Using DNA barcodes to connect adults and early life stages of marine fishes from the Yucatan Peninsula, Mexico: potential in fisheries management Martha Valdez-Moreno A,C , Lourdes Va ´squez-Yeomans A , Manuel Elı ´as-Gutie ´rrez A , Natalia V. Ivanova B and Paul D. N. Hebert B A El Colegio de la Frontera Sur, Avenida Centenario km 5.5, Chetumal 77014, Quintana Roo, Me ´xico. B Biodiversity Institute of Ontario, University of Guelph, Guelph, Ontario, N1G 2W1, Canada. C Corresponding author. Email: mvaldez@ecosur.mx Abstract. Barcoding has proven a useful tool in the rapid identification of all life stages of fish species. Such information is of critical importance for fisheries management and conservation, especially in high-diversity regions, such as Mexico’s marine waters, where more than 2200 species occur. The present study reports the barcode analysis of 1392 specimens from the Yucatan Peninsula, corresponding to 610 adults and juveniles, 757 larvae and 25 eggs, representing 181 species (179 teleosts and 2 rays), 136 genera and 74 families. Barcoding results revealed major range extensions and overlooked taxa, including three sympatric species of Albula (one likely undescribed) and a new taxon of Floridichthys. In total, six species of eggs and 34 species of larvae were identified through their barcode match with adults. These cases enabled the first discrimination of the larvae of four species of Eucinostomus, and new information about spawning locality and time was obtained from egg records for the hogfish, Lachnolaimus maximus, which is one of the most commercially important species in the Mexican Caribbean. Also, barcodes revealed mistakes in species recognition during a sport-fish contest. In the future, barcodes will help avoid similar errors and protect rare or endangered species, and will aid regulation of fisheries quotas. Additional keywords: COI, coxI, Cytochrome c oxidase, identification, mitochondrial DNA. Introduction Since the pioneering study by Ward et al. (2005) on DNA barcodes of Australian fishes, more than 20 fish-barcoding studies have been published, confirming the effectiveness of this technique in the identification of fishes in different regions. For example, Rock et al. (2008) analysed 35 fish species from the Scotia Sea, reporting that COI (cytochrome oxidase subunit I gene) provided effective species-level discrimination that was unaffected by broad geographic sampling. Schlei et al. (2008) reported similar results in their study of eight coregonine species with distributional overlap in Alaska. Barcode compar- isons of fish populations from North Atlantic, Mediterranean and Australasian waters showed no significant spatial genetic differentiation in 13 of 15 species, whereas two other species included two highly divergent barcode clusters, suggesting that they were each likely to be a species pair (Ward et al. 2008a). In a study examining patterns of sequence divergence among populations of 35 major commercial fish species from the opposite sides of the Indian Ocean (Australia, South Africa), Zemlak et al. (2009) concluded that current taxonomic systems substantially underestimate species diversity. Steinke et al. (2009) used DNA barcodes to examine 391 fish species important in the ornamental fish trade and also reported several cases of overlooked species. Of other genes, Cytochrome b has been the DNA marker most used in fish, proving highly reliable for species identification (Teletchea 2009). However, currently more than 7205 species of fish have been barcoded with the COI (Barcode of Life Database (BOLD), www.boldsystems.org, verified 24 November 2009). Aside from their role in species discovery and identification, DNA barcodes will play an increasingly important role in the identification of fish larvae and eggs. Egg-based identifications can play a major role in fisheries management (Fox et al. 2005). The early stages of most fish are unknown, and the larvae of different species are often quite similar (Leis and Carson- Ewart 2000; Richards 2006; Baldwin et al. 2009). Furthermore, larvae are so delicate that they are often badly damaged during collection, and fixation regularly causes the loss of pigmentation patterns important for identification (Smith 1995). An obvious solution to these problems involves the identification of egg and CSIRO PUBLISHING www.publish.csiro.au/journals/mfr Marine and Freshwater Research, 2010, 61, 665–671 Ó CSIRO 2010 10.1071/MF09222 1323-1650/10/060665