Aim: To identify potentially human-mediated biogeographic patterns in selection and adaptive tradeoffs affecting the evolution an over-exploited shellfish.

Location: Hawaiian Archipelago

Taxon: Mollusca, Gastropoda, Patellagastropoda, Nacellidae, Cellana exarata, ‘Opihi makaiauli

Methods: We surveyed phenotypic characters associated with temperature and predation avoidance across the entire species range and tested for differences in the relationship between these characters and latitude, on islands with and without humans.

Results: Among all limpets surveyed, there was a bimodal distribution in shell color (light, dark) and a parapatric pattern of shell coloration across the archipelago with lighter shells being prevalent on the uninhabited islands and darker, more camouflaged shells being prevalent on the inhabited islands.  On the cooler, uninhabited islands, all morphometric characters associated with thermal avoidance (surface area, height, and doming) increased with decreasing latitude. On the hotter, inhabited islands, however, shells were flatter, less variable, and less adapted for avoiding thermal stress than predation. 

Main Conclusions: The biogeographic patterns in shell phenotype and previous genetic studies suggest that the population is beginning to bifurcate in response to disruptive and directional selection as well as geographic isolation between the islands with and without humans. Decreased phenotypic and genetic diversity on the inhabited islands despite much larger populations of ‘opihi suggests a prominent historical bottleneck. The prevalence of maladaptive dark, flat phenotypes for thermal avoidance on the inhabited islands suggests that predation is a stronger selective force, driving adaptive tradeoffs in shape and color.  We propose that this is likely a case of fisheries-induced evolution and a millennium of harvesting is the most likely selective pressure driving the observed biogeographic patterns in shell morphology.  The flatter, darker shells will allow body temperatures to rise higher in direct sunlight, therefore we hypothesize that the thermal niche of ‘opihi is narrower on inhabited islands and will continue to narrow as Earth warms.  This study highlights the utility of using intraspecific biogeographic patterns in phenotype to identify adaptive tradeoffs in response to varying selective pressures and identify nascent ecologically driven lineage splitting.

Living and presumably healthy Cellana exarata were collected from eight Hawaiian Islands between 2012 and 2016: ‘Ōnū (ON, Gardner Pinnacles, Puha honu), Lalo (LA, La Perouse Pinnacles, Mokupāpapa), Mokumanamana (MM, Necker), Nihoa (NI), Kaua‘i (KA), O‘ahu (OA), Maui (MA), and Hawai‘i (HI, Big Island; Table 1; Kikiloi et al., 2017). The NWHI were accessed during intertidal monitoring cruises sponsored by Papahānaumokuākea Marine National Monument, and eight ‘opihi were collected from each of six, 1 cm size classes. In the MHI, shells of legal harvesting size (> 3 cm) were haphazardly collected along a 7 km section of coastline on Maui, two locations on Kaua‘i (Miloli‘i and Kekaha) and O‘ahu (Kaka‘ako, ‘Ᾱina Moana), and one location on each of the remaining islands. All sites were composed of natural basalt formations with the exceptions of the O‘ahu and Hawai‘i sites (basalt riprap) and Kekaha (emergent coral reef). Site selection was largely haphazard because shell morphology (at the level investigated here) was observed to be more homogenous within than between the NWHI and MHI. Images were taken of the lateral face of each individual with other measurements and measurements from images in csv files which are processed and analyzed using R code found https://github.com/jdselwyn/Opihi_Morphology.