3377 Ecology, 86(12), 2005, pp. 3377–3386  2005 by the Ecological Society of America LATITUDINAL VARIATION IN LIFE-HISTORY TRAITS IN EURASIAN PERCH ERIK HEIBO, 1,3 CARIN MAGNHAGEN, 1 AND LEIF ASBJøRN VøLLESTAD 2 1 The Swedish University of Agricultural Sciences (SLU), Department of Aquaculture, SE-901 83 Umea ˚, Sweden 2 University of Oslo, Department of Biology, Centre for Ecological and Evolutionary Synthesis, P.O. Box 1066 Blindern N-0316 Oslo, Norway Abstract. Few studies have examined multiple life-history traits across a latitudinal gradient to test whether variation in growth rate and mortality schedules induces trends predicted by life-history theory. We collected data for the following life-history traits for 75 Eurasian perch (Perca fluviatilis) populations: growth coefficient (K ) and asymptotic body length (L  ) from the von Bertalanffy growth model, size at ages one and two years, specific juvenile growth rate, instantaneous adult and juvenile mortality rates, life span, age and length at maturity, and reproductive life span and investment. All life-history traits except L  were significantly correlated with latitude. In general, growth rates, mortality rates, and reproductive investment decreased with latitude, whereas age at maturity, size at maturity, and life span increased with latitude. Populations could be grouped into two categories based on variation in L  : stunted (small sized) vs. piscivorous (large sized). Four trait–latitude relationships differed between these two types: the growth coefficient ( K ) and the juvenile growth rate were larger, and age and length at maturity were lower in the stunted populations compared with piscivorous populations. Perch from southern popula- tions tend to grow fast and experience high juvenile and adult mortality rates. As predicted from life-history theory, this selects for an early age and small size at maturity and relatively large investment in reproduction. The opposite pattern was found for northern populations. Key words: latitudinal cline; life-history variation; trade-offs; Perca fluviatilis L.; perch. INTRODUCTION In endothermic animals, such as birds and mammals, body size increases from south to north (Bergmann’s rule) (Ashton et al. 2000, Ashton 2002). Recently, it has also been documented that some ectotherms, such as squamate reptiles (but not chelonian reptiles), follow Bergmann’s rule (Ashton and Feldman 2003). Teleost fishes have also been reported to adhere to this rule (Lindsey 1966). However, for a number of North Amer- ican freshwater fish species, no such trend is found (Belk and Houston 2002). Instead, growth rate clearly decreases with latitude (Beverton 1987, Jonsson and L’Abe ´e-Lund 1993, Lobo ´n-Cervia ´ et al. 1996, Belk and Houston 2002). Most laboratory studies indeed show that ectothermic animals grow faster at high temper- atures, and that they exhibit smaller body sizes. This trend is due to phenotypic plasticity, and is often called the temperature–size rule (Angilletta and Dunhan 2003). Further, fish longevity has been shown to in- crease from south to north, suggesting that mortality rate decreases (Colby and Nepszy 1981, Beverton 1987, Gunderson and Dygert 1988). In addition, age at maturity increases with latitude (Colby and Nepszy 1981, Vøllestad 1992, Jonsson and L’Abe ´e-Lund 1993), whereas reproductive investment decreases Manuscript received 25 October 2004; revised 25 May 2005; accepted 31 May 2005. Corresponding Editor: K. O. Winemiller. 3 E-mail: erik.heibo@vabr.slu.se from south to north (Leggett and Carscadden 1978, Mann et al. 1984, Fleming and Gross 1990). Variation in life-history traits with latitude can be explained by phenotypic plasticity, or may reflect ge- netically determined adaptive differences caused by spatial variation in local selective pressures (Roff 2002). Life-history theory, in general, predicts that an increase in adult mortality will select for earlier mat- uration and increased reproductive effort, while an in- crease in juvenile mortality will select for the opposite (Gadgil and Bossert 1970, Schaffer 1974, Reznick et al. 1990, Hutchings 1993). Recently, life-history models have shown that small changes in environmental conditions can lead to abrupt changes in optimal life histories when size-dependent mortality is sufficiently strong (Taborsky et al. 2003). For example, such transitions may be associated with either early maturation and short inter-brood intervals or late maturation and long reproductive cycles. Thus, as environmental conditions change along a latitudinal gradient, we might not necessarily expect clinal vari- ation in other life-history traits if mortality is size de- pendent. Negative size-selective mortality (on small individuals) may select for phenotypes growing rapidly into a ‘‘size refuge’’ (i.e., large size; Taborsky et al. 2003). In fish, this may be achieved by postponing reproduction. When size-selective mortality on large individuals becomes sufficiently strong, an alternative