Differences in adaptive traits often correlate with environmental variation, but can arise from two different sources: differentiation of genes, or phenotypic plasticity, the capacity of a single genotype to encode multiple phenotypes. Distinguishing between these mechanisms is a central challenge in evolutionary biology. Hawaiian stream fishes provide a superb system for directly measuring the evolutionary mechanisms leading to changes in form and function in response to environmental variation. The five fish species native to Hawaiian streams are all amphidromous: larvae hatch upstream but are quickly swept to the ocean, where they develop up to six months before returning to freshwater. For three species, returns require climbs of vertical waterfalls tens of meters tall before reaching adult habitats. However, adult habitats differ strongly between the oldest (Kaua'i) and youngest (Hawai'i) islands. Hawaiian waterfalls are close to shore, placing a premium on climbing ability to escape non-climbing predators; in contrast, Kauaian falls are far inland, placing a premium on predator evasion as fish migrate upstream. Biomechanical models predict these contrasting demands will select for divergent body shapes. This project is a collaboration among a biomechanist, a population geneticist, a morphologist, and an aquatic ecologist to test the hypothesis that natural selection has promoted adaptive divergence in stream gobies in response to differing environmental challenges between Hawai'i and Kaua'i through a combination of genetic differentiation and adaptive phenotypic plasticity. A series of complementary studies on the climbing goby Sicyopterus stimpsoni and the non-climbing goby Stenogobius hawaiiensis will be conducted. Goals are to: (1) evaluate morphological divergence in juveniles and adults from several stream subpopulations on Hawai'i and Kaua'i, and test for correlations with environmental differences in streams; (2) evaluate the strength of selection imposed on gobies by climbing and predation, and test for performance differences between populations from Hawai'i and Kaua'i; (3) evaluate neutral genetic divergence in juvenile and adult S. stimpsoni and S. hawaiiensis from several streams on Hawai'i and Kaua'i, and test for differential patterns of gene flow in these species; (4) conduct common garden predator exposure experiments to evaluate levels of genetically based local adaptation and environmentally induced phenotypic plasticity among island subpopulations. This work will provide a model for research on the interactive mechanisms promoting the evolution of organismal form and function in response to environment. This project involves collaboration of research-focused (Clemson) and teaching-centered (St. Cloud State) institutions, and will promote integrative training in genetics, morphology, and functional biology for both undergraduate and graduate students, including local interns with the Hawai'i Division of Aquatic Resources (DAR). Cooperation with the DAR will inform water management decisions so that detrimental impacts of human stream use can be limited in Hawai'i, helping to protect these endemic freshwater fishes.
Hawaiian stream fishes provide a superb system for measuring the evolutionary mechanisms leading to changes in form and function in response to environmental variation. The five fish species native to Hawaiian streams are all amphidromous (Fig 1): larvae hatch upstream but are quickly swept to the ocean, where they develop up to six months before returning to freshwater. For three species, returns require climbs of tall waterfalls before reaching adult habitats. Larvae from different islands may mix in the ocean, potentially returning to islands that are different than the one where they hatched and leading to gene flow between island subpopulations. However, the environments to which juveniles return differ strongly between the oldest (Kaua’i) and youngest (Hawai’i) islands. Hawaiian waterfalls are close to shore, placing a premium on climbing ability to escape non-climbing predators; in contrast, Kauaian falls are far inland, placing a premium on predator evasion as fish migrate upstream (Fig 2). Biomechanical models predict these contrasting demands will select for different body shapes: streamlined bodies to reduce drag on Hawai’i, vs. tall bodies for generating thrust on Kaua’i. Can these different environments promote divergence in shape between fish from these islands, even in the face of gene flow? This project used stream fish species from the Hawaiian Islands to examine how natural selection and gene flow interact to influence how populations diverge in form and function in response to environmental variation. We first measured the extent of genetic and body shape (morphology) differences between subpopulations of S. stimpsoni from Hawai’i and Kaua’i, both in juvenile and adult fish. Despite limited genetic divergence, we found significant differences in morphology that correlate well with expectations for the functional demands found on each island (climbing vs. predation). Such shape differences could result as a developmental response to cues experienced when fish enter a specific environment (called phenotypic plasticity), or through the action of natural selection. Because fish arriving at each island are already different in shape, plasticity is an unlikely cause for shape differences between island subpopulations. We next tested the potential for natural selection to explain shape differences between fish from Hawai’i and Kaua’i. We measured the strength of selection imposed on juvenile S. stimpsoni by climbing and predation, and tested for differences in climbing performance between fish from Hawai’i and Kaua’i. Selection experiments showed that both climbing and predation impose significant selection on morphology. Patterns of selection that result from each pressure generally favor body shapes that are expected to improve performance in the face of that pressure, and provide evidence for at least a moderate functional tradeoff between the demands of climbing and predator evasion. Supporting this conclusion, climbing performance is inferior among S. stimpsoni from Kaua’i, which are slower while climbing and failed in a higher proportion of climbing trials than fish from Hawai’i. These results are consistent with the hypothesis that contrasting predominant selective pressures on Hawai’i (waterfalls) and Kaua’i (predators) contribute to divergence in body shape between fish subpopulations from these islands. They also suggest a need to consider differences in functional performance among stream fishes during water management decisions applied across the Hawaiian Islands. When we compared how natural selection imposed by climbing acted on juvenile S. stimpsoni from Hawai’i and Kaua’i, we found that even when they are exposed to the same agent of selection, groups that have adaptations to different local environments may experience different patterns of selection on body shape. Such differences can occur between islands and between subpopulations inhabiting different streams on the same island. However, the variety of selection patterns that result among such groups may all still favor enhanced performance against the applied selective pressure, just by favoring different combinations of traits. This may reflect "many-to-one mapping" between structure and function. Climbing by S. stimpsoni is distinctive, using movements of a sucker on the mouth as well as one on the belly. We found that the mouth moves very similarly during both climbing and feeding in this species, suggesting that the unusual climbing behavior of S. stimpsoni may have evolved by simply coopting movements from feeding. This process is known as exaptation, and could be another mechanism impacting the evolution of stream fish in habitats with extreme environmental pressures. This project involved collaboration by research-focused (Clemson) and teaching-centered (St. Cloud State) institutions, and promoted integrative training in genetics and functional biology for undergraduate and graduate students, including local interns with the Hawai’i Division of Aquatic Resources (DAR). Integration of these fields has raised awareness of the insights that each can provide to a broader audience. Moreover, cooperation with the DAR facilitated the use of our data on recruitment and survival of Hawaiian stream fishes to inform water management decisions so that detrimental impacts of human stream use can be limited in Hawai’i, helping to protect these native species.
