UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Performance of wild and domestic strains of diploid and triploid rainbow trout (Oncorhynchus mykiss)… Scott, Mark Adam 2012

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

Item Metadata

Download

Media
24-ubc_2012_fall_scott_mark.pdf [ 1009.04kB ]
Metadata
JSON: 24-1.0073196.json
JSON-LD: 24-1.0073196-ld.json
RDF/XML (Pretty): 24-1.0073196-rdf.xml
RDF/JSON: 24-1.0073196-rdf.json
Turtle: 24-1.0073196-turtle.txt
N-Triples: 24-1.0073196-rdf-ntriples.txt
Original Record: 24-1.0073196-source.json
Full Text
24-1.0073196-fulltext.txt
Citation
24-1.0073196.ris

Full Text

Performance of wild and domestic strains of diploid and triploid rainbow trout (Oncorhynchus mykiss) in response to environmental challenges by Mark Adam Scott  B.Sc., The University of British Columbia, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  September 2012  © Mark Adam Scott 2012  Abstract  To determine what may contribute to the poorer survival of triploid (3n) trout in lake stocking programs relative to their diploid (2n) counterparts, we compared whole animal performance in response to environmental challenges in juvenile 2n and 3n fish from four wild strains and one domestic strain of rainbow trout. Spanning four years (2008, 2009, 2010, and 2011), wild fish were caught from nature and spawned in-hatchery along with hatchery-reared domestic trout. Offspring from all strains were raised to eight months as both 2n and 3n and exposed to low oxygen, swimming, and high temperature challenges. The only measure of performance to show a consistent difference between 2n and 3n individuals across all strains was time to loss of equilibrium (LOE) as a result of hypoxia exposure (~10% air saturation, 16 torr). Triploid trout always showed a shorter time to LOE (by 15-86% depending on the strain) relative to their 2n counterparts, with the exception of lake reared trout which showed no significant differences between 2n and 3n time to LOE. Additionally, there were no consistent effects of ploidy on critical oxygen tension, ṀO2, critical swimming speed (Ucrit), critical thermal maxima (CTMax), or muscle enzyme activities. We observed significant effects of strain on all performance measures except for CTMax. In general, the Fraser Valley domestic strain had higher Ucrit, higher ṀO2, and greater muscle enzyme activities than did Blackwater, Tzenzaicut, and Pennask wild conspecifics, suggesting that domestication affects a variety of traits in addition to growth rates.  ii  Preface  Chapter 2 of this thesis is co-authored. The research described in Chapter 2 was conducted by Mark A. Scott under the supervision of Dr. Jeffrey G. Richards, with the exception of CTMax which was collected by Wallace Cheung (UBC) and growth rates, which were collected by the fish culturists at the Fraser Valley Trout Hatchery (Abbotsford, B.C.). I wrote all three chapters of this thesis and received editorial feedback from Drs. J Richards, P Schulte & R Devlin. All procedures involving animals were performed in accordance with protocols approved by the UBC Animal Care Committee, certificate A09-0611.  iii  Table of Contents  Abstract ................................................................................................................................................. ii Preface ................................................................................................................................................. iii Table of contents ...................................................................................................................................iv List of tables ..........................................................................................................................................vi List of figures ........................................................................................................................................ vii Acknowledgments ............................................................................................................................... viii Dedication .............................................................................................................................................ix 1  2  General introduction ........................................................................................................................ 1 1.1  Rainbow trout as a model organism ...................................................................................................... 1  1.2  Aquaculture and lake stocking ............................................................................................................... 1  1.3  Triploidy.................................................................................................................................................. 3  1.4  Performance in response to environmental challenges ........................................................................ 5 1.4.1 Do 2n and 3n differ in aerobic capacity? .................................................................................... 6 1.4.2 Do 2n and 3n differ in thermal tolerance? ................................................................................. 8 1.4.3 Do 2n and 3n differ in hypoxia tolerance? ................................................................................. 9  1.5  Strain and family effects ...................................................................................................................... 11  1.6  Improving performance in response to environmental challenges ..................................................... 12  1.7  Objectives ............................................................................................................................................. 13  Performance of wild and domestic strains of diploid and triploid rainbow trout (Oncorhynchus mykiss) in response to environmental challenges .................................................... 15 2.1  Introduction ......................................................................................................................................... 15  2.2  Materials and methods ........................................................................................................................ 17 2.2.1 Experimental animals ............................................................................................................... 17 2.2.2 Swimming performance ........................................................................................................... 19 2.2.3 Routine metabolic rate and critical oxygen tensions ............................................................... 20 iv  2.2.4 2.2.5 2.2.6 2.2.7 2.2.8  3  Thermal tolerance .................................................................................................................... 21 Hypoxia tolerance ..................................................................................................................... 22 Muscle enzyme activities .......................................................................................................... 24 Ploidy confirmation .................................................................................................................. 25 Statistical analysis ..................................................................................................................... 25  2.3  Results .................................................................................................................................................. 26 2.3.1 Swimming performance ........................................................................................................... 26 2.3.2 Routine metabolic rate ............................................................................................................. 27 2.3.3 Critical oxygen tensions ............................................................................................................ 28 2.3.4 Thermal tolerance .................................................................................................................... 28 2.3.5 Hypoxia tolerance ..................................................................................................................... 28 2.3.6 Muscle enzyme activities .......................................................................................................... 30 2.3.7 Specific growth rates ................................................................................................................ 31  2.4  Discussion ............................................................................................................................................. 32 2.4.1 Effects of ploidy on performance in response to environmental challenges........................... 32 2.4.2 Effects of strain on performance in response to environmental challenges ........................... 38 2.4.3 Recommendations and Conclusion .......................................................................................... 43  General discussion and conclusions ................................................................................................ 55 3.1  Overview .............................................................................................................................................. 55  3.2  The effects of ploidy and strain on performance in response to environmental challenges .............. 56  3.3  Future directions .................................................................................................................................. 58  References........................................................................................................................................... 60  v  List of tables  Table 2.1  Weights and lengths at time of analysis and correlation coefficients for selected measures of performance in response to environmental challenges of four strains of 2n and 3n rainbow trout ......................................................................................................... 45  Table 2.2  Critical thermal maximum of three strains of 2n and 3n rainbow trout..................................... 46  Table 2.3  Time to LOE of three strains of 2n and 3n adult lake rainbow trout .......................................... 47  Table 2.4  Muscle pyruvate kinase, citrate synthase, and lactate dehydrogenase enzyme activities per gram wet weight of four strains of 2n and 3n rainbow trout................................ 48  Table 2.5  Specific growth rates (weights and lengths) taken between 1110 and 1530 ATUS of four strains of 2n and 3n rainbow trout from the 2009 brood year ........................................... 49  vi  List of figures  Figure 2.1  Time to LOE of four strains of 2n and 3n rainbow trout........................................................... 50  Figure 2.2  Critical oxygen tensions of four strains of 2n and 3n rainbow trout ........................................ 51  Figure 2.3  Critical swimming speeds of four strains of 2n and 3n rainbow trout ..................................... 52  Figure 2.4  Mass-specific oxygen consumption rates of four strains of 2n and 3n rainbow trout............. 53  Figure 2.5  Time to LOE of two strains of 2n and 3n rainbow trout ........................................................... 54  vii  Acknowledgements  I would like to begin chronologically by thanking my undergraduate directed studies supervisor Dr. Jason Bystriansky for being the first person to give me the opportunity to participate in real research. I would also like to thank him for introducing me to his collaborator Dr. Trish Schulte, whose lab I had the privilege of working in for a second 448 term. Next, I would like to acknowledge the contributions of Trish to my graduate career for whatever initial arm twisting she did to convince Dr. Jeffrey Richards to take me on as a Masters student in his lab. It is to Jeff that I offer the most gratitude to for his patience and tolerance thruout the years and of course, for the titillating afternoon confabs. I must also acknowledge the essential contributions of my third committee member, Dr. Bob Devlin, and collaborator Dr. Rush Dhillon. I would additionally like to thank the following people that I worked with over the years on components of research directly related to my thesis: Tammy Rodela, Milica Mandic, Gigi Lau, Lili Yao, Andrew Thompson, Sara Northrup, Adrian Clarke, Dave Allen, Patrick Tamkee, Travis Van Leeuwen, Jodie Atkinson, and the four undergraduate students that I was fortunate to have the experience of working with (Jason Au, Derrick Groom, Benjamin Duchen, and Wallace Cheung). Finally, I would like to thank some individuals who were a source of advice, wisdom, and reassurance over the past few years: Anne Dalziel, Ben Speers-Roeche, Matthew Regan, Matt Casselman, Tim Healy, Ryan Shartau, Dan Baker, Graham Scott, Yuxiang Wang, Bob Shadwick, Colin Brauner, Bill Milsom and to the rest of the Zoology department for support whenever I gave a poster or talk.  viii  For my grandparents  ix  1  General introduction  1.1 Rainbow trout as a model organism Rainbow trout (Oncorhynchus mykiss) are salmonids native to the lakes and rivers of North America west of the Rocky Mountains. Unlike steelhead trout (which are members of the same species), they are typically not anadromous and do not make migrations to seawater. Rainbow trout in British Columbia are believed to have colonized the province after moving north from California following the retreat of ice after the last ice age (MuCuskar et al., 2000). Rainbow trout have been introduced to many geographic regions around the world for food and sport. The commercial importance of rainbow trout is likely the largest contributor to its widespread use as a model organism in physiological research and with their fast growth rates and good exercise performance (Houlihan and Laurent, 1987), rainbow trout are widely used in aquaculture and lake stocking programs.  1.2  Aquaculture and lake stocking Modern practices of aquaculture in North America began in the mid-1800’s (Milner, 1874) and  encompass the farming of aquatic organisms such as fish, molluscs, crustaceans, and aquatic plants. Fish farming is the most widespread form of aquaculture and usually involves the spawning and rearing of fish in a hatchery. The most abundantly farmed fish found in hatcheries around the world are salmon, carp, tilapia, and trout, which are all economically important. According to estimates from 2010, the total amount of rainbow trout harvested worldwide from fish farming was over 700,000 tonnes with a value of over $US 2.5 billion (FAO, 2010). The primary use of rainbow trout is for lake stocking which is required to mitigate the impacts of intense fishing activity on more vulnerable wild fish populations, particularly those in lakes and rivers near urban areas (Almodovar and Nicola, 1998; Almodovar et al., 2001a). The production of trout for  1  lake stocking involves the spawning of fish from brood stocks and rearing their offspring in a hatchery. These fish can be reared until they are fry, yearlings, or of catchable size before being stocked into lakes, depending on the demands of anglers. The Freshwater Fisheries Society of B.C. alone stocked more than 1,300 lakes between 2001 and 2004 (FFSBC, 2004). In addition to rainbow trout, other species of fish stocked in B.C. lakes are brook trout, cutthroat trout, and kokanee salmon. Domestication is an important component of fish farming and involves the artificial selection of fish for increased growth rates. From the early 1900’s to the present around 97% of aquatic species harvested from aquaculture have been domesticated (Duarte et al., 2007). Improved growth rates are a desirable characteristic of fish used in lake stocking programs because of the increased benefit to cost ratio associated with fast growing fish, which have been estimated to range from 5:1 to 50:1 (Gjerde, 1986). In addition to rainbow trout, selective breeding programs are common with many other species of fish important for aquaculture such as carp (Cyprinus carpio; Moav and Wohlfarth, 1976), channel catfish (Ictalurus punctatus; Bondari, 1983), and Atlantic salmon (Salmo salar; Gjedrem, 1979), among others. A consequence of stocking hatchery reared fish, particularly domesticated fish, into lakes is that they may interbreed with native populations, altering the genetic pool of resident fish. These hybridizations are known to threaten the naturally occurring genetic diversity of the native population (Machordom et al., 1999, 2000; Almodovar et al., 2001b), possibly resulting in a loss of biodiversity or phenotypic responses to environmental challenges. Common phenotypic responses that can be lost as a result of hybridization are resistance to disease as well as tolerance to stressors such as changing thermal and oxygen conditions (Ferguson, 2006).  2  1.3  Triploidy Since triploid (3n) trout were discovered in nature (Cueller and Uyeno, 1972) and the  development of technologies to induce triploidy in trout hatcheries that followed, 3n trout have served as the mainstay for lake stocking programs (Purdom, 1993; Benfey, 1999; Liu et al., 2001). Triploidy can occur naturally in salmonids but the rate of occurrence is very low (3-5%; Cuellar and Uyeno, 1972; Cormier et al. 1993). However, high occurrences of triploidy have been observed in a domesticated hatchery stock where six of eleven individuals of a particular family were found to be 3n (Thorgaard and gall, 1979). Beneficial qualities of triploidy in fish are enhanced flesh quality and increased growth rates at maturation (Galbreath et al., 1994; FFSBC, 2004), making them a desirable fish for lake stocking programs. However, the induction of triploidy can sometimes lead to mortality events throughout the hatchery rearing life stage, reduced growth rates, and high rates of jaw abnormalities (McGeachy et al., 1995; McCarthy et al., 1996; reviewed by Benfey, 2001). Triploidy can be artificially induced by one of several different shock treatments, which include thermal, chemical, pressure, and electrical shocks. These strategies for artificially inducing triploidy are not used only in salmonids (Chourrout, 1980; Thorgaard et al., 1981; Refstie et al., 1982) but have been successfully applied to zebrafish (Brachydanio rerio; Streisinger et al., 1981), three-spined stickleback (Gasterosteus aculeatus; Swarup, 1958), and tilapia (Tilapia aurea; Valenti, 1975), among others. Shock treatments can vary considerably in their success rates but the treatment with the highest efficiency and the one that is most commonly used is pressure shocking (Johnstone, 1985; Benfey et al., 1988). Standard practice for trout is to apply, at thirty minutes post-fertilization and at roughly 8-10oC, a 10,000 psi pressure shock for five minutes. When done correctly, this process disrupts spindle fibre formation during meiosis causing the chromosomes to misalign which results in retention of the second polar body (Chourrout, 1984; Benfey, 2001). The second female chromosome remaining intact along with the one chromosome from the male results in triploidy. 3  Since normal 2n and 3n individuals are externally morphologically identical, to ensure a high proportion of fish in a brood stock have had triploidy successfully induced it is necessary to confirm 3n status. There are two common ways in which 3n status is confirmed and both involve analysis of blood samples from a subset of individuals from the brood stock of interest. One relatively straightforward method is to measure erythrocyte dimensions which under a microscope are larger in 3n fish compared to 2n fish (Benfey, 1999). A preferred alternative to taking measurements of cell dimensions through a microscope is to use a Coulter Counter Channelyzer to measure erythrocyte volume which can take measurements of more than 200,000 cells from a single sample in less than five minutes (Benfey et al., 1984). Instead of measuring cell size, the second common way in which triploid status is confirmed is by measuring DNA content with flow cytometry (Thorgaard et al., 1982; Allen, 1983). Measuring stained nuclei with flow cytometry is rapid and several samples can be measured per minute. In comparison of the different methods of confirming 3n status, flow cytometry is preferred because of its high throughput and accuracy; although measuring cell volume is more common because it has comparable accuracy and the equipment required is cheaper and more easily accessed (Benfey et al., 1984; Benfey, 2001). The fundamental differences between 2n and 3n cells arises from the fact that 3n cells have additional genetic material which results in larger cells. Two consequences of larger cells are decreased surface area to volume ratios and at the tissue and organ level, fewer cell numbers (Benfey, 2001). It is thought that larger cells with lower surface area to volume ratios might have impaired nutrient and metabolite exchange both within cells and, depending on how cell shape might be affected, between cells (Benfey, 2001; Maxime, 2008). Several processes could be affected by impaired exchange such as protein synthesis and signal transduction cascades which at the whole organism level might negatively impact an individual’s ability to mediate responses to environmental challenges.  4  1.4  Performance in response to environmental challenges Rainbow trout are generally considered to be relatively sensitive to environmental challenges  including extreme temperature and hypoxia (Molony, 2001); however, some populations have been found to survive at temperatures ranging from 0.0oC to 29.8oC (Rogers and Griffiths, 1983; Currie et al., 1998). Although some populations of trout may be able to survive over large ranges in environmental conditions, in general, the range of conditions that yield optimal growth and reproduction is quite narrow (Peterson and Meador, 1994). Some common environmental challenges that cause mortality in lake stocked trout are high and low water temperatures. Secondary effects of elevated water temperatures include decreased oxygen levels (as warmer water holds less oxygen) as well as oxygen depletion from algal blooms that are more common at higher temperatures. Secondary effects of low water temperatures, specifically when ice cover is present, are low oxygen from the obstructed mixing of air and water and inhibited photosynthesis of water plants. If the extent to which light cannot penetrate the surface of the water is too severe, the death and ensuing decomposition of vegetation can contribute to further reducing oxygen in already hypoxic water. In addition to being subjected to seasonal bouts of extreme temperature and hypoxia, since trout are a predatory fish they must also forage for food while contending with suboptimal environmental conditions. Studies have shown that both temperature and hypoxia can negatively affect swimming performance and in some cases contribute to greater mortality (Jones, 1971; Hyndman, et al., 2003). Due to the wide range of environmental scenarios that lake stocked trout can be faced with, it is necessary to develop a broad characterization of performance in response to environmental challenges of hatchery reared trout to ensure that a lake stocking program can become sustainable. In order to ensure 3n rainbow trout lake stocking can be sustainable, many laboratory studies comparing 2n and 3n responses to environmental challenges have been performed (reviewed by Benfey, 1999 and Maxime, 2008; Benfey, 2001). Some aspects of triploidy are very well understood 5  such as induction methods, detection of triploidy, and growth but other aspects such as our understanding of basic 3n physiology is still developing. Although there is a general consensus that 3n fish show reduced tolerance to environmental challenges (reviewed by Maxime, 2008), there is a great deal of debate in the literature about the effects of triploidy on specific performance measures such as thermal tolerance and hypoxia tolerance. One area where there is agreement is on the differences between the hematology of 2n and 3n fish. Despite 2n and 3n fish having different cell sizes and cell numbers their percentage of total blood volume (hematocrit) is roughly the same (Sezaki et al., 1991; Parsons, 1993; Biron and Benfey, 1994). Conversely, the effects of triploidy on red blood cell hemoglobin are much less clear. Erythrocyte hemoglobin content is higher in 3n fish compared to 2n fish due to a larger cell volume (Aliah, et al., 1991; Sezaki et al., 1991; Parsons, 1993) but total blood hemoglobin has been reported by some studies to be similar in 2n and 3n trout (Aliah, et al., 1991; Stillwell and Benfey, 1994, 1996a) and in other studies it has been reported to be lower in 3n trout compared to 2n trout (Graham et al., 1985; Small and Randall, 1989; Parsons, 1993). It is important to identify if blood hemoglobin levels are different between 2n and 3n because of the role that oxygen uptake and transport plays in determining aerobic capacity (Bernier et al., 1996).  1.4.1 Do 2n and 3n trout differ in aerobic capacity? Aerobic capacity describes the level of aerobic fitness of fish and has been thoroughly characterised in trout (Altimiras et al., 2002; Stillwell and Benfey, 1997). It is a valuable indicator of performance in response to environmental challenges because it describes a fish’s ability to uptake and transport oxygen which is important under conditions of oxygen limitation (Aliah et al., 1991; Yamamoto and Iida, 1994). Studies measuring standard, routine, and active metabolic rates have all found no differences between 2n and 3n fish (Sezaki et al., 1991; Parsons, 1993; Yamamoto and Iida, 1994). Standard metabolic rate describes a theoretical baseline value of oxygen consumption that is 6  thought to represent the aerobic energy expenditure of homeostatic processes. Routine metabolic rate is a more biologically relevant measure and includes all of the components of standard metabolic rate but with the added aerobic energy expenditure of daily processes such as digestion and minimal levels of activity. Active metabolic rates similarly involve the aerobic energy expenditure of daily processes but with an added environmental performance challenge such as a swimming test. The difference between the maximum and minimum possible aerobic energy expenditure obtained for an individual is its aerobic scope. No difference between 2n and 3n in the above measures of metabolic rates suggests that 2n and 3n do not differ in aerobic capacity. Swimming performance is commonly used as an alternative measure of aerobic capacity. Critical swimming tests and exhaustive chases are two different kinds of assessments of swimming performance. Exhaustive chases tend to be high intensity and short duration which can make them good proxies for anaerobic capacity whereas critical swimming tests generally tend to be used as indicators of aerobic capacity. Critical swimming tests are initially of low intensity and ramp up over a relatively longer duration to a high intensity. Since this test ends in exhaustion there is an inherent anaerobic component to it but the component is assumed to be small because the high intensity swimming occurs over a relative short duration. Several studies have compared 2n and 3n critical swimming speeds (Virtanen et al., 1990; Sezaki et al., 1991; Stillwell and Benfey, 1996b) but despite there being conflicting findings of blood hemoglobin levels between 2n and 3n fish, there seems to be a general consensus that 2n and 3n fish do not differ in their aerobic capacity as assessed by swimming performance (reviewed by Benfey, 1999 and Maxime, 2008). However there have been studies suggesting otherwise. Virtanen et al. (1990) determined 3n compared to 2n rainbow trout have a reduced aerobic capacity based on the swelling of erythrocytes and the accumulation of anaerobic metabolites in 3n compared to 2n red blood cells following exercise.  7  1.4.2 Do 2n and 3n trout differ in thermal tolerance? Tolerance to extremes in temperature is commonly assessed in hatchery-reared trout because it can be an indicator of survival in new environments, especially environments prone to water temperature fluctuations (Molony, 2001). The two most common assessments of thermal tolerance are critical thermal minimum (CTMin) and CTMax which are generally ramp tests lasting from only a couple hours to over ten hours (Elliot and Elliot, 1995). They consist of altering the temperature of water surrounding a solitary fish or a group of fish from an acclimation temperature to an endpoint that is usually LOE or death (Currie et al., 1988). The rate of thermal change is usually held constant throughout the test and varies between studies from 0.1oC min-1 to 0.5oC min-1 (Boyd and Tucker, 1998). CTMax tests are more common than CTMin tests because the environments that trout are stocked in tend to be more prone to upper thermal extremes than lower thermal extremes. Due to the commercial importance of rainbow trout many studies have been done assessing CTMax, which averages 24-26oC for trout (Bidgood, 1980). Few studies have investigated the thermal tolerance of 2n and 3n rainbow trout, but two that have measured CTMax and did not find any differences in time to loss of equilibrium between 2n and 3n trout (Benfey et al., 1997; Galbreath et al., 2006). However, Hyndman et al. (2003) showed that there were differences in the ability of 2n and 3n trout to recover in warm water following exhaustive exercise with 3n trout showing a significantly higher rate of mortality than 2n trout. Similarly, Ojolick, et al. (1995) found that female 3n brook trout (Salvelinus fontinalis) did not grow or survive as well as their diploid counterparts when reared at high temperatures. In light of these contrasting findings from this limited number of studies comparing 2n and 3n CTMax and growth and survival at high rearing temperatures it is difficult to draw any conclusions about the effects of triploidy on thermal tolerance. However, one possible explanation could be that CTMax tests are either an exposure that is too severe or a duration that is too short to resolve subtle differences between 2n and 3n thermal tolerance. For 8  example, Wagner et al. (2001) observed greater variation between populations of cutthroat trout in loss of equilibrium in response to thermal challenges over a 96 hour test compared to a CTMax test. Perhaps comparisons of 2n and 3n thermal tolerance would benefit from a less severe exposure over an extended duration (Elliot and Elliot, 1995).  1.4.3 Do 2n and 3n trout differ in hypoxia tolerance? One of the most important factors in determining whether or not lake stocked trout will be successful in their new environment is dissolved oxygen levels in the water. For rainbow trout and other salmonids, water oxygen levels of less than 5.0-6.0 mg L-1 can result in mortality (Doudoroff and Shumway, 1970; Weithman and Haas, 1984; Molony, 2001) or impaired growth rates (MacConnell, 1989), depending on the temperature. Interestingly, one study has actually attributed annual economic loss to a county in Missouri (USA) of $267,000-$432,000 as a result of low concentrations of dissolved oxygen in a local lake (Weithman and Haas, 1984). Tolerance of low dissolved oxygen in rainbow trout has been well characterized (Dunn and Hochachka, 1986; Boutilier et al., 1988; Bernier et al., 1996). When faced with a low oxygen challenge, typical responses of fish are to attempt aquatic surface respiration, decrease locomotor activity, and increase ventilation rate (Gee and Gee, 1991; Nilsson et al., 1993; Chapmen et al., 1994). Rainbow trout are relatively intolerant of hypoxia and are not capable of more drastic physiological responses to hypoxia such as gill surface remodelling and metabolic rate suppression to increase the surface area of the gill available for oxygen uptake and to conserve energy, respectively (Nilsson, 2007; Richards, 2011). A common method of assessing hypoxia tolerance is to expose fish to a severe hypoxic challenge and measure the time at which they lose equilibrium (Chapmen et al., 1994; Yamamoto and Iida, 1994). The value of this type of test is that the hypoxia tolerance of many fish can be assessed at a single time all under the exact same experimental conditions. Another common high-throughput way to 9  assess hypoxia tolerance is to lower oxygen levels at a constant rate and measure the concentration at which fish LOE (Chapmen et al., 1994; Lilyestrom et al. 1999). An alternative method to measuring time to LOE for determining hypoxia tolerance is to measure Pcrit (Mandic, et al., 2008; Speers-Roesch et al., 2012), which has been described as a whole-animal measure of the ability of fish to acquire oxygen from their environment (Gannon et al., 1999). It is the point at which fish transition from an oxygen regulating strategy to an oxyconforming strategy. Fish with lower Pcrit values are able to maintain oxygen uptake at a lower environmental oxygen tension than fish with higher Pcrit values, indicating greater hypoxia tolerance (Chapman and McKenzie, 2009). The three methods briefly described above are all used to assess hypoxia tolerance in hatchery reared trout (Klar et al., 1979). There have been few studies comparing the hypoxia tolerance of 2n and 3n fish. However, Yamamoto and Iida (1994) using rainbow trout and Lilyestrom et al. (1999) with catfish hybrids (Ictalurus punctatus x Ictalurus furcatus) both found that 3n compared to 2n have reduced hypoxia tolerance. The authors attributed the impaired 3n tolerance to a reduced oxygen uptake and blood oxygen carrying capacity as a result of 3n fish having reduced blood hemoglobin levels (Parsons, 1993; Sadler et al., 2000). A reduction of blood hemoglobin could limit oxygen uptake during transit through the gills. Reduced gill surface area in 3n fish compared to 2n fish would further limit oxygen uptake from the environment, as has been found in 3n Atlantic salmon (Sadler et al., 2001). Although the study of 2n and 3n tolerance of hypoxia is still in its infancy, the evidence accumulated so far suggests that 3n have greater sensitivity to hypoxia than their 2n counterparts. To summarize, even though the majority of studies comparing 2n and 3n performance in response to environmental challenges only show significant differences between 2n and 3n in hypoxia tolerance (reviewed by Benfey, 1999 and Maxime, 2008), there have been studies showing reduced 3n swimming performance and thermal tolerance (Virtanen et al., 1990; Hyndman et al., 2003). Additionally, other studies have shown greater 3n mortality in natural environments (Cotter et al., 2000; 10  Koenig et al., 2011), in net pens (Withler et al., 1995), and under suboptimal rearing conditions (Ojolick et al., 1995; Hyndman et al., 2003; Altimiras et al., 2002). It is likely that the observed differential 2n/3n mortality in natural systems is multifaceted, with interactive effects of environmental stressors overwhelming the abilities of 3n fish to cope with environmental challenges. Since there is still relatively little known about the physiology of 3n fish, more research on 2n/3n performance in response to environmental challenges is warranted. To effectively maintain recreational fishing programs in natural lakes with fluctuating environmental conditions robust 3n fish with good tolerance to changing environmental conditions are needed.  1.5  Strain and family effects Strain refers to geographically isolated populations of fish that are usually locally adapted to the  specific suite of environmental conditions within which they reside. When the glaciers covering B.C. receded the previously inaccessible lakes were colonized by rainbow trout and other fish. As habitat conditions changed over the last 10,000 years, the rainbow trout living in these lakes have been forced to adapt with them. The varied habitats now present across the province have resulted in many genetically distinct populations of rainbow trout that vary in performance and tolerance traits (FFSBC, 2004; Clarke, 2012). Strain effects describe variation between populations whereas family effects can be used to describe the variation within a population. The important component of these locally adapted traits is that they are heritable. The most commonly described genetic-based traits are related to growth and resistance to disease (Nagai, et al., 2004; Gjerde, 2005; Silverstein el al., 2008). Intraspecific variation can be utilized to selectively breed strains with enhanced ability for a given trait. For example, different strains and families of a species sometimes differ in growth rates (Moav et al., 1964, 1975) and these differences in growth rates can be utilized to selectively breed strains with enhanced growth (Purdom, 1993). The desire to breed fish with higher growth rates has 11  progressed into the need to develop stocks with greater tolerance to environmental perturbations. For example, rainbow trout from Pennask lake (referred to as Pennask trout strain) that are stocked in B.C. have a greater ability to store lipids compared to other strains which makes them a suitable fish to stock in lakes that experience longer winters and cooler temperatures (Clarke, 2012). In addition to thinking about intraspecific variation in terms of its benefits for selective breeding programs, it must also be considered when making comparisons of 2n and 3n responses to environmental challenges. Intraspecific differences in traits such as thermal tolerance and swimming performance have been found in fish (Klar et al., 1979; Fangue et al., 2008; Dalziel and Schulte, 2012), so when no differences are observed between 2n and 3n swimming performance and thermal tolerance (Sezaki et al., 1991; Stillwell and Benfey, 1996b; Benfey et al., 1997; Galbreath et al., 2006) it might not necessarily reflect a lack of difference between 2n and 3n fish for that species but simply a lack of difference within that particular strain (Guo et al., 1990). Similarly, strain effects may also account for some of the discrepancies between studies that show differences in 2n/3n responses to environmental challenges and studies that show no difference in 2n/3n responses. Therefore, it is important to consider strain, and if possible, use multi-strain comparisons when assessing 2n/3n responses to environmental challenges (Benfey, 1999).  1.6  Improving performance in response to environmental challenges As a result of the commercial importance of rainbow trout and other hatchery-reared fish a lot  of work has been done to improve performance in response to environmental challenges of lake stocked trout. One suggestion to improve performance and tolerance has been to expose fish to an environmental challenge or training stage during the rearing process in order to improve performance later on (Davison, 1989; Farrell, et al., 1991; Currie et al., 1998). An alternative is to use selective breeding programs to introduce desirable tolerance traits into domesticated populations that have 12  already been bred for enhanced growth rates (Kirpichnikov et al., 1974; Babouchkine, 1987; Molony, 2001). This practice can similarly be carried out to improve the tolerance of 3n trout by artificially inducing triploidy in 2n stocks that have been bred for enhanced growth and tolerance. Ultimately, having 3n fish with comparable tolerance to their 2n counterparts would enable lake stocking programs to maintain the benefits of triploidy and alleviate the concerns about poorer 3n survival.  1.7  Objectives The widespread use of rainbow trout in aquaculture and lake stocking programs throughout the  world has been a result of its popularity for sport fishing and its utilization as a model organism. One issue that has arisen with lake stocking large numbers of hatchery-reared trout is the potential loss of biodiversity as a result of hybridization between hatchery reared and native populations. To combat this concern and to profit from the potential benefits of triploidy such as better flesh quality and improved growth rates at maturation, hatchery reared trout are made sterile by triploidization. An unfortunate consequence of artificially inducing triploidy in hatchery reared fish is that 3n fish compared to 2n fish have been observed to sometimes show greater mortality in natural environments or under suboptimal rearing conditions. In order to solve this problem an adequate understanding of why 3n fish respond poorer to environmental challenges compared to 2n fish is required. However, the literature available on 3n physiology is very limited and somewhat inconsistent. Therefore, it is necessary to conduct research assessing the performance of specific strains of 2n and 3n trout in response to environmental challenges so that we can better understand what may be contributing to poorer 3n survival in the wild. The objectives of this thesis are to determine if there is an environmental challenge where 3n trout consistently do poorer than their 2n counterparts across strains and years, to determine what strain of trout performs best in response to environmental challenges, and to understand how a 13  laboratory analysis of juvenile captive trout relates to adult trout reared in natural environments. Collectively, these findings should better inform fisheries biologists about environmental challenges that may impact 3n mortality and provide a comprehensive assessment of the responses to environmental challenges of strains that are currently stocked in British Columbia lakes.  14  2  2.1  Performance of wild and domestic strains of diploid and triploid rainbow trout (Oncorhynchus mykiss) in response to environmental challenges Introduction Since the discovery of triplod (3n) trout in nature (Cueller and Uyeno, 1972) and the subsequent  development of technologies to induce triploidy in trout hatcheries, 3n trout have served as a useful tool for aquaculture and fisheries management to keep lakes stocked with hatchery-reared trout and at the same time prevent hybridization with native trout. Sterile 3n trout are also attractive to fisheries biologists because they show enhanced flesh quality and increased growth rates at maturation (FFSBC, 2004). However, some studies have shown greater mortality of 3n fish in natural environments (Cotter et al., 2000; Koenig et al., 2011), in net pens (Withler et al., 1995), and under conditions of suboptimal rearing (Ojolick et al., 1995; Hyndman et al., 2003; Altimiras et al., 2002). To effectively maintain recreational fishing programs in natural lakes with fluctuating environmental conditions, robust 3n fish with good tolerance to changing environmental conditions are needed. Although there is a general consensus that 3n fish show reduced tolerance to environmental challenges (reviewed by Maxime, 2008), there is a great deal of debate in the literature about the effects of triploidy on performance in response to specific environmental challenges and even the effects of triploidy on basic physiology. For example, studies that analyzed growth rates in 2n and 3n salmonids show highly variable results with some studies showing 3n trout grow faster (Galbreath et al., 1994) while other studies show slower growth rates, (Cassani and Caton, 1986; McGeachy et al., 1995; McCarthy et al., 1996) and still other studies show no difference in growth rate between 2n and 3n Atlantic salmon (O’Flynn et al., 1997). Some studies have found 2n and 3n trout to have similar oxygen consumption rates (Aliah et al., 1991; Sezaki et al., 1991; Parsons, 1993) while others have found 3n trout to have lower oxygen consumption rates (Stillwell and Benfey, 1994, 1996a). Differences between 15  species, strains, and rearing conditions used in the above studies may present a confounding issue when trying to compare the effects of triploidy between studies, as strain differences may overwhelm the effects of ploidy (Guo et al., 1990). Numerous studies have shown that different strains and families of fish can respond differently to environmental challenges (Klar et al., 1979; Fangue et al., 2008; Dalziel and Schulte, 2012). For example, Klar et al. (1979) found differences between strains of rainbow trout in the oxygen level at which exercising females of different strains lost equilibrium. One strain (50.4±23.0 torr) was significantly higher than the other two (31.9±5.2 torr and 36.0±6.9 torr). But in addition to these studies, there are many examples where no differences were observed between strains and families of fish exposed to environmental challenges (Lee and Rinne, 1980; Wagner, et al., 2001; Johnson et al., 2004). For instance, Wagner et al. (2001) found no differences in tolerance to hypoxia between four stocks of cutthroat trout. Similarly, two multi-strain studies of rainbow trout found little to no differences in CTMax between strains (Lee and Rinne, 1980; Wagner et al., 2001). Since it is unclear how strain or family may impact assessments of performance in response to environmental challenges, it is necessary to employ a multi-strain approach when investigating differences between 2n and 3n responses to environmental challenges. To better understand why some 3n trout perform less well in natural environments compared to their 2n counterparts we have measured hypoxia tolerance (in three brood years as juveniles and once as adults from lakes), Pcrit, CTMax, Ucrit (in two brood years), routine ṀO2, and three key muscle enzymes of metabolism, across multiple years, in both wild and domestic strains, and in multiple families of 2n and 3n rainbow trout. Revenues from recreational trout fishing in Canada are vital to many regional communities so it is important to find the most efficient way possible to both keep lakes stocked with hatchery reared rainbow trout and to preserve the natural biodiversity in those water bodies. The goals of this study are: 1. to determine if there is an environmental challenge where 3n  16  trout consistently do poorer than their 2n counterparts across strains and years, 2. to determine what strain of trout performs best across assessments of environmental performance, and 3. to understand how our laboratory analysis on juvenile captive trout relates to adult trout reared in natural environments. Collectively, these findings should better inform fisheries biologists about environmental challenges that may impact triploid mortality and provide a comprehensive assessment of the environmental performance of strains that are currently stocked in British Columbia lakes.  2.2 Materials and methods 2.2.1 Experimental animals This study used three wild strains and one domesticated strain of rainbow trout (Oncorhynchus mykiss) that were named after their site of origin. Pennask (PN) and Tzenzaicut (TZ) are wild lakedwelling pelagic strains, Blackwater (BW) is a wild river-dwelling piscivorous strain, and Fraser Valley (FV) is a domesticated strain from the Fraser Valley Trout Hatchery (Abbotsford, British Columbia; FFSBC, 2004). A fourth wild strain, Carp Lake (CL), was added to this study later, but only used for analysis of hypoxia tolerance (see below). Carp Lake trout are mainly lake-dwelling and perform well in competitive and low productivity environments (Clarke, 2012). Breeding pairs of the wild strains were collected from their native environment in the spring, and crosses within strains were performed at the Fraser Valley Trout Hatchery. Crosses within the FV domestic strain were performed each fall from captive broodstock maintained at the Fraser Valley Trout Hatchery. This study was conducted over four years using two separate breeding strategies. In 2008, we bred six families for each of three wild strains (PN, TZ, and BW). Three families were left to develop as 2n and in three families we induced triploidy using hydrostatic pressure shock (methods described below). In 2008, we also bred three families of 2n FV (no 3n families of FV were generated in 2008). In 2009 we used the same breeding strategy as in 2008, generating three families for 2n and 3n strains 17  including three families of 2n and 3n FV. In 2010, we bred three families each of TZ, PN, and FV. We also bred nine families of BW but due to low fish numbers only used three. Unlike the 2008 and 2009 breeding strategies, the progeny of each family in 2010 were divided and half were allowed to develop as 2n and the other half of each family underwent hydrostatic pressure shock to induce triploidy. In 2011, we used the same breeding strategy as in 2010 but only generated three 2n and 3n families for each of TZ and CL. Triploidy was induced in-hatchery using the standard practice of a 30 minute post-fertilization incubation at 8 to 10oC followed by a pressure application of 10,000 psi for 5 minutes using a High Pressure Chamber (Aquatic Eco-Systems, Inc. Florida, USA). From fertilization to hatch, eggs were raised in heath trays for 6-8 weeks. Fry of each family were housed separately and fed to satiation several times a day. Each year, at 6-8 months post-fertilization when the fish weighed approximately 3-4g, roughly half the trout were transported to The University of British Columbia (UBC) and half were transported to research lakes Moss I (+49o 20’ 50.22”, -121o 40’ 52.72”) and Menzies (+50o 45’ 32.28”, 100o 28’ 15.01”) in British Columbia. In 2011, adult (two, three, and four years old from brood years 2008, 2009, and 2010, respectively) 2n and 3n BW, TZ, and PN were collected from these research lakes with trap and fyke nets and transported with oxygen supplied tanker trucks to UBC for analysis. One important difference between the two lakes that these fish were collected from is that they differ considerably in productivity. As a result, the trout captured from the more productive Menzies lake weighed, on average, six times more than the trout captured from the less productive Moss I lake. Holding conditions at UBC were the same for the 2008, 2009, and 2011 juveniles but different for the 2010 brood year. In 2008, 2009, and 2011, each group of fish from a particular family/strain/ploidy were housed separately (at a stocking density of roughly 3.0g L-1) in perforated 16L plastic buckets in a well-aerated recirculation system. In 2010, we used the visible implant elastomer (VIE) tagging system (Northwest Marine Technologies, Washington, USA) to apply identifying marks that  18  distinguished between family, strain, and ploidy. We combined all the families and ploidies from a single strain and held them in 2 replicate 600L tanks within the same recirculating system. The 2011 fish were also VIE tagged but 2n and 3n were held in separate tanks within the same recirculation system. Holding 2n and 3n apart from one another in 2008, 2009, and 2011 allowed us to minimize behavioural interactions between ploidies and housing 2n and 3n together in 2010 enabled minimizing tank effects between ploidies. The adult fish that were retrieved from lakes were all held in a single outdoor 5,750L round aquaria supplied with flow-through dechlorinated city tapwater. Strains and ploidies were differentiated by fin and maxillary clips. Families were not distinguishable. All fish, from all years, were held at 10 to 12oC and maintained on a natural photoperiod, fed once daily, and allowed at least three weeks recovery before undergoing experimental procedures. All holding and experimental procedures were approved by the UBC Animal Care Committee, certificate A09-0611. In 2008, we assessed swimming performance. In 2009, we measured swimming performance a second time, routine metabolic rate and critical oxygen tensions, thermal tolerance, muscle enzyme activities, and hypoxia tolerance. In 2010, we determined hypoxia tolerance a second time. In 2011, we assessed hypoxia tolerance twice more, once with the 2011 fish and a second time with adult lake fish of the previous breeding years.  2.2.2 Swimming performance Critical swimming speeds (Ucrit) were assessed using a step-test protocol similar to that of Farrell et al. (1992). In short, fish were fasted for at least 24 hours before the determination of Ucrit. Four individuals of the same family were transferred to a flow through 92.5L Loligo swim tunnel and given an hour to recover from handling stress at a swimming speed of 0.5BL s-1. Temperature within the swim tunnel was maintained at 10-12oC by controlling the rate of flow of 8-12oC dechlorinated city tap water through the tunnel. The front half of the tunnel was kept covered in black opaque plastic. The 19  tunnel speed was increased by 0.5BLs-1 in 2008 and 1.0 BLs-1 in 2009 every 10 minutes until the fish exhausted and fell to the back of the swim tunnel up against a grid. When an individual failed a first time it was removed from the rear grid by netting and placed back in the middle of the tunnel at a momentarily (~5 seconds) reduced speed. Once an individual failed a second time, swimming speed and time spent at that speed were recorded. Fork length was also measured and the adipose fin was clipped to prevent re-use, and the fish was returned to its holding tank. Virtually no mortalities were noted following the swimming trial (99% of individuals recovered). Critical swimming speed was calculated according to the formula described by Brett (1964) Ucrit = U + (t ti-1x Ui) where U is the speed of the previous step, Ui is the increase in speed of each step, t is the time swam at the step when the fish failed, and ti is the time interval for each step.  2.2.3 Routine metabolic rate and critical oxygen tensions Routine ṀO2 and Pcrit were determined using protocols outlined in Henriksson et al. (2008), with some modifications. Briefly, fish were fasted for 24 hours in their holding tank and then transferred to individual 100ml opaque glass cylindrical respirometers where they were fasted for another 24 hours overnight. Temperature within the respirometers was kept at 12oC with a recirculating water bath. Fish were allowed overnight to habituate to the respirometers under flow through, well-aerated conditions. An hour before trials began a NeoFOX Phase Measurement Systems FOXY oxygen probe (Ocean Optics, Florida, USA) was inserted into a small hole in the top of each respirometer. Probes were calibrated each time they were used to 100% - air and 0% (achieved with nitrogen gas). The trial began when water flow through the cylinder was shut off with the use of stopcocks. Mixing of water within the respirometer was aided by a magnetic stir bar confined to one end. Continuous readings of oxygen tensions within the respirometer were recorded throughout the trial until water oxygen tensions 20  reached ~10% air saturation when the trial was ended. At the end of each trial, each fish was removed from their respiometer, weight and length were recorded, the adipose fin was clipped to prevent reuse, and then the fish was returned to its holding tank. 96% of all individuals recovered. Oxygen consumption traces were generated for each fish using NeoFox Viewer Software and were used to calculate ṀO2 (µmol g-1wet weight h-1) over sequential 1 min intervals using the change in water Po2 and correcting for fish weight and respirometer volume. ṀO2 was calculated by the Fick Principle: Ṁo₂  =  [Δܲo₂ ⋅ αO₂ ⋅ volume] [ϐish	weight ⋅ time]  where ∆PO2 is the difference in PO 2 over the duration of the 1 min interval, αO2 is the solubility coefficient of O2 in water at 12oC (2.2µmol torr-1 L-1; Boutiliier et al., 1984), volume is the volume of water in the respirometer (L), fish weight is the mass of the individual in the respirometer (kg), and time is the duration of the interval (1 min). Routine ṀO2 was taken as the mean ṀO2 at a water PO 2 ranging from 40-140torr based on the lowest flat region of the plot. The water PO 2 that the ṀO2 was taken over varied between individuals based on the amount and intensity of activity of the fish within the respirometer, but most were taken within 60-120torr. Pcrit was determined using Regress software based on the BASIC program designed by Yeager and Ultsch (1989). Pcrit was determined as the intersection between slopes when ṀO2 remains constant with a declining Po2 and when ṀO2 begins to decline with a declining Po2.  2.2.4 Thermal tolerance Critical thermal maxima (CTMax) were determined based on a revised protocol described by Fangue et al. (2006). All fish were fasted for 24 hours before the determination of CTMax. For each trial, 21  a total of eight fish were placed into individual 1.5L containers that were held within a 12oC water bath. Fish were transferred an hour before trials began to allow fish to recover from any handling stress. The 1.5L containers were aerated with individual air stones and each were topped with a transparent plastic lid to keep the fish from escaping. Temperature in the bath was then increased at a constant rate of 0.3oC min-1 using two Lauda RM6 benchtop temperature controllers. Water throughout the bath was kept thoroughly mixed by placing two Repti Flo 200 (Art.# PT-2090, EXO TERRA, Montreal, CA) circulation pumps within the water bath. Temperature was increased until all the fish in the trial displayed a LOE for more than 2 seconds. Loss of equilibrium is the point at which a fish can no longer maintain dorso-ventral orientation. When an individual fish displayed LOE, temperature of the 1.5L container was recorded and fish were removed by hand from the experimental tank and placed into a well-aerated temporary holding tank for recovery. Fish were weighed, VIE elastomer tagged to prevent re-use then returned to holding tanks. 98% of all individuals recovered.  2.2.5 Hypoxia tolerance The time to LOE was determined using the protocols outlined in Mandic et al. (in press). Briefly, trout were fasted in their holding tanks for 24 hours before being transferred to the experimental tank where they were fasted for another 24 hours before determination of time to LOE began. The experimental apparatus for determining time to LOE for the juvenile fish consisted of a 60L glass aquarium covered in black plastic with the exception of one side which was used to view the group of fish. The experimental apparatus for determining time to LOE for the adult fish was an opaque 1,550L round insulated fiberglass aquaria which enabled viewing of the fish only from above. In 2009 and 2010, the experimental apparatus was half filled with water from the recirculating holding system and water temperature was maintained at 12oC by submersing the aquarium in recirculating water from the same holding system. This was similar in 2011 except temperature was 22  controlled by carrying out assessments in a temperature controlled environmental chamber. Temperature was not controlled during the analysis of time to LOE in adult fish, but temperature was monitored continuously throughout the trial and changed by less than 1oC over the duration of each hypoxia exposure trial. All fish were transferred from their holding tank to the experimental tank and allowed to recover from handling stress overnight under well-aerated conditions. For the analysis of fish from 2009, we subdivided the experimental apparatus in to four 2.3L mesh chambers and placed 6-8 individuals from a single family/strain/ploidy into each chamber. Three separate determinations of time to LOE were done for each family/strain/ploidy. For trout bred in 2010, there was no need to separate families/strains/ploidies because all fish were marked with an elastomer tag. Four 2n and four 3n fish from each of three families of one strain (24 fish) were assessed per trial. The trials were replicated once for a total sample size of 8 individuals per family/strain/ploidy. Time to LOE in one strain (comparing 2n and 3n trout) was concluded before moving on to the next. The order was BW, TZ, PN, and FV. For 2011, three individuals from each of the three TZ families both as 2n and 3n were assessed per trial. The trials were each replicated once for a total sample size of 6 per family/ploidy. For the adult fish, 1-3 individuals from each of the three strains from each of 2n and 3n from each of the two research lakes (24 fish) were assessed per trial. The trials were each replicated four times for a total sample size of ~8 individuals per strain/ploidy/research lake. To determine time to LOE, oxygen tensions were lowered over one hour from 100% air saturation to ~10% air saturation (16 torr, 2.1kPa) by bubbling nitrogen into the water. To better maintain constant oxygen levels clear packing bubble wrap was draped over the surface of the water which greatly reduced O2 ingress and prevented fish from performing aquatic surface respiration. Oxygen tension was monitored with 2-4 oxygen probes (same as above). Water oxygen levels were kept uniform throughout the apparatus by submerging 2 (in diagonal corners) Repti Flo 200 circulation pumps (as above).  23  Time to LOE was recorded as the time at which a fish was no longer able to maintain dorsoventral orientation for more than 5 seconds. Following LOE, fish were netted from the experimental tank and placed into a well-aerated temporary holding tank for recovery. Fish were weighed, adipose fin clipped to prevent re-use, and returned to holding tanks. 94-98% of all individuals recovered.  2.2.6 Muscle enzyme activities Fish were fasted for 48 hours before sampling. To sample tissues, fish were quickly netted from their holding tank and euthanized in an overdose of benzocaine (250mg l-1, Sigma-Aldrich, USA). Once a fish was euthanized, weight was recorded and muscle samples were collected from both sides of the fish between the base of the caudal fin and the base of the anal fin. Muscle samples were immediately frozen in liquid nitrogen and stored at -80oC until analysis. Muscle samples were prepared and enzyme analyses conducted using protocols similar to those described in Dalziel et al. (2011). Muscle samples were ground into a fine powder in a liquid nitrogen cooled mortar and pestle and ~20mg of frozen muscle was immediately sonicated in buffer (20mM Hepes, 5mM EDTA, and 0.1% Triton X-100, pH 7.0), using a Kontes micro-ultrasonic cell disruptor. Enzyme activities were assayed using a Molecular Devices SpectraMAX Plus spectrophotometer at 25oC. Activities were measured for citrate synthase (CS; EC 2.3.3.1), Pyruvate kinase (PK; EC 2.7.1.40), and lactate dehydrogenase (LDH; EC 1.1.1.27). Final well concentrations for each assay were as follows: CS (0.15mM DTNB, 0.30mM Acetyl-CoA, and 0.5mM Oxaloacetate in 50mM Tris, pH8.0), PK (0.5mM NADH, 5mM ADP, 100mM KCl, 10mM MgCl2, 10μM fructose 1,6-bisphosphate, 50U ml-1 LDH, and 5mM phosphoenol pyruvate in 50mM Tris, pH 7.4), and LDH (0.8mM NADH and 22.5 mM pyruvate in 50mM Tris, pH 7.4).  24  2.2.7 Ploidy confirmation The induction of triploidy for each family was confirmed in-hatchery for each brood year and we reconfirmed ploidy in our experimental fish from 2009, 2010, and 2011 using a subset from each family once all assessments had been concluded for that year. Triploid status was also re-confirmed with the adult fish. To determine ploidy, blood samples were collected by caudal severance from fish euthanized with an overdose of benzocaine (same as above). Our protocol was adapted from Arumuganathan and Earle (1991). Briefly, 10µl of blood was diluted into 500µl PI solution (0.0125ml citrate acid dextrose solution, 0.00425mg ribonuclease A, 0.005ml IGEPAL CA-630, and 0.025mg propidium iodide). After 30 minutes, 100µl of 4% paraformaldehye was added and the samples were stored at 4oC overnight. Ploidy was confirmed by measuring whole blood nuclear DNA content via flow cytometry on a FACSCalibur bench top flow cytometer (BD Biosciences, California, USA). The percent induction of triploidy that we measured was 93% in 2009, 91% for both the 2010 and lake reared trout, and 94% in 2011.  2.2.8 Statistical analysis Two different statistical analyses were used to test the influence of strain and ploidy on responses of 2n and 3n trout to environmental challenges. The 2008 (Ucrit), 2009 (Ucrit, ṀO2, Pcrit, CTMax, enzyme activities, and time to LOE), and lake reared trout data sets were analyzed with twoway analyses of variance (ANOVA) and the 2010 and 2011 time to LOE data sets were analyzed with a mixed-effects model nested one-way ANOVA. The need for two different statistical analyses arises from the two different breeding strategies used in this study. The two-way ANOVA was used on data sets where 2n and 3n trout were assumed to be unrelated and the mixed effects model nested one-way ANOVA was used on the two data sets where 2n and 3n trout were siblings. For the mixed-effects model, the nlme package in R (Pinheiro et al., 2009) was used with individual nested within family 25  (random effects) and family nested within strain and ploidy (fixed effect). Assumptions of homogeneity of variance and normality were met for both data sets. To detect pairwise differences Tukey’s honestly significant difference post hoc tests were done using the multcomp package in R (Hothorn et al., 2008). Since weight is an important covariate with physiological assessments, where there were significant differences in weight between strains we tested for a correlation between weight and the performance variable. For Ucrit (2008), LDH enzyme activity, and time to LOE (2010) there was a significant correlation with length or weight so the p-value and correlation coefficient are given for each family/strain/ploidy. It was not possible to conduct an analysis of covariance in these three instances since not all families showed a significant correlation. However, for all of our assessments the 2n and 3n counterparts of a given strain did not significantly differ from one another in weight or length. All results are presented as mean ± SEM.  2.3 Results 2.3.1 Swimming performance In the 2008 trout, two-way ANOVA revealed a significant effect of strain (P<0.0001) but no significant effect of ploidy (P=0.3950) on Ucrit (figure 2.1a). For strain, 2n FV trout (only 2n were available for comparison in 2008) and 2n and 3n PN trout had significantly higher Ucrit than 2n and 3n BW and TZ. There were significant differences in length between strains and when we considered body length of all individuals tested, there was a significant correlation between length and Ucrit (P<0.0001, r=-0.887; table 2.1). When each strain and ploidy is considered independently only 3n BW, TZ, PN, and 2n PN show significant correlations between length and Ucrit (table 2.1). The mean length of all trout from 2008 used for determining Ucrit was 12.6±0.2cm, n=188. At the level of family, there was large variation in Ucrit, with a coefficient of variation of 12.3%, which is slightly larger than the coefficient of  26  variation seen among strains and ploidies which was 10.4% (figure 2.1a). For the 2008 groups, strains and ploidies differed based on the mean lengths of the groups (e.g. BW and TZ were larger than and differed in Ucrit from the smaller PN and FV). In the 2009 trout, two-way ANOVA showed a significant effect of strain (P<0.0001) but no significant effect of ploidy (P=0.0959) on Ucrit (figure 2.1b). For strain, 2n and 3n FV domestic trout had a significantly higher Ucrit than all other strains. 2n and 3n BW and TZ Ucrits were significantly lower than 2n PN. In the 2009 trout, the mean length of all fish used for the determination of Ucrit was 6.43±0.01cm (n=192) and there was almost no significant variation in length between strains and ploidies except BW 3n (6.49±0.03cm) were significantly longer than 3n PN (6.34±0.03cm). There was no correlation between length and Ucrit (P=0.134). At the level of family, there was variation in the 2009 Ucrit that was comparable to the variation in the 2008 Ucrit, with a coefficient of variation of 12.8%, which is only slightly larger than the coefficient of variation seen among strains and ploidies which was 11.9% (figure 2.1b).  2.3.2 Routine metabolic rate Two-way ANOVA of the unpaired 2n and 3n trout from 2009 revealed a significant effect of strain (P<0.0001) but no significant effect of ploidy (P=0.3233) on ṀO2 (figure 2.2). For strain, 2n and 3n FV domestics had a higher routine metabolic rate than 2n and 3n BW, 3n TZ, and 2n PN. There were no significant differences in size between any strains or ploidies and the mean weight of all trout used for the determination of ṀO2 was 3.34±02g, n=192. At the level of family, there was moderate variation in ṀO2, with a coefficient of variation of 9.8%, which is larger than the coefficient of variation seen among strains and ploidies which was 6.7% (figure 2.2).  27  2.3.3 Critical oxygen tensions We found a significant effect of strain (P<0.0001) but no significant effect of ploidy (P=0.1210) on Pcrit. (Two-way ANOVA; figure 2.3). For strain, 2n and 3n BW and 2n FV had higher Pcrits than 3n TZ and 2n PN and the remaining strains and ploidies had intermediate Pcrit values. There were no significant differences in size between any strains or ploidies and the mean weight of all trout used for the determination of Pcrit was 3.34±02g, n=192. At the level of family, there was variation in Pcrit similar to that of ṀO2, with a coefficient of variation of 9.9%, which is larger than the coefficient of variation seen among strains and ploidies which was 7.4% (figure 2.3).  2.3.4 Thermal tolerance Two-way ANOVA revealed no significant differences between strains (P=0.76) but a significant difference in CTMax between ploidies (P=0.004; table 2.2), but the significant effect of ploidy was not consistent across strains. For ploidy, 2n TZ and PN trout had significantly higher CTMax than their respective 3n counterparts. The opposite was observed for BW, a post-hoc analysis showed that 2n BW CTMax was significantly lower than 3n BW. FV were not available for assessments of CTMax. There were no significant differences in size between any strains or ploidies and the mean weight of all trout used for the determination of CTMax was 2.4±05g, n=141. At the level of family, there was very little variation in CTMax, with a coefficient of variation of 1.0%, which is similar to the coefficient of variation seen among strains and ploidies which was 0.8% (table 2.2).  2.3.5 Hypoxia tolerance With the juvenile 2009 trout, a two-way ANOVA revealed there was both a significant effect of strain (P<0.0001) and ploidy (P<0.0001) on time to LOE at ~10% air saturation, (16 torr; figure 2.4a). For  28  strain, 2n BW had the longest time to LOE while 3n TZ and FV had the shortest time to LOE. Within any strain, 3n trout generally showed shorter times to LOE compared with 2n trout, but this effect was only significant with post-hoc analysis in TZ and FV. There was no significant variation in size between strains with the exception of 2n and 3n FV which weighed more than 2n and 3n BW, TZ, and PN (table 2.1). However there was no significant correlation between weight and time to LOE (P=0.382) between any strains or ploidies and the mean weight of all juvenile trout from 2009 used for the determination of time to LOE was 3.32±0.03g, n=620. At the level of family, there was substantial variation in time to LOE, with a coefficient of variation of 51.6%, which is larger than the coefficient of variation seen among strains and ploidies which was 46.2% (figure 2.4a). Analysis of time to LOE in the 2010 juvenile trout revealed similar results to those obtained from the 2009 juvenile trout. Mixed-effects model 1-way ANOVA, with family as a nested variable, revealed a significant effect of strain (P<0.0001) and ploidy (P<0.0001) on time to LOE (figure 2.4b). For strain, 2n and 3n FV domestics had the longest time to LOE while 2n and 3n PN were shorter and 2n and 3n BW and TZ had the shortest time to LOE. Within any strain, 3n fish generally showed shorter time to LOE compared with 2n fish, but a post hoc analysis did not reveal any significant differences. There was greater variation in size between the 2010 strains compared to the 2009 strains. The 2n and 3n FV weighed the most, the 2n and 3n BW and TZ weighed the least, and the 2n and 3n PN were intermediate (table 2.1). There was a significant correlation between weight and time to LOE (P<0.0001, r=0.532) overall, but there were no significant correlations between weight and time to LOE within any family/strain/ploidy and the mean weight of all trout used for the determination of time to LOE for the 2010 trout was 9.9±0.2g, n=288. At the level of family, there was substantial variation in time to LOE, more than in any of our other assessments of environmental performance (figure 2.4b). The coefficient of variation at the family level was 92.5%, which is much larger than the coefficient of variation seen among strains and ploidies which was 75.3%.  29  With the 2011 2n and 3n TZ and CL trout a mixed-effects model 1-way ANOVA, with family as a nested variable, revealed a significant effect of strain (P<0.0001) and ploidy (P=0.0144) on time to LOE (figure 2.5). For strain, 2n CL had the longest time to LOE whereas 3n PN had the shortest time to LOE. Within any strain, 3n fish generally showed shorter time to LOE compared with 2n fish, but this effect was only significant with post hoc analysis in CL. There were no significant differences in size between any strains or ploidies and the mean weight of all fish used for the determination of time to LOE on the 2011 trout was 4.5±0.2g, n=76. At the level of family, there was substantial variation in time to LOE, with a coefficient of variation of 41.6%, which is not much different than the coefficient of variation seen among strains and ploidies which was 41.3% (figure 2.5). Unlike the assessments of hypoxia tolerance for the 2009, 2010, and 2011 juvenile rainbow trout, no significant effect of strain (P=0.9046) nor ploidy (P=0.5584) on time to LOE (table 2.3) was seen with adult trout that lived in natural lakes for one to three years. There were no significant differences in LOE when individuals were divided and analyzed separately according to lake or combined and analyzed together so the data is presented with lakes combined (table 2.3). There were significant differences in weight between strains and ploidies and when we considered the weights of all individuals tested, there was a significant correlation between weight and time to LOE (P<0.0001, r=0.483; Table 2.1). When each strain and ploidy is considered independently, only 2n and 3n BW and 2n TZ show significant correlations between weight and time to LOE (table 2.3). The mean weight of all adult lake trout used for the determination of time to LOE was 468±47g, n=86. Families were not identified for assessments of hypoxia tolerance on adult lake fish.  2.3.6 Muscle enzyme activities A two-way ANOVA of PK, CS, and LDH muscle enzyme activities revealed no significant effects of ploidy (P=0.8202, 0.9326, 0.3233, respectively) but with all three enzymes we did see significant effects 30  of strain (P=0.0007, P<0.0001 and P<0.0001, respectively; table 2.4). For PK, 2n FV domestics had significantly higher enzyme activity than 2n TZ and PN and 3n BW. For CS, 2n and 3n FV domestics had higher enzyme activity than 3n TZ and 2n and 3n PN. For LDH, 2n and 3n FV domestics as well as 3n PN were not significantly different from each other or from any other strain and ploidy. 2n and 3n TZ however, had significantly higher LDH activity compared with 2n and 3n BW and 2n PN. There were significant differences in weight between strains and when we considered the weights of all individuals tested, there was a significant correlation between weight and LDH enzyme activity (P=0.0001, r=0.249) but not weight and PK or CS enzyme activity (P=0.116 and P=0.515, respectively; Table 2.1). When each strain and ploidy is considered independently only 2n PN and 3n BW and TZ show significant correlations between weight and time to LOE (table 2.1). The mean weight of all trout used for the determination of muscle enzyme activities was 1.69±0.03g, n=300. At the level of family, there was large variation in muscle enzyme activities, with a coefficient of variation for LDH, CS, and PK of 17.6%, 17.2%, and 18.3%, respectively; which is nearly twice as much as the coefficient of variation seen among strains and ploidies which was 9.8%, 12.4%, and 10.3%, respectively (table 2.4).  2.3.7 Specific growth rates Two-way ANOVA of 2n and 3n trout from the 2009 brood year revealed a significant difference between strains (P<0.0001, P<0.0001) but no significant difference in specific growth rates between ploidies for both length (P=0.629) and weight (P=0.598), respectively (table 2.5). Post-hoc analysis showed that FV domestics have higher specific growth rates for weight and length over 28 days compared to their wild conspecifics.  31  2.4 Discussion Many species of fish exhibit greater mortality as triploids in natural environments (Cotter et al., 2000; Koenig et al., 2011) and in net pens (Withler et al., 1995) compared to their 2n counterparts; however, there is still considerable variability between studies in the effects of triploidy on performance in the face of environmental challenges (reviewed by Benfey, 1999 and Maxime, 2008). The goal of the present study was to develop a better understanding of why 3n fish do not perform as well as their 2n counterparts in natural environments. We accomplished this goal by comparing hypoxia tolerance, Pcrit, ṀO2, thermal tolerance, muscle enzyme activities, and swimming performance between 2n and 3n rainbow trout from multiple families of multiple strains and, in hypoxia tolerance and swimming performance, over multiple years. Over three years of analysis, we also stocked two natural lakes with subsamples of the juvenile trout used for our laboratory analysis and recaptured them as adults for analysis of hypoxia tolerance. Together, these comparisons were used to address three specific objectives: 1. to determine if there is an environmental challenge where 3n trout consistently do poorer than their 2n counterparts across families, strains and years, 2. to determine what strain of trout performs best across assessments of environmental performance, and 3. to understand how our laboratory analysis on juvenile captive trout relates to adult trout reared in natural environments.  2.4.1 Effects of ploidy on performance in response to environmental challenges Of all the analyses performed, the only consistent effect of triploidy we observed across all five strains was that 3n trout are less tolerant of hypoxia than their 2n counterparts (figures 2.4a, 2.4b, 2.5). This effect was seen when we did (2010, 2011) and did not (2009) control for parentage and was also consistent across wild and domestic strains. Our comparisons of upper thermal tolerance for the 2009 brood year revealed a small but significant difference between 2n and 3n trout (table 2.2), but the effects of ploidy on CTMax did not have a consistent direction across strains. We also observed no 32  consistent effect of ploidy on Ucrit both with 2008 (figure 2.1a) and 2009 (figure 2.1b) brood stocks. Furthermore, our assessments of time to LOE at 16 torr oxygen revealed a significant effect of ploidy over three consecutive years (2009, figure 2.4a; 2010, figure 2.4b; 2011, figure 2.5). Our results of reduced 3n relative to 2n hypoxia tolerance are consistent with Lilyestrom et al. (1999) who found a significant effect of ploidy on the time to LOE of catfish hybrids (Ictalurus punctatus x Ictalurus furcatus) plunged into water at oxygen levels maintained at ~9 torr. They measured a time to LOE (min±SD) of 9.8±7.2 for 2n and 4.7±5.0 for 3n. Compared to our study, the shorter time to LOE is likely a result of using different species but it could also be due to differences in methodologies between our two studies. In our study, we used a less severe hypoxia exposure to allow for greater resolution for detecting differences between strains and ploidies. Our results are also consistent with Yamamoto and Iida (1994) who found only minor, yet significant, differences in the partial pressure of oxygen when 2n and 3n rainbow trout first showed LOE (2n, 23.2±2.7 torr and 3n, 26.1±0.8 torr), no significant difference when half showed LOE (2n, 23.2±0.4 torr and 3n, 23.4±0.4 torr), and a significant difference in the oxygen tensions at which all trout had LOE (2n, 17.5±0.2 torr and 3n, 22.2±0.0 torr). The increased sensitivity to hypoxia seen in the 3n trout was attributed to impaired oxygen uptake and blood oxygen carrying capacity as a result of 3n fish having larger red blood cells from the extra chromosome copy (Yamamoto and Iida 1994). Larger red blood cells in 3n fish results in a decreased surface area to volume ratio, which could limit oxygen binding to haemoglobin during transit through the gills. Triploid Atlantic salmon have been reported to have smaller gills than 2n Atlantic salmon (Sadler et al., 2001), which would further limit oxygen uptake from the environment. Oxygen uptake from the environment and delivery to tissues could be further impaired in 3n compared to 2n fish because of lower overall hemoglobin concentrations in the blood of 3n compared with 2n fish. Increases in red blood cell size are compensated for by having fewer cells in circulation  33  (Reviewed by Benfey, 1999). Having fewer cells in circulation may explain why 3n fish have been shown to sometimes have lower overall hemoglobin concentrations in the blood than 2n fish (Parsons, 1993; Sadler et al., 2000). With less hemoglobin to hold oxygen and a reduction in surface area to volume ratio in the red blood cells of 3n compared to 2n fish, oxygen delivery to tissues may also be impaired (Maxime, 2008). Clearly, there is evidence to suggest that 3n fish have compromised oxygen uptake and/or delivery to tissues and therefore we would predict that 3n trout would consistently show higher Pcrit values compared to 2n trout. Pcrit is the point at which fish transition from an oxygen regulating strategy to an oxyconforming strategy and as a result Pcrit has been proposed as a whole-animal measure of the ability to acquire oxygen from the environment (Gannon et al., 1999). Animals with lower Pcrit values are able to maintain oxygen uptake at a lower environmental oxygen tension than animals with higher Pcrit values and therefore, Pcrit is thought to be related to hypoxia tolerance (Chapman and McKenzie, 2009). Contrary to our prediction that 3n trout would show higher Pcrit values than their 2n counterparts, we found little variation in Pcrit between 2n and 3n trout (figure 2.3). This was surprising since 3n fish have been found to be more sensitive than their 2n counterparts to conditions of high oxygen demand and/or low oxygen availability (Blanc et al., 1992; Simon et al., 1993; and Johnstone, 1996). Additionally, a clear relationship has been demonstrated between Pcrit and hypoxia tolerance through studies comparing Pcrit and time to LOE in sculpins (Mandic et al., in press) and Pcrit and blood O2 transport in sharks (Speers-Roesch et al, 2012). However, not all studies show a relationship between Pcrit and hypoxia tolerance. Yao and Richards (in prep) found significant differences in time to LOE at 12 torr between species of Danio and Davario but with no variation in Pcrit between species. To our knowledge, this is the first time that Pcrit has been assessed in 3n fish and in light of Yao and  34  Richards (in prep) and the present study, researchers assessing hypoxia tolerance in 3n fish should be cautioned against assuming Pcrit is an accurate measure of hypoxia tolerance. Consistent with the lack of an effect of triploidy on Pcrit, we also found that there were no consistent differences in routine ṀO2 between 2n and 3n trout (figure 2.2), which is consistent with the majority of studies that have examined ṀO2 in 2n and 3n fish (Parsons, 1993; Yamamoto and Iida, 1994, Altamiras et al., 2002). These studies, along with our own data (figure 2.2) support the idea that triploidy does not hinder aerobic metabolism. In further support of this idea we observed no consistent effect of ploidy on Ucrit both with the 2008 (figure 2.1a) and 2009 (figure 2.1b) trout. Ucrit is commonly used to compare aerobic capacity between 2n and 3n fish (as reviewed by Benfey, 1999 and Maxime, 2008). Our findings of no effect of ploidy on Ucrit are consistent with the majority of other studies assessing critical swimming speeds in 3n fish (Parsons, 1993; Stillwell and Benfey, 1996a, 1997; Altamiras et al., 2002). The absence of a difference in muscle enzyme activities that we measured also supports the notion that 2n and 3n fish do not differ in aerobic capacity as CS activity can be used as a proxy for aerobic capacity (Dalziel et al. 2012; table 2.4). Our comparisons of upper thermal tolerance for the 2009 brood year revealed a significant difference between 2n and 3n trout (table 2.2); however, the effects of ploidy on CTMax were not consistent across strains. We found 2n TZ and PN had significantly higher CTMax than their respective 3n counterparts, but for BW, 3n had higher CTMax than 2n BW. Our findings differ from those of Benfey et al. (1997) and Galbreath et al. (2006) who found no differences in thermal tolerance between 2n and 3n trout. However, Hyndman et al. (2003) showed that there were differences in the ability of 2n and 3n trout to recover in warm water following exhaustive exercise with 3n trout showing a significantly higher rate of mortality than their 2n trout. These data suggest that differences in thermal tolerance between 2n and 3n trout may only emerge in response to interactive effects with other environmental factors and might not necessarily reflect differences in thermal tolerance per se. Fundamental 35  differences in 2n and 3n physiology such as increased diffusion distances in larger 3n cells (Maxime, 2008) may result in differences in coping strategies between 2n and 3n trout when exposed to an environmental stress or in recovery strategies after being subjected to an environmental challenge. This notion is supported by the idea that 3n fish may have reduced energy stores and/or increased rates of depletion when exposed to environmental challenges (Benfey, 1999). From their measurements of key muscle metabolites phosphocreatine and glycogen (among others), Hyndman et al. (2003) attributed poorer 3n tolerance to a reduced capacity to recruit anaerobic pathways at warmer temperatures. At least in muscle, our analyses of enzyme activities do not support this contention. If 3n tolerance was impaired as a result of reduced capacity to recruit anaerobic pathways than we may have expected to see lower PK and LDH enzyme activities in 3n than 2n trout, but we observed no differences in PK, CS, and LDH muscle enzyme activities between 2n and 3n from the 2009 brood year (table 2.4). To our knowledge this is the first time muscle enzymes have been measured in 3n trout. However, enzymes such as alkaline phosphatase, aldolase, and lactate dehydrogenase have been measured before in erythrocytes and were found to be elevated in 3n compared to 2n fish but the observed differences did not persist when the fact that 3n have decreased cell numbers was taken into account (Barker et al., 1983; Sezaki et al., 1983, 1988). To test whether our laboratory analysis performed on lab-reared juvenile trout translates to lake-reared trout we measured time to LOE in 2n and 3n from the same brood years as above (2008, 2009, and 2010) to see if lake-reared trout would show the same 2n/3n differences in hypoxia tolerance as trout analyzed in the laboratory (figure 2.4a, 2.4b, and 2.5). Interestingly, we did not observe any effect of ploidy on adult time to LOE at 16 torr (table 2.3). There are three possible explanations for the lack of consistency between our analysis on juvenile trout and the results of the lake-reared trout. Firstly, selective forces present in natural systems have already acted on these fish and the individuals we captured for our assessments of time to LOE represent the better performing 2n and 3n individuals  36  from each strain. Secondly, non-random capturing of adult trout as a result of trap net bias could have caused us to capture trout based on some parameter which may influence time to LOE. We captured fewer 3n BW, TZ, and PN compared to 2n (table 2.3) which may be as a result of differences in time spent scavenging or positioning in the water column and not necessarily differential 2n/3n mortality. Thirdly, there is much larger variation in confounding variables when doing assessments of performance in response to environmental challenges with adult lake trout compared to with juvenile captive trout. For example, adult trout can be more variable in size and reproductive status and lake trout can be more prone to conditions of deteriorating health. However, the likely largest contributor to the differences between assessments of performance in response to environmental challenges with captive juveniles and adult lake trout is the potential for the lake trout to have previously been exposed to environmental stressors. It is well documented that previous exposures, or training, can improve performance in response to environmental challenges (Woodward and Smith, 1985; Young and Chech, 1993a, 1993b; Schaefer and Ryan, 2006) and when using lake fish for assessments of performance in response to environmental challenges it is difficult to know what stressors different individuals have experienced. Rather than attributing differential 2n/3n mortality in natural systems to any one component of environmental performance or tolerance it is more likely that complex natural settings varying in a wide range of biotic and abiotic parameters that greater 3n mortality arises from a number of compounding factors. For example, low oxygen levels render triploids less able to compete for resources or avoid predators as it has been found that swimming performance in trout is reduced under hypoxic conditions (Jones, 1971). Additionally, low environmental oxygen levels may exacerbate the impacts of increased water temperature. It is generally accepted that when trout are exposed to elevated temperatures oxygen demand increases (reviewed by Portner, 2001), which at high enough temperatures can lead to temperature induced hypoxemia (Portner et al., 2004). Since (as mentioned  37  above) 3n fish have been found to be more impaired by conditions of high oxygen demand and/or low oxygen availability than their 2n counterparts it is likely that 3n sensitivity to hypoxia under high temperature conditions would be more pronounced than their 2n counterparts. A study comparing the interactive effects of hypoxia with a common secondary environmental challenge may help to elucidate lake characteristics that are more likely to cause greater 3n mortalities compared to their 2n counterparts. The main contribution of this study to the body of literature on 3n performance and tolerance is our consistent findings that 3n juvenile trout are less tolerant of hypoxia than 2n trout, which we have demonstrated across multiple strains and multiple years. These findings suggest that environmental oxygen challenges may contribute to poorer 3n performance in natural settings.  2.4.2 Effects of strain on performance in response to environmental challenges We found a significant effect of strain on all of our assessments of environmental performance with the exception of thermal tolerance (table 2.2). Three previous multi-strain studies using 2n trout also yielded little to no differences in CTMax between strains (Lee and Rinne, 1980; Wagner et al., 2001; Rodnick, et al., 2004), suggesting that a thermal tolerance study of less intensity for longer duration may better elucidate differences in thermal tolerance (Wagner et al., 2001). However, studies assessing intraspecific upper thermal tolerance in other species such as the common killifish (Fundulus heteroclites) and orangethroat darter (Etheostoma spectabile) have found significant differences between populations in CTMax (Strange et al., 2002; Fangue et al., 2006). It was also surprising that we did not see an effect of strain on thermal tolerance considering the varied ancestral life histories of our trout and the different selective forces on thermal tolerance that they would have experienced (FFSBC, 2004). With contrasting findings in the literature on intraspecific differences in CTMax it might be worthwhile to perform additional assessments of upper thermal tolerance on hatchery-reared rainbow 38  trout strains that are stocked into B.C. lakes, especially if water temperatures are expected to rise over the coming decades. We observed an effect of strain on hypoxia tolerance with juveniles (figures 2.4a, 2.4b, 2.5) but not with lake-reared adults (table 2.3). For the 2009 juvenile trout BW and PN had significantly higher time to LOE than TZ and FV whereas for the 2010 juveniles FV domestics had higher time to LOE than all other strains. For the 2011 juvenile trout TZ was the lowest again. Our finding of differences in hypoxia tolerance between strains is in contrast to the findings of Wagner et al. (2001) who found no differences in tolerance to hypoxia between four stocks of cutthroat trout and these authors attributed the lack of difference between stocks to a lack of natural selection for that trait. However, our findings are in agreement with Klar et al. (1979) who reported interstrain differences in dissolved oxygen tolerance in rainbow trout. The authors measured the concentration of dissolved oxygen (converted to partial pressure) at which exercising trout lost equilibrium and found one strain lost equilibrium at a significantly higher oxygen tension (50.4±23.0 torr) than the other two (31.9±5.2 torr and 36.0±6.9torr), suggesting that there is large variation in the oxygen partial pressure at which different strains of rainbow trout lose equilibrium. These authors attributed the intraspecific differences in hypoxia tolerance in rainbow trout to differences in LDH isozymes, which, in the future, should be examined in our trout. Assessments of Pcrit revealed a significant effect of strain (figure 2.3), with BW and FV having higher Pcrits than TZ and PN, but our Pcrit results do not explain the variation in time to LOE that we observed between strains. A possible explanation for TZ and PN having lower Pcrits than BW and FV may arise from TZ and PN having similar life history characteristics. Tzanzaicut and PN are both from lake environments which are more prone to hypoxia than rivers due, in part, to less mechanical mixing of oxygen between the air and the water (Giller and Malmqvist 1989). Although Pcrit does not always show a relationship with hypoxia tolerance (as described above) perhaps there have been greater selective 39  forces for oxygen extraction among TZ and PN compared to BW as a result of being historically from environments of potentially different selective pressures. Differences in Ucrit were observed between strains (figures 2.1a, 2.1b). BW and TZ Trout from 2008 and 2009 showed significantly lower Ucrits than did FV domestics with PN being intermediate. The marked difference between the FV strain and the other three may relate to FV being a domestic strain whereas the other three are wild. Perhaps fish with higher growth rates (as FV have been selected for over several generations; table 2.5) have higher swimming performance as was found in studies involving trained fish which grew faster than controls (Houlihan and Laurent, 1987; Farrell et al., 1990; Young and Cech, 1993a). However, the opposite has been found in juvenile rainbow trout (Gregory and Wood, 1998) which has been used to suggest that there is a trade-off between swimming performance and specific growth rate (Kolok and Oris, 1995). Our data does not support this contention which could possibly reflect superior swimming performance of the FV strain masks the trade-offs of higher growth rates from domestication. We observed differences in ṀO2 between strains (figure 2.2), similar to the differences we saw between strains in Ucrit. The FV domestics had higher routine metabolic rates than BW, TZ, and PN trout. The higher metabolic rate of the FV domestics is consistent with a previous study showing that conspecifics with higher growth rates have higher metabolic rates, (Arnott et al., 2006). Since it has been reported that rainbow trout with higher growth rates also have higher critical swimming speeds and aerobic scopes than those trout with lower growth rates (McKenzie et al., 2006), future studies with these trout should include measurements of maximum metabolic rate as a higher aerobic scope in the FV domestics would be a likely explanation for the observed higher FV critical swimming speeds. Muscle enzymes of glycolysis and the Krebs cycle were measured in an attempt to explain the variation that we observed between strains in swimming performance. The FV domestics’ enzyme activities were higher than their conspecifics for PK and CS but not LDH (table 2.4) which is consistent 40  with our ṀO2 data that FV have higher aerobic metabolism. However, the observed intermediate LDH levels are inconsistent with our Ucrit data where we may have expected FV to have higher LDH activities since higher LDH activities have been associated with higher anaerobic exercise performance (Sullivan and Somero, 1980). This suggests that anaerobic exercise performance contributed relatively little to our assessments of Ucrit because if anaerobic exercise were a large contributor to our Ucrit results we would have observed higher TZ Ucrits on account of their higher LDH activity. Our observed differences between the FV domestics and the wild strains in muscle PK and CS activity but not LDH activity suggests that FV may have an advantage over its wild conspecifics in aerobic but not anaerobic exercise performance. It has been found in Atlantic cod (Gadus morhua) that higher growth rates correlate with higher enzyme activities for glycolytic enzymes (PK and LDH) but not mitochondrial enzymes (CS) (Pelletier et al., 1994). Our findings for PK are consistent with their study but our findings for LDH and CS are inconsistent with their study which could be accounted for by differences in the relationship between enzyme activities and growth rates between rainbow trout and Atlantic cod. The differences we observed between strains in baseline levels of enzyme activities of aerobic and anaerobic energy production suggests that there may be differences in the capacities of these strains to metabolize energy stores and generate metabolic waste products. Different biochemical capacities to respond to environmental stressors could potentially explain some of the observed differences in responses to environmental challenges that we observed between strains. To better understand the differences between these strains further studies assessing the metabolite profiles of these fish is warranted. Domestication may provide benefits in addition to enhanced growth rates. We observed better performance in our domesticated strain compared to the three wild strains and, assuming our observed differences are a result of domestication and not simply strain this suggests that domestication acts on many traits to improve overall performance. With the FV domestics we measured higher critical 41  swimming speeds, time to LOE (2010), and routine metabolic rates which may overall suggest some advantages of using domesticated fish in lake stocking programs not just for the benefits of enhanced growth rates but also for the improved performance which we have demonstrated. However, this suggestion is in contrast with a study that showed domestication can impair tolerance to environmental challenges, specifically thermal tolerance, in trout (Carline and Machung, 2001). Although it is more likely that domestication will affect behavioural traits than traits of environmental tolerance (Leider et al., 1990), more studies need to be done on the effects of domestication on tolerance to environmental challenges. It is important to evaluate how different 2n and 3n families of a strain may differ in performance in response to environmental challenges because of the potential benefits of selective breeding at the family level. Johnson et al. (2004) suggests that better performing 2n families can be selected for breeding programs to improve the overall survival of 3n fry. An interesting comparison that Johnson et al. (2004) makes in his study with another study (Friars et al. 2001) is the relationship between domestication and variation in traits at the family level. The authors observed significant variation in growth rates and survival between families domesticated for only one generation in the study by Friars et al. (2001) but observed no significant variation in their own study, where families were domesticated for six generations. In the present study, a comparison of variation between families of the FV domestics and families of the wild strains did not reveal any effect of domestication on family variation for any performance measure which suggests that sufficient family variation is present in our domesticated strain to allow for selecting better performing 2n families for breeding programs which may improve 3n FV survival. Since the study of triploid physiology is still in its infancy, there is much that we cannot yet explain. For example, one 2n BW family from the 2010 assessment of time to LOE (figure 2.4b) was able to tolerate the hypoxia challenge for three times longer than the two other 2n BW families as well as its  42  3n siblings. It is possible that some of these differences can be explained by gene dosage effects (Devlin et al. 1988) and increased heterozygosity (Thorgaard et al. 1999), which are poorly understood and possibly vary across families. Additionally, the role that family plays in determining the response of strain to environmental challenges is also unclear. For example, all of our measurements the coefficient of variation was greater for between family comparisons than strain/ploidy comparisons which suggest that much of the differences that we observed between strains and ploidies are more likely to be as a result of family than strain, per se.  2.4.3 Recommendations and conclusion The comprehensive assessment of performance and tolerance undertaken in the present study allows us to make some informed recommendations for lake stocking programs. First and foremost, lakes where 3n fish are stocked should be oxygen profiled at least in the winter if the lake freezes over and in the summer to see if stratification occurs. Secondly, in hypoxia-prone lakes it might be advantageous to use an alternative strain than Tzenzaicut, which has been shown to be comparatively less tolerant of hypoxia compared to its conspecifics. Thirdly, the Fraser Valley domestics strain did not seem to be more sensitive to environmental challenges nor did the domesticated strain underperform in any of the performance measures. On the contrary, the domesticated strain performed better than the wild strains used in this study. Lastly, if the prevalence of hypoxia in bodies of water where fish are stocked is expected to rise in the future it may be worthwhile to invest in selective breeding programs where broods are generated from more environmentally tolerant families. Our data shows a large variation in environmental tolerance between families which suggests a high capacity to breed desirable tolerance traits into fish used in lake stocking programs here in British Columbia. In summary, we found that 3n trout consistently have poorer hypoxia tolerance compared to their 2n counterparts. Poorer 3n tolerance of hypoxia persisted across five strains of rainbow trout and 43  three different brood years. We also determined the FV domestic strain to be the most tolerant across assessments of environmental performance and the TZ strain to be less tolerant, with it consistently showing poor hypoxia tolerance. We did not observe similar differential 2n/3n tolerance of hypoxia in lake reared adult trout as we did with the three years of juveniles. Lastly, we evaluated variation in performance in response to environmental challenges at the level of family, strain, and ploidy and found that variation between families was greater than the variation between strains and ploidies for all performance measures which is promising for improving performance in response to environmental challenges of stocked trout through selective breeding programs.  44  Table 2.1. Weights and lengths at time of analysis and correlation coefficients for selected measures of performance in response to environmental challenges of four strains of 2n and 3n rainbow trout.  Sample sizes, weights or lengths, and correlation coefficients Ucrit 2008 Strain  Ploidy  n  Length (cm)  2n  24  14.5±0.1  3n  24  14.1±0.5  2n  24  13.6±0.2  3n  24  14.7±0.2  2n  24  10.8±0.2  3n  24  10.0±0.2  2n  24  10.7±0.0  Correlation  n  Weight (g)  P=0.172  50  1.84±0.05  B  P<0.0001, r=-0.887  50  1.69±0.05  B  P=0.113  30  1.74±0.05  B  P=0.0114, r=-0.507  30  1.60±0.05  A  P=0.0082, r=-0.527  30  1.38±0.07  A  P=0.0395, r=-0.423  30  A  P=0.177  40  TZ  FV 3n  Correlation  B  BW  PN  Muscle LDH enzyme activities 2009  40  Time to LOE 2009 n  Weight (g)  Time to LOE 2010 n  Weight (g)  A  24  8.1±0.2  A  24  8.0±0.2  A  24  8.1±0.3  A  24  8.1±0.3  A  24  10.6±0.3  A  24  10.5±0.3  B  24  16.3±0.5  B  24  16.4±0.5  C  P=0.064  99  3.2±0.1  C  P<0.0001, r=0.634  105  3.3±0.1  C  P=0.270  60  3.3±0.1  P=0.0002, r=0.643  56  3.3±0.1  AB  P=0.0299, r=0.404  72  3.1±0.1  A  P=0.0020, r=0.541  70  3.1±0.1  BC  P=0.261  87  3.7±0.1  B  P=0.140  71  3.6±0.1  BC  1.25±0.05 1.58±0.04  1.46±0.04  Time to LOE Lake fish Weight (g)  n  Correlation  A  22  725±105  CD  A  13  311.99±  A  22  449±81  A  4  B  15  B  10  AB  BC  D  926±63  ABC  351±102  A  139±63  P=0.0014, r=-0.637 P=0.0488, r=-0.555 P=0.0003, r=-0.696 P=0.4823 P=0.2146 P=0.1161  C  C  Means ± SE are presented. Superscript letters that differ within a single column denote significant differences between the means of individuals pooled from all families of a given strain and ploidy.  45  Table 2.2. Critical thermal maximum of three strains of 2n and 3n rainbow trout.  Critical thermal maxima n  o  2n  31  28.73±0.08  3n  32  29.12±0.08  2n  23  29.14±0.09B  3n  16  28.67±0.16A  2n  21  29.11±0.09B  3n  18  28.71±0.09A  Strain Ploidy  C A  BW  TZ  PN  B  Means ± SE are presented for the 2009 brood year. Superscript letters that differ within a single column denote significant differences between the means of individuals pooled from all families of a given strain and ploidy. There was no effect of strain (P=0.76) but there was an effect of ploidy (P=0.004) on CTMax.  46  Table 2.3. Time to LOE and percent recapture of three strains of 2n and 3n adult lake rainbow trout.  Time to LOE n  Percent recapture  min  2n  22  6.1  209±32  3n  13  4.7  271±35  2n  15  14.5  294±24  3n  10  4.4  220±52  2n  22  7.8  263±34  3n  4  3.7  217±33  Strain Ploidy  BW  TZ  PN  Means ± SE are presented for the adult fish. There was no effect of strain (P=0.9046) nor was there an effect of ploidy (P=0.5584) on time to LOE.  47  Table 2.4. Muscle pyruvate kinase, citrate synthase, and lactate dehydrogenase enzyme activities per gram wet weight of four strains of 2n and 3n rainbow trout.  Muscle enzyme activities PK Strain Ploidy  CS μmol min-1 g-1  n AB  PN  FV  AB  202±8  A  1.13±0.04  ABC  201±6  A  1.13±0.06  ABC  254±9  2n  50  3n  50  7.6±0.3  A  2n  30  7.0±0.5  A  3n  30  7.7±0.7AB  1.00±0.07A  244±12B  2n  30  7.5±0.5A  0.93±0.05A  196±10A  3n  30  7.9±0.4AB  1.00±0.04A  233±11AB  2n  40  9.7±0.5B  1.31±0.06C  236±8AB  3n  40  8.6±0.4AB  1.29±0.08BC  229±10AB  7.9±0.3  BW  TZ  LDH  1.07±0.05  B  Means ± SE are presented for the 2009 brood year. Superscript letters that differ within a single column denote significant differences between the means of individuals pooled from all families of a given strain and ploidy. For PK, CS, and LDH there were effects of strain (P=0.0007, P<0.0001, and P<0.0001) but no effects of ploiy (P=0.8202, 0.9326, and 0.3233) on enzyme activity.  48  Table 2.5. Specific growth rates (weights and lengths) taken between 1110 and 1530 ATUS of four strains of 2n and 3n rainbow trout from the 2009 brood year.  Specific growth rates Weight  Length  n  (g day-1)  (cm day-1)  2n  3  1.50±0.21AB  0.41±0.09  A  3n  3  1.53±0.14  AB  0.44±0.06  A  2n  3  2.15±0.24B  0.62±0.14AB  3n  3  2.33±0.30BC  0.70±0.11AB  2n  3  1.43±0.11A  0.47±0.03A  3n  3  1.35±0.16A  0.43±0.05A  2n  4  3.12±0.04C  0.98±0.02C  3n  4  2.88±0.09C  0.86±0.02B  Strain Ploidy  BW  TZ  PN  FV  Means ± SE are presented for the 2009 brood year. Letters denote significant differences between the means of families pooled from a given strain and ploidy, n= the number of families. There was a significant effect of strain for both length (P<0.0001) and weight (P<0.0001) but no effect ploidy (P=0.629, P=0.598), respectively, on specific growth rates.  49  Figure 2.1. Critical swimming speeds of four strains of 2n and 3n rainbow trout. Means ± SE are presented for 2008 (a) and 2009 (b) brood years. Circles represent different families. Closed circles are 2n and open are 3n. Box plot boundaries indicate 25th and 75th percentiles and whiskers denote 10th and 90th percentiles. Black bars are the mean and grey bars the median. Letters that differ denote significant differences between the means of individuals pooled from all families of a given strain and ploidy. For the 2008 crosses there was an effect on strain (P<0.0001) but no effect of ploidy (p=0.3950). For 2009 there was an effect of strain (P<0.0001) but no effect of ploidy (P=0.0959) on Ucrit.  50  Figure 2.2. Mass-specific oxygen consumption rates of four strains of 2n and 3n rainbow trout. Means ± SE are presented for the 2009 brood year. Circles represent different families. Closed circles are 2n and open are 3n. Box plot boundaries indicate 25th and 75th percentiles and whiskers denote 10th and 90th percentiles. Black bars are the mean and grey bars the median. Letters that differ denote significant differences between the means of individuals pooled from all families of a given strain and ploidy. There was an effect of strain (P<0.0001) but there was no effect of ploidy (P=0.3323) on mass-specific oxygen consumption.  51  Figure 2.3. Critical oxygen tensions of four strains of 2n and 3n rainbow trout. Means ± SE are presented for the 2009 brood year. Circles represent different families. Closed circles are 2n and open are 3n. Box plot boundaries indicate 25th and 75th percentiles and whiskers denote 10th and 90th percentiles. Black bars are the mean and grey bars the median. Letters that differ denote significant differences between the means of individuals pooled from all families of a given strain and ploidy. There was an effect of strain (P<0.0001) but there was no effect of ploidy (P=0.1210) on critical oxygen tensions.  52  Figure 2.4. Time to LOE of four strains of 2n and 3n rainbow trout. Means ± SE are presented for the 2009(a) and 2010(b) brood years. Circles in (a) represent families of unrelated parentage and shapes in (b) represent 2n and 3n siblings of a given strain. Closed shapes are 2n and open are 3n. Box plot boundaries indicate 25th and 75th percentiles and whiskers denote 10th and 90th percentiles. Black bars are the mean and grey bars the median. Letters that differ denote significant differences between the means of individuals pooled from all families of a given strain and ploidy. For both 2009 and 2010 there was an effect of strain (P<0.0001 and P<0.0001, respectively) and an effect of ploidy (P<0.0001 and P=0.0001, respectively) on time to LOE.  53  Figure 2.5. Time to LOE of two strains of 2n and 3n rainbow trout. Means ± SE are presented for the 2011 brood year. Shapes represent 2n and 3n siblings of given strain. Closed shapes are 2n and open are 3n. Box plot boundaries indicate 25th and 75th percentiles and whiskers denote 10th and 90th percentiles. Black bars are the mean and grey bars the median. Letters that differ denote significant differences between the means of the three families for both 2n and 3n. There was both a significant effect of strain (P<0.0001) and ploidy (P=0.144) on time to LOE.  54  3  General discussion and conclusions  3.1  Overview I performed a comprehensive laboratory assessment of the responses of wild and domestic  strains of 2n and 3n rainbow trout to environmental challenges, I hope that fisheries biologists will be able to make more informed decisions about what strain of 3n fish is appropriate for stocking into B.C. lakes. A better understanding of how triploids respond to environmental challenges will result in reduced 3n mortality because strains that are better suited for a particular habitat can be more appropriately selected for that habitat. The main goals of this thesis were to determine if there was an environmental challenge where 3n trout consistently did poorer than their 2n counterparts across strains and years, to determine what strain of trout performs best across assessments of environmental performance, and ultimately, to better understand what may be contributing to poorer 3n survival in the wild. This research adds to the limited number of studies in the literature on 3n performance in response to environmental challenges. Comparing 2n and 3n responses to environmental challenges is necessary because 3n fish have been observed to sometimes show greater mortality in natural environments or under suboptimal rearing conditions than their 2n counterparts. Despite sometimes showing greater mortality, triploids are used in lake stocking programs because they are sterile and have been shown to have better flesh quality and improved growth rates at maturation. An additional benefit of using sterile fish is that it prevents hybridization between hatchery reared and native populations which aids in preserving the natural biodiversity of the habitat. Part of what makes rainbow trout a suitable species for aquaculture is their fast growth rates and good exercise performance. These qualities along with being good for eating and fishing have lead to the introduction of rainbow trout into lake stocking programs all around the world.  55  3.2  The effects of ploidy and strain on performance in response to environmental challenges To my knowledge, only two studies have been done comparing 2n/3n hypoxia tolerance.  Yamamoto and Iida (1994) using rainbow trout and Lilyestrom et al. (1999) with catfish hybrids both found that 3n compared to 2n have reduced hypoxia tolerance. My findings are in agreement with these two studies. I showed that across a total of five strains and three years that 3n juvenile rainbow trout consistently have poorer tolerance of hypoxia compared to their 2n counterparts. This finding adds support to the suggestion in the literature that 3n compared to 2n have impaired tolerance of high oxygen demand/low oxygen availability conditions. However, my finding of no difference in Pcrit among four strains of 2n and 3n trout may suggests otherwise. Pcrit is described as a whole-animal measure of the ability of fish to acquire oxygen from their environment (Gannon et al., 1999). Despite there being studies equating Pcrit with hypoxia tolerance (Mandic, et al., 2008; Speers-Roesch et al., 2012), it is not always the case that a relationship is observed between the two. Yao and Richards (in prep.) found significant differences in time to LOE between many species of Danio and Devario but found no differences in Pcrit. Aerobic capacity describes the level of aerobic fitness of fish and has been thoroughly characterised in trout (Altimiras et al., 2002; Stillwell and Benfey, 1997). It is a valuable indicator of performance in response to environmental challenges because it describes a fish’s ability to uptake and transport oxygen which is important under conditions of oxygen limitation (Aliah et al., 1991; Yamamoto and Iida, 1994). Studies measuring standard, routine, and active metabolic rates have all found no differences between 2n and 3n fish (Sezaki et al., 1991; Parsons, 1993; Yamamoto and Iida, 1994). Consistent with the literature, I did not find any effect of ploidy on routine metabolic rates among four strains of 2n and 3n rainbow trout.  56  Similarly, I did not find any difference in swimming performance between 2n and 3n trout among four strains and across two years. Several studies have compared 2n and 3n critical swimming speeds (Sezaki et al., 1991; Stillwell and Benfey, 1996b) but only one has found 2n and 3n to differ in swimming performance (Virtanen et al., 1990). Since swimming performance is commonly used as an alternative measure of aerobic capacity, the routine metabolic rate and swimming performance results I produced are consistent with the notion that 2n and 3n do not differ in aerobic capacity. There are few studies that have measured CTMax but those that have reported no difference between 2n and 3n trout (Benfey et al., 1997; Galbreath et al., 2006). In contrast with these studies, I found a significant difference in CTMax among three strains of 2n and 3n rainbow trout. However, my findings of differential 2n/3n CTMax are consistent with other studies comparing 2n/3n thermal tolerance that used different methods. Hyndman et al. (2003) showed that there were differences in the ability of 2n and 3n trout to recover in warm water following exhaustive exercise with 3n trout showing a significantly higher rate of mortality than 2n trout. Similarly, Ojolick, et al. (1995) found that female triploid brook trout did not grow or survive as well as their diploid counterparts when reared at high temperatures. It should be noted that the direction of the difference between 2n and 3n trout in CTMax we measured was not the same for all three strains. The CTMax of 3n trout was reduced in the TZ and PN strains but was the opposite in the BW strain, 2n CTMax was reduced compared to 3n. In light of these details, my findings do not necessarily give credence towards impaired 3n thermal tolerance; rather, they highlight the importance of using a multi-strain approach when assessing 2n/3n responses to environmental challenges. With the exception of CTMax, I observed a significant effect strain on all assessments of performance in response to environmental challenges. Similarly, other studies have found no difference in multi-strain assessments of CTMax (Lee and Rinne, 1980; Wagner et al., 2001; Rodnick, et al., 2004), but differences in intra- and interstrain response to other environmental challenges (Klar et al., 1979;  57  Van Leeuwen et al., 2011; Dalziel and Schulte, 2012). Since my trout were reared under common garden conditions, multiple instances of significant variation between strains suggest that there is much heritable variation that can be utilized for better lake stocking strategies. Breeding programs involving strains with desirable traits related to growth and angler satisfaction, and strains with higher tolerance to environmental challenges should result in the production of robust stocks better suited for specific lakes. My results showing that the FV domestic strain has high growth and tolerance of hypoxia, and better swimming performance relative to other strains suggests that this strain is a good candidate for breeding programs. To summarize, the majority of studies comparing 2n and 3n responses to environmental challenges only show significant differences between 2n and 3n fish in hypoxia tolerance (reviewed by Benfey, 1999 and Maxime, 2008). My findings of reduced 3n compared to 2n hypoxia tolerance in multiple strains over multiple years are consistent with the literature. Similarly, my findings of no difference between 2n and 3n routine metabolic rates and swimming performance are also consistent with the literature. As for thermal tolerance, where there is no consensus of the effect of triploidy in the literature, my findings of both reduced and enhanced (depending on the strain) 3n compared to 2n CTMax only contribute to this uncertainty. To my knowledge, this is the first time that Pcrit has been determined in 3n fish and my findings of no difference between 2n and 3n rainbow trout may go contrary to the general consensus and suggest that 2n and 3n fish do not differ in their hypoxia tolerance or it may simply mean that 2n and 3n do not differ in their ability to extract oxygen from their environment.  3.3  Future directions Since natural environments where rainbow trout are stocked often undergo hypoxic events as a  result of changing temperatures it would be worthwhile to investigate the interactive effects of a 58  thermal challenge and a hypoxia challenge. It is possible that 3n trout would show greater mortality compared to their 2n counterparts because 3n fish have been found to be more impaired by conditions of high oxygen demand and/or low oxygen availability (Blanc et al., 1992; Simon et al., 1993; and Johnstone, 1996), especially since trout exposed to elevated water temperatures have been found to have increased oxygen demand (reviewed by Portner, 2001). Additionally, temperature and hypoxia tolerance have been shown to be related as high temperatures can lead to temperature-induced hypoxemia (Portner et al., 2004). Using a multi-strain approach in this future study might also reveal variation between strains to interactive tolerances. If a population is found to have higher interactive tolerances than others then it could be incorporated into a breeding strategy to generate brood stocks that are more suitable for lakes prone to both fluctuating thermal and oxygen regimes.  59  References Aliah, R. S., Inada, Y., Yamoaka, K., and Taniguchi, N. 1991. Effects of triploidy on hematological characteristics and oxygen consumption of ayu. Nip. Sui. Gak. 57: 833-836. Allen, S. K. Jr. 1983. Flow cytometry: assaying experimental polyploidy fish and shellfish. Aqua. 33: 317328. Almodovar, A. and Nicola, G. G. 1998. Assessment of a brown trout Salmo trutta population in the River Gallo (Central Spain): angling effects and management implications. Ital. J. Zoo. 65: 539-543. Almodovar, A., Nicola, G. G., and Suarez, J. 2001a. Effects of fishery management on populations of brown trout Salmo trutta in central Spain. In Freshwater Fish Conservation. Options for the Future (Collares-Pereira, M.J., Cowx, I., and Coelho, M.M., eds), Pp. 337-345. Oxford: Fishing News Books, Blackwell Science. Almodovar, A., Suarez, J., Nicola, J. J., and Nuevo, M. 2001b. Genetic introgression between wild and stocked brown trout in the Douro River basin, Spain. J. Fish Biol. 59(A): 68-74. Altimiras, J., Axelsson, M., Claireaux, G., Lefrancois, C., Mercier, C. and Farrell, A. P. 2002. Cardiorespiratory status of triploid brown trout during swimming at two acclimation temperatures. J. Fish Bio. 60(1): 102-116. Arnott, S. A., Chiba, S., and Conover, D. A. 2006. Evolution of intrinsic growth rate: metabolic costs drive trade-offs between growth and swimming performance in Menidia menidia. Evolution. 60, 6: 1269-1278. Arumuganathan, K. And Earle, E. D. 1991. Estimation of nuclear DNA content of plants by flow cytometry. Plant Mol. Biol. Rpt. 9: 229-241. Babouchkine, Y. P. 1987. La selection d’une carpe resistant a l’hiver. Aqua. 1: 447-454. Benfey, T. J. 1999. The physiology and behaviour of triploid fishes. Rev. Fish. Sci. 7: 39-67. Benfey, T. J. 2001. Use of sterile triploid Atlantic salmon (Salmo salar L.) for aquaculture in New Brunswick, Canada. ICES J. Mar. Sci. 58: 525-529. Benfey, T. J., Bosa, P. G., Richardson, N. L., and Donaldson, E. M. 1988. Effectiveness of a commercialscale pressure shocking device for producing triploid salmonids. Aqua. Eng. 7: 147-154. Benfey, T. J., McCabe, L. E., and Pepin, P. 1997. Critical thermal maxima of diploid and triploid brook charr, Salvenilus fontinalis. Envir. Biol. Fish. 49: 259-264. Benfey, T. J., Sutterlin, A. M., and Thompson, R. J. 1984. Use of erythrocyte measurements to identify triploid salmonids. Can. J. Fish. Aquat. Sci. 41: 980-984.  60  Bernier, N. J., Fuentes, J., and Randall, D. J. 1996. Adenosine receptor blockade and hypoxia-tolerance in rainbow trout and pacific hagfish. J. Exp. Biol. 199: 497-507. Bidgood, B. F. 1980. Tolerance of rainbow trout to direct changes in water temperature. Fish. Res. Rep. Fish Wildl. Div. 15: 11. Biron, M. And Benfey, T. J. 1994. Cortisol, glucose, and hematocrit changes during acute stress, cohort sampling, and the diel cycle in diploid and triploid trout (Salvelinus fontinalis Mitchill). Fish Physiol. Biochem. 13: 153-160. Blanc, J. M., Poisson, H., and Vallee, F. 1992. Survival, growth and sexual maturation of the triploid hybrid between rainbow trout and Arctic char. Aquat. Liv. Res. 5: 15-21. Bondari, K. 1983. Response to bi-directional selection for body weight in channel catfish. Aqua. 33: 7381. Boutilier, R. G., Dobson, G., Hoeger, U., and Randall, D. J. 1988. Acute exposure to graded levels of hypoxia in rainbow trout (Salmo gairdneri): metabolic and respiratory adaptations. 71: 69-82. Boyd, C. E., and Tucker, C. S. 1998. Pond aquaculture water quality management. Kluwer Academic Publishers, Boston. 700. Carline, R. F. and Machung, J. F. 2001. Critical thermal maxima of wild and domestic strains of trout. Trans. Am. Fish. Soc. 130(6): 1211-1216. Chapman, L. J., Kaufman, L. S., Chapman, C. A., and Mckenzie, F. E. 1994. Hypoxia tolerance in twelve species of east African cichlids: potential for low oxygen refugia in Lake Victoria. Cons. Biol. 9(5): 12741287. Chapman, L. J. and McKenzie, D. 2009. Behavioural responses and ecological consequences. Pp. 26-77 in J. G. Richards, A. P. Farrell, and C. J. Brauner, eds. Fish Physiology. Vol. 27. Hypoxia. Academic Press, San Diego, CA. Chourrout, D. 1984. Pressure-induced retention of second polar body and suppression of first cleavage in rainbow trout: production of all-triploids, all-tetraploids, and heterozygous and homozygous diploid gynogenetis. Aqua. 36: 111-126. Clarke, A. 2012. Rainbow trout strains stocked in British Columbia: matching fish characteristics with habitat to maximize fishing quality. BCO Sport Fish. 69: 20-26. Cormier, S. M., Neiheisel, T. W., Williams, D. E., and Tiersch, T. R. 1993. Natural occurrence of triploidy in a wild brown bullhead. Trans. Am. Fish. Soc. 122(3): 390-392. Cotter, D., O’Donovan, V., O’Maoileidigh, N. O., Rogan, G., Roche, N. and Wilkins, N. P. 2000. An evaluation of the use of triploid Atlantic salmon (Salmo salar L.) in minimising the impact of escaped farmed salmon on wild populations. Aqua. 186: 61-75.  61  Cuellar, O. and Uyeno, T. 1972. Triploidy in rainbow trout. Cyto. 11: 508-515. Currie, R. J., Bennet, W. A., and Beitinger, T. L. 1998. Critical thermal minima and maxima of three freshwater game-fish species acclimated to constant temperatures. Environ. Biol. Fish. 51: 198-200. Dalziel, A. C., Ou, M., and Schulte, P. M. 2012. Mechanisms underlying parallel reductions in aerobic capacity in non-migratory threespine stickleback (Gasterosteus aculeatus) populations. J. Exp. Bio. 215: 746-759. Dalziel, A. C. and Schulte, P. M. 2012. Correlates of prolonged swimming performance in F2 hybrids of migratory and non-migratory threespine stickleback ecotypes. J. Exp. Biol. Davison, W. 1989. Training and its effects on teleost fish. Comp. Biochem. Physiol. 94(A): 1-10. Devlin, R.H., Holm, D.G., and Grigliatti, T.A. 1988. The influence of whole-arm trisomy on gene expression in Drosophila. Gen. 118: 87-101. Duarte, C. M., Marba, N., and Holmer, M. 2007. Rapid domestication of marine species. Sci. 316(5823): 382-383. Dunn, J. F., and Hochachka, P. W. 1986. Metabolic responses of trout (Salmo gairdneri) to acute environmental hypoxia. J. Exp. Biol. 123: 229-242. Doudoroff, P. and Shumway, D. L. 1970. Dissolved oxygen requirements of freshwater fishes. FAO Fisheres Technical Paper no.86. Rome, Italy: Food and Agricultural Organization of the United Nations. Elliot, J. M. and Elliot, J. A. 1995. The effect of the rate of temperature increase on the critical thermal maximum for parr of Atlantic salmon and brown trout. J. Fish Biol. 47: 917-919. Fangue, N.A., Hofmeister, M., and Schulte, P.M. 2006. Intraspecific variation in thermal tolerance and heat shock protein gene expression in common killifish, Fundulus heteroclitus. J. Exp. Biol. 209: 28592872. FAO: Special fact sheets: Oncorhynchus mykiss (Walbaum, 1972) Fisheries and Aquaculture Dempartment. Accessed Aug. 30, 2012. Farrell, A. P., Johansen, J. A., and Suarez, R. K. 1991. Effects of exercise-training on cardiac performance and muscle enzymes in rainbow trout, Oncorhynchus mykiss. Fish Phys. Bioc. 9(4): 303312. Ferguson, A. 2006. Genetic impacts of stocking on indigenous brown trout populations. Environmental Agency. UK. Pp. 12-14. Freshwater Fisheries Society of British Columbia. 2004. Rainbow trout strains currently stocked in BC waters.  62  Galbreath, P. F., Adams, N. D., Sherrill, L. W., and Martin, T. H. 2006. Thermal tolerance of diploid versus triploid rainbow trout and brook trout assessed by time to chronic lethal maximum. Envir. Biol. Fish. 75: 183-193. Galbreath, P. F., St. Jean, W., Anderson, V., and Thorgaard, G. H. 1994. Freshwater performance of allfemale diploid and triploid Atlantic salmon. Aqua. 128: 41-49. Gallaugher, P., Axelsson, M., and Farrell, A. P. 1992. Swimming performance and haematological variables in splenectomized rainbow trout, Oncorhynchus mykiss. J. Exp. Biol. 171: 301-314. Gannon, A. T., Demarco, V. G., Morris, T., Wheatly, M. G., and Kao, Y. H. 1999. Oxygen uptake, critical oxygen tension, and available oxygen for three species of cave crayfishes. J. Crus. Bio. 19(2): 235-243. Gee, J. H., and Gee, P. A. 1991. Reaction of gobioid fishes to hypoxia tolerance: Buoyancy control and aquatic surface respiration. Cop. 17-28. Giller, S. and Malmqvist, B. 1998. The biology of stream and rivers. Oxford University Press, Oxford. Pp. 296. Gjedrem, T. 1979. Selection for growth rate and domestication in Atlantic salmon. Zeit. Tier. Zuch. 96: 56-59. Gjerde, B. 1986. Kvalitet: Fiskeoppdrett med framtid. Trygve Gjedrem. Gjerde, B. 2005. Design of fish breeding programs. Pp 173-195 in Selection and Breeding Programs in Aquaculture. T. Gjedrem, ed. Springer, New York, NY. Graham, M. S., Fletcher, G. L., and Benfey, T. J. 1985. Effect of triploidy on blood oxygen content of Atlantic salmon. Aqua. 50: 133-139. Gregory, T. R. and Wood, C. M. 1998. Individual variation and interrelationships between swimming performance, growth rate, and feeding in juvenile rainbow trout (Oncorhynchus mykiss). Can. J. Fish Aquat. Sci. 55: 1583-1590. Guo, X., Hershberger, W. K., and Myers, J. M. 1990. Growh and survival of intrastrain and interstrain rainbow trout (Oncorhynchus mykiss) triploids. J. World Aquac. Soc. 21: 250-256. Henriksson, P., Mandic, M. and Richards, J. G. 2008. The osmorespiratory compromise in sculpins: impaired gas exchange is associated with freshwater tolerance. Physiol. Biochem. Zool. 81: 310-319. Hothorn, T., Bretz, F., and Westfall, P. 2008. Simultaneous inference in general parametric models. Biom. J. 50: 346-363. Houlihan, D. F. and Laurent, P. Effects of exercise training on the performance, growth, and protein turnover of rainbow trout (Salmo gairdneri). Can. J. Fish. Aquat. Sci. 44: 1614-1621. Hughes, G. M. 1973. Respiratory responses to hypoxia in fish. Amer. Zool. 13: 475-489.  63  Hyndman, C. A., Kieffer, J. D., and Benfey, T. J. 2003. Physiology and survival of triploid brook trout following exhaustive exercise in warm water. Aqua. 221: 629-643. Johnson, R. M., Shrimpton, M. J., Heath, J. W., and Heath, D. D. 2004. Family, induction methodology and interaction effects on the performance of diploid and triploid Chinook salmon (Oncorhynchus tshawytscha). Aqua. 234: 123-142. Johnstone, R. 1985. Induction of triploidy in Atlantic salmon by heat shock. Aqua. 49: 133-139. Johnstone, R., 1996. Experience with salmonid sex reversal and triploidization technologies in the United Kingdom. Bull. Aquac. Assoc. Can. 96(2): 9-13. Johnstone, R., McLay, H. A., and Walsingham, M. V. 1991. Production and performance of triploid Atlantic salmon in Scotland. Can. Tech. Rep. Fish Aquat. Sci. 51(1): 25-30. Jones, D.R. 1971. The effects of hypoxia and anaemia on the swimming performance of rainbow trout (Salmo Gairdneri). J. Exp. Biol. 55: 541-551. Kirpichnikov, V. S., Ilyasov, J. I., Shart, L. A., et al. 1993. Selection of Krasnodar common carp (Cyprinus carpio L.) for resistance to dropsy: principal results and prospects. Aqua. 111: 7-20. Klar, G. T., Stalnaker, C. B., and Farley, T. M. 1979. Comparative physical and physiological performance of rainbow trout, Salmo gairdneri, of distinct lactate dehydrogenase B2 phenotypes. Comp. Bioc. Phys. 63(A): 229-235. Koenig, M., Kozfkay, J. and Meyer, K. 2011. Performance of diploid and triploid rainbow trout stocked in Idaho alpine lakes. Nor. Am. J. Fish. Mgmt. 31: 124-133. Kolok, A. S. and Oris, J. T. 1995. The relationship between specific growth rate and swimming performance in male fathead minnows (Pimephales promelas). Can. J. Zool. 73: 2165-2167. Lee, R. M. and Rinne, J. N. 1980. Critical thermal maxima of five trout species in the southwestern United States. Trans. Am. Fish Soc. 109: 632-635. Leider, S. A., Hulett, P. L., Loch, J. J., and Chilcote, M. W. 1990. Electrophoretic comparison of the reproductive success of naturally spawning transplanted and wild steelhead trout through the returning adult stage. Aqua. 88(3-4): 239-252. Lilyestrom, C. G., Wolters, W. R., Bury, D., Rezk, M., and Dunham, R. A. 1999. Growth, carcass traits, and oxygen tolerance of diploid and triploid catfish hybrids. Nor. Am. J. Aqua. 61: 293-303. Liu, S., Liu, Y., Zhou, G. et al. 2001. The formation of tetraploid stocks of red crucian carp x common carp hybrids as an effect of interspecific hybridization. Aqua. 192: 171-186. Machordom, A., Garcia-Marin, J., Sanz, L. N., Almodovar, A., and Pla, C. 1999. Allozyme diversity in brown trout (Salmo trutta) from Central Spain: Genetic consequences of restocking. Fresh. Biol. 41: 707717.  64  Machordom, A., Suarez, J., Almodovar, A. and Bautista, J. M. 2000. Genetic differentiation and phylogenetic relationships among Spanish brown trout (Salmo trutta) populations. Molec. Eco. 9: 13251338. MacConnell, E. 1989. Effects of water reuse on lake trout. Prog. Fish. Cult. 51: 33-37. Mandic, M., Todgham, A. E., and Richards, J. G. 2008. Mechanisms and evolution of hypoxia tolerance in fish. Proc. R. Soc. B. 276: 735-744. Marvin, D. E. and Heath, A. G. 1968. Cardiac and respiratory responses to gradual hypoxia in three ecologically distinct species of fresh-water fish. Comp. Biochem. Physiol. 21: 349-355. Maxime, V. 2008. The physiology of triploid fish: current knowledge and comparisons with diploid fish. Fish Fish. 9: 67-78. McCarthy, I. D., Carter, C. G., Houlihan, D. F., Johnstone, R., and Mitchell, A. I. 1996. The performance of all-female diploid and triploid Atlantic salmon smolts on transfer together to sea water. J. Fish. Biol. 48: 545-548. McCusker, M., Parkinson, E. A., and Taylor, E. B. 2000. Phylogenetic conservation units for rainbow trout in British Columbia. Province of British Columbia. Fisheries Management Report 112. McGeachy, S. A., Benfey, T. J., and Friars, G. W. 1995. Freshwater performance of triploid Atlantic salmon (Salmo salar) in New Brunswick aquaculture. Aqua. 137: 333-341. McKenzie, D. J., Pedersen, P. B., and Jokumsen, A. 2007. Aspects of respiratory physiology and energetic in rainbow trout (Oncorhynchus mykiss) families with different size-at-age condition factor. Aqua. 263: 280-294. Milner, J. W. 1874. The progress of fish-culture in the United States. United States Commission of Fish and Fisheries Report of the Commissioner for 1872 and 1873. 535-544. Moav, R. and Wohlfarth, G. 1976. Two-way selection for growth rate in the common carp (Cyprinus carpio L.). Gen. 82: 83-101. Moav, R., Wohlfarth, G., and Lahman, M. 1964. Genetic improvement of carp: VI. Growth rate of carp imported from Holland relative to Israeli carp and some crossbred progeny. Bam. 16: 142-149. Moav, R., Hulata, G., and Wohlfarth, G. 1975. Genetic differences between the Chinese and European races of common carp: I. Analysis of genotype-environment interactions for growth rate. Her. 34: 323340. Molony, B. 2001. Environmental requirements and tolerances of rainbow trout (Oncorhynchus mykiss) and brown trout (Salmo trutta) with special reference to Western Australia: A review. Fish. Res. Rep. West. Aust. 130: 1-28.  65  Nagai, T., Tamura, T., Iida, Y., and Yoneji, T. 2004. Differences in susceptibility to Flavobacterium psychrophilium among three stocks of ayu, Plecoglossus altivelis. Fish Pathol. 39: 159-164. Nilsson, G. E. 2007. Gill remodeling in fish – a new fashion or an ancient secret? J. Exp. Biol. 210: 24032409. Nilsson, G. E., Rosen, P., and Johansson, D. 1993. Anoxic depression of spontaneous locomotor activity in crucian carp quantified by computerized imaging technique. J. Exp. Biol. 180: 153-162. O’Flynn, F. M., McGeachy, S. A. Friars, G. W., Benfey, T. J., and Bailey, J. K. 1997. Comparisons of cultured triploid and diploid Atlantic salmon (Salmo salar L.). ICES J. Mar. Sci. 54: 1160-1165. Ojolick, E. J., Cusack, R., Benfey, T. J. & Kerr, S. R. 1995. Survival and growth of all-female diploid and triploid rainbow trout (Oncorhynchus mykiss) reared at chronic high temperature. Dev. 131(94): 177187. Parsons, G. R. 1993. Comparisons of triploid and diploid white crappies. Trans. Amer. Fish. Soc. 122: 237-243. Pelletier, D., Dutil, J. D., Blier, P., and Guderley, H. 1994. Relation between growth rate and metabolic organization of white muscle, liver, and digestive tract in cod, Gadus morhua. J. Comp. Phys. B: Biochem. Syst. Enviro. 164(3): 179-190. Peterson, M. S., and Meador, M. R. 1994. Effects of salinity on freshwater fishes in coastal plain drainages in the southeaster U.S. Rev. Fish. Sci. 2: 95-121. Pinheiro, J. C., Bates, D. M., Debroy, S., Sarkar, D., and the R Development Core Team. 2009. nlme: linear and nonlinear mixed effects models. R package version 3.1-96. Vienna, Austria: R Foundation for Statistical Computing. Portner, H. O. 2001. Climate change and temperature-dependent biogeography: oxygen limitation of thermal tolerance in animals. Naturwiss. 88(4): 137-146. Portner, H. O., Mark, F. C., and Bock, C. 2004. Oxygen limited thermal tolerance in fish?: Answers obtained by nuclear magnetic resonance techniques. Resp. Phys. Neuro. 141(3): 243-260. Purdom, C. E. 1993. Genetics and fish breeding. Capman & Hall, London. Refstie, T., Stoss, J., and Donaldson, E. M. 1982. Production of all-female coho salmon (Oncorhynchus kisutch) by diploid gynogenesis using irradiated sperm and cold shock. Aqua. 29: 67-82. Richards, J. G. 2011. Physiological behaviour and biochemical adaptations of intertidal fishes to hypoxia. J. Exp. Biol. 214: 191-199. Rodgers, D. W. and Griffiths, J. S. 1983. Effects of elevated thermal regimes on survival of rainbow trout (Salmo gairdneri). J. Great Lakes Res. 9: 421-424.  66  Rodnick, K. J., Gamperl, A. K., Lizars, K. R., Bennet, M. T., Rausch, R. N., and Keeley, E. R. 2004. Thermal tolerance and metabolic physiology among redband trout populations in south-eastern Oregon. 64(2): 310-335. Sadler, J., Pankkurst, P. M., and King H. R. 2001. High prevalence of skeletal deformity and reduced gill surface area in triploid Atlantic salmon (Salmo salar L.) Aqua. 198: 369-386. Sadler, J., Wells, R. M., Pankhurst, P. M., and Pankhurst, N. W. 2000. Blood oxygen transport, rheology and haematological responses to confinement stress in diploid and triploid Atlantic salmon, Salmo salar. Aqua. 184(3-4): 349-361. Schaefer, J. And Ryan, A. 2006. Developmental plasticity in the thermal tolerance of zebrafish Danio rerio. J Fish Biol. 69: 722-734. Sezaki, K., Watabe, S., Tsukamoto, K., and Hashimoto, K. 1991. Effects of increase in ploidy status on respiratory function of ginbuna, Carassius auratus langsdorfi (Cyprinidae). Comp. Biochem. Physiol. 99(A): 123-127. Silverstein, J. T., Vallejo, R. L., Palti, Y., et al. 2008. Rainbow trout resistance to bacterial cold-water disease is moderately heritable and is not adversely correlated with growth. J. Anim. Sci. 87(3): 860-867. Simon, D. C., Scalet, C. G., and Dillon, J.C. 1993. Field performance if triploid and diploid rainbow trout in South Dakota ponds. North Amer. J. Fish Mgmt. 13: 134-140. Small, S. A., and Randall, D. J. 1989. Effects of triploidy on swimming performance of coho salmon (Oncorhynchus kisutch). Can. J. Fish. Aquat. Sci. 46: 243-245. Speers-Roesch, B., Richards, J. G., Brauner, C. J., Farrell, A. P., Hickey, A. J., Wang, Y. S., and Renshaw, G. M. 2012. Hypoxia tolerance in elasmobranchs. I. Critical oxygen tension as a measure of blood oxygen transport during hypoxia exposure. J. Exp. Biol. 215(1): 93-102. Stillwell, E. J. and Benfey, T. J. 1994. Hemoglobin level, metabolic rate, and swimming performance in triploid brood trout (Salvelinus fontinalis). In: High Performance Fish (MacKinlay, D. D., Ed.). p. 288-293. Vancouver: Fish Physiology Association. Stillwell, E. J. and Benfey, T. J. 1996a. Hemoglobin level, metabolic rate, opercular abduction rate and swimming efficiency in female triploid brook trout (Salvelinus fontinalis). Fish Physiol. Biochem. 15: 377383. Stillwell, E. J. and Benfey, T. J. 1996b. The swimming performance of triploid brook trout (Salvelinus fontalis). Bull Aquac. Assoc. Can. 96(2): 41-43. Stillwell, E. J. And Benfey, T. J. 1997. The critical swimming velocity of diploid and triploid brook trout. J. Fish Biol. 51: 650-653. Strange, K. T., Vokoun, J. C. and Noltie, D. B. 2002. Thermal tolerance and growth differences in orangethroat darter (Etheostoma spectabile) from thermally contrasting adjoining streams. Am. Midl. Nat. 148: 120-128. 67  Streisinger, G., Walker, C., Dower, N., Knauber, D., and Singer, F. 1981. Production of clones of homozygous diploid zebrafish (Brachydanio rerio). Nat. 291: 293-296. Swarup, H. 1958. Production of triploidy on Gasterosteus aculeatus. J. Gen. 56: 129-142. Sullivan, K. M. and Somero, G. N. 1980. Enzyme activities of fish skeletal muscle as influenced by depth of occurrence and habits of feeding and locomotion. Mar. Bio. 60: 91-99. Thorgaard, G. H., Arbogast, D. N., Hendricks, J. D. Pereira, C. B., and Bailey, G. S. 1999. Tumor suppression in triploid trout. Aqua. Tox. 46: 121-126. Thorgaard, G. H., Jazwin, M. E., and Stier, A. R. 1981. Polyploidy induced by heat shock in rainbow trout. Trans. Am. Fish. Soc. 110: 546-550. Thorgaard, G. H. And Gall, G. A. E. 1979. Adult triploids in a rainbow trout family. Gen. 93: 961-973. Thorgaard, G. H., Rabinovitch, P. S., Shen, M. W., Gall, G. A. E., Propp, J., and Utter, F. M. 1982. Triploid rainbow trout identified by flow cytometry. Aqua. 29: 305-309. Valenti, R. F. 1975. Induced polyploidy in Tilapia aurea (Steindachner) by means of temperature shock treatment. J. Fish Biol. 7: 519-528. Van Leeuwen, T. E., Rosenfeld, J. S., and Richards, J. G. 2011. Adaptive trade-offs in juvenile salmonid metabolism associated with habitat partitioning between coho salmon and steelhead trout in coastal streams. J. Anim. Ecol. 80(5): 1012-1023. Virtanen, E., Forsman, L., and Sundby, A. 1990. Triploidy decreases the aerobic swimming capacity of rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 96(A, 1): 117-121. Wagner, E. J., Arndt, R. E., and Brough, M. 2001. Comparative tolerance of four stocks of cutthroat trout to extremes in temperature, salinity, and hypoxia. West Nor. Am. Nat. 61: 434-444. Weithman, A. S. And Haas, M. A. 1984. Effects of dissolved-oxygen depletion on the rainbow trout fishery in lake Taneycomo, Missouri. Trans. Am. Fish. Soc. 113: 109-124. Withler, R. E., Beacham, T. D., Solar, I. I., and Donaldson, E. M. 1995. Freshwater growth, smelting, and marine survival and growth of diploid and triploid coho salmon (Oncorhynchus kisutch). Aqua. 136: 91107. Woods III, L. C. 2001. Domestication and strain evaluation of striped bass (Morone saxatilis). Aqua. 202: 343-350. Woodward, J. J. and Smith, L. S. 1985. Exercise training and the stress response in rainbow trout (Salmo gairdneri). J. Fish Biol. 26: 435-447. Yamamoto, A. and Iida, T. 1994. Oxygen consumption and hypoxic tolerance of triploid rainbow trout. Fish Path. 29: 245-251.  68  Yeager, D. P. and Ultsh, G. R. 1989. Physiological regulation and conformation: a BASIC program for the determination of critical points. Phys. Zool. 62(4): 888-907. Young, P. S. and Cech, Jr., J. J. 1993a. Improved growth, swimming performance, and muscular development in exercise-conditioned young-of-the-year striped bass (Morone saxatilis). Can. J. Fish. Aquat. Sci. 50: 703-707. Young, P. S. and Cech, Jr., J. J. 1993b. Effects of exercise conditioning on the stress responses and recovery in cultured and wild young-of-the-year striped bass (Morone saxatilis). Can. J. Fish. Aquat. Sci. 50: 2094-2099.  69  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073196/manifest

Comment

Related Items