Migratory marine animals exploit resources in different oceanic regions at different life stages, but how they navigate to specific oceanic areas is poorly understood. A particular challenge is explaining how juvenile animals with no prior migratory experience are able to locate specific oceanic feeding habitats that are hundreds or thousands of kilometers from their natal sites. Although adults reproducing in the vicinity of favorable ocean currents can facilitate transport of their offspring to these habitats, variation in ocean circulation makes passive transport unreliable, and young animals probably take an active role in controlling their migratory trajectories. Here we experimentally demonstrate that juvenile Chinook salmon (Oncorhynchus tshawytscha) respond to magnetic fields like those at the latitudinal extremes of their ocean range by orienting in directions that would, in each case, lead toward their marine feeding grounds. We further show that fish use the combination of magnetic intensity and inclination angle to assess their geographic location. The "magnetic map" of salmon appears to be inherited, as the fish had no prior migratory experience. These results, paired with findings in sea turtles, imply that magnetic maps are phylogenetically widespread and likely explain the extraordinary navigational abilities evident in many long-distance underwater migrants.

Natural selection often results in profound differences in body shape among populations from divergent selective environments. Predation is a well-studied driver of divergence, with predators having a strong effect on the evolution of prey body shape, especially for traits related to escape behavior. Comparative studies, both at the population level and between species, show that the presence or absence of predators can alter prey morphology. Although this pattern is well documented in various species or population pairs, few studies have tested for similar patterns of body shape evolution at multiple stages of divergence within a taxonomic group. Here, we examine morphological divergence associated with predation environment in the livebearing fish genus Brachyrhaphis. We compare differences in body shape between populations of B. rhabdophora from different predation environments to differences in body shape between B. roseni and B. terrabensis (sister species) from predator and predator free habitats, respectively. We found that in each lineage, shape differed between predation environments, consistent with the hypothesis that locomotor function is optimized for either steady swimming (predator free) or escape behavior (predator). Although differences in body shape were greatest between B. roseni and B. terrabensis, we found that much of the total morphological diversification between these species had already been achieved within B. rhabdophora (29% in females and 47% in males). Interestingly, at both levels of divergence we found that early in ontogenetic development, females differed in shape between predation environments; however, as females matured, their body shapes converged on a similar phenotype, likely due to the constraints of pregnancy. Finally, we found that body shape varies with body size in a similar way, regardless of predation environment, in each lineage. Our findings are important because they provide evidence that the same source of selection can drive similar phenotypic divergence independently at multiple divergence levels.

Models of habitat selection often assume that organisms choose habitats based on their intrinsic quality, regardless of the position of these habitats relative to low-quality habitats in the landscape. We created a habitat matrix in which high-quality (predator-free) aquatic habitat patches were positioned adjacent to (predator-associated) or isolated from (control) patches with single or two species of caged predators. After 16 days of colonization, larval insect abundance was reduced by 50% on average in both the predator and predator-associated treatments relative to isolated controls. Effects were largely similar among predator treatments despite variation in number of predator species, predator biomass, and whether predators were native or nonnative. Importantly, the strength of effects did not depend on whether predators were physically present. These results demonstrate that predator cues can cascade with equal strength across ecological boundaries, indirectly altering community assembly via habitat selection in intrinsically high-quality habitats.

Predation can drive morphological divergence in prey populations, although examples of divergent selection are typically limited to nonreproductive individuals. In livebearing females, shape often changes drastically during pregnancy, reducing speed and mobility and enhancing susceptibility to predation. In the present study, we document morphological divergence among populations of nonreproductive female livebearing fish (Brachyrhaphis rhabdophora) in predator and nonpredator environments. We then test the hypothesis that shape differences among nonreproductive females are maintained among reproductive females between predator and nonpredator environments. Nonreproductive females in predator environments had larger caudal regions and more fusiform bodies than females in nonpredator environments; traits that are associated with burst speed in fish. Shape differences were maintained in reproductive females, although the magnitude of this difference declined relative to nonreproductive females, suggesting morphological convergence during pregnancy. Phenotypic change vector analysis revealed that females in predator environments became more similar to females in nonpredator environments in the transition from nonreproductive to reproductive. Furthermore, the level of reproductive allocation affected shape similarly between predator environments. These results suggest a life-history constraint on morphology, in which predator-driven morphological divergence among nonreproductive B. rhabdophora is not maintained at the same level during pregnancy.

Two experiments were conducted in vivo with eggs of rainbow trout Oncorhynchus mykiss to compare the bactericidal ability of four common disinfectants. A third test compared bacterial abundance estimation methods for fish eggs (use of a vortex mixer for agitating an egg versus rolling the egg across a petri dish). In the first test, the number of colony forming units (CFU) counted on enriched Ordahl's agar with tobramycin (EOT) or trypticase soy agar (TSA) was compared among eggs treated with various doses of iodine, hydrogen peroxide, formalin, or rock salt. A treatment of 1,667 mg of formalin/L of water and all iodine, salt, and hydrogen peroxide treatments had significantly fewer bacteria on EOT than did controls, but CFU counts for a formalin treatment of 500 or 1,000 mg/L did not. All chemical treatments significantly reduced CFU counts on TSA relative to controls except salt at 0.030 mg/L and formalin at 500 mg/L. The least growth was observed on iodine-treated eggs. In the second experiment, we evaluated the effect of keeping the eggs suspended (i.e., constantly tumbling) during disinfection to increase chemical contact. Treatments were (1) static application of iodine at 100 mg/L, (2) suspension in iodine at 100 mg/L, (3) static application of iodine at 500 mg/L, (4) suspension in iodine at 500 mg/L, (5) suspension in formalin at 2,000 mg/L, (6) suspension in hydrogen peroxide at 2,000 mg/L, (7) static control, and (8) suspended control. Bacterial abundance was significantly reduced in suspended eggs in some cases but not others. Formalin and hydrogen peroxide reduced bacterial abundance but were inferior to iodine in some cases. Each chemical treatment resulted in the survival of bacteria despite attempts to attain better chemical contact by egg suspension. Comparison of the methods used to estimate total CFU per egg indicated that agitation recovered 71-100% of the bacteria on the outside of the egg. Hatchery managers should be aware that not all bacteria are killed by chemical treatment of eggs, and therefore a significant number of pathogens could still enter a hatchery via the importation of treated eggs.