The metapopulation concept is appearing with increasing frequency in the marine population dynamics and genetics literature, though its applicability to marine systems remains an open question. Moreover, in recent years, the meaning of the term ‘metapopulation’ has become blurred, concomitant with its increasing use. In this paper, we summarize the concept of metapopulation dynamics and the associated theoretical assumptions. We call for a stricter definition and use of the term ‘metapopulation’, critically evaluate the applicability of metapopulation theory to marine population dynamics and its use in the related literature, and consider two published case‐studies that investigate metapopulation structuring in specific marine populations. Finally, we urge scientists to carefully articulate what is meant by the term ‘metapopulation’ and to use appropriate citations in the primary literature to circumvent the potential for nebulous (and possibly damaging) conclusions in the future.

Putative spawning waves of Atlantic herring Clupea harengus were collected from 4 locations and genetically compared using 9 microsatellite loci. Shelf-scale (100s km) spatial differences were measured (max F ST = 0.01, p < 0.001) and 1 sub-annual temporal difference (F ST = 0.0058, p = 0.001) was revealed between spawning fish (predominantly comprising the same year-class) collected, 6 d apart, from the same location (Devastation Shoal, coastal Nova Scotia). Herring in the second spawning collection at Devastation Shoal had a greater average length-at-age within the year-class (t-test, p < 0.001). The genetic and morphometric differences between the Devastation Shoal collections are assumed to reflect a replacement period of spawning fish of approximately 6 d or less (at least at this location). We offer 3 explanations to account for the observations: (1) genetically distinct sympatric subpopulations or spawning waves; (2) sub-annual (d) genetic patchiness; and (3) transient use of spawning grounds which may indicate that the assumption of philopatry (natal spawning-site fidelity) in herring is invalid.

Abstract.-The genetic variation among 17 spawning groups of Atlantic herring Clupea harengus was assessed at several temporal and spatial scales using nine tetranucleotide microsatellites. PanAtlantic samples were drawn from the Scotian Shelf, the Celtic Sea, the Baltic Sea, and coastal Iceland. Significant differentiation was observed between northeastern and northwestern Atlantic herring (F ST ϳ 0.065) and among northwestern Atlantic spawning groups (max F ST ϭ 0.014) at the spatial scale of the Scotian Shelf. Geographic distance among Scotian Shelf collections did not explain the pattern of genetic differentiation observed (e.g., lack of isolation by distance). The temporal proximity of collections (as measured by days between collections) explained 30% of the pairwise population differentiation (P ϭ 0.0025). Allele frequencies of replicate samples and year-class analyses demonstrate temporal stability at four locations.

Analysis of nine tetranucleotide microsatellite loci for Atlantic herring at five locations in the Northwest Atlantic including the Bras d'Or Lakes shows considerable genetic variation and significant population structure within the Coastal Nova Scotia management component, and among coastal populations and herring collected from Georges Bank. However, results are also consistent with gene flow across the Gulf of Maine. The magnitude of differentiation between the Bras d'Or Lakes sample and all others considered was sufficient to warrant further investigation. These data support the precautionary spawning‐ground based management approach implemented in this area.

Nine tetranucleotide microsatellite loci developed for Atlantic herring are presented. Loci were isolated using a magnetic bead hybridization selection protocol that yielded an 80% enrichment for tetranucleotide microsatellites. Fifty Atlantic herring from each of a North‐west and North‐east Atlantic spawning group were screened at each locus. Loci were polymorphic (12–56 alleles per locus) and exhibited high levels of observed heterozygosity (0.66–0.98). These loci are therefore suitable for population studies and the analysis of mating systems.