UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fluvial mountain whitefish (prosopium williamsoni) in the Upper Fraser River: a morphological, behavioural,… Troffe, Peter M. 2000

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

Item Metadata


831-ubc_2000-0292.pdf [ 2.42MB ]
JSON: 831-1.0089468.json
JSON-LD: 831-1.0089468-ld.json
RDF/XML (Pretty): 831-1.0089468-rdf.xml
RDF/JSON: 831-1.0089468-rdf.json
Turtle: 831-1.0089468-turtle.txt
N-Triples: 831-1.0089468-rdf-ntriples.txt
Original Record: 831-1.0089468-source.json
Full Text

Full Text

FLUVIAL MOUNTAIN WHITEFISH (Prosopium williamsoni) IN T H E UPPER FRASER RIVER: A MORPHOLOGICAL, BEHAVIOURAL, AND GENETIC COMPARISON OF FORAGING FORMS.  PETER M . TROFFE B . S c , University of British Columbia, 1994 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS F O R THE D E G R E E OF M A S T E R OF SCIENCE IN THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Zoology)  We accept the thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A April 2000  © Peter M . Troffe, 2000  In  presenting  degree  this thesis  in  partial fulfilment of  requirements  for  an  advanced  at the University of British Columbia, I agree that the Library shall make it  freely available for reference and study. copying  the  I further agree that permission for  of this thesis for scholarly purposes  department  or  by  his  or  her  representatives.  extensive  may be granted by the head of It  is  understood  that  copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of The University of British Columbia Vancouver, Canada  DE-6 (2/88)  Abstract: Members of the family Coregonidae are notoriously plastic in their morphology and life histories, but in British Columbia there is little evidence of the kind of variation in trophic structures seen elsewhere in North America and Europe. There is, however, one exception — the mountain whitefish, Prosopium williamsoni. Museum, and field collections of fluvial mountain whitefish from the upper Peace, Columbia, and Fraser river systems commonly contain two sympatry phenotypes of fluvial mountain whitefish. One form (the most common) is characterized by a short blunt snout while, the other form has a long slightly upturned snout. I refer to this latter from as the 'pinocchio' form. Individuals with the pinocchio nose are not confined to British Columbia but are also known from isolated populations in Utah, the upper Missouri system, and the Olympic Peninsula in Washington State. The phenotypes from the Upper Fraser system differ in morphological features usually associated with trophic adaptations including gill raker counts and cranial architecture. Furthermore, the forms exhibit different foraging behaviours in sympatry, suggesting they occupy different foraging niches. A mitochondrial D N A survey reveals that pinocchio and normal mountain whitefish from Upper Fraser River tributaries have significantly different haplotype frequency distributions. The nature of the haplotype variation suggests either asymmetrical reproductive isolation — with normal males avoiding pinocchio females — or strong selection against hybrid progeny.  iii  Table of Contents  Abstract..... List of Figures List of Tables Acknowledgements  ii iv v vi  1.0  CHAPTER I - Introduction  1  2.0 2.1  CHAPTER II - Nature of the variation Methods 2.11 Characters 2.12 Assumptions and Size Adjustment 2.13 Distribution of residuals 2.14 Multivariate Analysis 2.15 Gill rakers Results 2.21 Character residual distribution 2.22 Multivariate Analysis 2.23 Gill rakers  6 7 8 9 9 10 11 11 11 12  2.3  Morphometric survey (conclusions)  14  3.0 3.1  3.2 3.3  CHAPTER III - Genetic survey Methods 3.11 D N A extraction 3.12 Mitochondrial D N A analysis Results Genetic survey (conclusions)  15 15 16 19 20  4.0 4.1 4.11 4.2 4.3  CHAPTER IV - Foraging observations Methods Statistical analysis Results Foraging observations (conclusions)  21 22 24 27  5.0  CHPATER V - General Discussion  28  2.2  Literature cited Appendix 1 Appendix II  57 65 68  List of Figures: Figure 1: Distribution of mountain whitefish in North America  38  Figure 2: Head profiles of normal and pinocchio mountain whitefish phenotypes  39  Figure 3: Upper Fraser River, including Prince George, B C  40  Figure 4: Head vs. standard length regression residuals, Swift River  41  Figure 5: Maxillary vs. standard length regression residuals, Swift River  42  Figure 6: Snout vs. standard length regression residuals, Swift River  43  Figure 7: Distribution of head vs standard length residuals, Swift River  44  Figure 8 Distribution of maxillary vs standard length residuals, Swift River  45  Figure 9: Distribution of snout vs standard length residuals, Swift River  46  Figure 10: Scatter plot of Swift River cranial character principal component analysis....47 Figure 11: Hierarchical cluster analysis of Swift River principal component values  48  Figure 12: Scatter plot of combined drainage principal component analysis  49  Figure 13: Hierarchical cluster analysis of combined drainage P C A  50  Figure 14: Gill raker counts from Swift River mountain whitefish phenotypes  51  Figure 15: Gill raker counts from four upper Fraser River tributaries  52  Figure 16: Mitochondrial D N A haplotype arranged by life history and collection sites...53 Figure 17: Mitochondrial D N A haplotypes arranged by phenotype  54  Figure 18: Foraging preferences of mountain whitefish phenotypes in Dome Creek  55  Figure 19: Foraging rates of mountain whitefish phenotypes in Dome Creek  56  V  List of Tables: Table 1: Upper Fraser River mountain whitefish included in morphometric survey  34  Table 2: Upper Fraser River mountain whitefish included in gill raker survey  34  Table 3: Comparison of gill raker counts of mountain whitefish phenotypes  34  Table 4: A N O V A -gill raker counts among drainages and phenotypes Table 5: Tissue samples used in mitochondrial D N A analysis Table 6: Distribution of mountain whitefish mitochondrial D N A haplotypes  35 35 36  Table 7: Mitochondrial D N A haplotypes arranged by mountain whitefish phenotype.. ..36 Table 8: Foraging preferences of mountain whitefish phenotypes (Top site)  37  Table 9: Foraging preferences of mountain whitefish phenotypes (Bottom site)  37  vi  Acknowledgements: I would like to thank all of my committee members for their patience over the last four years. M y advisor Dr. J.D. McPhail sowed the seeds for this thesis and continued to be very generous with his time, and vast scientific experience. Dr. E.B. Taylor provided all aspects of assistance in the genetics lab. He was able to convince me that I should " put down the quill pen, and jump into the world of molecular markers". Many thanks to Dr. Peter Mylecreest and Ian Kusabs for their help in the sometimes-trying field conditions. Peter provided this study with excellent photographs and he was instrumental in the collection of many mountain whitefish specimens. Dr. C C . Lindsey was generous with his whitefish experiences during his luncheon rounds at the U B C ichthyology collections. Additional tissue samples from outside the Fraser Basin were provided by James Baxter, David O'Brien, and E.B. Taylor. The maps of mountain whitefish distribution in North America and mountain whitefish head diagrams are after Diana McPhail's drawings. The Natural History Curators at the Royal B.C. Museum provided valuable support during the final writing.  CHAPTER I 1.0 Introduction: The study of phenotypic, genetic and behavioural variation is a fundamental component of evolutionary biology. Although virtually any aspect of an organism's phenotype can vary, morphological variation often is the most conspicuous and easily assayed form of variation. In North Temperate freshwater fish, intraspecific variation in morphology is especially noticeable and well documented. Individuals within a population of the same size and sex normally vary in their body proportions and meristic counts, but in some cases, the phenotypic variation is strikingly discontinuous. This latter situation is the subject of this thesis. Among freshwater fish, discontinuous variation in structures associated with foraging are common (e.g. Robinson and Wilson 1996; Meyer 1989; Liem 1973) and found in a wide range of phyletic groups distributed throughout both the northern and southern hemispheres (Ruzzante at al. 1998; Taylor and McPhail 1999; Schluter 1996; Northcote 1988). Most examples of discontinuous morphological variations involve structures associated with foraging ecology such as gill rakers, mouth size, cranial architecture and dentition (Galis and Drucker 1996; Skulason and Smith 1995; Taylor and Bentzen 1993; Galat and Vicinich 1983). In some cases where discontinuous morphological variation is observed there are clear genetic or inherited differences between two or more sympatric forms (e.g. Skulason at al. 1993, 1989; Taylor and Bentzen 1993; Ferguson and Mason 1981). In other cases, the variation is present in the absence of detectable genetic differences and variations in morphology are thought to  2  represent examples of environmentally induced phenotypic plasticity (Meyer 1989; Liem 1973; Hindar and Jonsson 1992; Jonsson and Hindar 1982). One group of fish in which the presence of different trophic morphs is especially common in are the whitefish (Coregonidae). Coregonids are a variable, Holarctic family of freshwater and anadromous fish that are distributed primarily in glaciated areas. Whitefish are notoriously variable in their morphology and life histories and, so far, have eluded the attempts of taxonomists to establish clear species boundaries within the family (Svardson 1979; Lindsey 1981; Bernatchez at al.. 1999). In British Columbia there appear to be ten species of whitefish but, surprisingly, there is little evidence of the kind of discontinuous variation in trophic structures seen elsewhere in North America and Eurasia. There is, however, one noticeable exception — the mountain whitefish, Prosopium williamsdni. Prosopium differs from other genera of whitefish in that they are adapted primarily for life in rivers rather than lakes, and they reach their highest diversity in western North America (Norden 1970; McCart 1970). Among the North American species of Prosopium the mountain whitefish has the widest geographic distribution and occupies a variety of aquatic habitats — large silty rivers, small clear headwater streams, and small to large lakes. Mountain whitefish are distributed along the eastslope of the Sierra Nevada in California and Nevada, they are present in internal drainage systems in Utah, Idaho and Oregon, and are widespread in the Columbia, Fraser, and Skeena systems as well as in rivers draining the eastslope of the Rocky Mountains in Wyoming, Montana and Alberta (Fig. 1). They also occur in the Mackenzie system (Athabasca, Peace and Liard drainages) and extend down the mainstem Mackenzie River at least as far as its confluence with the  3  Bear River. Throughout this extensive geographic range the mountain whitefish is an interior species and only approaches the coast in a few large rivers that have cut through the coastal mountains (e.g., the Fraser, Skeena, and Stikine rivers). The species is absent from Vancouver Island and from rivers that rise in the coastal mountains. Again, however, there is an exception: mountain whitefish are abundant in the short coastal rivers that drain the west slope of the Olympic Peninsula (Fig. 1). The significance of the Olympic populations, and the presence of isolated populations in arid areas like eastern California and Utah, is that they establish that mountain whitefish likely survived glaciation in more than one glacial refugium (McPhail and Lindsey 1970, 1986). Thus, British Columbia may have been colonized postglacially by more than one isolated or genetic form of mountain whitefish. Relative to trout and salmon the life history of this species is not well known; however, it is clear that fluvial populations perform complex reproductive, overwintering, and feeding migrations (Davies and Thompson 1976; Northcote and Ennis 1994; McPhail and Troffe 1998). In rivers, the diet of both juveniles and adults consists mostly of the larvae and nymphs of aquatic insects. Unlike trout, mountain whitefish rarely take terrestrial insects and appear to feed mainly on drift and benthic materials (McHugh 1940). Observations of several fluvial mountain whitefish specimens from the upper Peace, Columbia, and Fraser systems housed in the University of British Columbia ichthyology museum collection suggests the presence of phenotypes that differ in cranial architecture, most noticeably snout length. One form (the most common) is characterized by a short, rather blunt snout. I refer to this form as the "normal" form (Fig.2). The  4  alternate form, occurs at approximately a 20% frequency in the upper Fraser River system, and has a long, slightly upturned snout (Fig. 2). The elongation of the snout becomes progressively prominent with increasing specimen size, and I refer to this long snout morph as the "pinocchio" form. Individuals with the elongate snout are not confined to British Columbia but are also known from isolated populations in Utah (Stalnaker at al. 1974), the upper Missouri system in Wyoming (Evermann 1892), and the Olympic Peninsula in Washington (P. Mongillo, Washington Dept. of Fish and Game, pers. comm.). The function of the elongate snout of some fluvial mountain whitefish populations is unknown and it is unclear if this characteristic is an inducible, polymorphic feature or if it has a heritable genetic foundation (e.g. Meyer 1989; Lindsey 1981). Stalnaker at al. (1974) suggested the long rostrum of some mountain whitefish is a male secondary sexual characteristic. This seems unlikely since, in museum collections, the trait occurs in adults of both sexes, and my field collections, made throughout the spring and summer contained long snouted individuals of both sexes, even during the fall breeding season. Thus, there is no evidence that the elongate snout is sexually dimorphic or associated with reproduction. Also, there are adults that are intermediate between the two forms. The first question I address in this thesis is — is there more than one morphological form of fluvial mountain whitefish in the upper Fraser Basin (i.e. is the variation in head morphology discontinuous or, represented by more than one clear group)? Given a positive answer to this question, I then asked if there are genetic differences between the forms using a survey of mitochondrial D N A restriction fragment length polymorphisms. Finally, because head shape and gill raker architecture differences suggest the possibility  5  of trophic differences, I address the question is there a relationship between phenotype and foraging behaviour?  6  C H A P T E R II 2.0 Nature of the Variation:  The first question posed in the introduction relates to the distribution of the morphological variation between and within fluvial mountain whitefish populations in upper Fraser River Basin, British Columbia. Specifically, is there more than one distinct group of mountain whitefish based on cranial characters commonly associated with trophic foraging efficiency, or is the morphological variation in these traits uniformly distributed? Furthermore, if morphological differences are revealed among or within populations are there any differences in gill raker number among phenotypes?  2.1 Methods  The data for my morphometric survey of mountain whitefish cranial characters came from 84 mountain whitefish specimens from three upper Fraser River tributaries. Adult specimens of similar standard lengths were used in the morphometric survey to minimize any potential differences in allometry (Table 1). The first two data sets are museum specimens; the Wright Creek, and Swift River specimens were collected during June, 1955 and June, 1956, respectively (BC 55-355, B C 56-344) and are currently housed within the University of British Columbia ichthyology museum. Wright Creek is a moderately slow, small clear water creek that meets the Salmon River near its confluence with the Fraser River, approximately 15 kilometeres north of Prince George, B C (Fig. 3). The Swift River is a fast water tributary of the Cottonwood River which has its confluence with the Fraser River approximately 80 kilometeres south of Prince George, B C (Fig 3). The presence of juvenile and adult  7  specimens in the Wright Creek, and Swift River collections (0+, 1+, 2+ and older) suggest that mountain whitefish were using these reaches as both foraging and rearing grounds during the 1955 and 1956 summer seasons. The third morphometric data set included in this survey of fluvial mountain whitefish cranial characters comes from mountain whitefish I collected in the autumn of 1995 at the confluence of the Nechako and Fraser rivers at Prince George, B C (Fig 3). Adult specimens were collected with 25-metere seine net sets at the rivers' confluence during seasonal reproductive congregations. The 25 specimens were individually labeled and preserved in 10% buffered formaldehyde solution for five days before they were fixed in 50% isopropanol for final storage.  2.11 Characters: Four characters were measured on each specimen: head length, maxillary length, snout length and standard length. Straight-line measurements were conducted with digital calipers accurate to the nearest 0.01 mm following the procedures of Hubbs and Lagler (1947). When necessary, finer measurements were made under the resolution of a dissecting microscope at three times magnification. A n estimate of measurement accuracy was made by remeasuring 25 Swift River specimens and comparing both the individual and total variation to original measurements in a one-way-ANOVA (Haas and McPhail 1991). Measurement error was low, with the variance ratio of original to the recollected measurements being close to one (0.98), suggesting that the measurement procedure is repeatable.  2.12 Assumptions and Size Adjustment: To be statistically robust multivariate morphometric analysis usually requires the number of specimens to exceed the number of characters used in the multivariate analysis (Reist 1985; Sharp at al. 1978). A total of 84 mountain whitefish form the three drainages were found to be suitable for the morphometric analysis (i.e. they were not twisted or warped during preservation). The morphometric measurements were collected in ASCII format and analyzed with Systat Version 8. Bivariate regressions of were made with LOGio transformed data (Zar 1984; Humphries at al. 1981; Bookstein at al. 1985). As indicated in the literature (Holt 1960; Scott and Crossman 1973) I assumed that the sex ratio of mountain whitefish is equal, stable, and that the species is not sexually dimorphic. There are colour and lateral tubercle differences between the sexes when they are in spawning condition, but these differences are slight and do not involve any of the characters I measured (Scott and Crossman 1973; McAfee 1966). Despite attempts to obtain specimens of equal size for comparison, specimen sizes varied both within and among populations and character measurements were correlated with specimen size and required size adjustment. The morphometric data sets were univariately adjusted for differences in allometry by regressing each of the three cranial characters against individual specimen standard length and collecting deviations from the regression line as residuals (Reist 1985). Thus, differences among character measurements are expressed as deviations (residuals) from the linear relationship describing the overall body size relationship for the character set. The residual variation results from two sources; character measurement error due to the observer, and biological effects representing the deviations of individual specimens from the overall size  relationship (Haas and McPhail 1991). Since assessment of measurement error was low, it can be assumed that the resulting residual variation represents a biological effect. The regression based morphometric technique of size adjustment is thought to be superior when compared to other univariate methods (e.g. Ratios - division of character by standard length) as it avoids problems resulting from non-linear relationships between the size adjusted and the original characters (Reist 1985; Haas 1988).  2.13 Distribution of residuals: Measurements of snout, maxillary and head length were taken on 38 Swift River specimens. The character measurements were regressed against individual specimen standard length as outlined in the aforementioned section outlining character size adjustment. The distribution of variance surrounding the regression line (residuals) of each bivariate plot was then represented in scatter plots and histograms to determine i f character residuals are uniformly or asymmetrically distributed about the mean regression line.  2.14 Multivariate Analysis:  Two separate variance-covariance principal component analysis (PCA) procedures were performed on the size corrected character matrices to determine if the cranial characters are represented by more than one morphometric group (Winans 1984; Reist 1985; Haas 1988). The first P C A examined the 38 Swift River specimens, while the second analysis involved a combined 84 specimens from all three drainages (Table 1).  10  After collecting the P C A loading scores, a hierarchical cluster analysis using Systat V . 8 was performed on first principal component loading scores (principle component expressing the majority of explained variance) from the Swift River and Combined drainage PCA. The cluster analysis was used to determine i f any homogenous groups could be assigned based on the measured characters in a method similar to Taylor and Bentzen (1993). Hierarchical cluster analysis begins with each individual case in a separate cluster and then combines clusters sequentially using a least squares method, reducing the number of cluster groups until the variance is minimized.  2.15 Gill rakers: Gill raker counts were made after dissecting the left primary gill arch of 55 adult mountain whitefish from four upper Fraser tributaries (Table 2). The specimens were sorted into morphological groups as suggested by the initial P C A and cluster analysis before the counts were made. A l l gill raker counts were made under a dissecting microscope (3x) and counts include all rudimentary rakers (McPhail 1992). To test for effects of locality and phenotype, gill raker counts were compared against each other with a two-way-ANOVA. Morphotype comparisons within a single drainage were compared univariately with a two-tailed student's t-test (assuming equal variance).  2.2 Results 2.21 Character residual distribution: The distribution of the head character residuals about the mean regression line suggests there are two separate groups of mountain whitefish within the Swift River collection. Each of the regressions (snout, maxillary and head length) exhibited high squared multiple R-values (Maxillary =0.809, Snout =0.816, Head =0.941) suggesting the regressions against standard length were robust. The distribution of each residual data set is asymmetrical when plotted against specimen standard length. In each case positive residual values cluster together in a group while the negative values were more widely distributed (Figs. 4, 5, 6). Furthermore, frequency plots describing the distribution of the residuals for each of the three adjusted characters about the mean regression line are represented by strikingly disrupted distributions, with the lowest frequency counts occurring near the mean regression line (Figs. 7, 8, 9).  2.22 Multivariate Analysis: The total explained variance in the Swift River P C A was 99%, with 94.8%, and 4.5% variance accounted for by the first and second principal components, respectively. This high level of explained variance suggests that inference made from this analysis is robust (Winans 1984). The component loading values were highest for snout length (0.973) followed by head length (0.960) and maxillary length (0.957). A scatter plot of principal component Factor 1 against Factor 2 suggests the data are represented by two distinct clusters with most of the separation being attributed to differences along the first principal component axis (Fig. 10). Hierarchical cluster analysis performed on the factor  1 principal component loading values separates the data into two distinct clusters (Fig. 11). The division between the two clusters occurs at the first branch of the cluster tree (Euclidean distance = 0.7). Scatter plots of the P C A loading including the combined specimens from Wright Creek, and the Swift and Nechako Rivers are similar to the single drainage Swift River PCA. The total explained variance in the combined drainage P C A was 99%, with 94.8%) and 5% variance accounted for by the first and second components respectively. Two clusters are represented in the PCA, separated by differences along the first principal component similar to the single drainage P C A (Fig. 12). Hierarchical cluster analysis performed on the first principal component loading values divides the specimens into two distinct clades at the first branch of the cluster (Euclidean distance = 0.35) (Fig. 13).  2.23 G i l l raker counts: Frequency plots of gill raker counts of pinocchio and normal fish from the Swift River specimens, and data from combined drainages, reveal that individuals with long snouts have on average more gill rakers than sympatric short nosed individuals. The differences in gill raker number between pinocchio and normal fish are represented by a disrupted distribution rather than merely describing differences along both tails of a single normal distribution (Figs. 14, 15) (Lindsey 1981,1988). Although there is variation among drainages, there are consistent differences in gill raker number between pinocchio and normal mountain whitefish (Figs. 14, 15). Twotailed t-tests show that, regardless of drainage, pinocchio mountain whitefish on average  13  have a significantly higher (22.4 ± 0.27) gill raker count than sympatric normal whitefish (20.1 ±0.3) (Table 3). Independent of drainage designation, analysis of variance indicates consistent differences in gill raker counts between pinocchio and normal fish (Table 4). There are also significant differences in the mean gill raker count between drainages independent of mountain whitefish phenotype (Table 4).  14 2.3 Morphometric survey (conclusions) The variation in cranial morphology and gill raker count suggests the presence of two distinct mountain whitefish morphotypes in the upper Fraser River, B.C. The distribution of the residuals from regressions of maxillary, head and snout lengths against individual specimen standard length also indicates that there are two morphotypes of mountain whitefish. In all characters examined, the magnitudes of residuals above and below the regression line, was higher than along the mean regression line where they would be expected if the residuals were distributed normally. Hierarchical cluster analysis of P C A loading scores from a single, and three combined drainages confirms the results examining the character regression residuals by assigning mountain whitefish to one of two constructed clades. Although there is variation in mountain whitefish gill raker number between upper Fraser Basin tributaries, there are consistently significant differences between the long snout 'pinocchio' and blunt snout 'normal' morphotypes of mountain whitefish. In all four drainages where gill raker number was surveyed pinocchios had, higher gill raker counts than their normal snouted sympatric counterparts. The significance of these differences in gill raker architecture will be discussed below. The first question posed in my introduction to this thesis was — Is there evidence of discontinuous morphometric variation in fluvial mountain whitefish populations in the Upper Fraser system? Clearly, the answer to this question is — Yes. The fact that different phenotype can be maintained in sympatry suggests there could also be genetic and behavioural differences among upper Fraser River mountain whitefish populations.  15  C H A P T E R III 3.0 Genetic survey: The second question posed in my introduction concerns the genetic relationship between the morphotypes outlined in the previous chapter. This question was addressed by survey of a hypervariable portion of the mitochondrial D N A genome of mountain whitefish in the upper Fraser River and adjacent drainages.  3.1 Methods In 1995, mountain whitefish tissue samples were collected from six upper Fraser tributaries, the Fraser mainstem, two drainages in the Peace system, and the Duncan River in the Columbia system. The tissue samples were obtained from both adult and juvenile fish collected by pole seine, beach seine, trapping, electroshocking and angling (Table 5). Juvenile fish were slit ventrally and stored whole in 95% EtOH. Adult fish were sorted according to morphology, their adipose fins clipped, and the tissues stored in 95% EtOH. An RFLP (restriction fragment length polymorphism) analysis was performed on the tissues samples to assess mitochondrial D N A variability among mountain whitefish in the upper Fraser Basin.  3.11 D N A Extraction: Mitochondrial D N A was extracted using a modified version of the Chloroform/Phenol procedure. Liver, heart, and peduncle muscle were dissected from preserved juvenile specimens. Whole adipose fins from adult specimens and 20-50 mg of tissue from juvenile specimens were blotted dry, and macerated before extraction. Weighed tissue samples were digested overnight in a buffered Pronase solution. The  16  digestion temperature was 37° C. Samples that were poorly digested after 8-10 hours were reincubated and, in some cases, additional aliquots of Pronase were added until the samples appeared completely digested. Aliquots of RNAse were added to the fully digested tissue samples, and the samples were then incubated at 37° C for one to three hours. Equal volumes of phenol and chloroform were added to the digested samples. They were then centrifuged and the aqueous phase removed in preparation for final D N A precipitation. Cold isopropanol was added to the aqueous extractions, they were then gently mixed and held at -20° C for 20 minutes to maximize the yield of D N A precipitate. The precipitated samples were centrifuged, the isopropanol aspirated, and the D N A pellets washed for 1-3 hours in ice cold 70% lab grade EtOH. After alcohol washing the samples were re-centrifuged, the EtOH aspirated, and D N A pellets resuspended in 75-150 ul of TE buffer (pH 8.0) and stored at -20 C. 0  The D N A concentration of the extracts was determined with a spectrophotometer. A volume of 3ul of the resuspended D N A precipitate solution was diluted in 247p.l of 0.5 x TBE buffer (pH 8.0). The diluted suspension was compared to a control standard in a Pharmacia spectrophotometer and D N A concentrations were calculated and recorded.  3.12 Mitochondrial D N A analysis: Dilution trials and reaction conditions were standardized and refined for the Polemerase Chain Reaction (PCR) amplification of the combined Cytochrome b, and Dloop portion of the mountain whitefish mtDNA genome. These portions of the  17  mitochondrial genome were surveyed because they are considered highly polymorphic and are easily assayed (Hall and Nawrocki 1995). To define a standardized reaction condition that yielded consistent, high quality chain replicated products, trial PCR reactions were conducted at a variety of initial concentrations of D N A and PCR reagents. The primers used in the PCR reaction are after Bernatchez and Osinov (1995); and 0.4 units of Taq Polymerases was used per 25 pi PCR reaction. The final concentrations of the standard reaction consisted of:  - O.lpg/ul of raw D N A extract - 0.8mM dNTP nucleotide mix - 0.6mM of HN20, CGlu (Primers A,B) - 2.0 p/ul Taq polymerase (added last) - 2.0mM Magnesium Chloride The D N A extract and the PCR reagents were combined and the reaction conducted in a Robocycler unit via the hot start method. The mtDNA loci were chain replicated for 30 cycles with each of the denaturing, annealing, and extension phases continuing for 1.5 minutes at 95, 55, and 72° C, respectively. After 30 amplification cycles the samples were held in a single extension phase for 5 minutes before final storage at 6°C. PCR products were qualified by staining aliquots of product with 1.5% Ethidium Bromide/ loading buffer solution. The stained products were electophoretically separated by running them between 60-75 volts for 1 to 1.5 hours on a 1% agarose gel loaded with a lkilobase (Kb) molecular weight standard ladder. The separated gel was placed under ultra-violet light and the fluorescent profiles photographed on Polaroid film.  18  As recommended by the vendor, (New England Biolabs) aliquots of high quality PCR products were digested in individual vials with one multi-hexameric (Sty /restriction endonucleases. The digested fragments were combined with loading buffer, loaded into, and electrophoretically resolved on, 1.5 % agarose gels immersed in T B E buffer at between 75 and 90 volts for 2.5 to 3.5 hours. Each gel was run with a 1Kb ladder as a molecular weight standard. The separated gels were stained for 20 minutes in an ethidium bromide solution, washed in 0.5x T B E buffer for a further 20 minutes and finally illuminated with U V light for Polaroid photography. A random selection of five samples were re-amplified via PCR, and re-digested to determine if there is any variability in the RFLP procedure. There were no detectable differences in the RFLP trials. This indicates the procedure is repeatable. Distinct endonuclease genotype (haplotype) patterns were identified and given composite codes. In an attempt to understand the distribution of mountain whitefish mtDNA haplotype frequencies in the upper Fraser system the haplotype patterns were grouped according to their collection location and life-history stage (Table 6). Contingency tests were conducted using the M O N T E algorithm in the R E A P statistical package to determine i f there was any association between fish morphology and haplotype frequency (McElroy at al. 1992). Using a priori knowledge the contingency tests were blocked into two groups: a universal sample containing all pinocchio and normal mountain whitefish tissue collections (Bowron, Willow, McGregor and Nechako rivers), and a localized comparison from the Nechako River where fish of both morphotypes were collected at the same time and place.  19  3.2 Results: The RFLP survey (PCR products cut with Styl) revealed three haplotype patterns. A l l of the mountain whitefish populations examined from the Peace (Chowade and Burnt rivers), Columbia (Duncan River), and upper Fraser rivers possessed some mix of these mtDNA haplotypes, however, the Burnt and McGregor rivers appear to contain only patterns A and C (Table 6). Haplotypes A and B were found throughout the entire sampling area (Table 6). The frequencies of these two common haplotypes (A and B) are relatively homogeneous throughout the upper Fraser and account for 97% of the mtDNA variation in upper Fraser mountain whitefish populations (Fig. 16). In the Nechako system, contingency tests indicate that there is a difference between the mtDNA haplotype frequencies found in pinocchio and normal whitefish (X = 8, p<0.005). Samples from the Nechako River show that normals are exclusively haplotype A , while pinocchios possess both patterns (A and B) at about equal frequency (Fig. 17; Table 7). This difference in haplotype frequencies between pinocchio and normals also occurs in the combined samples from the Bowron, Willow and Nechako rivers (Fig. 17; Table 7). The haplotype distribution in juvenile fish is similar to the distribution in adults suggesting that, although the individuals are too small to be identified by their phenotype, both pinocchio and normal haplotype patterns are present in juvenile samples (Fig. 16).  20  3.3 Genetic survey (conclusions): The mitochondrial RFLP evidence implies that mountain whitefish populations in British Columbia are not a genetically homogeneous group. The haplotypes present in the upper Fraser system appear to be shared amongst all the tributaries but the relative frequency of the haplotypes differs among localities. This suggests that there are some genetic differences among upper Fraser mountain whitefish populations. The haplotypes, and their frequencies in the pinocchio and normal morphotypes, are significantly different. Haplotypes A and B are equally frequent in pinocchios; while normal fish are almost exclusively haplotype A and never haplotype B. The sample size is small (14 pinocchio and 15 normals); nevertheless these haplotype differences imply that the pinocchio and normal populations belong to different reproductive groups. Thus, although it is not clear what mechanisms maintain these haplotype differences in sympatry, it is clear that there are genetic differences between the two morphotypes. Consequently, the answer to the second question posed in the introduction — are there genetic differences between the morphotypes? — is yes. What this means is discussed in the final chapter.  21  C H A P T E R IV 4.0 Foraging observations: The third and final question posed in the introduction concerns the functional significance of different fluvial mountain whitefish morphotypes. This question was addressed through in situ field observations on the foraging behaviour of the two morphs in Dome Creek, B.C.  4.1 Methods In late August 1996, two days of foraging observations were conducted on pinocchio and normal mountain whitefish in Dome Creek, a small tributary of the upper Fraser River (Fig.3). Seasonal surveys of Dome Creek suggested that this tributary, like many in the region, is used as both a summer foraging area, and a reproductive site by both pinocchio and normal mountain whitefish. The hydrology of Dome Creek is highly variable, with variable flow regimes and low visibility for the majority of the ice-free season. During select periods in early August thorough to September the water levels decrease and the water visibility improves allowing for safe snorkel observation. Under-water snorkel observations focusing on mountain whitefish foraging behaviours were made at two sites (designated Top site and Bottom site). At both sites adult pinocchios and normals were distinguishable under water at distances of up to 10 meters. Observations were conducted for 5 minute periods. Focal individuals alternated between pinocchio and normal whitefish. Approximate standard lengths of focal animals were determined by comparing them to two mountain whitefish that had been angled, measured and tagged two weeks earlier. The whitefish showed no obvious fright response  22  to the presence of a snorkeling observer and fish continued to forage until the observer approached closer than 2 meters. Even then, only sudden movements by the observer displaced the fish. The turbidity and flow regime of Dome Creek allowed for only two days of foraging data collection, on a total of 13 mountain whitefish; eight normal and five pinocchio adult whitefish. The low density of larger fish at the top and bottom observation pools did not allow for larger sample sizes. The number, and type, of foraging attempts made by the focal animal in a five minute period were recorded with a soft pencil on a polyvinyl slate. Foraging attempts were categorized as either benthic or in the water column. Benthic foraging attempts occurred when fish foraged actively amongst, or on, the substrate. Water column attempts were recorded when fish were seen to make foraging movements directed at items suspended in the moving water column. The frequency of unique benthic foraging behaviours (e.g. rock-rolling and substrate burrowing) were recorded along with the two standard foraging categories. The success rate during the foraging trials is unknown but it was assumed that that not all foraging attempts were fruitful.  4.11 Statistical analysis Foraging observations conducted on pinocchio and normal whitefish were standardized by observing individuals of both forms in alternate trials of equal duration. The estimated size range of observed animals was similar for both forms and the number of observation periods on pinocchio and normal fish at each site was nearly equal (Tables 8, 9). Mann-Whitney U-tests were performed on the foraging data to determine i f pinocchio and normal fish have different foraging rates for benthic and water column  23  prey. The number of benthic and water column foraging attempts were compared for each observation site individually, then the data were pooled to test for overall rate differences:  24  4.2 Results: In situ snorkel observations conducted in Dome Creek during August 1996 confirmed that pinocchios and normals forage together at the same time but in different ways. At each observation site pinocchios foraged alongside normal fish, however, the morphotypes exhibited different foraging behaviours. Normal mountain whitefish foraged only on drift in the water column. Pinocchio forms also foraged on drifting prey suspended in the water column but, in addition, they foraged by burrowing amongst the gravel substrate with their snouts (Table 8, 9; Fig. 18). Normal mountain whitefish oriented themselves on a plane horizontal to the substrate and about 30 cm above the bottom. They hold themselves in the current with tail movements and hold their pectoral fins against the body. Movement was limited, and most of the normal fish did not move more than a couple of meters during the five minute observation periods. A l l of the foraging attempts by normal mountain whitefish in both observation pools were made in the water column and directed at drift items the fish appeared to identify visually. Most foraging attempts involved fish moving to, and then striking at, items drifting no more than 30 cm from their holding position. After a strike attempt these fish returned to a horizontal holding posture facing into the current. Pinocchios appear to be more benthically oriented than normal fish. They hold in the current about 20 cm above the substrate but, unlike normals, they did not remain in a single holding area during the observation trials. Instead pinocchios moved, stopped and moved again in a saltatorial pattern. While holding, and cruising over the substrate, pinocchios extend their pectoral fins away from their bodies and tilt their heads towards the substrate.  25  Approximately one-half of the foraging attempts made by pinocchios were directed at drifting items suspended in the water column while they held themselves into the current like normal fish (Tables 8, 9; Fig. 18). Unlike the normal fish, however, pinocchios also actively foraged in and amongst the substrate (Tables 8. 9; Fig 18). Thus, pinocchios appeared to forage in two modes: one involves holding a position and focusing on foraging for drift with occasional probes at the substrate; the other mode occurred during saltatorial movements and seemed to be oriented solely at the bottom. Occasionally, while in a holding position, pinocchios would turn their eyes away from the oncoming current and, looking at the substrate, they would sink their snouts into the gravel and burrow through the gravel, or wedge their noses between interstices among rocks while thrusting with their caudal fins. Many pinocchios were seen "coughing-up" small gravel and sand after such benthic foraging episodes. When moving between holding positions pinocchios cruise over the substrate with their eyes oriented towards the bottom and slow to a stop before burrowing into the gravel. During several bottom foraging episodes, pinocchios used their elongated snouts to flip and roll rocks then struck quickly at the freshly exposed substrate. Mann-Whitney U - tests comparing the number of benthic foraging attempts normal and pinocchio fish made at both observation sites are significantly different (both U o.io(i) 4 , 3 12; U o.io(i) 4,2 8 ; U> 15 for both sites) (Tables 8, 9; Fig 18). There were =  =  significant differences (U o.os(i) 8,5 =32; U> 37) in the total number of water column foraging attempts made in the top pool, when compared to the bottom pool (Fig. 18). Pinocchios directed more foraging attempts at the benthos in the top pool, and picked more drift in the bottom pool. This suggests that pinocchios can exhibit slightly different  foraging behaviour in different habitats. Foraging rates of normal fish were similar at each observation site, and usually slightly elevated compared to their pinocchio counterparts (Fig. 19).  27  4.3 Foraging observations (conclusions): Observations of foraging behavior under natural conditions in Dome Creek, combined with the morphometric differences suggest that pinocchio and normal mountain whitefish have different foraging ecologies while in sympatry. The two forms of mountain whitefish did not differ in their location within the observation pools and did not appear to hold foraging territories. Normal fish made more foraging attempts than pinocchios, but the differences are slight and any differences between the morphs in foraging rates may be attributed to differences in foraging behaviours. The movements of the pinocchio fish were saltatorial, and directed at both the substrate and water column thereby affording less time for drift feeding than the more stationary and exclusively drift-feeding normal mountain whitefish morph. The elongate snout of the pinocchios was used to flip rocks, burrow into the substrate, and poke into the interstices between larger rocks. These benthic-oriented behaviours were not observed in the normal form of fluvial mountain whitefish.  28  CHAPTER V 5.0 General Discussion: Coregonid fish were subjected to multiple bouts of range fragmentation, divergence, and range expansions during the Pleistocene glaciations (Lindsey 1981; McPhail and Lindsey 1986; Bernatchez and Dodson 1991). Many species show unique local divergences, and there are several examples of multiple forms reported from the same lake (Bodaly 1979; Bird 1979). Most of these investigations have focused on the genus Coregonus and little attention has been directed at variation in the genus Prosopium. Members of Prosopium are distributed throughout Siberia and western North America and the oldest known fossils (Miocene) are from deposits in southwest Idaho and northeast Oregon (Smith 1975). This suggests that the genus probably has its evolutionary roots in western North America rather than in Siberia (Norden 1970; Vuorinen 1998). The mountain whitefish, Prosopium williamsoni, is considered by morphological and molecular analyses to be the least derived of the six known species within Prosopium (Norden 1970; Sajdak and Phillips 1997; Vuorinen 1998). Curiously, all recent morphological accounts of this species ignore the long snouted (pinocchio) form of mountain whitefish and treat the species as if it were homogenous throughout its wide western North American distribution. In contrast, older accounts (e.g. Evermann 1892; Jordan and Snyder 1909) recognized the long snouted morph and emphasized morphological variation within the species. The first question posed in the introduction of this thesis concerns the nature of the morphological variation of fluvial mountain whitefish observed in upper Fraser River system -— are there discrete forms of mountain whitefish in the system, or is the variation  29  continuously distributed? The P C A conducted on size-adjusted Swift River and two other upper Fraser tributary specimens suggests that the presence of two groups of mountain whitefish based on head length, maxillary length, and snout length. Hierarchical cluster analysis of the first principal component loading scores identifies two phenotypes of upper Fraser mountain whitefish. The phenotypic differences vary between collection sites, but one form has consistently longer cranial features than the other group. Whatever the cause, the consistent differences in head shape and mouth suggest differences in either microhabitat use or foraging behaviour in fluvial environments (e.g.; Liem 1973; Bodaly 1979; Taylor and Bentzen 1993; Gailis and Metz 1998). In fish, differences in gill raker number often reflect differences in trophic ecology, especially in coregonid species (Bodaly 1979; McPhail 1992; Gardener at al.. 1988). My data indicate that pinocchios have a higher gill raker counts than normal whitefish. It is unclear whether these differences in gill raker counts are an effect of exposure to different micro-habitats early in life or if there are heritable gill raker differences between the pinocchio and normal forms; however, most coregonid taxonomists consider differences in gill raker numbers to be a reliable indicator of genetic differences among whitefish (Svardson 1957; Lindsey 1981). Thus, these differences in cranial features, in conjunction with the disrupted distribution in gill raker counts, suggest that the two forms of mountain whitefish in upper Fraser River tributaries differ in their trophic ecology. The second question posed in the introduction concerns the genetic structuring of mountain whitefish in the upper Fraser system, specifically — are there genetic differences between the phenotypes? Again, the answer is yes.  30  The haplotype data compiled from mitochondria D N A restriction fragment length polymorphisms in upper Fraser River mountain whitefish indicate that the fish from different locations vary in haplotype frequencies. The presence of this diversity in the mitochondrial genome implies that these populations have been subjected to historical restrictions in gene flow that arose as a result of founder effects or population bottlenecks (Bernatchez and Dodson 1990, 1991). Interestingly, over ninty-seven percent of mountain whitefish mtDNA haplotype diversity in the upper Fraser is attributable to two haplotypes — A and B. The frequencies of these haplotypes appear to be fixed in the upper Fraser system, since they are independent of life-stage and location. Mitochondrial D N A haplotype frequency distributions also are significantly different in the two morphological forms — pinocchios and normals. Haplotypes A and B are equally represented in pinocchio fish, while normal fish were almost exclusively Haplotype A and never contained haplotype B. This difference in mtDNA haplotype frequencies in pinocchios and normals suggests the two forms are partially reproductively isolated even though they occur together in spawning congregations; Since mtDNA is passed mainly through the maternal line, crosses between normal males and pinocchio females should yield some normal fish with the B haplotype. Thus, the absence of haplotype B in the normal form implies either asymmetrical reproductive isolation — with normal males avoiding pinocchio females (positively assorted mating) — or strong selection against hybrid progeny of normal male and pinocchio female matings. Examples of reproductive isolation are common among closely related fish and amphibia taxa (Michalak and Rafinski 1999; Dgebuadze at al. 1999; Rocha-Olivares at  31  al. 1999). Most examples of reproductive isolation in fish occur in north temperate, species poor, lacustrine environments and involve morphological features associated with foraging behaviours (Wimberger 1994). The genetic data suggests the pinocchio and normal mountain forms of whitefish represent a possible example of asymmetric reproductive isolation in a fluvial environment. Further research is needed before any more speculation surrounding mountain whitefish reproductive isolation is disscussed. The third, and final question posed in the introduction concerns the function of the morphological differences observed in upper Fraser system mountain whitefish populations. Observations in Dome Creek clearly show that in sympatry the long-snouted pinocchio and blunt-snouted normal forms of mountain whitefish have different foraging behaviours. When foraging, normal mountain whitefish are less motile than pinocchios foraging exclusively on water-born drift. In contrast, pinocchios cruise the substrate in a saltatory fashion, poking their elongated snouts into interstices and often rolling over rocks. Pinocchios direct their foraging attempts about equally to the benthos and the water column, but their over-all foraging rate is similar to the normal fish. The differences between pinocchios and normals in body position and pectoral fin extension when holding in the current suggest that pinocchios are inducing downward forces on the anterior end of their bodies. This benthically oriented body angle, as well as differences in head shape, may produce differences in near point focus. Thus, allowing pinocchios to search the substrate for prey items while remaining in a position to forage on passing drift. The morphological differences, especially in rostral length and gill raker architecture, combined with the striking differences in their foraging tactics suggest that  32  these fluvial forms of mountain whitefish — pinocchios and normals — use overlapping, but slightly different foraging niches (Fausch at al. 1997; Skulason and Smith 1995). The combined evidence from my morphometric, genetic and foraging behaviour investigations of mountain whitefish suggest that sympatric pairs of whitefish are not restricted to lacustrine environments. The mountain whitefish in the upper Fraser system are functionally dimorphic and, although there is molecular evidence supporting reproductive isolation between the morphs, it is unclear how this dichotomy is maintained. It is clear, however, that the observed differences in mountain white fish are not exclusive to the Fraser Basin. The two forms occur throughout a large portion of the species' range and the pinocchio form bears a striking resemblance to Coregonus oregonius, a species from the McKenzie River, Oregon proposed by D. S. Jordan and J. Snyder in 1909. While it is clear that there are two forms of mountain whitefish in upper Fraser River tributaries, it is unclear where or how the pinocchio and normal forms originated. The presence of both pinocchio and normal populations of mountain whitefish on both the Olympic Peninsula, Washington, and above the barrier at Shoshone Falls on the Snake River, Idaho, suggest that the propensity to form this dichotomy probably predates the Wisconsinan glacial period. Thus, postglacial migrants into British Columbia may have carried with them the genetic capacity to produce the pinocchio form. The presence of interpopulation variation in the development of elongate snout implies that either a strong environmental influence or local adaptation is involved in the expression of the trait. Thus, the pinocchio and normal forms may have diverged in situ. The present study is only a beginning. Questions about the morphological features associated with the  33  pinocchio morphotype and the role of environment and genetics in their expression clearly require more research. Further life history, laboratory rearing experiments, foraging observations, and genetic studies may unravel these problems now that the presence of two forms has been documented.  34  Drainage  Collection date  NORMAL S L ± s e (n) 165 ± 15 (19) 171 ± 35 (11) 176 ± 35 (25) 55  n  PINOCCHIO S L ± s e (n) 171 ± 19 (11) 173 ± 32 (10) 182 ± 23 (13) 34  25 Nechako Oct./95 21 Wright June/56 38 June/55 Swift 84 n Table 1 Ranges of standard lengths (mm) and collection sizes of the upper Fraser mountain whitefish measured in the morphometric survey.  Drainage  Collection date  N O R M A L , (n)  PINOCCHIO, (n)  total (n)  10 5 5 Bowron 1996 12 6 6 Wright 1956 21 10 11 1955 Swift 12 6 6 Nechako 1995 27 55 28 total Table 2 Sample sizes (n) of the upper Fraser mountain whitefish used in comparison of gill raker number in pinocchio and normal froms.  Drainage  Gill raker count ± s.'e. (NORMAL) 20.8 + 0.4 20.0 ± 0.6 18.8±0.5 20.5 ± 0.5 20.1 ± 0 . 3  Gill raker count ±s.e. (PINOCCHIO) 22.5 ± 0.43 23.0 ±0.61 21.3 ± 0 . 5 22.6 ± 0.5 22.4 ± 0.27  p-value  0.0047 Swift Bowron 0.005 0.027 Wright 0.0078 Nechako mean Table 3 Comparison of pinocchio and normal mountain whitefish primary gill arch raker number from various upper Fraser tributaries.  35  source  sum of squares 69.26 21.48 3.117  degrees of freedom 1 3 3  mean square 69.26 7.16 1.039  Ecotype Drainage Ecotype* Drainage Error 85.1 46 1.85 Table 4 Result of analysis of variance comparing pinocchio and normal raker numbers across upper Fraser tributaries.  Drainage McGregor Willow Salmon Chilako Bowron Nechako Fraser mainstem Chowade (Peace) Burnt (Peace) Duncan (Columbia) Totals  Adults normal 1 2 11 • 14  Adults pinocchio 3 3 6 12  F-ratio  p-value  37.44 3.87 0.562  <0.001 0.015 0.643  mountain whitefish gill  Juveniles  Total  4 4 6 3 3 12 7 4 2 45  4 8 6 3 8 17 12 7 4 2 71  Table 5 Mountain whitefish tissue sample sizes and collection sites included in the mtDNA analysis.  36  Composite Haplotype  Total sample  A  51 74.7% 15 22.5% 2 2.8% 68  B C  upper Fraser tributaries (adults), (juveniles) 19 73% , 13 76.4% 7 27% , 3 17.6% 0 0% , 1 6% 26 , 17  upper Fraser mainstem 9 75% 3  25%  0  0%  Peace System (Chowade, Burnt) 6 85.7%, 3 75% 1 14.3% , 0 0% 0 0% , 1 25% 7 , 4  Columbia system (Duncan) 1 50% 1  50%  0  0%  12 2 Total Table 6 Distribution of D-loop/cyt. b RFLP haplotype patterns in adult and juvenile mountain whitefish populations in British Columbia.  Composite Haplotype pattern  Pinocchios, (Nechako)  Normal, (Nechako)  Pinocchios, (Total sample)  Normal, (Total sample)  A 3, 50% 11, 100% 6, 50% 14, 100% B 3, 50% 0 6, 50% 0 C 0 0 0 0 14 Totals 6 11 12 Table 7 Distribution of D-loop/cyt b RFLP haplotype patterns from pinocchio and normal mountain whitefish populations in the upper Fraser system.  37  Morphotype  Normal Pinocchio  S.L± s.e. (mm) 285± 54 273± 45  Sample size  observation trials 10  Benthic foraging attempts (±s.e) 0  Water column foraging attempts (±s.e) 13.6±2.7  4 3  10  6.5+1.1  2.5±2  Total ±s.e 13.6± 2.7 9.0±3 .1  Table 8 Standard length, sample sizes, number of observation trials and mean number of foraging attempts made by pinocchio and normal mountain whitefish in the top observation pool in Dome Creek.  Morphotype  Normal Pinocchio  S.L+ s.e. (mm) 250± 53 215+ 45  Sample size  observation trials  4 2  14  Benthic foraging attempts (+s.e) 0±0  Water column foraging attempts (+s.e) 12.2412.2  12  3.6±1.5  9.511.1  Total +s.e 12.2 +2.2 10.1 11.2  Table 9 Standard length, sample sizes, number of observation trials and mean number of foraging attempts made by pinocchio and normal mountain whitefish in the bottom pool observation pool in Dome Creek.  38  Figure 1 — Distribution of mountain whitefish, Prosopium williamsoni, in North America.  39  Figure 2 — Head profiles of normal and pinocchio phenotypes of mountain whitefish, Prosopium williamsoni.  NORMAL  PINOCCHIO  40  Figure 3 —Upper Fraser River Basin including Prince George, B.C.  Upper Fraser River  41  Figure 4 — Distribution of residuals collected from a head vs. standard length regression of Swift River mountain whitefish.  HEAD LENGTH  o \—  Z) OQ Cd \-  co Q  _i < Q  CO  LU CL  -0.04 h -0.05 2.0  2.1 2.2 2.3 2.4 STANDARD LENGTH  2.5  42  Figure 5 — Distribution of residuals collected from a maxillary vs. standard length regression of Swift River mountain whitefish.  MAXILLARY LENGTH 0.10 o  0.05h  CQ  S  CO Q  o.ooh  < Q  CO UJ CrT  -0.05h  2.0  2.1 2.2 2.3 2.4 STANDARD LENGTH  2.5  43  Figure 6 — Distribution of residuals collected from a snout vs. standard length regression of Swift River mountain whitefish.  SNOUT LENGTH 0.10 o IGO \—  CO ^ Q  0.05h  o.oof-  < Q  CO -0.05h LU  -0.10  2.0  2.1 2.2 2.3 2.4 STANDARD LENGTH  2.5  44  Figure 7 — Frequency distribution of residuals collected from a head vs. standard length regression of Swift River mountain whitefish.  HEAD LENGTH 16 12  H0.4  0.3 o "O  C  o O  8  o 3o 0.2 =>  "D CD  0.1 S 0.0 0 -0.050 -0.025 0.000 0.025 0.050 RESIDUAL DISTRIBUTION  gure 8 —Frequency distribution of residuals collected from a maxillary vs. standard length regression of Swift River mountain whitefish.  MAXILLARY LENGTH 15  0.3 10h  3 o o  0.2 5h  CTJ  0.1  0 -0.10  -0.05 0.00 0.05 RESIDUAL DISTRIBUTION  0.0 0.10  -  46  Figure 9 — Frequency distribution of residuals collected from a snout vs. standard length regression of Swift River mountain whitefish.  SNOUT LENGTH 15  0.3  TJ  —i  10h  O "O  o  H0.2  O o  2o "O CD  0.1  o -0.10  -0.05 0.00 0.05 RESIDUAL DISTRIBUTION  0.0 0.10  0CJ 03  ^  47  Figure 10 — Scatter plot of principal component analysis loading values for Swift River mountain whitefish cranial measurements.  Swift River  -  2  1 0 1 2 Factor 1 (94.8% Variance)  48  Figure 11 — Hierarchical cluster analysis of factor one principal component loading values of mountain whitefish cranial measurements from the Swift River.  Case Case Case Case Case Case Case Case Case Case Case Case Case Case ase ase Case Case ase ase Case Case Case Case Case Case Case Case Case Case Case Case Case ase ase Case Case Case  S S  B  19 36 2 37 14 16 15 1 30 11 24 5 35 22 38 10 9 17 31 33 23 21 28 4 3 32 12 8 26 18 6 20 34 13 29 7 25 27  0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Euclidean Distance  49  Figure 12 — Scatter plot of principal component analysis loading values of mountain whitefish cranial measurements from three upper Fraser River tributaries.  Combined Drainages  CD o C ro i_ ro  > un  CN o '  o CO LL  -  2 - 1 0 1 2 3 Factor 1 (94.9% Variance)  50  Figure 13 — Hierarchical cluster analysis of factor one principal component loading values of mountain whitefish cranial measurements from three upper Fraser River tributaries.  0.0  0.1 0.2 0.3 Euclidean Distance  0.4  51  Figure 14 — Frequency distribution of gill raker counts from Swift River mountain whitefish.  GILL R A K E R S  — I  E O  8  0.4  6  MORPHOTYPE J 0.3 • NORMAL 0.2 • PINOCCHIO  4  0.1  2  0.0  0 t  j  o  "D O O  "O  C D -f 00  Q)  —%  TJ —^  0.1  2  TJ  O "O  o  O o  0.2  4  o "O  0.3  6 8 18  20 22 24 26 GILLRAKER NUMBER  0.4 28  C D —\ DO CD  52  Figure 15 — Frequency distribution of gill raker counts from mountain whitefish from four upper Fraser River tributaries.  GILL R A K E R S 20 -0.3  15  g O  O  10h  -0.2  5  -0.1  0  -0.0  5h  -0.1  10h 15 20 15  o o "O  MORPHOTYPE  -0.2  rjj  TJ O "O  o ao "O  CD —^  iNORMAL  -0.3  u PINOCCHIO 20  CD  25  GILLRAKER NUMBER  30  LTJ  Q)  53  Figure 16 — Frequency distributions of composite mitochondrial D N A RFLP haplotypes from upper Fraser River mountain whitefish arranged by collection location and life-history stage.  • Haplotype A • Haplotype B • Haplotype C 90 -i 80 70 u  e 60 s u" 50  (Ll be tt,  * 40 30  OH  20 10 0  I Mainstem Fraser  Tributaries (adults)  Tributaries (juveniles)  Total Samples  54  Figure 17 — Frequency distribution of upper Fraser River mountain whitefish composite mitochondrial RFLP haplotypes arranged by phenotypes.  • Haplotype A • Haplotype B 120  n  Pinocchio Nechako R.  Normal Nechako R.  Pinocchio Combined drainage  Normal Combined draianges  55  Figure 18 — Foraging preferences of pinocchio and normal mountain whitefish from two observation sites in Dome Creek (Top site and Bottom site). a) TOP SITE  BENTHIC  WATER COLUMN  MORPHOTYPE  MORPHOTYPE  b) B O T T O M SITE  WATER COLUMN  BENTHIC ==• 20 •  > &  111  1  t—  o z o  1  -  •  $ £  MORPHOTYPE  o  MORPHOTYPE  56  Figure 19 — Foraging rate of pinocchio and normal mountain whitefish from two observation sites in Dome Creek (Top and Bottom site).  TOP SITE  MORPHOTYPE  BOTTOM SITE  MORPHOTYPE  Literature cited: Bernatchez, L. and J.J. Dodson. 1990. Allopatric origin of sympatric populations of lake whitefish (Coregonus clupeaformis) as revealed by mitochondrial-DNA restriction analysis. Evolution 44: 1263-1271. Bernatchez, L., and J.J. Dodson. 1991. Phylogeographic structure in mitochondrial D N A of the lake whitefish (Coregonus clupeaformis) and its relation to Pleistocene Glaciations. Evolution 45: 1016-1035. Bernatchez, L., and A . Osinov. 1995. Genetic diversity of trout (Genus Salmo) from its most eastern native range based on mitochondrial D N A and nuclear gene variation. Molecular Ecology 4: 285-298. Bernatchez, L., A . Chouinard, and G. Lu. 1999. Integrating molecular genetics and ecology in studies of adaptive radiation: whitefish, Coregonus sp., as a case study. Biological Journal of the Linnean Society of London. 68:173-194. Bird, F.H. 1979. Pygmy whitefish, Prosopium coulteri, in three lakes of the Copper River system in Alaska. J. Fish. Res. Board Can. 36: 468-470. Bodaly, R.A. 1979. Morphological and ecological divergence within the lake whitefish (Coregonus clupeaformis) species complex in the Yukon Territory. Fish. Res. Board Can. Bull. 36:1214-1222. Bookstein, F.L., B. Chernoff, R. Elder, J. Humphries, G . Smith, and R. Strauss. 1985. Morphometries in evolutionary biology. The Academy of Natural Sciences of Philadelphia. Special Publication 15.  58  Davies, R.W. and G.W. Thompson. 1976. Movements of the mountain whitefish (Prosopium williamsoni) in the Sheep River watershed. Alberta. J. Fish. Res. Board Can. 33:2395-2401. Dgebuadze, Y . Y . , M . V . Mina, S.S. Alesgev, and A.S. Golubtsov. 1999. Observation on reproduction of the lake Tana barbs. J. Fish Biology. 54:417-423. Evermann, B.W. 1892. A reconnaissance of the streams and lakes of Western Montana and Northwestern Wyoming. Bull. U.S. Fish. Comm. XF3-60. Fausch K.D., Nakano S., Kitano S. 1997. Experimentally induced foraging mode shift by sympatric charr in a Japanese mountain stream. Behavioral Ecology 8: 414-420. Ferguson, A., and F . M . Manson. 1981. Allozyme evidence for reproductively isolated sympatric populations of brown trout, Salmo trutta, in Lough Melvin, Ireland. J. Fish Biology 18: 629-42. Galat, D.L., and N . Vucinich. 1983. Food partitioning between young of the year of two sympatric tui chub morphs. Trans. Am. Fish. Soc. 112:486-497. Galis, F. and E.G. Drucker. 1996. Pharyngeal biting mechanisms in Centrachids and chiclids: Insights into key evolutionary innovation. J. Evol. Biology 9:641-670. Galis, F. and J.A.J. Metez. 1998. Why are there so many cichlid species? Trends Ecol. Evol. 13:1-2. Gardner, A.S., A.F. Walker, and R.B. Greer. 1988 Morphometric analysis of two ecologically distinct forms of Arctic charr, Salvelinus alpinus (L.), in Loch Rannoch, Scotland. J. of Fish Biology 32:901-910. Haas, G.R. 1988. The sytematics, zoogeography and evolution of Dolly Varden and bull trout in British Columbia. M.Sc. thesis, Dept.of Zoology, U.B.C., Vancouver, B.C.  59 Haas, G.R., and J.D. McPhail. 1991. Systematics and distributions of Dolly Varden (Salvelinus malmd) and Bull trout (Salvelinus confluentus) in North America. Can. J. Fish. Aquat. Sci. 48:2191-2211. Hall H.J., and L.W. Nawrocki. 1995. A rapid method of detecting mitochondrial D N A variation in the brown trout, Salmo trutta. J. of Fish Biology 46:360-364. Hindar, K., and B. Jonsson. 1992. Ecological polymorphism in Arctic charr. Biological Journal of the Linnean Society 48:63-74. Holt, R.D. 1960. Comparative morphometry of the Mountain Whitefish, Prosopium williamsoni: Copeia, 1960:192-2001. Hubbs, C.L., and K.F. Lagler. 1947. Fishes of the Great Lakes region. Univ. Michigan Pess, Ann Arbor, Michigan. Humphries, J.M., F.l. Bookstein, B. Chernoff, G.R. Smith, R.L. Elder, and S.G. Poss. 1981. Multivariate discrimination of shape in relation to size. Systematic Zoology 30:291-308. Jonsson, B., and K . Hindar. 1982. Reproductive strategy of dwarf and normal Arctic charr (Salvelinus alpinus) from Vangsvatnet Lake, Western Norway. Can. J. Fish. A q . Sci. 39:1404-1413. Jordan, D.S., and J.O. Snyder. 1909. Description of a new whitefish (Coregonus oregonius) from McKenzie river, Oregon. Proceedings of the United States National Museum X X X V I : 425-430. Lagler, Carl, F. 1956. Freshwater Fishery Biology. W.M.C. Brown Company, Dubuque, Iowa.  Liem, K.F. 1973. Evolutionary strategies and morphological innovations: Cichilid pharyngeal jaws. Syst. Zoology 22:245-441. Lindsey, C C . 1988. Factors controlling meristic variation, pp. 197-274. In D.J. Randall and W.S. Hoar, (eds.) Fish Physiology, vol XIB. Academic Press, New York. Lindsey, C C 1981. Stocks are chameleons: plasticity in gill rakers of coregonid fishes. Can. J. Fish. Aquatic Sci. 38:1497-1505. McAfee, W.R.I 966. Mountain whitefish. In L . A . Calhoun (ed.) Inlandfisheries management. California Department of Fish and Game, Bulletin: 299-303. McCart, P. 1970. Evidence for the existence of sibling species of pygmy whitefish (Prosopium coulteri) in three Alaskan lakes. In C C Lindsey and C S . Woods (ed.) Biology of Coregonid Fishes, pp 81-98.  McElroy, M.D., E. Bermingham, and I. Kornfeild. 1992. REAP: an integrated environment for manipulation and phylogenetic analysis of Restriction data. J of Heredity 83: 157-158. McHugh, J.L. 1940. Food of the Rocky Mountain Whitefish, Prosopium williamsoni (Girard). J. Fish. Res. Board Canada. 5:131-137. McPhail J.D. 1992. Ecology and evolution of sympatric sticklebacks (Gasterosteus): evidence for a species-pair in Paxton Lake , Texada Island, British Columbia. Can. J. Zoology 70: 361-369. McPhail, J.D. and C C Lindsey. 1970. Freshwater fishes of northwestern Canada and  Alaska. Fish. Res. Board Can. Bull. 173.  McPhail, J.D., and C.C. Lindsey. 1986. Zoogeography of the freshwater fishes of Cascadia. In Hocutt and E.O. Wiley. John Wiley and Sons (ed.) The zoogeography of North American freshwater fishes. 615-637.  McPhail, J.D., and P.M. Troffe. 1998. The mountain whitefish (Prosopium williamsoni): A potential indicator species for the Fraser system. Technical report submitted to the Fraser River Action Plan. Meyer, A . 1989. Plasticity in morphology and performance in the trophically polymorphic cichlid fish, Cichlasoma citrinellum. Dissertation Abstracts International, Part B : The Sciences and Engineering 50: 229p. Michalak, P. and J. Rafinski. 1999. Sexual isolation between two newt species (Triturus valgaris and T. montandoni, Amphibia, Salamanridae). Biological Journal of the  Linnean Society 67:343-352 Norden C.R. 1970. Evolution and distribution of the genus Prosopium. In C.C. Lindsey and C.S. Woods (ed.) Biology of CoregonidFishes. 81-98. Northcote, T.G. and G.L. Ennis. 1994. Mountain Whitefish Biology and Habitat use in Relation to Compensation and Improvement Possibilities. Reviews in Fisheries Science 2(4): 347-371. Northcote, T.G. 1988. Meristic variation in the New Zealand smelt, Retropinna retropinna, (Richardson): New questions for an old problem. Verhandlungen. Internationale Vereinigung fur theoretische und angewandte Limnologie/Proceedings International 23(3) Available as: Congress in New Zealand 1987. Proceedings, pp 1797-1804.  Reist, J.D. 1985. A n empirical evaluation of several univariate methods that adjust for size variation in morphometric data. Can. J. Zooi. 63: 1429-1439. Robinson, B.W., and D.S. Wilson. 1996. Genetic variation and phenotypic plasticity in a trophically polymorphic population of pumpkinseed sunfish (Lepomis gibbosus). Evolutionary Ecology 10: 631-652. Rocha-Olivares, A., R.H. Rosenbalt, and R.D. Vetter. Molecular evolution, systematics and zoogeography of the rockfishes subgenus Sebastomus (Sebastes, scorpaenidae) based on Mitochondrial Cytochrome b and control region sequences. Molecular Phylogenetics and Evolution 11:441-458. Ruzzante, D.E., S.J. Waide, V . E . Cussac, P.J. Macchi and M.F. Alonso. 1998. Trophic polymorphism, habitat and diet segregation in Percichthys trucha (Pisces: Percichthyidae) in the Andes. Biological Journal of the Linnean Society 65: 191214. Sajdak, S.L., and R.B. Phillips. 1997. Phylogenetic relationships among Coregonus species inferred from the D N A sequence of the first transcribed spacer (ITS1) of ribosomal D N A . Can. J. Fish. Aquat. Sci. 54:1494-1503. Schluter, D. 1996. Ecological speciation in postglacial fishes. Philosophical Transactions of the Royal Society of London 351: 807-814. Scott, W.B., and E.J. Crossman. 1973. Freshwater fishes of Canada. Fish. Res. Board Canada. Bull. 184:966. Sharp, J.C., K . W . Able and W.C. Leggett. 1978. Utility of meristic and morphometric characters for identification of capelin (Mallotus vdlous) stocks in Canadian Atlantic waters. J. Fish. Res. Board Can. 35: 124-130.  63  Skulason, S., D.L.G. Noakes, and S.S. Snorrason. 1989. Ontogeny of trophic morphology in four sympatric morphs of Arctic charr, Salvelinus alpinus, in Thingvallavatn, Iceland. Biological Journal of the Linnean Society 38:281-301. Skulason, S., and T.B. Smith. 1995. Resourse polymorphisms in vertebrates. Trends Ecol. Evol. 10:366-370. Skulason, S., S.S. Snorrason, D. Ota, and D.L.G. Noakes. 1993. Genetically based differences in foraging behaviour among sympatric morphs of Arctic charr (Piscces: Salmonidae). Animal Behaviour 45:1179-1192. Smith, G.R. 1975. Fish of the Pliocene Glenns Ferry formation, southwest Idaho. University of Michigan Papers in Palaeontology 14:1-68. Stalnaker, C.B., R.E. Gresswell, and R.E. Siefert. 1974. Early life history and feeding of young mountain whitefish. U.S. Environmental Protection Agency, Ecological Research Series, EPA-660/3-73-019. Svardson, G. 1957. The coregonid problem, vi. The Palearctic species and their intergrades. Rep. Inst. Freshwater Res. Drottningholm 38:261-356. Svardson, G. 1979. Speciation of Scandinavian Coregonus. Rep. Inst. Freshwater Res. Drottningholm 57:95p. Taylor, E.B., and P. Bentzen. 1993. Evidence for multiple origins and sympatric divergence of trophic ecotypes of smelt (Osmerus) in northeastern North America. Evolution 47:813-822. Taylor, E.B., and J.D. McPhail. 1999. Evolutionary history of an adaptive radiation in species pairs of threespine sticklebacks (Gastewsteus): insights from mitochondrial D N A . Biological Journal of the Linnean Society. 66: 271-291.  Vuorinen, J.A., R . A Bodaly, J.D. Riest, and M . Lucynski. 1998. Phylogeny of five Prosopium species with comparisons with other Coregonine fishes based on isozyme electrophoresis. J. Fish Biology. 53(5). Wimberger, P.H. 1994. Trophic polymorphisms, plasticity, and speciation in vertebrates. Theory and Application of Fish Feeding ecology. University of South Carolina Press, Columbus, South Carolina, pp. 19-44. Winans, G.A. 1984. Multivariate morphometric variability in Pacific salmon: Technical demonstration. Can. J. Fish. Aquat. Sci. 41:1150-1159. Zar, J.H. 1984. Biostatistical Analysis, 2 ed. Prentice-Hall, Inc., Englewood Cliffs, New Jersey.  65  Appendix I: Mountain whitefish, Prosopium williamsoni, mitochondrial RFLP haplotype patterns by drainage collection, (p) = pinocchio phenotype  Sty I Haplotype restriction fragments Haplotype A 762 bp—  Haplotype B  762bp —  995 bp — 1096bp —  Drainage McGregor 1 McGregor 2 McGregor 3 McGregor 4  Sample # 19 21 23 24  Stvl A A C A  Willow 1 Willow 3 Willow 4 (p) Willow 5 Willow 6 (p) Willow 7 Willow 8 Willow 9  wil 1 wil 3 25 26 P(2) B017 B018 B019  A A B A B A A A  Salmon 1 Salmon 2 Salmon 3 Salmon 6 Salmon 7 Salmon 8  9 10 11 90 91 92  B A A A A B  Chilako 1 Chilako 2 Chilako 3  chil2 chil 3 chil4  A B B  Fr. Fr. Fr. Fr.  33 48 59 60  B A A A  Mainstem 1 Mainstem 4 Mainstem 5 Mainstem 6  Haplotype C 762bp — 883 bp—  Fr. Mainstem 7 Fr. Mainstem 9 Fr. Mainstem 10 Fr. Mainstem 11 Fr. Mainstem 12 Fr. Mainstem 13 Fr. Mainstem 14 Fr. Mainstem 15  65 70 71 80 85 86 95 96  A B A A A A A B  Nechako 1 (p) Nechako 3 Nechako 4 Nechako 5 Nechako 6 Nechako 7 Nechako 8 (p) Nechako 9 Nechako 10 (p) Nechako 11 (p) Nechako 13 Nechako 14 Nechako 15 Nechako 16 Nechako 17 Nechako 18 (p) Nechako 19  DI D3 D4 D5 D6 D7 D8 D9 D10 Dll D13 D14 D15 D16 Nek 2 Nek 5 Nek 9  A A A B A A A A B A A A A A A B A  Bowron Bowron Bowron Bowron Bowron Bowron Bowron Bowron  Bow 1 Bow 2 Bow 3 Bow 4 Bow 5 Bw20 Bw21 Bw37  A A A A A A A B  Duncan 1 Duncan 2  D61 D62  A B  Chowade 1 Chowade 2 Chowade 3 Chowade 4 Chowade 5 Chowade 6  Chi Ch2 Ch3 Ch4 Ch5 Ch6  A B A A A A  1 2 3 4 5 6 (p) 7 (p) 8 (p)  Drainage Chowade 7  Ch7  Sty I A  Burnt Burnt Burnt Burnt  BI B2 B3 B4  A A A C  1 2 3 4  Sample #  Appendix II: Comment on Wright Creek: The University of British Columbia fish museum contains mountain whitefish samples from Wright Creek, a tributary to the Salmon River. The sample was collected in June, 1955 and contains mountain whitefish of all size classes, including young-of-theyear and large adults of both the pinocchio and normal forms. This collection suggests that at one time mountain whitefish used Wright Creek during summer months, presumably as an adult foraging site and a juvenile rearing area. In 1995 repeated poleseining and back-pack electroshocker surveys in Wright Creek — from the source to its confluence with the Salmon River— uncovered no mountain whitefish. The absence of adults, and especially young-of-the-year and juveniles suggest that the creek no longer supports a mountain whitefish population.  


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