Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A genetic and behavioural analysis of the distinctiveness of the Cultus pygmy sculpin (Cottus aleuticus)… Woodruff, Patricia 2010

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

Item Metadata

Download

Media
24-ubc_2010_fall_woodruff_patricia.pdf [ 1.77MB ]
Metadata
JSON: 24-1.0071112.json
JSON-LD: 24-1.0071112-ld.json
RDF/XML (Pretty): 24-1.0071112-rdf.xml
RDF/JSON: 24-1.0071112-rdf.json
Turtle: 24-1.0071112-turtle.txt
N-Triples: 24-1.0071112-rdf-ntriples.txt
Original Record: 24-1.0071112-source.json
Full Text
24-1.0071112-fulltext.txt
Citation
24-1.0071112.ris

Full Text

A GENETIC AND BEHAVIOURAL ANALYSIS OF THE DISTINCTIVENESS OF THE CULTUS PYGMY SCULPIN (COTTUS ALEUTICUS) AND IMPLICATIONS FOR ITS CONSERVATION  by  PATRICIA WOODRUFF B.Sc., The University of Victoria, 1998    A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE   In   THE FACULTY OF GRADUATE STUDIES  (Zoology)   THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)   August 2010  ©Patricia Woodruff, 2010  ii  Abstract The Cultus pygmy sculpin (Cottus aleuticus) is endemic to Cultus Lake, southwestern BC, and is listed as threatened under the federal Species at Risk Act (SARA). The Cultus pygmy sculpin was first discovered by Ricker in the 1930’s (Ricker 1960) and was described as a dwarf coastrange sculpin (C. aleuticus).  It matures at a smaller size than the “normal” C. aleuticus; has a lacustrine rather than a fluvial life history; and appears to undertake diurnal feeding migrations into the water column to feed on the abundance of Daphnia.  The Cultus pygmy sculpin also has different morphological features, including larger cephalic pores, shorter pelvic fins, and a greater number of pectoral fin rays, which may be advantageous for an open-water, planktivorous lifestyle. Little else, however, is known of the biology of this fish. I used mtDNA and microsatellite analyses to determine whether there is any genetic distinction between the pygmy sculpin and the coastrange sculpin.  Potential behavioural differences between the two forms were also examined by comparing vertical depth selection in the laboratory, with and without Cottus asper, a benthic predator found in Cultus Lake.  Results from combined microsatellite and mtDNA analyses indicate that there is no clear phylogeographic separation between pygmy and coastrange sculpin, but that there is genetic differentiation at the population level.  Pelagic sculpin from Lake Washington were compared with the Cultus pygmy sculpin and showed divergence from it, which suggests that the pelagic life history has evolved independently in each lake. The Cultus pygmy sculpin was found significantly higher in experimental depth selection tanks than the stream forms of C. aleuticus, but only when C. asper was also present in the tanks.  Taken together, molecular and behavioural results indicate that the Cultus pygmy sculpin is discrete from the coastrange sculpin, and validate its recognition as a iii  conservation unit distinct from “typical” C.aleuticus. The presence of abundant planktonic prey within the water column of Cultus Lake, combined with the presence of predators on the lake’s bottom, may be important features of the lake that have promoted the evolution of pelagic behaviour by the Cultus pygmy sculpin.  iv  Table of contents Abstract ........................................................................................................................... ii Table of contents ............................................................................................................ iv List of tables .................................................................................................................... v List of figures ................................................................................................................. vii Acknowledgements ......................................................................................................... ix Chapter 1: General introduction ...................................................................................... 1 The coastrange sculpin ............................................................................................... 2 Chapter 2: Evolutionary relationships and population structure of the Cultus pygmy sculpin, Cottus aleuticus ................................................................................................. 7 Introduction ................................................................................................................. 7 Materials and methods ...............................................................................................11 Study area ..............................................................................................................11 Sculpin samples .....................................................................................................11 Mitochondrial DNA ..................................................................................................12 Microsatellite analysis .............................................................................................13 Results .......................................................................................................................14 Mitochondrial DNA ..................................................................................................14 Microsatellite DNA ..................................................................................................16 Discussion ..................................................................................................................19 mtDNA inter-relationships .......................................................................................19 Microsatellite DNA differentiation ............................................................................21 Chapter 3: Habitat use and evidence for behavioural distinctiveness of the Cultus pygmy sculpin ................................................................................................................46 Introduction ................................................................................................................46 Habitat ....................................................................................................................47 Behaviour ...............................................................................................................48 Materials and methods ...............................................................................................50 Fish samples ..........................................................................................................50 Behavioural assays ................................................................................................52 Results .......................................................................................................................53 Habitats used by sculpins in Cultus Lake ...............................................................53 Depth selection of sculpins .....................................................................................54 Discussion ..................................................................................................................56 Future studies .........................................................................................................60 Chapter 4: Summary and general discussion ................................................................70 Further research .........................................................................................................75 References ....................................................................................................................78 Appendix 1. UTM coordinates of minnow trap sites in Cultus Lake. ..............................95 Appendix 2. List of lengths of Cultus pygmy sculpin captured in minnow traps in Cultus Lake, May-October 2008. ..............................................................................................97 Appendix 3. Animal care certificate used in conducting field sampling and laboratory experiments. ................................................................................................................101  v  List of tables Table 1. Location, species and number of samples of Cottus used for mtDNA D-loop analysis. .........................................................................................................................25 Table 2. Population location and number of samples of Cottus used for microsatellite analysis. .........................................................................................................................26 Table 3a. PCR strategy: repeat motifs and sequences of the primers used for microsatellite analysis. ...................................................................................................27 Table 3b. PCR strategy: the multiplex technique used for microsatellite analysis. ........27 Table 4. Number and range of sizes (in base pairs, bp) of alleles by locus and population for microsatellite analysis of Cottus samples. ...............................................28 Table 5.  Descriptive statistics by locus and population for microsatellite analysis of Cottus samples.  N is the number of samples, HE is the expected heterozygosity, HO is the observed heterozygosity, and AR is allelic richness. ................................................29 Table 6. Results of tests of Hardy-Weinberg Equilibium by locus and population for microsatellite analysis of Cottus samples.  Underlined probability values indicate significant differences from Hardy-Weinberg expectations. ...........................................30 Table 7. Results from tests of linkage disequilibrium for all populations, locus by locus, for microsatellite analysis of Cottus samples.  There was no significant result over all populations, following Bonnferroni correction. ...............................................................31 Table 8. FST values for eight populations, using eight loci for microsatellite analysis of various Cottus samples.  Underlined values are not significantly different from 0. The population in boldface represents the Cultus Lake pygmy sculpin. Samples in italics are found within the Cultus Lake drainage. Cheak = Cheakamus river, Mam = Mamquam river, Squam = Squamish river, Cultus = samples from fish caught by trawl and minnow trap in Cultus Lake, Wash = Lake Washington, Tributaries = Watt and Windfall creeks32 Table 9.  Assignment test results for individual C. aleuticus caught by minnow trap in Cultus Lake. Tribs=Tributaries (Watt and Windfall Creeks), Cultus=Cultus Lake pygmy population. “Percentage” refers to the relative level of confidence of assignment to the most likely population for each individual fish. ...............................................................33 Table 10.  Structure results for microsatellite analysis across the eight original sample localities (Cheakamus, Mamquam and Squamish Rivers; Cultus Lake and its three tributary streams: Frosst Creek and combined Watt and Windfall creeks; Lake Washington; and Norrish Creek). The K values are an average of ten separate runs. The value in bold and underlined represents the most likely number of populations. ....34 Table 11.  Structure results for microsatellite analysis, across the three localities within the Cultus Lake drainage (Cultus Lake; Frosst Creek; and combined Watt and Windfall creeks).  The K values are an average of ten separate runs.  The value in bold and underlined represents the most likely number of populations. .......................................35 Table 12. Analysis of molecular variance of microsatellite DNA allele frequencies.  The first analysis (Squamish vs. Cultus vs. others) compared the three Squamish drainage rivers (Cheakamus, Squamish and Mamquam) with Cultus Lake and its three tributaries (Frosst, Watt and Windfall creeks) and also with the other two systems (Norrish Creek and Lake Washington).  The second analysis compared Cultus pygmy sculpin (caught by trawl and minnow trap in Cultus Lake) with its three tributaries (Frosst, Watt and Windfall creeks). ............................................................................................................36 Table 13.  Locations and sample size of the populations of Cottus aleuticus used for the depth selection study. ....................................................................................................62 vi  Table 14.  Number of Cultus pygmy sculpins captured by month using stratified minnow traps in 2008, and the average length of the fish captured. ...........................................63 Table 15. The number of mature-sized (> 40 mm total length) sculpin caught in DFO mid-water sockeye trawls throughout the province of BC.  Of the 62 total lakes surveyed, only four caught sculpin that were the size of mature fish.  Due to the large number of sculpin caught in Cultus Lake in some of the trawls, lengths were not recorded for all fish, and so the number of mature fish caught in Cultus Lake is a conservative estimate. The sculpin caught in Pitt and Red Bluff lakes were not all identified to species, so it is possible that other species (i.e., C. asper) are included, and this number is likely an overestimate of mature C. aleuticus sampled from those lakes. .......................................................................................................................................64 Table 16.  The average depth and number of times off bottom, with standard deviation for each mean in brackets, shown by the Cultus pygmy sculpin (Cultus) and the C. aleuticus from the streams (Frosst = Frosst Creek; LCR = Little Campbell River). Experiment one used a “focal fish”, or one fish per tank; experiment two used a “focal fish” with a C. asper in the tank. ....................................................................................65   vii  List of figures Figure 1.  Global distribution of the coastrange sculpin based on 262 occurrence records obtained from the WFC (2006) and GBIF (2006) (from COSEWIC 2010). .......37 Figure 2.  A stream ecotype C. aleuticus (top), from Little Campbell River, compared with a Cultus pygmy sculpin (bottom).  Both were identified as mature females by the presence of eggs. ..........................................................................................................38 Figure 3. Cultus Lake bathymetric map (10 m intervals), modified from Shortreed (2007). Inset shows location of Cultus Lake within the boxed area in British Columbia. Mature Cottus were identified by the presence of eggs or testes. Trawl transects are denoted by T1-T4 (i.e., Trawls 1-4)................................................................................39 Figure 4. Phylogenetic analysis of mtDNA D-loop across 775 nucleotides under a GTR+I+G model of sequence evolution. Numbers represent the Bayesian posterior probability (%, 10,000 trees). Asper = Cottus asper. Sequences annotated with a filled square are Cultus Lake pygmy sculpin, those with an open square are C. aleuticus from Cultus Lake tributary streams. .......................................................................................40 Figure 5. Phylogenetic analysis of mtDNA D-loop across 340 nucleotides under a GTR+I+G model of sequence evolution. Numbers represent the Bayesian posterior probability (%, 10,000 trees). Asper = Cottus asper. Sequences annotated with a filled square are Cultus Lake pygmy sculpin, those with an open square are C. aleuticus from Cultus Lake tributary streams. .......................................................................................41 Figure 6.  Structure results across eight localities and K = 3 genetic groups:  green, blue, and red.   Each fish is represented by a thin vertical line composed of one to three colours each of which represents the proportional contribution of each genetic group to a fish’s genetic composition across eight microsatellite loci. In order from 1-8: 1 = Cheakamus River, 2 = Mamquam River, 3 = Squamish River, 4 = Cultus Lake pygmy sculpin (trawl and minnow trap fish combined), 5 = Frosst Creek, 6 = Lake Washington, 7 = the two tributary streams (Watt and Windfall creeks), and  8 = Norrish Creek. .......42 Figure 7.  Structure results for Cultus Lake drainage samples and K = 2 genetic groups: red and green shading.  Each fish is represented by a thin vertical line composed of one to two colours each of which represents the proportional contribution of each genetic group to a fish’s genetic composition across eight microsatellite loci. In order from 1-3: 1 = Cultus Lake pygmy sculpin (trawl and minnow trap fish combined), 2 = Frosst Creek, and 3 = the two tributary streams (Watt and Windfall creeks). .......................................43 Figure 8.  Factorial Correspondence Analysis (FCA) of variation across eight microsatellite loci among 350 Cottus samples (including C. aleuticus and the Cultus pygmy sculpin) collected from nine locations in 2006-2008.  The three Squamish drainage populations, from left to right in the Squamish circle, include the Mamquam River (blue), the Squamish River (white), and the Cheakamus River (yellow). The three Cultus drainage populations, from left to right in the Cultus circle, include the two tributary streams (Watt and Windfall creeks) (brown), Frosst Creek (pink), and the Cultus pygmy sculpin (grey). .........................................................................................44 Figure 9. Factorial Correspondence Analysis (FCA) of variation across eight microsatellite loci between Cultus pygmy sculpin (N=61; yellow); Frosst Creek C. aleuticus (N=30; blue); and the two tributary streams (Watt and Windfall Creeks) (N=55; white). ............................................................................................................................45 Figure 10.  Box plots of the distribution of the average depths by population of Cottus aleuticus, with no C. asper present.  Cultus = Cultus pygmy sculpin (n=25); Frosst = viii  Frosst Creek (n=32); LCR = Little Campbell River (n=27).  The thick horizontal line shows the median depth, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). ............66 Figure 11.  Box plots of the distribution of the number of times Cottus aleuticus were off the bottom, by population, with no C. asper present.  Cultus = Cultus pygmy sculpin (n=25); Frosst = Frosst Creek (n=32); LCR = Little Campbell River (n=27). The thick horizontal line shows the median count, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). ............................................................................................................................67 Figure 12.  Box plots of the distribution of the average depths of Cottus aleuticus by population, with C. asper present.  Cultus = Cultus pygmy sculpin (n=23); Frosst = Frosst Creek (n=31); LCR = Little Campbell River (n=12). The thick horizontal line shows the median depth, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). ............68 Figure 13.  Box plots of the distribution of the average number of times Cottus aleuticus were off the bottom, by population, with no C. asper present.  Cultus = Cultus pygmy sculpin (n=23); Frosst = Frosst Creek (n=31); LCR = Little Campbell River (n=12).  The thick horizontal line shows the median count, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). .............................................................................................................69  ix  Acknowledgements I would like to thank Greg Wilson and Thomas Woodruff for their assistance in the field and their continuing support.  I would also like to thank my parents for all their help.  I am grateful for all the support provided to me by my supervisor, Dr. Eric Taylor, and my committee members Dr. Mary Berbee, Dr. Darren Irwin, and Dr. Jordan Rosenfeld, as well as all the assistance given to me by other faculty and staff in the Department of Zoology.  This study could not have been completed without the samples and data provided by Jeremy Hume and Steve MacLellan of Fisheries and Oceans Canada.  Other samples were provided by Dr. Tom Quinn, Rob Dolighan, Tom Bansak, Andrew Whiteley, and Roger Tabor.  Thanks to all members of the Taylor lab, past and present, for the good times in the field and lab, and at social events.  1  Chapter 1: General introduction  British Columbia’s (BC) glacial history involved repeated glacial advances and retreats during the Pleistocene (~ the last 2 million years), and the province is home to a fish fauna that is relatively depauperate taxonomically; BC has only about 70 taxonomically recognized species (McPhail 2007).  During Pleistocene glacial advances the BC fish fauna was either eliminated or restricted to ice-free refugia peripheral to the main ice sheets. Consequently, the contemporary freshwater fish fauna of BC consists almost entirely of postglacial immigrants (McPhail and Carveth 1992).  Even after the climate began to warm and the glaciers started to retreat at the end of the Wisconsinan glaciation about 12,000 years ago, there were only a few connections between different drainages and these passages were only temporary (McPhail and Carveth 1992).  The geography of drainage systems and their interconnections heavily influenced dispersal and range expansion of freshwater fishes (Lindsey and McPhail 1986; McPhail and Lindsey 1986; Bernatchez and Wilson 1998).  Consequently, these historical and biological factors have resulted in many opportunities for allopatry between fish populations. These factors, coupled with the generally low species diversity and potentially unexploited niches, have probably promoted postglacial genetic divergence and speciation (McPhail and Carveth 1992; Taylor 1999).  The BC freshwater fish fauna, therefore, is characterized by relatively low taxonomic diversity, but contains many examples of highly divergent “ecotypes”, intraspecific phylogenetic groups, and biological species that have yet to be described formally (e.g. McPhail and Taylor 1999; McPhail 2007). One way in which adaptive radiation occurs is when species begin to exploit a variety of different resource types (Futuyma 1986).  Geographic isolation and new, 2  harsh ecological conditions present during deglaciation (McPhail and Carveth 1992) provided the opportunity to expand into new niches, possibly due to intraspecific competition for limited resources (Schluter 1993).  As morphology and habitat use are strongly correlated with each other within and between populations, different selection pressures can promote changes in morphology (Schluter 1993). These selection pressures are thought to have promoted diversification of various “species complexes” in freshwater fishes (Behnke 1972; Schluter and McPhail 1992). Many fishes, such as members of the Salmonidae (salmon, trout, char, grayling, and whitefishes) that inhabit post-glacial lakes, frequently exhibit two or more closely related, sympatric forms within a single taxon (Behnke 1972).  These differences could be due to a single species adopting different life history traits, or they could be separate species (Schluter 1993).  Freshwater fishes such as threespine sticklebacks (Gasterosteus spp.; Schluter 1993) and cichlids (Amphilophus spp.; Barluenga et al. 2006) have shown evidence of rapid speciation owing, in part, to interactions in sympatry.  My study focused on a distinctive population of a freshwater sculpin (Cottidae) currently only observed in open water habitats of one lake in southwestern BC that is thought to be another example of rapid, postglacial divergence. The population is presumed to have been recently derived from a geographically more widespread form that typically has a stream-dwelling and benthic life history (Scott and Crossman 1973).  The coastrange sculpin  The coastrange sculpin (Cottus aleuticus) is found in coastal watersheds of western North America from Bristol Bay in Alaska, south to San Luis Obispo, California 3  (Scott and Crossman 1973) (Figure 1). As is the case with all sculpins, it lacks a swim bladder and is, therefore, primarily benthic.  Except in a few large river systems such as the Fraser and Skeena rivers, it is typically restricted to within about 100 km of the coast. Sculpins invaded freshwater from the marine environment during the Pliocene or late Miocene (Yokoyama and Goto 2005).  In BC, C. aleuticus is found in all coastal watersheds, including Vancouver Island and the Queen Charlotte Islands (McPhail 2007).  It is believed that it survived the most recent glaciation in both northern and southern refugia, but all BC populations probably originated from a single southern refugium (McPhail and Lindsey 1970).  Due to its extensive range, it is likely that ecological specializations exist (e.g., Whitely et al. 2009).  Ricker (1960) described one such specialization: a population of dwarf coastrange sculpins known only from Cultus Lake, southwestern BC, and recognized informally as the “Cultus pygmy sculpin”. The young of the coastrange sculpin become planktonic in lakes or estuaries after hatching, and then take up a benthic lifestyle approximately one month later, before making their way back into the streams (Scott and Crossman 1973).  The coastrange sculpin favours a flowing water environment when they become bottom- dwelling (Scott and Crossman 1973).  While the larvae feed on microplankton during their pelagic stage, adult fluvial coastrange sculpin primarily forage on the nymphs and larvae of aquatic insects (McPhail 2007). The Cultus Lake pygmy sculpin is different from ”typical”  coastrange sculpin in that it appears to reside and mature at a smaller size in the lake’s pelagic environment (Cannings 1993; McPhail 2007; Figure 2).  A close relative of the coastrange sculpin, the prickly sculpin (C. asper) is a voracious piscivore (McPhail 2007) and is also found in Cultus Lake, but apparently only in the littoral and benthic habitats of the lake. 4  Coastrange sculpin found in Frosst Creek (a tributary of Cultus Lake) and Cultus pygmy sculpin appear to be smaller on average than C. aleuticus caught in Norrish Creek (lower Fraser River) and the Little Campbell River (Semiahmoo Bay just south of the Fraser River estuary).  Cultus pygmy sculpin are thought to occupy only the deeper, offshore waters of the lake, based on lack of observation of Cultus pygmy sculpins during shore and tributary sampling. In Ricker’s (1960) study, the Cultus pygmy sculpin was not captured in shore seining or in traps set for juvenile Pacific salmon (Oncorhynchus spp.), and was only found in significant numbers in the stomachs of bull trout (Salvelinus confluentus), which prefer deeper, offshore waters.  Since 1970, Fisheries and Oceans Canada (DFO) have captured sculpins in their pelagic trawling for sockeye salmon (Oncorhynchus nerka) in Cultus Lake (J. Hume, DFO, pers. comm.).  It has been proposed that the Cultus pygmy sculpin undertakes diurnal feeding migrations, following the zooplankton on which it feeds (McPhail 2007). Limnetic life histories are extremely rare in the Cottidae. In fact, only three species of the 250 or so described taxa are limnetic and these belong to two highly divergent genera endemic to Lake Baikal in central Siberia (Kontula et al. 2003). The Cultus Lake pygmy sculpin is a provincially red-listed species (meaning that it is an indigenous species listed as threatened in BC) as its range is restricted to one heavily utilised lake:  Cultus Lake received three million recreational visitors in 2009 (Cultus Lake Park Board minutes, September 2009).  The Cultus pygmy sculpin has also been recognized as a “designatable unit” (DU) separate from the coastrange sculpin by the Committee for the Status of Endangered Wildlife in Canada (COSEWIC, e.g. COSEWIC 2010). A designatable unit is an intraspecific population, or assemblage of populations, within a species that is recognized as separate from other members of 5  the taxon if it is both “discrete”, and if this discreteness is “significant” in terms of the evolutionary or ecological legacy of the taxon (COSEWIC 2009). Consequently, a DU can be recognized as a “wildlife species” for legal listing and protection under Canada’s Species at Risk Act (SARA). Various genetic, biogeographic, and ecological criteria are used to assess the “discrete” and “significance” criteria for DU recognition by COSEWIC (2009). The Cultus pygmy sculpin is legally listed under SARA on Schedule 1 as a threatened species, which means that recovery planning and critical habitat protection are mandatory and a recovery strategy for this DU has been completed (National Recovery Team for the Cultus Pygmy Sculpin 2007). Despite the status of the Cultus pygmy sculpin as a DU, the only studies on this fish were conducted by Ricker in 1932- 37 (Ricker 1960) and the status reports of COSEWIC (e.g. COSEWIC 2010).  Although COSEWIC (2010) reported some differences in morphology between Cultus pygmy sculpin and typical coastrange sculpin (primarily in cephalic pore sizes), there has been no rigorous assessment of the status of this fish as a valid DU under the discrete and significant criteria of COSEWIC. For instance, it is unclear how genetically divergent the lake form is from the stream-dwelling form in Cultus Lake. It is possible, for example, that the two forms are part of the same genetic population and that the lake form results from dispersal from the stream-spawning habitats of coastrange sculpin. Alternatively, the lake form may be a genetically discrete population(s) that, coupled with its ecological specializations, is behaving as a species distinct from coastrange sculpin. These possibilities have direct bearing on the status of the lake form as a distinct DU under COSEWIC and SARA (COSEWIC 2010), yet, the Cultus pygmy sculpin’s validity as a DU has never been objectively assessed against the criteria used by COSEWIC (COSEWIC 2009 Operations and Procedures Manual Appendix F5). This is an 6  important limitation because although the Cultus pygmy sculpin is currently listed by SARA, such listings are reviewed at least every 10 years and the status may change, and eligibility of a wildlife species as a DU is one basis on which the legal status may be altered. Further, there have been no ecological studies on this fish since Ricker’s (1960) original work, and little of its life history is known.  It is believed that the population is stable and that reproduction is sufficient to replace the population, although spawning has not been observed, and the location of spawning and fate of the newly-hatched young is not known.  The Cultus Lake situation could be unique and significant from a conservation standpoint in that the Cultus pygmy sculpin is apparently endemic to BC, and it provides an example of rapid diversification within a species.  A similar fish, however, has been found in Lake Washington near Seattle, WA (Larson and Brown 1975). The evolutionary origin of these similar ecotypes, however, is uncertain. This other pelagic sculpin could share a common ancestor with the Cultus pygmy sculpin or it could have arisen through parallel evolution. In summary, the eligibility of the Cultus pygmy sculpin as a DU under SARA as well as the evolutionary origins of this fish are uncertain. My thesis sought to address these uncertainties and had the following overall objectives: 1. Determine the genetic distinctiveness of the Cultus pygmy sculpin from C. aleuticus found in tributary streams to Cultus Lake and elsewhere throughout its range, through the use of mtDNA and microsatellite (nuclear) DNA. 2. Map the habitat of the Cultus pygmy sculpin. 3. Using Fisheries and Oceans Canada trawl catch data and lab experiments, determine whether there are any behavioural differences between the Cultus pygmy sculpin and typical stream-dwelling coastrange sculpin. 7  Chapter 2: Evolutionary relationships and population structure of the Cultus pygmy sculpin, Cottus aleuticus  Introduction  Over the last thirty years, there has been an increase in the use of molecular genetic information to assess the conservation status of intraspecific units (Avise 2004). Although the morphology of the Cultus pygmy sculpin has been examined (Ricker 1960; COSEWIC 2010), no genetic assessments of its distinctiveness from typical coastrange sculpin or of its evolutionary origin have been conducted. Genetic distinctiveness, including differences in mtDNA sequences or allele frequency differences in microsatellites, is one of the criteria used to determine whether a population is a Designtatable Unit (COSEWIC 2009).  My study used mitochondrial DNA (mtDNA) and microsatellite (nuclear) DNA to determine the phylogenetic relationships and population structure of the Cultus pygmy sculpin, in comparison with C. aleuticus. Mitochondrial DNA is a small, haploid molecule that is inherited maternally (Avise et al. 1987; Moritz et al. 1987; Allendorf and Luikart 2007).  The analysis of mtDNA can be used to obtain information about the relationships between individuals and populations (Moritz et al. 1987).  There is no recombination (Aquadro and Greenberg 1983; Avise et al. 1987; Allendorf and Luikart 2007), and so it provides a clear record of genealogical descent (Allendorf and Luikart 2007).  The ultimate source of variation in the mtDNA sequence is mutation (Aquadro and Greenberg 1983; Avise 2004), and evolution at the nucleotide sequence level is rapid compared to most single copy 8  nuclear loci (Avise et al. 1987), which often allows for the resolution of intraspecific variation (Yokoyama and Goto 2002).            Within the genus Cottus, mtDNA has been used on a broad scale to examine the relationships among species and related taxa (Kinzinger et al. 2005; Yokoyama and Goto 2005) and to explore the diversification of a monophyletic group (Kontula et al. 2003).   At the species level, mtDNA has been used to examine genetic fragmentation, variation and phylogeography in Cottus nozawae (Okumura et al. 1996; Yagami et al. 2002; Yokoyama and Goto 2002).  The mtDNA sequences that have been used for Cottus include the ATPase 8 and 6 genes, the cytochrome b gene (Kontula et al. 2003), the control region, and 12S rRNA (Yokoyama and Goto 2005). For my study, I used the displacement loop (D-loop) or the “control region”.  This sequence functions in replication, and can change very rapidly (Moritz et al. 1987).  Sequences in the control region have been used successfully in Cottus to estimate genetic differentiation (Okumura et al. 1996) and reveal phylogenetic relationships (Yokoyama and Goto 2002; Kontula et al. 2003). If a population is genetically distinct from other populations, this satisfies one of the criteria for recognition as a DU under SARA (Guidelines for recognizing Designatable Units, COSEWIC 2009).  For example, sequences from the control region of mtDNA were used to examine the phylogeography of the sand goby, Pomatoschistus minutus (Stefanni and Thorley 2003).  The Adriatic population of the sand goby was found to be highly distinct from other populations and was therefore determined to be an Evolutionarily Significant Unit, or ESU, a concept analogous to that of DUs for many jurisdictions outside Canada (Moritz 1994).  Sequences including those from the mtDNA control region were used to distinguish the deepwater sculpin (Myoxocephalus 9  thompsonii) from the fourhorn sculpin (M. quadricornis); these results also identified two separate DUs for the deepwater sculpin in Canada (COSEWIC 2006).         The mutation rate of mtDNA is typically more rapid than that of many nuclear DNA loci, which has proven useful for resolving major evolutionary units within species (Moritz 1994); however, this rate of mtDNA mutation is relatively too low to disclose processes on a more recent time scale, such as since around the end of the last glacial maximum (~ 15,000 years before present, McPhail and Lindsey 1986; Hanfling et al. 2002).  Comparatively, microsatellites are more rapidly evolving loci that can be more informative over such short time scales (Hanfling et al. 2002).  Microsatellites are variable, non-coding regions of nuclear DNA (Hansen et al. 2001), which are composed of tandemly-repeated simple sequences (e.g., CACACA, GTTTGTTT, Jarne and Lagoda 1996).  Microsatellite loci are commonly used to understand population structure within species (Slatkin 1995), because they are highly polymorphic, even in small populations, due to a high mutation rate and because it is relatively easy to assay many unlinked loci (Allendorf and Luikart 2007).  The evolutionary divergences within species are assessed and the distribution of genetic variation within and between populations is analyzed using F-statistics (Jarne and Lagoda 1996; Rico et al. 1996).  In populations that display low genetic diversity (e.g., Canadian polar bears, Ursus maritimus, Paetkau et al. 1995;  European bullhead, C. gobio, Hanfling et al. 2002), microsatellites have been used to examine genetic population structure with greater success than mtDNA.  Englbrecht et al. (1999) successfully developed microsatellite loci for C. gobio, and these loci were also variable in related taxa.  These loci have been used successfully to look at genetic subdivision among populations (Whiteley et al. 2009) and diversity within populations (Taylor et al. 2010) in C. aleuticus. 10         In this chapter, I used a combined approach to assess the genetic distinctiveness of the Cultus pygmy sculpin. First, I used mtDNA data to reconstruct the relationships between sequences of the Cultus pygmy sculpin relative to that of the coastrange sculpin. I expected that if the Cultus pygmy sculpin is a valid DU, it should represent a monophyletic assemblage distinct from the coastrange sculpin or, failing this, that it should be significantly different in terms of the frequencies of shared mtDNA haplotypes. To these ends, I compared the mtDNA D-loop sequences of Cultus pygmy sculpin with coastrange sculpin from the inlet stream and other tributaries to Cultus Lake. I also included the Lake Washington pelagic sculpin, as well as samples from throughout the range of C. aleuticus to get an idea of how much sculpin samples from within the Cultus Lake drainage differ from one another relative to sculpins sampled from outside the drainage and to assess how different the Cultus pygmy sculpin is from a similar phenotype in Lake Washington.      Second, I used eight microsatellite loci to determine whether the Cultus pygmy sculpin is a population genetically distinct (i.e., with different allele frequencies; COSEWIC 2009) from C. aleuticus from the inlet stream and other tributaries to Cultus Lake.  For comparison with other coastrange sculpin populations, I included samples from Norrish Creek in the lower Fraser River (which is also part of the Fraser system, but is located on the north side of the Fraser River); three different rivers in the Squamish River watershed (to provide information on population differentiation in a small, interconnected geographic area); and pelagic sculpin from Lake Washington (to compare two different pelagic sculpin populations).  The use of microsatellites should provide a more sensitive test of the genetic “discreteness” criterion of DU status for the 11  Cultus pygmy sculpin, given that they probably have diverged from C. aleuticus only postglacially.  Materials and methods  Study area  Cultus Lake is located approximately 10 km south of Chilliwack, in the eastern Fraser River Valley, BC (Shortreed 2007) (Figure 3).  The surface area of the lake is 6.3 km2 and it has a mean depth of 31 m, with a maximum depth of 44 m (Shortreed 2007). Frosst Creek is the lake’s inlet stream and largest tributary; the lake discharges into Sweltzer River, which itself empties into the Chilliwack River (a tributary of the lower Fraser River) approximately 3 km downstream from the lake (Shortreed 2007).  Cultus Lake is warm, monomictic, and thermally stratified from May-November (Shortreed 2007). Cultus Lake is mesotrophic and there has been an increase in lake temperature over the past 70 years which has been accompanied by an increase in productivity (Shortreed 2007).  Pelagic C. aleuticus are also found in Lake Washington, which is located at Seattle, WA, and has a maximum depth of 65.2 m, with an average depth of 32.9 m (Ikusemiju 1975).  Sculpin samples  A total of 46 different samples from 12 different locations were used to obtain mtDNA sequences, including three from C. asper that was used as the outgroup (Table 1).  Ten populations consisting of a total of 350 samples from nine different locations 12  were used for microsatellite analysis (Table 2).  Samples were captured by trawl, minnow trap or electrofishing.  Mitochondrial DNA  Tissue samples were subject to Qiagen spin column-based DNA extraction procedures and the DNA was stored at –20oC.  The entire control region was amplified by PCR using the primers L-Thr (5’-AGCTCAGCGYCAGAGCGCCGGTCTTGTAA-3’) and H12Sr5 (5’-TGATAATAAAGTCAGGACCAAG-3’) (Yokoyama and Goto 2002). Amplification reactions contained 50-300 ng of genomic DNA template (Ilves 2007). The control region of mtDNA was amplified in a 50 L PCR containing: 10 x buffer (New England Biolabs, NEB), 5 mM of dNTPs, 10 M of primers L-Thro and H12Sr5 (Yokoyama and Goto 2002), and 1.5 units of Taq DNA polymerase (NEB).  Thermal cycling conditions began with 3 minutes at 95oC, 1 minute at 58oC, 1 minute at 72oC, with 5 cycles of 1 minute at 94oC, 1 minute at 58oC, and 1 minute at 72oC, 33 cycles of 92oC for 45 seconds, 56oC for 30 seconds, and 72oC for 1 minute, followed by a final extension step of 72oC for 10 minutes. PCR products were purified using the Qiagen QIAquick PCR purification kit and sequenced at the NAPS unit at the University of British Columbia on an ABI Prism 377 automated sequencer using L-Thr as the sequencing primer. Sequences were edited using BioEdit v7.0.9.0 (Hall 1999) and aligned using ClustalX (Thompson et al. 1997). MrModeltest v2.3 (Posada and Crandall 1998) was used to select a model for sequence evolution employing the Akaike Information Criterion (AIC) to select the best model (Posada and Buckley 2004).  PAUP* (Swofford 2002) was used for neighbour-joining tree analysis of pairwise genetic distances. 13  MrBayes v3.1.2 (Huelsenbeck and Ronquist 2001) was used for Bayesian estimates of phylogeny, with 1,000,000 generations run. Due to three samples (Lake Washington and two from Alaska) having shorter sequences, the analysis was run once with the more complete sequences (using 775 base pairs) and then again using all the sequences, with the longer ones truncated to 340 base pairs. Microsatellite analysis  A total of 350 fish were genotyped at eight Cottus gobio dinucleotide microsatellite loci (Nolte et al. 2005; Whiteley et al. 2009; Taylor et al. 2010).  These loci were amplified by PCR using fluorescently-labelled primers (Table 3a) according to Qiagen’s Multiplex PCR protocol for amplification of microsatellite loci using multiplex PCR (Taylor et al. 2010).  An annealing temperature of 60oC and 25 cycles were used in the amplifications (Table 3b, Taylor et al. 2010).  The PCR products were then diluted 1:20 before being assayed using a CEQ 8000 Genetic Analysis System and software (Beckman Coulter) with CEQ DNA Size Standard Kit (400 base pairs) used as internal size standard (Taylor et al. 2010). Convert (Glaubitz 2004) was used to convert the raw microsatellite data into the formats required by the different programs.  Microchecker v2.2.3 (Van Oosterhout et al. 2004) was used to test for the presence of null alleles and to detect any technical problems with the microsatellite results.  To obtain descriptive statistics, Tools for Population Genetic Analysis (TFPGA) (Miller 1997) provided observed (HO) and expected (HE) heterozygosities, sample size (N) and number of alleles. Fstat (Goudet 2001) was used to determine allelic richness (AR), which is a measure of allelic diversity that takes sample size into account (Allendorf and Luikart 2007); Fstat was also used to compare allelic richness among groups.   Genepop v3.4 (Raymond and Rousset 1995) 14  was used to detect any deviation from Hardy Weinberg Equilibrium (HWE) and linkage disequilibrium, and also to test for differences in FST values and genotypic differentiation between all pairs of populations.  Geneclass v2.0 (Piry et al. 2004) was used to investigate the origins of the fish caught in minnow traps by comparing them to reference samples from all sample localities within Cultus Lake.  Structure v2.3.3 (Pritchard et al. 2000) was used to estimate the number of distinct genetic clusters in the microsatellite data.  The burnin period was 50,000 iterations while the Markov Chain Monte-Carlo (MCMC) length was also 50,000.  There were ten separate runs conducted for each population estimate (K), and the average of each log likelihood value was used for each K estimate.  All eight populations were first tested, and then a nested analysis including only the three populations within the Cultus Lake drainage was performed. Arlequin (Schneider et al. 2000) was used to produce an analysis of molecular variance (AMOVA) that compared all sculpin populations sampled and those within the Cultus Lake drainage.  Factorial Correspondence Analysis (FCA) in Genetix (Belkhir et al. 2001) was used to visualize similarity among the C. aleuticus multilocus microsatellite genotypes (Taylor et al. 2010). The sequential Bonferroni correction was used to determine significance for tests that involved many simultaneous tests (Rice 1989).   Results  Mitochondrial DNA   The best model selected by AIC using MrModelTest was determined to be GTR + I + G (General Time Reversible model (GTR) (Tavaré 1986), with proportion of 15  invariable sites (I) and gamma distribution (G)). The GTR model assumes that the probability of the data is independent of the placement of the root on the tree; that the relative rates of substitutions remain constant across the tree; and that the four nucleotides remain at the same frequencies (Waddell and Steel 1997).  The substitution rate varies among sites in a gamma distribution (with my data, the shape of the gamma distribution was 1.0277) (Li 1997).  The proportion of invariable sites is the expected frequency of sites that do not evolve (with my data, I = 0.7046) (Waddell and Steel 1997).  This model was used to develop the Bayesian trees with likelihood values (Figure 4).  Two samples, Cultus 214 and Frosst 73, were more closely related to the C. asper group than to C. aleuticus; these two fish were removed from further analysis. The Cultus pygmy sculpin samples formed a paraphyletic grouping, and did not form a monophyletic grouping distinct from C. aleuticus (Figure 4). The average pairwise genetic distance amongst all samples and the outgroup C. asper was 0.048. The average pairwise divergence between C. aleuticus sampled from outside the Cultus Lake drainage and Cultus pygmy sculpin sequences was 0.010. The average sequence divergence between Cultus pygmy sculpin and coastrange sculpin sampled from other areas of the Cultus Lake drainage was 0.005.  When the truncated sequences, including Lake Washington and Alaska sculpins, were used (Figure 5), the sequence divergence of the Lake Washington fish was 0.014 from all other C. aleuticus, and 0.013 from Cultus pygmy sculpin.  Sequence divergence of C. aleuticus sequences from Alaska were 0.009 both from Cultus pygmy sculpin and all other C. aleuticus. 16   Microsatellite DNA  Across loci there were between four and thirty-seven alleles detected, with individual populations having between four and twenty-four alleles at each locus (Table 4).  Cottus aleuticus from the Cultus Lake streams and the Cultus pygmy sculpin both had lower allelic richness (p < 0.05) when compared to other C. aleuticus populations (Table 5).  No population failed to meet Hardy-Weinberg assumptions across all loci (Table 6).  Only 12 out of 280 tests showed evidence of departures from linkage equilibrium, but such departures tended to be idiosyncratic and not concentrated at particular loci or within particular populations (Table 7).  The lack of linkage disequilibrium could be due to the number of low frequency alleles as well as the high amount of variation observed in the limited sample size; to observe linkage disequilibrium, it would be necessary to compare areas matched for allele frequency (Eberle et al. 2006).  Microchecker identified locus CottES19 to be a possible locus with null alleles, associated with cases of heterozygotic deficiency/homozygote excess.  The analyses were conducted with and without this locus, and no differences between results were observed; therefore, this locus was included in the analyses.   FST values across population comparisons ranged from 0.001 to 0.06 (Table 8), while the overall FST value was 0.03.  There was no significant difference in FST between coastrange sculpin sampled from Watt and Windfall creeks, and so the two populations were combined.  There was no significant difference between Cultus pygmy sculpin caught using trawl or minnow traps, and so those two groups were also combined (Table 8).  Consequently, subsequent tests consisted of three groups of fish from the Cultus Lake drainage: Cultus pygmy sculpin sampled from the lake, the 17  Windfall and Watt creeks combined samples (“tributaries”), and the Frosst Creek sample. There was a significant difference in FST values between Cultus Lake pygmy sculpin and sculpin from Frosst Creek and the tributary samples although these differences were much smaller than FST values between populations sampled from different watersheds (Table 8).  Assignment tests showed that approximately half of the fish caught by minnow traps were assigned to the Cultus pygmy population, while the other half were assigned to the tributary populations (Table 9). The results from the Structure analysis indicated that the most likely number of genetic groups across all samples was three (Table 10).  Group 1 consisted of Cheakamus, Mamquam, Squamish rivers and Norrish Creek; Group 2 consisted of Cultus Lake, Frosst, Watt and Windfall creeks; and Group 3 consisted of the Lake Washington sample (Figure 6).  The nested Structure analysis indicated that within the Cultus Lake drainage, the most likely number of populations was two (Table 11).  Some individuals in the Cultus pygmy sculpin population (Group 1) are more strongly assigned to one population, compared to the more symmetrical assignments seen in the sculpins found in Frosst Creek (Group 2) and Watt and Windfall creeks (Group 3; Table 11; Figure 7).  Therefore, there does appear to be a difference between the Cultus pygmy sculpin population with the stream populations and population structure is typically concluded to be present when the proportions of the sample assigned to each population is asymmetric (Hubisz et al. 2009). The hierarchical analysis of genetic structure using AMOVA showed that about 2.0% (P < 0.001) of the total genetic variability could be attributed to differences among the Squamish River, Cultus, and “other” (Norrish Creek, Lake Washington) drainages 18  and the vast majority of remaining variation resided within populations (Table 12). The subtle genetic differences between Cultus pygmy sculpin and other C. aleuticus within the Cultus Lake drainage were also indicated when the AMOVA analyses were conducted within the Cultus Lake drainage alone. Here, Cultus pygmy sculpin were partitioned between trawl and minnow trap samples and were compared to stream fish partitioned across Windfall, Watt, and Frosst creeks. In the Cultus Lake drainage, only 1.2% of the variation (P > 0.05) was attributable to differences between lake and stream sculpins (Table 12). The FCA analysis amongst all samples (Figure 8) showed that the populations from the same drainages grouped together, while the Lake Washington and Norrish Creek populations formed two separate clusters.  These results agreed with the Structure analysis (Figure 6), in that the Lake Washington population was different from the other populations, while Norrish Creek was intermediate between the Squamish and Cultus drainage populations.  When the analysis was restricted just to the Cultus Lake samples, the FCA also suggested that there is genetic differentiation between the Cultus pygmy sculpin and the Frosst Creek and tributary sculpins (Figure 9).  The three populations do not form completely distinct clusters, but do show some segregation in FCA space (i.e., there is little overlap between the populations).  This association between genetic identity and geographic location is similar to the results from the Structure analysis (Figure 7), in that the three populations are distinct from each other, but the amount of divergence between the three populations is low.  19   Discussion  mtDNA inter-relationships  The results from the mtDNA analysis did not show genetic differentiation between the Cultus pygmy sculpin and other C. aleuticus populations.  Fish from northern populations in North America that recolonized habitats from southern refugia have lower genetic diversity, as well as broader geographic distributions than southern populations (Bernatchez and Wilson 1998; Hewitt 2000).  Cultus Lake (and nearby Lake Washington) only became accessible to fishes within the last 8 – 10, 000 years, after the Wisconsinan ice sheets retreated (COSEWIC 2010).  The recent emergence of the only known habitats presently occupied by pygmy populations of coastrange sculpin implies that these fish have had a postglacial origin as well. My mtDNA results are consistent with this idea. Sequence divergence estimates between Cultus pygmy sculpin and parapatric coastrange sculpin were very low and consistent with differences between populations that have only recently diverged from one another (e.g., Bernatchez and Wilson 1998; Taylor and McPhail 1999). In addition, the mtDNA haplotypes of Cultus Lake and coastrange sculpin were not reciprocally monophyletic, again, suggesting a recent divergence. The phylogeny of mtDNA in recently diverged species may be affected by retained ancestral polymorphisms in that after a divergence event, the relationships among the alleles proceed from polyphyly, through various paraphyletic conditions to monophyly, as ancestral polymorphisms are replaced by derived states (Moritz 1994). If the parent population is larger than the daughter population, some lineages in the parent population may share a more recent common ancestor with those in the 20  daughter population than in the rest of the parent population (Paetkau 1999).  Other species, including polar bears (U. maritimus, Talbot and Shields 1996), song sparrows (Melospiza melodia, Zink and Dittman 1993), and juncos (Junco spp., Mila et al. 2007) have demonstrated differences in phenotypes, which were not evident in their mtDNA haplotypes. To qualify as a Designatable Unit under COSEWIC guidelines, a population needs to show genetic distinctiveness, including, but not limited to, differences in DNA sequences and microsatellite allele frequencies (COSEWIC 2009).  For a population to be significant in terms of neutral loci, there must be qualitative genetic differences at slowly evolving markers (COSEWIC 2009).  The relatively recent, postglacial divergence of the Cultus pygmy sculpin from the parapatric C. aleuticus, as demonstrated by the mtDNA analysis, is such that there are no marked genetic differences that would qualify the Cultus Lake population as significant.  Therefore, I used microsatellite loci to look at population-level differentiation, as well as differences in behaviour and habitat.   Interestingly, of the forty-three samples that I identified by morphology as C. aleuticus, two samples were designated as C. asper by mtDNA.  While I used multiple criteria in order to identify each individual to species (Scott and Crossman 1973; McPhail 2007), it is possible that those two specimens could have been misidentified (i.e., if anal fin rays had been broken off).  It is also possible that this mismatch could result from hybridization, specifically C. asper females mating with C. aleuticus males (Taylor et al. 2010).  One of these fish was from Frosst Creek; the sampling site in Frosst Creek had a predominantly riffle habitat, preferred by C. aleuticus (Taylor 1963), but small numbers of C. asper were also caught during sampling.  The low numbers of 21  C. asper in the area would influence hybridization dynamics in that one might expect hybridization to be biased in favour of C. asper females mating with C. aleuticus males (Wirtz 1999; Taylor et al. 2010). The second fish was from Cultus Lake; C. asper is present in large numbers in all areas throughout the lake, and matures at a much larger size than C. aleuticus and the Cultus pygmy sculpin (McPhail 2007).  Hybridization can result from differences in sizes at maturity, with small males of one species mating with larger females (Wirtz 1999; Taylor et al. 2010). While hybridization events between sympatric species pairs of Cottus have been reported elsewhere in North America (McPhail 2007), the only other report of C. aleuticus x C. asper hybridization has been from the Cheakamus River (Taylor et al. 2010).  As exemplified by the suggested cases of hybridization described above, due to the peculiarities of mtDNA inheritance (no recombination, maternally-inherited), mtDNA-based phylogenies may not reflect the situation in the rest of the genome and I conducted microsatellite DNA analyses to address this limitation (Coyne and Orr 2004).  Microsatellite DNA differentiation   The FST values of the Cultus pygmy sculpin in comparison with sculpins from tributary streams within the drainage were less than 0.05, but statistically significant and exhibit genetic differentiation at the population level.  The samples were characterized by abundant genetic variation within samples, but low levels of divergence between samples. These general patterns are consistent with the few other studies that have been conducted on freshwater sculpins. For instance, Whitely et al. (2009) found little evidence for genetic subdivision in the Alaska populations separated by about 50 km 22  (FST < 0.001), while Taylor et al. (2010) found the fish in the three Squamish Rivers to have an overall FST of 0.003.  Consequently, compared with the low values observed in other systems, the differences that I have documented between Cultus pygmy sculpin and tributary fish are relatively large. There are several reasons why sculpins may exhibit a general pattern of low interpopulation variability. For instance, as newly hatched larvae, the coastrange sculpin (and many other species) have a pelagic dispersal stage in which they enter a larger body of water (such as an estuary, lake or ocean) and remain in the water column for up to a month before re-entering freshwater streams (Mason and Machidori 1976). This highly dispersive stage has been suggested to constrain the development of population subdivision at neutral loci (Whiteley et al. 2009).  Combined with the relatively recent time for colonization and the elevated level of dispersal, the amount of gene flow can be high enough to obscure the effects of genetic drift in driving molecular divergence at neutral loci (Whiteley et al. 2009).  The low level of sequence divergence between the coastrange sculpins in BC and Alaska also suggests that they survived the last glaciation in one refugium (McPhail and Lindsey 1970; Mila et al. 2007). Lake Washington fish, however, were highly distinct in the microsatellite analyses and this suggests that they have had a separate origin from those in Cultus Lake.  If so, this would suggest that the two known occurrences of pygmy sculpin are a result of parallel evolution, in that similar traits have evolved in similar habitats (e.g., Rundle et al. 2000). Both the Cultus pygmy sculpin and the Lake Washington pelagic sculpin live in the open waters of productive lakes, and have enlarged cephalic pores, shorter pelvic fins, and a higher number of pectoral fin rays (Larson and Brown 1975; COSEWIC 2010).  Other northwestern North American fishes have exhibited parallel evolution after glaciation, 23  including sockeye/kokanee (Oncorhynchus nerka, Taylor et al. 1996), threespine stickleback species pairs (Gasterosteus aculeatus, Taylor and McPhail 1999), and dwarf and normal lake whitefish (Coregonus clupeaformis) pairs (Pigeon et al. 1997). The FST values and results from Structure and the FCA showed that the Cultus Lake pygmy sculpin were genetically differentiated from the adjacent stream samples. The one apparently anomalous result, however, was that sculpins that I captured in minnow traps were found by the assignment tests to be comprised of both Cultus Lake and fish from the two tributaries, Watt and Windfall creeks.  The fish that were assigned to the tributaries had been sampled in July and tended to be from traps located near the shoreline, when the tributaries were either dry or the depth and flow were both greatly reduced.  One possible explanation for this result is that these fish actually belonged to the stream populations, but they had taken refuge in the lake.  Despite this spatial overlap in the summer, it is unlikely that there is gene flow between the stream and lake fish; the stream fish lay their eggs in spring, generally from late March to early April (Ricker 1960), allowing time for the young to hatch before the streams dry up, while the Cultus pygmy sculpin fish have a summer spawning season (mature fish were caught May–late July; also Ricker 1960). Alternatively, the assignment of some minnow trap fish both to Cultus Lake and tributary populations could stem from the relatively low assignment scores of the fish identified to the Cultus Lake population. These low assignment scores could be a result of the lower sample size of the Cultus reference population (n=30, compared with the tributary population, n=55).  In populations with low FST, good assignment scores require larger samples of individuals and loci (Cornuet et al. 1999; Bjornstad and Roed 2002).  The low FST between the Cultus Lake and tributary populations, combined with the high level of diversity observed in each 24  population, would mean that more of the variation would be captured with a larger reference population and would thus result in higher assignment scores (in this case, for the tributary population).   In summary, this chapter has presented evidence that Cultus pygmy sculpin shows genetic differentiation (at least at microsatellite DNA loci) from stream-dwelling fish and that this divergence has occurred postglacially. In addition, my data present the first evidence that the pygmy forms in Cultus Lake and Lake Washington have had independent origins and represent another example of parallel evolution in postglacial fishes. In the next chapter, I discuss the results from behavioural assays conducted in laboratory conditions and I then use Fisheries and Oceans Canada trawl data to define potential habitat requirements of pelagic sculpin. 25  Table 1. Location, species and number of samples of Cottus used for mtDNA D-loop analysis.  Location Species Watershed code Drainage N Cultus Lake Cottus aleuticus 100-065700-09700-13300 Chilliwack 14  C. asper 100-065700-09700-13300 Chilliwack 1 Frosst Creek C. aleuticus 100-065700-09700-13300- 9760 Chilliwack 10 Windfall Creek C. aleuticus 100-065700-09700-13300- 5370 Chilliwack 5 Watt Creek C. aleuticus 100-065700-09700-13300- 9100 Chilliwack 5 Sweltzer River C. aleuticus 100-065700-09700-13300 Chilliwack 1 Lake Washington C. aleuticus 17110012 Lake Washington 1 Norrish Creek C. aleuticus 100-06400 Harrison 1 Alouette Reservoir C. aleuticus 100-026700-06000 Lower Fraser 1 Skeena River C. aleuticus 400 Skeena 1 Bella Coola River C. aleuticus 910-290700 Bella Coola 2 Bartlett River, Alaska C. aleuticus  Glacier Bay 2 Kinbasket Reservoir C. asper 300 Columbia 2  26   Table 2. Population location and number of samples of Cottus used for microsatellite analysis.  Location Watershed code Drainage N Cheakamus River 900-097600-12900 Squamish 67 Mamquam River 900-097600-05100 Squamish 48 Squamish River 900-097600 Squamish 30 Cultus Lake (trawl caught) 100-065700-09700-13300 Chilliwack 30 Frosst Creek 100-065700-09700-13300-9760 Chilliwack 30 Lake Washington 17110012 Lake Washington 30 Windfall Creek 100-065700-09700-13300-5370 Chilliwack 25 Watt Creek 100-065700-09700-13300-9100 Chilliwack 30 Norrish Creek 100-06400 Harrison 29 Cultus Lake (minnow trap caught) 100-065700-09700-13300 Chilliwack 31  27   Table 3a. PCR strategy: repeat motifs and sequences of the primers used for microsatellite analysis.  Locus Repeat motif Primer sequences (5’-3’) Size bp Cott100 (TG)15 CTCATCGTGGTTTGATCGGTG CCGAGCGTGAGTCAGGCGTG 176 CottES19 (CT)11 CCCCTTCCGTGAGCGACG ATATCACAGCACTTTTGGGGATG 144 Cott127 (AC)14 TTCAGCTGCCTGTGTAAGGC TGTATGGAGCACTTGGGCTG 194 Cott687 (CT)17 TGTACCTAGTGAGCCTGCTG CGTCAGCATCTACTGGGCAG 144 LCE29 (TG)9(T)(TG)22(A)(GT)6 GTGGGGAGAATGACGGATGG TACACATGCATTTGGATTGACC 221 Cott635 (TC)8 AGAAGGGCAACATAAGGGCC GAGGCGAGTGCGATGCCACC 169 Cott255 (CT)10 TCACTACAGCCAGGTGTCTG GCATGTGCATGCCTCCAGAG 172 Cott582 (GT)39 AGCTCTGGATGAGGACTGTC TCTGTGAGGTTCAGTGCGAC 233  Table 3b. PCR strategy: the multiplex technique used for microsatellite analysis.  Multiplex Locus Beckman dye Range of sizes in base pairs Quadplex I Cott100 Cy5 165-233  CottES19 Cy5 130-166  Cott127 Cy5.5 183-211  Cott687 Cy5.5 129-153 Diplex I LCE29 Cy5.5 167-269  Cott635 Cy5 166-352 Diplex II Cott255 Cy5.5 148-214  Cott582 Cy5 173-207  28   Table 4. Number and range of sizes (in base pairs, bp) of alleles by locus and population for microsatellite analysis of Cottus samples.  Locus Cheakamus River Squamish River Mamquam River Cultus Lake trawl Frosst Creek Cott100 22(165-229) 16 (169-231) 21 (165-233) 11 (179-209) 10 (173-209) CottES19 12(132-166) 10 (132-158) 11 (132-166) 10 (130-158) 6 (132-146) Cott127 7(183-203) 5 (183-191) 6 (183-193) 5 (183-193) 5 (183-203) Cott687 8(129-147) 7 (131-153) 6 (131-145) 7 (137-145) 4 (139-145) Cott255 11(148-176) 7 (152-178) 9 (152-184) 4 (164-174) 7 (162-204) Cott635 14 (166-352) 11 (166-240) 13 (166-352) 11 (166-266) 10 (166-352) LCE29 23 (173-247) 21 (173-241) 24 (167-247) 14 (173-259) 9 (181-249) Cott582 9 (173-195) 6 (177-189) 8 (173-191) 7 (173-189) 6 (177-189)  Locus Lake Washington Windfall Creek Watt Creek Norrish Creek Cultus Lake traps Cott100 17 (173-229) 12 (175-209) 13 (173-213) 15 (169-225) 14 (175-211) CottES19 6 (132-158) 5 (132-146) 9 (132-158) 11 (132-162) 8 (132-154) Cott127 8 (183-211) 5 (183-193) 5 (183-195) 5 (183-191) 5 (183-195) Cott687 6 (129-145) 5 (137-145) 3 (139-145) 4 (131-145) 4 (139-145) Cott255 9 (162-192) 5 (162-174) 7 (162-184) 10 (152-214) 6 (152-174) Cott635 13 (166-320) 8 (166-352) 11 (166-236) 8 (166-236)  13 (166-352) LCE29 12 (173-259) 8 (181-249) 10 (179-249) 15 (173-269) 11 (175-251) Cott582 8 (177-191) 8 (175-189) 8 (177-195) 10 (173-207) 7 (177-189) 29  Table 5.  Descriptive statistics by locus and population for microsatellite analysis of Cottus samples.  N is the number of samples, HE is the expected heterozygosity, HO is the observed heterozygosity, and AR is allelic richness. Population Cott100 CottES19 Cott127 Cott687 LCE29 Cott635 Cott582 Cott255 Ave. Cheakamus N 66 67 65 66 61 64 66 63 65 HE 0.908 0.800 0.745 0.777 0.878 0.856 0.784 0.838 0.823 Ho 0.879 0.672 0.754 0.758 0.787 0.844 0.788 0.762 0.780 AR 13.850 8.055 5.955 6.580 13.872 9.993 6.496 8.461 9.158 Mamquam N 48 48 48 48 44 47 48 46 47 HE 0.919 0.823 0.720 0.734 0.893 0.873 0.821 0.812 0.824 Ho 0.833 0.750 0.688 0.729 0.909 0.766 0.708 0.848 0.779 AR 14.555 8.512 4.898 5.643 15.371 10.461 6.902 7.325 9.208 Squamish N 31 31 30 30 32 32 31 31 31 HE 0.882 0.830 0.618 0.796 0.904 0.868 0.805 0.775 0.810 Ho 0.774 0.613 0.733 0.733 0.906 0.750 0.742 0.839 0.761 AR 12.657 8.584 4.739 6.523 16.098 10.061 5.911 6.059 8.829 Cultus trawl N 27 27 27 27 27 28 30 30 28 HE 0.865 0.826 0.521 0.620 0.821 0.830 0.757 0.748 0.749 Ho 0.852 0.556 0.630 0.704 0.778 0.786 0.600 0.933 0.730 AR 9.985 8.501 4.254 4.579 11.162 9.567 6.302 4.000 7.294 Frosst N 27 27 21 21 28 29 27 26 26 HE 0.844 0.758 0.571 0.514 0.722 0.669 0.748 0.771 0.700 Ho 0.815 0.407 0.524 0.760 0.750 0.759 0.704 0.923 0.676 AR 9.049 5.861 4.614 3.935 7.371 7.347 5.495 6.192 6.233 Lake Wash N 26 25 21 25 27 27 28 27 26 HE 0.891 0.786 0.722 0.658 0.783 0.865 0.792 0.826 0.790 Ho 0.808 0.800 0.524 0.760 0.852 0.852 0.679 0.889 0.770 AR 13.835 5.679 7.206 5.040 9.512 11.059 7.630 8.315 8.535 Windfall N 23 24 23 23 22 23 23 23 23 HE 0.876 0.751 0.637 0.598 0.721 0.622 0.710 0.738 0.707 Ho 0.957 0.583 0.609 0.478 0.864 0.652 0.609 0.609 0.670 AR 10.883 4.708 4.478 4.475 7.223 7.187 7.090 4.936 6.373 Watt N 29 28 29 29 28 28 30 29 29 HE 0.878 0.783 0.596 0.561 0.736 0.656 0.779 0.754 0.718 Ho 0.931 0.607 0.655 0.690 0.750 0.714 0.667 0.724 0.717 AR 10.839 7.515 4.169 2.975 7.829 8.660 6.914 6.103 6.876 Norrish N 26 28 24 26 17 19 27 25 24 HE 0.896 0.847 0.578 0.654 0.863 0.850 0.833 0.810 0.791 Ho 0.923 0.714 0.625 0.615 0.824 0.842 0.593 0.720 0.732 AR 12.802 9.582 4.412 3.885 15.000 7.982 8.468 8.593 8.841 Cultus traps N 27 28 26 26 24 26 24 24 26 HE 0.892 0.769 0.663 0.570 0.771 0.792 0.736 0.760 0.744 Ho 0.963 0.607 0.769 0.539 0.667 0.846 0.708 0.792 0.736 AR 12.168 7.037 4.617 3.617 9.219 10.420 6.315 5.627 7.378 30  Table 6. Results of tests of Hardy-Weinberg Equilibium by locus and population for microsatellite analysis of Cottus samples.  Underlined probability values indicate significant differences from Hardy-Weinberg expectations.  Population Cott100 CottES19 Cott127 Cott687 LCE29 Cott635 Cott582 Cott255 Cheakamus 0.5628 0.0333 0.7998 0.2940 0.3135 0.0466 0.4442 0.1249 Squamish 0.3926 0.2005 0.7923 0.6288 0.3620 0.0149 0.0900 0.5229 Mamquam 0.2071 0.0014 0.3019 0.0848 0.7268 0.2310 0.1623 0.9311 Trawl 0.5396 0.0024 0.5419 0.4077 0.7882 0.0392 0.0445 0.1181 Frosst 0.2417 0.0002 0.5498 0.3199 0.4550 0.4240 0.0017 0.3131 Lake Wash 0.0124 0.0889 0.0924 0.1924 0.4880 0.3292 0.0237 0.2422 Windfall 0.9103 0.1130 0.8011 0.1720 0.7390 0.8853 0.1839 0.4222 Watt 0.1301 0.0181 1.0000 0.4333 0.6977 0.6532 0.0738 0.1526 Norrish 0.6473 0.0472 0.6549 0.8342 0.4457 0.1885 0.0270 0.0209 Traps 0.4819 0.1223 0.3585 0.7819 0.3594 0.5462 0.5091 0.5056  31   Table 7. Results from tests of linkage disequilibrium for all populations, locus by locus, for microsatellite analysis of Cottus samples.  There was no significant result over all populations, following Bonnferroni correction.  Locus  Locus P value Cott100 CottES19 0.071 Cott100 Cott127 0.380 CottES19 Cott127 0.895 Cott100 Cott687 0.881 CottES19 Cott687 0.361 Cott127 Cott687 0.036 Cott100 LCE29 0.987 CottES19 LCE29 0.956 Cott127 LCE29 0.980 Cott687 LCE29 0.416 Cott100 Cott635 0.986 CottES19 Cott635 0.943 Cott127 Cott635 0.999 Cott687 Cott635 0.423 LCE29 Cott635 0.900 Cott100 Cott582 0.813 CottES19 Cott582 0.992 Cott127 Cott582 0.555 Cott687 Cott582 0.959 LCE29 Cott582 0.648 Cott635 Cott582 0.854 Cott100 Cott255 0.567 CottES19 Cott255 0.666 Cott127 Cott255 0.981 Cott687 Cott255 0.306 LCE29 Cott255 0.948 Cott635 Cott255 0.702 Cott582 Cott255 0.430 32   Table 8. FST values for eight populations, using eight loci for microsatellite analysis of various Cottus samples.  Underlined values are not significantly different from 0. The population in boldface represents the Cultus Lake pygmy sculpin. Samples in italics are found within the Cultus Lake drainage. Cheak = Cheakamus river, Mam = Mamquam river, Squam = Squamish river, Cultus = samples from fish caught by trawl and minnow trap in Cultus Lake, Wash = Lake Washington, Tributaries = Watt and Windfall creeks  Population Cheak Mam Squam Cultus Frosst Wash Tributaries  Mamquam 0.0037 Squamish 0.0010 0.0030 Cultus  0.0384 0.0277 0.0445 Frosst Creek 0.0475 0.0438 0.0577 0.0111 Lake Washington 0.0425 0.0314 0.0459 0.0405 0.0552 Tributaries 0.0511 0.0466 0.0536 0.0078 0.0069 0.0655 Norrish 0.0113 0.0062 0.0106 0.0222 0.0346 0.0367 0.0414        33  Table 9.  Assignment test results for individual C. aleuticus caught by minnow trap in Cultus Lake. Tribs=Tributaries (Watt and Windfall Creeks), Cultus=Cultus Lake pygmy population. “Percentage” refers to the relative level of confidence of assignment to the most likely population for each individual fish. Most likely population Percentage Month captured Cultus 55.7 May Cultus 51.2 May Cultus 85.2 May Cultus 62.3 May Cultus 86.4 June Cultus 55.3 July Cultus 65.5 July Cultus 91.0 July Cultus 92.4 July Cultus 74.6 July Cultus 69.5 July Cultus 57.6 July Cultus 83.2 July Cultus 59.3 July Tribs 90.9 July Tribs 89.9 July Tribs 98.2 July Tribs 95.6 July Tribs 76.4 July Tribs 96.7 July Tribs 61.4 July Tribs 94.0 July Tribs 94.8 July Tribs 99.2 July Tribs 76.7 July Tribs 68.1 July Tribs 96.0 July Tribs 86.1 July Tribs 91.5 July Tribs 59.0 July 34  Table 10.  Structure results for microsatellite analysis across the eight original sample localities (Cheakamus, Mamquam and Squamish Rivers; Cultus Lake and its three tributary streams: Frosst Creek and combined Watt and Windfall creeks; Lake Washington; and Norrish Creek). The K values are an average of ten separate runs. The value in bold and underlined represents the most likely number of populations.  K Log likelihood 1 -10597.9 2 -10282.6 3 -10167.1 4 -10204.3 5 -10291.8 6 -10357.8 7 -10393.0 8 -10444.2 9 -10424.5 10 -10559.8  35   Table 11.  Structure results for microsatellite analysis, across the three localities within the Cultus Lake drainage (Cultus Lake; Frosst Creek; and combined Watt and Windfall creeks).  The K values are an average of ten separate runs.  The value in bold and underlined represents the most likely number of populations.   K Log likelihood 1 -3611.67 2 -3585.12 3 -3622.12 4 -3641.41 5 -3673.02 6 -3655.73 7 -3633.07 8 -3631.91 9 -3655.49 10 -3662.75   36    Table 12. Analysis of molecular variance of microsatellite DNA allele frequencies.  The first analysis (Squamish vs. Cultus vs. others) compared the three Squamish drainage rivers (Cheakamus, Squamish and Mamquam) with Cultus Lake and its three tributaries (Frosst, Watt and Windfall creeks) and also with the other two systems (Norrish Creek and Lake Washington).  The second analysis compared Cultus pygmy sculpin (caught by trawl and minnow trap in Cultus Lake) with its three tributaries (Frosst, Watt and Windfall creeks).  Grouping hypothesis Variance component Sum of squares Percent variation P value Squamish vs. Cultus vs. other Among groups 25.625 2.10 0.0029  Among populations within groups 19.576 0.46 0.0000  Within populations 1462.821 97.45 0.0000 Cultus lake vs. streams Between groups 4.602 1.24 0.3402  Among populations within groups 3.313 0.00 0.0166  Within populations 477.133 98.77 0.0098  37   Figure 1.  Global distribution of the coastrange sculpin based on 262 occurrence records obtained from the WFC (2006) and GBIF (2006) (from COSEWIC 2010). North Pacific Ocean Canada USA 38   Figure 2.  A stream ecotype C. aleuticus (top), from Little Campbell River, compared with a Cultus pygmy sculpin (bottom).  Both were identified as mature females by the presence of eggs. 39    Figure 3. Cultus Lake bathymetric map (10 m intervals), modified from Shortreed (2007). Inset shows location of Cultus Lake within the boxed area in British Columbia. Mature Cottus were identified by the presence of eggs or testes. Trawl transects are denoted by T1-T4 (i.e., Trawls 1-4). 40   Figure 4. Phylogenetic analysis of mtDNA D-loop across 775 nucleotides under a GTR+I+G model of sequence evolution. Numbers represent the Bayesian posterior probability (%, 10,000 trees). Asper = Cottus asper. Sequences annotated with a filled square are Cultus pygmy sculpin, those with an open square are C. aleuticus from Cultus Lake tributary streams. 41   Figure 5. Phylogenetic analysis of mtDNA D-loop across 340 nucleotides under a GTR+I+G model of sequence evolution. Numbers represent the Bayesian posterior probability (%, 10,000 trees). Asper = Cottus asper. Sequences annotated with a filled square are Cultus pygmy sculpin, those with an open square are C. aleuticus from Cultus Lake tributary streams. 42     Figure 6.  Structure results across eight localities and K = 3 genetic groups:  green, blue, and red.  Each fish is represented by a thin vertical line composed of one to three colours each of which represents the proportional contribution of each genetic group to a fish’s genetic composition across eight microsatellite loci. In order from 1-8: 1 = Cheakamus River, 2 = Mamquam River, 3 = Squamish River, 4 = Cultus pygmy sculpin (trawl and minnow trap fish combined), 5 = Frosst Creek, 6 = Lake Washington, 7 = the two tributary streams (Watt and Windfall creeks), and  8 = Norrish Creek.  43     Figure 7.  Structure results for Cultus Lake drainage samples and K = 2 genetic groups: red and green shading.  Each fish is represented by a thin vertical line composed of one to two colours each of which represents the proportional contribution of each genetic group to a fish’s genetic composition across eight microsatellite loci. In order from 1-3: 1 = Cultus Lake pygmy sculpin (trawl and minnow trap fish combined), 2 = Frosst Creek, and 3 = the two tributary streams (Watt and Windfall creeks).  44    Figure 8.  Factorial Correspondence Analysis (FCA) of variation across eight microsatellite loci among 350 Cottus samples (including C. aleuticus and the Cultus pygmy sculpin) collected from nine locations in 2006-2008.  The three Squamish drainage populations, from left to right in the Squamish circle, include the Mamquam River (blue), the Squamish River (white), and the Cheakamus River (yellow). The three Cultus drainage populations, from left to right in the Cultus circle, include the two tributary streams (Watt and Windfall creeks) (brown), Frosst Creek (pink), and the Cultus pygmy sculpin (grey). 45    Figure 9. Factorial Correspondence Analysis (FCA) of variation across eight microsatellite loci between Cultus pygmy sculpin (N=61; yellow); Frosst Creek C. aleuticus (N=30; blue); and the two tributary streams (Watt and Windfall Creeks) (N=55; white). 46  Chapter 3: Habitat use and evidence for behavioural distinctiveness of the Cultus pygmy sculpin  Introduction   For a population to be a designatable unit (DU) under Canada’s Species-at-Risk Act (SARA), it needs to have attributes that make it “discrete” and “evolutionarily significant” relative to other populations (COSEWIC 2009).  One of the main criteria used to assess a DU is evidence of genetic distinctiveness, including inherited quantitative traits and neutral genetic markers (COSEWIC 2009).  Quantitative traits include differences in morphology, life history and behaviour. These are important to assay when possible because molecular markers do not measure all potential genetic differences (e.g., behaviour was used to show reproductive isolation in cichlid fishes, Stauffer et al. 1995).  Once a population is considered discrete, the significance of such discreteness to the evolutionary legacy of the species is then assessed. One way to assess the significance is if the discrete population persists in “an ecological setting that is unusual or unique to the species, such that it is likely to have given rise to local adaptations” (COSEWIC 2009).  For my study, I compared habitat data from Cultus Lake with other lakes within the C. aleuticus range, and also conducted depth selection experiments in the lab to test the hypothesis that the limnetic life history of Cultus pygmy sculpin is associated with distinct habitat features of Cultus Lake and with differences in behaviour that distinguish Cultus pygmy sculpins from typical stream- dwelling forms. 47   Habitat  Sculpins (Cottus spp., or “cottids”) are an important component of freshwater ecosystems in northwestern North America (Tabor et al. 2007), and in some habitats can be the most abundant species of fish (Bond 1963). For example, 73% of the fish biomass in Lake Washington has been estimated to be comprised of cottids (Eggers et al. 1978).  When more than one species is present, sculpins tend to be spatially segregated; this segregation can occur both in different stream habitats (i.e., Taylor 1963) and at different depths throughout a lake (Madenjian and Bunnell 2008).  Coastrange sculpin (C. aleuticus) are most commonly found in riffle habitat in streams (Taylor 1963; Mason and Machidori 1976; Ringstad 1982; White and Harvey 1999), but have also been captured along the shores of lakes (Tabor et al. 2007).  In streams, prickly sculpin (C. asper) tend to predominate in pool habitats, but are also found in the deep benthic habitats of lakes (Tabor et al. 2007).  It is believed that both species have a planktonic larval stage; however, previous studies on both species have been limited to a few streams in coastal BC, and much remains unknown (McPhail 2007).  The range of C. aleuticus extends along the mainland coast of BC, where the lakes tend to have a low dissolved solid content, which suggests that they are oligotrophic (Northcote and Larkin 1963).  Total dissolved solid content of the water appears to be the most important single factor determining the general level of productivity in lakes throughout BC (Northcote and Larkin 1963).  Pelagic sculpin, however, have been found in the deep offshore areas of productive lakes (i.e., Lake Washington and Cultus Lake; Arhonditsis et al. 2003; Shortreed 2007), in the presence of large numbers of C. asper (Tabor et al. 2007). 48   Behaviour  An important goal of evolutionary biology is to investigate the processes that lead to the divergence of natural populations along differential axes of resource exploitation (Schluter 1993; Rogers et al. 2002).  Adaptive radiation (phenotypic divergence often leading to speciation) is the outcome of divergent natural selection, which can be promoted by resource heterogeneity and competitive interactions (Schluter 1993; Taylor and McPhail 1999; Rogers et al. 2002).  The development and maintenance of specializations in water-column use and diet are important in the understanding of the initial stages of population divergence in freshwater fishes (Schluter 1993; Rogers et al. 2002).  Adaptive behavioural differences have been observed between populations in terms of habitat isolation and predator avoidance (e.g., Rogers and Bernatchez 2007).   In contrast to C. aleuticus, the Cultus pygmy sculpin has only been found in the deep, offshore areas of Cultus Lake, similar to the pelagic C. aleuticus found in Lake Washington (Ikusemiju 1975; Larson and Brown 1975; Tabor et al. 2007).  Ikusemiju (1975) suggested that the Lake Washington pelagic sculpins use the water column for feeding on planktonic invertebrates and predator avoidance.  Cultus Lake has the highest productivity of any sockeye salmon nursery lake in B.C., with an unusually high proportion of Daphnia that are present throughout the year and occur below the thermocline (Shortreed 2007).  Daphnia spp. are planktonic cladocerans and are the main food item of the Cultus pygmy sculpin (Ricker 1960).  The deep benthic areas of Cultus Lake also support C. asper, a bottom-dwelling ambush feeder (Broadway and Moyle 1978) that feeds on smaller fish (Gadomski and Parsley 2005; McPhail 2007). The morphology of the Cultus pygmy sculpin includes larger cephalic pore sizes, 49  shorter pelvic fins, and higher pectoral fin ray counts, which could be adaptations for sensory perception and for maintaining position in the water column during a planktivorous lifestyle (COSEWIC 2010). Similar morphological characteristics are exhibited by pelagic C. aleuticus in Lake Washington (Larson and Brown 1975) and by the highly specialized limnetic sculpins of Lake Baikal (genera Comephorus and Cottocomephorus, Kontula et al. 2003).  Behavioural differentiation can be influenced by the niches occupied by different ecotypes (i.e., limnetic vs. benthic, stream vs. lake) and the difference in diet that can result (Rogers et al. 2002).  The limnetic ecotype present in many freshwater fishes is generally smaller, feeds on pelagic prey, and should therefore occupy higher positions in a water column (e.g., McLaughlin et al. 1994, Rogers et al. 2002).  The Cultus pygmy sculpin is smaller and consumes Daphnia, a pelagic invertebrate (Ricker 1960), and thus I predicted that it should show behavioural differences that are associated with greater use of the water column in Cultus Lake  relative to typical stream-dwelling forms of C. aleuticus. The Cultus pygmy sculpin is listed as threatened under SARA, but its qualification as a valid DU has been more inferred than demonstrated. Behavioural studies have been used to distinguish between species, and populations that exhibit distinctive behaviours have qualified as Evolutionarily Significant Units (ESU, analogous to DU; e.g., Stauffer et al. 1995).  My study used data collected from the field and behavioural assays in the lab to determine whether the Cultus pygmy sculpin does inhabit a unique ecological setting, with different behaviour than C. aleuticus found in 50  streams. The data in this chapter, therefore, complement the molecular genetic assessments of distinctiveness that I conducted in Chapter 2.  My first objective was to determine the general habitat distribution of the Cultus pygmy sculpin in Cultus Lake.  There have been no studies conducted on finding the habitat of the Cultus pygmy sculpin; Ricker (1960) believed the pygmy sculpin to be found offshore because it was only found in the stomachs of a deepwater, offshore fish (bull trout).  The Cultus pygmy sculpin has been captured as incidental bycatch in DFO trawl surveys; these surveys were focussed on catching sockeye salmon and also swept a distance of approximately 1.5 km, so the precise location of capture would be unknown. My second objective was to determine whether there were any behavioural differences between the lacustrine Cultus pygmy sculpin and the fluvial coastrange sculpin.  I examined DFO trawl survey records, to determine whether there were any other examples of pelagic sculpin in other sockeye lakes in BC.  I also conducted a behavioural study of depth selection in the laboratory, to determine whether any behavioural differences observed in the field could be quantified.  Materials and methods  Fish samples   Fish were collected from the offshore area of Cultus Lake, and from Frosst Creek and the Little Campbell River (Table 13, Figure 3, Appendix 1).  Frosst Creek is the main inlet to Cultus Lake (Figure 3) and the Little Campbell River flows into Semiahmoo Bay, just south of the Fraser River estuary.  The streams were sampled using an 51  electrofisher (Smith-Root Model 12-B).  The shoreline was sampled using minnow traps deployed from shore, while the pelagic area was sampled using minnow traps suspended off the bottom of the lake by a float, 10 m apart, and at depths that ranged from 10-40 m.  The minnow traps were baited with sockeye salmon (as sockeye salmon carcasses are present in the lake at spawning time) and Cyalume yellow 12 hour “glow sticks” which have been shown to help attract deepwater sculpin, Myoxocephalus thompsonii (Sheldon et al. 2008).  The collection localities were determined using a GPS unit (Garmin 12 GPS) and mapped using ArcMap in ArcGIS (ESRI) (Figure 3). Mature female fish were identified by the presence of eggs.  One mature male was identified by the presence of testes. The Little Campbell River was chosen as a population outside Cultus Lake that represents the typical stream-dwelling form of C. aleuticus; as the river flows directly into the Pacific Ocean, it is unlikely there is any gene flow between sculpins from Little Campbell River and the Cultus Lake drainage. To sample fish for the behavioural assays, the minnow traps in Cultus Lake were deployed in the same area as that swept by the trawl net on the Fisheries and Oceans Canada (DFO) juvenile sockeye salmon trawl surveys (Figure 3), and the fish were captured in May, September and October. I did not sample fish during low water periods in the summer in order to minimize the capture of tributary fish that appear to take refuge in the lake when their streams dry-up (Chapter 2).  Live fish were brought back to an environmental chamber in the Biological Studies building at the University of British Columbia (UBC), and were kept in 136 litre aquaria, with approximately 10-15 fish per tank.  Fish from Cultus Lake, Frosst Creek, and the Little Campbell River were maintained in separate aquaria at 14-16 degrees C for about one month before the behavioural experiments began. 52   Behavioural assays  Two “drop tanks” 1.8 m in height and 0.3 m square were used for the depth selection experiments.  For all experiments, the tanks were placed in an environmental chamber with white blinds on three sides (Houtman and Dill 1994).  The temperature in the environment chamber was maintained at 14-16 degrees C. In the first experiment, each trial consisted of one fish being placed in each of the two drop tanks (the “focal” fish).  These fish were allowed to acclimate to the tanks for approximately 30 minutes while the camera was set up, and were then filmed with a web cam (Logitech QuickCam) attached to a laptop (Toshiba Windows XP) for 30 minutes.  In the second experiment, one sculpin from one of the three populations and one C. asper were placed in a tank together. The prickly sculpin (C. asper) will predate on smaller sculpins and can out-compete them for territory (McPhail 2007).  The average size (total length) of the Cultus pygmy sculpins used in the experiments when they were alone was 55 mm (SD = 4.2) and their average size when studied with C. asper was 62 mm (SD = 8.3) (the increase in size was due to growth while in the laboratory aquaria, and due to mortality of the smaller fish). The average size of the Frosst Creek sculpins used in the experiments when they were alone was 62 mm (SD = 8.4) and their average size when studied with C. asper was 74 mm (SD = 5.5). The average size of the Little Campbell River sculpins used in the experiments when they were alone was 86 mm (SD = 11.8) and their average size when studied with C. asper was 85 mm (SD = 10.0) (the smaller average size was due to mortality). The average size of C. asper used in the experiments with the three sculpin populations was 112 mm (SD = 2.1). In all experiments and trials, the depth at which each focal fish was observed was recorded 53  every minute, and then averaged for each fish across 30 measurements per trial (Rogers and Bernatchez 2007).  For each observation, I also recorded the number of times a focal fish was observed to be off the bottom of the tank, regardless of the actual depth selected. The paired data (without predator – with predator) for depth selected and counts off the bottom for each population from the two experiments were compared using the paired t-test (Whitlock and Schluter 2009) in R v2.8.1 (R Development Core Team 2008).  A single factor ANOVA comparing the differences between treatments among the three populations was used to determine whether the depth selection and number of times off the bottom were significantly different among the treatment groups of fish (Whitlock and Schluter 2009).  The Tukey-Kramer test was used to determine which pairs of populations had behaviours significantly different from the others (Whitlock and Schluter 2009).  Results  Habitats used by sculpins in Cultus Lake Cultus pygmy sculpin were caught in minnow traps set offshore in the benthic habitat rather than in the littoral habitat of the lake (Figure 3, Appendix 1).  Most fish were captured either on the bottom or within 10 m of the bottom, with some fish being caught up to 30 m above the bottom (Appendix 2). Cultus pygmy sculpin were captured every month from May-October, 2008 (Table 14, Appendix 2).  The highest catches occurred in July and August, most likely due to recruitment from young-of-the-year as the average total length of the fish was lowest during this period (Table 14).  Prickly 54  sculpin (C. asper) were also captured in minnow traps, along with the Cultus pygmy sculpin and, in two cases, large prickly sculpin had pygmy sculpin protruding from their mouths.  Mature Cultus Lake pygmy sculpin, including eight females and one male, were captured from May-July, 2008 (Figure 3). Midwater trawl surveys for sockeye were conducted by DFO in 62 lakes in BC (eg., Hume and MacLellan 2000 for methods). These data indicated that in only three of these lakes were mature-sized C. aleuticus fish (i.e., > 40 mm; Ruzycki et al. 1998) caught in the water column (Table 15).  Sculpins were caught more often in Cultus Lake, and the catches consisted of a range of size and age classes, compared to the other two lakes.  The species identification of sculpin in Red Bluff Lake has not been confirmed, however, this lake is within the C. aleuticus range.  Both Red Bluff and Pitt Lakes are deeper than Cultus Lake (80 m and 143 m maximum depths, respectively) and are also less productive (Slaney 1988; Bodtker et al. 2007), making them less likely to support a pelagic population.  The number of mature-sized fish in Cultus Lake is a conservative estimate, as only an average and range of sizes were included, due to the large number of sculpin caught in each trawl; the lengths of all fish were recorded for the other systems, due to the small number of sculpins caught per survey, and so these numbers are an exact count.  Large numbers of mature-sized sculpin have only been caught in the water column of Cultus Lake, compared with the other sockeye lakes surveyed by DFO.  Depth selection of sculpins When the fish were first introduced to the tank, their behaviour was variable, ranging from immediately heading to the bottom and staying there, to swimming around 55  at different depths in the tank.  No trends in initial behaviour were observed for any population (i.e., each individual fish seemed to have its own behavioural pattern). There was no apparent difference in behaviour between 0-30 minutes (after camera set-up) and 30-90 minutes, so the filming took place in the first thirty minutes (Rogers et al. 2002; Rogers and Bernatchez 2007). In comparisons within each population, the Cultus pygmy sculpin population did not show a significant difference in its average depth selection (mean difference =- 10.76cm, degrees of freedom (df) = 22, sample size (n) = 23, p = 0.3020) or the number of times off the bottom (mean difference = 0.78, df = 22, n = 23, p = 0.75) when the predator was present or absent (Table 16, Figures 10-13).  However, Frosst Creek and Little Campbell River fish exhibited significantly lower depth selection in the presence of the predator and also went off the bottom a significantly fewer number of times (Table 16, Figures 10-13).  Frosst Creek fish had a more significant change in behaviour (mean difference in depth = 24.35, df = 30, n = 31, p = 0.0067; mean difference in times off the bottom = 7.52, df = 30, n = 31, p = 0.00011) than did the fish from Little Campbell River (mean difference in depth = 2.63, df = 11, n = 12, p = 0.038; mean difference in times off the bottom = 1.92, df = 11, n = 12, p = 0.043). The differences in depth distribution and number of times fish were observed off the bottom were significantly different between the three populations (p = 0.0049 and p = 0.029, respectively).  Cultus pygmy sculpin and fish from Frosst Creek had the most significant differences in their depths (p = 0.0038) and the time spent off the bottom (p = 0.033).  Fish from Little Campbell River also differed from Cultus pygmy sculpin in depth (p = 0.024) and number of times off the bottom (p = 0.038).  However, the behavioural 56  changes between Frosst Creek and Little Campbell River fish were not significantly different for their depth and time off the bottom (p = 0.22 and p = 0.20, respectively).  Discussion  Pelagic coastrange sculpin have been found in two productive lakes.  Northcote and Larkin (1956) found a positive relationship between abundances of plankton and fish with the amount of total dissolved solid content in lake waters, with the relationship strongest in lakes of low mean depth (Northcote and Larkin 1963).  The majority of coastal lakes in BC have low dissolved solids and a low level of oxygen depletion, which characterizes oligotrophic lakes (Northcote and Larkin 1963).  Cultus Lake, however, has the highest amount of total dissolved solids recorded of any BC sockeye salmon nursery lake (Shortreed 2007) and is classified as mesotrophic.  By comparing Lake Washington and Cultus Lake with other lakes found within the range of C. aleuticus, it appears that pelagic sculpins tend to be found in shallow (< 70 m maximum depth, ~ 30 m mean depth), productive (mesotrophic) lakes. The low number of such lakes along the BC coastline that overlap with the distribution of C.aleuticus perhaps explains why pelagic sculpins are found only in Cultus Lake in BC. Studies comparing different ecotypes have found differences in depth selection or habitat in correlation with preferred prey (Schluter 1993; Skulason et al. 1996; Rogers et al. 2002).  Rogers et al. (2002) found that dwarf whitefish (Coregonus clupeaformis) stayed higher off the bottom in laboratory experiments than the normal form; this result was expected, as the dwarf ecotype was smaller and had a diet of limnetic zooplankton prey.  When a benthic predator (C. asper) was introduced to the tank, the smaller Cultus Lake sculpin did have a higher average depth selection than 57  the other two populations of fish and remained up in the water column instead of decreasing its depth and movement.  Smaller benthic fish in streams will leave deeper pools and head to shallower water in the presence of predators (Magoulick 2004); smaller fish also tend to get displaced from refuges by larger species (Rahel and Stein 1988; McNeely et al. 1990).  In comparison to the stream populations, the different behaviour of the Cultus pygmy sculpin could, therefore, be a result of a response to move to shallow (closer to the surface) water to escape predation, or a response to being forced off the bottom by the larger and perhaps more aggressive prickly sculpin. Larger competitors and predators are both capable of forcing smaller fish out of deep water habitats (Schlosser 1987; Lonzarich and Quinn 1995); fish are found to occupy deeper habitats as they increase in size (Lonzarich and Quinn 1995), which could explain why the Cultus pygmy sculpin, which were the smallest fish tested, were found more often in the water column than fish from the other, larger populations. Adaptation to local environments is a major factor promoting divergence among populations (Taylor 1991; Cousyn et al. 2001; Kawecki and Ebert 2004).  For example, changes in fish predation levels have been shown to increase genetic differentiation in a predator-avoidance trait in Daphnia magna (e.g., Cousyn et al. 2001).  Predator presence has also induced a change in morphology in Crucian carp (Carassius carassius), resulting in an increased locomotor response (e.g., Domenici et al. 2007). Differences between benthic-limnetic morphs in a variety of species complexes have been shown to be genetically based (Schluter 1993; Skulason et al. 1996; Rogers et al. 2002).  The morphological differences in the Cultus pygmy sculpin (COSEWIC 2010), combined with the predation pressures unique to their lacustrine life history, could result 58  in the observed difference in escape response when confronted with a larger, benthic predator. Behavioural characters can be used to identify different taxa (Trewavas 1983) and to estimate the phylogenetic relationships amongst taxa (Esteve and McLennan 2007).  In some cases, behaviour has been shown to be as useful as morphology in estimating phylogeny (de Queiroz and Wimberger 1993) and can be used to resolve ambiguities in molecular data (Esteve and McLennan 2007).  Ecologically significant traits, such as changes in behaviour, can lead to divergence through natural selection (Schluter 2001), and can be present even when there is no difference in allele or haplotype frequency at neutral loci between populations (e.g., Legge et al. 1996; Fraser and Bernatchez 2001).  Individual genetic strains of paradise fish (Macropodus opercularis) have been characterized by stable, heritable differences in behaviour (Gervai and Csanyi 1985).  The Cultus pygmy sculpin displayed a different behaviour in the presence of a predator than two different stream populations of C. aleuticus, which could indicate an adaptive difference to its environment and associated genetic change. Fish such as sculpins that have cryptic colouration typically respond to predators by reducing their movements (McNeely et al. 1990).  The saddle patches on freshwater sculpins (Armbruster and Page 1996), combined with their ability to change colour to match their substrate (Houtman and Dill 1994; Whiteley et al. 2009), suggest that sculpins rely on disruptive colouration and crypsis to reduce predation pressure. The coastrange sculpins from the two stream populations that I tested had low average depth and stayed on the bottom when C. asper was present, as would be expected from a fish that relies on cryptic behaviour to escape predation. In contrast to this response, Cultus pygmy sculpin went up into the water column when a predator (C. 59  asper) was present.  As C. asper is a larger sculpin, it would likely be able to outcompete the Cultus pygmy sculpin for any refuge on the bottom and is also a known predator on Cultus Lake sculpin (McNeely et al. 1990).  In fact, it is unusual for coastrange sculpin (C. aleuticus) to be found in large numbers in habitat occupied by prickly sculpin (C. asper); in streams, the two species tend to spatially segregate, with prickly sculpin favouring areas such as pools with slower flow, while coastrange sculpin are more commonly found in the faster flowing, riffle habitat (Taylor 1963).  In lakes, C. aleuticus can be found along the shoreline; however, smaller pelagic sculpin have been found in the deep benthic habitat of Lake Washington (Tabor et al. 2007).  Similar to the Lake Washington pelagic sculpin, Cultus pygmy sculpin are also found in the offshore benthic habitat, and may be captured in minnow traps along with prickly sculpin.  My data and these observations suggest, therefore, that in addition to lake productivity an important factor in the evolution of a limnetic life history of the Cultus pygmy sculpin could be predation/competition from sympatric prickly sculpin. 60   Future studies  Several additional lines of investigation could provide more precise information on behaviour and habitat use. For instance, the behavioural depth selection experiment tested for possible differences between stream- and lake-dwelling fish and the results were quite variable within treatment groups.  One way to eliminate the large amount of variance would be to take measures at thirty second (or possibly even smaller) intervals, as this method would provide more measurements of depth to reduce variance.  Increasing the number of individuals per population would also reduce the standard error (Whitlock and Schluter 2009).  Previous work on escape response in sculpin involved changing the colour of the substrate to something that sculpin were unable to match (Houtman and Dill 1994); the only substrate used in my experiment was grey, the colour of sand or gravel, which would promote cryptic behaviour by sculpin and would have constrained their willingness to leave the bottom (Houtman and Dill 1994). In addition, during my surveys Cultus pygmy sculpin were only captured in the offshore areas of Cultus Lake, which agreed with the findings of Ricker (1960).  Mature fish were also captured, and appeared to be mainly found closer to the shore, and at shallower depths (similar to the findings of Ruzycki et al. 1998); the dates and locations of these captures can assist in the search for spawning locations which, along with precise spawning timing, is still unknown.  My work focussed on determining the status of the Cultus pygmy sculpin as a designatable unit; critical habitat still needs to be identified to allow recovery planning.  It is also unknown whether larval Cultus pygmy sculpins have a pelagic stage in the water column (as many stream-dwelling Cottus 61  aleuticus do), or are more benthically-oriented as juveniles; this information is needed to determine the habitats used by all life stages, so all stages can be protected. In conclusion, the Cultus pygmy sculpin has only been found in one lake in BC, which has a high level of productivity.  The Cultus pygmy sculpin uses the water column both for feeding on Daphnia, a pelagic invertebrate, and probably as a refuge from the benthic and predacious C. asper, instead of relying only on its cryptic colouration.  The Cultus pygmy sculpin is found in an unusual ecological setting, and displays a distinctive behavioural response, different from that of the stream C. aleuticus, to a common predator (the prickly sculpin); the data from this chapter strongly suggests that the Cultus pygmy sculpin qualifies as a DU under SARA. 62  Table 13.  Locations and sample size of the populations of Cottus aleuticus used for the depth selection study. Location Watershed code Drainage N Cultus Lake 100-065700-09700- 13300 Chilliwack 25 Frosst Creek 100-065700-09700- 13300-9760 Chilliwack 32 Little Campbell River 900-000500 Semiahmoo Bay 27  63  Table 14.  Number of Cultus pygmy sculpins captured by month using stratified minnow traps in 2008, and the average length of the fish captured.  Month Number of sculpins caught Average total length (mm, range) May 22 50.73 (43-65) June 28 47.21 (30-59) July 53 33.74 (18-58) August 44 39.05 (15-63) September 13 56.08 (44-62) October 7 52.29 (47-58)  64    Table 15. The number of mature-sized (> 40 mm total length) sculpin caught in DFO mid-water sockeye trawls throughout the province of BC.  Of the 62 total lakes surveyed, only four caught sculpin that were the size of mature fish.  Due to the large number of sculpin caught in Cultus Lake in some of the trawls, lengths were not recorded for all fish, and so the number of mature fish caught in Cultus Lake is a conservative estimate. The sculpin caught in Pitt and Red Bluff lakes were not all identified to species, so it is possible that other species (i.e., C. asper) are included, and this number is likely an overestimate of mature C. aleuticus sampled from those lakes.  Lake name Total number of surveys Number of surveys where sculpin caught Number of mature-sized sculpin caught Cultus 89 64 48 Pitt 41 16 3 Red Bluff 1 1 4 65  Table 16.  The average depth and number of times off bottom, with standard deviation for each mean in brackets, shown by the Cultus pygmy sculpin (Cultus) and the C. aleuticus from the streams (Frosst = Frosst Creek; LCR = Little Campbell River). Experiment one used a “focal fish”, or one fish per tank; experiment two used a “focal fish” with a C. asper in the tank.   Population  N Average depth (focal fish; cm off bottom) Average number of times off bottom (focal fish) Average depth (focal fish with C. asper; cm off bottom) Average number of times off bottom (focal fish with C. asper) Cultus       25 16.0 (20.5) 7.8 (7.6) 23.1 (41.4) 6.3 (7.6) Frosst       32 28.9 (34.1) 9.8 (8.9) 5.5 (10.4) 2.6 (3.4) LCR          27 5.4 (7.4) 4.6 (5.8) 0.5 (1.3) 0.8 (2.1)   66      Figure 10.  Box plots of the distribution of the average depths by population of Cottus aleuticus, with no C. asper present.  Cultus = Cultus pygmy sculpin (n=25); Frosst = Frosst Creek (n=32); LCR = Little Campbell River (n=27).  The thick horizontal line shows the median depth, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). 67    Figure 11.  Box plots of the distribution of the number of times Cottus aleuticus were off the bottom, by population, with no C. asper present.  Cultus = Cultus pygmy sculpin (n=25); Frosst = Frosst Creek (n=32); LCR = Little Campbell River (n=27). The thick horizontal line shows the median count, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). 68   Figure 12.  Box plots of the distribution of the average depths of Cottus aleuticus by population, with C. asper present.  Cultus = Cultus pygmy sculpin (n=23); Frosst = Frosst Creek (n=31); LCR = Little Campbell River (n=12). The thick horizontal line shows the median depth, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). 69    Figure 13.  Box plots of the distribution of the average number of times Cottus aleuticus were off the bottom, by population, with C. asper present.  Cultus = Cultus pygmy sculpin (n=23); Frosst = Frosst Creek (n=31); LCR = Little Campbell River (n=12).  The thick horizontal line shows the median count, with the top and bottom of the box showing, respectively, the 75th and 25th percentiles (Crawley 2007).  The dashed lines represent approximately two standard deviations, with the circles representing outliers (Crawley 2007). 70  Chapter 4: Summary and general discussion    Due to the repeated glacial advances and retreats during the Pleistocene era, BC has very low taxonomic diversity in its freshwater fish fauna; however, there are many examples of highly divergent “ecotypes” and intraspecific phylogroups (e.g., McPhail and Taylor 1999; McPhail 2007).  Fish that recolonized habitat following deglaciation typically have lower levels of molecular neutral genetic diversity than populations in unglaciated portions of their geographic ranges (Bernatchez and Wilson 1998; Hewitt 2000).  The low sequence divergence estimates between Cultus pygmy sculpin and coastrange sculpin, combined with the paraphyletic mtDNA haplotypes, suggest a recent postglacial divergence. Differentiation at the microsatellite level is common in species that have diverged since the last glaciations (i.e., Icelandic Arctic char, Salvelinus alpinus, Gislason et al. 1999; Salvelinus spp., Salmenkova et al. 2005), while other species show some divergence at the mtDNA level (i.e., whitefish, Coregonus clupeaformis, Pigeon et al. 1997; sticklebacks, Gasterosteus aculeatus in both BC, Taylor and McPhail 1999; and Iceland, Olafsdottir et al. 2007). The FST values of the Cultus pygmy sculpin in comparison with other C. aleuticus, were low, but statistically significant, and demonstrated that the Cultus pygmy sculpin is genetically distinct at the population level.  This result could also be visualized with the differentiation observed in the factorial correspondence analysis (Figure 9). The microsatellite analyses showed that the pelagic sculpin found in Lake Washington were highly distinct, which suggests that the similar adaptations to pelagic lifestyles (i.e., larger cephalic pores, shorter pelvic fins, higher numbers of pectoral fin rays) found in both Lake Washington and Cultus Lake fish are a result of parallel 71  evolution. Similar morphological and ecological shifts have evolved multiple times under similar circumstances (Schluter and McPhail 1993); in fish evolving in postglacial lakes, the resources are partitioned such that one form is limnetic, feeding on plankton in open water, while the other form is benthic, feeding on macroinvertebrates in the benthos (Schluter and McPhail 1993; Robinson and Wilson 1994; Taylor 1999).  Similar to previous studies with other fish species (i.e., Schluter 1993; Olafsdottir et al. 2007; Bernatchez et al. 2010), C. aleuticus has evolved into a dwarf pelagic form at least twice, if not three times (Ricker 1960; Ikusemiju 1975; Larson and Brown 1975).  This diversification is common in novel or newly emergent habitats such as in the Holarctic, which suggests that species can expand their niche in the absence of competitors (Robinson and Wilson 1994).  Similar to the limnetic Arctic char (Salvelinus alpinus) morphs of Lake Thingvallavatn, the Cultus pygmy sculpin uses the benthos but feeds primarily on zooplankton in the open water (Malmquist et al. 1992; Robinson and Wilson 1994). The interspecific competition with the larger C. asper could contribute to ecological character displacement observed in the Cultus pygmy sculpin (Schluter and McPhail 1993).  Sculpin species tend to be spatially segregated: for example, C. aleuticus is found in riffle habitat in streams and along the shorelines of lakes, while C. asper is found in stream pools and the deep benthic habitat of lakes (Tabor et al. 2007); it is only in the absence of C. asper that coastrange sculpin are found in pools (McPhail 2007). The Cultus pygmy sculpin, however, has only been found in the offshore benthic habitat of Cultus Lake, similar to the pelagic sculpin found in Lake Washington (Ikusemiju 1975, Larson and Brown 1975).  Both lakes are highly productive, in comparison to most other coastal lakes in BC, with Cultus Lake also having large numbers of Daphnia (an 72  invertebrate food source) available year-round (Shortreed 2007). This constant food source, that also occurs below the thermocline within the colder water preferred by fish (Shortreed 2007), coupled with the large number of a potential competitor/predator on the bottom, could be an important factor in the adaptation and persistence of the Cultus pygmy sculpin.  The lower productivity and greater depths of most coastal lakes in the C. aleuticus range could mean that a similar niche for a pelagic sculpin is not available in other lakes, although there is always the possibility that other as yet unsampled lakes contain pelagic sculpins (e.g., Gow et al. 2008).  As the history of Lake Washington has shown, however, increasing amounts of anthropogenic nutrients can result in eutrophication and a plankton community dominated by inedible cyanobacteria (Edmondson 1994; Arhonditsis et al. 2003).  This increase in productivity causes a decrease in water quality (Arhonditsis et al. 2003) and a decline in fish diversity (Lopez- Rojas and Bonilla-Rivero 2000).  A similar process could threaten the persistence of the Cultus pygmy sculpin (see below).  As fishes recolonized habitats after glaciations, new and empty niches may have provided novel ecological opportunities and promoted divergence among populations in resource exploitation (Schluter 1993; Rogers et al. 2002). This process could explain how pelagic sculpin, such as the Cultus pygmy sculpin, developed specializations in water-column use for feeding and predator avoidance (Ikusemiju 1975; Rogers and Bernatchez 2007).  One difference in behaviour was observed when Cultus pygmy sculpin were placed in a tank with a benthic predator commonly found in Cultus Lake (C. asper): while stream-dwelling C. aleuticus decreased their movement and may have relied more heavily on cryptic colouration for protection, Cultus pygmy sculpin swam up 73  into the water column, exhibiting a different escape response, even though they are also cryptically coloured.   In summary, my thesis has combined information on molecular genetic variation, habitat use, and behaviour to compare the Cultus pygmy sculpin to its presumed ancestral form of typical stream-dwelling Cottus aleuticus. My data has contributed to our understanding of the unusual environment that the Cultus pygmy sculpin inhabits, the pelagic areas of Cultus Lake, in contrast to the flowing, riffle habitat in streams where C. aleuticus are most commonly found.  I also demonstrated that the Cultus pygmy sculpin is genetically distinct from the coastrange sculpins found in the lake’s tributary streams.  Although the colouration of the Cultus pygmy sculpin allows it to take advantage of cryptic behaviour to reduce predation pressure, the Cultus pygmy sculpin also appears to make greater use of the water column, a presumably adaptive response to the large numbers of predaceous C. asper found in the benthic areas of Cultus Lake. The genetic distinction of the Cultus pygmy sculpin and the differences in its habitat and behaviour, coupled with its morphological traits that are plausibly interpreted as adaptations to a pelagic lifestyle (COSEWIC 2010), strongly substantiate and support the Designatable Unit status for this population under Canada’s Species at Risk Act. The status of the Cultus pygmy sculpin was recently (2010) assessed by COSEWIC as “threatened”, a designation that was also applied in 2000.  The COSEWIC status report describes several potential threats to the Cultus pygmy sculpin (COSEWIC 2010).  It is unknown how the declining numbers of sockeye salmon in Cultus Lake (O. nerka) could impact the Cultus pygmy sculpin.  It is believed that sockeye salmon fry and Cultus pygmy sculpin could compete for zooplankton prey; 74  however, returning sockeye adults can contribute marine nutrients such as phosphorus and nitrogen to nursery lakes (Gende et al. 2002), which can increase zooplankton abundance (COSEWIC 2010).  Shortreed (2007) found an increase in lake productivity since the 1930’s, despite the declining number of sockeye salmon adults and the marine derived nutrients that they add to the system.  In years of low sockeye salmon abundance, predators shift from consuming sockeye salmon towards other prey species, including sculpin species (Ricker 1941). Consequently, the dramatic declines of sockeye salmon could result in an increase in predation on the Cultus pygmy sculpin. Alternatively, if the major predator on Cultus pygmy sculpin is C. asper, and if the other piscivorous fish target the larger and more abundant prickly sculpin then predation on the Cultus pygmy sculpin could decrease. The effect of northern pikeminnow (Ptychocheilus oregonensis) on the persistence of Cultus pygmy sculpin is unknown.  Ricker (1960) did not find any Cultus pygmy sculpin in the stomachs of northern pikeminnow, but since his study in the 1930s, there has been an increase in northern pikeminnow and a decrease in sockeye salmon.  The main predator of Cultus pygmy sculpin was bull trout (Salvelinus confluentus; Ricker 1960), and the current population size of this piscivore is unknown. Fishing pressure on its spawning grounds may have reduced the bull trout numbers in Cultus Lake (J.D. McPhail, pers. comm.), but this decrease has not been substantiated. At this time, the only documented exotic species in Cultus Lake is Eurasian milfoil (Myriophyllum spicatum) and its only suspected effect has been to provide increased cover, possibly allowing northern pikeminnow numbers to increase.  The lower Fraser Valley, however, has large numbers of invasive species (Nowosad 2010), including brown bullhead (Ameiurus nebulosus), which was responsible for the 75  extinction of the threespine stickleback species pair in Hadley Lake, BC (COSEWIC 2001).  It is possible that these invasive species could expand their range into the Cultus Lake watershed, or that species such as yellow perch (Perca flavescens) and bass (Micropterus spp.), favoured by anglers, could be illegally introduced into Cultus Lake (see Dunphy 2006 and Koopmans 2006 for examples of other instances of illegal introductions), where they could prey upon or compete with Cultus pygmy sculpin. The large number of visitors to Cultus Lake, and the increasing development of the area, have the potential to increase eutrophication, another potential threat to the Cultus pygmy sculpin.  The productivity of the lake has increased, suggesting that anthropogenic factors are contributing nutrients (average lake temperature has also increased).  This is in spite of a possible reduction in the annual influx of marine nutrients due to a decline in the number of sockeye salmon returning to Cultus Lake. Poor sewage control and additions from agricultural fertilizers can increase nutrients (Schubert et al. 2002), which can result in algal blooms and the eutrophication of the lake, resulting in declining water quality and a decrease in plankton and fish communities (Edmondson 1994).   Further research   At least two critical questions remain regarding the biology of Cultus Lake sculpin with relevance to its conservation. First, the abundance of Cultus pygmy sculpin and how it varies within and between years is unknown. The only available “guessestimate”, of 3,000 - 10,000 individuals, comes from Cannings (1993) and has no apparent quantitative basis.  I found the numbers from trawl UTM coordinates from DFO surveys 76  to be too low and too variable for use in estimating the Cultus pygmy sculpin population. The trawls are designed to target sockeye salmon.  The pygmy sculpin is only opportunistic bycatch and the potential habitat is underrepresented in the sampling. Better methods to estimate the population could include catch per unit effort on minnow trap sampling and mark-recapture studies.  The Cultus pygmy sculpin has shown up on the echosounding equipment of the DFO researchers, suggesting that it could be detected with hydroacoustic surveys (J. Hume, pers. comm.).  The Cultus pygmy sculpin numbers show seasonal variation (Chapter 3, COSEWIC 2010), possibly due to pulses of reproduction from different parts of the lake, as well as sculpin from seasonal streams taking refuge in the lake.  It would therefore be useful to conduct a multi-year study, in order to capture the variation over time, and to incorporate ageing and population assignment studies.  Second, because the Cultus pygmy sculpin is listed under SARA as a Schedule 1 species, its critical habitat must be defined and protected.  As defined in the Species at Risk Act (2001), “critical habitat” is “the habitat that is necessary for the survival or recovery of a listed wildlife species, and that is identified as the species’ critical habitat in the recovery strategy or in an action plan for the species”.  This habitat would include spawning and rearing areas of the lake, as well as foraging and refuge areas.  While I have captured mature fish in the lake, the actual spawning sites have not been determined.  More intensive sampling of the sites where I captured mature fish, at the same time of year, could help determine whether those sites are spawning areas (i.e., if large numbers of mature fish are caught there, both males and females, or if any of the fish captured appear to have recently spawned).  To help determine fertilized egg incubation areas, it would be necessary to continue to sample the spawning areas, in 77  an attempt to capture the earliest developmental stages and to determine whether Cultus pygmy sculpin fry remain on the benthos as juveniles or are pelagic as larvae. Confirming that the main prey of Cultus pygmy sculpin is Daphnia, and determining areas of the lake with the highest Daphnia concentrations and whether those areas are correlated with large numbers of Cultus pygmy sculpin would help determine feeding areas.  Finally, mapping the benthic composition of the lake bottom and comparing that to locations where Cultus pygmy sculpin are found could help define the type of habitat required (i.e., do Cultus pygmy sculpin prefer rocky areas, or sandy areas with vegetation for cover, spawning, etc.).  The easiest way to protect these areas would be to protect the whole lake, including the shoreline and benthic areas, possibly by adding them to the already existing Cultus Lake Provincial Park.  It is important to protect the inlet streams to ensure the water quality remains high and to monitor the shoreline developments to minimize any sedimentation and improper sewage treatment or storage.  Protecting the water quality of the lake and the integrity of the benthos would ensure that the habitat continues to support the persistence of the Cultus pygmy sculpin. Although sculpin are an integral part of the freshwater fish fauna of BC, few studies have examined possible local adaptations within a sculpin species.  My research used field, molecular and behavioural studies to contribute to our knowledge of the postglacial divergences of freshwater fishes in Canada. 78  References Allendorf, F.W. and G. Luikart. 2007. Conservation and the Genetics of Populations. Blackwell Publishing Ltd. Aquadro, C.F. and B.D. Greenberg. 1983. Human mitochondrial DNA variation and evolution: analysis of nucleotide sequences from seven individuals. Genetics 103: 287-312. Arhonditsis, G., M.T. Brett and J. Frodge. 2003. Environmental control and limnological impacts of a large recurrent spring bloom in Lake Washington, USA. Environmental Management 31: 603-618. Avise, J.C. 2004. Molecular Markers, Natural History and Evolution. Sinauer Associates, Ltd., USA. pp.684 Avise, J.C., J. Arnold, R.M. Ball, E. Bermingham, T. Lamb, J.E. Neigel, C.A. Reeb, and N.C. Saunders. 1987. Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematic. Am. Rev. Ecol. Sys. 18: 489-522 Barluenga, M., K.N. Stolting, Q.Salzburger, M. Muschick, and A. Meyer. 2006. Sympatric speciation in Nicaraguan crater lake cichlid fish. Nature 439: 719-723. Beere, M.C. 2002. A reconnaissance of Tuya Lake, August 7-13, 2002. Skeena Fisheries Report 135, BC Ministry of Water, Land and Air Protection. Behnke, R.J. 1972. The systematics of salmonid fishes of recently glaciated lakes. Journal of the Fisheries Research Board of Canada 29:639-671. Belkhir, K. P. Borsa, N. Chikhi, N. Raufaste, and F. Bonhomme. 2001.GENETIX 4.03, logiciel sous Windows TM pour la génétique des populations. in. Laboratoire Génome, Populations,Interactions, CNRS UMR 5000, Université de Montpellier II, Montpellier, France. 79  Bernatchez, L., S. Renaut, A.R. Whiteley, N. Derome, J. Jeukens, L. Landry, G. Lu, A.W. Nolte, K. Ostbye, S.M. Rogers and J. St-Cyr. 2010. On the origin of species: insights from the ecological genomics of lake whitefish. Phil. Trans. R. Soc. B. 365: 1783-1800. Bernatchez, L. and C.C. Wilson. 1998. Comparative phylogeography of Nearctic and Palearctic fishes. Molecular Ecology 7: 431-452. Bjornstad, G. and K.H. Roed. 2002. Evaluation of factors affecting individual assignment precision using microsatellite data from horse breeds and simulated breed crosses. Animal Genetics 33: 264-270. Bodtker, K.M., R.M. Peterman and M.J. Bradford. 2007. Accounting for uncertainty in estimates of escapement goals for Fraser River sockeye salmon based on productivity of nursery lakes in British Columbia, Canada. North American Journal of Fisheries Management 27: 286-302. Bond, C.E. 1963. Distribution and ecology of freshwater sculpin, genus Cottus, in Oregon. Doctoral dissertation. University of Michigan, Ann Arbor. Broadway, J.E. and P.B. Moyle. 1978. Aspects of the ecology of the prickly sculpin, Cottus asper, Richardson, a persistent native species in Clear Lake, Lake County, California. Environmental Biology of Fishes 3:327-343. Cannings, S.G. 1993. Rare freshwater fish of British Columbia. B.C. Ministry of Environment, Lands and Parks. Fisheries Branch. Conservation Data Centre. Victoria, B.C., Canada Cornuet, J.-M., S. Piry, G. Luikart, A. Estoup, and M. Solignac. 1999. New methods employing multilocus genotypes to select or exclude populations as origins of individuals. Genetics 153: 1989-2000. 80  COSEWIC. 2001. Status of the stickleback species pair, Gasterosteus spp., in Hadley Lake, Lasqueti Island, British Columbia. Canadian Field Naturalist, 115(4): 579-583. COSEWIC. 2003. Operations and procedures manual, November 2003 update. Committee on the Status of Endangered Wildlife in Canada, Ottawa, ON. 44 pp. + appendices. COSEWIC 2006. COSEWIC assessment and update status report on the deepwater sculpin Myoxocephalus thompsonii (Western and Great Lakes-Western St. Lawrence populations) in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa vii + 39pp. (www.sararegistry.gc.ca/status/status_e.cfm) COSEWIC. 2009. Guidelines for recognizing designatable units. Committee on the Status of Endangered Wildlife in Canada, Ottawa, ON. COSEWIC 2010. Draft status report on Cultus pygmy sculpin (Cottus sp.). Unpublished report for the Committee on the Status of Endangered Wildlife in Canada (COSEWIC). www.cosewic.gc.ca Cousyn, C., L. De Meester, J.K. Colbourne, L. Brendonck, D. Verschuren and F. Volckaert. 2001. Rapid, local adaptation of zooplankton behavior to changes in predation pressure in the absence of neutral genetic changes. PNAS 98: 6256-6260. Coyne, J.A. and H.A. Orr. 2004. Speciation. Sinauer Associates, Sunderland, MA. Crawley, M.J. 2007. The R Book. John Wiley & Sons Ltd., England. De Queiroz, A. and P.H. Wimberger. 1993. The usefulness of behaviour for phylogeny estimation: levels of homoplasy in behavioural and morphological characters. Evolution 47: 46-60 81  Domenici, P., H. Turesson, J. Brodersen and C. Bronmark. 2007. Predator-induced morphology enhances escape locomotion in crucian carp. Proc. R. Soc. B 275: 195- 201. Dunphy, M. 2006. It came from the pond. Georgia Straight Magazine, Vancouver, British Columbia. Web site: http://www.straight.com/content.cfm?id=18704 [Accessed July 2006]. Eberle, M.A., M.J. Rieder, L. Kruglyak and D.A. Nickerson. 2006. Allele frequency matching between SNPs reveals an excess of linkage disequilibrium in genic regions of the human genome. PLOS Genetics 2(9): 1319-1327. Edmondson, W.T. 1994. Sixty years of Lake Washington: a curricumlum vitae. Lake and Reservoir Management 10: 75-84. Eggers, D.M., N.W. Bartoo, N.A. Rickard, R.E. Nelson, R.C. Wissmar, R.L. Burgner, and H. Devol. 1978. The Lake Washington ecosystem: the perspective from the fish community production and forage base. Journal of the Fisheries Research Board of Canada 35: 1553-1571. Englbrecht, C.C., C.R. Largiader, B. Hanfling, and D. Tautz. 1999. Isolation and characterization of polymorphic microsatellite loci from Cottus gobio L. (Osteichthyes) and their applicability to related taxa. Molecular Ecology 8: 1957- 1969. Esteve, M. and D.A. McLennen. 2007. The phylogeny of Oncorhynchus (Euteleostei: Salmonidae) based on behavioural and life history characters. Copeia 2007: 520- 533. Fraser, D.L. and L. Bernatchez. 2001. Adaptive evolutionary conservation: towards a unified concept for defining conservation units. Molecular Ecology 10: 2741-2752. 82  Futuyma, D.J. 1986. Evolutionary Biology. Second edition. Sinauer Associates, Sunderland, Massachusetts, USA. Gadomski, D.M. and M.J. Parsley. 2005. Effects of turbidity, light level, and cover on predation of white sturgeon larvae by prickly sculpins. Transactions of the American Fisheries Society 134: 369-374. GBIF. 2006. Data Portal, Global Biodiversity Information Facility, University of Copenhagen, Denmark. Website: http://www.gbif.org/ [Accessed July 2006]. Gende, S. M., R. T. Edwards, M. F. Willson and M. S. Wipfli. 2002. Pacific salmon in aquatic and terrestrial ecosystems. BioScience 52: 917-928. Gervai, J. and V. Csanyi. 1985. Behavior-genetic analysis of the paradise fish, Macropodus opercularis. I. Characterization of the behavioural responses of inbred strains in novel environments: a factor analysis. Behavior Genetics 15: 503-519. Gislason, D., M.M. Ferguson, S. Skulason and S.S. Snorrason. 1999. Rapid and coupled phenotypic and genetic divergence in Icelandic Arctic char (Salvelinus alpines). Can. J. Fish. Aquat. Sci. 56: 2229-2234. Glaubitz J.C. (2004) CONVERT: A user-friendly program to reformat diploid genotypic data for commonly used population genetic software packages. Molecular Ecology Notes 4: 309-310. Goto, A. 1990. Alternative life-history styles of Japanese freshwater sculpins revisited. Environmental Biology of Fishes 28: 101-112. Goudet, J. 2001. FSTAT version 2.9.3.1 Updated from Goudet, J. 1995. J. Heredity 86:485-486. 83  Gow, J.L., C.A. Peichel, and E.B. Taylor. 2007. Ecological selection against hybrids in natural populations of sympatric threespine sticklebacks. J. Evol. Biol. 20(6): 2173- 2180. Gow, J.L., S.M. Rogers, M. Jackson and D. Schluter. 2008. Ecological predictions lead to the discovery of a benthic-limnetic sympatric species pair of threespine stickleback in Little Quarry Lake, British Columbia. Canadian Journal of Zoology 86: 564-571. Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41:95-98. Hanfling, B. B. Hellemans, F.A.M. Volckaert, and G.R. Carvalho. 2002. Late glacial history of the cold-adapted freshwater fish Cottus gobio, revealed by microsatellites. Molecular Ecology 11:1717-1729. Hansen, M.M., E. Kenchington, and E.E. Nielsen. 2001. Assigning individual fish to populations using microsatellite markers. Fish and Fisheries 2: 93-112. Hartl, D.L. and A.G. Clark. 2007. Principles of Population Genetics. Sinauer Associates, Inc. Sunderland, MA, USA. Heard, W.R. 1965. Limnetic cottid larvae and their utilization as food by juvenile sockeye salmon. Trans. Am. Fish. Soc. 94:191-193. Hewitt, G. 2000. The genetic legacy of the Quaternary ice ages. Nature 405:907-913. Houtman, R. and L.M. Dill. 1994. The influence of substrate colour on the alarm response of tidepool sculpins (Oligocottus maculosus; Pisces, Cottidae). Ethology 96: 147-154. 84  Hubisz, M.J., D. Falush, M. Stephens and J.K. Pritchard. 2009. Inferring weak population structure with the assistance of sample group information. Molecular Ecology Resources 9: 1322-1332. Huelsenbeck, J. P., and F. Ronquist. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17:754-755. Hume, J.M.B. and S.G. MacLellan. 2000. An assessment of the juvenile sockeye salmon (Oncorhynchus nerka) populations of Babine Lake. Can. Tech. Rep. Fish. Aquat. Sci. 2327:37p. Ikusemiju, K. 1975. Aspects of the ecology and life history of the sculpin, Cottus aleuticus (Gilbert), in Lake Washington. J. Fish. Biol 7:235-245. Ilves, K.L. 2007. Molecular systematics and biogeography of the holarctic smelt family Osmeridae (Pisces). Ph.D. thesis, University of British Columbia. Jarne, P. and P.J.L. Lagoda. 1996. Microsatellites, from molecules to populations and back. TREE 11: 424-429. Kinzinger, A.P., R.M. Wood, and D.A. Neely. 2005. Molecular systematic of the Genus Cottus (Scorpaeniformes: Cottidae). Copeia 2: 303-311. Kontula, T. S.V. Kirilchik, and R. Vainola. 2003. Endemic diversification of the monophyletic cottoid fish species flock in Lake Baikal explored with mtDNA sequencing. Molecular Phylogenetics and Evolution 27: 143-155. Koopmans, R. 2006. Alien fish species threaten the ecosystems they invade. Freshwater Fisheries Society of British Columbia, Victoria, British Columbia.  Web site: http://www.gofishbc.com/tips_articles/perch.htm [Accessed July 2006]. 85  Larson, K.W. and G.W. Brown Jr. 1975. Systematic status of a midwater population of freshwater sculpin (Cottus) from Lake Washington, Seattle, Washington. Journal of the Fisheries Research Board of Canada 32: 21-28. Legge, J.T., R. Roush, R. DeSalle, A.P. Vogler and B. May. 1996. Genetic criteria for establishing evolutionarily significant units in Cryan’s buckmoth. Conservation Biology 10:85-98. Li, W-H. 1997. Molecular Evolution. Sinauer Associates, Inc. USA. 487pp. Lindsey, C.C. and J.D. McPhail. 1986. Zoogeography of fishes of the Yukon and Mackenzie basins. In: Zoogeography of North American freshwater fishes. (Hocutt, C.H. and E.O. Wiley, eds.) Wiley, New York, USA. 639-674. Lonzarich, D.G. and T.P. Quinn. 1995. Experimental evidence for the effect of depth and structure on the distribution, growth, and survival of stream fishes. Canadian Journal of Zoology 73: 2223-2230. Lopez-Rojas, H. and A.L. Bonilla-Rivero. 2000. Anthropogenically induced fish diversity reduction in Lake Valencia Basin, Venezuela. Biodiversity and Conservation 9: 757- 765. Madenjian, C.P. and D.B. Bunnell. 2008. Depth distribution dynamics of the sculpin community in Lake Michigan. Transactions of the American Fisheries Society 137: 1346-1357. Magoulick, D.D. 2004. Effects of predation risk on habitat selection by water column fish, benthic fish and crayfish in stream pools. Hydrobiologia 527: 209-221. Malmquist, H.J., S.S. Snorrason, S. Skulason, B. Jonsson, O.T. Sandlund and P.M. Jonasson. 1992. Diet differentiation in polymorphic arctic charr in Thingvallavatn, Iceland. Journal of Animal Ecology 61: 21-35. 86  Mason, J.C. and S. Machidori. 1976. Populations of sympatric sculpins, Cottus aleuticus and Cottus asper, in four adjacent salmon-producing coastal streams on Vancouver Island, B.C. Fish Bull. 74: 131-141. McAllister, D.E. and C.C. Lindsey. 1961. Systematics of the freshwater sculpins (Cottus) of British Columbia. Bull. Nat. Mus. Canada Contrib. Zool. (1959) 172:66- 89. McLaughlin, R.L., J.W.A. Grant, and D.L. Kramer. 1994. Foraging movements in relation to morphology, water-column use, and diet for recently emerged brook trout (Salvelinus fontinalis) in still-water pools. Can. J. Fish. Aquat. Sci. 51: 268-279. McNeely, D.L., B.N. Futrell, and A. Sih. 1990. An experimental study on the effects of crayfish on the predator-prey interaction between bass and sculpin. Oecologia 85: 69-73. McPhail, J.D. 2007. The Freshwater Fishes of British Columbia. University of Alberta Press, Edmonton. McPhail, J.D. and R. Carveth. 1992. A foundation for conservation: the nature and origin of the freshwater fish fauna of British Columbia. Unpublished report to the Fisheries Branch, B.C. Ministry of the Environment, Lands and Parks, Victoria, B.C. McPhail, J.D. and C.C. Lindsey. 1970. Freshwater Fishes of Northwestern Canada and Alaska. Bulletin of the Fisheries Research Board of Canada 173. McPhail, J.D. and C.C. Lindsey. 1986. Zoogeography of freshwater fishes of Cascadia (the Columbia system and rivers north to the Stikine). In: Zoogeography of North American freshwater fishes. (Hocutt, C.H. and E.O. Wiley, eds.) Wiley, New York, USA. 615-637. 87  McPhail, J.D. and E.B. Taylor. 1999. Morphological and genetic variation in northwestern longnose suckers, Catostomus catostomus: the Salish sucker problem. Copeia 1999: 884-893. Mila, B., J.E. McCormack, G. Castaneda, R.K. Wayne, and T. B. Smith. 2007. Recent postglacial range expansion drives the rapid diversification of a songbird lineage in the genus Junco. Proc. R. Soc. B 274: 2653-2660. Miller, M.P. 1997. Tools for Population Genetic Analysis. Version 1.3: A Windows program for the analysis of allozyme and molecular population genetic data. Department of Biological Sciences, Northern Arizona University, Flagstaff. Moritz, C. 1994. Defining “evolutionarily significant units” for conservation. Trends in Ecology and Evolution 9: 373-375. Moritz, C., T.E. Dowling, and W.M. Brown. 1987. Evolution of animal mitochondrial DNA: relevance for population biology and systematic. Am.Rev. Ecol. Sys. 18: 269- 292. National Recovery Team for Cultus Pygmy Sculpin. 2007. Recovery Strategy for Cultus Pygmy Sculpin (Cottus sp.) in Canada. Species at Risk Act Recovery Strategy Series, Fisheries and Oceans Canada, Ottawa, v + 21 pp. Nolte, A.W., K.C. Stemshorn, and D. Tautz. 2005. Direct cloning of microsatellite loci from Cottus gobio through a simplified enrichment procedure. Molecular Ecology Notes 5: 628-636. Northcote, T.G. and P.A. Larkin.1956. Indices of Productivity in British Columbia lakes. J. Fish. Res. Bd. Canada 13:515-540. Northcote, T.G. and P.A. Larkin. 1963. Western Canada IN Frey, D.G. Limnology in North America. The Univeristy of Wisconsin Press, USA. 88  Nowosad, D. 2010. Conservation ecology and genetics of the brassy minnow (Hybognathus hankinsoni) in British Columbia. MSc thesis, Dept. of Zoology, University of British Columbia, Vancouver, BC. Okumura, N., Y. Hasebe, and A. Goto. 1996. Genetic variations of Cottus nozawae populations from five tributaries of the Shiribetsu River of southern Hokkaido, Japan. Ichthyol. Res. 43: 65-71. Olafsdottir, G.A., S.S. Snorrason and M.G. Ritchie. 2007. Postglacial intra-lacustrine divergence of Icelandic threesping stickleback morphs in three neovolcanic lakes. J. Evol. Biol. 20: 1870-1881. Paetkau, D. 1999. Using genetics to identify intraspecific conservation units: a critique of current methods. Conservation Biology 13 (6): 1507-1509. Paetkau, D., W. Calvert, I. Stirling, and C. Strobeck. 1995. Microsatellite analysis of population structure in Canadian polar bears. Molecular Ecology 4: 347-354. Pigeon, D., Chouinard, A. and L. Bernatchez. 1997. Multiple modes of speciation involved in the parallel evolution of sympatric morphotypes of lake whitefish (Coregonus cleupeaformis, Salmonidae). Evolution 51: 196-205. Piry S, Alapetite A, Cornuet, J.-M., Paetkau D, Baudouin, L., Estoup, A. (2004) GeneClass2: A Software for Genetic Assignment and First-Generation Migrant Detection. Journal of Heredity 95:536-539. Posada, D., and T. R. Buckley. 2004. Model selection and model averaging in phylogenetics: advantages of Akaike Information Criterion and Bayesian approaches over likelihood ratio tests. Syst. Biol. 53:793-808. Posada, D., and K. A. Crandall. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:817-818. 89  Pritchard, K., M. Stephens, and P. Donnelly. 2000.Inference of population structure using multilocus genotype data. Genetics 155: 945-959. R Development Core Team. 2008. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051- 07-0, URL http://www.R-project.org Raymond, M. and F. Rousset. 1995. Genepop (version 3.2):population genetics software for exact tests and ecumenicism. J. Hered. 86:248-249. Rice, W.R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. Ricker, W. E. 1941. The consumption of young sockeye salmon by predaceous fish. Journal of the Fisheries Research Board of Canada 5(3): 293-313. Ricker, W.E. 1960. A population of dwarf coastrange sculpins (Cottus aleuticus). Journal of the Fisheries Research Board of Canada 17:929-932. Rico, C., I. Rico, and G. Hewitt. 1996. 470 million years of conservation of microsatellite loci among fish species. Proc. R. Soc. Lond. B 263: 549-557. Ringstad, N. 1982. Carnation Creek watershed project freshwater sculpins: genus Cottus, a review. Pages 219-239 in G.F. Hartman, editor. Proceedings of the Carnation Creek workshop: a ten-year review, Malaspina College, Nanaimo, B.C. Robinson, B.W. and D.S. Wilson. 1994. Character release and displacement in fishes: a neglected literature. The American Naturalist 144: 596-627. Rogers, S.M. and L.Bernatchez. 2007. The genetic architecture of ecological speciation and the association with signatures of selection in natural lake whitefish (Coregonus sp. Salmonidae) species pairs. Mol. Biol. Evol. 24(6): 1423-1438. 90  Rogers, S.M., V. Gagnon, and L. Bernatchez. 2002. Genetically based phenotype- environment association for swimming behaviour in lakewhitefish ecotypes (Coregonus clupeaformis Mitchell). Evolution 56(11): 2322-2329. Rundle, H.D., L. Nagel, J.W. Boughman, and D. Schluter. 2000. Natural selection and parallel speciation in sympatric sticklebacks. Science 287: 306-308. Ruzycki, J.R., W.A. Wurtsbaugh, and C. Lay. 1998. Reproductive ecology and early life history of a lacustrine sculpin, Cottus extensus (Teleostei, Cottidae). Environmental Biology of Fishes 53:117-127. Salmenkova, E.A., V.T. Omel’chenko, O.A. Radchenko, N.V. Gordeeva, G.A. Rubtsova and N.S. Romanov. 2005. Genetic divergence of chars of the genus Salvelinus from Kronotsky Lake (Kamchatka Peninsula). Russian Journal of Genetics 41: 897-906. Schlosser, I.J. 1987. The role of predation in age- and size-related habitat use by stream fishes. Ecology 68: 651-659. Schluter, D. 1993. Adaptive radiation in sticklebacks: size, shape and habitat use efficiencies. Ecology 74(3):699-709. Schluter, D. 2001. Ecology and the origin of species. Trends in Ecology and Evolution 16: 372-380. Schluter, D. and J.D. McPhail. 1992. Ecological character displacement and speciation in sticklebacks. American Naturalist 140:85-108. Schluter, D. and J.D. McPhail. 1993. Character displacement and replicate adaptive radiation. TREE 8: 197-200. Schneider, S., D. Roessli, and L. Excoffier. 2000. Arlequin ver. 2.000: A software for population genetics data analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland. 91  Schubert, N. D., T. D. Beacham, A. J. Cass, T. E. Cone, B. P. Fanos, M. Foy, J. H. Gable, J. A. Grout, J. M. B. Hume, M. Johnson, K. F. Morton, K. S. Shortreed, M. J. Staley, and R. E. Withler. 2002. Status of Cultus Lake sockeye salmon. Canadian Science Advisory Secretariat, Research Document 2002/064. Scott, W.B. and E.J. Crossman. 1973. Freshwater Fishes of Canada. Galt House Publications, Ltd. Oakville, Ontario, Canada. Sheldon, T.A., N.E. Mandrak, and N.R. Lovejoy. 2008. Biogeography of the deepwater sculpin (Myoxocephalus thompsonii), a Nearctic glacial relict. Can. J. Zool. 86: 108- 115. Shortreed, K.S. 2007. Limnology of Cultus Lake, British Columbia. Can. Tech. Rep. Fish. Aquat. Sci. 2753:vi + 85p. Skulason, S., S.S. Snorrason, D.L.G. Noakes and M.M. Ferguson. 1996. Genetic basis of life history variations among sympatric morphs of Arctic char, Salvelinus alpines. Can. J. Fish. Aquat. Sci. 53: 1807-1813. Slaney, T.L. 1988. Biophysical evaluation of 15 lakes on the North coast of British Columbia. Prepared for the Department of Fisheries and Oceans, Vancouver, BC. Slatkin, M. 1995. A measure of population subdivision based on microsatellite allele frequencies. Genetics 139: 457-462. Stauffer, J.R., N.J. Bowers, K.R. McKaye, and T.D. Kocher. 1995. Evolutionarily significant units among cichlid fishes: the role of behavioural studies. American Fisheries Society Symposium 17: 227-244. Stefanni, S. and J.L. Thorley. 2003. Mitochondrial DNA phylogeography reveals the existence of an Evolutionarily Significant Unit of the sand goby Pomatoschistus 92  minutus in the Adriatic (Eastern Mediterranean). Molecular Phylogenetics and Evolution 28: 601-609. Swofford, D. L. 2002. PAUP*:phylogenetic analysis using parsimony. Ver. 4.0b10. Sinauer Associates, Sunderland, MA. Tabor, R.A., K.L. Fresh, D.K. Paige, E.J. Warner, and R.J. Peters. 2007. Distribution and habitat use of cottids in the Lake Washington basin. American Fisheries Society Symposium 53: 135-150. Talbot, S.L. and G.F. Shields. 1996. Phylogeography of brown bears (Ursus arctos) OF Alaska and paraphyly within the Ursidae. Molecular Phylogenetics and Evolution 5: 477-494. Tavaré, S. 1986. Some probabilistic and statistical problems in the analysis of DNA sequences. Lectures on Mathematics in the Life Sciences 17: 57-86. Taylor, E.B. 1991. A review of local adaptation in Salmonidae, with particular reference to Pacific and Atlantic salmon. Aquaculture 98: 185-207. Taylor, E.B. 1999. Species pairs of north temperate freshwater fishes: evolution, taxonomy, and conservation. Reviews in Fish Biology and Fisheries 9: 299-334. Taylor, E.B. 2006. How to identify putative designatable units (PDUs) for COSEWIC. Taylor, E.B., C.J. Foote, and C.C. Wood. 1996. Molecular genetic evidence for parallel life history evolution within a Pacific salmon (sockeye salmon and kokanee, Oncorhynchus nerka). Evolution 50: 401-416. Taylor, E.B., J.L. Gow, P. Tamkee, and P.Woodruff. 2010. Molecular species identification and population structure analysis of sculpin (Pisces: Cottidae) species in the Cheakamus River, British Columbia. Prepared for Triton Environmental Consultants, Ltd., Richmond, BC 93  Taylor, E.B. and J.D. McPhail. 1999. Evolutionary history of an adaptive radiation in species pairs of threespine sticklebacks Gasterosteus: insights from mitochondrial DNA. Biol. J. Linn. Soc. 66: 271-291. Taylor, G.D. 1963. Distribution and habitat responses of the coastrange sculpin (Cottus aleuticus) and prickly sculpin (Cottus asper) in the Little Campbell River, British Columbia. M.Sc thesis, University of British Columbia. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The ClustalX-Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882. Trewavas, E. 1983. Tilapiine fishes of the genera Sarotherodon, Oreochromis, and Danakilia. British Museum (Natural History) Publication No. 878. London. Van Oosterhout, C., W.F. Hutchinson, D.P.M. Wills, and P. Shipley. 2004. Micro- checker: software for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4:535-538. Waddell, P.J. and M.A. Steel. 1997. General time reversible distances with unequal rates across sites: mixing Г and inverse Gaussian distributions with invariant sites. Molecular Phylogenetics and Evolution 8 (3): 398-414. WFC. 2006. Fishbase: a Global Information Source about Fishes, World Fisheries Centre, Penang, Malaysia. Website: http://www.fishbase.org [Accessed July 2006]. White, J.L. and B.C. Harvey. 1999. Habitat separation of prickly sculpin, Cottus asper, and coastrange sculpin, Cottus aleuticus, in the mainstem Smith River, northwestern California. Copeia 1999(2): 371-375. 94  Whiteley, A.R., S.M. Gende, A.J. Gharrett, and D.A. Tallmon. 2009. Background matching and color-change plasticity in colonizing freshwater sculpin populations following rapid deglaciation. Evolution 63(6):1519-1529. Whitlock, M.C. and D. Schluter. 2009. The Analysis of Biological Data. Roberts and Company Publishers, Colorado, USA. Wirtz, P. 1999. Mother species-father species: unidirectional hybridization in animals with female choice. Animal Behaviour 58: 1-12. Yagami, T., R. Yokoyama, and A. Goto. 2002. Genetic fragmentation of populations of the fluvial sculpin Cottus nozawae (Pisces: Cottidae) at the southern margin of its native range. Can. J. Zool. 80: 873-881. Yokoyama, R. and A. Goto. 2002. Phylogeography of a freshwater sculpin, Cottus nozawae, from the northeastern part of Honshu Island, Japan. Ichthyol. Res. 49:147- 155 Yokoyama, R. and A. Goto. 2005. Evolutionary history of freshwater sculpins, genus Cottus (Teleostei; Cottidae) and related taxa, as inferred from mitochondrial DNA phylogeny. Molecular Phylogenetics and Evolution 36:654-668. Zink, R.M. and D.L. Dittmann. 1993. Gene flow, refugia, and evolution of geographic variation in the song sparrow (Melospiza melodia). Evolution 47(3): 717-729. 95  Appendix 1. UTM coordinates of minnow trap sites in Cultus Lake. Date (YYYYMMDD) Trap number Bottom depth (m) Zone Easting Northing 20070507 1 n/a 10 572500 5432754 20070507 2 n/a 10 572499 5432732 20070507 3 n/a 10 572506 5432721 20070507 4 n/a 10 572492 5432710 20070507 5 n/a 10 572509 5432712 20070507 6 n/a 10 572505 5432729 20070507 7 n/a 10 572523 5432721 20070507 8 n/a 10 573730 5432939 20070705 1 n/a 10 574297 5432726 20070705 2 n/a 10 572742 5432015 20070705 3 n/a 10 573918 5432395 20070705 4 n/a 10 575009 5433803 20070705 5 n/a 10 574236 5436031 20070705 6 n/a 10 575667 5434402 20070705 7 n/a 10 572246 5432421 20070705 8 n/a 10 574403 5432832 20070705 9 n/a 10 574923 5433580 20070705 10 n/a 10 575004 5433794 20070705 11 n/a 10 575670 5434478 20071105 1 n/a 10 572642 5432603 20071105 2 n/a 10 572517 5432493 20071105 3 n/a 10 572589 5432181 20071105 4 n/a 10 572938 5432052 20071105 5 n/a 10 573915 5435434 20071105 6 n/a 10 57361 5434362 20071105 7 n/a 10 574664 5433437 20071105 8 n/a 10 574625 5434723 20080513 1 n/a 10 573004 5432643 20080513 2 n/a 10 573306 5432893 20080513 3 n/a 10 573996 5433632 20080513 4 n/a 10 574106 5433259 20080513 5 n/a 10 574299 5434138 20080513 6 n/a 10 574435 5434426 20080513 7 n/a 10 574520 5434957 20080514 1 n/a 10 572672 5432531 20080514 2 n/a 10 572784 5432606 20080514 3 n/a 10 574133 5433243 20080514 4 n/a 10 572707 5432448 20080514 6 n/a 10 574446 5434418 20080514 5 n/a 10 572609 5432443 20080514 7 n/a 10 572611 5432363  96  Appendix 1 cont. UTM coordinates of minnow trap sites in Cultus Lake.  Date (YYYYMMDD) Trap number Bottom depth Zone Easting Northing 20080614 1  10 572479 5432456 20080614 2 32 10 572678 5432164 20080614 3 33 10 573604 5434472 20080614 4 38 10 573604 5434472 20080614 5 40 10 573932 5435344 20080614 6 38 10 575049 5434840 20080614 7 38 10 574662 5434143 20080726 1 40 10 572543 5432522 20080726 2 43 10 573009 5432550 20080726 3 42 10 573393 5433506 20080726 4 25 10 574135 5433528 20080726 5 41 10 574123 5434907 20080726 6 36 10 574823 5434362 20080726 7 40 10 574323 5435086 20080821 1 37 10 572499 5432430 20080821 2 43 10 573008 5432852 20080821 3 40 10 573656 5433574 20080821 4 28 10 575530 5434628 20080821 5 23 10 574563 5433642 20080821 6 41.5 10 573555 5432561 20080821 7 30.3 10 572542 5432158 20080927 1 38 10 572663 5432390 20080927 2 40 10 572683 5432514 20080927 3 35 10 572547 5432320 20080927 4 32.4 10 573774 5433564 20080927 5 26 10 574507 5433698 20080927 6 32 10 574518 5435357 20080927 7 39 10 572562 5432551 20080927 8 30 10 572660 5432149 20081024 1 35 10 572463 5432389 20081024 2 33 10 572422 5432440 20081024 3 35 10 572507 5432482 20081024 4 37 10 572555 5432392 20081024 5 36 10 572628 5432263 20081024 6 34 10 574149 5433217 20081024 7 38.4 10 574191 5433720 20081024 8 34 10 574891 5434293  97   Appendix 2. List of lengths of Cultus pygmy sculpin captured in minnow traps in Cultus Lake, May-October 2008. Date of capture Trap number Depth of capture (m) Depth from bottom (m) Total length (mm) May 15 4 40 n/a 54 May 15 1 30 n/a 45 May 15 7 10 n/a 48 May 15 2 30 n/a 53 May 15 1 40 n/a 45 May 15 1 40 n/a 45 May 15 3 40 n/a 43 May 15 6 40 n/a 47 May 15 2 40 n/a 51 May 15 2 40 n/a 45 May 15 2 40 n/a 48 May 15 2 40 n/a 46 May 15 2 40 n/a 49 May 15 5 40 n/a 50 May 15 5 40 n/a 65 May 15 5 40 n/a 45 May 15 5 40 n/a 60 May 15 5 40 n/a 54 May 15 7 40 n/a 53 May 15 7 40 n/a 59 May 15 7 40 n/a 58 May 15 7 40 n/a 53 June 15 1 40 n/a 35 June 15 6 20 18 50 June 15 7 20 18 55 June 15 3 30 3 30 June 15 3 30 3 50 June 15 3 30 3 52 June 15 3 30 3 48 June 15 7 40 0 54 June 15 7 40 0 47 June 15 6 40 0 53 June 15 6 40 0 46 June 15 6 40 0 47 June 15 6 40 0 45 June 15 6 30 8 47 June 15 6 30 8 58 June 15 6 30 8 45 June 15 6 30 8 30 June 15 6 30 8 49 June 15 1 40 n/a 56 June 15 1 40 n/a 46   98  Appendix 2 cont. List of lengths of Cultus pygmy sculpin captured in minnow traps in Cultus Lake, May-October 2008.  Date of capture Trap number Depth of capture (m) Depth from bottom (m) Total length (mm) June 15 1 40 n/a 50 June 15 1 40 n/a 54 June 15 1 40 n/a 35 June 15 3 40 0 48 June 15 3 40 0 45 June 15 3 40 0 54 June 15 2 30 2 34 June 15 5 40 0 59 July 26 7 30 0 40 July 26 2 20 13 25 July 26 5 40 1 35 July 26 4 40 0 23 July 26 3 30 12 28 July 26 4 30 0 50 July 26 4 30 0 37 July 26 4 30 0 30 July 26 4 30 0 34 July 26 4 30 0 35 July 26 4 30 0 21 July 26 4 30 0 21 July 26 4 30 0 20 July 26 4 30 0 20 July 26 4 30 0 20 July 26 4 30 0 18 July 26 4 30 0 20 July 26 1 40 0 50 July 26 1 40 0 43 July 26 1 40 0 25 July 26 1 40 0 21 July 26 7 30 10 47 July 26 7 30 10 45 July 26 7 30 10 42 July 26 7 30 10 53 July 26 7 30 10 58 July 26 7 30 10 35 July 26 7 30 10 40 July 26 7 30 10 35 July 26 7 30 10 31 July 26 7 30 10 34 July 26 1 30 10 43 July 26 1 30 10 36 July 26 1 30 10 45 July 26 7 30 10 42    99  Appendix 2 cont. List of lengths of Cultus pygmy sculpin captured in minnow traps in Cultus Lake, May-October 2008.  Date of capture Trap number Depth of capture (m) Depth from bottom (m) Total length (mm) July 26 7 30 10 41 July 26 7 30 10 37 July 26 7 30 10 35 July 26 7 30 10 35 July 26 7 30 10 34 July 26 7 30 10 35 July 26 7 30 10 36 July 26 7 30 10 29 July 26 7 30 10 34 July 26 7 30 10 30 July 26 7 30 10 20 July 26 7 30 10 20 July 26 1 30 10 26 July 26 1 30 10 38 July 26 1 30 10 32 July 26 1 30 10 35 July 26 1 30 10 33 July 26 1 30 10 36 August 22 4 40 0 24 August 22 3 10 30 58 August 22 4 30 0 39 August 22 3 40 0 54 August 22 3 40 0 57 August 22 3 40 0 25 August 22 5 30 0 29 August 22 5 30 0 18 August 22 5 30 0 30 August 22 5 30 0 23 August 22 5 30 0 21 August 22 5 30 0 16 August 22 5 30 0 20 August 22 5 30 0 25 August 22 5 30 0 15 August 22 3 30 10 38 August 22 3 30 10 50 August 22 3 30 10 63 August 22 3 30 10 35 August 22 3 30 10 45 August 22 3 30 10 57 August 22 3 30 10 35 100  Appendix 2 cont. List of lengths of Cultus pygmy sculpin captured in minnow traps in Cultus Lake, May-October 2008.  Date of capture Trap number Depth of capture (m) Depth from bottom (m) Total length (mm) August 22 3 30 10 58 August 22 3 30 10 50 August 22 3 30 10 35 August 22 3 30 10 62 August 22 3 30 10 47 August 22 3 30 10 52 August 22 3 30 10 36 August 22 3 30 10 30 August 22 3 30 10 31 August 22 1 40 0 25 August 22 1 40 0 30 August 22 1 40 0 60 August 22 1 40 0 21 August 22 1 40 0 35 August 22 1 40 0 45 August 22 1 40 0 56 August 22 1 40 0 47 August 22 1 40 0 25 August 22 1 40 0 35 August 22 1 40 0 51 August 22 1 40 0 54 August 22 1 40 0 56 September 28 6 30 2 58 September 28 6 40 0 62 September 28 6 40 0 59 September 28 4 40 0 58 September 28 7 20 19 55 September 28 2 40 0 56 September 28 2 40 0 44 September 28 5 30 0 53 September 28 5 30 0 59 September 28 5 30 0 57 September 28 5 40 0 55 September 28 5 40 0 55 September 28 5 40 0 58 October 25 5 40 0 52 October 25 6 30 4 53 October 25 6 30 4 47 October 25 6 40 0 52 October 25 6 40 0 49 October 25 8 30 4 58 October 25 8 40 0 55  101  Appendix 3. Animal care certificate used in conducting field sampling and laboratory experiments.   102  Appendix 3 cont.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0071112/manifest

Comment

Related Items