Reproductive biology of the Cisalpine pike (Esox cisalpinus Bianco & Delmastro, 2011) in Trasimeno Lake

The Cisalpine pike Esox cisalpinus is an endemic species to Northern and Central Italy, recently separated from the transalpine populations of northern pike Esox lucius L. 1758, based on morphology and genetics (Bianco and Delmastro, 2011; Lucentini et al., 2011). The information about the biology of E. cisalpinus is very limited and the trends of population are still unknown. Hence, currently the species is “Not Evaluated” according to the IUCN Red List of Threatened Species and it is listed as “Data Deficient” in the Italian IUCN Red List (Rondinidi et al., 2013). In the Trasimeno Lake (43°9’11’’ N and 12°15’E), the largest laminar lake in Italy, E. cisalpinus is one of the few native species of the fish community, mainly dominated by non-native species (Pompei et al., 2012). A strong reduction of the abundances of E. cisalpinus in the Trasimeno Lake occurred in the recent years primarily due to different factors relating to non-native species introduction: competition for food with largemouth bass Micropterus salmoides (Lacépedè, 1802), negative interaction with Carassius sp. and hybridization with E. lucius (Lorenzoni et al., 2002; Lorenzoni and Ghetti, 2011). Even the degradation of breeding areas contributed to the E. cisalpinus decline and actually its reproduction is widely supported by controlled spawning. The aims of this study were to provide more information on reproductive biology of the population of E. cisalpinus inhabiting Trasimeno Lake by analyzing some reproductive traits such as Gonadosomatic Index, fecundity, egg size and their temporal trend over a period of 5 years. Females of E. cisalpinus were collected from 2010 to 2014 (except 2013) during the reproductive period (February-March) by the staff of Ichthyogenic Centre of Trasimeno Lake. All fish were measured (TL, cm) weighted (W, g) and a sample of scales was removed for age determination (Bagenal and Tesch 1985). The total weight of eggs squeezed from each female was measured (Wg). Gonadosomatic Index (GSI) was calculated as: GSI=100(Wg/W). A subsample of eggs was weighted (Wgn) and the eggs in the subsample were counted (Ns) to obtain fecundity (F=NsWgn-1). Relative Fecundity (RF) was expressed as the number of eggs produced (F) per unit of body weight: RF=1000(FW-1). The diameters of 10 randomly selected eggs from each ovary were measured by means of the image processing program software ImageJ and the mean value (δ, cm) for each female was used in the subsequent analysis. In order to analyse whether F, RF and δ varied with the increase size, the following regressions were calculated: log10F=a+b log10TL; RF=a+bTL; δ=a+bTL. The mean annual values of F, RF and δ were compared by Analysis of Covariance (ANCOVA) (using TL as covariate mean); moreover variations in these parameters over time were evaluated using linear regression analysis. Body condition was determined using Relative Weight Wr=100(W’/Ws) (Murphy et al., 1991), where W’ is the somatic weight ad Ws is the Standard Weight calculated as follows: log10Ws=– 5.37+3.06 log10TL (Willis, 1991). Variations in condition over time were evaluated using a linear regression analysis. Total sample consisted of 292 females ranging in size from 28 to 103 cm (2+ -14+ age classes). Descriptive statistic of GSI, F, FR, δ and Wr is reported in Table 1. Fecundity was positively correlated with TL (r2=0.67, r=0.82, p<0.01) (Fig. 1A); a positive significant relationship emerged between diameter of eggs and size of females (r2=0.16; r=0.40; p<0.01) (Fig. 1B), whereas relative fecundity significantly decreased with the increasing of fish size (r2=0.03, r=‐0.17; p<0.05) (Fig. 1C), even if low variance is justified by the model. No differences emerged in F among years at ANCOVA (F=2.29, p>0.05; covariate mean: TL=61.03 cm). On the contrary, RF varied remarkably from year to year (ANCOVA: F=3.57, p<0.05; covariate mean: TL=61.03 cm), without a clear temporal trend in the regression analysis (r2<0.01; r=-0.02; p=0.74). The great variability of RF values should be justified by the high unpredictability of environmental condition (i.e. water level, water temperature) that characterize the Trasimeno Lake (Lorenzoni et al. 2002). The mean diameter of eggs showed a marked decrease from 2010 to 2012, with a slight increase in 2014 (ANCOVA: F=48.56, p<0.01; covariate mean: TL=61.03 cm); however the regression analysis highlights a general decrease over time (r2=0.03; r=-0.17, p<0.01) (Fig. 2A). A similar pattern was revealed from the regression between Wr and years (r2=0.02; r=-0.15, p<0.01) (Fig. 2B). In many fish species, egg size, which correlates with female size or condition, has been found to contribute to egg quality (Kamler, 1992). The reduction of the eggs diameter ultimately results in the quality of the offspring, since the amount and composition of the yolk have a great influence on the development of the larvae (Laine and Rajasilta, 1999). Condition of spawners is known to have significant effects on reproductive potential (Laine and Rajasilta, 1999), but it also influences the viability of eggs (Kjørsvik et al., 1990), thus the decrease in eggs size may be correlated to the worsening of condition during years that seems to affect E. cisalpinus in Trasimeno Lake. The worsening of both condition and quality of the eggs make necessary the continuation of the monitoring study to safeguard the conservation of the population. The preliminary results reported in this research increase the knowledge on the biological characteristic of E. cisalpinus and should assist decision-making in relation to the future management of this endemic species