DNA barcodes of fish of the Scotia Sea, Antarctica indicate priority groups for taxonomic and systematics focus J. ROCK 1 *, F.O. COSTA 1 , D.I. WALKER 1 , A.W. NORTH 1 , W.F. HUTCHINSON 2 and G.R. CARVALHO 1 1 Molecular Ecology & Fisheries Genetics Laboratory, School of Biological Sciences, University of Wales, Bangor LL57 2UW, UK 2 Department of Biological Sciences, University of Hull, Hull HU6 7RX, UK *j.rock@bangor.ac.uk Abstract: We analysed cytochrome oxidase I (COI) barcodes for 35 putative fish species collected in the Scotia Sea, and compared the resultant molecular data with field-based morphological identifications, and additional sequence data obtained from GenBank and the Barcode of Life Data System (BOLD). There was high congruence between morphological and molecular classification, and COI provided effective species-level discrimination for nearly all putative species. No effect of geographic sampling was observed for COI sequence variation. For two families, including the Liparidae and Zoarcidae, for which morphological field identification was unable to resolve taxonomy, DNA barcoding revealed significant species-level divergence. However, the dataset lacked sufficient sensitivity for resolving species within the Bathydraco and Artedidraco genera. Analysis of cytochrome b for these two genera also failed to resolve taxonomic identity. The data are discussed in relation to emergent priorities for additional taxonomic studies. We emphasize the utility of DNA barcoding in providing a valuable taxonomic framework for fundamental population studies through assigning life history stages or other morphologically ambiguous samples to parental species. Received 19 July 2007, accepted 4 December 2007 Key words: COI, cyt b, Liparidae, Myctophidae, Notothenioidei, Zoarcidae Introduction The Scotia Sea encompasses a broadly dispersed arc of islands from Shag Rocks and South Georgia that lie in close proximity to the Polar Front, to the south-easterly distributed South Sandwich Islands, the high latitude South Orkney Islands and more westerly South Shetland Islands lying near the Antarctic Peninsula. Due to the broad latitudinal range of the Scotia Sea, as well as the influence of two distinct hydrographic regimes, including the Antarctic Circumpolar Current in the north and Weddell Sea convergence in the south, the region encompasses both low- and high-Antarctic fish fauna. Taxonomic exploration of fish species in the Scotia Sea began as early as the 19th century, though studies until recently have been heavily biased towards species around South Georgia (reviewed in Kock & Jones 2005). Commercial harvesting of fish in this region did not begin until 1977 but by the mid-1980s management of the resource was undertaken through The Convention on the Conservation of Antarctic Marine Living Resources (CCAMLR). Both research and fishing ventures have yielded in-depth survey data on the species composition of fish within the Scotia Sea. The majority of shelf species (77–92%), and indeed the greatest biomass (95% of all individuals) belong to the suborder Notothenioidei (Kock & Jones 2005). This assemblage includes five families, all of which are well represented in the Scotia Sea: Nototheniidae, Channichthyidae, Harpagiferidae, Artedidraconidae and Bathydraconidae. However, we currently do not have a measure of the accuracy of species identifications using conventional morphological criteria. Species identification is a major fisheries issue and has been targeted, for example, by the Food and Agriculture Organization (FAO) programme since the 1960s. Aggregate data from 1950–2002 shows that up to 33% of the fisheries catch failed to be identified to species (Lleonart et al. 2006). In fact, species resolution in capture fisheries statistics has been declining, apparently influenced by the imprecision in the statistics from south-east Asia (Lleonart et al. 2006). Indeed, the distribution of our taxonomic knowledge of fish is highly biased among taxa and regions of the globe. Whereas some regions benefit from centuries of research and catch statistics, leading to a solid taxonomic knowledge, other areas such as the Southern Ocean have only recently received commercial and/or scientific attention. The classification of taxonomically cryptic fish and fish products is of particular importance where sampling access is at a premium. There can be few more difficult or costly regions to access than the Southern Ocean and the opportunity to fish in these waters is further constrained by the fact that they are seasonally difficult to navigate and are also largely uncharted for depth and bottom features. Southern Ocean waters are thereby typically unworkable for many traditional fishing methods such as bottom trawling, thus limiting the amount of scientific data that can be collected. 253 Antarctic Science 20 (3), 253–262 (2008) & Antarctic Science Ltd 2008 Printed in the UK doi: 10.1017/S0954102008001120 https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0954102008001120 Downloaded from https://www.cambridge.org/core. IP address: 207.241.231.80, on 25 Jul 2018 at 00:42:16, subject to the Cambridge Core terms of use, available at