Baseline demographics of a non‐native lake trout population and inferences for suppression from sensitivity‐elasticity analyses

Management agencies in several western states of the United States are implementing suppression programmes to control non‐native lake trout, Salvelinus namaycush (Walbaum), for the conservation of native species. This study was implemented to ascertain the population demographics of an expanding lake trout population and use those data to construct an age‐structured model to inform suppression efforts. Population projection matrices were used to model population growth and identify age or stage classes with the greatest influence on population growth. The size and age structure of lake trout sampled was skewed towards juveniles, indicating strong recruitment and a growing population. Matrix‐model simulations corroborated the observed size and age structure, as the lake trout population was predicted to grow exponentially (λ = 1.35, 95% CL: 1.25–1.43) with no suppression efforts. Elasticity analysis of matrix models indicated the relative contribution of survival rates to population growth among immature age classes was equal from age 0 to age at first maturity, but immature survival rates contributed more than adult survival and fertility rates. These results emphasise the importance of targeting juvenile lake trout for suppression efforts during exponential growth in recently established populations.     

Seasonal movements of non‐native lake trout in a connected lake and river system

Abstract Non‐native lake trout, Salvelinus namaycush (Walbaum), threaten native salmonid populations in the western United States. Effective management of lake trout requires understanding movements within connected lake and river systems. This study determined the seasonal movements of subadult lake trout in the Flathead River upstream of Flathead Lake, Montana, USA using radio telemetry. The spatiotemporal distribution of lake trout in the river was related to water temperature. Lake trout were detected in the river primarily during autumn, winter and spring, when water temperatures were cool. By contrast, fewer were detected when temperatures were warmest during summer and during high spring flows. Downriver movements to Flathead Lake occurred throughout autumn and winter when water temperature decreased below 5 °C, and in late spring as water temperature rose towards 15 °C and river discharge declined following spring runoff. Upriver movements occurred primarily in October, which coincided with migrations of prey fishes. These results suggest that lake trout are capable of moving throughout connected river and lake systems (up to 230 km) and that warm water temperatures function as an impediment to occupancy of the river during summer. Controlling source populations and maintaining natural water temperatures may be effective management strategies for reducing the spread of non‐native lake trout.     

Lake trout (Salvelinus namaycush) otoliths indicate effects of climate and lake morphology on growth patterns in Arctic lakes

Climate change is occurring rapidly in the Arctic, and an improved understanding of the response of aquatic biota and ecosystems will be important for this data-limited region. Here, we applied biochronology techniques and mixed-effects modelling to assess relationships among growth increments found on lake trout (Salvelinus namaycush) otoliths (N = 49) captured from 13 lakes on the Arctic Coastal Plain of northern Alaska, observed and modelled climate patterns, and individual-level fish and lake characteristics. We found that annual growth varied by year, fish growth slowed significantly as individuals aged, and females grew faster than males. Lake trout had higher growth in flow-through lakes relative to lakes that were perennially or seasonally connected. Annual growth was positively correlated with observed air temperature measurements from a local weather station for the period 1998–2014, but no clear warming trend was evident for this period. Modelled August air temperatures from 1978–2014 predicted lake trout annual growth (root mean squared error = 0.045 mm) and indicated increasing temperatures and annual lake trout growth over the period 1950–2014. This study demonstrated that biochronology techniques can reconstruct recent climate patterns and provide a better understanding of trends in Arctic lake ecosystems under a changing climate.     

Modelling individual variability in growth of bull trout in the Walla Walla River Basin using a hierarchical von Bertalanffy growth model

We examined growth in length of fluvial bull trout (Salvelinus confluentus) in the Walla Walla River Basin, Washington and Oregon. Our objectives were to quantify individual variability in growth; examine growth within and among years, life history forms, life stages and sexes; and estimate von Bertalanffy growth parameters. Individual variability was evaluated by modelling asymptotic length (L_∞) and the growth coefficient (k) as random variables. All models were fit with Bayesian methods and were evaluated for fit by the deviance information criterion. By incorporating individual variability, population‐level estimates of L_∞ and k appeared appropriate and estimated growth trajectories for specific bull trout fit individual observed patterns in growth. Growth trajectories and positive correlation between individual estimates of L_∞ and k suggest that some individuals grow at a faster rate and reach a larger maximum size than other individuals and those differences are maintained throughout life. Selected models suggest that fluvial migrants have higher estimates of L_∞ and k than residents, but there were only slight differences in parameter estimates among migrants from two adjacent spawning populations in the Walla Walla River Basin, as well as between males and females. Growth rates increased for fluvial migrants after subadult emigration. Individual variability in growth is consistent with the life history diversity assumed essential for bull trout population persistence. Quantifying this variability is important for modelling population dynamics and viability to conserve this threatened species.     

Developing bioelectrical impedance analysis methods for age‐0 brook trout

Year class strength of many fishes often is determined by survival through the first winter. Increased fat reserves improve survival and overall cohort success. Bioelectrical impedance analysis (BIA) methods are established for estimating proximate composition of adult brook trout, Salvelinus fontinalis (Mitchill), but none have been developed for early life stages. Small‐fish BIA would provide information about age‐0 percent dry weight, a proxy for fat reserves, allowing for better prediction of cohort success. The objective of this study was to develop BIA methods that provide reliable estimates of percent dry weight for age‐0 brook trout. BIA measurements were taken at seven anatomical locations from 48 to 115 mm fish. A model developed using BIA measures taken by subdermal needle electrodes precisely predicted percent dry weight (best model, RMSE = 1.03, R² = 0.86). Although lacking support, as determined by information theoretical analysis, BIA measured with non‐invasive external rod electrodes also precisely predicted percent dry weight (RMSE = 1.09, R² = 0.85). Models developed using two electrode locations performed better than models developed with only one location. For small brook trout, a dorsal to ventral pre‐dorsal fin electrode location should be used in conjunction with a dorsal total length location when measuring BIA to predict percent dry weight.     

The effects of a prolonged photoperiod and light source on growth,sexual maturation,fin condition,and vulnerability to fungal disease in brook trout Salvelinus fontinalis

The effects of an artificially prolonged photoperiod on growth, survival rate, colouration, and sexual maturation in brook trout Salvelinus fontinalis during pre‐ and post‐spawning periods from 21 June to 06 November were investigated. Fish of mean initial weight ?150 g were reared at ambient photoperiod as well as with an artificially prolonged photoperiod produced by either a light‐emitting diode or a metal‐halide light. The fish groups subjected to a prolonged period of artificial light grew significantly larger and had a higher survival rate (p < 0.05), regardless of sex, and showed lower occurrence of fungal disease compared to controls reared in the natural photoperiod. We found a significantly higher number of sexually mature fish in the control groups compared with experimental groups. The increased photoperiod effectively delayed gonad development and increased somatic growth in both male and female brook trout, and also increased resistance to fungal disease. The increased photoperiod produced an observable difference in fish colouration, with control groups exhibiting more intensive spawning colouration.     

Stable isotopes and gut content show diet overlap among native and introduced piscivores in a large oligotrophic lake

Abstract – In past dietary studies kokanee Oncorhynchus nerka were prominent in the diet of Pend Oreille Lake's large piscivores: native bull trout Salvelinus confluentus, cutthroat trout O. clarki and northern pikeminnow Ptychocheilus oregonensis, and introduced lake trout S. namaycush and Kamloops rainbow trout O. mykiss gairdneri. However, kokanee have declined to 10–20% of their former abundance. We therefore initiated this study to understand current predation demands on kokanee and diet overlap among piscivores, using gut content samples and analysis of stable nitrogen (δ¹⁵N) and carbon (δ¹³C) isotopes from the lake's fish and invertebrate community. In gut content samples, kokanee were the main prey item of large [i.e., ≥400 mm total length (TL)] bull and lake trout; a conclusion that was affirmed by stable isotope analysis. Rainbow trout >500 mm TL consumed mostly kokanee, thus there was a high degree of diet overlap among large bull, lake and rainbow trout. Small (i.e., <400 mm TL) rainbow and cutthroat trout diets overlapped, and were composed mostly of littoral benthic invertebrates. However, gut content and stable isotope analysis did not accord for 400–500 mm TL rainbow trout, small lake trout, and large cutthroat trout. In these instances, a linear mixing model using stable isotope results predicted kokanee consumption for each species, but no kokanee were identified in rainbow or lake trout gut content. Gut content and stable isotope analysis of native northern pikeminnow indicated a diet of mostly littoral benthic invertebrates at smaller (100–150 mm TL) lengths, with kokanee becoming more prominent in the diet of individuals >300 mm TL. Percent of kokanee in the diet of northern pikeminnow has declined from a prior study; otherwise piscivore diets have apparently remained unchanged. In this study, judgments as to the feeding of some piscvores, based on gut content alone, would be tenuous because of small sample sizes, but stable isotope analysis provided an efficient means for confirming diets.     

Assessing the food web impacts of an anadromous Arctic charr introduction to a sub‐Arctic watershed using stable isotopes

Anadromous Arctic charr, Salvelinus alpinus (L.), was introduced to a sub‐Arctic river–lake system near the village of Kujjuuaq, Nunavik, and the stable isotope values and diets of key resident fish species were used to assess changes in feeding patterns. Stable isotope values for most species did not differ significantly between the pre‐ and post‐introduction periods, with observed shifts being within the bounds of expected natural variation. Lake chub, Couesius plumbeus (Agassiz), were the single species to show a difference between study periods, with a small but significant increase in δ¹⁵N. No significant post‐introduction changes were seen in lake trout, Salvelinus namaycush (Walbaum), omnivory or in any of the assessed quantitative food web metrics. Gut contents of major fish species similarly showed significant temporal overlap between the pre‐ and post‐introduction periods, and there was no significant change in species' weight–length relationships. The minor ecological impact was interpreted in relation to the availability of open niches exploitable by ecological generalists such as Arctic charr. The explanation accords with the known habitat and feeding flexibility of Arctic charr and the ecological immaturity of sub‐Arctic lakes known to have driven adaptive variation among Arctic charr. Findings suggest that anadromous Arctic charr may be introduced at moderate densities to other sub‐Arctic watersheds without major negative food web consequences for other resident fish species.     

Stable isotope evaluation of population‐ and individual‐level diet variability in a large,oligotrophic lake with non‐native lake trout

Non‐native piscivores can alter food web dynamics; therefore, evaluating interspecific relationships is vital for conservation and management of ecosystems with introduced fishes. Priest Lake, Idaho, supports a number of introduced species, including lake trout Salvelinus namaycush, brook trout S. fontinalis and opossum shrimp Mysis diluviana. In this study, we used stable isotopes (δ¹³C and δ¹⁵N) to describe the food web structure of Priest Lake and to test hypotheses about apparent patterns in lake trout growth. We found that isotopic niches of species using pelagic‐origin carbon did not overlap with those using more littoral‐origin carbon. Species using more littoral‐origin carbon, such as brook trout and westslope cutthroat trout Oncorhynchus clarki lewisi, exhibited a high degree of isotopic niche overlap and high intrapopulation variability in resource use. Although we hypothesised that lake trout would experience an ontogenetic diet shift, no such patterns were apparent in isotopic signatures. Lake trout growth rates were not associated with patterns in δ¹⁵N, indicating that variation in adult body composition may not be related to adult diet. Understanding trophic relationships at both the individual and species levels provides a more complete understanding of food webs altered by non‐native species.     

Habitat overlap of juvenile and adult lake trout of Great Bear Lake: Evidence for lack of a predation gradient?

A range of organisms, from plankton to fish, commonly shift their habitat distributions horizontally or vertically due to predation risk. Juvenile lake trout, Salvelinus namaycush, are generally viewed as occupying deep areas of lakes to decrease predation pressure from adults. In contrast, we found that juvenile lake trout from Great Bear Lake, NT, Canada, occupied a variety of habitats and from shallow to deep depths (0–150 m), overlapping with adult lake trout. No evidence occurred for a length depth‐based segregation (e.g., ontogenetic shift). Genetic variation was also similar among juveniles in the different depth zones. However, isotopic niches and C:N ratios among juveniles showed some variability in niche widths and positions for individuals caught from the 51–150 m zone compared to juvenile individuals caught from 0–20 m and 21–50 m zones. The uniformly distributed adult lake trout in Great Bear Lake may evenly distribute predation pressure (including cannibalism) across shallow‐ and deep‐water habitats more than in other lakes. As a result, juveniles may respond to differences in foraging opportunities rather than predation risks. Juvenile lake trout did not appear to conform to the general pattern of juveniles seeking a deep‐water refuge to reduce predation risks. In contrast, juvenile lake trout of Great Bear Lake displayed broad resource use across all depths and habitats.     

Trophic interactions between introduced lake trout (Salvelinus namaycush) and native Arctic charr (S. alpinus) in a large Fennoscandian subarctic lake

Introduced fishes may have major impacts on community structure and ecosystem function due to competitive and predatory interactions with native species. For example, introduced lake trout (Salvelinus namaycush) has been shown to replace native salmonids and induce major trophic cascades in some North American lakes, but few studies have investigated trophic interactions between lake trout and closely related native Arctic charr (S. alpinus) outside the natural distribution of the former species. We used stomach content and stable isotope analyses to investigate trophic interactions between introduced lake trout and native Arctic charr in large subarctic Lake Inarijärvi in northern Finland. Both salmonids had predominantly piscivorous diets at >280 mm total length and were mainly caught from the deep profundal zone. However, lake trout had a more generalist diet and showed higher reliance on littoral prey fish than Arctic charr, whose diet consisted mainly of pelagic planktivorous coregonids. According to length at age and condition data, lake trout showed slightly faster growth but lower condition than Arctic charr. The results indicate that introduced lake trout may to some extent compete with and prey upon native Arctic charr, but currently have only a minor if any impact on native fishes and food web structure in Inarijärvi. Future monitoring is essential to observe potential changes in trophic interactions between lake trout and Arctic charr in Inarijärvi, as well as in other European lakes where the two salmonids currently coexist.     

Diel microhabitat use of age-0 bull trout in Indian Creek, Washington

Abstract – We studied diel microhabitat use at the focal point of age‐0 bull trout, Salvelinus confluentus, in Indian Creek, Washington during mid‐summer and fall of 1997. Microhabitat variables included water depth and velocity, distance from the stream bottom, habitat and refuge use, substrate type, and substrate embeddedness. Age‐0 fish were located over fines and gravel substrates in shallow, low‐velocity water near stream margins, but were located in shallower water at night. Bull trout were highly associated with loose substrate, and used the substrate interstices for refuge cover. Diurnal bull trout counts decreased and no age‐0 fish were observed after 15 September at water temperatures below 6.1 °C. Nocturnal counts remained relatively constant throughout the study. Our results suggest that age‐0 bull trout surveys be conducted at night when summer water temperatures begin to decline.     

Reliance on lakes by salmon,trout and charr (Oncorhynchus,Salmo and Salvelinus): An evaluation of spawning habitats,rearing strategies and trophic polymorphisms

Salmonid fishes may reside within or migrate between stream and lake habitats, or undergo anadromous migrations between freshwater and the ocean. While the degree of anadromy of salmonids has been thoroughly compared, no analogous review has examined the degree of lake use. To assess the extent of reliance on lake habitat in this family, we considered 16 species of salmon, trout and charr from the genera Oncorhynchus, Salmo and Salvelinus, comparing their (a) use of lakes as spawning habitat, (b) rearing strategies in lakes, and (c) occurrence and diversity of lacustrine trophic polymorphism. In identifying the primary life‐history patterns of each species and exploring the lesser‐known lacustrine behaviours, we found that the extent of reliance on lakes exhibits a negative association with the degree of anadromy. Oncorhynchus rely least on lakes, Salmo to an intermediate level and Salvelinus the most, opposite of the general prevalence of anadromy among these genera. Lakes are critical to adfluvial and lake‐resident salmonids, but they also support anadromous and fluvial life histories by providing spawning, rearing, overwintering and/or summer refuge habitat. Adfluviality, although a non‐anadromous life history, consists of similar migration‐related traits and behaviours as anadromy, including the parr–smolt transformation, sex‐biased patterns of migration and residency, and the presence of precocious males. Lakes support life‐history variants, reproductive ecotypes and trophic morphs unique to lacustrine habitat. Therefore, conservation of salmonids is dependent on maintaining the diversity and quality of their habitats, including lakes.     

Population‐level effects of egg predation on a native planktivore in a large freshwater lake

Using a 37‐year recruitment time series, we uncovered a field pattern revealing a strong, inverse relationship between bloater Coregonus hoyi recruitment success and slimy sculpin Cottus cognatus biomass in Lake Michigan (United States), one of the largest freshwater lakes of the world. Given that slimy sculpins (and deepwater sculpin Myoxocephalus thompsonii) are known egg predators that spatiotemporally overlap with incubating bloater eggs, we used recently published data on sculpin diets and daily ration to model annual bloater egg consumption by sculpins for the 1973–2010 year‐classes. Although several strong year‐classes were produced in the late 1980s when the proportion of eggs consumed by slimy sculpins was extremely low (i.e., <0.001) and several weak year‐classes were produced when the proportion of bloater eggs consumed was at its highest (i.e., >0.10–1.0), egg predation failed to explain why recruitment was weak for the 1995–2005 year‐classes when the proportion consumed was also low (i.e., <0.02). We concluded that egg predation by slimy and deepwater sculpins could have limited bloater recruitment in some years, but that some undetermined factor was more important in many other years. Given that slimy sculpin densities are influenced by piscivorous lake trout Salvelinus namaycush, the restoration of which in Lake Michigan has lagged behind those in lakes Superior and Huron, our study highlights the importance of an ecosystem perspective when considering population dynamics of fishes.     

Does habitat occupancy by lake trout and lake whitefish in large lakes match published thermal habitat envelopes?

Habitat occupancy patterns of lake trout (Salvelinus namaycush) and lake whitefish (Coregonus clupeaformis) in two large Canadian Shield lakes were modelled based on detections of fish from repeated depth‐stratified surveys over several summers. Lake trout and lake whitefish consistently occupied sites outside traditional thermal envelopes and were not detected at some sites within these ranges. This included the metalimnion and shallow epilimnion for lake trout and lake whitefish in Lake Opeongo. Physical habitat covariates were not important in defining lake trout habitat in both lakes. Physical habitat as represented by the hardness/softness gradient based on acoustic substrate surveys was important for lake whitefish in Lake Opeongo but not in Smoke Lake. In addition, thermal envelopes for lake whitefish differed between the lakes possibly because of differences in substrate slope. The wash zone of lakes, where the thermocline contacts the substrate, appears to be a physical habitat feature for lake whitefish in some lakes. Lake whitefish also exhibited diurnal activity behaviour that was reflected through greater detection rates in the morning versus the afternoon. By accounting for imperfect detection, true estimated overall occupancy of lake trout and lake whitefish increased 0.15–0.30 over naïve occupancy. Thermal habitat envelopes for lake trout and lake whitefish are wider than previously thought. Lake trout occupied a consistent thermal habitat envelope while lake whitefish varied between lakes likely because of lake specific differences in basin morphology and wash zone.     

The effect of timing of extended photoperiod on growth and maturity of brook trout (Salvelinus fontinalis)

The influence of timing of extended photoperiods on growth and maturity of brook trout was investigated in a 112‐day experiment. The fish with mean initial weight of ~192 g were reared under four light regimes: one control group with natural ambient photoperiod and three groups exposed to an 18L:6D regime initiated at days 1, 23 or 46 of the growth trial. Light‐emitting diodes, with intensity of 250–1000 lux, depending on the distance from the light source, were used for extending light periods. There was a positive effect of prolonged day length on fish growth (p < 0.05), and a delay in gonad development and sexual maturity. Significantly higher numbers of sexually mature fish were found among controls groups, regardless of sex. Survival rate was not affected by light regime. This study demonstrated that the short‐term expansion of the photo period delayed maturation and increased the growth rate of brook trout.     

Bioenergetic evaluation of diel vertical migration by bull trout (Salvelinus confluentus) in a thermally stratified reservoir

Many species living in deeper lentic ecosystems exhibit daily movements that cycle through the water column, generally referred to as diel vertical migration (DVM). In this study, we applied bioenergetics modelling to evaluate growth as a hypothesis to explain DVM by bull trout (Salvelinus confluentus) in a thermally stratified reservoir (Ross Lake, WA, USA) during the peak of thermal stratification in July and August. Bioenergetics model parameters were derived from observed vertical distributions of temperature, prey and bull trout. Field sampling confirmed that bull trout prey almost exclusively on recently introduced redside shiner (Richardsonius balteatus). Model predictions revealed that deeper (>25 m) DVMs commonly exhibited by bull trout during peak thermal stratification cannot be explained by maximising growth. Survival, another common explanation for DVM, may have influenced bull trout depth use, but observations suggest there may be additional drivers of DVM. We propose these deeper summertime excursions may be partly explained by an alternative hypothesis: the importance of colder water for gametogenesis. In Ross Lake, reliance of bull trout on warm water prey (redside shiner) for consumption and growth poses a potential trade‐off with the need for colder water for gametogenesis.     

Fine‐scale acoustic telemetry reveals unexpected lake trout,Salvelinus namaycush,spawning habitats in northern Lake Huron,North America

Previous studies of lake trout, Salvelinus namaycush, spawning habitat in the Laurentian Great Lakes have used time‐ and labour‐intensive survey methods and have focused on areas with historic observations of spawning aggregations and on habitats prejudged by researchers to be suitable for spawning. As an alternative, we used fine‐scale acoustic telemetry to locate, describe and compare lake trout spawning habitats. Adult lake trout were implanted with acoustic transmitters and tracked during five consecutive spawning seasons in a 19–27 km² region of the Drummond Island Refuge, Lake Huron, using the VEMCO Positioning System. Acoustic telemetry revealed discrete areas of aggregation on at least five reefs in the study area, subsequently confirmed by divers to contain deposited eggs. Notably, several identified spawning sites would likely not have been discovered using traditional methods because either they were too small and obscure to stand out on a bathymetric map or because they did not conform to the conceptual model of spawning habitat held by many biologists. Our most unique observation was egg deposition in gravel and rubble substrates located at the base of and beneath overhanging edges of large boulders. Spawning sites typically comprised <10% of the reef area and were used consistently over the 5‐year study. Evaluation of habitat selection from the perspective of fish behaviour through use of acoustic transmitters offers potential to expand current conceptual models of critical spawning habitat.     

Otolith size differences during early life of dwarf and cannibal Arctic char (Salvelinus alpinus)

In Arctic and alpine lakes, Arctic char [Salvelinus alpinus (L.)] often form two distinct morphs: invertebrate feeders (‘dwarfs’) and piscivores (‘giant or cannibals’). Here, we test for early life history growth variation in dimorphic Arctic char as a proximate explanation for the observed life history variation between the two forms. Char were sampled in 11 alpine and Arctic Norwegian lakes. Dwarfs (defined as sexually mature char less than 15 cm long; N = 304) had a mean total length of 105 mm, whereas the typical cannibal (body length above 20 cm; N = 153) was 388 mm long. A positive correlation between egg size and otolith hatching ring were ascertained in a separate hatching experiment with brown trout (Salmo trutta) and it is assumed that this relationship also is valid for Arctic char, and otolith size was used as a proxy for length. Otolith hatching ring from Arctic char cannibals was larger (mean ± SD; 187 ± 24 μm) than those from dwarfs (mean ± SD; 164 ± 23 μm). There were only minor size differences between dwarfs and cannibals during the next three years, after which dwarfs usually matured. Two mutually nonexclusive, proximate explanations for the differentiation into separate morphs (dwarfs and cannibals) are therefore maternal effects and/or genetic based differentiation. The high catchability of large piscivorous char and low production in alpine and Arctic lake ecosystems may make these stocks particularly vulnerable to overexploitation.     

Alternative routes to piscivory: Contrasting growth trajectories in brown trout (Salmo trutta) ecotypes exhibiting contrasting life history strategies

Large and long‐lived piscivorous brown trout, Salmo trutta, colloquially known as ferox trout, have been described from a number of oligotrophic lakes in Britain and Ireland. The “ferox” life history strategy is associated with accelerated growth following an ontogenetic switch to piscivory and extended longevity (up to 23 years in the UK). Thus, ferox trout often reach much larger sizes and older ages than sympatric lacustrine invertebrate‐feeding trout. Conventional models suggest that S. trutta adopting this life history strategy grow slowly before a size threshold is reached, after which, this gape‐limited predator undergoes a diet switch to a highly nutritional prey source (fish) resulting in a measurable growth acceleration. This conventional model of ferox trout growth was tested by comparing growth trajectories and age structures of ferox trout and sympatric invertebrate‐feeding trout in multiple lake systems in Scotland. In two of the three lakes examined, fish displaying alternative life history strategies, but living in sympatry, exhibited distinctly different growth trajectories. In the third lake, a similar pattern of growth was observed between trophic groups. Piscivorous trout were significantly older than sympatric invertebrate‐feeding trout at all sites, but ultimate body size was greater in only two of three sites. This study demonstrates that there are multiple ontogenetic growth pathways to achieving piscivory in S. trutta and that the adoption of a piscivorous diet may be a factor contributing to the extension of lifespan.