Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biomechanical assessment of adaptive radiation in threespine sticklebacks: (gasterosteus spp.) Law, Tara 1994

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

Item Metadata


831-ubc_1994-0280.pdf [ 1.09MB ]
JSON: 831-1.0087532.json
JSON-LD: 831-1.0087532-ld.json
RDF/XML (Pretty): 831-1.0087532-rdf.xml
RDF/JSON: 831-1.0087532-rdf.json
Turtle: 831-1.0087532-turtle.txt
N-Triples: 831-1.0087532-rdf-ntriples.txt
Original Record: 831-1.0087532-source.json
Full Text

Full Text

BIOMECHANICAL ASSESSMENT OF ADAPTIVE RADIATIONIN THREESPINE STICKLEBACKS (Gasterosteus spp.)byTARA LAWB.Sc., The University of British Columbia, 1989A THESIS SUBMI El ED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Zoology)We accept this thesis as conformingTHE UNIVERSITY OF BRITISH COLUMBIAApril 1994© Tara Law, 1994in presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. 1 further agree that permissionfor extensivecopying of this thesis for scholarly purposes may be granted by thehead of mydepartment or by his or her representatives. it is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)________________Department of /o. ,‘The University of British ColumbiaVancouver, CanadaDate 2 . /7 VDE-6 (2/88)iiABSTRACTThis study compares the morphological characteristics and swimming performancesof a sympatric species pair of threespine stickleback in order to gain an understanding of theprocesses involved in their divergence. The fishes are young (13,000 years old),morphologically distinct, and inhabit different niches in the lake (Paxton Lake, TexadaIsland, British Columbia). Experiments were conducted to compare the steady swimmingand escape fast-start performances of these fishes. I tested predictions of two hypotheses: 1)Selection has acted on the steady swimming ability of limnetics, and therefore I predictedthat limnetics have greater steady swimming performance than benthics. 2) Thespecialization for steady swimming has compromised the fast-start performance of limnetics;consequently I predicted that benthics have greater fast-start performance than limnetics.As predicted, limnetics had greater steady swimming performance than benthics. Themean regression of the logarithm of fatigue time (F.T.) on swimming speed (U, in bodylengths/second) for limnetics [Log (ET.) = 5.24 - 0.46U] was significantly higher (p <0.05)than that for benthics [Log (F.T.) = 3.77 - 0.43U]. The streamlined shape (to reduce drag),larger pectoral muscles (to provide power), and larger pectoral fin areas (to provide thrust)all contribute to the higher performance of limnetics. However, the fast-start performance ofthe two fishes was not significantly different for any linear parameter (pooled mean valuesfor total fast-start: duration 0.049 s, distance 0.034 m, maximum velocity 1.10 m/s, andmaximum acceleration 133.9 mIs2). With large, caudally placed fins limnetics achieved equalfast-start performance to the deeper bodied benthics.The prevailing view of functional morphology in fishes is that adaptations for highperformance in one swimming mode compromise those for high performance in the otherswimming modes. For example, optimal characteristics for fast-starts are thought to trade-offagainst optimal traits for manoeuvering and steady swimming. However, both of theiiisticklebacks studied here were capable of good manoeuverability and fast-start performance.The propulsive systems of these fish aie decoupled; consequently there is no trade-offbetween adaptations for these swimming modes. Limnetics achieved high performance in allthree swimming categories with no compromise between good steady swimming and fast-start ability.The differences in prolonged swimming performance are linked to morphological andhabitat differences. I suggest that selection due to differential resource use has lead todivergence of body form and therefore of steady swimming performance. However, theresults of this study also suggest that selection from predators has lead to maintenance ofhigh performance in fast-start swimming. Predation pressure is probably similar on thesefishes; consequently selection would favour traits which enhance high escape fast-startperformance in both species.ivTABLE OF CONTENTSI.Absact.iiII. List of Tables viIII. List of Figures viiIV. Acknowledgements viiiV. Chapter 1 - General Introduction 1VI. Chapter 2- Steady Swimming PerformanceA. Introduction 5B. Materials and Methods1. Fish Collection and Maintenance 62. Swimming Performance 73. Morphological Measurements 84. Drag Tests 105. Statistical Tests 12C. Results1. Swimming Performance 132. Morphological Measurements 143. Drag Tests 14D. Discussion 15VII. Chapter 3 - Escape Fast-Start PerformanceA. Introduction 27B. Materials and Methods1. Fish Collection and Maintenance 282. Fast-Start Performance 283. Film Analysis 294. Morphological Measurements 315. Statistical Tests 32C. Results1. Fast-Start Performance 322. Kinematics 333. Morphological Measurements 34VD. Discussion .35VIII. Chapter 4- General Discussion and Conclusions 46IX. Bibliography 50viLIST OF TABLESTable I - Swimming Endurance at Absolute Speeds .20Table II - Morphological Measurements: Morphometrics 21Table III - Escape Fast-Start Performance 40Table IV - Morphological Measurements: Fin Areas and 41Positions of Maximum Mass and Added MassviiLIST OF FIGURESFigure 1 - The Endurance of Sticklebacks .22Figure 2 23Part A - Drawings of Paxton Lake Limnetic and BenthicThreespine Sticklebacks.Part B - Body Depth Along the Length of theStickleback.Figure 3 - The Drag of Sticklebacks 24Figure 4 - The Coefficient of Drag for Sticklebacks 25Figure 5 - The Endurance of a Variety of Small Fishes 26Figure 6- The Escape Fast-Start Performance 42of Stickiebacks.Figure 7 - Use of Three Dimensional Space During Escape 43Fast-Starts of Sticklebacks.Figure 8 - Kinematics of Escape Fast-Starts 44of Sticklebacks.Figure 9 45Part A - Drawings of Paxton Lake Sticklebacks.Part B - The Distributions of the Fish’s Mass andAdded Mass Along the Length of the BodyviiiACKNOWLEDGEMENTSThis study could not have been completed without the advice and support of manypeople. I would like to thank my family for providing moral and financial support. I thankmy supervisor, Robert Blake, for all that he has taught me over these years. I also thank themembers of my committee for support, advice, and pep talks: J. M. Gosline, J. D. McPhail,and D. Schiuter. I would especially like to thank D. Schiuter for all his help in the collectionof the animals, and the design of the experiments. I thank: T. Day, T. Hatfield, M. Linden,and L. Nagel for their help with animal collection and advice on animal care, and L. Nagelfor providing the drawings in Figs. 2 and 9. Thanks also go to W. Neill for providing theenvironmental chamber space necessary for this project, as well as identifying the calanoidcopepods to species and helping with the culturing of these animals. I thank W. Milsom forproviding access to computing facilities. An enormous number of people provided usefuldiscussions on this project, in particular I would like to thank: C. Brauner, P. Domenici, M.Eymann, D. Ghan, D. Harper, S. Inglis, M. Kasapi, F. Kroon, C. Ouimet, and J. Harris. Ithank R. Andrews for being both my harshest critic and my greatest supporter.I would like to dedicate this thesis to the staff of the Department of Zoology at theUniversity of British Columbia. Without their help none of us would be able to carry on ourscientific investigations.1CHAPTER 1General IntroductionAlthough understanding the mechanisms involved in speciation is a central problemin evolutionary biology, the processes thought to be responsible for adaptive radiation are notvery well understood (Schiuter 1993). Inferential studies can correctly define selectiveprocesses only if the investigation includes a knowledge of both the organism’s ecology(Clarke 1978) and a particular trait’s functional significance (Bock 1980). In addition, theprocesses involved can only be inferred from the study of incipient or recent species (Mayr1978). Therefore, situations where speciation has occurred recently, and where the ecologyof the animals and functional significance of their characteristics are known are ideal forinvestigations of the mechanisms involved in the divergence of species.The majority of studies investigating the causes of diversity among fish haveconcentrated on trophic morphology and its correlation with food type (Bentzen and McPhail1984; Ehlinger and Wilson 1988; Larson 1976; Lavin and McPhail 1985, 1986; Liem 1974;McPhail 1984, 1992; Schiuter 1993, 1994; Schluter and McPhail 1992; Wainwright 1988;Werner 1977). These studies have shown that morphology and resource use are correlated(Liem 1974; Larson 1976; Lavin and McPhail 1985; Schiuter and McPhail 1992; Wainwright1988), that morphology and resource use efficiency are correlated (Bentzen and McPhail1984; Larson 1976; Schiuter 1993, 1994; Schluter and McPhail 1992; Werner 1977), and thatresource use efficiency and growth (a fitness component) are correlated (Schiuter 1994).Factors other than the manipulation and ingestion of prey can affect the fish’s fitness. Theabilities to find prey and escape predators also contribute to the fitness of an animal (Videler1993). Success in these activities is largely determined by the fish’s swimming abilities2(Webb 1984a). The contribution of various features of the fish’s body to its swimmingperformance can be determined using a biomechanical approach (Blake 1983; Webb 1984a;Weihs and Webb 1984).The prevailing view of functional morphology in fishes is that there is mutualexclusion of optimum body characteristics for three different modes of swimming: transient(burst swimming), steady swimming (cruising), and manoeuvering (Webb 1984a).Characteristics that enhance performance in a particular swimming mode are thought tocompromise performance in the others. The design elements resulting in the best transient-swimming performance, such as large body depth and a large percentage of the body mass asanaerobic muscle, are costly during steady swimming, whereas characteristics such asstreamlining, and a high proportion of the body’s muscle mass made up of aerobic muscle areimportant for steady swimming (Blake 1983; Lindsey 1978; Webb 1975). In addition, thelarge surface areas of the body and fins of manoeuvering specialists increase drag, whichhinders steady swimming performance, and their shortened body decreases the availablespace for the muscle mass needed for burst swimming (Webb 1984a). Therefore, fish havebeen classified as being steady swimming, fast-start or manoeuvering specialists, orgeneralists (Webb 1984a). Generalists have good performance at more than one swimmingmode but superior performance in none.To test the applicability of this model, the swimming performances of a variety offishes with differing body designs and lifestyles have been compared (Domenici and Blake1991; Harper and Blake 1990; Kasapi et a!. 1993). In general, these studies have comparedunrelated or distantly related fishes. This study is a comparison of the body characteristicsand swimming performances of closely related fishes. A sympatric species pair of threespinestickleback (Gasterosteus spp.) from Paxton Lake, Texada Island, British Columbia, wasstudied in order to gain an understanding of the processes involved in their divergence.3A number of studies have investigated the diversity of morphologies of Gasterosteusspp. in the southwestern region of British Columbia (Baumgartner et al. 1988; Bentzen andMcPhail 1984; Bentzen et a!. 1984; Larson 1976; Lavin and McPhail 1985, 1986; McPhail1984, 1992; Schiuter 1993; Schiuter and McPhail 1992). This species complex provides anopportunity to study recent divergence because all the species are endemic to coastal watersystems that did not form until 13,000 years ago (Mathews et al. 1970).These studies have shown morphological differences between anadromous andstream sticklebacks (Taylor and McPhail 1986), sticklebacks from different lakes (Lavin andMcPhail 1985, 1986; Schluter and McPhail 1992), and sticklebacks within the same lake(Baumgartner et al. 1988; Bentzen and McPhail 1984; Bentzen et a!. 1984; Larson 1976;McPhail 1984, 1992; Schiuter 1993; Schluter and McPhail 1992). Morphological differencesare particularly pronounced between species occurring in the same lake (Schluter andMcPhail 1992). The species pair found in Paxton Lake, British Columbia, was chosen for thisstudy as an example of closely related fishes with pronounced differences in morphology.Evidence that these are a species pair and not a single polymorphic species includethe observations that morphological differences are retained when fish are bred in thelaboratory (Larson 1976; McPhail 1992), that hybrids are morphologically intermediate(McPhail 1992; Schluter 1993), and that the morphological and allelic frequencies betweenthe forms have been stable for over 20 generations (McPhail 1992). The differences betweenthe sympatric fishes are therefore thought to be genetically based and there appears to belittle gene flow between the two fishes (McPhail 1992; Schluter 1993).Previous studies of this sympatric species pair have investigated spatial and trophicsegregation (Larson 1976), morphological and genetic differences (McPhail 1992), trophicmorphology and feeding efficiency (Schluter 1993), and feeding efficiency and growth(Schiuter 1994). The two fishes occupy different regions of the lake and have been referredto as benthic and limnetic, according to their primary habitat. Benthics live near the bottom4in weedy areas, and feed primarily on large insect larvae (Larson 1976; Schiuter 1993;Schluter and McPhail 1992). Limnetics form large schools in the pelagic region of the lake,and feed primarily on zooplankton (Larson 1976; Schluter 1993; Schluter and McPhail1992). The benthic region is a spatially complex habitat where precise manoeuvering isimportant, whereas the pelagic region is an open habitat where steady swimming isimportantHere I have considered two main functions of fish swimming: cruising to locate prey(Chapter 2), and fast-starts to escape predators (Chapter 3). I tested predictions of twohypotheses: 1) Selection has acted on the steady swimming ability of limnetics and thereforeI predict that limnetics will have greater steady swimming performance than benthics. 2) Thespecialization for steady swimming has compromised the fast-start performance of limneticsand therefore I predict that benthics will have greater fast-start performance than limnetics.5CHAPTER 2Steady Swimming PerformanceINTRODUCTIONThe ability to swim quickly and steadily for a long time is important for fish that arepelagic (Wardle 1977). Pelagic fish feed on prey that usually have patchy distributions in anopen environment, and routinely swim for prolonged periods, searching for these patches(Horwood and Cushing 1977). Pelagic fish are specialists for steady swimming and havesuperior performance to fish that only swim occasionally (Webb 1984a). The body design ofthese pelagic fish is considered to be optimal for steady swimming and includes suchcharacteristics as streamlining to reduce drag, stiffening of the body to minimize movementof the head, and a high percentage of aerobic fibers in the body’s muscle mass (Blake 1983;Lindsey 1978; Webb 1975). Most pelagic fish swim by undulations of the body and caudal‘fin; consequently previous studies of steady swimming performance have concentrated onsuch fish (e.g. Beamish 1984; Brett 1965; Taylor and McPhail 1985; Tsukamoto et a!. 1975;Tsukuyki and Williscroft 1977).Paired fin swimming is advantageous for manoeuverability at low speeds butdisadvantageous for high speed, steady swimming (Blake 1983; Weihs and Webb 1984).Butterfly fish and angelfish have characteristics thought to enhance manoeuverability, but thelarge surface areas of the body and fins impose a high price in increased drag for steadyswimming (Webb 1984a). In addition, these fish tend to have a shortened body whichdecreases the available space for the aerobic muscle mass needed for high speed cruising. A6few studies have investigated the steady swimming ability of fish which swim using pectoralfins, but these have concentrated on low speed swimming (Dom et al. 1979; Stahlberg andPeckman 1987; Taylor and McPhail 1986; Whoriskey and Wootton 1987).This part of my investigation compares the steady swimming performances of the twospecies of threespine stickleback (Gasterosteus spp.) that live sympathcally in Paxton Lake.Sticklebacks use paired pectoral fins for routine locomotion, mostly by rowing (Taylor andMcPhail 1986). Therefore, these fish should have good manoeuverability but poor steadyswimming performance. However, many sticklebacks are anadromous, migrating greatdistances to breeding grounds (Wootton 1976, 1984), and others (e.g. limnetics) must swimfor prolonged periods while searching for patches of planktonic prey. Limnetics live in anopen water habitat and are more fusiform than benthics; therefore my hypothesis is thatselection has acted to increase the steady swimming performance of limnetics but notbenthics. Based on this hypothesis I have predicted that limnetics have greater endurance andexperience less drag than benthics.MATERIALS AND METHODSFish Collection and MaintenanceFish were collected from Paxton Lake, Texada Island, British Columbia (17-ha lake,490 42’ 30” N, 1240 31’ 30” W) using minnow traps and dip nets. Only juvenile and adult fishwere used (i.e. no young of the year). Fish were transported to the lab where they werevisually separated into the two species based on size, shape, and colour (Fig. 2 A) and placedinto holding tanks.7Four 163 litre holding tanks were each divided into three equal sections withlimestone gravel bottom cover (to maintain pH 7.8 - 8.2, similar to Paxton Lake water;Larson 1972), plants, air stone, mini water filter, and 14 Watt full spectrum light on a naturaldawn to dusk light schedule (6 am to 10 pm). The tanks were kept in an environmentchamber to maintain a constant temperature of 15° ± 1°C. Holding tank water consisted offresh, dechlorinated water and 3 % salt. Two fish of each species were placed into eachsection of the four holding tanks and held for one week prior to testing. During the initialweek, sick or dead fish were removed, and replaced with healthy fish. All fish wereindividually identified by colour patterns or slight size differences and assigned a number.Fish were fed live brine shrimp nauplii, live calanoid copepods (Diaptomus oregonensis,from Paxton Lake), and either previously frozen adult brine shrimp or blood worms, once perday during the initial week.Swimming PerformanceFeeding ceased one day prior to testing to ensure a post-absorptive state (Niimi andBeamish 1974). The experiments were carried out at 15° ± 1°C in a 12 liter flow tank(working section 10 cm x 60 cm x 10 cm) with an adjustable speed controlled propellermotor (Cole-Parmer model 4555-3). Water velocity was monitored with a small velocityprobe and meter (Nixon Streamflo 422, ± 1 cmfs). Water (pretreated for a minimum of threedays with limestone, 3 %o salt, and aeration) was replaced daily to ensure similar conditionsas in the holding tanks. The flow was smoothed by a straw grid at the upstream end of theworking section. A downstream screen prevented the fish from being swept into thepropeller.For each trial the water speed was set, then the selected fish was added. The observersat close to the tank with a stopwatch and measured the time to fatigue, or up to an arbitrarily8defined cutoff time of sixty minutes, at which time the fish was removed and returned to theholding tanks. Fatigue time was defined as the time until a fish ceased swimming and fellback against the downstream screen.Every fish (benthic n = 10, limnetic n = 9) was tested once each at five speeds: 27.0,36.0, 40.5,44.5, and 53.5 cm/s, and allowed 48 hours rest between trials. The regression offatigue time on swimming speed was calculated for every fish and the mean slopes andintercepts of the two species were compared. Trials where the fish were still swimming at thecut off time of 60 minutes were included as this gave conservative estimates of the regressionslopes.Hoar and Randall (1978) define prolonged swimming as a range of speeds betweenburst and sustained swimming, where the swimming period lasts between 15 seconds and200 minutes and if maintained will end in fatigue. For both limnetics and benthics the speedswere chosen to allow a range of speeds to be tested that were within this definition.Morphological MeasurementsFish were killed by an overdose of MS222, dried externally with a paper towel, andweighed with an analytical balance (Mettler Scale model PK300 ± 0.001 g). Standardlengths (L = tip of rostrum to end of caudal peduncle) were measured with vernier calipers (±0.005 cm).The centre of mass (CM) of a fish was determined by suspending the fish from themouth and marking the vertical line of gravity and repeating this procedure for suspensionfrom the cloaca. The CM was the point where the two lines cross. The CM as a proportion ofthe length was the distance along the straight body from the tip of the rostrum to the CM, andthe distance from the fishes dorsal surface to the CM was the CM as a proportion of bodydepth. The body depth at the CM was measured with vernier calipers (±0.005 cm).9To determine the position of maximum body depth, depth was measured at intervalsof 10% of total length (0.1L) from the tip of rostrum to the trailing edge of caudal fin, whilethe fish was submerged in water in a petri dish to ensure full extension of the fins.Pectoral fin areas were measured by painting the fins with methylene blue andblotting the splayed fins onto paper. The resulting fin prints were cut Out and weighed(Mettler model M3 ± 0.001 mg). The mass was then converted to area using a regressionobtained by weighing known areas of the same paper. Surface area (Sw) was measured bywrapping the body of the fish with plastic wrap (Glad Cling Wrap), weighing the resultantplastic piece, and converting the weight to area (Varley 1989), as done for the fin areas. Inthe calculation of fin areas as percentages of total surface area, all surfaces of the fins wereincluded, and the fins were considered to have zero thickness.The right pectoral muscles, both abductor and adductor, easily distinguished fromsurrounding muscle by their red colour, were dissected out of the fish and weighed together.There is a significant difference in the size of the two species (McPhail 1992), andbody proportions are size dependent; therefore, morphological measurements were correctedfor size using a regression method for adjusting treatment means (Steel and Tome 1980).This adjustment uses the linear regressions of the log of each variable on the log of standardlength. The adjustment equation is:A —YjJk=Yjk-Bjk(L-5.O) (1)where is the adjusted measurement of the jth variable for the 1th individual in the kthpopulation, jk is the sample mean of the jth variable in the kth population, BJk is thecoefficient of allometry for the jth variable on standard length for the kth population, and Lik -5.0 is the standard length of the jth individual minus the standardizing size. In Gasterosteus itis conventional to use 5.0 cm as a standardizing size for comparison (Hagen and Gilbertson1973; Lavin and McPhail 1985; McPhail 1983). In addition, 5.0 cm is approximately half-10way between the mean benthic length (5.4 cm) and the mean limnetic length (4.5 cm).Therefore, the adjustment affects the two species an equal amount.Submerged weight was measured with the fish first squeezed under the water toremove air bubbles, and then weighed with an electronic balance equipped with an adaptorwhich suspended the fish in the fluid (Mettler PK300 with manufacturer’s submerged weightadapter ± 0.001 g). The density of the fish (p’) was calculated usingWs=Wa(1P/P’) (2)(Gal and Blake 1987) where W is the submerged weight, Wa the weight of the fish in air,and p is the water density. Fish density was calculated as:= _p_ (3)WaWsThe sex of the fish and the presence of any external or internal parasites wererecorded. Fish that were obviously gravid or parasitized were not used.Drag TestsTeminal velocity estimates were obtained by dropping dead fish down a glasscolumn (30 x 32 x 120 cm) filled with water at room temperature, 20°C (water density =0.998 glcm2,kinematic viscosity 0.01004 cm2/s, Vogel 1981). The rigid-body assumptionfor dead-drag drop tank tests does not apply to fish that swim using body undulationsbecause of additional pressure and friction drag associated with repeated bending, but fishthat swim using paired fins (e.g. sticklebacks) hold their body rigid. Therefore, dead-dragmeasurements can be assumed to give a good indication of true swimming drag (Blake1983).To obtain a relationship between drag and velocity for every fish, each fish wasdropped at a number of different weights. Submerged weights of the fish were increased by11inserting small lead balls through the mouth without extending the pharyngeal pouch.Submerged weights were measured as described above. For each fish at a given weight, threeseparate trials were conducted and the results averaged.To achieve a vertical descent, a dart flight (wetted surface area = 20.11 cm2) wasattached to the fish on the end of a shaft made from thin wire (diameter 0.07 cm) which wasinserted through the caudal peduncle, parallel to the spinal column. The dart flight wasattached 10 cm from the trailing edge of the fish’s caudal fin to reduce interference. Inaddition, the pectoral, dorsal, and anal fins were removed, flush with the body, to eliminatefin flutter, and to ensure a vertical descent (Blake 1983; Varley 1989).Two lines on the outside of the column (55.5 cm apart) marked start and end fortiming. The column was filmed with a video camera (Panasonic CCTV camera model WVBL600) connected to a 60 frame/second VHS recorder (J.V.C. video cassette recorder modelB2-3200U). A 1/100 second counter (Panasonic model WJ-810) fed directly into the recorderas part of the image. The time used in the calculation of terminal velocity was the differencebetween the counter time when the dart flight passed the top mark line and the counter timewhen the flight crossed the bottom mark line. Terminal velocity was always reached prior tothe dart flight crossing the start line.A drag-terminal velocity calibration curve for the dart flight and shaft was alsodetermined. Submerged weights were increased by rolling thin sheets of lead around theleading end of the flight shaft. The regression of drag on terminal velocity for the fish wascalculated by subtracting the curve for the flight from the curve for the flight attached to thefish.At terminal velocity, the submerged weight of the fish is equal to the drag force, D:W = D = O.5PSWU2CD (4)(Gal and Blake 1987) where U is the velocity, S is wetted surface area, and CD is the dragcoefficient. Therefore, the drag coefficient can be calculated as:12CDt 2D (5)pSU2Reynolds number, Re, can be calculated from:Re=LII (6)1)(Vogel 1981) where L is length, U is velocity and 1.) is the kinematic viscosity of the fluid.To remove the effect of variation in terminal velocities between fish, the drag at 10velocities (.250, .270, .300, .360, .405, .445, .490, .535, and .600 m/s) was calculated foreach fish using the regression equation for drag on terminal velocity for that particular fish.These drag data were then corrected for size using the adjustment technique described above.Drag was then compared between the fish at each velocity and the mean regression for dragon velocity calculated for each species. Coefficients of drag were calculated using the sizeadjusted drag values. Therefore, direct comparisons of coefficients of drag at each Reynoldsnumber were possible.Statistical TestsAll data were compared using t-tests except the position of maximum body depthalong the length of the fish which was compared using a Mann-Whitney U test. Percentageand proportion data were arcsin-square root transformed before testing with t-tests. All nullhypotheses were rejected at p <0.05 and all means are presented ± 2 S.E.13RESULTSSwimming PerformanceBoth fishes primarily used pectoral fin locomotion, although occasional caudal finthrusts to maintain position at higher speeds were observed. Limnetics collapsed their dorsaland anal fins, raising them only during quick bursts, whereas benthics had raised finscontinuously.There was no difference in swimming performance between males and femaleswithin a species (p > 0.05), so the data for a species were pooled.All limnetics (n = 9), and all but one benthic (total n = 10), were able to swim for theentire 60 minute thai period at 27 cm/s. Limnetics swam for longer at all other speeds thanthe benthics (Table I). The convention is to show endurance data as the logarithm of fatiguetime compared to speed in body lengths/second, L/s (Brett 1965). Using this convention,limnetics showed superior performance to benthics (Fig. 1). The regressions of fatigue timeon swimming speed for all fish were significant (p <0.05), and the r2 values ranged from0.87 - 0.99 with the mean for both fishes at 0.95. The mean intercept of the regression forlimnetics (5.24 ± 1.18) was significantly (p < 0.05) higher than that for benthics (3.77 ±0.44). Although the slopes were not significantly different (benthic -0.43 ± 0.06, limnetic= -0.46 ± 0.12), the limnetic slope was slightly steeper than the benthic slope which causedconvergence of the regression lines at higher speeds. The benthic mean regression equationwasLog (F.T.) = 3.77 - 0.43U (7)and the limnetic mean regression wasLog (F.T.) = 5.24 - 0.46U. (8)14Where F.T. is fatigue time (minutes) and U is swimming speed (body lengths/second). Therewas no allometric effect on endurance for benthics or limnetics (p > 0.05).Morphological MeasurementsAlthough benthics are larger, heavier, and deeper bodied than limnetics, the fishwhole body density and the pectoral muscle mass for the two species were not significantlydifferent (Table II). The pectoral fin area as a percentage of the total surface area wassignificantly larger (p <0.05) for limnetics 12.2% ± 0.03%) than benthics (7.8% ± 0.02%).Pectoral fin muscle mass as a percentage of the total body mass was also greater for limnetics(3.0% ± 0.005%) than benthics (1.3% ± 0.004%). The position of maximum depth for thetwo fish also differed. Benthics were deepest at 0.3L, and limnetics at 0.6L, the region wherethe dorsal and anal fins both contribute to depth (Fig. 2).Size corrected measurements showed large differences between the two fishes.Although benthics have greater mass and surface area than limnetics, their pectoral musclemass and pectoral fin area were significantly smaller. There was no relationship (p > 0.05)between density and size for either of these fishes.Drag TestsThe size corrected drag was significantly higher (p <0.05) for benthics than limneticsat all speeds (Fig. 3). Drag coefficients for benthics were also greater than those for limneticsat all Reynolds numbers (Fig. 4).15DISCUSSIONAs predicted, limnetics had superior steady swimming performance compared tobenthics. The endurance of limnetics ranged from 1.5 to 7.5 times that of benthics. Althoughlimnetics had greater performance at all speeds, the difference between the fishes was less athigher speeds.The steady swimming ability of limnetics relates to the lifestyle of these fish.Limnetics are open water fish that feed primarily on calanoid copepod zooplankton (Schiuter1993). Zooplankton populations are typically found in patches throughout lakes (Richersonet al. 1977; Steele 1977). Thus, limnetics must travel relatively large distances in search ofprey and therefore must have good steady swimming ability.The results of the morphological measurements indicate that there is significantdifferentiation with respect to morphology. This differentiation was characterized by bothallometric and shape variation. Benthics are larger, heavier, more robust fish, with largersurface areas. Limnetics are smaller, more fusiform fish, with smaller surface areas. Sizecorrected measurements showed that the pectoral fin area and pectoral muscle mass of thesmaller fish (limnetic) were 2 and 3 times greater, respectively, than for the larger fish(benthic).The morphological characteristics considered to be advantageous for steadyswimming include streamlining to minimize drag, stiffening of the body to minimize recoil,and a high percentage of the body’s mass made up of aerobic muscle (Webb 1984a). Thereare differences between the two fishes for all of these characteristics.Drop tank tests showed that the streamlined shape of limnetics results in lower drag.The size corrected drag was 2 times less for the limnetic fish than for the benthic fish. In16addition to having a streamlined body to reduce drag, limnetics collapse their fins, therebyreducing fin flutter and surface area. The position of maximum body depth for limnetics isfar back on the body (Table II and Fig. 2), and the depth is mostly due to the dorsal and analfin heights. Thus, by collapsing the fins limnetics reduce their maximum body depth, therebyreducing drag. Benthics, however, do not gain the advantage of decreasing maximum bodydepth by collapsing their fms because the position of maximum depth for these fish is dueentirely to body depth. Therefore, collapsing the fins would not significantly reduce the dragexperienced by benthics.Limnetics have a pelvic girdle and lateral plates which may increase stiffness forprolonged swimming. The pelvic girdle is absent in the benthics and the number of lateralplates is greatly reduced (McPhail 1992). Although pectoral fin swimmers do not bend theirbodies while swimming, if muscular effort is required to maintain a rigid body, the greaterbody stiffness of limnetics would enhance steady swimming by reducing the muscularenergy needed to hold the body rigid. Thus the lower body stiffness of benthics may reducetheir steady swimming performance.The pectoral muscles of the sticklebacks make up a relatively large percentage oftheir total body mass, an advantage for endurance during prolonged swimming. Limneticshave larger pectoral muscles, and a greater percentage of their body mass as pectoral musclemass than benthics. The relative amount of aerobic muscle in the body is an importantcharacteristic for swimming performance (Webb 1984a). The pectoral muscles dissectedfrom the sticklebacks were red in colour and are likely composed of aerobic muscle fibers.Boddeke et al. (1959) found that the pectoral muscles of the River Bullhead, a labriformswimmer, were entirely made up of red aerobic fibres. During the routine, steady swimmingof the sticklebacks, the pectoral muscles were used almost exclusively. Therefore, thisactivity should be aerobic. However, Luiker and Stevens (1992) found white fibers mixedwith the red fibers in the pectoral muscles of the sunfish (Lepomis gibbosus). The prolonged17swimming of sticklebacks may be entirely aerobic, or partially anaerobic. A completehistochemical analysis of the pectoral muscles of these fishes is necessary to determine theextent of the aerobic contribution to prolonged swimming.Wardle (1977) considered fast sustained cruising only an advantage for pelagicfeeders. Such steady swimming ability is a significant determinate of the feeding success ofthese fish (Videler 1993). The streamlined body of the limnetic reduces the drag experienced,and the larger pectoral fins and pectoral muscles increase the thrust per stroke, therebyenhancing steady swimming performance. Benthics, however, live in a complex vegetatedhabitat where they feed by hovering above a prey item (e.g. insect larvae) on, or in, thesediment, and swim slowly closer to suck in a mouthful of substrate and prey (Anker 1978).Finding prey requires careful examination of the substrate rather than searching large areasby swimming. Therefore, superior steady swimming performance is not necessary for thefeeding success of benthics.There are difficulties in comparing the swimming performance results of this study toothers because the majority of researchers have determined critical speed, rather thanendurance at a range of speeds. Farlinger and Beamish (1977) have shown that the methodsused to determine critical speed, in particular the time increments for increasing speed,greatly affect maximum endurance. At the end of step tests, fish swim rapidly and probablyanaerobically. Whereas, during endurance tests fish swim relatively slowly and mostprobably aerobically (Kolok 1992). Step tests pre-exhaust fish to anaerobic levels; therefore,endurance tests provide a more accurate measure of a fish’s ability to swim for a length oftime at a certain speed (Kolok 1992). Step-tests to determine critical speeds involve priorpartial fatigue of the fish, before the test velocity is reached. Consequently, these tests tend togive underestimates of the endurance performance of fish. In addition, endurance trials at avariety of speeds allow for comparisons of different fishes at several speeds, as well as anindication of the differences within one fish type at a range of speeds. The use in this study18of endurance trials at a variety of speeds for each fish showed that limnetics had greaterendurance than benthics over the entire range of prolonged swimming speeds (Fig. 1).Figure 5 shows the steady swimming performance of a variety of fishes. All the fishesshown here are between three and ten cm in length. Thus, the confounding factors ofallometric effects are reduced. Both fishes which use body/caudal fin undulations, and fisheswhich use pectoral fins for locomotion are shown. Since the two modes of swimming aredifferent in movement, muscle use, and drag augmentation, it is difficult to interpret theexact meaning of comparative results. Comparisons with other pectoral fin swimmers isperhaps more appropriate. In general, benthic steady swimming performance could beconsidered average to poor, and limnetic performance considered superior to other fishes,both pectoral and caudal fin swimmers.Several researchers have investigated the steady swimming performance ofGasterosteus spp. from a variety of different locales (Stahlberg and Peckman 1987; Taylorand McPhail 1986; Whoriskey and Wootton 1987). Taylor and McPhail (1986) studied twodifferent morph types of threespine sticklebacks and found that the streamlined anadromousfish had longer fatigue times than the more robust stream fish. Although their results were inthe same direction as the results of this study, the swimming performance of the limneticswas greater than that for the anadromous sticklebacks (Fig. 5). The performance of streamsticklebacks studied by Stahlberg and Peckman (1987) was lower than that of benthics (Fig.5). However, the stream sticklebacks studied by Whoriskey and Wootton (1987) had anintermediate performance between that for limnetics and benthics (Fig. 5). In general, itappears that the limnetics are at the extreme in the stickleback gradation of morphologies forsuperiority in steady swimming performance and that the benthics are towards the extremefor poor performance.The superior prolonged swimming performance of limnetics compared to benthicscan be explained by their morphological differences. There was a 1.5 to 7.5 times difference19in swimming endurance, and, although probably not directly additive, the differences in drag(limnetics 2 X less), pectoral muscle mass (limnetics 3 X more), and pectoral fin area(limnetics 2 X greater) more than account for the greater performance of limnetics. Thestreamlined shape reduces drag, the large pectoral muscles increase the power available forthrust, and the large pectoral fins increase the area generating thrust thereby increasing theperformance of limnetics. These results support my hypothesis that divergent selection hasled to a high steady swimming performance of limnetics compared to benthics.TABLE I - Swimming Endurance at Absolute Speeds 20Swimming Endurance in Minutes (±2 S.E.Speed (cmls’ Benthic Limnetic36.0 12.58 (11.8) 44.46 (15.6) *40.5 3.15 (1.9) 16.45 (12.4) *44.5 1.73 (1.1) 13.16 (1.4) *53.5 0.82 (0.2) 1.17 (0.5) ** significantly different at p <0.05Benthic n = 10 Limnetic n =9TABLETI-MorphologicalMeasurements:Morphometrics(±2S.E.)ParameterBenthicLimneticMass(g)1.904(0.34)0.991(0.16)*Length(cm)5.41(0.22)4.51(0.16)*Density(glcm3)1.063(0.011)1.072(0.021)PectoralFinArea(cm2)0.42(0.06)0.32(0.03)*TotalSurfaceArea(cm2)21.38(2.9)10.41(0.84)*PectoralMuscleMass(g)0.019(0.004)0.016(0.003)DepthatCentreofMass(cm)1.09(0.04)0.88(0.06)*PositionofCentreof MassAlongDepth0.56D(0.04)0.55D(0.06)PositionofCentreof MassAlongLength0.44L(0.02)0.45L(0.04)PositionofMaximumDepthAlongLength0.3L0.6L*SizeCorrectedMass(g)1.758(0.037)1.006(0.005)*SizeCorrectedPectoralFinArea(cm2)0.26(0.02)0.41(0.03)*SizeCorrectedTotalSurfaceArea(cm2)20.87(0.08)10.51(0.04)*SizeCorrectedPectoralMuscleMass(g)0.010(0.003)0.034(0.006)**significantlydifferentatp<0.05Benthicsn=10Limneticsn=10IIIII100_D•D•••r-/)..E100..E.,- H. .tJ.—0Do0.1I•III•I468101214SwimmingSpeed(bodylengths/second)Figure1-Theenduranceofsticklebacks.Shownarethebenthic(o)meanregressionLog(F.T.) =3.77-.43Uandthelimnetic(•)meanregressionLog(F.T.)=5.24-.46U,whereF.T.=fatiguetime(minutes)andU=swimmingspeed(bodylengths/second).N)23AB1. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Proportion of LengthFigure 2. Part A - Drawings of Paxton Lake limnetic and benthicthreespine stickleback (modified from Schiuter 1993).Part B - Body depth along the length of the stickleback. From therostral tip (O.OL) to the trailing edge of the caudal fm (1.OL), bodydepth is shown for benthics ( ) and limnetics (- — —), todetermine the position of maximum body depth.Benthic••••0.005-0.004--bi0.003-r0.002-0•0.001—-I••I•II•I0.250.300.350.400.450.500.550.60Velocity(mis)Figure3-Thedragofsticklebacks.Shownaretheregressionsfor benthics()D=O.113V’59andlimnetic(——)D=O.006V1.54overthevelocityrangeusedforendurancetrials,whereD=drag(Newtons)andV=velocity(m/s).0.04.0•1•10.035c0.03000.0250.020..___0-ci)..——0.015—I•I•I••II12500150001750020000225002500027500ReynoldsNumberFigure4-Thecoefficientof dragfor sticklebacks. Shownarethe benthicmeanregression()Cd=Reandthelimneticmeanregression(——)Cd=1.42Re,whereCd=coefficientof dragandRe=Reynoldsnumber.1%.)(7’II—-IICl,..QSwimmingSpeed(bodylengths/second)Figure5-Theenduranceofavarietyofsmallfishes.Allfishesarebetween3and10cminlength.Solidsymbolsrepresentfishwhichswimusingbody/caudalfmundulationsandhollowsymbolsrepresent fishwhichswimusingpectoralfms.Shownarethelimneticmeanregression(-),thebenthicmeanregression()streamsticklebacks(v)andanadromoussticklebacks(0)(TaylorandMcPhail1986),streamsticklebacks()(WhoriskeyandWootton1987),streamsticklebacks(c)(StahlbergandPeckman1987),pumpkinseed, Lepomisgibbosus(0)(BrettandSutherland1964), blacksmith, Chromispunctipinnus(C>),(Dorneta!.1979),cohosalmon, Oncorhynchus kisutch,fromcoastal streams(A),andfrominteriorstreams(v)(TaylorandMcPhail1985),goldfish, Carassiuscarassius (•)(Tsukamotoeta!.1975), pacificsardine,Sardinopssagax(•)(Beamish1984),andrainbowtrout,Oncorhynchus gairdneri(•)(TsuyukiandWiffiscroft1977).27CHAPTER 3Escape Fast-Start PerformanceINTRODUCTIONFast-starts are used by many fish to catch evasive prey or to escape from predators(Harper and Blake 1988; Rand and Lauder 1981; Webb 1984b, Webb and Skadsen 1980).Success in these behaviours, particularly predator evasion, has a direct impact on fitness(Videler 1993). Therefore, selection for characteristics which enhance fast-start performanceare under intense selective pressure in fishes under high predation risk (Domenici and Blake1991; Frith and Blake 1991; Harper and Blake 1990; Kasapi et al. 1993; Webb 1975; Weihs1973). Investigation of the morphology and performance of fish during escape fast-starts cangive an indication of the intensity of predation pressure.The ability of fish to escape predators may depend upon linear performance, such asvelocity and acceleration (Domenici and Blake 1993; Harper and Blake 1988, 1990; Vinyard1982; Webb 1976; Weihs and Webb 1984), accurate timing (Eaton and Hackett 1984), andturning ability (Howland 1974; Webb 1982). Adaptations for effective fast-start performanceare thought to include a large proportion of white muscle relative to red, large caudal fin andbody depth for producing thrust, and high body flexibility (Blake 1983; Webb 1984a; Weihs1973).Most investigators have concentrated on the fast-start performance of fishes thatswim using body/caudal fin undulations (Dubois et al. 1976; Eaton et al. 1977; Frith 1990:Harper and Blake 1990; Webb 1976, 1978). Specialization for median or paired finswimming has been hypothesized to impair burst swimming performance (Webb 1984a).However, high fast-start performances by median fin swimmers has been shown in the28knifefish, Xenomystus nigri (Kasapi et a!. 1993), and by paired fin swimmers, in theangelfish, Pterophyllum eimekei (Domenici and Blake 1991). These fishes are laterallycompressed and have large body depth and high flexibility which enables them to achievehigh fast-start performance. Sticklebacks are more fusiform in shape (Fig. 2 A) and thus areexpected to have poor fast-start performance compared to angelfish or knifefish. Limneticshave a streamlined body which reduces drag and aids steady swimming (Chapt. 2), butshould lower thrust production during fast-starts (Blake 1983; Webb 1984a; Weihs 1973).Benthics are larger, deeper bodied fish (Chapt. 2), and thus could produce greater thrust thanlimnetics during fast-starts. My hypothesis is that the morphological adaptations to steadyswimming in limnetics will impair their fast-start performance; therefore, benthics shouldhave higher fast-start performance than limnetics.MATERIALS AND METHODSFish Collection and MaintenanceFish were collected at the same time and kept in the same holding tanks as describedin Chapter 2.Fast-Start PerformanceFish were habituated to the bright lights (3 Berkey Coloran Halide 650W bulbs)needed for filming by turning the lights on during feeding every day for one week prior totesting. Feeding was stopped the day before testing. Nine litres of water (pretreated for a29minimum of three days with limestone, 3%o salt, and aeration) was replaced daily in theexperimental tank. Experiments were conducted at 150 ± 1 OC.Single, previously identified fish were transferred to the experimental glass tank (24cm X 29 cm X 14.4 cm) and allowed to habituate for one hour prior to being startled.Attached to the back and bottom of the tank were one cm square reference grids. Black papercovered the sides of the experimental tank so that the fish could not see the approachingstimulus. A mirror angled at 45° over the tank allowed the top view of the fish to be filmed.Profile view through the front of the tank, and top view from the mirror, were filmedsimultaneously (Redlake Locam camera Model 51 with Sun-Dionar 16 zoom lens at 400frames/second using Kodak 7250 colour 400 ASA tungsten high speed reversal film). Theimage on the film was split by a black tape line on the back, top edge of the tank. A 1 meterpole with a rubber ball on the end was struck against the side of the tank to elicit the escaperesponse. In all escapes analyzed, fish never touched the walls of the tank. The order inwhich the fish were tested was altered for each set of trials and all fish were allowed at leastone day rest between trials. An escape response was successfully elicited for all fish tested.Film AnalysisOne fast-stan was analyzed for each fish tested. Sequences were projected (Photooptical data analyzer, Photographic Analysis Ltd. model 224A) with frame by frame advanceonto a 450 mirror, and from the mirror onto a horizontal paper. The image was magnified 2.3to 3 times. This reduced the total error (measurement + framing rate) to approximately ±13% for maximum acceleration and made negligible the error for average acceleration,velocity, and distance (Harper and Blake 1989).Escape responses have been shown to be a fixed action pattern consisting of twostages, stage 1 (Si) which consists of a unilateral contraction of the axial muscles, bending30the fish into a C shape, followed by stage 2 (S2), a strong propulsive stroke of the tail in theopposite direction (Gillette 1987). A third stage in which the fish forms another curve orcoasts is possible (Webb and Blake 1985; Weihs 1973). Conventionally, studies havefocused on the first two stages (Domenici and Blake 1991; Eaton and Hackett 1984; Kasapiet al. 1993; Webb 1976, 1978). Therefore, this study included only Si and S2 for analysis.The durations of Si and S2 were defined by the changes in turn direction of the head.For each escape response the number of frames analyzed included 5 frames beforethe stimulus (start of Si) to 5 frames after S2. For each frame a flexible ruler was used tomeasure the apparent length of the fish from the top view. The position of the centre of mass(CM) along the fish’s length was then calculated for that particular fish and the positionmarked on the paper, along with the position of the head and the end of the caudal pedunclein the top view. Using a parallel ruler, a line was drawn from the CM on the top view to theprofile view. The depth of the fish at this line was measured. The position of CM along thefish’s depth was then calculated and marked on the paper. The position of CM on the straightbody of the fish is conventionally used as a point of reference for analysis of escapeperformance (Domenici and Blake 1991; Kasapi eta!. 1993).The centre of mass, tip of head, and end of caudal peduncle for both views wererecorded frame by frame. These points were later analyzed on a digitizing pad (GTCO type,0.61 m x 0.91 m) connected to a computer (80286 AT-compatible). Data were thentransferred to another 286 AT-compatible computer for further analysis.Velocity and acceleration during the two stages were derived by means of a five pointmoving regression (Lanczos 1956) on distance-time data. Vertical (z-axis), forward (x-axis,parallel with front glass of tank) and lateral (y-axis, parallel with the side glass of the tank)displacements of the centre of mass were combined using vector sums to expressdisplacement in three dimensions. The displacement of the centre of mass was calculatedbetween frames using the equation:31D =(x2+yz)°5 (9)(Shenk 1984) where D is the displacement of the centre of mass between the positions atsubsequent frames, and x, y, and z are the displacements of the centre of mass between thesame frames in the x, y, and z directions, respectively.The pitch angle was measured on the profile view as the angle between a line throughthe longitudinal axis of the fish and the horizontal grid on the back of the tank, parallel to thebottom of the tank. The turning angle was determined from the top view of the fish from aline through the same portion of the fish used to determine pitch and a horizontal line(bottom grid, parallel to front of tank). The pitch measured from profile views (apparentpitch, ‘y) is affected by the orientation of the fish along the y-axis and is therefore not the truepitch angle. Therefore, the instantaneous turning angle, t, at each frame was also measured.The actual pitch angle, P, was then determined by:P = arctan(tanyx cost) (10)(Kasapi et a!. 1993).Turning radius was measured from sequence tracings of the movement of CM.Turning radius is presented as a proportion of body length.Morphological MeasurementsMeasurements of weight, standard length, position of CM, fin areas, and surface areawere taken as described in Chapter 2. Morphological measurements were again adjusted to astandard size (5.0 cm) as described in Chapter 2.The cross-sectional depth of the body was measured with vernier calipers (± 0.005cm) at intervals of 10% of total length (0. 1L), while the fish was submerged in water in apetri dish to ensure full extension of the fins. The fish body was then cut into ten sections of32equal length (0.1L), and each section was weighed (Mettler model M3 ± 0.001 mg) todetermine the distribution of mass along the length of the fish’s body.Added mass (Ma), the mass of water entrained by the fish as it moves, was calculatedusing:Ma0.25P1td213 (ii)(Blake 1983), where p is the density of water, d is the body depth of the section, and 13 is ashape-dependent constant. For most fish cross-sections 13 is close to 1 (Blake 1983).The sex of the fish, and the presence of any external or internal parasites wererecorded. Fish which were obviously gravid or parasitized were not used.Statistical TestsAll data were compared using t-tests except the positions of maximum mass andadded mass along the length of the fish, which were compared using Mann-Whitney U tests.Percentage and proportion data were arcsin-square root transformed before testing with ttests. All null hypotheses were rejected at p <0.05 and all means are presented ±2 S.E.RESULTSFast-Start PerformanceLinear PerformanceAll fast starts were of the double bend type (Domenici and Blake 1991) and involvedthe formation of a C-shape by the body of the fish (Si), followed by a contralateral bend(S2), turning the head in the opposite direction.33Data for both sexes within a species were pooled because comparisons of males(benthic n =4, limnetic n =9) and females (benthic n =7, limnetic n = 3) within a species forall parameters in Si and S2 showed no significant differences (p > 0.05).No significant differences (p > 0.05) between limnetics (n = 12) and benthics (n = 11)were found for any of the linear parameters measured (Table III, Fig. 6). In general, fishtravelled for longer during Si than during S2, accelerated and travelled faster during S2 thanSi. The highest recorded total distance for a fast-start was 0.06 m, the highest velocity was1.59 m/s, and the highest acceleration was 215.59 m/s2. There were no allometric effects onduration, distance, velocity, or acceleration (p > 0.05).Although the distances travelled during escape responses were not different for thetwo species (Table Ill), benthics travelled in the lateral (y-axis) and forward (x-axis)directions significantly (p < 0.05) more than in the vertical (z-axis) direction, whereaslimnetics used all three equally (Figure 7).KinematicsBenthics had a tighter (p <0.05) turning radius for their length (T.R./L = 0.064 ±0.014) than did limnetics (0.087 ± 0.018). The tighter turns of the benthics are seen ontracings of body movement during escape fast-starts (Fig. 8). There was a positivecorrelation between fixed time average velocity and turning radii (r = 0.58, p <0.05).Pitch angles were defined as positive if the fish’s head pointed up and negative if thefish’s head pointed down. The initial pitch angle of benthics was significantly (p < 0.05)more positive than the pitch angles at the ends of Si and S2. There was no significantdifference (p > 0.05) between the pitch angles at the ends of Si and S2 for benthics. Thegreatest change occurred during Si, where the benthic fish changed from a positive to anegative pitch angle. Although the CM of the benthics only moved in the horizontal34directions (lateral, y-axis and forward, x-axis) during Si and S2, the head of the benthicspitched downward during Si. Limnetics showed no significant differences (p > 0.05)between initial, Si, and S2 pitch angles. Benthics had a significantly more downward pitch atthe end of S2 than limnetics (p <0.05), and all benthics had a downward final escape path,whereas 3 limnetics had an upward escape path (all others went down).Morphological MeasurementsThere was no difference between males and females within either species for any ofthe morphological measurements; therefore, the data for a given species were pooled.Benthics were heavier than limnetics, and their distribution of mass (Mf) along theirbody length differed (Fig. 9). Benthics were heaviest at 30% of total length (0.3L), theposition of maximum depth, whereas limnetics were heaviest at 0.4L where the pectoralmuscles are located (Table IV, Fig. 9).Although the size corrected added masses (Ma) for the two fishes were notsignificantly different (p > 0.05), the distributions of Ma along the length of the fish’s bodydid differ (Fig. 9). Benthics had the greatest Ma at 0.3L, whereas the position of maximumMa for limnetics was at 0.6L (Table IV). The maximum Ma did not include fins for benthics,whereas the maximum Ma for limnetics was influenced by the dorsal and anal fins.Although benthics were larger fish and therefore had larger fins, in relation to totalsurface area limnetics had larger(p<0.05) dorsal fins (4.8% ± 0.08%) and anal fins (4.1% ±0.09%) than benthics (dorsal fin = 4.2% ± 0.02%; anal fin = 3.2% ± 0.03%). The percentageof the surface area comprised of fin area for limnetics was 28.6% (± 0.14%), and for benthicsit was 2 1.46% (± 0.12%).35When the fin areas were corrected for size differences (Table IV), limnetics hadsignificantly (p <0.05) larger anal fins and smaller caudal fms than benthics. The dorsal finareas were not significantly different (p > 0.05).DISCUSSIONContrary to the prediction that benthics should have higher fast-start performancethan limnetics, the two fishes have equal performance. This result is consistent with theprobability of these fishes encountering the same predators. The major predators ofsticklebacks include piscivorous fish and diving birds (Wootton 1984). Larson (1972) foundcutthroat trout in Paxton Lake, and loons are present on the lake (personal observation).These predators probably have access to both habitats used by the two species. Therefore, itseems likely that limnetics and benthics are exposed to predation pressure from the samepredators and thus must share the ability to escape.This study provides a comparison of two pectoral fin propulsion specialists. Theresults show that this specialization does not impair performance in body/caudal fin burstswimming. Harper and Blake (1990) found that pike achieve maximum accelerations of157.8 m/s2. Some median/paired fin specialists can also achieve high accelerations duringescape fast-starts (Domenici and Blake 1991; Kasapi et al. 1993). Angelfish, Pterophyllumeimekei, are capable of accelerating at 79.0 m/s2 (Domenici and Blake 1991), and knifefish,Xenomystus nigri at 127.9 m/s2 (Kasapi et a!. 1993). These fishes are thought to bespecialists for both slow speed manoeuverabiity and high speed fast-starts (Domenici andBlake 1991; Kasapi et al. 1993). Both angelfish and knifefish are laterally compressed,36which provides a large body depth for thrust during fast-starts. The sticklebacks in this studyare more fusiform in shape, and it was therefore predicted that they should have poor fast-start performance. However, these sticklebacks achieve an average maximum acceleration of133.9 m/s2, and the highest recorded acceleration was 215.9 m/s2. Although sticklebacks arepaired fin specialists and are fusiform in shape, they achieve accelerations comparable withfast-start specialists. Therefore, specialization for paired fin swimming does not compromisebody/caudal fm propulsion during fast-starts. The pectoral muscles and pectoral fms used forsteady swimming do not interfere with, or compromise the myotomal muscles and caudal finused during fast-starts. These two swimming modes use decoupled propulsion systems.Consequently, high performance in both these modes is not mutually exclusive (Domenici1993; Kasapi et al. 1993).The effect of size on fish swimming performance is widely documented (Bainbridge1959; Wardle 1975, 1977; Webb 1976, 1977). During escape responses, larger fish attainhigher velocities than smaller ones. This is because fast-start durations are longer in largerfish, whereas, acceleration is size independent (Webb 1976). When considered within a fixedtime, however, both distance iravelled and maximum velocity are size independent (Webb1976). These comparisons are among fishes of very different body sizes (Bainbridge 1959;Wardle 1975; 1977; Webb 1976). In this study, the fishes were relatively similar in size.Distance travelled, velocity, and acceleration were all found to be independent of size forthese sticklebacks.Weihs (1973) hypothesized that body morphology is a key determinant of maximumacceleration of fish. Morphological characteristics of pike, such as a large surface areacaudally and a high percentage of body mass as muscle, are considered favourable for highthrust production during fast-starts (Webb 1984a). Harper and Blake (1990) reported thatpike achieve high acceleration during fast-starts. Intraspecific comparisons of fast-startperformance for coho salmon (Taylor and McPhail 1985) and threespine stickleback (Taylor37and McPhail 1986) showed that fish with greater body depth and caudal fin depth are capableof greater maximum and average velocities.The caudally placed anal and dorsal fins of pike increase fast-start performance (Frithand Blake 1991). Thrust from the caudal fm and the body section which contains the dorsaland anal fins account for > 90% of total thrust (Frith and Blake 1991). The combination ofhigher velocities and angles, and larger depths in these sections all contribute to higherthrust. The posterior placement of dorsal and anal fins in pike contributes 26% of total thrust(Frith 1990). The fins increase the depth of section and therefore the added mass, and alsogenerate lift forces (Weihs 1973). Webb (1978) found that for trout, caudally placed dorsaland anal fins contribute approximately 27% to total thrust. In addition, Webb et al. (1992)found that the low power produced by the caudal fin of the gar was compensated for by themedian fins, which were caudally positioned.The dorsal fm of limnetic sticklebacks is placed further back on the body, directlyabove the anal fin , and the anal fin surface area is enlarged in comparison to benthics. Thisincreases the body depth for limnetics and therefore increases added mass. The position ofgreatest added mass for limnetics is at 0.6L, where the depth is due to the dorsal and anal finheights. The limnetic’s enlarged anal fin and caudally placed dorsal fin increase added masswhere the majority of thrust is produced, thereby enhancing fast-start performance.Among fishes, depth is often enhanced to increase the added mass associated withacceleration reaction propulsion forces (Domenici and Blake 1991; Kasapi et al. 1993;Weihs 1989). Fins are more effective structures for increasing added mass than is body depth(Weihs 1989). This is because fins do not greatly increase mass which can hinder fast-startperformance (deBuffrénil et al. 1985; Webb and deBuffrénil 1990; Webb and Skadsen1980), but do increase added mass which benefits performance (Weihs 1989). Limneticsachieve a high fast-start performance by having large, caudally placed fins, and benthicsachieve the same performance by having large body depth.38It has been suggested that turning radius is a relevant parameter in predator-preyinteractions (Howland 1974; Webb 1976; Weihs and Webb 1984). Transient swimmingperformance has been defined as the ratio of the minimum turning radius of the center ofmass to the length of the body (Webb 1984a). The smaller the ratio, the better theperformance (Webb 1984a). Domenici and Blake (1991) found that angelfish had a turningradius to body length ratio (T.R.JL) of 0.065 indicating a very flexible body and high fast-start performance. Benthics have a smaller T.R.IL (0.064) than limnetics (0.087), indicatinggreater flexibility than limnetics during fast-starts. A low value of turning radius is beneficialin complex environments such as the benthic region of the lake. However, in the open watercolumn a tight turning radius would not be as important as the ability to move swiftly awayfrom a predator. There is a trade-off between average escape velocity and the tightness of theturning radius for these sticklebacks, suggesting that some fish elude predators by achievinghigh velocities and others by changing direction sharply.Active defenses, such as acceleration and speed are reduced in those species withpassive defense mechanisms, such as camouflage and armor (Eaton et at. 1977; Webb et at.1992). Passive defence such as armor may decrease the importance of speed in fleeing forcover (Webb et a!. 1992). The sticklebacks have an elaborate defensive armor that has beenlinked to predation pressure (Gross 1978; Hagen and Gilbertson 1973; Reimchen 1980; Reist1980a, 1980b). Therefore, these fishes might be expected to have reduced fast-startperformance. That they have such high performance is interesting. Perhaps the damageinflicted during an encounter with a predator, even though the fish may escape, isdisadvantageous to the fitness of the fish. Therefore, the ability to avoid such encounters maybe under selection. Another possibility is that more than one predator is having an effect onthe evolution of these fishes. One predator may be deterred by the presence of spines whileanother is not. Therefore, the sticklebacks would need both the presence of spines and theability to escape quickly to successfully avoid both predators.39Morphological responses to differential predation pressure have been suggested byvarious researchers (Larson 1976; McPhail 1977; Moodie and Reimchen 1976; Reimchen1980; Reist 1980a, 1980b). Burst swimming performance appears to be directly related to theability to evade predators (Taylor and McPhail 1985; Webb 1981; Webb and Corolla 1981),and body shape is an important variable in defining levels of evasion. Bronmark and Miner(1992) showed that a predator can induce a phenotypical change in body morphology,specifically an increase in body depth to avoid gape limited predators. Fish specialized formanoeuvering have large body depth (Webb 1990). Benthics, which live in the complexhabitat of the littoral zone, are deeper bodied than the pelagic limnetics, possibly to aid inmanoeuverability. The number and length of spines is greater for limnetics than benthics(McPhail 1992). Therefore, benthics may have deeper bodies to avoid gape limited predatorsand increase manoeuverability, whereas limnetics have numerous, large spines to repelpredators and a slender shape to reduce drag during steady swimming.With large, caudally placed fins limnetics have achieved the same thrust capabilitiesfor fast-starts as benthics. These results suggest directional selection towards high escapefast-start performance for both species. Maintenance of high performance in these fishes isconsistent with the likelihood of encounters with the same predators.Table ifi - Escape Fast-Start Performance (±2 S.E.) 40average maximum average maximumduration distance velocity velocity acceleration acceleration(s) (m) (mis) (mis) (m/s) (m/s2)STAGE 1Benthic .029 .017 .54 .79 4.94 96.45(.004) (.003) (.06) (.11) (9.78) (24.39)Limnetic .027 .018 .60 .88 4.01 92.75(.003) (.003) (.06) (.09) (9.58 (23.98)Pooled .028 .017 .57 .84 4.46 94.52(.003) (.002) (.04) (.07) (6.69) (16.73)STAGE 2Benthic .021 .017 .81 1.07 15.19 121.13(.008) (.006) (.15) (.18) (13.08) (24.12)Limnetic .021 .017 .86 1.06 19.04 127.99(.003) (.004) (.18) (.18) (9.72) (28.70)Pooled .021 .017 .84 1.07 17.20 124.71(.004) (.003) (.12) (.13) (7.91) (18.53)TOTAL (Si + S2)Benthic .051 .034 .66 1.10 9.38 127.93(.008) (.007) (.07) (.17) (7.16) (24.31)Limnetic .048 .034 .70 1.09 10.09 139.41(.004) (.006) (.10) (.16) (6.77) (22.33)Pooled .049 .034 .68 1.10 9.75 133.92(.004) (.004) (.06) (.12) (4.81) (16.27)FIXED TIME (0.03 s)Benthic .018 .56 .86 8.61 101.07(.002) (.07) (.12) (9.92) (21.32)Limnetic .020 .61 .93 7.74 101.19(.002) (.08) (.12) (10.68) (18.94)Pooled .019 .59 .89 8.16 101.13(.002) (.05) (.09) (7.15) (13.87)Comparisons of all parameters not significant, p > 0.05Benthic n = 11 Limnetic n 12TableIV-MorphologicalMeasurements:FinAreasandPositionsofMaximumMassandAddedMass(±2S.E.)ParameterBenthicLimneticAnal FinArea(cm2)0.34(0.05)0.22(0.04)*Dorsal FinArea(cm2)0.44(0.05)0.25(0.03)*CaudalFinArea(cm2)0.78(0.08)0.39(0.05)*Positionof MaximumMassAlongLength0.3L0.4L*PositionofMaximumAddedMassAlongLength0.3L0.6L*SizeCorrectedAnalFinArea(cm2)0.18(0.16)0.38(0.05)*SizeCorrectedDorsalFinArea(cm2)0.27(0.06)0.34(0.03)SizeCorrectedCaudalFinArea(cm2)0.64(0.05)0.53(0.05)**significant at p<0.05Benthicn=10Limneticn=100.750.500.25150100-100Limnetic0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07Time (s) 420.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07Time (s) 6 - The escape fast-start performance of sticklebacks. Shownare the distance travelled, velocity and acceleration over time. Thestacked graphs on the left are for a limnetic, and those on the rightfor a benthic. I0.011.251.005000-50500C)C)0.0300.0300.0250.0250.0200.020_-4———0.0150.015C)C)•0.010-0.010rd)•i:50.0050.0050. threedimensional spaceduringescapefast-startsof sticklebacks.Shownarethedistancestravelledovertimeintheforwarddirection,(),lateral(-—andvertical(- -—-)directions foralimnetic(left)andabenthic(right)threespinestickleback.1,If“yr/7 I,I--,//I-’/——I-...,/—III/// / /// I/ / //—____—---..w422BenthicFigureS-Kinematicsofescapefast-startsforsticklebacks.Tracings,magnifiedI.5X, ofthemovementof alimnetic(left)andabenthic(nght)threespinesticklebackduringfast-startescapes.Arrowheads representtheheadof thefish,andfilledcirclesrepresentthecentreofmass.Numbersindicatesequential tracings aridareatO.005sintervals.61661810F 122426Limnetic2845C,,1)AB0. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Proportion of LengthFigure 9. Part A - Drawings of Paxton Lake limnetic and benthicthreespine stickleback (modified from Schiuter 1993).Part B - The distribution of mass and added mass along the lengthof the fish’s body. The distribution of mass for benthics ( ) andlimnetics (- -- — -), and the distribution of added mass for benthics— ) and limnetics (- - — -) are shown to determine the positionsof maximum mass and maximum added mass for these fishes.46CHAPTER 4General Discussion and ConclusionsWebb (1984a) classified fish swimming styles into three broad categories: steadyswimming (cruising), transient swimming (fast-starts), and manoeuvering. He argued thatspecialist fish which excel in one of these categories sacrifice performance in the others,while generalist fish perform moderately well in all of these swimming modes but havesuperior performance in none (Webb 1984a). Optimal morphological characteristics for oneparticular swimming function are thought to interfere or trade-off with the others. However,both of the sticklebacks of this study were capable of superior manoeuverability and fast-start performance. Accelerating and manoeuvering imply axial locomotion and paired finlocomotion, respectively. These systems are decoupled; the propulsors for paired finswimming do not impair those for fast-starts (Domenici 1993; Kasapi et al. 1993). Thepectoral muscles and fins used for paired fin manoeuvering are separate from the myotomalmuscles and body/caudal fm sections used for fast-starts. Therefore, there is no compromisebetween adaptations for manoeuverability and burst swimming. In addition, manoeuveringspecialists can vary greatly in overall body form and still maintain high hydromechanicalefficiency and thrust, whereas specialists for body/caudal fin propulsion are constrained to afew optimal designs (Blake 1994, in prep.).Limnetics achieved high performance in all three swimming categories. These fishhave a streamlined body to reduce drag during steady swimming, large pectoral fins andpectoral muscles for steady swimming and manocuvering, and large, caudally placed fins forfast-starts. This is the first time a fish has been shown to excel in all three swimmingcategories. The compromise between good steady swimming and fast-start ability is avoided47with large, caudally placed fins to increase depth for fast-starts on an otherwise streamlinedbody.Movement through water is energetically expensive and organisms must trade-off thecosts of locomotion with the benefits of the behaviour, such as encountering prey (Ware1978). Foraging models assume that surplus power or net energy gain per unit time ismaximized, and yet the cost of locomotion is often ignored (Ware 1978, 1982). A relativelylarge proportion of the daily energy budget of fishes is allocated to swimming (Videler1993). Parrotfish spend 50% of their waking time swimming (Videler 1993) and pike spendapproximately 20% (Lucas et al. 1991). Therefore, a large fraction of the energy budget offish is required to fuel locomotion. Characteristics which decrease the cost of locomotionwould make available surplus energy for reproduction and thereby enhance fitness (Blake1983). Therefore, traits which decrease locomotor costs, such as streamlining to reduce drag,are under selective pressure.Several studies have shown that swimming performance and growth (a fitnesscomponent) are linked (Davison and Goldspink 1977; East and Magnan 1987; Greer Walkerand Emerson 1978; Koch and Wieser 1983). East and Magnan (1987) found a negativerelationship between swimming speed and growth for brook char (Salvelinus fontinalis).Experiments with Salmo trutta (Davison and Goldspink 1977) and with Oncorhychus mykiss(Greer Walker and Emerson 1978) showed a similar negative relationship between growthand swimming activity. Growth has many positive fitness consequences in fish (Schiuter1994), such as higher overwinter survival (Conover 1992; Shuter and Post 1990), higherfecundity (Bagenal 1978; Schluter 1994), and earlier breeding (Schultz et al. 1991). Inaddition to locomotion and body growth relationships, gonadal growth and swimmingactivity have been shown to be correlated (Koch and Weiser 1983). Koch and Wieser (1983)found that while gonadal growth was occurring, swimming activity was reduced, indicating atrade-off between energy for swimming and energy for gonad synthesis. These results48suggest that swimming performance influences the fimess of fishes. Therefore, thecharacteristics that enhance performance are probably under selective pressure.How body features affect the cost of locomotion is fundamental to biomechanicalstudies (Blake 1983; Frith and Blake 1991; Harper and Blake 1988; Lindsey 1978; Videler1993; Webb 1984a, 1984b, 1976; Weihs 1973, Weihs and Webb 1984). Morphologicalcharacteristics can be related to selective pressures by knowledge of the abilities of the fishand the functional significance of its morphological traits (Bock 1980). Differences inmorphology that have a genetic basis can therefore be linked with the underlying processesof adaptive radiation by studies of the biomechanical significance of the divergent traits.In this study, knowledge of the distinctive morphological traits of the fishes, alongwith information on their swimming abilities can be used to link the morphologies of thesefishes with the selective forces involved in their divergence. There are body shapedifferences that relate to differential selection for the methods of locating prey. Limneticsfeed on zooplankton in the open water column (Larson 1972, 1976; Schiuter 1993; Schiuterand McPhail 1992) and must have good cruising ability to locate their prey. Selection onthese fish has favoured traits that enhance steady swimming ability. Limnetics also exploitthe complex habitat of the littoral zone during the breeding season (Larson 1976; Schiuter1992) and are also adapted for good manoeuverability. Benthics live in the complex habitatthroughout their lives (Larson 1976; Schiuter 1992). Selection on benthics has acted towardshigh manoeuverability but not good steady swimming ability. Selection due to differentialresource use has probably lead to the divergence of body form of these fishes.Competition for resources has been shown to lead to character displacement (FjeldsA1983; Grant and Grant 1989; Schiuter 1988; Schiuter and Grant 1984; Schiuter and McPhail1992; Smith 1990; Werner 1977). Schiuter and McPhail (1992) presented evidence ofcharacter displacement among species of threespine sticklebacks from British Columbia andtheir results suggest that competition for food has played a critical role in the divergence49between species. Schiuter (1994) showed that habitat use efficiency by these sticklebacks iscorrelated with growth, indicating a relationship between morphology, habitat use, andfitness. My study supports the conclusion that differential resource use has lead to divergencein the morphological characteristics of these sticklebacks.The results of my study also suggest that selective pressure from predators has lead toconsistent high escape fast-start performance. These fish use different methods to achieve thesame goal of having a large body depth caudally to produce thrust for high performance ofescape fast-starts. This similarity of performance suggests that the selection pressure frompredators is, or has been very strong on both of these fishes. Recent investigations haveshown a consistency among fishes to have high escape fast-start performance (Domenici andBlake 1991; Kasapi et al. 1993). Selection pressure from predation can therefore lead toconvergence on morphologies that enhance thrust for high fast-start performance.In conclusion, this sympatric species pair of threespine stickleback has differentialswimming performances that are a result of their morphological differences. Habitat use andmorphological characteristics are linked by the influence that different traits have onswimming abilities. The morphological characteristics of these fishes have been underselective pressures from both resource competition and predation. Competition for resourceshas lead to character displacement and divergent selection while predation has lead todirectional selection towards high escape fast-start performance. Studies of the processesinvolved in adaptive radiation of species should consider the possibility that differentcharacteristics in diverging species may be the result of a combination of divergent andconvergent selective pressures acting on separate traits.50BIBLIOGRAPHYAnker, G.C. 1978. Analyses on respiration and feeding movements of the three-spinedstickleback, Gasterosteus aculeatus L. Neth. 3. Zool. 28(3-4):485-523Bagenal, T.B. 1978. Aspects of fish fecundity. In Ecology of Freshwater Fish Production.S.D. Gerking ed. Wilery, New York pp. 75-101Bainbridge, R. 1959. Speed and stamina in three fish. Exp. Biol. 37(1):129-154Baumgartner, J.V., Bell, M.A., and Weinberg, P.H. 1988. Body form differences between theEnos Lake species pair of threespine sticklebacks (Gasterosteus aculeatus complex).Can. 3. Zool. 66:467-474Beamish, F.W.H. 1984. Swimming performance of three southwest Pacific fishes. Mar. Biol.79:311-3 13Bentzen, P. and McPhail, J.D. 1984. Ecology and evolution of sympatric sticklebacks(Gasterosteus): specialization for alternative trophic niches in the Enos Lake speciespair. Can. J. Zool. 62:2280-2286Bentzen, P., Ridgway, M.S., and McPhail, J.D. 1984. Ecology and evolution of sympatricsticklebacks (Gasterosteus): spatial segregation and seasonal habitat shifts in theEnos Lake species pair. Can. J. Zool. 62:2436-2439Blake, R.W. 1983. Fish Locomotion. Cambridge University Press. Cambridge, UKBock, W.J. 1980. The definition and recognition of biological adaptation. Amer. Zool.20:217-227Boddeke, R., Slijper, E.J., and Van Der Stelt, A. 1959. Histological characteristics of thebody musculature of fishes in connection with their mode of life.Proc. K. Ned. Akad. Wet. 62:576-588Brett, J.R. 1965. The relation of size to rate of oxygen consumption and sustained swimmingspeed of sockeye salmon (Oncorhynchus nerka). J. Fish. Res. Bd. Can.22(6):1491-1501Brett, J.R. and Sutherland, D.B. 1964. Respiratory Metabolism of Pumpkinseed (Lepomisgibbosus) in relation to swimming speed. 3. Fish. Res. Bd. Can. 22(2):405-409Bronmark, C. and Miner, J.G. 1992. Predator-induced phenotypical change in bodymorphology in crucian carp. Science 258:1348-1350Clarke, B. 1978. Some contributions of snails to the development of ecological genetics. InEcological Genetics: The Interface. P.F. Brussard ed. Springer Verlag, New York. pp159- 170.Conover, D.O. 1992. Seasonality and the scheduling of life history at different latitudes.J. Fish. Biol. 41B: 161-178.51Davison, W. and Goldspink, G. 1977. The effect of prolonged exercise on the lateralmusculature of the brown trout (Salmo trutta). J. Exp. Biol. 70:1-12Domenici, P. 1993. The Fast-Start Swimming of Angelfish, Pterophyllum eimekei. Ph.D.University of British Columbia, Vancouver l37ppDomenici, P. and Blake, R.W. 1991. The kinematics and performance of the escape responsein the angelfish (Pterophyllum eimekei). J. Exp.Biol. 156:187-205Dorn, P., Johnson, L., and Darby, C. 1979. The swimming performance of nine species ofcommon California inshore fishes. Trans. Amer. Fish. Soc. 108:366-372DuBois, A.B., Giovanni, A.C., and Fox, R.S. 1976. Locomotion of bluefish. J. Exp. Zool.195:223-236East, P. and Magnan, P. 1987. The effect of locomotor activity on the growth of brook charr,Salvelinusfontinalis Mitchell. Can. J. Zool. 65:843-846Eaton, R.C., Bombardieri, R.A., and Meyer, D.L. 1977. The Mauthner-initiated startleresponse in teleost fish. J. Exp. Biol. 66:65-8 1Eaton, R.C. and Hackett, J.T. 1984. The role of the Mauthner cell in fast-starts involvingescape in teleost fishes. In Neural Mechanisms of Startle Behaviour. Plenum Press.New York. pp. 2 13-266Ehlinger, T.J. and Wilson, D.S. 1988. Complex foraging polymorphism in bluegill sunfish.Proc. Nat. Acad. Sci. USA 85:1878-1882Farlinger, S. and Beamish, F.W.H. 1977. Effects of time and velocity increments on thecritical swimming speed of largemouth bass (Micropterus salmoides).Trans. Amer. Fish. Soc. 106:436-439FjeldsA, J. 1983. Ecological character displacement and character release in Ibis 125:463-48 1Frith, H.R. 1990. Energetics of Fast-Starts in Northern Pike, Esox lucius. Ph. D. Universityof British Columbia, Vancouver 149ppFrith, H.R. and Blake, R.W. 1991. Mechanics of the startle response in the northern pike,Esox lucius. Can. J. Zool. 69(11):2831-2839Gal, J.M. and Blake, R.W. 1987. Hydrodynamic drag of two frog species: Hymenochirusboettgeri and Rana pipiens. Can. J. Zool. 65(5): 1085-1090Gillette, R. 1987. The role of neural command in fixed action patterns of behaviour. In Aimsand Methods in Neuroetholov. Manchester University Press. Manchester. pp. 46-79Grant, B.R. and Grant, P.R. 1989. Natural selection in a population of Darwin’s finches.Amer. Nat. 133(3):377-39352Greer Walker, M. and Emerson, L. 1978. Sustained swimming speeds and myotomal musclefunction in the trout, Salmo gairdneri. 3. Fish. Biol. 13:475-481Gross, H.P. 1978. Natural selection by predators on the defensive apparatus of the threespined stickleback, Gasterosteus aculeatus L. Can. J. Zool. 56:398-4 13Hagen, D.W. and Gilbertson, L.G. 1973. Selective predation and the intensity of selectionacting upon the lateral plates of threespine sticklebacks. Heredity 30(3):273-287Harper, D.G. and Blake, R.W. 1988. Energetics of piscivorous predator-prey interactions.J. Theor. Biol. 134:59-76Harper, D.G. and Blake, R.W. 1989. On the error involved in high speed film when used toevaluate maximum accelerations of fish. Can. J. Zool. 67(8):1929-1936Harper, D.G. and Blake, R.W. 1990. Fast-start performance of rainbow trout Salmo gairdneriand northern pike Esox lucius. J. Exp. Biol. 150:321-342Hoar, W.S. and Randall, D.J. 1978. Fish Physiology. Academic Press. New YorkHorwood, J.W. and Cushing, D.H. 1977. Spatial distributions and ecology of pelagic fish. InSpatial Patterns in Plankton Communities. Plenum Press. New Yorkpp. 355-383Howland, H.C. 1974. Optimal strategies for predator avoidance: The relative importance ofspeed and manoeuvrability. J. Theor. Biol. 47:333-350Kasapi, M.A., Domenici, P., Blake, R.W., and Harper, D. 1993. The kinematics andperformance of escape responses of the knifefish Xenomystus nigri. Can. 3. Zool.71(1): 189-195Koch, F. and Wieser, W. 1983. Partitioning of energy in fish: Can reduction of swimmingactivity compensate for the cost of reproduction? J. Exp. Biol. 107:141-146Kolok, A.S. 1992. Morphological and physiological correlates with swimming performancein juvenile largemouth bass. J. Amer. Physiol. Soc. R1042-R1048Lanczos, C. 1956. Applied analysis. Prentice Hall, Englewood Cliffs, New Jersey.Larson, G.L. 1972. Social behavior and feeding ability of two phenotypes of Gasterosteusaculeatus in relation to their spatial and trophic segregation in a temperate lake.Ph. D. University of British Columbia, Vancouver ll3ppLarson, G.L. 1976. Social behavior and feeding ability of two phenotypes of Gasterosteusaculeatus in relation to their spatial and trophic segregation in a temperate lake.Can. J. Zool. 54(2):107-121Lavin, P.A. and McPhail, J.D. 1985. The evolution of freshwater diversity in the threespinestickleback (Gasterosteus aculeatus): site-specific differentiation of trophicmorphology. Can. J. Zool. 63:2632-263853Lavin, P.A. and McPhail, J.D. 1986. Adaptive divergence of trophic phenotype amongfreshwater populations of threespine sticldeback (Gasterosteus aculeatus).Can. J. Fish. Aquat. Sci. 43:2455-2463Liem, K.F. 1974. Evolutionary strategies and morphological innovations: Cichlid pharyngealjaws. Syst. Zool. 22:425-441Lindsey, C.C. 1978. Form, Function, and Locomotory Habits in Fish. In Fish Physiology.Academic Press. New York, pp. 1-100Lucas, M.C., Priede, I.G., Armstrong, J.D., Gindy, A.N.Z., and de Vera, L. 1991. Directmeasurements of metabolism, activity and feeding behaviour of pike, Esox lucius L.,in the wild, by the use of heart rate telemetry. J. Fish. Biol. 39:325-345Luiker, E.A. and Stevens, E.D. 1992. Effect of stimulus frequency and duty cycle on forceand work in fish muscle. Can. J. Zool. 70:1135-1139Mathews, W.H., Fyles, J.G., and Nasmith, H.W. 1970. Postglacial crustal movements insouthwestern British Columbia and adjacent Washington state. Can. I. Earth Sci.7:690-702Mayr, E. 1978. Animal Species and Evolution. Harvard University Press. Cambridge, MAMcPhail, J.D. 1977. Inherited interpopulation differences in size at first reproduction inthreespine stickleback, Gasterosteus aculeatus L. Heredity 38(1) :53-60McPhail, J.D. 1984. Ecology and evolution of sympatric sticklebacks (Gasterosteus):morphological and genetic evidence for a species pair in Enos Lake, BritishColumbia. Can. J. Zool. 62:1402-1408McPhail, J.D. 1992. Ecology and evolution of sympatric sticklebacks (Gasterosteus):evidence for a species-pair in Paxton Lake, Texada Island, British Columbia.Can. J. Zool. 70(2):361-369Moodie, G.E.E. and Reimchen, T.E. 1976. Phenetic variation and habitat differences inGasterosteus populations of the Queen Charlotte Islands. Syst. Zool. 25:49-61Niimi, A.J. and Beamish, F.W.H. 1974. Bioenergetics and growth of largemouth bass(Micropterus salmoides). Can. J. Zool. 52:447-456Rand, D.M. and Lauder, G.V. 1981. Prey capture in the chain pickerel, Esox niger:Correlations between feeding and locomotor behavior. Can. I. Zool. 59(6):1072-1078Reimchen, T.E. 1980. Spine deficiency and polymorphism in a population of Gasterosteusaculeatus: an adaptation to predators? Can. J. Zool. 58: 1232-1244Reist, J.D. 1980a. Predation upon pelvic phenotypes of brook stickleback, Culaeainconstans, by selected invertebrates. Can. J. Zool. 58:1253-1258Reist, J.D. 1980b. Selective predation upon pelvic phenotypes of brook stickleback, Culaeainconstans, by northern pike, Esox lucius. Can. J. Zool. 58:1245-125254Richerson, P.J., Powell, T.M., Leigh-Abbott, M.R., and Coil, J.A. 1977. Spatialheterogeneity in closed basins. In Spatial Patterns in Plankton Communities. PlenumPress. New York, NY. pp. 239-276Schiuter, D. 1988. Character displacement and the adaptive divergence of finches on islandsand continents. Amer. Nat. 131(6):799-824Schiuter, D. 1993. Adaptive radiation in sticklebacks: size, shape, and habitat use efficiency.Ecol. 74(3):699-709Schiuter, D. 1994. Adaptive radiation in sticklebacks: Trade-offs in feeding performance andgrowth. Ecol. In PressSchluter, D. and Grant, P.R. 1984. Determinants of morphological patterns in communitiesof Darwin’s finches. Amer. Nat. 123(2):175-196Schiuter, D. and McPhail, J.D. 1992. Ecological character displacement and speciation insticklebacks. Amer. Nat. 140(1): 85-108Schultz, E.T., Clifton, L.M., and Warner, R.R. 1991. Energetic constraints and size-basedtactics: the adaptive significance of breeding-schedule variation in a marine fish(Embiotocidae: Micrometrus minimus). Am. Nat. 138:1408-1430.Shenk, A. 1984. Calculus and Analytic Geometry. Scott, Foresman and Company. Glenview,Illinois. 1131 ppShuter, B., and Post, J.R. 1990. Climate, population variability, and the zoogeography oftemperate fishes. Trans. Am. Fish. Soc. 119: 3 14-336.Smith, T.B. 1990. Natural selection on bill characters in the two bill morphs of the Africanfinch (Pyrenestes ostrinus. Evol. 44(4):832-842Stahlberg, S. and Peckman, P. 1987. The critical swimming speed of small Teleost fishspecies in a flume. Arch. Hydrobiol. 110(2):179-193Steel, R.G.D. and Torrie, J.H. 1980. Principles and Procedures of Statistics: ABiometrical Approach. McGraw-Hill Inc. New York, NY.Steele, J.H. 1977. Some comments on plankton patches. In Spatial Patterns in PlanktonCommunities. Plenum Press. New York pp. 1-20Taylor, E.B. and McPhail, J.D. 1985. Variation in burst and prolonged swimmingperformance among British Columbia populations of coho salmon, Oncorhynchuskisutch. Can. J. Fish. Aquat. Sci. 42:2029-2033Taylor, E.B. and McPhail, J.D. 1986. Prolonged and burst swimming in anadromous andfreshwater threespine stickleback, Gasterosteus aculeatus. Can. J. Zool. 64:416-420Tsukamoto, K., Kajihara, T., and Nishiwaki, M. 1975. Swimming ability of fish.Bull. Jap. Soc. Sci. Fish. 41(2):167-17455Tsuyuki, H. and Williscroft, S.N. 1977. Swimming stamina differences betweengenotypically distinct forms of rainbow (Salmo gairdneri) and steelhead trout.J. Fish. Res. Bd. Can. 34:996-1003Varley, M.R. 1989. Functional design and swimming energetics of the freshwater pufferfish,Tetraodonfluviatilis. M. Sc. University of British Columbia, Vancouver 184Videler, J.J. 1993. Fish Swimming. Chapman and Hall Inc. London.Vinyard, G.L. 1982. Variable kinematics of Sacramento perch (Archoplites interruptus)capturing evasive and nonevasive prey. Can. J. Fish. Aquat. Sci. 39:208-211Vogel, S. 1981. Life in Moving Fluids. Princeton University Press. Princeton, NJ.Wainwright, P.C. 1988. Morphology and ecology: Functional basis of feeding constraints inCaribbean Labrid fishes. Ecol. 69(3):635-645Wardle, C.S. 1975. Limit of fish swimming speed. Nature 255:725-727Wardle, C.S. 1977. Effects of size on the swimming speeds of fish. In Scale Effects inAnimal Locomotion. Academic Press. New York pp. 299-3 13Ware, D.M. 1978. Bioenergetics of pelagic fish: Theoretical change in swimming speed andration with body size. J. Fish. Res. Bd. Can. 35:220-228Ware, D.M. 1982. Power and evolutionary fitness of teleosts. Can. J. Fish. Aquat. Sci.39:3- 13Webb, P.W. 1975. Acceleration performance of rainbow trout Salmo gairdneri and greensunfish Lepomis cyanellus. J. Exp. Biol. 63:45 1-465Webb, P.W. 1976. The effect of size on the fast-start performance of rainbow trout Salmogairdneri, and a consideration of piscivorous predator-prey interactions. J. Exp. Biol.65:157-177Webb, P.W. 1977. Effects of size on performance and energetics of fish. In Scale Effects inAnimal Locomotion. Academic Press. New York pp. 315-331Webb, P.W. 1978. Fast-start performance and body form in seven species of teleost fish.J. Exp. Biol. 74:211-226Webb, P.W. 1981. The effect of the bottom on the fast-start of flatfish Citharichthysstigmaeus. U.S. Fish. Bull. 79(2):271-276Webb, P.W. 1982. Fast-start resistance of trout. J. Exp. Biol. 96:93-106Webb, P.W. 1984a. Form and function in fish swimming. Sci. Amer. 251(1):58-68Webb, P.W. 1984b. Body form, locomotion and foraging in aquatic vertebrates. Amer. Zool.24: 107-12056Webb, P.W. 1990. How does benthic living affect body volume, tissue composition, anddensity of fishes? Can. 3. Zool. 68:1250-1255Webb, P.W., and Blake, R.W. 1985. Swimming. In Functional Vertebrate Morphology. M.Hildebrand , D.M. Bramble, K. Liem, and D.B. Wake eds. pp. 110-128Webb, P.W. and Corolla, R.T. 1981. Burst swimming performance of northern anchovy,Engraulis mordax, larvae. U. S. Fish. Bull. 79(1):143-150Webb, P.W. and deBuffrénil, V. 1990. Locomotion in the biology of large aquaticvertebrates. Trans. Amer. Fish. Soc. 119:629-641Webb, P.W. and Skadsen, J.M. 1980. Strike tactics of Esox. Can. J. Zool. 58: 1462-1469Webb, P.W., Hardy, D.H., and Mehi, V.L. 1992. The effect of armored skin on theswimming of longnose gar, Lepisosteus osseus. Can. J. Zool. 70:1173-1179Weihs, D. 1973. The mechanism of rapid starting of slender fish. Biorh. 10:343-350Weihs, D. 1989. Design features and mechanics of axial locomotion in fish. Amer. Zool.29:151-160Weihs, D. and Webb, P.W. 1984. Optimal avoidance and evasion tactics in predator-preyinteractions. J. Theor. Biol. 106:189-206Werner, E.E. 1977. Species packing and niche complementarity in three sunfishes.Amer. Nat. 111(979):553-578Whoriskey, F.G. and Wootton, R.J. 1987. The swimming endurance of threespinesticklebacks, Gasterosteus aculeatus L., from the Afron Rheidol, Wales. J. Fish Biol.30:335-339Wootton, R.J. 1976. The Biology of the Sticklebacks. Academic Press. New YorkWootton, R.J. 1984. A Functional Biology of Sticklebacks. Croom Helm Ltd.Sidney, Australia.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items