UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Physiological basis of growth-performance trade-offs : insights from different strains of rainbow trout Allen, David William 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2015_february_allen_david.pdf [ 1.12MB ]
JSON: 24-1.0135623.json
JSON-LD: 24-1.0135623-ld.json
RDF/XML (Pretty): 24-1.0135623-rdf.xml
RDF/JSON: 24-1.0135623-rdf.json
Turtle: 24-1.0135623-turtle.txt
N-Triples: 24-1.0135623-rdf-ntriples.txt
Original Record: 24-1.0135623-source.json
Full Text

Full Text

PHYSIOLOGICAL BASIS OF GROWTH-PERFOMANCE TRADE-OFFS: INSIGHTS FROM DIFFERENT STRAINS OF RAINBOW TROUT  by David William Allen  B.Sc., McGill University, 2006  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2014  © David William Allen, 2014 ii  Abstract  Growth rate is a fundamental life history trait common to all taxa. Despite its importance, the underlying physiological mechanisms and constraints associated with growth remain poorly understood. The purpose of my research was to explore the physiological correlates and trade-offs associated with high growth. I examined a suite of physiological variables related to growth including metabolic rate, digestive capacity, and tissue energy content. Three strains of juvenile rainbow trout (Oncorhynchus mykiss) were chosen based on known differences in ecology and growth. Fish from each strain were assigned to one of three food ration treatments: (i) satiation over eight hours every day, (ii) fed 1% of body mass over eight hours every day, and (iii) complete deprivation of food. A wide range of growth rates were observed within and across all strains. I found that fast growing rainbow trout (hatchery strain) had higher standard metabolic rates and lower maximum metabolic rates and aerobic scopes, suggesting that high growth rate results in a reduced capacity to do metabolic work. I also found that trout with high growth rates, generally, had larger gastrointestinal tracts, higher maximum food consumption rate, and higher growth efficiency. Lipid content and water content had opposing correlations with growth rate; larger individuals with higher growth rates had a higher body lipid content, while water content was highest amongst smaller individuals. I show that there is a suite of physiological traits that correlate with growth rate,  that these traits are affected by both genotype (strain) and environment (food ration), and that these traits appear to be consistently traded off against other correlates of performance.  iii  Preface The research presented in this thesis was designed, carried out, and analyzed by myself, the author, David Allen. Guidance and advice regarding hypotheses, experimental design, data analysis, and writing edits was kindly provided by Dr. Jordan Rosenfeld and Dr. Jeffrey Richards. This project was funded by an NSERC discovery grant awarded to Dr. Jordan Rosenfeld. All experimental protocols presented in this thesis were approved by UBC’s Animal Care Committee under Animal Use Protocol #AUP A09-0611. iv  Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables ............................................................................................................................... vii List of Figures ............................................................................................................................. viii List of Abbreviations ................................................................................................................... ix Acknowledgements ........................................................................................................................x Chapter 1: General Introduction .................................................................................................1 1.1 Rainbow trout in British Columbia: social context and ecological variation among natural and hatchery stocks ......................................................................................................... 1 1.2 Ecological context: why study growth and what limits it? ............................................. 2 1.3 Bioenergetics of growth .................................................................................................. 4 1.4 Metabolic processes affected by growth ......................................................................... 5 1.5 Organ-system trade-offs and digestive strategies to maximise growth .......................... 7 1.6 Trade-offs ........................................................................................................................ 9 1.7 Domestication and its effect on rainbow trout .............................................................. 12 1.8 Objectives and predictions ............................................................................................ 12 Chapter 2: Physiological basis of growth-performance trade-offs: insights from different strains of rainbow trout ...............................................................................................................14 2.1 Introduction ................................................................................................................... 14 2.2 Materials and methods .................................................................................................. 18 v  2.2.1 Experimental design.................................................................................................. 18 2.2.2 Maximum metabolic rate .......................................................................................... 21 2.2.3 Standard metabolic rate............................................................................................. 22 2.2.4 Gastrointestinal tract morphology............................................................................. 23 2.2.5 Lipid and water content ............................................................................................ 24 2.2.6 Statistical analysis ..................................................................................................... 25 2.3 Results ........................................................................................................................... 26 2.3.1 Specific growth rate (SGR) ....................................................................................... 26 2.3.2 Standard metabolic rate (SMR) ................................................................................ 27 2.3.3 Maximum metabolic rate (MMR) ............................................................................. 28 2.3.4 Aerobic scope (AS) ................................................................................................... 28 2.3.5 Whole body water content ........................................................................................ 29 2.3.6 Whole body lipid content .......................................................................................... 29 2.3.7 Gastrointestinal (GI) tract morphology ..................................................................... 29 2.3.8 Maximum food consumption .................................................................................... 30 2.3.9 Growth efficiency ..................................................................................................... 31 2.3.10 Hierarchical partitioning ....................................................................................... 31 2.3.11 Seasonal effect ...................................................................................................... 31 2.4 Discussion ..................................................................................................................... 32 2.4.1 Physiological trade-offs associated with rapid growth ............................................. 32 2.4.2 What is the physiological basis of rapid growth? ..................................................... 37 2.4.3 Relative magnitude of genetic (fixed) vs. flexible (plastic) effects on growth and metabolism ............................................................................................................................ 40 vi  2.4.4 Conclusions, caveats, and implications .................................................................... 41 Chapter 3: General Discussion ...................................................................................................52 3.1 Physiological mechanisms that promote high growth rate ........................................... 52 3.2 How metabolic rates relate to growth rate .................................................................... 54 3.3 Physiological trade-offs associated with growth .......................................................... 54 3.4 General caveats ............................................................................................................. 57 3.5 General conclusions and applications ........................................................................... 58 Bibliography .................................................................................................................................60 Appendix A: Data Tables ............................................................................................................69 A.1 Fraser Valley strain data ............................................................................................... 69 A.2 Pennask Lake strain data ............................................................................................... 70 A.3 Tzenzaicut Lake strain data .......................................................................................... 71  vii  List of Tables  Table 1.  Parameter estimates using AICc model selection for standard metabolic rate, maximum metabolic rate, aerobic scope, specific growth rate, lipid content, water content, growth efficiency, gastrointestinal tract size, and maximum food consumption for three different strains of rainbow trout............................................................................................................................. 43 Table 2.  Means, sample sizes, and standard deviations of anterior intestine length and stomach mass............................................................................................................................................... 44 Table 3.  Hierarchical partitioning estimates of the independent and joint contribution that strain, ration, and mass have on variation in specific growth rate, standard metabolic rate, maximum metabolic rate, aerobic scope, lipid content, water content, growth efficiency, gastrointestinal tract size, and maximum food consumption ................................................................................. 45   viii  List of Figures  Figure 1. The relationship between specific growth rate and mass, standard metabolic rate and mass, maximum metabolic rate and mass, and aerobic scope and mass for three different strains of rainbow trout on three different food rations............................................................................ 46 Figure 2. The relationship between body water content and mass, body lipid content and mass, and gastrointestinal tract size and mass for three different strains of rainbow trout on three different food rations..................................................................................................................... 48 Figure 3. The relationship between maximum food consumption and mass, and growth efficiency and mass for three different strains of rainbow trout .................................................................... 49 Figure 4. Hierarchical partitioning of the variance explained by strain, ration, and mass on specific growth rate, standard metabolic rate, maximum metabolic rate, aerobic scope, lipid content, water content, gastrointestinal tract mass, and growth efficiency for three different strains of rainbow trout on three different food rations ................................................................ 50 Figure 5. The relationship between specific growth rate and mass for 3 different strains of rainbow trout housed in groups of 5 to 6 ...................................................................................... 51    ix  List of Abbreviations  ANOVA Analysis of variance ANCOVA Analysis of co-variance AS  Aerobic scope GI  Gastrointestinal RMR  Routine metabolic rate SDA  Specific dynamic action SGR  Specific growth rate SMR  Standard Metabolic Rate MMR  Maximum Metabolic Rate FV  Fraser Valley rainbow trout strain Tz  Tzenzaicut Lake rainbow trout strain Pn  Pennask Lake rainbow trout strain H  Satiation ration M  Mid (1 % body mass day-1) ration S  Starvation/fasting ration   x  Acknowledgements I owe quite a bit to Dr. Jordan Rosenfeld for his patience, advice, and feedback. I am very grateful to have had such an approachable adviser who always had the time to meet with me and answer my questions. Jordan’s guidance and encouragement was invaluable to me. I will not soon forget the hours spent together in a cramped office discussing papers. I had the good fortune of having a second adviser, Dr. Jeffrey Richards. There was no cruising through a meeting with Jeff; every meeting was intellectually challenging and rewarding. I am indebted to Jeff for providing mentorship and advice when it was needed most. I had a lot of help taking care of my fish. I would like to thank Taylor Gibbons, Andrew Thompson, Christine Verhile, and Patrick Tamkee, without whom I would have gone crazy during my experiments. My lab mates from both the Rosenfeld and Richards lab could not have provided a better support group. Lab meetings, and lab outings were always a pleasure. I was never lacking for friends or for help. Dr. Eric Taylor, my final committee member, was helpful in providing input into the design of the experiments, and in giving editorial feedback and encouragement. A great deal of help with R and statistical analysis came from Dr. Sam Yeaman, and Dr. Thor Veen, both of whom were happy to help while at UBC, or while out skiing. Early on, there was a field component to my thesis. Campfires and trapping fish on the Sunshine Coast and Vancouver Island with Seth Rudman, Justin Lenze, Luke Trevan, and Anne Cayer-Huard will forever be a fond memory for me. Lastly I would like to thank Mrs.Vivien Allen, my very talented editor/sister-in-law who provided me with a great deal of editorial feedback. 1  Chapter 1: General Introduction  1.1 Rainbow trout in British Columbia: social context and ecological variation among natural and hatchery stocks Sport fishing in British Columbia (BC) is a major tourist attraction and revenue generator for the province. In response to high fishing pressure, the Freshwater Fisheries Society of BC (FFSBC) stocks around 800 lakes with over eight million salmonid fishes, principally rainbow trout (Oncorhynchus mykiss), cutthroat trout (O. clarkii) and brook char (Salvelinus fontinalis) annually. Stocked rainbow trout come from roughly eight genetically distinct strains that have been shown to differ in their responses to environmental and social challenges. Because lakes vary in their local conditions (e.g., temperature, productivity, pH, etc.) the FFSBC attempts to stock each lake with a stain that will be compatible with the environment. The FFSBC gets most of its brood stock from the wild each year, which means that many of its stocks are subject to minimal domestication selection. However, fully domesticated stocks of rainbow trout (i.e., adult brood stock maintained in the hatchery through many generations) do exist and are characterised by rapid growth, bold behaviour, and increased mortality upon release when exposed to predation (Biro et al. 2004a).  Populations of wild rainbow trout vary in their ecology and feeding habits. Some populations are piscivorous, some are insectivorous, while others remain opportunistic generalists. Different feeding ecology, including quality and abundance of prey, leads to differences in a number of physical attributes, including, but not limited to, growth rate and maximum body size. In this thesis I examine the variation in growth rate and a suite of associated physiological attributes amongst distinct populations of rainbow trout under varying levels of 2  food availability. In an effort to span a range of growth rates, three populations of rainbow trout were selected for this study; (i) a domesticated strain, Fraser Valley domestics, which are known to have high growth rates and to be opportunistic feeders upon release; (ii) Tzenzaicut Lake “wild” strain, which are known to have a moderate growth rate and to be piscivorous; (iii) and Pennask Lake “wild” strain, which have a relatively slow growth rate and are largely insectivorous.  1.2 Ecological context: why study growth and what limits it? Growth rates differ between taxa, species, populations, and individuals. Growth rate is an emergent life history trait that is determined, in large part, by local adaptation to both biotic and abiotic processes such as food availability, foraging risk, competition, temperature, and seasonal time constraints. Both absolute growth rate and environmental plasticity in growth are subject to local selection that may differentiate species and populations. Although growth represents a key life history trait that varies among individuals, populations, and species, and is subject to strong selection, the factors that influence growth are poorly understood. Growth also exerts a strong influence on energy metabolism through ontogeny, and as a result differences in growth rate have a large effect on standard metabolic rate (SMR) in juvenile fish. Consequently, my thesis focuses on the physiological mechanisms whereby juvenile fish achieve high growth rates and the physiological performance attributes and potential trade-offs associated with high growth. Life history theory often groups life into two stages: pre and post reproductive maturity (Stearns 1992). High juvenile growth rate is beneficial because it can minimise the time where an animal is exposed to risk of mortality before being able to reproduce. In fishes, juvenile growth can determine age and size at maturity, for example slow growing animals often mature at larger 3  sizes which offers greater fecundity (Gotthard 2001). In contrast, faster growers often mature at a smaller size limiting time spent pre-reproduction. Age and size at maturity is a fundamental life history trait and by studying juvenile growth trajectories we may develop a better understanding of the evolution of life history patterns. Another advantage of studying juvenile growth is that the juvenile life stage is characterised by the highest growth rates; therefore, it is at this stage where differences in growth should be most pronounced. Resource variation in the wild means that realised growth rate is often below maximal growth (Arendt 1997; Nylin & Gotthard 1998; Gotthard 2001). Balancing selection acts to produce growth rates that are optimal (typically intermediate) in any given ecological context, but generally well below the physiological maximum that could be achieved under artificial selection (Dmitriew 2011). For example, high growth rates are often associated with elevated mortality due to increased time spent foraging under threat of predation (Werner & Hall 1988; Werner & Anholt 1993; Abrahams & Sutterlin 1999), which leads to an optimal growth rate that is the result of maximising the fitness costs and benefits associated with foraging and growth.  The period of growth for animals in temperate environments is also constrained by the duration of their growing season. Often animals are faced with seasonal minimum size thresholds that are essential to maintain adequate fitness (e.g. a minimum overwintering size, or a minimum size required to reach sexual maturation for the breeding season). Atlantic silversides (Menidia menidia), for instance, are found up and down the east coast of North America; northern populations have a much shorter growing season (the time period over which water temperatures are suitable for growth) than their southern counterparts, yet they achieve the same overwintering size. Common garden experiments have shown that northern populations achieve higher growth rates when fed ad libitum due to increased food consumption and conversion efficiency 4  (Billerbeck, Schultz & Conover 2000). Atlantic salmon smolts have also demonstrated variation in growth across a latitudinal gradient (Nicieza, Reyes-Gavilán & Braña 1994b). In both cases, the northern populations have a shorter growing season, which is compensated for by increasing growth (an example of countergradient variation (Levins 1968)). 1.3 Bioenergetics of growth Growth can be modeled using a simple bioenergetic approach (modified from Nisbet et al. 2012): 𝐺 = 𝐶 −𝑀𝐴 − 𝑆 − 𝐹 − 𝑈                                                              (1) where G is growth, C is food consumption, M is metabolism due to maintenance, A is a dimensionless factor for activity, S is the cost of food digestion (specific dynamic action; SDA), F is egestion, and U is excretion. Another way to look at it is that the energy assimilated from food is equal to the summation of energy used by all metabolic processes, or that input is equal to output. A bioenergetic framework is fundamental to modelling whole-organism physiology and growth energetics; however, it also provides a useful framework for understanding the intersection of ecology, energetics, and adaptation.  Individuals or species may adaptively differentiate by partitioning consumed energy differently between activity costs and growth; or organisms may differentiate based on absolute food consumption.  By increasing food consumption (input), the energy available for all other processes including growth and activity also increases. For example, steelhead trout (rainbow trout with an anadromous life cycle) and coho salmon (O. kisutch) often inhabit the same streams, but due to their earlier hatching date, juvenile coho salmon are larger and therefore able to dominate pool habitats that are characterised by slow flowing water (Hartman 1965). Steelhead trout, are forced into fast 5  flowing riffle habitat where energy expenditure is higher, but so is the exposure to food (Young 2001; Grossman et al. 2002; Rosenfeld et al. 2014a). Steelhead trout adopt an energy maximising strategy whereby they eat more food, but spend more energy on metabolism (Van Leeuwen, Rosenfeld & Richards 2011). In this case, the local ecology (interspecific competition) drives physiological traits (metabolism) that are governed by energetic input and demand. 1.4 Metabolic processes affected by growth Equation 1 illustrates the sensitivity of growth to the suite of factors on the right hand side of the equation.  A given growth rate can, in principle, be achieved through any combination of changes in food consumption, resting metabolism, active metabolism, or digestion efficiency. However, it is unrealistic to think that all strategies to achieve growth variation are equally feasible. The key question, from a physiology-ecology perspective, is whether or not changes in growth are achieved through consistent covaration among specific components of the energy budget that also correspond with particular adaptive trade-offs along a discrete continuum of physical and ecological traits.  Within this context, I consider how the different metabolic compartments of equation 1 influence, or are influenced by, growth. Standard metabolic rate (SMR) is the lowest metabolic rate that an animal can exhibit, and is synonymous with basal metabolic rate (BMR) in endotherms. Standard metabolic rate is typically measured as the rate of oxygen consumed by a resting animal and represents the summation of a broad suite of physiological processes that have been shown to vary between species, populations, and individuals. Standard metabolic rate can account for a large percentage of the energy budget (Finstad et al. 2007) and is indirectly implicated in the metabolic cost of growth (Wieser 1994): 6  SMR is by definition maintenance metabolism, and is associated with processes of cellular maintenance, protein turnover, and any other bodily function not directly associated with physical activity or growth. Metabolism scales allometrically to be higher in juveniles primarily because of the overhead costs of maintaining a larger digestive tract to facilitate growth (Rosenfeld et al. 2014b). Similarly, elevated ration may trigger a plastic upregulation of digestive tract size and activity, leading to elevated metabolism in growing fish; strictly speaking, these are additional overhead costs of growth rather than an elevation in SMR, which cannot be measured in growing animals. Here I refer to SMR measured in growing animals as apparent SMR (Rosenfeld et al. 2014b). High SMR has also been correlated with high levels of aggression which can promote growth when food can be monopolised (Ryer & Olla 1996; Cutts, Metcalfe & Caylor 1998), but which is also energetically wasteful when food is absent or cannot be monopolised (Vøllestad & Quinn 2003). Maximum metabolic rate (MMR) is a key performance diagnostic that represents the maximum amount of aerobic energy that an animal can produce. Maximum metabolic rate is ultimately limited by mitochondrial density and the body’s ability to deliver oxygen to the mitochondria. Maximum metabolic rate represents the high end of an animal’s aerobic limit, and it is correlated with maximum swimming speed and other performance measures (Van Leeuwen et al. 2011). Maximum metabolic rate is measured in fish most often using a ramped swimming test where a fish is placed in a swim tunnel and the flow rate is increased until the fish can no longer keep up with the flow. One drawback of this method is that it is time intensive unless fish are swum in groups, in which case there are other confounding factors including drafting and social interactions. Another commonly used method is to use a chase protocol (see Chapter 2 methods). While it can be argued that a chase protocol does not give true MMR, but rather an 7  exhaustive metabolic rate, it has been shown that chase protocols can yield metabolic rates higher than those attained from swim protocols (Reidy et al. 1995) and may in fact be more ecologically relevant. Aerobic scope (AS) is the rate of aerobic energy production available for an animal to do work and it is calculated as the difference between MMR and SMR. Aerobic scope is correlated with performance measures like maximum prolonged swimming speed (Reidy, Kerr & Nelson 2000; Van Leeuwen et al. 2011). Aerobic scope decreases when either MMR is decreased or SMR is increased, which makes AS sensitive to growth rate and SDA. This highlights the potential for trade-offs among components of the energy budget and the anatomical and ecological features that underlie them. For instance, increased growth may come at the cost of decreased swimming ability, increasing the vulnerability of faster-growing individuals or populations to predation (Billerbeck, Lankford & Conover 2001; Biro et al. 2004a). 1.5 Organ-system trade-offs and digestive strategies to maximise growth Food varies in its availability, quality, and ease of handling. It is well-established in ecology that specializing on one prey resource or foraging mode may involve behavioural or morphological adaptations that preclude exploiting other resources effectively; there can also be corresponding differences in digestive strategy (Rosenfeld et al. 2014b). The digestive strategy that is employed by an animal will depend on the energy and nutrient density of the prey that it is adapted to consuming, and on the costs of prey acquisition including but not limited to foraging effort, risk of predation, and inter- or intra-species competition. Maximum food consumption capacity is a key dimension differentiating digestive strategies, and also may underlie differences in growth rate.  For example, domesticated rainbow trout fed a set ration via automatic feeders have been shown to grow at the same rate as wild rainbow trout. However, when fed ad libitum 8  these same domesticated fish grew faster than wild trout (Valente et al. 2001).  The ability to consume large quantities of food when it is available can be important as well, because food is spatially and temporally heterogeneous which favours the ability of an animal to capitalise on unpredictable food pulses (Armstrong & Schindler 2011). An increased ability to store excess food when the opportunity arises can be the difference between a positive and negative energy budget, so retaining the ability to process quantities of food that surpass the daily average is beneficial (Armstrong & Schindler 2011), but can also be energetically costly due to the high metabolic maintenance costs of digestive tissue (Karasov, Martínez del Rio & Caviedes-Vidal 2011). A large stomach for example would be beneficial because it gives an animal the ability to store excess food without forcing a shorter gut passage time. Food quality has two defining attributes: energy content and bulk (Orlando, Brown & Whelan 2009).  Low quality food is defined as having a set energy benefit, but a lot of bulk, as opposed to a high quality food that would have the same set energy benefit, but with very little bulk. Bulk affects the digestive strategy by filling up the gastrointestinal (GI) tract, limiting the amount of food that can be further ingested. A strategy to get around the negative effects of bulky food is to speed up passage time. By shortening the passage time, an animal can eat more of the low quality food while stripping off only the most easily absorbed, labile nutrients. For this strategy to work, the low quality resource must by plentiful. Yellow-rumped warblers (Dendroica coronata) provide a good example of this strategy. A Yellow-rumped warbler will modify its food passage time depending on the quality of the food available to it (Afik & Karasov 1995). When presented with a high bulk, low quality food (fruit), the Yellow-rumped warbler adopts a fast food passage time that is associated with decreased extraction efficiency. 9  When acclimated to a high quality diet (seeds), the Yellow-rumped warbler adopts a long food passage time and an associated high extraction efficiency (Afik & Karasov 1995). Digestive efficiency is the  amount of energy or nutrients absorbed from a given amount of food and it can be improved by increasing the rate of digestion (hydrolysis) and/or the rate of absorption (nutrients leaving the lumen of the GI tract to enter the blood stream), and/or by lengthening food passage time (mouth to anus transit time) (Orlando et al. 2009). Digestion happens primarily in the stomach and is aided by digestive enzymes and stomach contractions that mix the stomach contents (Carey, Kanwisher & Stevens 1984). The rate of absorption can be increased by increasing the areas of the GI tract where absorption occurs, namely the small intestine for most animals, as well as the pyloric caeca in fish (Grosell, Farrell & Brauner 2010).  Bergot, Blanc & Escaffre (1981), showed a positive correlation between the number of pyloric caeca in rainbow trout and growth rate, and Stevens & Devlin (2005) found that growth hormone transgenic coho salmon with a high growth rate had an increased surface area of the lumen of their pyloric caeca; the pyloric caeca are important for nutrient absorption, and seem to play an important role in rapid growth of salmonids. Lastly, increasing the time available for digestion and absorption to occur (the passage time) will also increase digestive efficiency. 1.6 Trade-offs Although all fish share a common basic life history (eggs, larvae, juvenile, and adult), there is huge variation in adult body size, fecundity, longevity, and size at age (growth) among populations and species.  This variation reflects trade-offs among life history traits and indicates that the combination of traits that maximizes fitness depends on environment and prey type, among other constraints (e.g. phylogeny, etc.). For example, there are trade-offs associated with high juvenile growth rate. If there were no such trade-offs then growth would be maximised at 10  all times in order to minimise time taken to reach reproductive maturity. Perhaps the best documented trade-off with high growth is the increased risk of predation associated with increased time spent foraging (Houston, McNamara & Hutchinson 1993; Ali, Nicieza & Wootton 2003; Biro et al. 2004a).  Swimming performance in fish has been suggested to trade-off with high growth rates. In the case of the Atlantic silverside, where countergradient variation in growth rate was found, it was observed that northern populations had reduced swim performance (Billerbeck et al. 2001). Both burst and prolonged swimming speeds were reduced. The same was seen in hatchery-reared salmonids where selection for high growth rates is associated with reduced swim performance (Duthie 1987; Magnan et al. 2009; Van Leeuwen et al. 2011).  Starvation resistance, the ability to resist weight loss or death during times of food scarcity, has been shown to be negatively correlated with growth rates (Dupont-Prinet et al. 2010). Two mechanisms have been proposed to explain this: (i) high growth rate requires a high metabolic rate, which will burn up energy reserves faster, and (ii) fast growers invest less energy in storage tissues with high energy content; when faced with starvation they will use somatic tissue for energy (Gotthard 2001).  Damselflies (Lestes viridis) that have been manipulated to have high growth rates exhibit both of these mechanisms, they have higher metabolic rates that burn up their reduced energy stores quickly, i.e., they exhibit a trade-off between energy allocation to storage versus structural growth (Stoks, Block & McPeek 2006). Similarly, Drosophila strains that have been selected for starvation resistance show reduced growth rates and increased lipid levels (Chippindale, Chu & Rose 1996). European sea bass (Dicentrarchus labrax) with high growth rates are less starvation tolerant than European sea bass with low growth rates (Dupont-Prinet et al. 2010). However, in this case metabolic rates of starvation 11  resistant individuals were not found to be different from fast growers (storage content was not tested). Digestion rate can be the limiting factor for energy/nutrient acquisition (Booth 1990; Hart & Gill 1992). When food is in excess, it can be advantageous to increase food consumption and decrease passage time in order to attain more labile nutrients in a shorter period of time. Higher food consumption and shorter passage time are often associated with lower extraction efficiency (Afik & Karasov 1995), which would in effect decrease growth efficiency, while at the same time maximise nutrient/energy assimilation under high throughput (Millidine, Armstrong & Metcalfe 2009). A trade-off between growth rate and growth efficiency is likely to exist amongst populations that are adapted to high food availability. The metabolic machinery required for high growth rate, in particular the GI tract, is metabolically expensive to build and to maintain (Martin & Fuhrman 1955; Cant, McBride & Croom 1996). The GI tract is so expensive to maintain that some animals, such as snakes, actually allow the lining of the lumen to atrophy between feedings (Secor, Stein & Diamond 1994; Secor & Diamond 2000), resulting in a lower metabolic rate that conserves energy between intermittent feeding bouts. This strategy would not work for animals whose growth is limited by their rate of digestion, where maintaining digestive capacity is central to rapid growth. However, energy to invest in different organ systems is limited and, when energy is invested in organ systems that promote growth (e.g., GI tract, the liver, etc.), less energy will be available for other organ systems such as the cardiovascular or musculo-skeletal system. I suggest that aerobic scope will be limited amongst fast growing fish due to two underlying mechanisms: (i) an elevated SMR due to rapid growth, and maintenance of the machinery required for rapid 12  growth; and (ii) a proportionally lower investment in the cardiovascular system that will lead to a decreased MMR. 1.7 Domestication and its effect on rainbow trout Domesticated fish will not have the same selective pressures acting on them as fish in the wild. Being reared at high densities, without predation, in a controlled environment with abundant food will select for high growth rates, bold behaviour, and large adult body size associated with rapid juvenile growth trajectories (Johnsson & Abrahams 1991; Fleming et al. 2002). High growth rates will allow a fish to be more dominant, leading to high food consumption when housed in high densities. Bold behaviour is rewarded, as the fish that are not afraid of predation will be the first to the surface where most food is presented in hatcheries (Biro et al. 2004a). One attribute that will likely see weaker selection is the need for minimum energy content to survive overwintering. Swimming speed of rainbow trout has been found to be lower for domesticated strains (Magnan et al. 2009). Furthermore, it has been demonstrated that domesticated fish are poorer at predator avoidance than wild strains (Berejikian 1995). Domesticated fish provide an ideal model to study the trade-offs associated with selection for high growth since, in a controlled environment, many of the trade-offs normally associated with rapid growth are relaxed and will manifest without repercussion.  1.8 Objectives and predictions My thesis focuses on understanding the relationship between growth variation and trade-offs among metabolic performance attributes of different strains of rainbow trout.  The main objectives are to (i) understand how high growth rates are achieved by looking at digestive and bioenergetic strategies employed by domestic and wild strains of rainbow trout; and (ii) to 13  understand the nature of the adaptive trade-offs involved in growing quickly (i.e., in terms of low versus high metabolic rate, energy conversion efficiency, or investment in different organ systems). My specific predictions were that fast growing strain(s) and higher food ration treatment(s) would result in a high standard metabolic rate, a low maximum metabolic rate, and low growth efficiency which would be correlated with high maximum food consumption and a large mass-specific digestive tract.  I also predicted that the high growth strain(s) would show a trade-off between growth and starvation resistance; i.e. that high growth strain(s) would perform well when food is abundant but lose weight more quickly when deprived of food.  My predictions were that a domesticated rainbow trout strain would have the largest performance trade-offs (between high and low food availability), and that performance trade-offs in wild strains would exist to a lesser degree in proportion to their lower growth rate.  This work will help to better understand the metabolic constraints around increasing growth rate in both wild and domesticated populations. Not all possible combinations on a phenotypic landscape are realisable; I hope to provide insight into what combinations of traits and trade-offs are realistic in salmnoids, and potentially other taxa.  This will help better inform the trade-off axes that differentiate natural populations, and provide potential guidance for future breeding programs by informing the ecological consequences of selection for higher growth.  14  Chapter 2: Physiological basis of growth-performance trade-offs: insights from different strains of rainbow trout  2.1 Introduction Growth rate is a fundamental life-history trait that differentiates among species, populations, and individuals (Arendt 1997). Growth rates evolve in response to local biotic and abiotic selection pressures, leading to an optimal growth rate that confers maximum fitness (Arendt 1997; Dmitriew 2011) based on trade-offs between growth and other life-history traits. Maximising food intake versus minimising predation risk is a key behaviourally-mediated trade-off selecting against higher growth (Metcalfe, Huntingford & Thorpe 1987; Biro et al. 2004a) and results directly from increased predation associated with elevated foraging (Walters & Juanes 1993; Werner & Hall 1988; Jones & Rydell 1994). However, associated trade-offs often involve physiological components that, while equally important, are more cryptic and poorly understood. For example, it has been suggested that a large capacity for growth when food is abundant is associated with reduced resistance to starvation (measured as weight loss, or survival; Scharf, Filin & Ovadia 2009; Dupont-Prinet et al. 2010; Killen, Marras & McKenzie 2011) and that this is due to the increased metabolic demand of physiological and biochemical machinery associated with high growth (Bochdansky et al. 2005; Stoks et al. 2006; Scharf et al. 2009). Alternatively, a trade-off between starvation resistance and growth can also be based on investing in lipid stores versus  somatic tissue (Chippindale, Chu & Rose 1996; Gotthard 2001; Post & Parkinson 2001), independent of growth effects on maintenance costs during fasting.  15  In addition to potential trade-offs between growth and predation risk or starvation resistance, performance-related physiological trade-offs have also been implicated in fish. (Billerbeck et al. 2001; Arnott, Chiba & Conover 2006; Álvarez & Metcalfe 2007). Maximum metabolic rate (MMR) represents the peak oxygen consumption capacity of an individual, and is likely limited by the cardiovascular system’s ability to deliver oxygen to mitochondria in muscles. The difference between MMR and oxygen demand for maintenance metabolism (SMR) is aerobic scope (AS), which is the residual capacity of an organism to do metabolic work after maintenance costs are accounted for. High growth typically elevates metabolism (Wieser 1994), leading to reduced AS and a diminished potential for peak activity (i.e., reduced maximum swimming speed; (Van Leeuwen et al. 2011)). For example, selection for rapid growth to attain overwinter size thresholds in northern populations of Atlantic silverside (Menidia menidia) has evolved at the cost of reduced maximum swimming speed, leading to a trade-off between growth and vulnerability to predation (Billerbeck et al. 2001), independent of foraging duration. In addition, there is the potential for selection on growth to drive differences in anatomical trade-offs among organ systems (e.g. Kotrschal et al. 2013). A greater investment in an enhanced digestive system to support high growth may necessitate a relatively decreased investment in other organ systems (e.g. cardio-vascular, musculo-skeletal, cerebral, etc.) that translates into reduced swimming performance (Billerbeck et al. 2001). In general, these observations suggest that selection for higher growth may result in investment in organ systems that promote growth (gastro intestinal (GI) tract, liver, etc.) at the expense of organ systems that enhance cardiovascular performance or other functions. Aside from behavioural attributes that appear to increase net energy intake (e.g. increased foraging duration or reduced activity costs), physiological pathways to increased growth involve 16  increasing satiation capacity (maximum ration), and/or increasing growth efficiency (body mass gained per unit mass ingested (Valente et al. 2001). Higher food consumption is associated with increased gut capacity (Karasov et al. 2011; Armstrong & Schindler 2011) and evacuation rates (Afik & Karasov 1995). Increased growth efficiency on the other hand requires improved digestive performance (to extract more energy per unit food ingested), or lowering energy density of somatic tissue (Post & Parkinson 2001), or the energetic cost of tissue synthesis, to create more tissue per unit energy consumed (Nicieza, Reiriz & Braña 1994). Increased net digestive performance is typically associated with either a greater intensity of paracellular transport (Caviedes-Vidal et al. 2008), or increased size and internal surface area of the GI tract. Consequently, digestive tract size and activity often matches fixed genetic differences in growth and food consumption between fast and slow growing strains of rodents (Konarzewski & Diamond 1995; Sadowska, Gebczynski & Konarzewski 2013) or fish (Stevens & Devlin 2005), analogous to short-term plastic up-regulation of GI tract size and function in response to prey consumption (e.g., Secor et al. 1994). To test for metabolic, anatomical, and performance-related trade-offs associated with high growth, I contrasted growth and metabolism among three genetically distinct strains of rainbow trout that differ in their maximum capacity for growth and ecological origin. The fast growing strain (Fraser Valley, FV) is a domesticated hatchery strain selected for high growth and large adult body size (Freshwater Fisheries Society of BC 2004). Juveniles of this strain have been shown to tolerate higher predation risk at the cost of lower survival (Biro et al. 2004a) and were used as an exaggerated analogue of the higher growth strains that occur in nature. The two remaining strains (Tzenzaicut Lake, TZ, and Pennask Lake, PN) were hatchery raised, but from wild stock; Tzenzaicut Lake trout are largely lake-dwelling piscivores with intermediate growth, 17  and Pennask Lake trout are primarily lake-dwelling insectivores with the slowest growth of all three strains (Freshwater Fisheries Society of BC 2004). To assess whether strains differed in trade-offs related to growth, individuals of all three strains were reared on starvation, intermediate, and satiation rations for 21 days and traits associated with food consumption and energy flow (maximum ration, growth, growth efficiency), metabolic performance (SMR, MMR, AS), digestive system investment (gastrointestinal tract size), and tissue energy content (whole body lipid content, whole body water content) were measured throughout the experiment. I hypothesized that rainbow trout would manifest an integrated suite of trade-offs among metabolism, organ systems, and starvation resistance along a fast-slow metabolic continuum (Lovegrove 2003; Reale et al. 2010), and that fixed differences among strains would parallel the plastic metabolic responses of starved vs. satiated fish on the same metabolic continuum.  My main objectives were to (i) understand how high growth rates are achieved at a gross physiological level and to understand the consequences of selection on growth for emergent whole-body metabolism; (ii) to understand the nature of the adaptive trade-offs involved in growing quickly (i.e. in terms of low vs. high metabolic rate, energy conversion efficiency, or investment in different organ systems); and (iii) to compare the magnitude of genetic differences in growth among populations with the plastic phenotypic growth response to environmental variation (i.e. ration size). I make the case that selection on growth is associated with a suite of integrated physiological, anatomical, ecological, and performance-related trade-offs that represent one of the major adaptive axes differentiating individuals and populations (Reale et al. 2010; Careau & Garland 2012). Specific predictions were that the fastest growing strain and treatments would manifest a higher standard metabolic rate, a decreased maximum metabolic rate, a lower growth efficiency associated with high maximum food consumption, and the largest 18  mass-specific digestive tract, causing the high-growth strain to perform well when food is abundant but lose weight more quickly under starvation. Expectations for differentiation in resource-related reaction norms were that the domesticated trout strain under the strongest growth selection would demonstrate the largest performance trade-offs (i.e., between high and low rations), but that the same patterns would manifest in the wild strains in proportion to their differences in growth rate. These hypotheses address general uncertainties and expectations for how selection on growth may constrain emergent phenotype by trading off against other performance attributes that collectively affect fitness. 2.2 Materials and methods 2.2.1 Experimental design Juvenile rainbow trout were acquired from the Freshwater Fisheries Society of British Columbia, Abbotsford hatchery, in January 2012. Fish were held in outdoor, freshwater flow through tanks for one to two months prior to experiments. Twelve days before treatments were initiated, fish were transferred to an environmentally regulated room where the temperature was set at 13.5 ° C, with a day:night cycle of 12:12 hours. Housing tanks consisted of 240 cm by 60 cm troughs with a recirculating water supply (with negligible water velocities) that were subdivided into 16 compartments with corrugated plastic and mesh screen. Fish were initially housed in groups of eight for the first five days, after which they were individually marked using elastomer tags, and housed individually to minimise any effects of dominance on growth and metabolism. Once fish were individually housed, they were placed on a food ration of 1% of their wet weight (slightly above maintenance). After seven days on the 1% ration, fish were randomly assigned to three ration levels and maintained on these rations for 21 days. The rations were: high (H) ration where fish were fed to satiation, mid (M) ration where fish were fed 1% of 19  their wet weight per day, and starvation (S) ration, where fish were fasted. Fish were fed 1.2mm pellet commercial trout chow (BioPro, Bio-Oregon, Washington, USA) via automatic feeders that distributed food at a constant rate over 8 hours. Satiation ration was estimated from a pilot experiment, with an added safety margin of 2% body mass. Due to time and space constraints, experiments with each of the three strains of rainbow trout (Fraser Valley, Tzenzaicut Lake, and Pennask Lake genotypes, or FV, TZ, PN) were performed in sequential trials over several months (March 7 – June 13, 2012). Thirty-two fish (11 H ration, 11 M ration, 10 S ration) from each strain were used and each strain was further divided into four groups of eight with start dates staggered by one day to allow time for performance measurements on individuals. Multiple breeding pairs from each strain were used; two pairs from the Fraser Valley strain, three pairs for the Tzenzaicut Lake strain, and three pairs from the Pennask Lake strain. To assess whether any differences in growth among strains (and by implication metabolism) could be due to a seasonal effect, two groups of five to six fish from each strain were reared simultaneously during the final four weeks of the main experiment on a ration of 2% of their body weight per day. Specific growth rate (SGR) of these grouped fish and the temporally staggered treatments from the main experiment were compared to evaluate the potential for seasonal effects on growth. Aside from temporally staggering populations, all population were subject to common garden conditions of temperature and photoperiod.  Length and weight of fish were measured when they were first housed individually (day 0), and every 7 days thereafter up to and including when they were sacrificed at day 28 for final lipid and water measurements. Ration (i.e. 1 % of body mass or satiation) was recalculated after each weighing. Mean mass ± SD of fish at the end of the pre-treatment acclimation period were 5.18 +/- 0.788 (FV), 3.72 +/- 0.899 (Tz), and 3.11 +/- 0.661 (Pn). Initial masses of Pennasks and 20  Tzenzaicuts did not differ across ration group (Tukey HSD (honest significant difference); P > 0.62), while Fraser Valley’s had a higher mass for every ration group (Tukey HSD; P < 0.02). Food consumption estimates were recorded on days 14, 21, and 28. To estimate food consumption, the bottom of tanks were siphoned to remove excess food and faeces, then automatic feeders were loaded with a known amount of food and fish were left to feed for 24 hours at which point the tanks were again siphoned. Care was taken to separate food from faeces, and the unconsumed food was dried and weighed. Mass-specific food consumption (𝐹𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑, as a % of wet body mass) was calculated as: 𝐹𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑  =  𝐹𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 − 𝐹𝑅𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔𝑀𝐹𝑖𝑠ℎ x 100 Where 𝐹𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑 is the dry food consumed (% body mass), 𝐹𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 is the dry mass (g) of fish food fed to the fish, 𝐹𝑅𝑒𝑚𝑎𝑖𝑛𝑖𝑛𝑔 is the dry mass (g) of food not eaten by the fish, and 𝑀𝐹𝑖𝑠ℎ is the wet mass (g) of the fish. Food consumption of fish on the high ration is considered to be their maximum food consumption, i.e. the maximum amount of food a fish can eat in a day.  Growth efficiency was calculated as: 𝐺𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = 𝑀𝐺𝑎𝑖𝑛𝑒𝑑𝐹𝐶𝑜𝑛𝑠𝑢𝑚𝑒𝑑 Where  𝑀𝐺𝑎𝑖𝑛𝑒𝑑 is the average daily mass increase (% wet body mass). Growth efficiency was calculated every 7 days using growth data from that week. However all growth efficiency data presented is an average from the entire experiment (3 measurements per fish). Specific growth rate (SGR) was calculated as: 𝜇 =𝑙𝑛 𝑀𝐹𝑖𝑛𝑎𝑙 − 𝑙𝑛 𝑀𝐼𝑛𝑖𝑡𝑖𝑎𝑙𝑡 21  Where 𝜇 is SGR (% body mass day-1), 𝑙𝑛 𝑀𝐹𝑖𝑛𝑎𝑙 is the natural logarithm of the final body mass (g), 𝑙𝑛 𝑀𝐼𝑛𝑖𝑡𝑖𝑎𝑙 is the natural logarithm of the initial body mass (g), and 𝑡 is time (days). Metabolic rates were measured on day 28 after 21 days on the treatment ration, as described below. 2.2.2 Maximum metabolic rate Fish tanks were vacuumed to remove any excess food or debris within the first 1.5 hours of the light cycle on the day of MMR measurement. It is possible that fish ate some residual food off the bottom of the tank early in the morning of the protocol, and that fish experienced some SDA costs; however, this should not affect our MMR values (Alsop and Wood, 1997). To control for possible effects of circadian rhythm, MMR was measured 5-8 hours after the daily light cycle commenced. Maximum metabolic rate was obtained by chasing individual fish to exhaustion in a 5 gallon bucket, fish were deemed exhausted after a minimum of five minutes of chasing, or when they no longer struggled when their tail was pinched and they were rolled over onto their backs. Immediately after being chased, each fish was placed in a closed respirometer appropriate to the size of the fish (26 mm diameter plastic tubes 8-13 cm long for 2-16 g fish) at a constant temperature of 13.8°C ± 0.5 SD. To ensure adequate mixing, a small stir bar was placed in the respirometer. Oxygen tension in the respirometer was recorded using a fluorescence based optical sensor (NeoFox oxygen probes, Ocean Optics, Dunedin, Florida, USA), and was terminated when oxygen tension dropped to ~60% of air saturated water, which usually took around 5 minutes. To account for lag in the response time of the probes, MMR was determined by using the 60 second period over which the rate of change of oxygen content in the respirometer was maximised. Oxygen consumption (?̇?𝑂2) was calculated as (Ege & Krogh 1914): 22  ?̇?𝑂2 =𝑉𝑊 · ∆𝐶𝑊𝑂2∆𝑡 Where ?̇?𝑂2 is the rate of oxygen consumption (µmol O2 hr-1), 𝑉𝑊 is the volume of water in the respirometer, ∆𝐶𝑊𝑂2 is the change in oxygen tension in the respirometer, and ∆𝑡 is the time period over which the drop in oxygen tension was measured. Concentration was calculated by correcting for PO2 (partial pressure of oxygen) using barometric pressure and multiplying by the solubility coefficient of water at the appropriate temperature, αO2 (µmol O2 L-1 kPa-1). Mass specific oxygen consumption was calculated as oxygen consumption divided by the mass of the fish (grams). 2.2.3 Standard metabolic rate Standard metabolic rate was measured following MMR in all cases except for Fraser Valley fish on high ration. The first three Fraser Valley fish on high ration died when placed in the flow-through respirometer while recovering from the MMR protocol. The reason for these mortalities is unknown. The fish did not die from asphyxiation as the recorded O2 tension did not drop below 70% saturation for these individuals. The remaining Fraser Valley fish on high ration had their MMRs recorded after the SMR protocol to avoid further mortalities.  Upon completion of the MMR protocol described above, recovering fish were placed into individual 13 cm long by 2.4 - 3.2 cm diameter glass flow-through respirometers that were attached to an aerated recirculating water system. Respirometers were individually covered with black plastic so that fish in adjacent respirometers could not see one another (eight fish were run simultaneously in parallel respirometers each night). Flow rates were adjusted to the oxygen demand of individual fish using a peristaltic pump to achieve oxygen tensions of 80% to 90% saturation. If oxygen tensions fell below 50% saturation during readings then data were discarded. Oxygen tension was recorded in the outflow from each respirometer and flow rate was 23  determined by weighing the amount of water discharged in one minute. Fish were placed in the respirometer in the afternoon, and were allowed to acclimate to the respirometer for 13-16 hours before their SMR was determined the next morning. Standard metabolic rate was recorded as the lowest average ?̇?𝑂2 recorded over a half hour period within 5 hours of the start of the day cycle in the environmental chamber. A pilot experiment indicated that this was the time at which the lowest ?̇?𝑂2was observed, which was consistent with previous findings (Van Leeuwen et al. 2011). Data were discarded if there was any indication of activity in the respirometer (identified by spikes in ?̇?𝑂2). Oxygen consumption was calculated as: ?̇?𝑂2 = 𝑉𝑊 · ∆𝐶𝑊𝑂2 Where 𝑉𝑊 is the flow rate through the respirometer (l·h-1), and ∆𝐶𝑊𝑂2 is the change in O2 content between the inflow and outflow water.  There is some contention around whether a true standard metabolic rate can be measured in fishes and whether there will always be some degree of energy expenditure above SMR. I refer to our metabolic rate results as SMR and not simply a resting metabolic rate due to the decline in metabolic rate observed in the final hours of the night cycle, along with strict guidelines set out a priori to reject any data where there was evidence of activity.  2.2.4 Gastrointestinal tract morphology At the end of the SMR readings fish were anesthetised with a lethal dose of MS-222. The GI tract (esophagus to anus) was then removed and excess fat and organs attached to the GI tract were returned to the body cavity. Wet mass of the GI tract was recorded, then stomach length and anterior intestine length were measured using digital calipers. The GI tract was then preserved in Karnovsky’s fixative and the body was frozen at -80°C for future lipid analysis.  24  Pyloric caeca were counted on a subset of fish using a dissection microscope; caeca were individually removed from the GI tract using a razor blade and tweezers and then counted. 2.2.5 Lipid and water content Whole fish (less the GI tract) were ground in a liquid nitrogen cooled mortar and pestle, taking care to keep the sample under nitrogen to prevent introduction of water. Three subsamples of ground tissue from each fish were placed in snap top microtubes and frozen at -80 °C for later analysis. Lipid extraction was performed using a modified Folch extraction with a sulphophosphovanillin assay (Barnes & Blackstock 1973). 40-60 mg of ground tissue was added to 3 ml of chloroform:methanol (2:1 v/v) which was then mixed using a sonicator, then 0.6 ml of NaCl solution was added and the solution was vortexed. After 10 minutes in the centrifuge at 1500 x g, a 1 ml aliquot of the lower (chloroform) layer containing the extracted lipid was transferred to another test tube where it was evaporated under N2. One ml of sulphuric acid was added to the residue in the tubes, vortexed, then placed in a boiling water bath for 15 minutes. 25𝜇l aliquots were then transferred to clean test tubes which had 4 ml of phosphovanillin colour reagent. Absorbance of the final solution was compared to that of a cholesterol standard prepared following the same procedure as above. All samples were run in triplicate, and an average absorbance of the three subsamples was calculated for each fish.  To measure water content, tissue subsamples were allowed to thaw inside of microtubes, then samples of 50-100 mg of tissue were weighed on pre-weighed individual aluminum foil cups. Samples were then placed in a drying oven at 60°C for 6 days. Dry samples were weighed, and percent water content was calculated by dividing the difference in wet and dry weights by the wet weight of the given sample. 25  2.2.6 Statistical analysis To test for differences in growth performance and metabolic attributes among strains, and to evaluate potential ration-dependent trade-offs, the effects of strain and ration on SMR, MMR, SGR, AS, growth efficiency, lipid content, water content, GI tract mass, stomach length, anterior intestine length, and caeca number were all modeled using the AICcmodavg package in R, which uses relative Akiake weights to rank models. Candidate linear models included strain, ration, and all interactions. Because of size variation among fish and known mass-dependence of most response variables, final weight was included as a covariate. A delta AICs value >2 is considered to show substantial evidence for the given model (Burnham & Anderson 2002); when there was more than one candidate model that fell within 2 delta AIC of the top model, these models where compared using an F-ratio test, and where applicable the most parsimonious model was accepted. Maximum food consumption was modeled in the same way except ration was dropped from candidate models. After a best model was selected (Table 1), an ANCOVA or ANOVA test was performed to determine statistical significance of the main effects, followed by a Tukey test (glht command from the multcomp package in R) to test for statistically significant differences between treatment groups. Strain and ration were treated as random effects. When a strain by ration interaction was included in the model, data were analysed separately by treatment. Differences in initial and final weights were also assessed using a Tukey test. To quantify the relative magnitude of strain (i.e. fixed) and ration (i.e. plastic or environmentally induced) effects on metabolism and growth attributes, the relative explanatory contributions of strain, ration, and weight were analysed using hierarchal partitioning (Chevan & Sutherland 1991; Nally 1996). Hierarchal partitioning is a technique for partitioning the variance explained by multiple explanatory variables in a manner that is not dependant on variable order 26  in a model statement. All possible models that include the explanatory variables are considered, and each variable’s explanatory power is assessed based on the average increased predictability that a model has when that variable is included. All possible models, including ones with interaction terms, are considered. Conceptually, hierarchal partitioning assigns relative importance to each explanatory variable based on its average contribution to the coefficient of determination (R2). Many researchers correct for metabolic scaling by choosing an appropriate scaling exponent (e.g. 0.75), then comparing residuals. While this is the preferred method to account for allometric scaling of metabolic rates, it hinges on the assumption that individual and population metabolic rates scale to the same exponent. One objective of this research was to investigate the causes of variation in metabolic rate between populations; I expected populations to scale differentially and thus did not feel justified assuming homogenous scaling then analysing residuals. Instead, individual mass was used as a covariate in all analyses as noted above. All statistics were performed using R version 3.0.0 statistical software (R Core Team 2013). Type 3 sum of squares (SAS Institute 1989) were used where applicable for assessing statistical significance. 2.3 Results 2.3.1 Specific growth rate (SGR) Ration had a significant effect on SGR (ANCOVA; F2,81 = 326.8, P < 0.001; Fig. 1a). Within a strain, SGR was always highest for individuals on the satiation ration, followed by fish on mid ration, and lowest for starved fish (Tukey HSD; P < 0.001 for differences among all rations for each strain; Fig. 1a), a group that showed negative growth. There was a significant effect of strain on growth (ANCOVA; F2,81 = 8.33, P < 0.001), with domestics exhibiting the 27  highest SGR at both high (Turkey HSD; P < 0.001) and mid rations (Turkey HSD; P < 0.001) when compared to either of the wild strains. The Tzenzaicut Lake strain grew marginally, but insignificantly faster than the Pennask Lake strain at high (Turkey HSD; P = 0.76) and mid (Turkey HSD; P = 0.112) ration. Contrary to expectations, the domestic strain did not lose weight fastest under starvation, and the Tzenzaicut Lake strain lost weight faster than both the domestics (Tukey HSD; P = 0.036) and the Pennask Lake strain (Tukey HSD; P = 0.002). Specific growth rate also showed a significant strain by ration interaction (ANCOVA; F4,81 = 17.05, P < 0.001), with growth of domestics responding most strongly to increased ration (Fig. 1a). Specific growth rate increased significantly with mass (ANCOVA; F1,81 = 7.61, P = 0.007), which was unexpected because the typical allometry of growth predicts that larger individuals should have a lower mass specific growth rate; however, this result is biased due to SGR and mass not being independent of one another. Final mass of individuals tended to match growth rates, i.e., final mass was highest for domestics and fish on the highest ration. 2.3.2 Standard metabolic rate (SMR) SMR trended to increase with increasing ration across all strains (Fig. 1b), but the main effect of ration was not significant (ANCOVA; F2,68 = 1.68, P = 0.193). There was a significant strain effect (ANCOVA; F2,68 = 3.14, P = 0.049), with domestics showing an elevated SMR relative to the wild strains at every ration level (Fig. 1b). However, there was very little difference in SMR between the two wild strains at any given ration. There was also a significant ration by mass interaction (ANCOVA; F2,68 = 4.99, P = 0.009), with mass-independent SMR at satiation, but with the expected decline in mass-specific SMR at mid and starvation rations (Fig. 1b). There was also a general trend for faster growers (e.g. the domestic strain and high ration treatments) to have higher SMRs. 28  2.3.3 Maximum metabolic rate (MMR) Maximum metabolic rate was negatively related to ration across all strains (ANOVA; F2,81 = 22.00, P < 0.001; Fig. 1c), with fish on the highest ration showing the lowest MMR; starved fish had the highest MMR (Tukey HSD; all ration levels significantly different at P < 0.03). There was also a significant effect of strain on MMR (ANOVA; F2,81 = 25.36, all strains differ at P < 0.001), with the ranking of strain reversed relative to SMR: Fraser Valley domestics had the lowest MMRs, followed by Tzenzaicut Lake trout (Tukey HSD; P < 0.001) who had lower MMRs than Pennask Lake trout (Tukey HSD; P < 0.001). Relative ranking was consistent across ration levels, even under starvation (Fig. 1c). There was a general trend for fast growing fish to have relatively lower MMRs across both ration treatments and strains. 2.3.4 Aerobic scope (AS) Trends in AS generally mirrored trends in MMR (Fig. 1d). Aerobic scope declined with increasing ration (ANCOVA; F2,66 = 3.23, P = 0.046), such that starved fish tended to have higher AS than fish on high ration (Tukey HSD; P = 0.052) or mid ration (Tukey HSD; P = 0.057), although this trend was marginally insignificant. Strain was also not significant (ANCOVA; F2,66 = 2.76, P = 0.070), although the ranking of AS among strains (AS of domestics lower than Tzenzaicut Lake trout (Tukey HSD; P = 0.084) and Pennask Lake trout (Tukey HSD; P = 0.084)) was the same as for MMR. In general, AS declined with increasing growth across strains, rations, and mass (ANCOVA; F1,66 = 4.05, P = 0.048). The marginal significance of treatment effects on AS, despite strong effects on SMR and MMR, is likely driven by the smaller sample size for AS (n = 72 for AS, vs. n = 76 and 86 for SMR and MMR, respectively). Because AS = MMR-SMR, discarding either an SMR or MMR observation for quality control reasons results in a lost AS observation. 29  2.3.5 Whole body water content Whole body water content was influenced by ration (ANCOVA; F2,78 = 7.05, P = 0.002), strain (F2,78 = 3.77, P = 0.027), and mass (ANCOVA; F1,78 = 16.11, P < 0.001). Water content was lowest at satiation, and increased with starvation across all strains (Fig. 2a). At every ration level, domestics appear to have higher water content than either of the wild strains. Water content generally decreased with body size for all strains, inversely mirroring the increase in lipid content noted below. There was a significant ration by mass interaction (ANCOVA; F2,78 = 3.36, P = 0.039), suggesting that mass effects on water content (i.e. increased water content in small fish) were most pronounced under starvation. 2.3.6 Whole body lipid content Lipid content varied systematically with ration (ANCOVA; F2,80 = 13.86, P < 0.001), strain (ANCOVA; F2,80 = 3.15, P = 0.047), and mass (ANCOVA; F1,80 = 9.87, P = 0.002). High ration treatments had the highest lipid content, while those on mid and starvation ration were similar (Fig. 2b). Lipid content generally increased with body size irrespective of ration, although there was a significant strain by mass interaction (ANCOVA; F2,80 = 4.28, P = 0.017). The slowest growing strain (Pennask Lake) had the lowest lipid content, which did not appear to change with body size, suggesting genetic differences in allocation of energy to storage among strains. 2.3.7 Gastrointestinal (GI) tract morphology GI tract mass differed with both strain (ANCOVA; F2,78 = 16.08, P < 0.001) and ration (ANCOVA; F2,78 = 25.10, P < 0.001). The domestic strain had a 25 - 50% larger GI tract than either the Tzenzaicut or Pennask strains at all ration levels (Tukey HSD; P < 0.002 for all comparisons; Fig. 2c), indicating a greater investment in organs related to growth and food 30  consumption. There was a significant strain by ration interaction such that Tzenzaicut Lake trout had a larger GI tract than Pennask Lake trout at mid (Tukey HSD; P = 0.002) and high (Tukey HSD; P < 0.05) rations, but not at starvation ration. Within strains the GI tract size was always largest for fish at high ration (Tukey HSD; P < 0.001 for each comparison). In general, the faster growing strains and ration levels had larger GI tracts (Fig. 2c), indicating both fixed and plastic responses to food consumption. The anterior intestine, stomach, and pyloric caeca were examined to assess their relative contribution to overall GI tract size. Faster growing strains had longer anterior intestines (measured as a percentage of body length; ANCOVA; F2,81 = 5.07, P =0.008; Table 2), but ration showed no effect (ANCOVA; F2,81 = 0.67, P =0.51). Stomach mass (measured as a percent of wet body mass) was greater for fish belonging to faster growing strains (ANCOVA; F2,82 = 24.06, P < 0.001; Table 2) and ration levels (ANOVA; F2,82 = 43.04, P < 0.001; Table 2). Caeca count had no significant relationship with strain (ANOVA; F2,8 =2.46, P = 0.14) or ration (ANOVA; F2,8 =0.15, P = 0.86), but the sample size was limited to a 13 fish pilot experiment for caeca counts.  2.3.8 Maximum food consumption   The Fraser Valley domestic strain had a significantly elevated maximum food consumption (ANOVA; F2,29 = 34.32, P <0.001) relative to either Tzenzaicut Lake trout (Tukey HSD; P < 0.001) or Pennask Lake trout (Tukey HSD; P < 0.001; Fig. 3a), which did not significantly differ in maximum food consumption (Tukey HSD; P = 0.83). Maximum food consumption showed no significant relationship with size across the range of observed body mass (3-16 g), despite the expectation of a mass-specific decline. 31  2.3.9 Growth efficiency Average growth efficiency was always higher for groups on mid rather than high ration (ANCOVA; F1,55 = 8.65, P = 0.005;Fig. 3b), and efficiency increased significantly with mass (ANCOVA; F1,55 = 4.43, P = 0.039), suggesting that faster-growing individuals were more efficient. There were also significant strain effects on growth efficiency (ANCOVA; F2,55 = 21.39, P < 0.001), with domestics showing a higher growth efficiency than either Tzenzaicut Lake trout (Tukey HSD; P < 0.001) or Pennask Lake trout (Tukey HSD; P = 0.001) at mid ration. At high ration the domestics trended towards the highest growth efficiency, although differences between the strains were not significant. There was also a significant strain by ration interaction (ANCOVA; F2,55 = 6.61, P = 0.003), indicating that the elevated efficiency of domestic fish was most pronounced at mid-ration.  2.3.10 Hierarchical partitioning The relative contributions of strain and ration to energetic and physiological responses were attribute-dependent (Table 3, Fig. 4). Although fixed genetic effects (strain) were significant for all attributes, variation explained by strain exceeded that explained by ration only for MMR and growth efficiency (Fig. 4). Ration could account for the majority of the explained variance in growth rate, SMR, lipid content, and water content. Strain effects were comparable to plastic responses for MMR, AS, and GI tract mass. Mass’ effect on the variation observed in our dependant variables was never below 20% of the variation explained by strain, ration, and mass combined.  2.3.11 Seasonal effect When all three strains were grown simultaneously, I observed the same relative pattern in growth rates, i.e., the Fraser Valley domestic fish grew the fastest, followed by the Tzenzaicut 32  Lake strain, while the Pennask Lake strain grew the slowest (Fig. 5). This suggests that the different growth rates observed in the rest of the study are real, and not just a seasonal artefact of different hatching times among strains.   2.4 Discussion 2.4.1 Physiological trade-offs associated with rapid growth By explicitly comparing rainbow trout strains with different ecological origins, growth, and metabolic attributes, I was able to evaluate some of the key physiological and anatomical trade-offs thought to be associated with rapid growth. In particular, I focused on two competing hypotheses: a trade-off between growth and performance (scope for activity), and a trade-off between growth and starvation resistance, i.e., the ability to achieve high growth under nutrient enrichment vs. the ability to minimise weight loss during famine (Dupont-Prinet et al. 2010). Higher growth has been demonstrated to trade-off against lipid storage in diverse taxa (e.g. fruit flies; Djawdan, Rose & Bradley 1997, oysters; Bayne 2004). If strains have traded-off growth against increased lipid storage and starvation resistance, the expectation would be crossed reaction norms where the fastest growing strain would achieve the highest growth at high ration (enriched environment), while the slowest growing strain (Pennask) would be the most resistant to weight loss during starvation. Evidence for this idea was at best equivocal. The fastest growing strain, domestics, did not show greater weight loss under starvation, despite having a higher SMR which should consume energy reserves faster. It is possible that domestic fish had a higher lipid reserve associated with their larger size which mitigated the effect of higher SMR, although this is not apparent in the lipid-body mass regression. In contrast, the two size matched 33  wild strains followed the expected pattern of crossed reaction norms, where the strain with the higher growth (Tz) lost more weight under starvation. However, this cannot be unambiguously attributed to differences in SMR or lipid content since neither significantly differed between the wild strains, although high variation amongst individuals may limit power to detect meaningful differences in SMR between wild strains that might be consistent with a growth-starvation resistance trade-off. It is also possible that differences in starvation tolerance are due to differences in activity (routine metabolic rate, RMR) or stress responses from periodic handling (Careau et al. 2008; Burton et al. 2011). There was much stronger evidence for a trade-off between growth and maximum aerobic capacity among strains. The fast-growing domestic rainbow trout had the lowest aerobic scope, reflecting their elevated SMR and lower MMR. Metabolic differences between the wild strains were also consistent with the relative ranking of their growth rates, with faster growing Tzenzaicut Lake trout having a lower MMR and AS than slower growing Pennask Lake trout. The plausibility of this trade-off is enhanced by the similar reduction in AS of fish on high rations. Aerobic Scope is highly correlated with prolonged swim performance (Farrell et al. 2008), which has been demonstrated to increase survival under moderate to high predation risk (Billerbeck et al. 2001). Lower AS as a consequence of higher growth has been previously demonstrated among wild fish populations (e.g. counter gradient growth variation in Atlantic silversides; Arnott et al. 2006). Reduced AS and swimming performance in domesticated strains has also been well documented (Magnan et al. 2009; Van Leeuwen et al. 2011), suggesting that a growth-performance trade-off among populations is likely widespread, at least among fishes. The most likely mechanism for a negative effect of higher growth on AS is through an increase in apparent SMR. Elevated metabolism associated with growth is a general phenomenon 34  among faster growing strains or animals on higher rations (Wieser 1994), and is well established in fishes (e.g. Arnott et al. 2006; Van Leeuwen et al. 2011). An increase in size of the GI tract in faster-growing strains, as observed in this study, also provides a partial mechanism for elevated SMR leading to reduced AS, since the GI tract and associated visceral organs have a high mass-specific metabolic rate that contributes disproportionately to resting metabolism (Konarzewski & Diamond 1995; Chappell, Bech & Buttemer 1999). Alternatively, increased GI tract mass may simply be correlated with elevated whole-body costs of tissue synthesis and metabolism in faster growing fishes, rather than oxygen demand from additional GI tissue being a direct cause of elevated SMR. A trade-off between growth and AS may also be mediated by an organ-system trade-off, i.e. a reduced investment in the cardiovascular system in favour of a larger GI tract to process food in faster growing strains. This hypothesis is consistent with the positive ranking of GI tract size and growth among the domestic and wild rainbow trout strains, although I did not measure heart size, and therefore have no direct evidence for this potential organ system trade-off. However, the limited number of studies that have tested for a trade-off between investment in the GI and circulatory tissue have failed to find any negative correlation between heart and digestive organ mass (e.g. sparrows, (Chappell et al. 1999); high vs. low growth strains of mice, (Sadowska et al. 2013)). In fact, Norin and Malte (2012) found a positive correlation between heart and intestine/stomach mass among individual brown trout (Salmo trutta), suggesting that overall selection for higher growth and metabolism along a general slow vs. high pace-of-life gradient (Reale et al. 2010) may override any trade-off between digestive and circulatory systems. This does not necessarily negate the general potential for organ-system trade-offs related to growth; for example, recent selection experiments on brain sizes among guppies have demonstrated that larger brain size comes at the cost of a reduced investment in size of the 35  digestive tract (Kotrschal et al. 2013). However, the picture that emerges is that organ system trade-offs, while potentially important, may be quite variable (Reale et al. 2010; Careau & Garland 2012). Differentiation along environmental gradients is a pervasive feature in evolution that promotes diversity and co-existence among taxa (Young 2001). Physiological constraints may limit the scope and combination of phenotypes that are optimal in any environment (Kirkpatrick & Lofsvold 1992), such that divergent taxa may display consistent covariance among traits on environmental gradients (an “integrated phenotype”; Pigliucci 2003). A slow to fast metabolic continuum has been proposed as a dominant adaptive gradient differentiating organisms (Lovegrove 2003); this concept is now referred to as the pace-of-life syndrome, where suites of behavioural, physiological, anatomical, and metabolic traits are correlated along a slow-fast metabolic continuum (Reale et al. 2010). Core predictions of the pace-of-life syndrome are positive correlations among growth, metabolism, and bold behaviour (Álvarez & Nicieza 2005; Biro & Stamps 2010; Reale et al. 2010). Trait variation among strains of rainbow trout that differed in growth in this study were broadly consistent with expectations from the pace-of-life syndrome in that the fastest growing strain (FV) exhibited the highest maintenance (but not peak) metabolism. A logical extension of the pace-of-life syndrome is to include energetic efficiency of resource use. Growth efficiency typically decreases with increasing ration and growth, and in organisms with higher metabolic rates. For example, Tucker and Rasmussen (1999) found that growth efficiencies and growth rate were inversely related within juvenile Atlantic salmon (Salmo salar), and also between Atlantic salmon and brook trout. Millidine, Armstrong & Metcalfe (2009) also found that juvenile Atlantic salmon with higher SMR had higher costs of 36  digestion, presumably as a trade-off with faster gut clearance to facilitate greater maximum food consumption and growth. Collectively, these studies suggest that a trade-off exists between maximising food consumption and growth at one end of a metabolic continuum (energy maximisers) and maximising the efficiency of resource use (efficiency maximisers) at the other. Energy maximising (Schoener 1971) in this sense involves fish adopting a high metabolic rate, maximising gross energy intake to maximise growth, and consequently elevating foraging and overall activity levels. In contrast, an efficiency maximising strategy involves maximising efficiency of available resource use in lower productivity environments, which may entail reducing activity costs or adopting a sit-and-wait type of foraging strategy (e.g. Sotiropoulos, Nislow & Ross 2006), and a lower metabolic rate. Contrary to expectation, domestic fish adopting an energy maximising strategy experienced no penalty of decreased growth efficiency, and in fact exhibited the highest growth efficiency among strains, i.e., they maximised both energy intake and efficiency. It appears that domestics can mitigate the energetic costs of an elevated metabolism associated with increased food consumption by being more efficient at absorbing nutrients from their food. Alternatively (or additionally), domestic fish may be increasing their apparent growth efficiency by reducing the energy density of their tissue (e.g., by having a higher water content) or by reducing their costs of tissue synthesis (Wieser 1994). McCarthy, Houlihan and Carter (1994) found that faster growing juvenile rainbow trout had lower protein turnover rates (maintenance costs) and therefore higher growth efficiency. The elevated SMRs of Fraser Valley domestics suggests that they do not adopt this strategy to maximise growth efficiency, but given uncertainty in the contribution of growth and maintenance costs to their measured SMR, it is possible that domestics have reduced true maintenance costs that support higher growth. Regardless, the elevated growth efficiency of 37  domestics demonstrates that high food consumption and efficiency are not intrinsically incompatible, although the direct relevance to trade-offs in wild populations remains somewhat unclear. It remains possible that pervasive relaxation of selection in hatcheries allows trade-offs that are unlikely in natural populations. 2.4.2 What is the physiological basis of rapid growth? Optimal growth rates are ultimately determined by trade-offs between growth and survival (Arendt 1997), which will vary with environment or the life-history strategy adopted by different genotypes or species. Food-rich environments with low predation risk, like hatcheries, result in selection for high growth rates (Petersson et al. 1996; Fleming et al. 2002) because increased food consumption is decoupled from exposure to predation, and high predation mortality on bold individuals is relaxed (Biro & Stamps 2010). Conversely, high predation risk for small size classes in the wild may also select for fast juvenile growth so that juveniles can minimise exposure to gape-limited predators (Urban 2007). Irrespective of the selective environment, growth rate will also be limited by resource supply (environment) and the food consumption and energy processing capacities of the organism. Unsurprisingly, the domestic strain showed the highest growth, which was achieved through a combination of increased maximum food consumption, increased growth efficiency, and decreased tissue energy content. Increased maximum food consumption has been observed in other high-growth strains, ranging in taxonomic diversity from Atlantic silversides (Billerbeck et al. 2000) to oysters (Bayne 2004), and is the most direct adaptation to increase growth (Bayne 2004). Increased growth efficiency requires increased extraction efficiency, or a reduction in either active metabolic rate, energy density of tissue, cost of tissue synthesis, or maintenance (Bayne 2000, 2004). Increased extraction efficiency (the amount of nutrients extracted from ones diet) 38  necessitates either more intense uptake per unit gut area (Caviedes-Vidal et al. 2008), or an increase in gut passage time allowing for more time to extract nutrients (Afik & Karasov 1995). However, increasing gut passage time will reduce throughput (maximum sustained ration; Nicieza et al. 1994), unless the animal also increases the volume of the digestive tract. While I did not measure passage time, this mechanism is supported by the increase in GI tract size in the domestic (FV) and the faster growing of the two wild strains (Tz). A larger surface area for absorption will also increase extraction efficiency, which in rainbow trout occurs primarily in the anterior intestine and pyloric caeca (Bergot et al. 1981; Stevens & Devlin 2005). Relative mass of the anterior intestines and caeca have been shown to increase in transgenic Atlantic salmon engineered for rapid growth (Stevens, Wagner & Sutterlin 1999). Similarly, faster growing full-sib rainbow trout that are larger at age have also been shown to have a greater number of pyloric caeca and increased growth efficiency (Bergot et al. 1981). Although a positive correlation between higher growth strains and anterior intestine length appears to exist, there was no visible correlation between caecal number and growth.  This is not entirely surprising, however, because absorption is sensitive to both surface area and digestive intensity along the GI tract (Caviedes-Vidal et al. 2008). In addition to selecting for increased absorptive capacity, increased selection on growth should result in increased stomach size to maximise capacity to capitalize on heterogeneous prey availability, allowing for gorging when food is in excess (Armstrong & Schindler 2011). I found that high growth strains and fish on high rations both tended to have marginally larger stomachs (measured as a percentage of body weight), supporting both a plastic as well as a genetic basis for differences in stomach capacity. However, GI tract mass in domestic trout was not significantly different between starvation and mid ration, lending support to the inference that 39  domestic fish always invest more energy in growth-promoting organs even when energy is limited, suggesting an overall up-regulation of digestive capacity. Finally, increased water content of the domestic strain suggests that lower tissue energy density (and potentially lower costs of tissue synthesis) also contributed to higher somatic growth. Post and Parkinson (2001) observed that juvenile trout have a high water content when small, and transition to lower water and increased lipid content as they grow. Post and Parkinson (2001) suggested that this trade-off between water and lipid maximises somatic growth when small, minimising vulnerability to gape limited predators, but maximises likelihood of overwinter survival by transitioning to lipid storage as autumn approaches (Clark 1994; Post & Parkinson 2001). The pattern of substitution of water for lipid as juveniles grow is apparent in my data; however, there is variation among strains, i.e., Pennask Lake trout showed no increase in lipid at larger size, despite their low overall somatic growth rates. The effects of size on water content is most pronounced under starvation, suggesting that maintaining body size (i.e. by increasing water content) is most important for small fish, and persists even for the hatchery strain where predation risk on juveniles is presumably very low. This suggests that dominance and competition for resources may also select for high water content as a strategy to grow quickly, because dominance hierarchies continue to play a significant role in the ecology of hatchery raised domesticated fish (Abbott, Dunbrack & Orr 1985).  Mass-specific growth of juvenile rainbow trout (i.e. daily growth as a percent of body weight) also showed some unexpected allometries. Mass-specific growth should decline as organisms grow. However, mass-specific growth consistently increased with size, which is not predicted from classic metabolic allometry (predicted slope of -0.25). This may partly be an artefact of correlation between axes (specific growth rate and mass), i.e. individuals with higher 40  growth will also have larger mass at the end of the growth period, elevating the slope of the line. It could also be an artefact of using a fixed pellet size (1.2 mm) that was relatively easier to digest for larger fish with a longer gut retention time (Zhu et al. 2001). Alternatively, growth rates of salmonids may have an allometry that deviates from that of typical animals. Allometry of SMR in salmonids is higher than average (i.e. mass exponent of ~0.92 (Bohlin et al. 1994)), which is significantly higher than the average allometry of 0.75.    2.4.3 Relative magnitude of genetic (fixed) vs. flexible (plastic) effects on growth and metabolism Unsurprisingly, ration was responsible for much of the variation in metabolic attributes like apparent SMR, growth, and body composition (lipid and water content). Growth is primarily dependent on ration, and the positive effect of ration on apparent SMR is well-documented (e.g. Wieser 1994; Van Leeuwen et al. 2011). However, the magnitude of strain (genetic) effects on the potential for active metabolism (MMR and AS), size of the GI tract, and in particular growth efficiency were comparable to or exceeded the effects of ration. Significant genetic differences between strains (albeit inferred from phenotypic differences in common garden experiments) on all measured attributes indicates that variation among genotypes is multi-dimensional, but the largest effects seemed to hinge on food consumption and processing (GI tract size and growth efficiency), and the apparent negative associations with active metabolic potential (MMR and AS).  The large differences in growth and metabolic attributes across the three strains I examined need to be tempered by the observation that one of them was artificially selected for high growth, which may have increased the range of attribute variation relative to wild 41  populations. Nevertheless, the two wild populations also differed in growth and other attributes, although to a lesser extent. Insofar as the two wild strains both represent viable stocking options for sport fishing in lakes, they do not necessarily represent the full range of variation in growth among wild stocks. Many genetically distinct and slow growing rainbow trout stocks occur in high elevation lakes or low productivity headwater streams, as well as fast-growing large lake piscivores (Keeley, Parkinson & Taylor 2005), and it is conceivable that the range in growth rates among natural rainbow trout populations is comparable to or greater than those represented in this study. Both plastic and parental effects cannot be entirely ruled out as contributors to the genetic variation that was observed. While every effort was made to achieve common garden conditions, and seasonal timing effects appeared to be minimal, plastic responses to slightly different environmental factors (e.g. temperature in outdoor flow through tanks pre-experiment) could play a role in development, potentially affecting later life stages (Desai & Hales 1997). Although multiple breeding pairs were used for each population, it is quite likely that the phenotype of the broodstock deviates to some degree from the  population average; in this case observed differences among strains would still be genetically based, but with some component due to deviation of the parental genotype from the population mean.  2.4.4 Conclusions, caveats, and implications Animals with high growth have been shown to have elevated SMR or BMR, high food consumption, and larger GI tracts. This is the first study to my knowledge that demonstrates a common evolvability of all these traits in fish. Further, I have demonstrated that these same traits do not only vary among individuals, but differ between genetically distinct populations with 42  different growth rates. I suggest that rainbow trout with high growth rates maintain a high digestive capacity at the cost of reduced MMR and aerobic scope, and by implication swim performance, supporting a general growth vs. performance trade-off at the population level. One limitation of our study was the inability to perfectly size match the different groups of fish. Due to allometric scaling of morphological and physiological traits, I must interpret my results with caution.  Domesticated fish provide a good model for studying the trade-offs associated with high growth rate if we assume that they evolve the same trade-offs that occur under natural selection. While many of the measured traits did not differ statistically between my two wild strains, all of my conclusions, if not noted otherwise, were supported by the differences seen in the wild strains; the trends seen between fast growing domestics and the wild strains were generally similar (just greater in magnitude) to the trends seen between the moderate and slow growing wild strains. Physiological trade-offs associated with growth do exist, and are a key part of integrated morphological, behavioural, and physiological change along selection gradients (Reale et al. 2010; Careau & Garland 2012). While ecological trade-offs dominate the literature (e.g., growth vs. predation risk trade-offs), physiological trade-offs are equally important to understand and are often the underlying causes of the less cryptic ecological trade-offs (e.g., decreased aerobic scope leading to higher vulnerability to predation). Studying physiological correlates of growth provides insight into an important dimension of the adaptive landscape that can ultimately help us understand the nature of species adaptation and differentiation.43  Dependent variable Strain Ration Log mass Strain x Ration Strain x Log mass ration x Log mass Intercept n F P value Log SMR 6.1e-02 (FV) -4.09e-04 (H) -0.243 x Log mass - - 0.254 x Log mass (H) 0.795 76 11 <0.001  0 (Tz) 0.115 (M)  - - -7.18e-03 x Log mass (M)      3.87e-03 (Pn) 0 (S)  - - 0 x Log mass (S)     Log MMR -9.39e-02 (FV) -0.125 (H) - - - - 1.41 86 22.4 <0.001  0 (Tz) -8.15e-02 (M) - - - -      3.94e-02 (Pn) 0 (S) - - - -     Log AS -9.12e-02 (FV) -0.12 (H) -0.249 x Log mass - - - 1.44 72 16.2 <0.001  0 (Tz) -8.1e-02 (M)  - - -      1.73e-02 (Pn) 0 (S)  - - -     Log SGR + 1.8 5.77e-02 (FV) 0.524 (H) 0.113 x Log mass 9.21e-02 (FV:H) 2.28e-03 (FV:M) 0 (FV:S) - - 1.14e-02 91 471 <0.001  0 (Tz) 0.364 (M)  0 (Tz:H) 0 (Tz:M) 0 (Tz:S) - -      6e-02 (Pn) 0 (S)  -0.104 (Pn:H) -7.69e-02 (Pn:M) 0 (Pn:S) - -     Lipid Content -0.646 (FV) 1.02 (H) 2.31 x Log mass - 0.581 x Log mass (FV) - 1.64 88 31.4 <0.001  0 (Tz) 0.124 (M)  - 0 x Log mass (Tz) -      0.958 (Pn) 0 (S)  - -2.17 x Log mass (Pn) -     Water Content 1.01 (FV) -3.49 (H) -6.88 x Log mass - - 3.77 x Log mass (H) 80.8 86 19.8 <0.001  0 (Tz) -2.75 (M)  - - 3.96 x Log mass (M)      5.13e-02 (Pn) 0 (S)  - - 0 x Log mass (S)     Growth Efficiency 0.602 (FV) -0.264 (H) 0.535 x Log mass -0.427 (FV:H) 0 (FV:M) - - 0.764 62 36.3 <0.001  0 (Tz) 0(M)  0 (Tz:H) 0 (Tz:M) - -      -3.62e-02 (Pn) -  -8.01e-02 (Pn:H) 0 (Pn:M) - -     GI Tract 1.31 (FV) 1.96 (H) - 0.592 (FV:H) -0.592 (FV:M) 0 (FV:S) - - 3.19 87 40.8 <0.001  0 (Tz) 1.24 (M) - 0 (Tz:H) 0 (Tz:M) 0 (Tz:S) - -      5.01e-02 (Pn) 0 (S) - -0.656 (Pn:H) -0.747 (Pn:M) 0 (Pn:S) - -     Max Consumption 1.04 (FV) - - - - - 2.65 32 34.3 <0.001  0 (Tz) - - - - -      -3.33e-02 (Pn) - - - - -      Table 1: The effect of strain (Fraser Valley (FV), Tzenzaicut Lake (Tz), Pennask Lake (Pn)), ration (High (H), Mid (M), Starvation(S)), mass, and their interactions on logSMR (Standard Metabolic Rate - µmol O2 h-1), logMMR (Maximum Metabolic Rate - umol O2 h-1), logAS (Aerobic Scope - umol O2 h-1), log SGR+1.8 ( Specific Growth Rate - % body mass per day), whole body lipid content (% of wet mass), whole body water content (% of wet mass), growth efficiency (g eaten per g body mass gained), gastrointestinal (GI) tract mass (% of wet mass), and maximum food consumption (% wet mass per day). Example of how to calculate logSGR +1.8 of a FV on H ration: log10(SGR + 1.8) = 0.0577 + 0.524 + 0.113 x Log10(wet mass) + 0.0921 + 0.0144  Dependent variable Strain Ration Mean n SD Anterior Intestine Length FV  21.9 30 4.04  Tz  21.1 26 3.71  Pn  17.6 31 3.52       Stomach Mass FV  15.2 30 1.95  Tz  14.2 26 1.79  Pn  13.2 31 1.26   H 15.5 29 1.65   M 14.3 29 1.48   S 12.7 29 1.28  Table 2: Means, sample sizes (n), and standard deviations (SD) of anterior intestine length (as a percent of total body length) and stomach mass (as a percent of total body mass). Only data found to be associated with a significant effect of either strain (Fraser Valley (FV), Tzenzaicut Lake (Tz), and Pennask Lake (Pn)) or ration (High (H), Mid (M), and Starvation (S)) are presented.    45   Dependent Variable Explanatory Variable Independent Contribution (%) Contribution Joint Contribution (%)     Log SGR + 1.8 Strain 3.3 0.5  Ration 37.5 19.8  Log mass 21.1 17.8     Log SMR Strain 4.7 -3.1  Ration 55 16.9  Log mass 15.9 10.6     Log MMR Strain 16.9 11.1  Ration 14.7 9.1  Log mass 24.4 23.8     Log AS Strain 13.7 8.2  Ration 15.8 9.6  Log mass 28.6 24.2     Lipid Content Strain 6.2 4.3  Ration 25.2 16.8  Log mass 26.7 20.8     Water Content Strain 4.7 -3.4  Ration 35.2 23.2  Log mass 25.8 14.5     GI Tract Mass Strain 14.4 10.8  Ration 16.7 11.3  Log mass 23.2 23.7     Growth Efficiency Strain 41.3 17.5  Ration 23.4 -2.6  Log mass 15.6 4.7  Table 3: The percentage of total variation that can be explained from the independent and joint contributions of strain, ration, and mass, for specific growth rate (SGR), standard metabolic rate (SMR), maximum metabolic rate (MMR), aerobic scope (AS), whole body lipid content, whole body water content, gastrointestinal (GI) tract mass, and growth efficiency. Independent and joint contributions were obtained through hierarchical partitioning.  46  47  Figure 1 (previous page): Strain and mass effects on specific growth rate (a), standard metabolic rate (b), maximum metabolic rate (c), and aerobic scope (d), at high (left column), mid (middle column) and starvation (right column) rations. Multivariate regression lines represent best fits from AICc model selection (see Table 1 for model details). Each data point represents an individual fish. Squares and solid lines represent the Fraser Valley strain, triangles and dashed lines represent Tzenzaicut Lake strain, and circles and dotted lines represent Pennask Lake strain. Horizontal lines indicate no significant main effect of mass in the best model (e.g. MMR).  48   Figure 2: Relationships between mass and whole body water content (a), whole body lipid content (b), and gastrointestinal (GI) tract size (c), at high (left column), mid (middle column) and starvation (right column) rations. Multivariate regression lines represent best fits from AICc model selection (see Table 1 for model details). Each data point represents an individual fish. Squares and solid lines represent the Fraser Valley strain, triangles and dashed lines represent the Tzenzaicut Lake strain, and circles and dotted lines represent the Pennask Lake strain.  49   Figure 3: Relationships between mass and maximum food consumption (a) and growth efficiency (b). Multivariate regression lines represent best fits from AICc model selection (Table 1). Each data point represents an individual fish. Squares and solid lines represent the Fraser Valley strain, triangles and dashed lines represent the Tzenzaicut Lake strain, and circles and dotted lines represent the Pennask Lake strain. Open symbols represent fish on high ration, grey filled symbols represent fish on mid ration. 50   Figure 4: The percentage of the total variance that can be explained by strain (black), ration (grey), and mass (white) for specific growth rate (SGR), standard metabolic rate (SMR), maximum metabolic rate (MMR), aerobic scope (AS), whole body lipid content (Lipid), whole body water content (Water), gastrointestinal tract size as a percent of body mass (GI mass), and growth efficiency (GE).  Variance in each category represents the sum of independent and joint effects which were obtained through hierarchical partitioning.  51   Figure 5: The relationship between specific growth rate (SGR) and mass for fish held in groups of 5-6 during the final two weeks of the experiment. Each data point represents the average growth rate of fish found in one tank. Squares represent the Fraser Valley strain, triangles represent the Tzenzaicut Lake strain, and circles represent the Pennask Lake strain.52  Chapter 3: General Discussion The objectives of my research were to (i) understand how high growth rates are achieved and the consequences for emergent whole-body metabolism; (ii) to understand the nature of adaptive trade-offs involved in rapid growth, specifically looking at metabolic rate, energy conversion efficiency, and investment in different organ systems. In this discussion, I will briefly recap the literature associated with my objectives and how the results from my research add to this literature. Furthermore, I suggest areas where further research could be directed.  3.1 Physiological mechanisms that promote high growth rate Total food consumption and growth efficiency govern growth rate. Food consumption in nature varies depending on the cost of acquisition, quality, and energy demand. Cost of acquisition includes energy used during foraging and potential predation risk; quality can be thought of as the ratio of labile nutrients to bulk; and demand will depend on the energy requirements of the animal (Karasov et al. 2011). In the lab, cost of acquisition and food quality can be held constant, meaning that observed food consumption, when fed to satiation, is expected to reflect energy demand. Interestingly, the two wild strains of rainbow trout in my study had similar maximum food consumption yet showed different growth rates, the explanation for this being that Tzenzaicut Lake trout had a marginally (but not significantly) higher growth efficiency than Pennask Lake trout.  Animals with high food consumption often show decreased growth efficiency (Afik & Karasov 1995). I had hypothesised that one of the consequences of higher growth in domesticated fish would be lower growth efficiency compared to the two wild strains and that this would be particularly pronounced at satiation. Contrary to expectations, domesticated fish had higher growth efficiency at both high and mid ration levels. McCarthy, Houlihan & Carter 53  (1994) showed rainbow trout with high growth rates had a constant rate of protein turnover whereas slow growers demonstrated an increased rate of protein turnover with increasing ration level. The decreased growth efficiency of slowly growing fish was attributed to a combination of increased protein turnover, higher quality tissue, or lower water content. It is possible that the Fraser Valley domestic strain has a fixed lower rate of protein turnover regardless of ration level. Further investigation into rates of protein turnover (tissue maintenance costs) may provide more insight into differences in growth efficiency. The cost of tissue synthesis can be reduced by decreasing the energy content of newly formed tissue.  Post & Parkinson (2001) suggested that the inverse relationship that is often seen between lipid and water content in young fish may allow small rainbow trout to increase body size at a very low energetic cost. I demonstrated that the domestic strain had higher water content for a given size than the two wild strains, suggesting that they are reducing the energy content of their tissues and as a result can achieve higher growth rates. Unfortunately, the results from the whole body lipid content assay are hard to interpret due to a significant strain by log-mass affect. While the domestic fish had higher lipid content, this may be solely due to allometry (i.e. they were larger on average than other strains).  Many animals modulate their gastrointestinal (GI) tract in response to food consumption (Secor et al. 1994). Domesticated fish had a larger GI tract than the wild fish. In particular, the stomachs and anterior intestines of domestic fish were larger than the slower growing wild strains. Interestingly, the size of the GI tract was not only dependant on the strain, but also on ration, implying that there are plastic responses to food consumption that can modulate the size and, possibly, efficiency of the GI tract. 54  I found that domesticated rainbow trout implement three different strategies to achieve high growth rates: (i) they eat more food, (ii) they grow tissues that have lower energy content, and (iii) they have enlarged GI tracts. I suggest that the degree of increased food consumption and reduced energy content of the domestic strain would have negative fitness consequences in the wild and therefore are not realised by the faster growing wild strain. The mechanism underlying the differences in growth rates among wild strains remains unclear, but interestingly, although not necessarily statistically significant, most trends observed in the domesticated fish were also seen in the faster growing of the two wild strains, i.e. Tzenzaicut Lake trout had larger GI tracts, higher lipid content, and higher maximum food consumption.  3.2 How metabolic rates relate to growth rate  Metabolic rates are known to vary with growth rate (Álvarez & Nicieza 2005). When food is available in excess, high standard metabolic rate (SMR) is correlated with high growth rate and dominance (Metcalfe 1998; Yamamoto, Ueda & Higashi 1998), but when food is scarce high SMR has been shown to correlate with low growth rates, probably because it is energetically wasteful (Álvarez & Nicieza 2005). I found that high growth rate was correlated with high SMR and low maximum metabolic rate (MMR), resulting in a low aerobic scope (AS). I found that starved fish showed no consistent pattern of weight loss based on SMR, or MMR. The data suggest that metabolic rates have both a genetic (strain) and plastic (ration) relationship with growth rate; increased growth rate due to either strain or ration has detrimental consequences for metabolic rate. 3.3 Physiological trade-offs associated with growth One major motivation behind my research was to identify the physiological trade-offs associated with growth in rainbow trout and to what extent these trade-offs differ between 55  distinct, locally adapted populations. There are several well documented trade-offs that I did not explore (predation risk vs. foraging behaviour trade-off, habitat selection along a productivity gradient, aggressive dominance vs. growth trade-off), most of which are based on behaviour and/or ecology. After an exhaustive search of the literature, I came up with four candidate physiologically based trade-offs, each of which I will address below. Animals that have high SMR and low energy reserves should show a reduction in starvation resistance measured as weight loss or time to death (Chippindale et al. 1996; Gotthard 2001; Stoks et al. 2006). Animals that show a high propensity for growth are characterised by high SMR and often they have low energy stores (i.e. investing surplus energy in somatic structure rather than energy reserves; Gotthard 2001). There is a wealth of literature that suggests a trade-off exists between high growth rate and low starvation resistance (Chippindale et al. 1996; Scharf et al. 2009; Dupont-Prinet et al. 2010); however, my results do not provide strong support for this. The significant strain by ration interaction for lipid content does indicate that domestics lose lipid somewhat faster under starvation (or gain it more quickly at satiation). The domesticated strain also had the highest SMR, but size matching issues complicate interpretation of strain effects on lipid content (energy reserve). In any event, the domestics did not show the most rapid weight loss. One possible explanation for this, as well as a potential subject for future research, is that SMR may not be indicative of the true metabolic energy expenditure of my fish, and that instead routine metabolic rate should be investigated as it will give a more reliable measure of the energy used during starvation (Burton et al. 2011). Another possibility is that the domesticated fish were less stressed in my experimental system than the wild strains; stressed animals tend to lose weight faster when starved (Sloman et al. 2000). 56  Aerobic scope is directly related to prolonged locomotor performance (Farrell et al. 2008) which is often traded-off for high growth rates (Billerbeck et al. 2001). While I did not directly test swim performance, I did find that groups with high growth rate had lower AS, which came about due to an increased SMR and a decreased MMR. My study confirms that aerobic performance can be limited by genetic affects between populations (Billerbeck et al. 2001; Arnott et al. 2006), as well as environmental affects between rations.   When food is abundant, food passage time (time from mouth to anus) will often decrease resulting in a correlative decrease in extraction efficiency. The digestive strategy employed by a population will depend on local food type, its quality, its availability, and associated handling costs, be they internal (digestive, absorptive, etc.) or external (foraging risk, energy expenditure associated with prey capture, etc.).  I had predicted that domestic fish, which are adapted to a low cost and high food availability environment, would have a lower growth efficiency compared to wild strains. Presumably the domestics would benefit from a short food passage time associated with a lower extraction efficiency, which would allow them to consume more food. By contrast, domestic fish had higher growth efficiencies than the wild strains on both mid and high rations. On the other hand, when comparing within a strain, but between rations, I did see the expected trend: fish that ate less had a higher growth efficiency. While organs account for only a small fraction of the overall mass of many animals, they can be metabolically very expensive and can contribute disproportionately to SMR or basal metabolic rate (BMR) (Konarzewski & Diamond 1995). Liver size is often correlated with GI tract size and whole body BMR (Konarzewski & Diamond 1995; Rosenfeld et al. 2014b); digestion is energetically expensive and the cost of maintaining a high capacity for growth appears to come at the price of a large GI tract with a high metabolic demand (Secor et al. 1994; 57  Secor & Diamond 2000; Armstrong & Bond 2013). Further, organ growth itself is not cheap (Secor 2001), given that energy for growth is limited, I propose that a reduction in MMR amongst fish with large GI tracts is in part due to a relative reduction in cardiovascular system investment. Future research could focus on the energetic costs associated with growth and maintenance of the GI tract and cardiovascular system in rainbow trout.   3.4 General caveats Unfortunately, given the nature of my research, it was impossible to perfectly size-match all fish. Interpretation of the results was—and should be—done with caution when comparing across strains that differ in size, especially when considering metabolic rates which are highly affected by allometric scaling (Rosenfeld et al. 2014b). Because the different strains have different growth rates, strains that were size-matched were not the exact same age (time since hatching). When the option was present, size-matching was prioritized over age-matching.  It is also possible that some of the differences seen in digestive performance are due to adaptation to different food types. Domestic fish have been reared on commercial trout chow and so they should be adapted to it. On the other hand, Tzenzaicut Lake trout are predominantly piscivores as adults, while Pennask Lake trout are insectivores. We might expect Pennask Lake trout to have a longer gut passage time in order to allow digestive and absorptive enzymes more time to process hard to handle content like chitin (Karasov et al. 2011). However, commercial trout chow is designed to be readily digestible; differences that were observed should be due to overall GI tract efficiency and not the ability of one strain to digest some particular food type better than the others. 58  3.5 General conclusions and applications How can my research help future stocking programs in BC? Quite a bit is already known about the different strains and how their survival rate changes in different lakes depending on both predation and productivity (e.g., Biro et al. 2004a; b). My research suggests underlying reasons for why we see differential survival; the domestic strain has an almost insatiable appetite that will increase susceptibility to predation, and Tzenzaicut Lake trout will lose weight faster when placed in an ecosystem where food is limited. When survival is not a concern, but the varied experiences of anglers is, we can consider that although Pennask Lake trout may show a general lower growth rate, slower growth is associated with a higher AS which may translate in a better “fight” for a given size. If on the other hand, anglers prefer large fish, then the fast growing domesticated strain may be the strain of choice. However, if survival of domestics is extremely low, then the angler return per stocked fish may suffer. There has been much emphasis on individuals and their associated differences in metabolic rate and digestive strategy, but this is the first study of its kind where populations that are known to differ in their growth rate are compared and at the same time are tested to see if the same traits that promote high growth between environments (in this case ration level) also promote high growth between populations. I have demonstrated that there are a common set of evolvable traits that promote growth at both the genetic level, and through a plastic response to food ration. These traits include maximum food consumption, growth efficiency, GI tract size, whole body water content, and metabolic rates. While the most pronounced differences in my response variable were found in the domestic strain, many of the general trends between domestics and wild strains were seen between the faster growing wild strain and the slower growing one. This is to be expected given that the difference in growth rate seen between the 59  domestic and wilds strains was much greater than that seen between the two wild strains. Overall, my research suggests common adaptive trade-offs in response to variation in food availability within and between populations. 60  Bibliography  Abbott, J.C., Dunbrack, R.L. & Orr, C.D. (1985) The interaction of size and experience in dominance relationships of juvenile steelhead trout (Salmo gairdneri). Behaviour, 92, 241–253. Abrahams, M.V. & Sutterlin, A. (1999) The foraging and antipredator behaviour of growth-enhanced transgenic Atlantic salmon. Animal Behaviour, 58, 933–942. Afik, D. & Karasov, W.H. (1995) The trade-offs between digestion rate and efficiency in warblers and their ecological implications. Ecology, 76, 2247–2257. Ali, M., Nicieza, A. & Wootton, R.J. (2003) Compensatory growth in fishes: a response to growth depression. Fish and fisheries, 4, 147–190. Álvarez, D. & Metcalfe, N.B. (2007) The tradeoff between catch-up growth and escape speed: variation between habitats in the cost of compensation. Oikos, 116, 1144–1151. Álvarez, D. & Nicieza, A.G. (2005) Is metabolic rate a reliable predictor of growth and survival of brown trout ( Salmo trutta ) in the wild? Canadian Journal of Fisheries and Aquatic Sciences, 62, 643–649. Arendt, J.D. (1997) Adaptive intrinsic growth rates: an integration across taxa. Quarterly Review of Biology, 149–177. Armstrong, J.B. & Bond, M.H. (2013) Phenotype flexibility in wild fish: Dolly Varden regulate assimilative capacity to capitalize on annual pulsed subsidies. Journal of Animal Ecology, 82, 966–975. Armstrong, J.B. & Schindler, D.E. (2011) Excess digestive capacity in predators reflects a life of feast and famine. Nature, 476, 84–87. Arnott, S.A., Chiba, S. & Conover, D.O. (2006) Evolution of intrinsic growth rate: metabolic costs drive trade-offs between growth and swimming performance in Menidia menidia. Evolution, 60, 1269–1278. Barnes, H. & Blackstock, J. (1973) Estimation of lipids in marine animals and tissues: Detailed investigation of the sulphophosphovanilun method for “total” lipids. Journal of Experimental Marine Biology and Ecology, 12, 103–118. Bayne, B.L. (2000) Relations between variable rates of growth, metabolic costs and growth efficiencies in individual Sydney rock oysters (Saccostrea commercialis). Journal of Experimental Marine Biology and Ecology, 251, 185–203. 61  Bayne, B.L. (2004) Phenotypic flexibility and physiological tradeoffs in the feeding and growth of marine bivalve molluscs. Integrative and Comparative Biology, 44, 425–432. Berejikian, B.A. (1995) The effects of hatchery and wild ancestry and experience on the relative ability of steelhead trout fry (Oncorhynchus mykiss) to avoid a benthic predator. Canadian Journal of Fisheries and Aquatic Sciences, 52, 2476–2482. Bergot, P., Blanc, J.M. & Escaffre, A.M. (1981) Relationship between number of pyloric caeca and growth in rainbow trout (Salmo gairdneri Richardson). Aquaculture, 22, 81–96. Billerbeck, J.M., Lankford, T.E. & Conover, D.O. (2001) Evolution of intrinsic growth and energy acquisition rates. I. Trade-offs with swimming performance in Menidia menidia. Evolution, 55, 1863–1872. Billerbeck, J.M., Schultz, E.T. & Conover, D.O. (2000) Adaptive variation in energy acquisition and allocation among latitudinal populations of the Atlantic silverside. Oecologia, 122, 210–219. Biro, P.A., Abrahams, M.V., Post, J.R. & Parkinson, E.A. (2004a) Predators select against high growth rates and risk–taking behaviour in domestic trout populations. Proceedings of the Royal Society of London. Series B: Biological Sciences, 271, 2233. Biro, P.A., Morton, A.E., Post, J.R. & Parkinson, E.A. (2004b) Over-winter lipid depletion and mortality of age-0 rainbow trout (Oncorhynchus mykiss). Canadian journal of fisheries and aquatic sciences, 61, 1513–1519. Biro, P.A. & Stamps, J.A. (2010) Do consistent individual differences in metabolic rate promote consistent individual differences in behavior? Trends in Ecology & Evolution, 25, 653–659. Bochdansky, A.B., Grønkjær, P., Herra, T.P. & Leggett, W.C. (2005) Experimental evidence for selection against fish larvae with high metabolic rates in a food limited environment. Marine Biology, 147, 1413–1417. Bohlin, T., Dellefors, C., Faremo, U. & Johlander, A. (1994) The energetic equivalence hypothesis and the relation between population density and body size in stream-living salmonids. The American Naturalist, 143, 478–493. Booth, D.J. (1990) Effect of water temperature on stomach evacuation rates, and estimation of daily food intake of bluegill sunfish (Lepomis macrochirus Rafinesque). Canadian journal of zoology, 68, 591–595. Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer. 62  Burton, T., Killen, S.S., Armstrong, J.D. & Metcalfe, N.B. (2011) What causes intraspecific variation in resting metabolic rate and what are its ecological consequences? Proceedings of the Royal Society B: Biological Sciences, 278, 3465–3473. Cant, J.P., McBride, B.W. & Croom, W.J. (1996) The regulation of intestinal metabolism and its impact on whole animal energetics. Journal of Animal Science, 74, 2541–2553. Careau, V. & Garland, T. (2012) Performance, personality, and energetics: correlation, causation, and mechanism. Physiological and Biochemical Zoology, 85, 543–571. Careau, V., Thomas, D., Humphries, M.M. & Réale, D. (2008) Energy metabolism and animal personality. Oikos, 117, 641–653. Carey, F.G., Kanwisher, J.W. & Stevens, E.D. (1984) Bluefin tuna warm their viscera during digestion. Journal of Experimental Biology, 109, 1–20. Caviedes-Vidal, E., Karasov, W.H., Chediack, J.G., Fasulo, V., Cruz-Neto, A.P. & Otani, L. (2008) Paracellular absorption: a bat breaks the mammal paradigm. PLoS ONE, 3, e1425. Chappell, M.A., Bech, C. & Buttemer, W.A. (1999) The relationship of central and peripheral organ masses to aerobic performance variation in house sparrows. Journal of Experimental Biology, 202, 2269–2279. Chevan, A. & Sutherland, M. (1991) Hierarchical partitioning. The American Statistician, 45, 90–96. Chippindale, A.K., Chu, T.J.F. & Rose, M.R. (1996) Complex trade-offs and the evolution of starvation esistance in Drosophila melanogaster. Evolution, 50, 753–766. Clark, C.W. (1994) Antipredator behavior and the asset-protection principle. Behavioral Ecology, 5, 159–170. Cutts, C.J., Betcalfe, N.B. & Caylor, A.C. (1998) Aggression and growth depression in juvenile Atlantic salmon: the consequences of individual variation in standard metabolic rate. Journal of Fish Biology, 52, 1026–1037. Desai, M. & Hales, C.N. (1997) Role of fetal and infant growth in programming metabolism in later life. Biological Reviews of the Cambridge Philosophical Society, 72, 329–348. Djawdan, M., Rose, M.R. & Bradley, T.J. (1997) Does selection for stress resistance lower metabolic rate? Ecology, 78, 828–837. Dmitriew, C.M. (2011) The evolution of growth trajectories: what limits growth rate? Biological Reviews, 86, 97–116. Dupont-Prinet, A., Chatain, B., Grima, L., Vandeputte, M., Claireaux, G. & McKenzie, D.J. (2010) Physiological mechanisms underlying a trade-off between growth rate and 63  tolerance of feed deprivation in the European sea bass (Dicentrarchus labrax). The Journal of Experimental Biology, 213, 1143–1152. Duthie, G.G. (1987) Observations of poor swimming performance among hatchery-reared rainbow trout, Salmo gairdneri. Environmental biology of fishes, 18, 309–311. Ege, R. & Krogh, A. (1914) On the Relation between the Temperature and the Respiratory Exchange in Fishes. Internationale Revue der gesamten Hydrobiologie und Hydrographie, 7, 48–55. Farrell, A.P., Hinch, S.G., Cooke, S.J., Patterson, D.A., Crossin, G.T., Lapointe, M. & Mathes, M.T. (2008) Pacific salmon in hot water: applying aerobic scope models and biotelemetry to predict the success of spawning migrations. Physiological and Biochemical Zoology, 81, 697–709. Finstad, A.G., Forseth, T., Ugedal, O. & NæSje, T.F. (2007) Metabolic rate, behaviour and winter performance in juvenile Atlantic salmon. Functional Ecology, 21, 905–912. Fleming, I.A., Agustsson, T., Finstad, B., Johnsson, J.I. & Björnsson, B.T. (2002) Effects of domestication on growth physiology and endocrinology of Atlantic salmon ( Salmo salar ). Canadian Journal of Fisheries and Aquatic Sciences, 59, 1323–1330. Freshwater Fisheries Society of BC. (2004) Rainbow trout strains currently stocked  in BC waters. Gotthard, K. (2001) Growth strategies of ectothermic animals in temperate environments. Environment and animal development, 287–304. Grosell, M., Farrell, A.P. & Brauner, C.J. (2010) Fish Physiology: The Multifunctional Gut of Fish. Academic Press. Grossman, G.D., Rincon, P.A., Farr, M.D. & Ratajczak, R.E. (2002) A new optimal foraging model predicts habitat use by drift‐feeding stream minnows. Ecology of Freshwater Fish, 11, 2–10. Hart, P.J.B. & Gill, A.B. (1992) Constraints on prey size selection by the three-spined stickleback: energy requirements and the capacity and fullness of the gut. Journal of Fish Biology, 40, 205–218. Hartman, G.F. (1965) The role of behavior in the ecology and interaction of underyearling coho salmon (Oncorhynchus kisutch) and steelhead trout (Salmo gairdneri). Journal of the Fisheries Board of Canada, 22, 1035–1081. Houston, A.I., McNamara, J.M. & Hutchinson, J.M.C. (1993) General results concerning the trade-off between gaining energy and avoiding predation. Philosophical Transactions of the Royal Society B: Biological Sciences, 341, 375–397. 64  Johnsson, J.I. & Abrahams, M.V. (1991) Interbreeding with domestic strain increases foraging under threat of predation in juvenile steelhead trout (Oncorhynchus mykiss): an experimental study. Canadian Journal of Fisheries and Aquatic Sciences, 48, 243–247. Karasov, W.H., Martínez del Rio, C. & Caviedes-Vidal, E. (2011) Ecological physiology of diet and digestive systems. Annual Review of Physiology, 73, 69–93. Keeley, E.R., Parkinson, E.A. & Taylor, E.B. (2005) Ecotypic differentiation of native rainbow trout ( Oncorhynchus mykiss ) populations from British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 62, 1523–1539. Killen, S.S., Marras, S. & McKenzie, D.J. (2011) Fuel, fasting, fear: routine metabolic rate and food deprivation exert synergistic effects on risk-taking in individual juvenile European sea bass: Metabolic rate, food deprivation, and risk-taking in sea bass. Journal of Animal Ecology, 80, 1024–1033. Kirkpatrick, M. & Lofsvold, D. (1992) Measuring selection and constraint in the evolution of growth. Evolution, 954–971. Konarzewski, M. & Diamond, J. (1995) Evolution of basal metabolic rate and organ masses in laboratory mice. Evolution, 49, 1239–1248. Kotrschal, A., Rogell, B., Bundsen, A., Svensson, B., Zajitschek, S., Brannstrom, I., Immler, S., Maklakov, A.A. & Kolm, N. (2013) Artificial selection on relative brain size in the guppy reveals costs and benefits of evolving a larger brain. Current Biology, 23, 168–171. Van Leeuwen, T.E., Rosenfeld, J.S. & Richards, J.G. (2011) Adaptive trade-offs in juvenile salmonid metabolism associated with habitat partitioning between coho salmon and steelhead trout in coastal streams. Journal of Animal Ecology, 80, 1012–1023. Levins, R. (1968) Evolution in Changing Environments: Some Theoretical Explorations. Princeton University Press. Lovegrove, B.G. (2003) The influence of climate on the basal metabolic rate of small mammals: a slow-fast metabolic continuum. Journal of Comparative Physiology B, 173, 87–112. Magnan, P., Reinbold, D., Thorgaard, G.H. & Carter, P.A. (2009) Reduced swimming performance and increased growth in domesticated rainbow trout, Oncorhynchus mykiss. Canadian Journal of Fisheries and Aquatic Sciences, 66, 1025–1032. Martin, A.W. & Fuhrman, F.A. (1955) The relationship between summated tissue respiration and metabolic rate in the mouse and dog. Physiological Zoology, 28, 18–34. McCarthy, I.D., Houlihan, D.F. & Carter, C.G. (1994) Individual variation in protein turnover and growth efficiency in rainbow trout, Oncorhynchus mykiss (Walbaum). Proceedings of the Royal Society of London. Series B: Biological Sciences, 257, 141–147. 65  Metcalfe, N.B. (1998) The interaction between behavior and physiology in determining life history patterns in Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences, 55, 93–103. Metcalfe, N.B., Huntingford, F.A. & Thorpe, J.E. (1987) The influence of predation risk on the feeding motivation and foraging strategy of juvenile Atlantic salmon. Animal Behaviour, 35, 901–911. Millidine, K.J., Armstrong, J.D. & Metcalfe, N.B. (2009) Juvenile salmon with high standard metabolic rates have higher energy costs but can process meals faster. Proceedings of the Royal Society B: Biological Sciences, 276, 2103–2108. Nally, R.M. (1996) Hierarchical partitioning as an interpretative tool in multivariate inference. Australian Journal of Ecology, 21, 224–228. Nicieza, A.G., Reiriz, L. & Braña, F. (1994a) Variation in digestive performance between geographically disjunct populations of Atlantic salmon: countergradient in passage time and digestion rate. Oecologia, 99, 243–251. Nicieza, A.G., Reyes-Gavilán, F.G. & Braña, F. (1994b) Differentiation in juvenile growth and bimodality patterns between northern and southern populations of Atlantic salmon (Salmo salar L.). Canadian Journal of Zoology, 72, 1603–1610. Nisbet, R.M., Jusup, M., Klanjscek, T. & Pecquerie, L. (2012) Integrating dynamic energy budget (DEB) theory with traditional bioenergetic models. Journal of Experimental Biology, 215, 1246–1246. Norin, T. & Malte, H. (2012) Intraspecific variation in aerobic metabolic rate of fish: relations with organ size and enzyme activity in brown trout. Physiological and Biochemical Zoology, 85, 645–656. Nylin, S. & Gotthard, K. (1998) Plasticity in life-history traits. Annual review of entomology, 43, 63–83. Orlando, P.A., Brown, J.S. & Whelan, C.J. (2009) Co-adaptations of feeding behaviours and gut modulation as a mechanism of co-existence. Evolutionary Ecology Research, 11, 541–560. Petersson, E., Jaurvi, T., Steffner, N.G. & Ragnarsson, B. (1996) The effect of domestication on some life history traits of sea trout and Atlantic salmon. Journal of Fish Biology, 48, 776–791. Pigliucci, M. (2003) Phenotypic integration: studying the ecology and evolution of complex phenotypes. Ecology Letters, 6, 265–272. Post, J.R. & Parkinson, E.A. (2001) Energy allocation strategy in young fish: allometry and survival. Ecology, 82, 1040–1051. 66  R Core Team. (2013) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Reale, D., Garant, D., Humphries, M.M., Bergeron, P., Careau, V. & Montiglio, P.-O. (2010) Personality and the emergence of the pace-of-life syndrome concept at the population level. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 4051–4063. Reidy, S.P., Kerr, S.R. & Nelson, J.A. (2000) Aerobic and anaerobic swimming performance of individual Atlantic cod. Journal of Experimental Biology, 203, 347–357. Reidy, S.P., Nelson, J.A., Tang, Y. & Kerr, S.R. (1995) Post-exercise metabolic rate in Atlantic cod and its dependence upon the method of exhaustion. Journal of Fish Biology, 47, 377–386. Rosenfeld, J.S., Bouwes, N., Wall, C.E. & Naman, S.M. (2014a) Successes, failures, and opportunities in the practical application of drift-foraging models. Environmental Biology of Fishes, 97, 551–574. Rosenfeld, J., Van Leeuwen, T., Richards, J. & Allen, D. (2014b) Relationship between growth and standard metabolic rate: measurement artefacts and implications for habitat use and life-history adaptation in salmonids. Journal of Animal Ecology, n/a–n/a. Ryer, C.H. & Olla, B.L. (1996) Growth depensation and aggression in laboratory reared coho salmon: the effect of food distribution and ration size. Journal of Fish Biology, 48, 686–694. Sadowska, J., Gebczynski, A.K. & Konarzewski, M. (2013) Basal metabolic rate is positively correlated with parental investment in laboratory mice. Proceedings of the Royal Society B: Biological Sciences, 280, 20122576–20122576. SAS Institute. (1989) SAS/STAT User’s Guide: Version 6. SAS Institute Cary, North Carolina. Scharf, I., Filin, I. & Ovadia, O. (2009) A trade-off between growth and starvation endurance in a pit-building antlion. Oecologia, 160, 453–460. Schoener, T.W. (1971) Theory of feeding strategies. Annual Review of Ecology and Systematics, 2, 369–404. Secor, S.M. (2001) Regulation of digestive performance: a proposed adaptive response. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 128, 563–575. Secor, S.M. & Diamond, J.M. (2000) Evolution of regulatory responses to feeding in snakes. Physiological and Biochemical Zoology, 73, 123–141. 67  Secor, S.M., Stein, E.D. & Diamond, J. (1994) Rapid upregulation of snake intestine in response to feeding: a new model of intestinal adaptation. American Journal of Physiology - Gastrointestinal and Liver Physiology, 266, G695–G705. Sloman, K.A., Motherwell, G., O’Connor, K.I. & Taylor, A.C. (2000) The effect of social stress on the standard metabolic rate (SMR) of brown trout, Salmo trutta. Fish Physiology and Biochemistry, 23, 49–53. Sotiropoulos, J.C., Nislow, K.H. & Ross, M.R. (2006) Brook trout, Salvelinus fontinalis, microhabitat selection and diet under low summer stream flows. Fisheries Management and Ecology, 13, 149–155. Stearns, S.C. (1992) The Evolution of Life Histories. Oxford University Press Oxford. Stevens, E.D. & Devlin, R.H. (2005) Gut size in GH-transgenic coho salmon is enhanced by both the GHtransgene and increased food intake. Journal of Fish Biology, 66, 1633–1648. Stevens, E.D., Wagner, G.N. & Sutterlin, A. (1999) Gut morphology in growth hormone transgenic Atlantic salmon. Journal of Fish Biology, 55, 517–526. Stoks, R., Block, M.D. & McPeek, M.A. (2006) Physiological costs of compensatory growth in a damselfly. Ecology, 87, 1566–1574. Tucker, S. & Rasmussen, J.B. (1999) Using 137Cs to measure and compare bioenergetic budgets of juvenile Atlantic salmon (Salmo salar) and brook trout (Salvelinus fontinalis) in the field. Canadian Journal of Fisheries and Aquatic Sciences, 56, 875–887. Urban, M.C. (2007) The growth-predation risk trade-off under a growing gape-limited predation threat. Ecology, 88, 2587–2597. Valente, L.M.P., Fauconneau, B., Gomes, E.F.S. & Boujard, T. (2001) Feed intake and growth of fast and slow growing strains of rainbow trout (Oncorhynchus mykiss) fed by automatic feeders or by self-feeders. Aquaculture, 195, 121–131. Vøllestad, L.A. & Quinn, T.P. (2003) Trade-off between growth rate and aggression in juvenile coho salmon, Oncorhynchus kisutch. Animal Behaviour, 66, 561–568. Walters, C.J. & Juanes, F. (1993) Recruitment limitation as a consequence of natural selection for use of restricted feeding habitats and predation risk taking by juvenile fishes. Canadian Journal of Fisheries and Aquatic Sciences, 50, 2058–2070. Werner, E.E. & Anholt, B.R. (1993) Ecological consequences of the trade-off between growth and mortality rates mediated by foraging activity. The American Naturalist, 142, 242–272. 68  Werner, E.E. & Hall, D.J. (1988) Ontogenetic habitat shifts in bluegill: the foraging rate-predation risk trade-off. Ecology, 69, 1352–1366. Wieser, W. (1994) Cost of growth in cells and organisms: general rules and comparative aspects. Biological Reviews, 69, 1–33. Yamamoto, T., Ueda, H. & Higashi, S. (1998) Correlation among dominance status, metabolic rate and otolith size in masu salmon. Journal of Fish Biology, 52, 281–290. Young, K.A. (2001) Habitat diversity and species diversity: testing the competition hypothesis with juvenile salmonids. Oikos, 95, 87–93. Zhu, S., Chen, S., Hardy, R.W. & Barrows, F.T. (2001) Digestibility, growth and excretion response of rainbow trout (Oncorhynchus mykiss Walbaum) to feeds of different ingredient particle sizes. Aquaculture Research, 32, 885–893. 69  Appendix A: Data Tables A.1 Fraser Valley strain data Fish ID Ration Initial length (mm) Initial mass (g) Final length (mm) Final mass (g) Specific growth rate (% body mass day-1) Standard metablolic rate (µmol O2 hr-1) Maximum metabolic rate (µmol O2 hr-1) Aerobic scope (µmol O2 hr-1) Gastrointestinal tract mass (g) Stomach length (mm) Anterior intestine length (mm) Dry food eaten (% wet mass day-1) Feed conversion  (wet mass gain / dry mass eaten) Pyloric caeca count Water content (% wet mass) Lipid content (% wet mass) O1O2 H 81 6.2 105 16.13 4.55 - 15.27 - 1.26 19 29 3.44 1.32 - 74.90 7.22 O3B4 M 84 5.97 93 8.89 1.90 6.01 16.05 10.05 0.47 13 18.5 0.88 2.16 - 77.48 3.07 Y3B4 S 80 5.53 82 5.36 -0.15 4.88 20.00 15.12 0.26 11 14 0.00 - - 76.81 3.14 O4B4 H 79 5.39 105 15.76 5.11 - 10.05 - 1.42 20 34.5 3.68 1.39 - 75.32 4.84 O1O3 M 76 5 83 6.87 1.51 5.23 14.95 9.72 0.32 9 18 0.91 1.66 - 76.45 3.51 O1Y2 S 73 4.72 73 4.47 -0.26 5.33 23.94 18.61 0.23 11 16 0.00 - - 77.80 2.49 O2B4 H 73 4.45 94 11.29 4.43 - 15.45 - 1.07 18 28 4.09 1.08 - 76.01 4.60 O1O4 M 71 4.13 79 5.74 1.57 - 22.32 - 0.26 12.5 13.5 0.91 1.73 - 76.12 3.39 Y2B4 S 76 4.68 77 4.36 -0.34 4.25 18.18 13.92 0.21 11.31 14.73 0.00 - - 78.27 2.62 O1R1 H 71 3.93 91 9.47 4.19 7.85 19.49 11.64 0.62 14.8 22.22 3.12 1.34 - 74.80 5.15 O1Y4 M 83 6.73 93 9.14 1.46 4.38 19.19 14.81 0.51 14.7 19.83 0.91 1.60 - 76.43 3.43 O1Y1 S 84 6.92 84 6.16 -0.55 4.74 19.11 14.37 0.24 10.45 16.92 0.00 - - 76.78 3.26 O1Y3 H 80 5.89 112 14.13 4.17 6.62 16.31 9.69 0.85 15.91 22.58 3.69 1.13 55 73.98 5.92 Y1B4 M 78 5.56 87 7.68 1.54 4.45 19.70 15.24 0.35 12.76 16.54 0.90 1.72 - 76.52 3.88 R1B4 S 77 5.16 79 4.79 -0.35 4.54 17.31 12.77 0.23 10.52 12.51 0.00 - - 77.29 3.05 B4Y4 H 75 5.05 112 14.42 5.00 - 18.15 - 1.04 17.05 22.93 3.91 1.28 - 76.94 4.79 O1B1 M 77 5.23 85 7.21 1.53 6.31 22.33 16.02 0.36 13.46 14.63 0.89 1.72 - 75.68 3.61 R4B4 S 77 4.73 76 4.4 -0.34 5.26 21.67 16.41 0.19 9.76 18.98 0.00 - - 77.14 3.01 R2B4 H 74 4.39 100 13.73 5.43 - - - - - - 4.42 1.23 - - - O1R4 S 75 5.69 77 5.27 -0.37 5.50 19.62 14.12 0.23 11.4 15.68 0.00 - - 75.64 4.76 R3B4 H 79 5.43 104 15.65 5.04 7.37 11.67 4.29 1 16.98 27.28 3.92 1.29 - 74.35 5.36 O1B2 M 75 5.33 85 8.02 1.95 6.75 13.35 6.60 0.48 14.69 17.35 0.83 2.34 - 76.63 4.51 O1R3 S 77 5.2 77 4.58 -0.60 4.58 20.48 15.90 0.19 9.53 13.69 0.00 - - 77.31 2.84 B124 H 80 5.51 108 16.75 5.29 7.17 - - 1.04 19.04 29.69 3.69 1.43 - 74.32 5.22 O1B3 M 76 4.79 81 6.36 1.35 7.09 18.24 11.15 0.32 12 20.44 0.91 1.49 - 75.95 3.48 B12O1 S 74 4.53 74 4.04 -0.55 4.85 21.78 16.94 0.16 10.76 15.58 0.00 - - 77.86 2.61 B1B4 H 74 4.45 98 13.03 5.12 7.85 - - 0.83 17.11 23.02 3.59 1.43 - 73.84 4.83 B2B4 M 72 4.1 80 5.9 1.73 - 16.54 - 0.32 12.86 14.54 0.88 1.97 - 76.23 3.12 B3B4 S 71 3.77 70 3.41 -0.48 5.63 20.35 14.73 0.16 10.12 13.87 0.00 - - 78.87 2.28 O1B23 H 81 6.27 101 13.9 3.79 7.48 - - 0.76 15.06 24.48 3.07 1.24 56 73.16 5.54 B234 M 80 5.7 88 8.11 1.68 5.38 19.66 14.28 0.44 13.47 19.17 0.88 1.90 - 77.45 2.99  70   A.2 Pennask Lake strain data Fish ID Ration Initial length (mm) Initial mass (g) Final length (mm) Final mass (g) Specific growth rate (% body mass day-1) Standard metablolic rate (µmol O2 hr-1) Maximum metabolic rate (µmol O2 hr-1) Aerobic scope (µmol O2 hr-1) Gastrointestinal tract mass (g) Stomach length (mm) Anterior intestine length (mm) Dry food eaten (% wet mass day-1) Feed conversion  (wet mass gain / dry mass eaten) Pyloric caeca count Water content (% wet mass) Lipid content (% wet mass) R1B2 H 77 4.3 87 6.28 1.80 5.74 23.53 17.80 0.29 11.97 19.04 2.19 0.82 - 76.28 3.61 O1P4 M 74 3.61 75 4.219 0.74 5.97 23.97 18.01 0.15 10.55 12.15 0.83 0.89 - 77.74 1.76 B1R4 S 72 3.382 71 3.05 -0.49 5.46 22.00 16.53 0.1 8.14 10.4 0.00 - - 77.44 2.63 R1B4 H 70 3.399 83 5.867 2.60 6.72 14.86 8.14 0.3 11.66 18.31 2.35 1.11 56 75.35 4.76 R1B3 M 68 2.768 72 3.43 1.02 6.17 21.71 15.55 0.16 11.61 19.62 0.85 1.20 - 77.97 1.85 O2R4 S 66 2.821 66 2.63 -0.33 4.04 31.21 27.17 0.08 7.87 11.09 0.00 - - 78.06 2.92 B3R4 H 64 2.41 70 3.42 1.67 5.88 23.88 18.00 0.15 9.98 12.3 3.11 0.54 - 75.89 4.10 B2R4 M 61 2.075 64 2.512 0.91 6.50 21.02 14.52 0.09 7.84 12.78 0.93 0.98 - 76.25 3.82 R1O4 S 63 2.313 63 2.07 -0.53 6.02 32.82 26.80 0.07 7.19 10.18 0.00 - - 79.51 2.10 R1O2 H 64 2.261 69 3.41 1.96 - 21.32 - 0.15 8.6 11.6 2.21 0.89 - 75.80 3.76 O3P4 M 73 3.848 76 4.35 0.58 6.01 22.69 16.68 0.17 10.41 11.43 0.64 0.91 - 75.09 3.07 R1Y2 S 74 3.734 73 3.36 -0.50 4.70 29.20 24.51 0.11 9.53 11.93 0.00 - - 78.61 2.82 B24P4 H 66 2.823 73 4.08 1.75 5.76 23.91 18.15 0.18 9.93 17.79 2.34 0.75 - 75.68 4.07 B4P4 M 67 2.682 69 3.24 0.90 - 18.71 - 0.11 9.42 9.88 0.73 1.23 - 77.75 2.77 R4O4 S 66 2.695 65 2.43 -0.49 4.68 28.69 24.01 0.06 8.08 12.21 0.00 - - 78.18 2.88 R1O3 H 66 2.735 75 4.22 2.07 6.72 20.55 13.83 0.21 11.08 13.43 2.87 0.72 - 76.60 3.53 R1G2 M 71 3.492 77 4.28 0.97 5.39 22.19 16.80 0.16 10.67 13.83 0.90 1.07 - 76.00 2.80 R3P4 S 65 2.538 64 2.29 -0.49 4.89 27.20 22.31 0.08 7.96 10.37 0.00 - - 77.11 2.54 R1R4 H 61 2.048 68 3.18 2.10 7.24 18.90 11.67 0.17 9.66 11.06 3.03 0.69 - 75.71 3.42 R1G3 S 79 4.738 78 4.277 -0.49 - 35.64 - 0.14 9.39 12.03 0.00 - - 76.50 3.01 G2P4 H 74 3.772 81 5.423 1.73 6.15 - - 0.24 11.52 11.51 2.70 0.64 - 74.81 2.92 R1R2 M 73 3.723 78 4.521 0.92 - 30.21 - 0.19 10.22 12.98 0.82 1.12 - 76.80 2.66 G4P4 S 69 3.182 69 2.93 -0.39 4.52 24.84 20.32 0.1 8.18 11.31 0.00 - - 75.25 2.65 B23R4 H 69 3.12 76 4.37 1.60 - 14.25 - 0.18 11.01 12.87 2.99 0.54 - 74.69 3.60 B14P4 M 68 3.205 70 3.43 0.32 6.70 23.98 17.28 0.12 7.94 18.03 0.70 0.46 - 75.12 3.56 G3P4 S 69 3.127 67 2.78 -0.56 4.76 24.94 20.19 0.09 7.48 8.58 0.00 - - 76.51 2.32 O14P4 H 69 3.06 76 4.6 1.94 6.91 30.05 23.14 0.21 11.44 10.97 2.70 0.72 - 76.16 3.67 G14R4 M 66 2.663 71 3.47 1.26 6.54 24.25 17.71 0.13 9.1 12.59 0.92 1.37 - 76.65 2.72 O23P4 S 64 2.506 64 2.26 -0.49 5.64 32.67 27.03 0.08 7.49 9.51 0.00 - - 78.04 2.41 G23P4 H 75 4.057 82 5.53 1.47 5.38 26.23 20.85 0.2 11.73 10.94 2.30 0.64 47 74.16 3.50 G1R4 M 73 3.468 76 4.28 1.00 5.90 18.87 12.97 0.13 9.39 15.16 0.92 1.09 - 76.76 3.18 71  A.3 Tzenzaicut Lake strain data Fish ID Ration Initial length (mm) Initial mass (g) Final length (mm) Final mass (g) Specific growth rate (% body mass day-1) Standard metablolic rate (µmol O2 hr-1) Maximum metabolic rate (µmol O2 hr-1) Aerobic scope (µmol O2 hr-1) Gastrointestinal tract mass (g) Stomach length (mm) Anterior intestine length (mm) Dry food eaten (% wet mass day-1) Feed conversion  (wet mass gain / dry mass eaten) Pyloric caeca count Water content (% wet mass) Lipid content (% wet mass) G1B2 M 81 5.23 87 6.521 1.05 4.65 21.89 17.24 0.27 13.12 25.97 0.92 1.14 - 75.17 3.88 O1B4 M 74 3.548 78 4.258 0.87 4.75 21.29 16.54 0.185 11.15 17.01 0.86 1.01 73 76.17 3.51 B1O2 S 68 2.985 68 2.647 -0.57 5.21 22.75 17.54 0.089 9.19 12.49 0.00 - - 76.60 2.95 G1B4 H 67 2.587 73 3.808 1.84 5.83 23.54 17.71 0.176 11.8 16.86 2.74 0.67 - 74.72 3.84 O1B3 M 67 2.309 70 2.879 1.05 6.07 20.95 14.88 0.132 10.66 12.84 0.99 1.06 - 77.22 2.26 G1R2 S 68 2.841 68 2.435 -0.73 6.00 30.60 24.60 0.073 8.91 14.63 0.00 - - 79.66 2.06 G1R3 H 68 2.534 77 4.625 2.87 8.59 18.75 9.85 0.238 12.19 17.44 2.42 1.18 - 75.45 4.08 R1O2 M 82 5.197 88 6.379 0.98 5.81 19.04 13.23 0.285 12.92 14.13 0.92 1.06 - 75.25 3.65 O1R4a S 82 4.985 83 4.535 -0.45 4.79 23.12 18.34 0.154 10.69 17.69 0.00 - 58 - 2.98 O1R2 H 78 4.18 89 6.907 2.39 6.38 22.78 16.40 0.387 13.06 23.18 2.72 0.88 62 74.63 4.38 O2R3 M 77 4.163 81 5.159 1.02 6.20 21.37 15.17 0.221 11.38 14.02 0.93 1.10 51 76.66 3.43 G1R4 S 74 3.718 74 3.212 -0.70 6.01 31.54 25.52 0.094 7.2 16.19 0.00 - - 78.56 2.77 O1R3 H 72 3.492 85 6.386 2.87 9.09 17.10 7.70 0.327 12.68 24.74 3.25 0.88 - 73.11 4.85 G1O3 M 73 3.455 76 4.248 0.98 6.46 20.24 13.78 0.176 10.65 14.35 0.92 1.07 - 75.91 3.09 O1O4 S 72 3.281 71 2.891 -0.60 4.36 22.05 17.68 0.088 9.17 11.89 0.00 - - 77.26 2.67 O1O2 H 72 3.354 80 4.898 1.80 4.50 21.46 16.96 0.242 12.3 17.75 2.33 0.77 69 75.39 4.75 G1O4 M 72 3.207 75 3.857 0.88 4.91 25.98 21.07 0.194 10.96 16.6 0.91 0.97 - 76.10 4.46 O2O3 H 75 3.965 90 7.984 3.33 - 20.41 - - - - 2.59 1.29 - 76.66 3.39 O1O3 S 75 3.813 74 3.359 -0.60 4.01 27.05 23.03 0.107 9.01 14.01 0.00 - 60 77.78 1.84 O12Y2 S 73 3.71 72 3.189 -0.72 4.08 21.39 17.32 0.101 9.16 13.27 0.00 - - 77.10 3.95 G1Y2 H 77 3.953 84 6.152 2.11 6.08 20.95 14.87 0.295 13.23 16.4 2.51 0.84 - 73.93 4.41 G1Y4 M 74 3.804 78 4.79 1.10 - 16.90 - - - - 0.92 1.19 - - - O1Y3 S 73 3.53 72 3.126 -0.58 4.43 24.64 20.21 0.107 9.25 18.79 0.00 - - 77.10 2.58 Y1O2 H 74 3.529 81 5.278 1.92 - 12.92 - - - - 2.62 0.73 - - - O1Y4 M 68 2.887 72 3.645 1.11 4.86 22.53 17.67 0.169 11.37 12.58 0.93 1.19 - 76.46 3.15 G1Y3 S 69 2.266 69 1.932 -0.76 5.24 30.61 25.37 0.062 7.13 11.13 0.00 - 66 80.23 2.24 O2Y3 H 84 5.636 96 9.25 2.36 5.71 22.48 16.77 0.512 16.14 18.73 2.47 0.96 - 74.22 5.06 O1R4b M 79 4.432 85 5.735 1.23 4.89 18.99 14.10 0.243 12.05 20.77 0.88 1.39 64 - 3.23  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items