The Oregon Coastal Subprovince, a new biogeographic subprovince for primary freshwater fishes in Oregon

The Pacific Northwest has a relatively low diversity of primary freshwater fishes with most of the endemism and diversity in the Columbia River and Klamath River. However, the Oregon Coastal Subprovince defined as the coastal rivers from Miami River in the north to Sixes River in the south, has a relatively diverse primary freshwater fish fauna, and, potentially unrecognized endemism. The species diversity and endemism of these systems is not clear because their taxa are allopatric members of more wide ranging taxa with some recognized as distinct species while others are not. The presence of primary freshwater fishes in the Oregon Coastal Subprovince has been explained as either due to their origin in the coastal river systems or dispersal to coastal rivers from the Willamette River. The goals of this study were: 1) to describe fishes in the genera Catostomus and Ptychocheilus in the Oregon Coastal Subprovince using morphological data and mitochondrial cytochrome b sequences; 2) to investigate the relationships of fishes in the Oregon Coastal Subprovince to nearby provinces; and 3) to address competing distribution theories. In this study, I re-described C. tsiltcoosensis, C. macrocheilus, C. rimiculus, P. umpquae and P. oregonensis and recognized C. sp A (Coquille River) and C. sp. B (Rogue River). Catostomus tsiltcoosensis and C. sp. A tend to have higher counts of infraorbital pores and fewer dorsal fin rays than C. macrocheilus. Catostomus tsiltcoosensis had six fixed base pair differences from C. macrocheilus. Catostomus sp. A had 14 fixed base pair differences in cytochrome b from C. macrocheilus and C. tsiltcoosensis and had a narrower body width at base of the pectoral fin than C. tsiltcoosensis. In the cytochrome b phylogeny, C. macrocheilus (Columbia) was sister to C. tsiltcoosensis (Siuslaw River, Umpqua River and Coos River) and C. sp. A (Coquille River) was sister to C. macrocheilus and C. tsiltcoosensis plus C. columbianus and C. tahoensis. The Rogue River C. sp B was recognized as a separate species from C. rimiculus because it had higher counts of vertebrae anterior to the dorsal fin and had nine fixed base pair differences from C. rimiculus. Although Catostomus sp. B was previously placed in C. rimiculus, phylogenetic analysis showed C. rimiculus was more closely related to other catostomids in the Klamath Basin than to C. sp. B. This was likely caused by hybridization among four different species of suckers in the Klamath system. Ptychocheilus oregonensis tended to have fewer scales around the caudal peduncle, fewer scales above the lateral line, fewer transverse scales, deeper body depth at the origin of the dorsal fin, and shallower caudal peduncle than P. umpquae. Ptychocheilus umpquae had 15 fixed base pair differences from P. oregonensis. Based on phylogenetic analysis, P. oregonensis (Columbia - Willamette River) was sister to the P. umpquae (Siuslaw and Umpqua) and P. grandis (Sacramento) was sister to both. If C. tsiltcoosensis and C sp. A are considered separate species, the Oregon Coastal Sub-Province has 62.5% endemism. This suggests that it is another important area of endemism in the Pacific Northwest. Based on the suckers and pikeminnow phylogenies, two common nodes (Siuslaw- Umpqua vs. Willamette-Columbia and Sacramento vs. Willamette-Columbia-Siuslaw- Umpqua) were found in sucker and pikeminnow phylogenies. If pikeminnow and suckers shared a common history, two vicariant events (Sacramento from Willamette – Columbia – Umpqua - Siuslaw and Willamette - Columbia from Siuslaw - Umpqua) were responsible for such pattern. On the other hand, if the two groups had separate histories, the phylogeny of the suckers also suggested two additional vicariant events (Coquille from Willamette – Columbia – Umpqua - Siuslaw and Willamette - Columbia from Siuslaw - Umpqua). Similar to the sucker phylogeny, the phylogeny of the pike minnow suggested a vicariance pattern. The estimated divergence times among taxa were supported by geological evidence.