venom components from across the identified clades using a combination of transcriptomics and proteomics. We conducted selection-based analyses in the context of predicted three-dimensional properties of the venom proteins and found support for both hypotheses. Consistent with the two-speed hypothesis, we found a prevalence of negative selection across all proteins. Consistent with the RAVER hypothesis, we found evidence of positive selection on solvent-exposed residues, with structural and less-exposed residues showing stronger signal for negative selection. Through the use of phylogenetics, transcriptomics, proteomics, and selection-based analyses, we were able to describe the evolution of venom from an ancient venomous lineage and support principles of protein evolution that directly relate to multigene family evolution.