Comparison of scale and otolith age readings for trahira, Hoplias malabaricus (Bloch, 1794), from Parana´ River, Argentina By I. E. Lozano 1 , S. Llamazares Vegh 1 , A. A. Doma´nico 2 and A. Espinach Ros 1 1 Direccio ´n de Pesca Continental, Ministerio de Agricultura Ganader  ıa y Pesca, Ciudad Auto ´noma de Buenos Aires, Argentina; 2 Comisi  on de Investigaciones Cient  ıficas y T ecnicas, La Plata, Buenos Aires, Argentina Summary The aim of this work was to compare age determinations and precision using two deposition structures of trahira Hoplias malabaricus (Bloch, 1794): the scales, which are most fre- quently used, and otoliths (lapilli). The length-age relation- ships were obtained with both structures and compared with results from previous studies. The 163 sets of trahira otoliths (lapilli) and scales were 17–46 cm standard length (SL) from Cayast a (Santa Fe) and Islas Lechiguanas (Entre R  ıos), Paran a River, Argentina. Three independent readings of each structure were conducted. An age bias plot was performed to compare age estimations from scales and otoliths. To assess the precision of age determinations using both structures, the percent agreement among readers for both structures and the coefficient of variation were calculated (%CV). The age-length relationships were plotted and fitted with the von Bertalanffy growth function for both structures and compared with previ- ous works. Age readings recorded for scales were lower than those recorded for otoliths for ages above or equal to 3 years. Percent agreement among readers was higher than 80% for otoliths and less than 65% for scales. The%CV obtained for scales was 20% for young fish and 17% for adults. For oto- liths the %CV was 7% for young fish and 3% for adults. The %CV obtained for scales was over the recommended limit (>7.6%). The von Bertalanffy parameters for scales were L inf = 45.80 mm; k = 0.29; t 0 = À1.34 and for otoliths were L inf = 40.76 mm; k = 0.39; t 0 = À1.05. Precision of age esti- mations assessed from the percent agreement and the coeffi- cient of variation indicates that the scales of the trahira are inappropriate to estimate age in population studies for juvenile and adult specimens. Introduction Accurate age estimates are fundamental for an appropriate fisheries management. Age determination allows calculation of growth rate, mortality rate and productivity (Campana, 2001). Fish retain a record of growth in several anatomical structures such as: scales, otoliths and bones (Chambers and Miller, 1995). The process of estimating fish age incorporates two major sources of error (Campana, 2001) (i) a process error associated with the structure being examined; not all bony structures in fish form a complete growth sequence throughout the lifetime of the animal, nor do all axes within a given structure show a complete growth record (Beamish, 1979), and (ii) error due to the element of subjectivity required for all age estimations. Historically, the major part of previously published studies of age and growth of species from South America fisheries were made with scales. In the case of trahira there were no published otolith studies (Ram  ırez, 1963; Paiva, 1974; Barbieri, 1989; Dom anico et al., 1993; Dom anico, 1998; Grosman et al., 2004; Tordecilla-Petro et al., 2005; Balboni et al., 2011). The trahira (Hoplias malabaricus) (Bloch, 1794) is an ich- thyophagous species with a wide geographical distribution of 11°N – 35°S; 85° – 35°W (Fowler, 1950; Reis et al., 2003). In South America (Teresa et al., 2010; Carvalho et al., 2011; Pedroza et al., 2012; Volcan et al., 2012), the species inhabits most of the hydrographic basins except those to the west of the Andes and in Patagonia. The species is frequently caught for human consumption in freshwater lagoons of Argentina (Grosman et al., 2004), Colombia (Bentancur-V asquez et al., 2004) and Uruguay (Crossa, 1994; Amestoy, 2001). In Argentina, the trahira occurs in lotic environments (lower and mid Paran a River, lower Uruguay River and pampasean rivers and streams) where it is caught for commercial and recreational purposes (Espinach Ros and Sanchez, 2007). The objective of the present work is to compare age deter- minations and precision using two deposition structures of trahira: scales, which are most frequently used, and otoliths (lapilli). Length-age relationships obtained with both struc- tures and compared with results obtained in previous studies are also analyzed. Materials and methods Specimens were caught in the Paran a River (Argentina) in the framework of Proyecto de Evaluaci on del Recurso S abalo en el rio Paran a, between Cayast a (Santa Fe prov- ince) and Islas Lechiguanas (Entre R  ıos province), Argentina. Fish were captured between March 2007 and August 2009, whereby 163 sets of otoliths (lapilli) and scales from 17 to 46 cm standard length (SL) trahira were ana- lyzed. Young fish were defined as smaller than 23.16 cm SL for males and 21.53 cm SL for females (pers. comm.). Two otoliths and 10 to 20 scales were taken from each specimen. Selected for this analysis were lapilli otoliths that present some advantages with respect to the sagittae otoliths, which are too small, fragile and difficult to find. The asteri- scii otoliths are the largest but are also fragile and cannot withstand the sanding procedure. Scale readings followed the criterion outlined by Ram  ırez (1963): ten or more scales per fish were removed from the left flank (between dorsal and pectoral fin), rinsed in water with a soft brush and mounted between glass slides with the external side up. They were then placed under a stereoscopic microscope (10X) illuminated from below for the estimation J. Appl. Ichthyol. (2013), 1–4 © 2013 Blackwell Verlag GmbH ISSN 0175–8659 Received: May 11, 2013 Accepted: June 6, 2013 doi: 10.1111/jai.12317 Applied Ichthyology Journal of