Fine-scale population structure in a deep-sea teleost (orange roughy, Hoplostethus atlanticus)

Microsatellite and otolith chemistry variability were analysed to assess fine scale genetic structure in the deep-sea teleost orange roughy (Hoplostethus atlanticus). The Porcupine Bank located on the continental shelf west of Ireland, comprises a complex system of mounds and flat areas that are broken up by canyons. Orange roughy form spawning aggregations on mounds and flat areas, and were heavily fished until the resource was depleted. By analysing adults in spawning condition and juvenile orange roughy from six mounds and one flat area, shallow but significant genetic population structure was evident (F_ST=0.0031, D_est across loci=0.0306 and G-test). Most of the structure was accounted for by inclusion of a sample from the flats (six of ten significant pairwise F_ST estimates and G-tests, and five of the highest D_est estimates included the flat sample). While the flat sample contributed most to the genetic structure, there was still significant (albeit weaker) structure among mound samples. The observed structure was supported by otolith analyses. Fish caught as late juveniles in either the flat or mound areas showed consistent differences in chemistry at the otolith core throughout the initial 10 years of growth, which could indicate site fidelity. We hypothesise that seafloor topographic structures (mounds and flats) may provide discrete spawning areas for orange roughy and that the limited gene flow between these spawning areas is insufficient to counteract genetic drift.     

High-resolution records of location and stratigraphic provenance from the rare earth element composition of fossil bones

Bone apatite acts as a natural, timed sampling device, scavenging trace elements from local pore waters over timescales of ca. 1-50 ka. The rare earth element (REE) and U/Th composition of fossil bones reflects associated pore water compositions during the period of recrystallisation. The REE composition of fossil bones is controlled by partitioning of REE between pore waters and particle surfaces, and the REE composition of fossil bones reflects the REE composition of pore waters which vary spatially and temporally. Light REE are preferentially sorped onto particle surfaces, thus the high La/Yb values seen in many bones from coastal marine and aeolian environments are best explained by release of REE from light REE-enriched particles to local pore waters and subsequent immobilisation in recrystallising bones. The REE compositions of bones recovered from pedogenically altered diatomite sediments of the Olorgesailie Formation of southern Kenya vary over spatial scales of less than 10 m. Location accounts for 48% of the observed variation in bone chemistry and bones recovered from eight discrete excavations within the same time-equivalent stratigraphic layer can be assigned to their excavation location with >70% accuracy based on a discriminant analysis of REE, U, and Th composition. Despite this within-layer variation, bones recovered from different stratigraphic horizons within the Olorgesailie Formation can also be distinguished on the basis of their trace element composition. Bones recovered from four stratigraphic horizons spanning ca. 0.5 million years were assigned to their correct stratigraphic layer with >90% accuracy. Where sedimentological conditions are favourable, the trace element composition of fossil bone may be used to test stratigraphic provenance and burial location in excavated bone with a temporal resolution of <10 ka and a spatial resolution of <10 m. The trace element composition of fossil bone may also be used to investigate the accumulation history of vertebrate assemblages and to reconstruct pore water variability across land surfaces.