The purpose of this study was to test the hypothesis that the genetic diversity of commercially significant species of King Crabs (Lithodes spp.) along the south-eastern Pacific (SEP) comprises different independent evolutionary units (IEUs) with spatially isolated distribution. Nine localities from inner and open waters along the SEP Chilean coast (39 degrees S-55 degrees S) were sampled. We analyzed sequences from 173 individuals for the mitochondrial gene Cytochrome oxidase I (COX-I), 151 individuals for the Internal Transcribed Spacer 1 (ITS) and 135 for the structural ribosomal RNA (28S). Genetic delimitation was performed through three analytical methods: ABGD, GMYC, and its Bayesian implementation, bGMYC. Bayesian phylogenetic analyses and haplotype networks were also performed. Divergence time between clades was assessed for the COX-I marker and estimated from known evolutionary rates for this marker in other crustacean species and fossil calibration from other Anomuran species. Delimitation analyses, phylogenetic analyses, and mitochondrial haplotype networks suggested the presence of two deeply divergent mitochondrial lineages of Lithodes in the SEP, referred to as Clade1 and Clade 2. Nuclear markers showed low phylogenetic resolution and therefore were unsuitable for molecular species delimitation. Divergence time analysis of the mitochondrial lineages suggests a separation between Clades of approximately 2.3 Mya. The divergence time obtained suggested that Pliocene glaciations and deglaciations cycles could be involved in hybridization events between Lithodes IEUs at southern tip of South American coasts. The different frequencies of Lithodes haplotypes in inner and open water environments along SEP coasts could be explained by events such as the last glacial maximum or by differences in the adaptation of each clade to different environments. These findings support the necessity of evaluating the taxonomic status of Lithodes individuals found along SEP coasts under an integrative taxonomy approach or through markers with other evolution rates than those already used.