Species boundaries are not always straightforward, especially during early stages of divergence, when gene flow may still take place. Nipponocypris koreanus, a widely distributed freshwater fish species endemic to Korea, was recently classified as a separate species from its relative Nipponocypris temminckii. However, morphological differences between the two species are ambiguous and genetic differences between and within species remain largely unexplored. In this study, using mitochondrial DNA cytochrome oxidase I (COI), control region (CR) and our developed 9 microsatellite loci, we determined phylogenetic relationships and population genetic structure of N. koreanus among geographically separated six river basins in comparison with N. temminckii. We found that N. koreanus evolved relatively well-separated clades in each basin, and that some morphologically N. koreanus individuals were genetically N. temminckii, suggesting the possibility of natural hybridization between the two species. Nipponocypris koreanus individuals from the Nakdong and Seomjin Rivers are the most closely related to N. temminckii, and they probably represent cryptic species complex as they showed a genetic distance of 2.2~2.9%, although N. koreanus had a distinct genetic structure. Comparisons of the whole mitogenomes from different rivers further support the notion that the Nakdong and Seomjin River populations constitute cryptic species in N. koreanus rather than a single species. The results of this study will shed an intriguing insight into understanding how Korean freshwater fishes have evolved in response to geographically isolated major river/stream environments. The evolutionary mechanisms underpinning the cryptic species diversity in freshwater systems are worth to be further studied.

Bitterling fishes evolve an idiosyncratic symbiosis with freshwater mussels, in which they are obligated to spawn in the gills of mussels for reproduction. In recent years, freshwater mussel populations have been drastically diminishing, due to accelerating anthropogenic impacts, which can be large threats to the risk of bitterling’s extinction cascade (i.e. ‘coextinction’). The host mussel size may be an important factor driving the adaptation and evolution of bitterling’s reproductive phenotypes. Here we examined the host size preference and morphological adaptation of female bitterling to the host size from 17 localities at the Han River in Korea. Using our developed molecular-based species identification for bitterling’s eggs/larvae inside the mussels, we further determined the spawning patterns of seven bitterling species. Mean length of spawned mussels (N=453) was significantly larger than that of unspawned mussels (N=1,814), suggesting that bitterling prefers to use larger hosts as a spawning ground. Spawning probability was clearly greater as mussel size increases. Results of our reciprocal transplant experiments do provide some evidence supporting the ‘bitterling’s larger host preference’ hypothesis. Interspecific competition appeared to be intense as two fish species often spawned eggs in the same mussel individuals simultaneously. Longer ovipositor and more elongated egg may evolve in females of Tanakia signifier in response to larger host environments. The observed bitterling’s spawning preference for large-sized mussels may evolve perhaps because of the fitness advantage in relation to the offspring survival. Our findings further inform on the development of effective conservation and management strategy for the endangered bitterling fishes.