1 COSEWIC Special Report

2

3 Designatable units at an appropriate scale for

4 the Lake Whitefish (Coregonus clupeaformis)

5 in Canada

6 prepared for

7 COMMITTEE ON THE STATUS OF ENDANGERED WILDLIFE IN 8 CANADA

9 By

10

11

12 Sean M. Rogers 13 Department of Zoology 14 University of British Columbia 15

16 Funding provided by Environment Canada

17

18

19

20 Final Report Submitted November 3, 2008

i 21 Table of Contents

22 Executive Summary ...... 1 23 I. General Introduction...... 1 24 Description and biology of Lake Whitefish...... 3 25 Distribution of Lake Whitefish ...... 4 26 Taxonomy of Lake Whitefish ...... 4 27 II. Key for establishing designatable units for COSEWIC ...... 6 28 1. The putative designatable unit (PDU) is a distinct taxonomic entity or qualifies as a 29 distinct biological species? ...... 6 30 1.1. Taxonomic entities in the Lake Whitefish species complex...... 6 31 1.1.1 Coregonus lavaretus and Coregonus pidschian ...... 7 32 1.1.2 Coregonus nelsonii...... 8 33 1.1.3 Coregonus huntsmani...... 8 34 1.2. Are these taxonomic groups distinct biological species? ...... 9 35 1.2.1 Limnetic and benthic Lake Whitefish...... 9 36 1.2.2 Distribution and characteristics of limnetic and benthic pairs in Canada...... 10 37 1.2.2.1 Squanga Lake, Yukon Territory...... 11 38 1.2.2.2 Little Teslin Lake, Yukon ...... 12 39 1.2.2.3 Dezadeash Lake, Yukon ...... 13 40 1.2.2.4 Teenah Lake, Yukon...... 13 41 1.2.2.5 Hanson Lake, Yukon...... 13 42 1.2.2.6 Dragon Lake, British Columbia...... 14 43 1.2.2.7 Great Slave Lake, NWT...... 14 44 1.2.2.8 Lake Athabasca, NWT...... 15 45 1.2.2.9 Lake Opeongo, Ontario...... 15 46 1.2.2.10 Como Lake, Ontario...... 16 47 1.2.2.11 Lake Superior, Ontario...... 16 48 1.2.2.12 Lake Simcoe, Ontario ...... 17 49 1.2.2.13 Témiscouata Lake, Quebec...... 17 50 1.2.2.14 East Lake, Quebec...... 18 51 1.2.2.15 Caniapiscau Lake, Quebec...... 18 52 1.2.2.16 Lac Outardes II, Quebec ...... 19 53 1.2.2.17 Lac Manicouagan V, Quebec...... 19 54 1.2.3 Hybridization of C. clupeaformis with other Coregonineae...... 20 55 1.2.4 Species status of Coregonus clupeaformis...... 20 56 1.3. Summary...... 21 57 2. The PDU represents a major phylogenetic grouping separate from other groupings 58 within the taxon in question? ...... 22 59 2.1. Glacial History of Canada and Postglacial dispersal of Lake Whitefish...... 22 60 2.2. Phylogeography of the Lake Whitefish species complex...... 23 61 2.2.1 Beringian Glacial Race: ...... 24 62 2.2.2 Nahanni Glacial Race: ...... 25 63 2.2.3 Mississippian Glacial Race:...... 26 64 2.2.4 Atlantic Glacial Lineage: ...... 27 65 2.2.5 Acadian Glacial Race:...... 28 66 2.3. Population Subdivision within Glacial Races:...... 28

ii 67 2.5. Summary...... 29 68 3. The PDU has distinctive traits that (1) represent local adaptation and (2) identifies the 69 PDU as not ecologically interchangeable with other known PDUs within the species, or 70 (3) identifies the PDU as an irreplaceable component of Canada’s biodiversity? ...... 30 71 3.1. Evidence of distinctive traits that represent local adaptation in the species pairs .....30 72 3.1.1 Distinctive Traits:...... 30 73 3.1.1.1 Morphology...... 30 74 3.1.1.2 Behaviour...... 32 75 3.1.1.3 Physiology and performance...... 32 76 3.1.1.4 Gene expression differences ...... 33 77 3.1.2 Evidence for local adaptation:...... 33 78 3.2. Identification of the limnetic and benthic species pairs as not ecologically 79 interchangeable with other known PDUs within the species...... 35 80 3.2.1 Independent origins of the species pairs ...... 35 81 3.2.2 Evidence of reproductive isolation ...... 37 82 3.3. Are limnetic and benthic species pairs an irreplaceable components of Canadian 83 biodiversity?...... 37 84 3.3. Consideration of the Mira River population ...... 38 85 3.5. Summary...... 39 86 4. The PDU represents a major range disjunction?...... 39 87 4.1. Summary...... 39 88 5. The PDU inhabits a different biogeographic zones? ...... 40 89 5. 1 Summary...... 40 90 6. Concluding Summary ...... 41 91 Acknowledgements...... 42 92 Literature Cited ...... 43 93 Table 1...... 66 94 List of Figures...... 69 95 Figure 1...... 70 96 Figure 2 ...... 71 97 Figure 3 ...... 72 98 Figure 4 ...... 73 99 Figure 5 ...... 74 100

101

102

iii 103 Executive Summary 104

105 The Lake Whitefish (Coregonus clupeaformis, Mitchill) is a broadly distributed 106 freshwater fish in Canada. It is referred to as the ‘Coregonus clupeaformis species 107 complex’ in recognition of taxonomic problems existing across the country. The objective 108 of this special report was to carry out an evaluation of designatable units (DUs) for 109 Coregonus clupeaformis in Canada. This report examines the species complex with respect 110 to genetics, ecology, morphology, distribution, range disjunction, as well as presence in 111 different biogeographic zones for over eighty populations. These sources of information 112 were used to evaluate and identify putative DUs (PDUs) under COSEWIC’s guidelines for 113 recognizing DUs below the species level.

114 Thirty two DUs were identified in the Lake Whitefish species complex. The 115 Eurasian Lake Whitefish (Coregonus lavaretus) was identified as a distinct taxonomic 116 entity and designated as a DU. Seven limnetic and benthic sympatric species pairs were 117 recognized as distinct biological species and designated as 14 DUs. Three additional 118 limnetic and benthic species pairs were given provisional status (6 DUs). Lake Whitefish 119 populations were recognized as encompassing five genetically distinct groups reflecting 120 geographic isolation during the Pleistocene glaciations. National Freshwater Biogeographic 121 Zones (NFBZs) that capture these significant major phylogeographic groupings of the 122 species complex were identified as 11 DUs. These PDUs will require closer inspection and 123 this report should be considered as a first approximation before COSEWIC assessments can 124 be undertaken.

125

126 I. General Introduction 127

128 Coregonines are cold-water fishes common throughout the Holarctic in North 129 America, Europe and Asia (Lindsey and Woods 1970). They support important 130 commercial and recreational fisheries and are the focus of significant worldwide 131 aquaculture operations (Eckmann et al. 1996). Coregonines are a dynamic example of 132 evolutionary change, with numerous species, sub-species, and forms evolving during and 133 after the Pleistocene glaciations throughout their entire distribution (e.g., Kirkpatrick and 134 Selander 1979, Morin et al. 1982, Bodaly et al. 1991, Vuorinen et al. 1993, Bodaly et al. 135 1994, Svaerdson 1998, Turgeon et al. 1999, Bernatchez 2004, McDermid et al. 2005, 136 Ostbye et al. 2006). Their broad distribution and successful colonization of lake and river 137 environments following the retreat of glacial ice has contributed to the significant interest 138 of Coregonines as a model system for understanding evolution. However, the broad 139 distribution and dynamic evolutionary histories of many coregonine species have also

1 140 resulted in inconsistent taxonomies and disagreements in the nomenclature, commonly 141 referred to as the 'Coregonine problem' (Svärdson 1957, 1965 Lindsey et al. 1970, Scott 142 and Crossman 1973, Nikolsky and Reshetnikov 1970 , Douglas et al. 2005, McPhail 2007).

143 The Lake Whitefish (Coregonus clupeaformis) is a broadly distributed freshwater 144 Coregonine in Canada (Scott and Crossman 1973). The species is the most valuable 145 commercial freshwater fish in the country (Bodaly 1986, DFO 2005) and has been studied 146 extensively as an important aquaculture resource (Flüchter 1980, Bodaly 1986, Drouin et al. 147 1986, Zitzow and Millard 1988, Harris and Hulsman 1991, Champigneulle and Rojas- 148 Beltran 1990). Variation among populations across its distribution has resulted in a 149 confusing taxonomic picture referred to as the ‘Coregonus clupeaformis complex’ (McPhail 150 and Lindsey 1970, Scott and Crossman 1973). An explicit evaluation of designatable units 151 (DUs) is therefore required for Coregonus clupeaformis within Canada before COSEWIC 152 assessments can be undertaken.

153 The objectives of this special report were to: (1) assess Canadian Lake Whitefish 154 populations to determine if they meet the guidelines for DUs, 2) for each PDU, assess the 155 criteria that substantiated the delineation, 3) when an assessment cannot be made, determine 156 the information lacking that compromised the identification, and (4) synthesize overall 157 results in a context that facilitates the evaluation of various conservation scenarios towards 158 ultimately adopting appropriate DUs for use in subsequent status assessments for Lake 159 Whitefish as a whole.

160 The following methodology was used to identify DUs. First, an extensive Lake 161 Whitefish literature was reviewed for relevant information about the species with respect to 162 genetics, ecology, morphology, distribution, range disjunction, and presence in different 163 biogeographic zones. A comprehensive list of the current Canadian Lake Whitefish 164 populations requiring DU consideration was generated from this literature (Table 1). When 165 lakes were sampled multiple times, data from all studies were integrated towards assessing 166 DU status (e.g., multiple DNA markers in phylogenetic studies). These populations were 167 then applied to a key for assessing PDUs (Taylor 2006), with criteria based on those in the 168 COSEWIC Guidelines for Recognizing DUs below the species level 169 (http://www.cosewic.gc.ca/eng/). The key was applied as a series of questions to attempt 170 identification of DUs that have not yet been proposed and to establish whether the 171 identification of Lake Whitefish DUs is necessary within a complex of populations, i.e., 172 evaluated for DU status at one time. The final result is a DU decision chart and summary 173 illustrating the proposed DUs for the Lake Whitefish species complex in Canada (Table 1, 174 Figure 2).

175 Specific cases of DUs will require closer inspection and therefore the key is 176 considered as a first approximation. For instance, Taylor (2006) suggested that if a 177 diagnostic character was identified in step 3 of the key, but was resolved after only limited

2 178 sampling, DU status might be deferred or made provisional until the diagnostic nature of 179 the trait was confirmed after more extensive sampling. Alternatively, if the presence of a 180 DU was suspected based on inductive/deductive reasoning, but for which no direct 181 evidence of distinction was available, that group may be give “provisional” or “deferred” 182 DU status. In these cases, provisional is defined as a tentative DU status, reserved for 183 populations that meet only some of the data requirements specified in the key. Deferred 184 refers to a postponement in DU status, reserved for populations where the presence of a DU 185 was suspected based on inductive/deductive reasoning, or no published data was available. 186 At the end of each question in the key, evidence for DU status was summarized for use in a 187 “Lake Whitefish Decision Chart” for reference (Figure 2).

188

189 Description and biology of Lake Whitefish 190

191 Lake Whitefish (Class: Actinopterygii, Order: Salmoniformes, Family: Salmonidae, 192 Sub-family: Coregoninae, Genus: Coregonus, Species: clupeaformis, as the Greek name 193 clupeaformis implies, are “herring shaped”, referring to their short head but elongate, 194 laterally compressed body (Scott and Crossman 1973) (Figure 1a). Common names for the 195 species include “le Grand Corégone” and “humpback whitefish” (McPhail and Lindsey 196 1970, Scott and Crossman 1973, Alt 1979). Lake Whitefish are silvery (but in some lakes 197 can have a relatively dark green/brown dorsal surface) and attain an average length of 380 198 mm (Scott and Crossman 1973). They possess a single, soft dorsal fin plus an adipose fin, 199 and have a scaly process at the base of each pelvic fin (McPhail and Lindsey 1970) (Figure 200 1a). They have 70 to 97 large, cycloid scales along the lateral line. An inferior mouth 201 reflects an adaptation to bottom feeding and distinguishes Lake Whitefish from cisco 202 species within the same genus (McPhail and Lindsey 1970, Scott and Crossman 1973, 203 Figure 1b). Distinguishing characters from other closely related whitefish (e.g., broad 204 whitefish, Coregonus nasus, Pallus, or Mountain Whitefish, Prosopium williamsoni, 205 Pennant) include a concave, smaller profile of the head and a smaller ratio of maxillary 206 length to interorbital width (Lindsey 1962, McPhail and Lindsey 1970, Scott and Crossman 207 1973, Figure 1c). The Lake Whitefish can typically be distinguished from the broad 208 whitefish by its smaller size, longer gill rakers and a longer snout (Figure 1d).

209 Lake Whitefish prefer cold water (8 – 14 degrees Celsius) and spawn in the fall, 210 with more northern populations generally spawning earlier (Scott and Crossman 1973, 211 McPhail 2007). Females are deeper bodied than males, while males have significantly 212 longer jaws and pectoral/pelvic fins (Lindsey 1963b, Casselman and Schulte-Hostedde 213 2004). Breeding males develop epidermal structures (tubercles) that protrude as sharp 214 bumps from the scales and are located repeatedly along the dorsal and lateral sides shortly 215 before their spawning season (Wedekind et al. 2001). Several species of benthic organisms

3 216 make up the diet of Lake Whitefish, including copepods, cladocerans, chironomid larvae, 217 amphipods, gastropods, and sometimes small fish (Brown and Taylor 1992, Davis and 218 Todd 1998, Chouinard et al. 1996, Cucin and Faber 1985, Landry et al. 2007).

219 Distribution of Lake Whitefish 220

221 Coregonines (including whitefishes, vendace, ciscoes) are distributed over cooler 222 regions of the northern hemisphere, reflecting both the species’ preference for coldwater 223 habitats as well as their dispersal ability. In Canada, the broad distribution of Coregonus 224 clupeaformis extends from Yukon Territory to Labrador. Lake Whitefish inhabit the lakes 225 and rivers of every province except PEI and Newfoundland, extending southward into the 226 New England and Great Lake States (USA) (Figure 3). The northern limit of its range is 227 Victoria Island in the NWT, near Cambridge Bay and the Arctic Archipelago. In these 228 northern regions (e.g., Ungava, the Hudson Bay region, and Arctic Ocean drainages in the 229 North West Territories, the species also enters brackish water (Scott and Crossman 1973). 230 Collections from Canadian lakes have been extensive (Table 1, Figure 3). For example, 231 Bodaly et al. (1992b) report an impressive 119 lake and river populations sampled between 232 1968 and 1986. Scott and Crossman (1973) report over 166 locations known to inhabit the 233 species in Canada and the United States. Genetic data have been collected from over 100 234 Canadian lakes and rivers, rendering this species uniquely informative as a model system 235 reflecting the Canadian zoogeographic history of postglacial freshwater fish dispersal 236 (Figure 3). This list is also by no means exhaustive, with many more bodies of water 237 inhabited by native populations (see McPhail 2007).

238 Additionally, there have been several instances of Lake Whitefish introductions into 239 several lakes. For example, McPhail (2007) identified 20 introduced populations in 240 southern British Columbia, of which nine failed. Most of the failures were in small lakes, 241 while introductions into larger, deeper lakes appear to have been largely successful (e.g., 242 Shuswap, Okanagan, Kootenay, and Arrow lakes) (McPhail 2007). This pattern of 243 introductions has also been recorded in Ontario (Lasenby et al. 2001) and Nova Scotia 244 (DFO 2006). Scott and Crossman (1964) also report an introduced population in Hogan’s 245 Pond, Newfoundland, originally from Lake Erie.

246 Taxonomy of Lake Whitefish 247

248 During the Pleistocene glaciation (1 800 000 to 12 000 years before present) 249 Coregonines were isolated in meltwater lakes and rivers around the margins of ice in the 250 northern hemisphere (Lindsey 1970, McPhail and Lindsey 1970). This extended period of 251 allopatric isolation between populations in areas free of glacial ice led to significant levels 252 of evolutionary divergence, complicating resolution of taxonomic relationships (Bernatchez

4 253 and Wilson 1998, Reist et al. 1998). Moreover, morphological characters commonly used to 254 describe the species are also highly subject to modification by the conditions under which 255 fish develop, so the same population may appear quite different among lakes (Lindsey et al. 256 1970, Loch, 1974, Woodger, 1976, Lindsey, 1981, Sandlund, 1992). The broad distribution 257 of the species has consequently led to the same Coregonine species having been given 258 different names in North America, Russia, and Europe (McPhail and Lindsey, Lindsey 259 1988). For this reason, it has been argued that species names and descriptions remain 260 tentative until more information is available across their distribution (McPhail and Lindsey 261 1970, McPhail 2007).

262 Phylogenetic studies have been largely successful in resolving the Coregonine tree 263 at the species level (Bodaly et al. 1991, Bernatchez and Dodson 1991, Bernatchez et al. 264 1991a, Bernatchez and Dodson 1994). There are 28 extant species recognized in the sub- 265 family Coregonineae. The major grouping of whitefishes in this family include three 266 genera; Stenodus, Prosopium, and Coregonus (Reshetnikov 1988). Genetic data suggests 267 that while Prosopium diverged in the order of 10 million years ago (mya) during the 268 Miocene – early Pliocene period, Stenodus is more closely related sharing a common 269 ancestor with Coregonus approximately 2.5 mya (Bernatchez et al. 1991a). The entire 270 radiation of extant Coregonine species has been rapid, occurring during and since the 271 Pleistocene glaciation dating back approximately 1.5 mya (Bernatchez et al. 1991a). 272 Phylogenetic analyses based on different DNA markers have shown congruence in their 273 findings with respect to branching patterns, suggesting that the Coregonine tree is an 274 accurate representation of the evolutionary history of the sub-family Coregonineae. (Bodaly 275 et al. 1991b, Bernatchez et al. 1991a)

276 One of the most complex nodes of the tree represents two widespread forms of Lake 277 Whitefish in the northern hemisphere, Coregonus clupeaformis in North America and 278 Coregonus lavaretus in Eurasia (Bodaly et al. 1991, Bernatchez and Wilson 1998, Reist et 279 al. 1998, McPhail 2007). Several lines of evidence from phylogenetic analyses suggest that 280 Beringia (Northwestern North America) was recolonized by populations of Coregonus 281 lavaretus that dispersed extensively in Eurasia and into Beringia, and subsequently merged 282 with an endemic Beringian whitefish group (Bodaly 1977, Bodaly et al. 1991, Bernatchez 283 and Dodson 1994). However, two reproductively isolated populations, one Eurasian and 284 one Beringian, persist in the region.

285 In North America, the Lake Whitefish was first described by Mitchill in 1818 as 286 Salmo clupeaformis (from Lake Huron) (Koelz 1929, Lindsey et al. 1970, Scott and 287 Crossman 1973). Three taxonomically described species currently fall within the 288 Coregonus clupeaformis species complex; (C. clupeaformis, Mitchill; C. pidschian, 289 Gmelin; and C. nelsonii, Bean) (Lindsey 1963a, McPhail and Lindsey 1970, Alt 1971, 290 Lindsey 1988, Bean 1884). Details about the description, distribution, and PDU status of 291 these additional species within the complex are described in section 1.1.1.

5 292 The taxonomic issues in whitefish are not unique to North American Coregonus 293 clupeaformis complex. Indeed, taxonomic confusion also persists for Coregonus lavaretus 294 in regions of Europe and Russia (Bodaly et al. 1994, Beaumont et al. 1995, Svaerdson 295 1998, Douglas et al. 2005). Several populations of Coregonus lavaretus in these regions 296 exhibit phylogeographic patterns of reproductive isolation as well as ecological speciation 297 similar to their counterparts in North America (Luczynski et al. 1995, Sandlund et al. 1995, 298 Reist et al. 1998, Svaerdson 1998, Bernatchez and Wilson 1998, Bernatchez et al. 1999, 299 McDermid et al. 2005, Ostbye et al. 2005, Ostbye et al. 2006). These studies suggest that 300 similar ecological and evolutionary processes reported in C. clupeaformis, namely isolation 301 during periods of glaciation followed by ecological opportunity in post glacial 302 environments, have also resulted in cases of rapid population divergence in C. lavaretus 303 (Ostbye et al. 2006). Consequently, the criteria used to assign DUs in C. clupeaformis may 304 be useful for other Coregonine populations. This report seeks to identify designatable units 305 at the appropriate level only in Canadian Lake Whitefish populations, therefore only 306 taxonomic designations (past and present) for whitefish within the complex will be 307 considered within this assessment.

308

309 II. Key for establishing designatable units for COSEWIC

310 1. The putative designatable unit (PDU) is a distinct taxonomic entity or qualifies as a 311 distinct biological species?

312 1.1. Taxonomic entities in the Lake Whitefish species complex 313

314 Lake Whitefish exhibit tremendous phenotypic variation in morphological and life 315 history characteristics. Consequently, the species has been split into several distinct, albeit 316 phenotypically based taxa as a result of these differences (McPhail and Lindsey 1970, Loch 317 1974, Woodger 1976, Lindsey and McPhail 1986, Lindsey 1988). Gill rakers have 318 traditionally been the marker used to address taxonomy in most Coregonine populations 319 (Lindsey 1981). As gill raker variation is heritable, it was thought to give generalized 320 information about the genetic status of the population concerned (Svärdson 1957, Svärdson 321 1965). For example, North American members of the genus Coregonus that are not ciscoes 322 (i.e., have fewer than 33 gill rakers and have the snout overhanging the tip of the lower jaw, 323 Figure 1b) were deemed either broad whitefish, Coregonus nasus, (classified as having 324 short gill rakers and a wide head) or “humpback” whitefishes, Coregonus clupeaformis 325 (McPhail and Lindsey 1970). Despite the perceived utility of this morphological marker 326 towards differentiating populations of whitefish, interpretations of Coregonus taxonomic 327 classification have remained confusing, contributing to its general classification as a species 328 complex (McPhail and Linsdey 1970, Woodger 1976, Alt 1979, Lindsey 1988). Below,

6 329 species that remain taxonomically classified within the Lake Whitefish species complex are 330 described. It is important to note that most of these species have been identified primarily 331 on the basis of gill raker and other morphometric data. Thus, the current genetic and 332 historical information that may support or refute their classification as a distinct biological 333 species warranting DU consideration is also discussed.

334

335 1.1.1 Coregonus lavaretus and Coregonus pidschian 336

337 Lake Whitefish in northwestern Canada consist of some populations that colonized 338 the lakes and rives postglacially from Siberia. These Lake Whitefish have been classified 339 as C. pidschian and are routinely described as part of the C. clupeaformis species complex 340 (low modal gill raker counts, 20-22, named by Bean 1788 as Salmo pidschian from the Ob 341 River in Siberia) (Lindsey 1963a, Lindsey 1963b, Alt 1979, Page and Burr 1991, 342 Mecklenberg et al. 2002). The distribution of the Siberian Lake Whitefish extends 343 westward at least as far as the Ob River in Siberia and possibly into Europe (Walters 1955, 344 Bodaly et al. 1994, Politov et al. 1999, Politov et al. 2000). It is thought the Siberian Lake 345 Whitefish lived in huge glacial lakes that persisted through most of Siberia during the 346 Pleistocene (Denton and Hughes 1981, Lindsey and McPhail 1986) and were able to 347 disperse to northwestern North America via Beringian land bridges that opened during 348 glacial maxima (Lindsey and McPhail 1986).

349 Allozyme and mtDNA phylogeographic analyses have since demonstrated that 350 populations originating from the Bering Sea and inhabiting northwestern Canada (and 351 Alaska) are much more closely related to Eurasian populations of whitefish (C. lavaretus) 352 than to any other North American populations (Bodaly et al. 1991b, Bodaly et al. 1994, 353 Bernatchez et al. 1991a, Bernatchez and Dodson 1994, Sajdak and Phillips 1997). In fact, 354 Bernatchez et al. 1991 found that Coregonus pidschian sampled from Siberia had mtDNA 355 signatures identical or very closely related to those observed in C. lavaretus from Europe. 356 Consequently, populations of C. pidschian described in Northwestern America are most 357 likely conspecific with Eurasian Coregonus lavaretus and not Coregonus clupeaformis. If 358 this is indeed the case then the name C. lavaretus (Linnæus 1758) should have priority over 359 C. pidschian (Gmelin 1789).

360 Such observations have important consequences for our understanding of the Lake 361 Whitefish species complex in Canada. Conspecific populations of Coregonus pidschian 362 (currently part of the Coregonus clupeaformis complex) with C. lavaretus suggest that there 363 are two distinct species of Lake Whitefish inhabiting northwestern North America, (C. 364 lavaretus/C. pidschian in the upper Liard River and C. clupeaformis in Mackenzie River 365 region, Figure 2) (Smith and Todd 1984, Sajdak and Phillips 1997, McPhail 2007).

7 366 This Beringian taxonomic grouping, with close affinities to Eurasian populations of 367 C. lavaretus/C. pidschian should be considered a DU under C. clupeaformis until a more 368 thorough assessment can be made. Future assessments to resolve this DU will clearly 369 require additional genetic and morphometric data (Table 1, Figure 2).

370

371 1.1.2 Coregonus nelsonii 372

373 C. nelsonii was described as distinct from C. clupeaformis due to different gill raker 374 counts from Nulato on the Yukon River in northwestern North America (Bean 1884. 375 Lindsey 1963a). Lake Whitefish classified as nelsonii have now been reported for most of 376 the Yukon River, Paxson Lake, the Copper River system, and sporadically around the coast 377 of northern Alaska, in the Mackenzie River delta and in the Anderson River. Because of the 378 intermediate number of gill rakers, C. nelsonii has also been hypothesized to represent 379 hybrid populations between C. clupeaformis and C. pidschian/C. lavaretus (McPhail and 380 Lindsey 1970). The most recent genetic evidence suggests that Coregonus nelsonii is likely 381 conspecific with the true Beringian Lake Whitefish glacial race (Bernatchez et al. 1999, 382 McPhail 2007) and therefore does not comprise a distinct taxonomic group in this context 383 (but see Beringian glacial race DU below).

384

385 1.1.3 Coregonus huntsmani 386

387 The endangered, anadromous Atlantic whitefish, Coregonus huntsmani, inhabits the 388 Tusket and Petite Riviere watersheds in Nova Scotia. The terminal mouth of the Atlantic 389 whitefish suggests that it is not the progenitor of C. clupeaformis (McPhail and Lindsey 390 1970, but see Figure 1a). Lateral line scale numbers differs (average of 94 in Atlantic 391 whitefish versus an average of 77 in Lake Whitefish, as does the number of vertebrae 392 (average of 65 in Atlantic whitefish versus 61 in Lake Whitefish, Edge et al. 1991). Atlantic 393 whitefish are also purported to have smaller teeth than Lake Whitefish (Edge et al. 1991, 394 Hasselman 2003). Genetically, Atlantic whitefish differ from both Lake Whitefish and 395 cisco (Bernatchez et al. 1991b, Murray 2005). In fact, Bernatchez et al. (1991) found that 396 among 21 taxa C. huntsmani was the most distant in the assemblage compared to all other 397 Coregonus species, including ciscoes. Thus, C. huntsmani is as divergent as the genus 398 Stenodus, suggesting this species represents an ancestral form (Behnke 1972). 399 Consequently, C. huntsmani should be treated under its own assessment and is not 400 considered further here.

401

8 402 1.2. Are these taxonomic groups distinct biological species? 403

404 1.2.1 Limnetic and benthic Lake Whitefish 405

406 Several sympatric pairs of Lake Whitefish inhabit northern temperate lakes in 407 Canada (e.g., Lindsey 1963b, Fenderson 1964; Bodaly 1977, Bodaly et al. 1992a, 408 Bernatchez et al. 1996). These so-called “species pairs” are characterized by a derived form 409 that is typically slower growing, matures at a much earlier age and size, and lives in the 410 limnetic zone of lakes. This is compared to a larger benthic form that grows faster and to a 411 larger size, matures at a later age, and lives sympatrically within the benthic zone of lakes. 412 For these reasons, the derived populations are often referred to as ‘limnetic’ or ‘dwarf’ 413 because of the distinct differences in life-history, behavioural, and morphological characters 414 associated with the use of trophic resources (Fenderson 1964, Bernatchez et al. 1999).

415 How did limnetic and benthic Lake Whitefish evolve? Ecological opportunity (e.g., 416 the absence of competitors) as well as character displacement, is thought to have led to the 417 repeated parallel evolution and reproductive isolation of limnetic populations diverging in 418 sympatry with the respective niches of these environments (Vuorinen et al. 1993, Schluter 419 1996, Taylor 1999, Schluter 2000, Pigeon et al. 1997, Bernatchez 2004, Rogers and 420 Bernatchez 2006). Genetic evidence supports the observation that the limnetic form has 421 evolved repeatedly and independently in several lakes, leading to the hypothesis that the 422 ecology of the lakes has had a substantial influence on the evolution of these species pairs 423 (Lu and Bernatchez 1999). Furthermore, limnetic whitefish are only found in sympatry with 424 the benthic population and only in the absence of other limnetic coregonine fishes, such as 425 the cisco (Coregonus artedii), suggesting that divergent natural selection resulting from 426 ecological opportunity within the limnetic niche is the most parsimonious explanation for 427 the evolution of the derived limnetic whitefish within the last 10 000 years (Pigeon et al. 428 1997, Bernatchez et al. 1999).

429 The limnetic and benthic Lake Whitefish dichotomy is now supported by several 430 genetically-based phenotype environment associations (Bernatchez 2004). A phenotype- 431 environment association is a trait that (1) differs between populations, (2) has a clear utility, 432 (3) is genetically based, and (4) is under the influence of divergent natural selection, 433 thereby fulfilling the “adaptive” criterion. Identifying these traits is a key step towards 434 understanding the variation underlying differences between the species pairs and the action 435 of selection maintaining their divergence in sympatry (Schluter 2000). Recent evidence 436 suggests that Fenderson (1964) accurately described the evolution of dwarfism in Lake 437 Whitefish as a primarily physiological adaptation. These adaptations, namely slower 438 growth and earlier maturation, have survival value in the face of adverse conditions (e.g., 439 increased predation) of the limnetic environment (Fenderson 1964, Chouinard and

9 440 Bernatchez 1998, Rogers and Bernatchez 2005, Rogers and Bernatchez 2007). Earlier 441 maturation at a smaller size is under genetic control and adaptive in postglacial 442 environments, because limnetic fish that mature at a younger age likely have a higher 443 probability of reproducing, thereby increasing fitness. Indeed, gene flow is significantly 444 reduced at the genes underlying growth and fecundity, indicating selection has been 445 maintain differences between the species pairs at these adaptive traits (Rogers and 446 Bernatchez 2007). Overall, these studies suggest that limnetic and benthic Lake Whitefish 447 species pairs fulfil the definition of a biological species as groups of interbreeding natural 448 populations that are reproductively isolated from other such groups (Mayr 1963).

449 However, the biological significance of phenotypic and genetic variation in these 450 species pairs must be considered in light of potential limitations in lakes inhabited by 451 species pairs that are currently “data deficient” (see Section 3). In section 1.2.2, the 452 distribution and characteristics of species pairs are provided with potential DU designation. 453 When considering the relevance of PDUs as “distinct biological species”, knowledge of 454 sufficient barriers to gene flow between species will be considered adequate for DU status 455 according to the key. The key further states that under the assumption that two or more 456 populations of a single taxonomic unit are found in reproductive sympatry and demonstrate 457 significant reproductive isolation from one another, they should be considered as valid 458 biological species even with the same taxonomic designation (Taylor 2006). Under this 459 assumption, each species pair qualifies as a DU that is distinct from one another, as well as 460 from all other populations or DUs of the species. This is required because they will have 461 different habitat requirements, behvaiour, physiology, and other biological features that are 462 necessary to take into account in subsequent assessments. When some of the data to assess 463 this is lacking, the putative species pair will be given provisional DU status until a more 464 thorough assessment can be made. When only inferential reasoning or limited unpublished 465 data is found to support their existence, the putative species pair will be given deferred DU 466 status.

467

468 1.2.2 Distribution and characteristics of limnetic and benthic pairs in Canada 469

470 In Canada, limnetic and benthic species pairs inhabit at least 17 lakes from the 471 Yukon to Labrador (Lindsey, 1970, McPhail and Lindsey 1970, Bernatchez and Dodson 472 1990) (Figure 3). However, these species pairs also exist in at least 22 lakes in the 473 headwaters of the St. John River watershed, but most of these are in northern Maine 474 (Fenderson 1964, Kirkpatrick and Selander 1979, Bernatchez et al. 1999). The available 475 data for each species pair varies, but overall demonstrates parallel patterns of adaptive 476 divergence within these environments (Table 1).

10 477

478 1.2.2.1 Squanga Lake, Yukon Territory 479 60° 28' 60 N, 133° 37' 60 W

480

481 Limnetic and benthic Squanga Lake Whitefish differ in gill raker count, distance 482 between gill rakers, depth selection, morphology (size of the head and size of their fins 483 relative to their body), diet, and spawning times (Lindsey 1963b, Bodaly 1979, Lindsey 484 and McPhail 1986, Bodaly et al. 1987, Bodaly et al. 1988, Bodaly et al. 1992, Bodaly 485 2007). However, they do not fall within the limnetic and benthic dichotomy with respect to 486 size. Lindsey (1963) reported that, even after handling hundreds of specimens, it was 487 impossible to assign a fish without first examining the gill rakers. Despite this, catch reports 488 from gill nets suggest that the proportion of catch in the limnetic and benthic zones of the 489 lake are highly correlated, with low gill raker fish comprising over 75% of the bottom catch 490 and the pelagic high gill raker species comprising 61% of the limnetic catch (Bernatchez et 491 al. 1996). The presence of this species appears to be associated with the absence of the 492 Least Cisco (Coregonus sardinella). Thus, certain aspects of the Squanga Lake limnetic and 493 banthic pair are consistent with the limnetic/benthic dichotomy observed in other parts of 494 Canada.

495 Squanga Lake Whitefish are of Beringian glacial origin, but the whitefish in this 496 lake represent a case where C. lavaretus/pidschain and Beringian C. clupeaformis overlap. 497 This is supported by genetic data where the existence of sympatric pairs is best explained 498 by the secondary contact of two monophyletic whitefish groups that evolved in allopatry 499 during the last glaciation events (Bernatchez and Dodson 1994, Bernatchez et al. 1996). 500 These results therefore support polyphyletic origins of limnetic and benthic populations, 501 meaning the species pairs among lakes have evolved independently. Others have suggested 502 that two distinct species inhabit the lake, C. lavaretus/pidschian and C. clupeaformis 503 (McPhail and Lindsey 1970). Gill raker counts in Squanga Lake are higher than what is 504 typically found in C. lavaretus/pidschian, but this may be due to character displacement 505 from its low gill raker C. clupeaformis counterpart within the lake.

506 Elevated admixture of Beringian mtDNA and nuclear genes has been found between 507 the species pair (Bodaly et al. 1992, Bernatchez et al. 1996). Nei’s genetic distance between 508 these high and low gill raker populations based on allozymes is low (0.02). In fact, the 509 weakest support for reproductive isolation between Squanga Lake Whitefish ecotypes is 510 observed in Squanga Lake, which exhibit significant but mild genetic differences at the 511 PGM-2* locus only (Bodaly et al. 1992, Bernatchez et al. 1996). Nevertheless, the high and 512 low gill raker dichotomy, in association with significant segregation for different spawning 513 habitats, along with levels of gene flow that vary depending on the degree of trophic

11 514 specialization, is suggestive of partial reproductive isolation (Bodaly et al. 1988, 515 Bernatchez et al. 1996, Bodaly 2007).

516 It is also important to note that the high gill raker limnetic population is apparently 517 unable to coexist with ciscoes, which may be more effective consumers of zooplankton. 518 Therefore, introduction of ciscoes to a lake inhabited by Squanga Whitefish is suspected to 519 result in the elimination of the limnetic whitefish. Similarly, a limnetic whitefish population 520 could be significantly reduced by the addition of a piscivorous (fish-eating) fish to its lake 521 (SARA 2006), consistent with other species pairs in Canada. The Squanga Lake Whitefish 522 is currently listed by COSEWIC as ‘special concern’.

523

524 Overall, based on the evidence of morphological and life history differentiation, this 525 species pair represents reproductively distinct populations (Lindsey 1963, Bodaly 1979, 526 Bodaly et al. 1988, Bodaly et al. 1992, Bernatchez et al. 1996, Bodaly 2007). Consequently, 527 both the high gill raker and the low gill raker Squanga Lake Whitefish species pair warrant 528 DU status. However, as detailed in section 1.1.1, more information is needed on the 529 evolutionary history of this species pair, particularly to confirm the possibility of two 530 distinct species (C. lavaretus and C. clupeaformis), the former exhibiting derived 531 characters such as high gill raker number possibly due to evolutionary character 532 displacement (Table 1, Figure 2).

533

534 1.2.2.2 Little Teslin Lake, Yukon 535 60° 28' 60 N, 133° 37' 60 W

536

537 The Little Teslin species pair differ significantly with respect to morphology (gill 538 raker) and depth selection. Depth selection of high and low gill raker populations is highly 539 significant, with the high gill raker population comprising over 98% of the surface catch 540 (Bernatchez et al. 1996). This appears to be associated with the absence of the Least Cisco 541 (Coregonus sardinella) in the four southern Yukon lakes where this species occurs.

542 Genetically, the more limnetic form is fixed for the Eurasian clade, whereas the 543 Beringian clade predominates in the benthic form (Bodaly et al. 1992, Bernatchez et al. 544 1996). This indicates that, as in Squanga Lake, their sympatric occurrence originated from 545 the secondary contact of two monophyletic groups of whitefish that evolved allopatrically 546 (Bernatchez et al 1996, Bernatchez et al. 1999).

12 547 Given the observed adaptive divergence and significant reproductive isolation 548 (Bodaly 1979, Bernatchez et al. 1996, the Little Teslin species pair warrant DU status 549 (Table 1, Figure 2)

550 1.2.2.3 Dezadeash Lake, Yukon 551

552 Dezadeash Lake is a separate catchment than Squanga or Little Teslin Lake, but the 553 species pair inhabiting Dezadeash Lake also differ primarily with respect to depth selection 554 and gill raker number (Lindsey 1963b). Gene flow is most restricted between the species 555 pairs in this lake compared to other lakes harbouring the species pairs in the Yukon 556 (Bernatchez et al. 1996). Both high and low gill raker populations were associated with the 557 Eurasian clade of the Beringian glacial race (Bernatchez et al. 1996). This was congruent 558 with results based on isozymes and suggested a distinct evolutionary origin for ecotypes of 559 Dezadeash Lake compared to those found in the other two lakes (Bodaly et al. 1992b).

560 Based on the adaptive trait divergence and significant reproductive isolation 561 (Bodaly 1979, Bernatchez et al. 1996, the Dezadeash Lake species pair warrants DU status 562 (Table 1, Figure 2).

563

564 1.2.2.4 Teenah Lake, Yukon 565

566 This lake, south of Squanga lake, is known to inhabit the species pairs (Bodaly 567 1979, Bernatchez et al. 1996). It has not been included in the same genetic surveys and 568 therefore remains data deficient.

569 Based on the existing information for the Squanga Lake species pair, Lake Whitefish 570 inhabiting Teenah Lake should be granted deferred DU status until further genetic and 571 phenotypic information can confirm their status.

572

573 1.2.2.5 Hanson Lake, Yukon 574 64.01°N 135.35°W

575

576 This lake was reported to have been inhabited by the Squanga Lake Whitefish. Their 577 extinction occurred when the Hanson Lake fish fauna were poisoned in 1963 with 578 toxaphene to facilitate the introduction of rainbow trout for sportfishing (Vetter et al. 579 1999).

13 580

581 1.2.2.6 Dragon Lake, British Columbia 582 52°59′N, 122°29′W

583

584 This lake was hypothesized to have a slow growing and a fast growing whitefish species 585 pair (McPhail and Lindsey 1970, McPhail 2007). However, this population was also 586 eradicated to promote trout fishing (McPhail and Lindsey 1970).

587

588 1.2.2.7 Lower Liard River, British Columbia

589

590 This system is reported to potentially be inhabited by a slow-growing and fast 591 growing species pair (McLeod et al. 1979). As discussed in section 1.1.1, it is possible that 592 the Liard River might inhabit a migratory population of Lake Whitefish or that the system 593 represents contact zone between Eurasian and North American clades (McPhail 2007)

594 Overall, there appears to be some evidence for differential growth between the 595 putative populations (McLeod et al. 1979). However, the data are not clearly supported by 596 phenotypic of genetic analyses. Therefore, deferred DU status should be given to the 597 potential species pair in this system until further information is available.

598 Although no mtDNA has been collected, allozymes have also suggested that this 599 system is a zone of secondary contact between Beringian, Mississippian, Nahanni, and 600 Eurasian Lake Whitefish (Table 1, Foote et al. 1992). This renders the Liard River system a 601 very important and unique system towards resolving questions about the species status of 602 these glacial races and the potential role for evolutionary reinforcement within this system.

603

604 1.2.2.7 Great Slave Lake, NWT 605 61°40’N, 114°00’W

606

607 This is the second largest lake in the North West Territories (behind Great Bear 608 Lake) and the deepest lake in North America (maximum depth = 614m). These Lake 609 Whitefish are of Mississippian glacial origin (Foote et al. 1992). Rawson (1947) reported a 610 dark, rounded form in sympatry with the benthic form, although the phenotypic data are 611 inadequate and the genetic data are unknown (Scott and Crossman 1973).

14 612 As these traits are not part of the known suite of phenotype-environment 613 associations evolving in limnetic-benthic species pairs, this putative species pair is data 614 deficient and does not currently warrant deferred DU status.

615

616 1.2.2.8 Lake Athabasca, NWT 617 59°16’N, 109°27’W

618

619 Lake Whitefish in Athabasca are of Mississippian glacial origin (Foote et al. 1992). 620 Rawson (1947) also reported a dark, distinctly rounded form in sympatry with the benthic 621 form in this lake although the phenotypic data are inadequate and the genetic data are 622 unknown (Scott and Crossman 1973).

623 As these traits are not part of the known suite of phenotype-environment 624 associations evolving in limnetic-benthic species pairs, this putative species pair is data 625 deficient and does not currently warrant DU status.

626

627 1.2.2.9 Lake Opeongo, Ontario 628 45°42’N, 78°24’W

629

630 Lake Opeongo is Algonquin Park’s biggest lake with an area of 5800 ha, a mean 631 depth of 14.6 m (maximum depth 49.4 m). Kennedy (1943) discovered that the lake 632 contains a limnetic and benthic species pair. Limnetic whitefish apparently grow slower 633 than benthics in the lake, although Kennedy (1943) reported that 1+ limnetic whitefish were 634 109 mm long while their benthic counterparts were 78 mm long at the same age. The 635 species pair also differs with respect to age at maturity and fecundity (Kennedy 1943, 636 Ihssen et al. 1981, Cucin and Faber 1985).

637 Although genetic evidence from Lake Opeongo is suggestive of differences between 638 limnetic and benthics, the focus of the study was a stock assessment in comparison with 639 allopatric populations in Ontario (Ihssen et al. 1981). Therefore, the level of reproductive 640 isolation between limnetics and benthics in this environment remains inconclusive.

641 Based on the divergence in adaptive growth rate, age at maturity and fecundity 642 between limnetic and benthics in a direction parallel to other species pairs, the Lake 643 Whitefish in Lake Opeongo warrants DU status. However, in the absence of genetic data 644 and compared to other species pairs in Canada, a provisional DU status may be more

15 645 appropriate until genetic information can confirm phylogenetic grouping and/or 646 reproductive isolation (Table 1, Figure 2).

647

648 1.2.2.10 Como Lake, Ontario 649 47°55’N, 83°30’W

650

651 Located in northern Ontario, Como Lake has a surface area of 1596 ha and a mean 652 depth of 9.4 m. Significant differences in body morphometry between limnetic and benthics 653 were found for five of nine meristic characters and 14 of 19 morphometric characters, 654 consistent with the magnitude and direction of parallel adaptive trait divergence in other 655 limnetic and benthic species pairs (Bodaly et al. 1991, Vuorinen et al. 1993).

656 Genetic differences in the frequency of mtDNA haplotypes has shown that Como 657 Lake Whitefish have diverged into a genetically distinct, reproductively isolated limnetic 658 and benthic species pair (Vuorinen et al. 1993). The mtDNA evidence also demonstrated 659 that the benthic Lake Whitefish are most likely of Atlantic glacial origin (a largely 660 diagnostic mtDNA haplotype found in southern Québec) while the limnetic population is of 661 Mississippian glacial origin (Bodaly et al. 1991, Vuorinen et al. 1993).

662 Given the adaptive trait divergence between the species pairs and the secondary 663 contact between genetically distinct Atlantic and Mississippian glacial PDUs, both limnetic 664 and benthic Como Lake Whitefish warrant DU status (Table 1, Figure 2).

665

666 1.2.2.11 Lake Superior, Ontario 667 47.7°N – 87.5°W

668

669 Edsal (1960) reported a species pair within Lake Superior but phenotypic and 670 genetic data were inadequate. A lack of phenotypic and genetic data for the putative 671 species pair do not support DU status for Lake Whitefish inhabiting Lake Superior.

672 It is noteworthy that species interactions between the invasive zebra mussel and the 673 decline of Lake Whitefish in Lake Superior have resulted in significant life history changes 674 for the Lake Whitefish within this ecosystem (Nalepa et al. 2005). There have been 675 observations that declines in abundance of the benthic amphipod, Diporeia spp., have 676 resulted in Lake Whitefish adopting a more limnetic life history (Nalepa et al. 2005, 677 Bernatchez 2005). Given the rapid evolution of the derived limnetic species upon

16 678 colonization of a limnetic environment, predictions can be made about the evolutionary 679 changes this alternative life history will have on Lake Superior whitefish. Namely, this 680 interaction may cause evolutionary changes that lead to a slower growing population that 681 matures earlier.

682

683 1.2.2.12 Lake Simcoe, Ontario 684 44°26′12″N, 79°20′21″W

685

686 Lake Whitefish are an important recreational fishery within this lake which has been 687 heavily stocked with whitefish since 1982 (Lasenby et al. 2001). MacCrimmon and Skobe 688 (1970) reported a sympatric species pair of Lake Whitefish but subsequent evidence 689 remains to be gathered (Scott and Crossman 1973). A lack of phenotypic and genetic data 690 for the putative species pair do not support DU status for Lake Whitefish inhabiting Lake 691 Superior.

692

693 1.2.2.13 Témiscouata Lake, Quebec 694 47°36′00″N, 68°45′00″W

695

696 Lac Témiscouata is a part of a series of lakes in the Allagash Basin and Saint John 697 River Watershed in south-eastern Quebec. The lake contains one of the most extensively 698 studied species pairs of Lake Whitefish. Significant adaptive trait divergence has evolved 699 between populations, with the limnetics attaining an average size at maturity of 188 mm 700 while the benthic population attains an average size of 238 mm at maturity. Limnetic and 701 benthics within this lake also differ with respect to gill rakers, behaviour, growth, and 702 spawning habitat (Lu et al. 1999). Limnetic forms in this lake spawn in a groundwater 703 stream feeding the lake while benthics spawn in the lake (Lu and Bernatchez 1998).

704 Significant genetic differentiation and reproductive isolation exists between 705 Témiscouata limnetic and benthic (Lu et al. 1999). Once believed to be of primarily 706 Acadian glacial origin (Bernatchez et al. 1990), subsequent genetic analyses have revealed 707 substantial hybridization between Atlantic and Mississipian glacial races in the benthic 708 population while the limnetic are primarily of Acadian origin (Lu et al. 2001, Rogers 2001)

709 Based on the significant suite of adaptive trait divergence and genetic differentiation 710 in Témiscouata Lake Whitefish, both the limnetic and benthic populations warrant DU 711 status in this lake (Table 1, Figure 2).

17 712

713 1.2.2.14 East Lake, Quebec 714 47°11′00″N, 69°33′00″W

715

716 Phenotypic differentiation in size between limnetic and benthics in East Lake is 717 highly significant, with limnetic fish averaging 153 +/ – 17 mm at maturity while benthics 718 attain an average size at maturity of 285 +/ – 67 mm (Lu et al. 2001). East Lake remains 719 one of the least phenotypically differentiated species pairs with respect to gill raker 720 variation, but the extent of morphological differentiation (other than adult size) between 721 ecotypes is significantly associated with trophic use (Chouinard et al. 1996, Bernatchez et 722 al. 1999). More pronounced morphological differentiation between ecotypes in other lakes 723 (e.g., Cliff Lake) translated into stronger trophic niche partitioning during periods of food 724 depletion than in East Lake, where limnetic and benthic fish feed mainly on planktonic and 725 epibenthic prey, respectively (Chouinard et al. 1996, Chouinard and Bernatchez 1998, 726 Bernatchez et al. 1999, Landry et al. 2007). These results have been important towards 727 demonstrating a functional link between morphology and trophic use (Bernatchez et al. 728 1999). These studies have also demonstrated that the persistence of ecological opportunity 729 for differential resource use throughout their ontogeny may be the primary selective force 730 promoting the extent of specialization reached.

731 Lake Whitefish in East Lake also support the hypothesis for an intralacustrine origin 732 of the species pairs (Pigeon et al. 1997). Both limnetic and benthic Lake Whitefish are of 733 Acadian glacial origin, more closely related to each other than other populations at both 734 mtDNA and Amplified Fragment Length Polymorphism (AFLP) DNA markers 735 (Bernatchez and Dodson 1990, Campbell et al. 2003, Campbell and Bernatchez 2004). The 736 mtDNA analysis supports a single founding population characterized by an original 737 haplotype (Bernatchez and Dodson 1990), followed by a novel mutation of another 738 haplotype following colonization. Genetic studies between the species pair have identified 739 candidate genes associated with growth that exhibit parallel adaptive divergence in other 740 limnetic-benthic species pairs (Rogers and Bernatchez 2005, Rogers et al. 2007).

741 Altogether, the adaptive trait divergence and genetically distinct ecotypes of 742 monophyletic origin suggests that both the limnetic and benthic Lake Whitefish of East 743 Lake should be granted DU status (Table 1, Figure 2).

744

745 1.2.2.15 Caniapiscau Lake, Quebec 746

18 747 Fortin and Gendron (1990) reported limnetic and benthic species pairs three Québec 748 lakes. The first, Caniapiscau Lake, exhibits phenotypic differences in growth, maturity, and 749 spawning. A limnetic/benthic phenotypic dichotomy in this phylogenetically distinct 750 assemblage supported the hypothesis that similar phenotypic patterns evolved in parallel in 751 separate, yet closely related lineages (Pigeon et al. 1997). Genetic evidence for 752 reproductive isolation is insignificant with mtDNA (Pigeon et al. 1997, but given the young 753 intralacustrine divergence of the species this genetic marker may not have had enough 754 resolution and should be followed up with more appropriate genetic markers such as 755 microsatellite DNA). Indeed, recent research on other north temperate fishes has pointed 756 out the possibility that this region in central Quebec may have been colonized by distinct 757 ‘subglacial’ races (not strictly mtDNA glacial races, but very distinct lineages within a 758 ‘traditional’ mtDNA race) originating from within a Mississippian race in each species 759 (Fraser and Bernatchez 2005).

760 Overall, Taylor (2006) suggested that if a diagnostic character or suite of characters 761 was identified with limited sampling, DU status might be deferred or made provisional until 762 the diagnostic nature of the trait was confirmed after more extensive sampling. This fits the 763 current situation with the Caniapiscau Lake species pair. Given the parallel patterns of 764 adaptive divergence in this limnetic and benthic dichotomy, these limnetic and benthic 765 populations should be given provisional DU status (Table 1, Figure 2)

766

767 1.2.2.16 Lac Outardes II, Quebec 768

769 Fortin and Gendron (1990) reported a limnetic and benthic species pair in Lac 770 Outardes II. This putative species pair exhibited phenotypic differences in growth, maturity, 771 and spawning date. This group is similar to Caniapiscan Lake in that there is evidence for 772 only the Mississipian glacial race in the lake (Bernatchez and Dodson, 1991). Genetic 773 evidence for reproductive isolation was also insignificant with mtDNA (Pigeon et al. 1997).

774 DU status for this species pair should be provisional until the diagnostic nature of 775 the traits is confirmed after more extensive sampling. Given the parallel patterns of adaptive 776 divergence in this limnetic and benthic dichotomy, these populations should be given 777 provisional DU status (Table 1, Figure 2)

778

779 1.2.2.17 Lac Manicouagan V, Quebec 780

19 781 Fortin and Gendron (1990) reported a limnetic and benthic species pair in Lac 782 Manicouagan V. This putative species pair also exhibited phenotypic differences in growth, 783 maturity, and spawning date. This group is similar to Caniapiscan Lake and Lac Outardes II 784 in that only the Mississipian glacial race is found in this lake (Bernatchez and Dodson, 785 1991). Genetic evidence for reproductive isolation is also insignificant with mtDNA 786 (Pigeon et al. 1997).

787 DU status should be provisional until the diagnostic nature of the trait was 788 confirmed after more extensive sampling. Given the parallel patterns of adaptive divergence 789 in this limnetic and benthic dichotomy, these populations should be given provisional DU 790 status (Table 1, Figure 2).

791 1.2.3 Hybridization of C. clupeaformis with other Coregonineae 792

793 Hybridization between C. clupeaformis and other Coregonines is rare but continues 794 to occur with several species in certain habitats (Smith 1992, Reist, 1992). For example, 795 hybrids with C. artedii, so-called “mule whitefish”, have been documented in Lake Erie and 796 the northwest (Koelz 1929, Scott and Crossman 1973). Mule whitefish possibly exhibit 797 heterosis with respect to growth rate (McPhail and Lindsey 1970). Several other cases of 798 hybridization have also been reported in the northwestern part of their range with inconnu 799 whitefish (Stenodus leucichthys), least cisco, (Coregonus sardinella), and Arctic cisco 800 (Coregonus autumnalis) (Alt 1971, Nelson and Paetz 1992, Reist et al. 1992).

801

802 1.2.4 Species status of Coregonus clupeaformis 803

804 Most studies suggest retaining the species status of C. clupeaformis populations 805 until more data are available (Walters 1955, Lindsey et al. 1970, Smith and Todd 1984, 806 McPhail 2007). Biogeographic and experimental data will be necessary to resolve these 807 issues. For example, C. clupeaformis has been taxonomically considered as the same 808 species as C. lavaretus (Walters 1955, Reshetnikov 1963). Although these species show 809 distinct phylogenetic groupings (Bodaly et al. 1991, Bernatchez et al. 1994, Bernatchez and 810 Wilson 1998), direct evidence for the extent of reproductive barriers between the two is 811 unknown, leading some to suggest that perhaps the name C. lavaretus be applied to all Lake 812 Whitefish populations in the northern hemisphere (Bernatchez and Wilson 1998). 813 However, indirect evidence from genetic distance data is indicative of reduced 814 hybridization and introgression when the species do overlap, suggesting that reproductive 815 isolation may be considerable.

20 816 Overall, attempts to delineate subspecies status remain premature (McPhail and 817 Lindsey 1970, Lindsey 1988, Bernatchez and Wilson 1998). Importantly, the utility of gill 818 rakers to resolve the species’ status has been largely replaced by genetic measures of 819 population differentiation. The patterns of modal gill raker counts across Canada remain 820 compelling, with western Lake Whitefish populations exhibiting low modal counts (20-22, 821 conforming to observations for C. lavaretus/C. pidschian, described by Gmelin in 1788) 822 while eastern and southern populations have higher modal counts (26+, agreeing with 823 Mitchill’s classification for C. clupeaformis). Overall, variation in this character in some 824 populations was perhaps the first line of evidence that a combination of historical and 825 contemporary evolutionary or environmental forces may have been driving observed 826 differences (Scott and Crossman 1964, Loch 1974, Woodger 1976). For instance, 827 comparable ranges of gill raker variation may be generated within each of the species 828 classified under the C. clupeaformis complex via natural selection and environmental 829 influences (Scott and Crossman 1964, Woodger 1976, Lindsey 1981).

830 It has been well established that traditional taxonomic methods employing 831 phenotypic variation should not be the primary criterion used to reflect the evolutionary 832 history or distinctness of Lake Whitefish populations (Lindsey 1981). This has been 833 established in other systems with similar taxonomic issues (e.g., Osmerus sp., Taylor and 834 Bentzen 1993, Bernatchez et al. 1999, whereby the use of genetic data for resolving 835 taxonomic issues has increased our knowledge of evolutionary history.

836 1.3. Summary 837

838 • Many proposed species within C. clupeaformis complex have been identified on the 839 basis of morphological variation, leading to taxonomic designations not fully 840 supported by genetic evidence. 841

842 • Genetic evidence strongly supports the possibility that two species, C. lavaretus and 843 C. clupeaformis, inhabit British Columbia and the Yukon. 844

845 • Provisional DU status should be given to populations that represent distinct 846 Coregonus lavaretus species (Figure 2). 847

848 • Genetic evidence suggests that C. lavaretus inhabits the upper Yukon River and, 849 potentially the upper Liard River system. A more thorough analysis of this latter 850 system is needed. 851

21 852 • Coregonus huntsmani, already listed on the endangered species list, was considered 853 here in the context of the Lake Whitefish species complex based on previous 854 associations with C. clupeaformis. Under the criteria of the key, this population’s 855 taxomonic position places it outside the scope of the current assessment. 856 857 • Under the assumption that two or more populations of a single taxonomic unit found 858 in reproductive sympatry and demonstrating significant reproductive isolation from 859 one another are valid biological species even with the same taxonomic designation 860 (Taylor 2006), each limnetic and benthic pair of Lake Whitefish, distinct from other 861 such pairs, qualifies as a DU. 862 863 • Of the 17 lakes known to be inhabited by the putative species pairs, seven warrant 864 DU status for the limnetic and benthic species pairs inhabiting these lakes, leading 865 to 14 DU designations (Figure 2). 866 867 • In three lakes, where the species pairs are purported to exist, but locally adapted 868 traits have not been assessed, or genetic differentiation between limnetic and benthic 869 is unknown, provisional DU status was granted until the diagnostic nature of 870 adaptive traits is confirmed after more extensive sampling (6 provisional DUs) 871 (Figure 2). 872 873 • Three additional lakes, purported to be inhabited by species pairs, remain 874 completely data deficient and were identified as deferred DUs. 875 876 • The seven PDU species pairs of limnetic and benthic Lake Whitefish are distributed 877 throughout Canada but mainly concentrated in two regions of secondary contact 878 between PDUs representing glacial races (Yukon and southern Quebec) (Figure 3) 879 880 • The addition or invasion of other limnetic species (e.g., cisco or rainbow smelt) will 881 likely precipitate the extinction of the limnetic Lake Whitefish. 882

883 2. The PDU represents a major phylogenetic grouping separate from other groupings 884 within the taxon in question?

885 2.1. Glacial History of Canada and Postglacial dispersal of Lake Whitefish 886

887 The Pleistocene ice age was arguably the most significant event in the history of 888 most extant northern organisms. Nowhere is this more apparent than for organisms 889 inhabiting the Canadian landscape (McPhail and Lindsey 1970, Denton and Hughes 1981, 890 Bodaly et al. 1991, Pielou 1991, Hewitt 1996, Schluter 1996, Dawson 2002, Power 2002, 891 Curry 2007). North American icesheets were larger than those of Europe and Asia 892 combined, covering most of the country. Ancient freshwater habitats were destroyed by the 893 advancing glaciers, displacing or eradicating local populations (Pielou 1991).

22 894 Each glaciation spanned approximately 100 000 years, with the interglacial periods 895 consisting of a duration of 10-12000 years (Dawson 2002). Subsequent retreat of the most 896 recent glacial period (between 15000 and 8000 years ago) was equally a major 897 environmental change. Meltwater formed large proglacial lakes affecting the dispersal of 898 aquatic organisms. These lakes were larger than any existing lakes and changed frequently 899 as their size and volume were largely determined by the rate of melting. For example, 900 geological radiocarbon evidence from the Strait of Georgia has shown that approximately 901 10 000 years ago, a catastrophic release of impounded glacial lake water and melt water 902 flooded the lower Fraser River Valley and the Strait of Georgia, leaving a 22cm layer of 903 clay in the Strait of Georgia and likely resulting in less brackish surface water for a period 904 of years (Conway et al. 2001). This may have contributed to the dispersal ability of several 905 freshwater aquatic organisms able to tolerate the reduced salinity of the strait. Events such 906 as these leave important signatures in the geological records for reconstructing the 907 zoogeography of the time.

908 Glaciations affected Lake Whitefish populations in two major ways. First, 909 advancing ice sheets resulted in a reduction of available habitat for the species. Second, the 910 ice sheets resulted in extreme geographic and temporal isolation of whitefish populations. 911 In fact, as discussed below, Lake Whitefish could have only survived in a limited range of 912 habitats that were not iced over during the repeated advances and retreats of glacial ice 913 during the Pleistocene. These areas were limited to regions not affected by the glacial 914 advances, providing refuge for terrestrial and aquatic organisms. In North America, three 915 large areas served as the primary refugia for organisms during the Wisconsin glaciation 916 (Pielou 1991). These areas included an ice-free corridor that formed a northward pointing 917 peninsula of ice – free land just east of the Rocky Mountains. Beringia, consisting of most 918 of Alaska and the Yukon Territory, formed a huge refuge along the beds of two shallow 919 seas – the Bering and the Chukchi seas. Third, the coastal plains, east of the ice sheet, 920 which are now submerged and form the continental shelves of the Atlantic Provinces. 921 Several smaller refugia, so-called nunataks, persisted primarily on coasts and at the summit 922 of mountain peaks. Several nunataks have been identified, including but not limited to; 923 Vancouver Island, the Queen Charlotte Islands, and Cape Breton, Nova Scotia. Coastal 924 refugia are believed to have been common around the Gulf of Saint Lawrence, the Gaspé 925 Peninsula, and the coast of Newfoundland. Coastal refugia and the ecosystems within them 926 would have migrated slowly inland as the level rose and the ice retreated. Because the 927 current distribution of Lake Whitefish matches the areas covered by the North American 928 glaciers, phylogeographic approaches have been highly suitable towards elucidating the 929 postglacial history of this species complex.

930

931 2.2. Phylogeography of the Lake Whitefish species complex 932

23 933 In the following section the existence of five major glacial lineages within C. 934 clupeaformis is discussed (Figure 3). Extensive genetic resources have been developed 935 within the species complex over the years which are ideal towards addressing 936 phylogeographic issues (Tsuyuki et al. 1966, Lindsey et al. 1970, Hamada et al. 1997, 937 Sajdak and Phillips 1997, Vuorinen et al. 1998, Sendek 1999, Bernatchez et al. 1999, 938 Politov et al. 2000, McDermid et al. 2005). Phylogenetic and genetic characteristics for 939 each of these glacial lineages and their relationship to one another in the context of 940 establishing DUs are described. Several important differences within each glacial race that 941 will need to be considered for assessing status of each and particularly for deciding the 942 relative importance for future assessments of each DU are also described.

943

944 2.2.1 Beringian Glacial Race: 945

946 This group of whitefish inhabits Alaska and Yukon and Northwest territories and is 947 believed to have survived in the Beringian refuge during the last ice age (Lindsey 1975, 948 Lindsey and McPhail 1986). Fish collected in these areas consisted largely of two 949 significantly distinct mtDNA groupings; A and B (Bootstrap support > 80%, Bernatchez 950 and Dodson 1991) (Figure 3). As discussed in section 1.1.1, one of these groupings is 951 representative of C. lavaretus originating from Siberia when the land masses connected. 952 The second grouping represents a population of C. clupeaformis that survived the 953 Pleistocene in the Beringian refugia. The percentage of mtDNA divergence between these 954 clades A and B suggests that these two populations diverged in the order of 360 000 years 955 ago, representing the deepest phylogenetic split compared to all glacial races, with the 956 major genetic break between whitefish inside and outside of Beringia during the Kansan ice 957 advance. Interestingly, C. lavaretus/pidschian and Beringian C. clupeaformis overlap and 958 are sympatric in only three regions; the Yukon River, Chatanika Lake, and Squanga Lake. 959 Within these regions, there is some evidence of reproductive isolation between populations 960 (Lindsey 1963b).

961 The effective population size (Ne) for the Beringian Lake Whitefish was estimated 962 to be 72 000 (Bernatchez and Dodson 1990), although exact values are not likely accurate 963 (Fraser et al. 2007). However, the interesting aspect with this estimate is that the effective 964 population size is three times higher than the rest of Canada, indicating that Lake Whitefish 965 were more abundant in this zone during the ice age despite the relatively small surface area 966 in this refuge. The genetic evidence offers support for the hypothesis that this region was a 967 glacial refuge for fish fauna during the Pleistocene ice age. Bernatchez and Dodson (1990) 968 suggest other reasons why such a major difference may persist between Beringia and other 969 regions of the country, including the role of bottlenecks (Bernatchez et al. 1989), and gene 970 flow between Alaska and Eurasia (Lindsey and McPhail 1986). Evidence from mtDNA

24 971 suggests that this clade reached an eastern limit of the lower Mackenzie River where they 972 overlap with Lake Whitefish from the Mississippian glacial race. This is in contrast with 973 evidence from allozymes that Beringian whitefish dispersed towards central Canada 974 (Franzin and Clayton 1977).

975 Beringian Lake Whitefish have a high level of genetic differentiation compared to 976 other assemblages (e.g., eight mtDNA clones alone exist in the Yukon and Chatanika 977 rivers) (Bernatchez and Dodson 1990, McDermid et al. 2005). The discontinuity of 978 haplotypes has also been corroborated by other genetic data, namely allozyme frequencies 979 (Franzin and Clayton 1977, Foote et al. 1992). Overall, significant differences in genetic 980 diversity exist between Beringian Lake Whitefish and the rest of North America 981 (McDermid et al. 2005).

982 Sequence divergence measured among the 46 mtDNA haplotypes comprising the 983 Beringian Lake Whitefish and all other glacial lineages is high (1.15%) and supported by 984 bootstrap values greater than 75% in two separate studies (Bernatchez and Dodson 1991, 985 Bernatchez and Dodson 1994). This is compared to interspecies measures of sequence 986 divergence of 1.8% between C. nasus and C. clupeaformis at the same genetic markers. 987 Thus, there is little doubt that these whitefish populations corresponding to the Beringian 988 glacial lineage represents a DU (Beringian DU) under COSEWIC’s guidelines (Table 1, 989 Figure 2).

990

991 2.2.2 Nahanni Glacial Race: 992

993 Lake Whitefish may have also survived during the ice age in an area of what is 994 currently Nahanni National Park in the Northwest Territories (NWT) (Figure 3). This area 995 was found to be ice-free during the Wisconsinan and Illinoisan glacial periods (Ford 1974). 996 Geological evidence suggests there was a ‘corridor’ between the Mackenzie mountains and 997 the Alaskan slope which could have provided an important refuge for organisms isolated 998 from the Beringia glacial refuge by the Mackenzie mountains (Prest 1970). Foote et al. 999 (1992) examined 43 populations covering all of these western regions and found that a 1000 distinct assemblage, compared to the Beringian and Mississippian glacial races, inhabited 1001 waters in the southwest corner of NWT, central BC, in lakes of the lower Liard, Tetcela, 1002 Fraser, and upper Peace rivers as well as the Talbot River. Although the genetic distance 1003 between the Nahanni and its neigbouring glacial races was low (0.047, presumably 1004 significant although p-values were not estimated, private alleles (genetic variants not found 1005 anywhere but in this region) were present only within this Nahanni group. Importantly, 1006 there were no clines in allele frequencies at these markers but rather breaks in the 1007 distribution, supporting the hypothesis that this group is genetically distinct from the other

25 1008 glacial refugia. Evidence that Arctic grayling (Thymallus arcticus) and lake trout 1009 (Salvelinus namaycush) also appear to have similar patterns of genetic divergence for 1010 populations within the Nahanni further support the hypothesis that this region is distinct 1011 from Beringian populations (Foote 1979, Wilson and Hebert 1998, Stamford and Taylor 1012 2004, McDermid et al. 2005).

1013 Interestingly, not all of the variation is related to isolation of fish in separate refugia, 1014 but possibly reflects population subdivision that occurred within the Beringian refugium 1015 during the ice age. Overall, there is evidence to suggest that populations comprising the 1016 Nahanni groups are likely distinct from other glacial races, but because of continued gene 1017 flow with the Beringian Lake Whitefish and high genetic similarity to the Mississippian 1018 glacial race, this group should not be considered a distinct phylogenetic grouping until more 1019 suitable genetic data can be obtained for resolving these questions. Nevertheless, 1020 populations within the Nahanni grouping should be given provisional DU status under 1021 COSEWIC’s guidelines, following the evidence for diagnostic alleles at allozyme loci from 1022 Foote et al. (1992) (Table 1, Figure 2).

1023

1024 2.2.3 Mississippian Glacial Race: 1025

1026 This was the biggest refugium responsible for most postglacial Lake Whitefish 1027 colonizations. Many species were thought to have survived in the Mississippian River 1028 drainage basin, in the southern, non-glaciated part of the continent (Crossman and 1029 McAllister 1986, Pielou 1991). Large proglacial lakes, such as Lake Agassiz, covered over 1030 350 000 km2, four times larger than Lake Superior, the largest freshwater lake in the world. 1031 Lake Agassiz drained south into the Mississippi River valley until recession of glacial ice 1032 10 700 years ago led to the lake spilling eastward. It is likely that the majority of fishes 1033 living in the interior of Canada today originated from Lake Agassiz (Crossman and 1034 McAllister 1986). This lake was also purported to be linked to and extend as far as Beringia 1035 at times.

1036 Indeed, mtDNA evidence suggests that many Canadian Lake Whitefish populations 1037 must have survived in the Mississippian refuge (Group C, clonal line 1, Bernatchez and 1038 Dodson 1991), diverging approximately 360 000 ya. Postglacial dispersal within this clade 1039 extends as far northwest as the Arctic Red River where Mississippian Lake Whitefish 1040 overlap with the Beringia glacial race (McDermid et al. 2005). This is the only known area 1041 of secondary contact with the Beringia glacial race (Rempel and Smith 1998). In eastern 1042 Canada, the Mississippian glacial race overlaps with the Atlantic glacial lineage (see 1043 1.2.2.13) in several lakes in southern Quebec and Maine. Secondary contact with the

26 1044 Acadian glacial race is limited to a single lake sampled in south-eastern Quebec (Lac 1045 Témiscouata).

1046 The majority of Mississippian populations (over 90%) contain the same mtDNA 1047 haplotype (Bernatchez et al. 1991). Thus, descendants of a single lineage recolonized over 1048 5 000 000 km2, from the Yukon to Labrador. This, the phylogeny of Mississippian Lake 1049 Whitefish is complex. Indeed, the node separating this glacial race from Beringian Lake 1050 Whitefish and other Coregonines has been well supported by both mtDNA bootstrap values 1051 (97%, Bernatchez and Dodson 1991, 1994, Lu et al. 2001) and allozyme data (Bodaly et al. 1052 1992a). Lake Whitefish from the Mississipian glacial race should therefore be considered a 1053 significant DU under COSEWIC’s guidelines. However, within this DU, there are distinct 1054 assemblages exhibiting disjunct, localized distributions (see Atlantic and Acadian glacial 1055 races below) that may warrant additional DU status (Table 1, Figure 2).

1056

1057 2.2.4 Atlantic Glacial Lineage: 1058

1059 Populations in southeastern Quebec and Maine likely survived in an Atlantic 1060 refugium south of the Wisconsinian ice sheet (Schmidt 1986, Underhill 1986, Bernatchez et 1061 al. 1991, Lu et al. 2001, Curry 2007). Bernatchez and Dodson (1991) found that this group 1062 comprised a sub-assemblage of the group C clade largely of the Mississippian glacial race, 1063 also supported by genetic differences in allozyme frequencies (Bodaly et al. 1992a) (Figure 1064 3). Whitefish that survived in the Atlantic glacial refuge share two mtDNA lineages with 1065 anadromous populations from the northern Quebec peninsula and in the St. Lawrence 1066 River. These mtNA haplotypes are absent everywhere else, suggesting that postglacial 1067 colonization in this region may have been from coastal populations that persisted in 1068 brackish water. As mentioned, this is possible if sufficient glacial melting resulted in lower 1069 salinity (de Vernal and Hillaire-Marcel 2000).

1070 The Atlantic glacial race diverged from the Mississippian glacial race only 18- 1071 75000 ya according to the evidence from mtDNA (Bernatchez et al. 1991). Secondary 1072 contact occurs with the Acadian glacial race in three eastern lake populations, one of which 1073 (Lac Témiscouata, Table 1) while the other two are located in Maine (Table 1). The 1074 remaining Lake Whitefish populations of known Atlantic glacial race origin have been 1075 sampled in Maine (e.g., limnetic ecotype of Usaskis Lake and Ross Lake). Further mtDNA 1076 analysis of populations in this region show significant phylogenetic divergence of the 1077 Atlantic from the Acadian or Mississipian glacial races (bootstrap support = 93%, Lu et al. 1078 2001). Furthermore, an analysis of variance of Mississippian, Atlantic, and Acadian 1079 populations in Eastern Canada found that between 65 and 82% of the total genetic variance 1080 among sympatric and allopatric populations could be attributed to genetic differences

27 1081 among the eastern glacial lineages (Lu et al. 2001). In some of these lakes (e.g., Lac 1082 Témiscouata), there is also evidence for reproductive isolation between 1083 Mississippian/Atlantic and Acadian glacial races (Rogers 2001, Lu et al. 2001). Overall, 1084 these results suggest the Atlantic glacial race is a distinct phylogenetic group, indicating 1085 that this assemblage be considered by COSEWIC as a DU (Table 1, Figure 2).

1086

1087 2.2.5 Acadian Glacial Race: 1088

1089 Both mtDNA (Bernatchez and Dodson 1991) and allozyme (Bodaly et al. 1992) 1090 data support evidence for a distinct glacial refuge lineage in eastern North America (Figure 1091 3). This glacial race inhabits lakes in Maine, the Gaspé Peninsula, and New Brunswick and 1092 probably survived glaciation in a northeast banks refugium on the coastal plains of 1093 northeastern North America (Schmidt 1986, Pielou 1991, Curry 2007).

1094 The mtDNA haplotypes (group D in Bernatchez and Dodson 1991, 1994) suggest 1095 that this group diverged from the Mississippian/Atlantic clade approximately 150 000 years 1096 ago. This split from the Mississippian clade is highly significant (bootstrap value = 89%, 1097 Bernatchez and Dodson 1994; bootstrap value = 79%, Lu et al. 2001). Consequently, these 1098 Eastern Canada Acadian glacial race populations represent a distinct DU (Table 1, Figure 1099 2).

1100 2.3. Population Subdivision within Glacial Races: 1101

1102 COSEWIC (2006) recognizes that distinct phylogenetic groupings should be 1103 inferred with genetic markers that are capable resolving population subdivision at an 1104 appropriate scale (e.g., mtDNA, nuclear genes). Population subdivision of Lake Whitefish 1105 has been studied within some glacial races, namely those of Mississippian origin of the 1106 Great Lakes Region (Imhof et al. 1980, Casselman et al. 1981, Ihssen et al. 1981, Stott et al. 1107 2004). Studies exploring allozyme variation (e.g., Imhof et al. 1980, Ihssen et al. 1981, 1108 Casselman et al. 1981) supported the mtDNA evidence that Lake Whitefish from these 1109 regions had a single origin. While microsatellites have further resolved this variation on a 1110 microgeographic scale, there has been no evidence that any of these populations are 1111 reproductively isolated or locally adapted to the extent necessary for defining these 1112 populations as PDUs. This lack of population subdivision includes the Lake Simcoe 1113 population, which was designated distinct stock status by COSEWIC in 1987 (deemed data 1114 deficient, COSEWIC 2005). However, the genetic data of Ihssen et al. 1981 show that the 1115 whitefish of Lake Simcoe are not significantly different from the Lake Huron or Lake 1116 Ontario populations (see COSEWIC 2005a). The Mira River population in Nova Scotia was

28 1117 also deemed data deficient by COSEWIC in 2000 (Goodchild 1999 (unpublished report). 1118 Mira Lake Whitefish fall within the Acadian DU designation based on phylogeny 1119 (Bernatchez and Dodson 1994). The Mira Lake population is considered in the context of 1120 the distinctive and rare traits in later sections of this report (see section 3.4).

1121 2.5. Summary 1122

1123 • Isolation of Lake Whitefish during glacial refugia resulted in significant allopatric 1124 divergence during the Pleistocene ice age 1125

1126 • Genetic evidence (mtDNA and allozymes) from Lake Whitefish inhabiting over 100 1127 populations across Canada strongly supports five major phylogeographic groupings 1128 (Figure 3) 1129

1130 • Lake Whitefish from the Beringia refuge should be granted DU status based in the 1131 distinct phylogenetic grouping and diagnostic haplotypes found for Lake Whitefish 1132 in this region (Table 1, Figure 2) 1133

1134 • The Nahanni warrants provisional DU status until more appropriate genetic markers 1135 can establish the significance of this population subdivision with respect to its split 1136 from the Beringian during the Pleistocene (Table 1, Figure 2) 1137

1138 • Lake Whitefish from the Mississippian refuge should be granted DU status based in 1139 the distinct phylogenetic grouping and diagnostic haplotypes found for Lake 1140 Whitefish in this region (Table 1, Figure 2) 1141

1142 • Lake Whitefish from the Atlantic refuge should be granted DU status based in the 1143 distinct phylogenetic grouping and diagnostic haplotypes found for Lake Whitefish 1144 in this region (Table 1, Figure 2) 1145

1146 • Lake Whitefish from the Acadian glacial race are a distinct phylogenetic group and 1147 warrants DU status (Table 1, Figure 2) 1148

1149 • Population subdivision within the Mississippian (Manitoba and Ontario, the Great 1150 Lakes) and Acadian glacial races (Mira River) reflect recent postglacial population 1151 divergence and significant population subdivision. Although local selection may be 1152 influencing the observed structure, this has not been established. Consequently,

29 1153 these populations should jointly be considered within the Mississipian/Atlantic and 1154 Acadian DUs unless further genetic or ecological evidence should prove otherwise. 1155

1156 3. The PDU has distinctive traits that (1) represent local adaptation and (2) identifies 1157 the PDU as not ecologically interchangeable with other known PDUs within the 1158 species, or (3) identifies the PDU as an irreplaceable component of Canada’s 1159 biodiversity?

1160 3.1. Evidence of distinctive traits that represent local adaptation in the species pairs

1161 3.1.1 Distinctive Traits: 1162

1163 Sympatric species pairs composed of limnetic and benthic forms display differences 1164 in life-history, behavioural, and morphological characters associated with the use of trophic 1165 resources (Fenderson 1964, Bernatchez et al. 1999). In this section the most commonly 1166 studied characters are provided to assess (i) variation in adaptive traits between species 1167 pairs, ii) that the variation underlying adaptive traits is genetically controlled, and (iii) that 1168 the variation in adaptive traits is under the influence of divergent natural selection (thereby 1169 fulfilling the adaptive criterion).

1170 3.1.1.1 Morphology 1171

1172 In the majority of sympatric pair DUs (Lake Opeongo, Como Lake, Témiscouata 1173 Lake, East Lake), a tremendous dichotomy in size between sexually mature fish is the 1174 primary trait distinguishing the species pairs within a lake environment (Figure 4a). 1175 Limnetic Lake Whitefish are rarely over 200 mm in length and 100g in weight. A sexually 1176 mature benthic Lake Whitefish, on the other hand, is almost always greater than 200 mm 1177 (usually bigger than 400 mm) and typically weighs more than 1000g. Fenderson (1964) 1178 reported that limnetic whitefish from Cliff Lake, Maine, ranged from 163 mm (age 1+) to 1179 193 mm (age 5+) while benthic Lake Whitefish varied from 178 mm (age 1+) to 465 mm 1180 (age 12+). This was consistent with additional measurements in Cliff Lake where limnetic 1181 whitefish were on average 168 mm and benthic whitefish were on average 285 mm (Lu and 1182 Bernatchez 1999). Interestingly, although size-at-age was not known in Lu et al. (1999, 1183 those benthic Lake Whitefish were significantly smaller than those reported 30 years earlier 1184 in Fenderson (1964), suggesting that older benthic Lake Whitefish are not as frequent in 1185 this population, are growing slower, or maturing earlier. Size at maturity can differ greatly 1186 among lakes inhabiting the species pairs. In Lake Opeongo, limnetic Lake Whitefish are 1187 smaller than those in Cliff Lake, ranging from 109 mm (age 1+) to 134 mm (age 5+) while 1188 their benthic counterparts exhibit tremendous variation, from 78 mm (age 1+) to 456 mm

30 1189 (age 14+). However, species pairs of Lake Whitefish in Squanga Lake, Yukon, completely 1190 overlap with respect to size (Figure 4b, Lindsey 1963b)

1191 Gill rakers have been found to contribute the most to significant morphological 1192 differences between lakes (Fenderson 1964, Kirkpatrick and Selander 1979, Bernatchez et 1193 al. 1996, Lu et al. 1999). Several studies on the adaptive divergence of northern temperate 1194 particulate feeding fishes have identified gill-raker characteristics as a key trait between 1195 diverging populations (Schluter and McPhail 1993, Robinson and Wilson 1994); Lu and 1196 Bernatchez, 1999, Landry et al. 2007). Gill rakers, the protuberences along the gill arches, 1197 are functional for sieving ingested prey while directing fluid movement within the buccal 1198 cavity; (Sanderson et al. 1991). Thus, gill raker number and spacing is believed to influence 1199 the efficiency of the retention of small prey (Schluter and McPhail, 1993; Robinson and 1200 Wilson, 1994; Budy and Haddix 2005) by potentially regulating fluid dynamics in the 1201 buccal cavity (Sanderson et al. 2001). Landry et al. (2007) discovered that a narrow length 1202 distribution of the planktonic invertebrate community resulted in a higher predation rate by 1203 limnetic whitefish. This was interpreted as evidence for resource limitation in the 1204 zooplanktonic structure, leading potentially to an increase in competition, which could 1205 ultimately result in a higher relative fitness for individuals with a higher efficiency in 1206 planktonic prey retention. To this end, Lindsey (1963b) and Bodaly (1979) revealed that 1207 limnetic ecotypes in southern Yukon possess more gill rakers than their counterpart benthic 1208 ecotypes. For example, limnetics have 27 or more while benthics have 26 or fewer (Lindsey 1209 1963). However, this varies tremendously and appears to be associated with the degree of 1210 trophic resource overlap between the species (Bodaly 2007). In East Lake, for example, 1211 there is really no difference between limnetic and benthic with respect to gill raker counts 1212 (limnetic Lake Whitefish have an average of 25.5 while benthics have an average of 26, Lu 1213 et al. 1999). This finding is consistent with other Lake Whitefish species pairs in Canada 1214 and the United States (Kennedy 1943, Vuorinen et al. 1993, Lu et al. 1999, Bernatchez 1215 2004, Landry et al. 2007).

1216 Kennedy (1943) found differences in the number of scales in the lateral line, a 1217 finding consistent with Lu et al. (1999), but the utility of this trait is unclear. Other 1218 morphological variables rarely discriminate between populations. For example, Lu et al. 1219 (1999) found that of 19 morphometric variables measured in six sympatric populations, 1220 only four variables discriminated between limnetic and benthic populations; adipose fin 1221 length, pectoral fin length, caudal peduncle length, and maxillary width. Further studies 1222 have not found any of these traits to be under divergent selection between limnetics and 1223 benthics in nature (Rogers et al. 2002, Bernatchez 2004), although the latter three traits are 1224 known to be associated with trophic use efficiency in some fishes (Webb 1984). Caudal 1225 peduncle depth to length ratio was also found to differentiate limnetic and benthic in the 1226 Allagash Basin, where differences in blood types have also been documented (Fenderson 1227 1964).

31 1228

1229 3.1.1.2 Behaviour 1230

1231 Adaptive trait differences have been observed for behavioural traits associated with 1232 habitat isolation and predator avoidance. Rogers et al. (2002) studied the genetic basis of 1233 three traits associated with swimming behaviours; depth selection, burst or ‘fast start’ 1234 swimming, and biased directional turns and found these behaviours differed between 1235 limnetic and benthic species pairs raised in the same environment, indicating a strong 1236 genetic basis for these traits. Analysis of the genes underlying these traits in natural 1237 populations found a significant reduction of gene flow at these genes, demonstrating that 1238 natural selection maintains differences in behavioural trait variation between the species 1239 pairs, at least in the environments examined thus far (Rogers and Bernatchez 2007).

1240

1241 3.1.1.3 Physiology and performance 1242

1243 Limnetic whitefish mature as early as one year old and seldom exceed 20 cm in 1244 length and 100 g in weight while benthic whitefish mature at an older age (greater than two 1245 years old) and commonly exceed 40 cm and 1000 g. Benthic benthic individuals typically 1246 reach sexual maturity at age 4 and may have a lifespan of up to 12 years, although Kennedy 1247 reported benthic individuals of 14 years old in Lake Opeongo, Ontario. Limnetic 1248 individuals reach sexual maturity earlier and typically do not live past 5 years (Kennedy 1249 1943, Fenderson 1964).

1250 Significant differences in size at maturity are associated with differences in growth 1251 (Bidgood 1973, Bidgood 1974, Beauchamp et al. 2004). Rogers and Bernatchez (2005) 1252 grew limnetic and benthic species pairs in the same environment and found that the limnetic 1253 group grew slower than the benthic group, leading to significant differences in size as early 1254 as age 1+. These differences in growth are likely due to differences in the bioenergetics of 1255 the species pairs as shown by Trudel et al. (2001). Under natural conditions, limnetic 1256 whitefish consume 40-50% more food than benthic ecotypes, yet their respective 1257 conversion efficiency of these resources is reported to be 2-3 times lower than the benthic 1258 ecotype (Trudel et al. 2001). These results were consistent with observations for lake cisco 1259 (Coregonus artedii, another coregonine species that feed in a limnetic environment, Trudel 1260 et al. 2001). Altogether, these findings demonstrate that inhabiting a limnetic environment 1261 has led to substantial bioenergetic differences in metabolism to the limnetic form. 1262 Differential growth is more difficult to measure in the field than size at maturity but could

32 1263 be an ideal trait to diagnose the species pairs if it was possible to rear specimens in a 1264 controlled environment.

1265

1266 3.1.1.4 Gene expression differences 1267

1268 Genes are the elemental units of adaptation. Genomic methods (e.g., microarrays, 1269 pyrosequencing) can measure levels of gene expression for thousands of genes, highlighting 1270 distinctive genes underlying local adaptation that can differentiate populations. Levels of 1271 differential gene expression have been measured for limnetic and benthic species pairs 1272 collected from two lakes in the Allagash Basin (Indian Lake and Cliff Lake, Maine) 1273 (Derome et al. 2006). Gene expression for over 130 genes in white muscle differed between 1274 limnetics and benthics in both lakes. Sixteen candidate genes (1.35%), belonging to 1275 energetic metabolism and regulation of muscle contraction, showed true parallelism of 1276 expression between the species pairs.

1277

1278 3.1.2 Evidence for local adaptation: 1279

1280 This key considers local adaptation in the strict sense in that variation in the trait is 1281 genetically controlled and influenced by divergent selection in distinct environments 1282 (Taylor 2006). Two methods have been applied to test the hypothesis that natural selection 1283 is maintaining differences between limnetic and benthic Lake Whitefish at the adaptive 1284 traits known to influence their survival in the limnetic and benthic trophic niches in nature.

1285 These methods include testing departures from neutrality at quantitative phenotypes (QST – 1286 FST comparisons, Rogers et al. 2002, Bernatchez 2004) and genome scans (Campbell and 1287 Bernatchez 2004, Rogers and Bernatchez 2005, 2007).

1288

1289 QST – FST

1290

1291 Testing whether traits are under divergent natural selection can be tested by

1292 comparing the extent of differentiation at phenotypic traits (QST) with that of neutral 1293 expectations (quantified at neutral molecular markers, FST) (Spitze 1993). Under the 1294 influence of neutral evolutionary forces (migration, mutation and drift) the among- 1295 population proportion of total genetic variance in phenotypic traits is expected to equal that 1296 of ‘neutrally evolving’ nuclear markers (Lande 1992). The prediction is that divergent

33 1297 selection will cause QST to be larger than that expected from neutral expectations. QST 1298 analyses based on the use of phenotypic variance as a surrogate for additive genetic 1299 variance must be interpreted cautiously. However, estimates derived from phenotypic and

1300 genotypic variance have not differed in their general patterns of FST-QST relationships 1301 (Merila and Crnokrak 2001, Lynch et al. 1999, Schluter 2000, Merilä and Crnokrak 2001, 1302 Bernatchez 2004) suggests that the approach based on phenotypic variance is not strongly 1303 biased.

1304 In Lake Whitefish, upper departures of QST from neutral expectations have been 1305 detected in behaviour, growth, and morphology suggesting these traits are locally adapted 1306 (Rogers et al. 2002, Bernatchez 2004, Rogers and Bernatchez 2005). Depth selection and 1307 gill-raker counts most strongly deviated from neutral expectations compared to other

1308 morphological traits. Bernatchez (2004) estimated QST for 18 morphological characters and 1309 found that gill raker counts were the only character between limnetic and benthic whitefish 1310 to differ significantly from neutral expectations suggesting that it most likely evolved under

1311 directional selective pressures. QST results should be interpreted with caution given that 1312 phenotypic rather than genetic variance was used. Yet, they strongly suggest that 1313 differences in behaviour, growth, and gill raker counts are driven by divergent natural 1314 selection, particularly when parallel departures from neutral expectations occur in multiple 1315 environments.

1316

1317 Genome Scans

1318

1319 A central challenge in adaptive variation is to identify genes underlying ecologically 1320 important traits and describe the fitness consequences of naturally occurring variation at 1321 these loci (Stinchcombe and Hoekstra 2007). The “genome scan” approach partially 1322 circumvent these problems through the simultaneous study of a large number of genetic 1323 markers to better understand the action of evolutionary forces on variation among 1324 populations at specific genetic markers or genes. Such studies hypothesize that the action of 1325 divergent selection should reduce gene flow at genomic regions implicated in adaptation 1326 and speciation. Empirical and simulation data strongly suggest that detecting loci subjected 1327 to directional selection is feasible on the basis of genetic differentiation estimates among 1328 genetic markers (Storz 2005, Campbell and Bernatchez 2004). In these cases, “outlier loci” 1329 exhibit much greater genetic distances than one would expect from neutral evolution alone. 1330 Genome scans must confirm the function and role of these outliers to demonstrate that the 1331 increased molecular divergence is due to selection (Luikart et al. 2003).

1332 Lake Whitefish represent one of the first empirical examples of how genome scans 1333 identified the most likely candidate genetic markers to be under the influence of local

34 1334 adaptation (Campbell and Bernatchez 2004). Rogers and Bernatchez (2005, 2007) were 1335 able to demonstrate that genetic markers associated with adaptive traits such as swimming 1336 behaviour (habitat selection, predator avoidance, growth rate, morphology (condition factor 1337 and gill rakers), and life history (onset of maturity and fecundity) were outliers between 1338 limnetics and benthics in nature, providing strong support for the hypothesis that divergent 1339 natural selection is currently maintaining adaptive differentiation and promoting ecological 1340 speciation in Lake Whitefish species pairs.

1341

1342 3.2. Identification of the limnetic and benthic species pairs as not ecologically 1343 interchangeable with other known PDUs within the species 1344

1345 Recent ecological non-exchangeability may be indicative of adaptive divergence 1346 and population persistence (Crandall et al. 2000). Genetic distinctiveness alone, however, 1347 does not imply that adaptive divergence has occurred. This is the challenge associated with 1348 making the link between ecological differentiation and heritable genetic variation in the 1349 wild (Fraser and Bernatchez 2001). To answer these questions, the origin of the species 1350 pairs and the underlying forces responsible need to be considered. If species pairs have 1351 arisen independently and more than once, and we can associate parallel adaptive traits with 1352 this divergence, then this supports the hypothesis that limnetic and benthic species pairs are 1353 not ecologically interchangeable with other PDUs within the species.

1354 3.2.1 Independent origins of the species pairs 1355

1356 Replicate evolution of the species pairs, a common phenomenon among other north 1357 temperate species, suggests that natural selection is the main evolutionary force driving the 1358 divergence of these forms (Taylor 1999, Robinson and Schluter 2000, Schluter and Nagel 1359 1995, Rogers and Bernatchez 2007). However, multiple modes of speciation are involved 1360 in the divergence. Allopatric divergence of glacial races followed by subsequent secondary 1361 contact and reinforcement has been found in some lakes (e.g., Lac Témiscouta). The best 1362 genetic and ecological evidence for the allopatric scenario has been found for Lake 1363 Whitefish inhabiting Cliff Lake (St. John River Drainage, Maine, USA). Limnetic and 1364 benthic species pairs in this lake are the result of secondary contact between previously 1365 allopatric Atlantic (benthic) and Acadian (limnetic) whitefish glacial races, both of benthic 1366 phenotype when in allopatry, now coinhabiting the lake (Bernatchez and Dodson 1990). 1367 Parallel adaptive radiation and ecological speciation has also led to divergence of the 1368 species pairs in lakes colonized by a single glacial race (e.g., the Acadian glacial race that 1369 colonized East Lake). Thus, both allopatric and sympatric modes of speciation are 1370 independently involved in the evolution of the species pairs (Bernatchez et al. 1996).

35 1371 Differentiation between limnetics and benthics has been shown to be associated 1372 with parallel genetic signatures of selection in natural populations, providing support for the 1373 hypothesis that divergent natural selection is currently maintaining adaptive differentiation 1374 and promoting ecological speciation in some of these Lake Whitefish species pairs, 1375 regardless of the historically contingent factors (Rogers and Bernatchez 2007). This is 1376 further supported by associations between neutral genetic data and the degree of trophic 1377 specialization. Chouinard and Bernatchez (1998) demonstrated a correlation with trophic 1378 niche overlap and gene flow between ecotypes. Lakes inhabiting the limnetic and benthic 1379 species pairs have significantly higher gene flow in the presence of overlapping trophic 1380 niches when compared to lakes with less gene flow. Traits that differentiate the two 1381 ecotypes are directly related to the specialization and availability of trophic niches, leading 1382 to a negative correlation between the extent of gene flow and morphological specialization 1383 between species pairs among lakes (Chouinard and Bernatchez 1998, Chouinard et al. 1996, 1384 Lu and Bernatchez 1999)

1385 Limnetic Lake Whitefish are only found in sympatry with the benthic ecotype and 1386 only in the absence of other limnetic coregonine fishes such as the cisco (Coregonus 1387 artedii), suggesting that divergent natural selection resulting from ecological pressures 1388 within the benthic niche may explain the evolution of the derived limnetic whitefish 1389 (Pigeon et al. 1997). This has been supported by evidence for convergence in the 1390 expression of genes associated with the limnetic life history (metabolism and swimming 1391 efficiency) among coregonine species. Derome and Bernatchez (2006) tested whether the 1392 same genes associated with adaptive divergence in limnetic Lake Whitefish were also 1393 implicated C. artedii. They found that both limnetic Lake Whitefish and cisco were over- 1394 expressing the same candidate genes modulating swimming activity. Natural selection in 1395 the limnetic niche appears to act on the same genetic variation in both species. However, 1396 even greater upregulation in cisco suggested that this species had a greater physiological 1397 potential in the limnetic niche, potentially explaining why the addition of a cisco to lakes 1398 inhabiting the limnetic species would likely result in its extinction (Derome and Bernatchez 1399 2006).

1400 Overall, there is substantial evidence that ecological processes are a factor for the 1401 extent of reproductive isolation and contribute to the sustained differentiation between the 1402 ecotypes (Bernatchez et al. 1996, Pigeon et al. 1997, Lu et al. 1999, Rogers and Bernatchez 1403 2007, Landry et al. 2007). Moreover, it has been demonstrated that divergent selection 1404 reduces gene flow at ecologically relevant genes associated with adaptive traits such as 1405 growth in multiple limnetic and benthic species pairs (Rogers and Bernatchez 2005). 1406 Adaptive traits exhibit much less gene exchange than other regions of the genome, 1407 supporting the hypothesis that divergent selection maintains adaptive differentiation despite 1408 the effects of gene flow (Rogers and Bernatchez 2007). Although there is no single trait 1409 distinguishes the PDU, these genetic data provide good evidence that a particular suite of

36 1410 adaptations make it very unlikely that the PDU could be replaced by recolonization or 1411 deliberate introductions from another population, consistent with the criteria set forth by 1412 Taylor (2006). However, despite the fact that similar ecological processes are involved in 1413 the parallel evolution of the species pairs, it remains difficult to establish the degree of 1414 ecological interchangeability with other PDUs within the species.

1415

1416 3.2.2 Evidence of reproductive isolation 1417

1418 Controlled crosses between limnetic and benthic species pairs have demonstrated 1419 significant intrinsic and extrinsic postzygotic isolation (Lu 1998, Rogers and Bernatchez 1420 2006). Hybrid inviability under experimental conditions is up to five time higher in hybrids 1421 than what is observed in the pure crosses. However, this experimental design in both of 1422 these studies employed crosses originating from different glacial races which may have had 1423 a significant impact in the degree of reproductive isolation. More recent studies have 1424 isolated the genetic factors associated with this decrease in fitness and found that progeny 1425 from hybrid backcross families either died during development or hatched at a sub-optimal 1426 time, suggesting that both genetic incompatibilities and extrinsic postzygotic isolation 1427 contribute to reproductive barriers (Rogers and Bernatchez 2006). This is consistent with 1428 observations that hybrids between diverging populations are rare (1-3%) within their 1429 natural habitat (Bodaly 1979b). More studies on the degree of reproductive isolation 1430 between glacial races and species pairs are clearly needed.

1431

1432 3.3. Are limnetic and benthic species pairs an irreplaceable components of Canadian 1433 biodiversity? 1434

1435 Lake Whitefish were among the first fish species to colonize the Canadian 1436 landscape. Less than 10 000 years later they now comprise some of the youngest biological 1437 species on earth (Bernatchez et al. 1999). The evolution of these species pairs reflects rapid 1438 adaptation following colonization of a novel niche not inhabited by the ancestral form 1439 within postglacial environments (Rogers and Bernatchez 2007). The derived Eurasian 1440 limnetic form is also an important prey resource for larger salmonid predators and in some 1441 cases is their main source of food (Kahilainen et al. 2002, 2003). Predation of limnetic 1442 Lake Whitefish from large salmonid predators such as lake trout is common (Dave Basely, 1443 Maine Fish and Wildlife, personal communication), but this has not been resolved in detail.

1444 Unique scenarios of population divergence between some species pairs (e.g., 1445 Squanga Lake Whitefish, Bodaly 2007) suggest that other historically contingent or

37 1446 ecological factors may affect population divergence in certain northern temperate limnetic 1447 and benthic environments. Namely, the role of historical contingency, due in part to 1448 isolation in glacial refugia, has also been shown to play a significant role in the evolution of 1449 reproductive isolation in many species pairs (Lu et al. 2001, Rogers et al. 2001). Historical 1450 contingency refers to the original assemblage, locally specific and unique selection 1451 pressures, and locally unique stochastic influences (e.g., founder effects, drift, extinctions), 1452 that may have influenced the evolutionary history of a lineage. In cases where inviability is 1453 associated with divergence in allopatry rather than sympatry, historical contingency may 1454 have primed subsequent ecological determinism (Taylor and McPhail 2000, Fraser, 2005, 1455 Rogers, 2001, Rogers and Bernatchez 2006). This is important to acknowledge given the 1456 mandate of COSEWIC to maximize the probability of protecting biological variation within 1457 a species. This requires considering both historical and ecological evolutionary forces that 1458 have given rise to isolated lineages (Fraser and Bernatchez 2001). Because evolutionary 1459 divergence in limnetic and benthic Lake Whitefish will be determined by the conditions of 1460 the lake (Bernatchez 2004), standing genetic variation of the colonizers (Rogers and 1461 Bernatchez 2005, Barrett and Schluter 2008), along with historically contingent factors that 1462 promote their differentiation (Lu et al. 2001, Rogers et al. 2001, Rogers and Bernatchez 1463 2006), the limnetic ecotype and the conditions that led to its divergence may differ from 1464 one environment to the next. Given the data, it would likely be unwise to assume that if one 1465 limnetic population was wiped out it could be created anew. This has been reinforced in 1466 similar species pairs such as threespine stickleback (Gasterosteus aculeatus), whereby 1467 alterations of natural habitat, such as an invasive species, resulted in the so-called 1468 “speciation in reverse” or collapse of a species pair (Taylor et al. 2006). Indeed, the 1469 introduction of ciscoes in Squanga whitefish lakes are suspected to have resulted in the 1470 extinction of the limnetic whitefish. Similarly, a whitefish population could be significantly 1471 reduced by the addition of a piscivorous fish to its lake (SARA 2006, both factors leading 1472 COSEWIC to list the Squanga Lake Whitefish as a population of ‘special concern’. Overall, 1473 limnetic and benthic Lake Whitefish species pairs embody the process of rapid evolutionary 1474 change but their extinction in a significant number of environments is a stark reminder that 1475 conservation efforts may ultimately be necessary for their persistence.

1476

1477 3.3. Consideration of the Mira River population 1478

1479 This small population of Lake Whitefish in Nova Scotia is found in a restricted 1480 range of the Mira and Salmon rivers (less that 100 km2) and have low lateral line scale and 1481 gill raker counts (Edge et al. 1991). The Mira River population is listed by COSEWIC as 1482 data deficient (Goodchild, 1999). Consequently, there is insufficient information to assess 1483 ecological interchangability of the Mira River with other populations.

38 1484

1485 3.5. Summary 1486

1487 • There is no single phenotypic trait that can be used to identify the species pairs but 1488 instead a suite of phenotype-environment associations that have been empirically 1489 tested in the lab and in nature among certain lakes inhabiting the species pairs. 1490

1491 • In lakes where the species pairs are purported to exist but phenotype-environment 1492 associations have not been assessed, or genetic differentiation between limnetic and 1493 benthic is unknown, DU status should be deferred until these data can be provided. 1494

1495 • Although ecological processes are associated with adaptive divergence, historically 1496 contingent reductions of introgression may differ among glacial races and play 1497 distinct roles during evolution in sympatry. 1498

1499 4. The PDU represents a major range disjunction? 1500

1501 Taylor (2006) defined major range disjunctions as two or more groups of 1502 populations separated widely by naturally unoccupied areas. Generally, this would include 1503 an area across which natural dispersal is not observed nor expected. Disjunctions may 1504 therefore reflect evolutionarily significant events caused by major geological transitions 1505 that now separate the species (e.g., ice sheets, change in sea level, and formation of 1506 mountains).

1507 Lake Whitefish are the most broadly distributed freshwater fish in Canada (Figure 1508 3). Lindsey (1970) noted that at least three sympatric pairs of whitefish species pairs 1509 exhibited range disjunction (east, west, and central Canada) as evidence that these have 1510 evolved independently. This was also considered as some of the first evidence that these 1511 populations survived in separate glacial refugia. Pigeon et al. (1997) supported these 1512 disjunctions between species pairs and proposed that five regions; St. John River Drainage, 1513 northern Quebec and Labrador, Ontario, Yukon, and British Columbia were disjunct and 1514 therefore Lake Whitefish species pairs must have evolved independently.

1515 4.1. Summary 1516

1517 Based on section 2.1, the disjunct regions proposed by Lindsey (1970) and Pigeon et 1518 al. (1997) also reflect evolutionarily significant sunderings caused by the glaciations. 1519 Arguably, these disjunctions do not contribute to the existing criteria under which Lake

39 1520 Whitefish glacial races and species pairs are considered as PDUs, and are already depicted 1521 in the major phylogeographic groupings. Consequently, major range disjunctions alone 1522 need not be identified as PDUs. Given the limitations of applying disjunction or population 1523 separation to lacustrine species, and the purpose of this report as testing DU criteria in a 1524 representative species, there is an obvious caveat regarding geographic separation as it 1525 applies to habitat-restricted species. Although the criteria regarding disjunctions is 1526 valuable; it is worth noting that for obligatory lacustrine species historical water 1527 connections such as glacial meltwater lakes may be more relevant than contemporary 1528 watershed boundaries

1529

1530 5. The PDU inhabits a different biogeographic zones? 1531

1532 National Freshwater Ecological Areas represent different eco-geographic regions 1533 within Canada (Figure 5). These regions may be relevant to Lake Whitefish if they depict 1534 the previously described glacial history within these phylogeographic zones. Fourteen 1535 biogeographic zones have been defined in Canada (Mandrak 2003). This scheme is meant 1536 to capture the major divisions within named taxa (i.e., the phylogeographic groupings of 1537 Lake Whitefish). Indeed, the Lake Whitefish species complex exists in eleven of these 1538 biogeographic zones with the exception of the Pacific Islands (Lake Whitefish are only 1539 indigenous to mainland BC), the Arctic Archipelago, and the Atlantic Islands (although 1540 Lake Whitefish were introduced into some lakes in Newfoundland, Scott and Crossman 1541 (1964)) (Figure 5).

1542 The Mississippian Glacial Race is captured by eight ecoregions, the largest number 1543 of all the putative phylogeographic DUs, while the Atlantic PDU consists of only one 1544 ecoregion, Lac Témiscouata in the Maritime NFEA (Table 1, Figure 2, Figure 5). Over 1545 thirty populations from the Mississippian, Beringian, and Nahanni groups are represented in 1546 the Western Arctic NFEA, the largest number of Lake Whitefish populations within a 1547 single NFEA. At the other end of the range, only one Mississippian Lake Whitefish 1548 population is found in the Missouri ecoregion (Table 1). However, these samples do not 1549 represent the entire distribution of the species within these regions and Lake Whitefish 1550 likely exist in other lakes and rivers within these areas.

1551

1552 5. 1 Summary 1553

1554 NFEA are identified as significant DUs that capture the significant major 1555 phylogeographic groupings of the species complex and the major range disjunctions

40 1556 previously described. The eleven NFEAs designated as DU for the Lake Whitefish species 1557 complex are: Maritimes, Eastern Arctic, Southern Hudson Bay-James Bay, Saskatchewan- 1558 Nelson, Western Hudson Bay, Yukon, Missouri, Eastern St. Lawrence, Great Lakes- 1559 Western St. Lawrence, Pacific and Western Arctic (Table 1, Figure 2).

1560

1561 6. Concluding Summary 1562

1563 • The taxonomic status, biology, and life history of the Lake Whitefish has been a 1564 source of taxonomic confusion for over one hundred years. Lake Whitefish have 1565 been described as several different species due to the tremendous variation it 1566 exhibits across its range. One species included in the Lake Whitefish species 1567 complex, Coregonus pidschian, has a genetic signature almost identical to the 1568 Eurasian whitefish, Coregonus lavaretus. Consequently, this species should be 1569 considered as a separate DU within the Lake Whitefish species complex. 1570

1571 • Ecological opportunity and divergent natural selection within postglacial lakes has 1572 led to the repeated evolution of a derived species that inhabits the limnetic zone of 1573 lakes and does not randomly interbreed with their benthic counterparts, an 1574 evolutionary pattern repeated regardless of allopatric or sympatric origin. Ecological 1575 speciation of seven Canadian Lake Whitefish populations meet the guidelines for 1576 designation of DUs (Figure 2) 1577

1578 • Lake Whitefish populations across Canada were significantly impacted by the 1579 Pleistocene glaciations. The phylogeography of the species across Canada 1580 demonstrates how thousands of years of separation within glacial refugia located in 1581 different areas of the country have led to contemporary reproductive isolation 1582 between populations that can be detected genetically. Four of these Canadian Lake 1583 Whitefish glacial races separated by major phylogeographic groupings caused by 1584 isolation during the Pleistocene ice age meet the guidelines for DU while one 1585 (Nahanni, Figure 2) should be granted provisional DU status. 1586 1587 • Adoption of NFEA DUs that encompass both the major phylogeographic groupings 1588 and the species pairs may facilitate conservation scenarios while addressing regional 1589 situations towards ultimately adopting appropriate DUs for use in subsequent status 1590 assessments for Lake Whitefish as a whole 1591 1592 • Conserving these differences among species pairs and glacial races as designatable 1593 units, fundamental to the biodiversity of the species, will help ensure that the 1594 evolutionary legacy of this species complex is protected in Canada. 1595 1596

41 1597 Acknowledgements 1598

1599 S.M. Rogers would like to thank E.B. Taylor and members of the Freshwater Fishes 1600 Specialist Subcommittee, as well as P. Tamkee, for comments that greatly improved the 1601 quality of this special report. We would also like to thank the COSEWIC Secretariat. 1602 Funding for the preparation of this special report was provided by Environment Canada

1603

1604

1605

1606

1607

1608

1609

1610

1611

1612

1613

1614

1615

1616

1617

1618

1619

1620

1621

1622

42 1623 Literature Cited 1624

1625 Alt, K. 1971. Occurrence of hybrids between inconnu, Stenodus leucichthys nelma (Pallas), 1626 and humpback whitefish, Coregonus pidschian (Linnaeus) in Chatanika river, Alaska. 1627 Transactions of the American Fisheries Society 100: 362 – 365.

1628

1629 Alt, K.T. 1979. Contributions to the life history of the humpback whitefish in Alaska. 1630 Transactions of the American Fisheries Society 108: 156 – 160.

1631

1632 Barrett, R.D.H., and Schluter, D. 2008. Adaptation from standing genetic variation. Trends 1633 in Ecology and Evolution 23: 38 – 44.

1634

1635 Bean, T.H. 1884. Description of a new species of whitefish (Coregonus nelsonii) from 1636 Alaska. Proceedings of the U.S. National Museum. Case Reference number 16047.

1637

1638 Beauchamp, K.C., Collins, N.C., and Henderson, B.A. 2004. Covariation of growth and 1639 maturation of Lake Whitefish (Coregonus clupeaformis). Journal of Great Lakes Research 1640 30: 451 – 460.

1641

1642 Beaumont, A.R., Bray, J., Murphy, J.M., and Winfield, I.J. 1995. Genetics of whitefish and 1643 vendace in England and Wales. Journal of Fish Biology 46: 880 – 890.

1644

1645 Behnke, R.J. 1972. Systematics of Salmonid Fishes of Recently Glaciated Lakes. Journal of 1646 the Fisheries Research Board of Canada 29: 639

1647 .

1648 Bernatchez, L. 2004. Ecological Theory of Adaptive Radiation: An empirical assessment 1649 from Corigonine fishes (Salmoniformes). In Evolution Illuminated: Salmon and Their 1650 Relatives. Edited by A.P. Hendry and S. Stearns. Oxford University Press, Oxford, U.K. pp. 1651 176 – 207.

1652

43 1653 Bernatchez, L. 2005. On the role of natural selection in promoting population divergence in 1654 Lake Whitefish (Coregonus clupeaformis): relevance for population management in the 1655 Great Lakes. In Proceedings of a workshop on the dynamics of Lake Whitefish (Coregonus 1656 clupeaformis) and the amphipod Diporeia spp. in the Great Lakes. Ann Arbor, MI. Edited 1657 by Great Lakes Fisheries Commission.

1658

1659 Bernatchez, L., Chouinard, A., and Lu, G. 1999. Integrating molecular genetics and ecology 1660 in studies of adaptive radiation: whitefish, Coregonus sp., as a case study. Biological 1661 Journal Of The Linnean Society 68: 173 – 194.

1662

1663 Bernatchez, L., Colombani, F., and Dodson, J.J. 1991a. Phylogenetic-Relationships among 1664 the Subfamily Coregoninae as Revealed by Mitochondrial-DNA Restriction Analysis. 1665 Journal of Fish Biology 39: 283 – 290.

1666

1667 Bernatchez, L., and Dodson, J.J. 1990. Allopatric Origin of Sympatric Populations of Lake 1668 Whitefish (Coregonus clupeaformis) as Revealed by Mitochondrial-DNA Restriction 1669 Analysis. Evolution 44: 1263 – 1271.

1670

1671 Bernatchez, L., and Dodson, J.J. 1991. Phylogeographic structure in mitochondrial-DNA of 1672 the Lake Whitefish (Coregonus clupeaformis) and its relation to Pleistocene glaciations. 1673 Evolution 45: 1016 – 1035.

1674

1675 Bernatchez, L., and Dodson, J.J. 1994. Phylogenetic relationships among Palearctic and 1676 Nearctic whitefish (Coregonus sp.) populations as revealed by mitochondrial DNA 1677 variation. 51: 240 – 251.

1678

1679 Bernatchez, L., Dodson, J.J., and Boivin, S. 1989. Population bottlenecks: Influence on 1680 mitochondrial DNA diversity and its effect in coregonine stock discrimination. Journal of 1681 Fish Biology. 35: 233 – 244.

1682

44 1683 Bernatchez, L., Edge, T.A., Dodson, J.J., and Qadri, S.U. 1991b. Mitochondrial-DNA and 1684 Isozyme Electrophoretic Analyses of the Endangered Acadian Whitefish, Coregonus- 1685 Huntsmani Scott, 1987. Canadian Journal of Zoology 69: 311 – 316.

1686

1687 Bernatchez, L., Vuorinen, J.A., Bodaly, R.A., and Dodson, J.J. 1996. Genetic evidence for 1688 reproductive isolation and multiple origins of sympatric trophic ecotypes of whitefish 1689 (Coregonus clupeaformis). Evolution 50: 624 – 635.

1690

1691 Bernatchez, L., and Wilson, C. 1998. Comparative phylogeography of nearctic and 1692 palearctic fishes. Molecular Ecology 7: 431 – 452.

1693

1694 Bidgood, B.F. 1973. Divergent Growth in two Lake Whitefish (Coregonus clupeaformis) 1695 Populations. Journal of the Fisheries Research Board of Canada 30: 1683 – 1696.

1696

1697 Bidgood, B.F. 1974. Reproductive Potential of 2 Lake Whitefish (Coregonus clupeaformis) 1698 Populations. Journal of the Fisheries Research Board of Canada 31: 1631 – 1639.

1699

1700 Bodaly, R.A. 1977. Evolutionary divergence between currently sympatric Lake Whitefish, 1701 Coregonus clupeaformis. Ph.D. thesis, University of Manitoba, Winnipeg.

1702

1703 Bodaly, R.A. 1979. Morphological and Ecological Divergence within the Lake Whitefish 1704 (Coregonus clupeaformis) Species Complex in Yukon Territory. Journal of the Fisheries 1705 Research Board of Canada 36: 1214 – 1222.

1706

1707 Bodaly, R.A. 1986. Biology, exploitation and culture of Coregonine fishes in Canada. Arch. 1708 Hydrobiol. Beih., Ergebn. Limnol. 22: 1 – 30.

1709

1710 Bodaly, R.A., Vuorinen, J., Reshetnikov, Y.S., and Reist, J.D. 1994. Genetic relationships 1711 of five species of Coregonine fishes from Siberia [translated from Russian]. Journal of 1712 Ichthyology 34: 117 – 130.

45 1713

1714 Bodaly, R.A. 2007. COSEWIC Status Report on Squanga whitefish Coregonus sp. Edited 1715 by Freshwater Fisheries Specialist Subcommittee. Available on request from the COSEWIC 1716 Secretariat (COSEWIC/[email protected]).

1717

1718 Bodaly, R.A., Clayton, J.W., and Lindsey, C.C. 1987. COSEWIC status report on the 1719 Squanga whitefish Coregonus sp. in Canada. Edited by Freshwater Fisheries Specialist 1720 Subcommittee. Available on request from the COSEWIC Secretariat 1721 (COSEWIC/[email protected]). p. iv + 32 pp.

1722

1723 Bodaly, R.A., Clayton, J.W., and Lindsey, C.C. 1988. Status of the Squanga whitefish, 1724 Coregonus sp., in the Yukon Territory, Canada. Canadian Field Naturalist 102: 114 – 125.

1725

1726 Bodaly, R.A., Clayton, J.W., Lindsey, C.C., and Vuorinen, J. 1992. Evolution of Lake 1727 Whitefish (Coregonus clupeaformis) in North-America During the Pleistocene – Genetic 1728 Differentiation between Sympatric Populations. 49: 769 – 779.

1729

1730 Bodaly, R.A., Vuorinen, J., and Macins, V. 1991. Sympatric presence of limnetic and 1731 benthic forms of the Lake Whitefish, Coregonus clupeaformis, in Como Lake, Ontario. 1732 105: 87 – 90.

1733

1734 Bodaly, R.A., Vuorinen, J., Reshetnikov, Y.S., and Reist, J.D. 1994. Genetic relationships 1735 of five species of Coregonine fishes from Siberia [translated from Russian]. Journal of 1736 Ichthyology 34: 117 – 130.

1737

1738 Bodaly, R.A., Vuorinen, J., Ward, R.D., Luczynski, M., and Reist, J.D. 1991b. Genetic 1739 comparisons of new and old world Coregonine fishes. Journal of Fish Biology 38: 37-51.

1740

1741 Brown, R.W., and Taylor, W.W. 1992. Effects of Egg Composition and Prey Density on the 1742 Larval Growth and Survival of Lake Whitefish (Coregonus clupeaformis Mitchill). Journal 1743 of Fish Biology 40: 381 – 394.

46 1744

1745 Budy, P., and Haddix, T. 2005. Zooplankton Size Selection Relative to Gill Raker Spacing 1746 in Rainbow Trout. Transactions of the American Fisheries Society 134: 1228 – 1235.

1747

1748 Campbell, D., and Bernatchez, L. 2004. Genomic scan using AFLP markers as a means to 1749 assess the role of directional selection in the divergence of sympatric whitefish ecotypes. 1750 Molecular Biology and Evolution 21: 945 – 956.

1751

1752 Campbell, D., Duchesne, P., and Bernatchez, L. 2003. AFLP utility for population 1753 assignment studies: analytical investigation and empirical comparison with microsatellites. 1754 Molecular Ecology 12: 1979 – 1991.

1755

1756 Casselman, J.M., Collins, J.J., Crossman, E.J., Ihssen, P.E., and Spangler, G.R. 1981. Lake 1757 Whitefish (Coregonus clupeaformis) Stocks of the Ontario Waters of Lake Huron. 38: 1772 1758 – 1789.

1759

1760 Casselman, S.J., and A.I. Schulte-Hostedde. 2004. Reproductive roles predict sexual 1761 dimorphism in internal and external morphology of Lake Whitefish (Coregonus 1762 clupeaformis). Ecology of Freshwater Fish 13: 217 – 222.

1763

1764 Champigneulle, A., and Rojas-Beltran, R. 1990. First attempts to optimize the mass rearing 1765 of whitefish (Coregonus lavaretus Linnaeus) larvae from Lémen and Bourget lakes 1766 (France) in tanks and cages. Aquatic Living Resources 3: 217 – 228.

1767

1768 Chouinard, A., and Bernatchez, L. 1998. A study of trophic niche partitioning between 1769 larval populations of reproductively isolated whitefish (Coregonus sp.) ecotypes. Journal of 1770 Fish Biology 53: 1231 – 1242.

1771

1772 Chouinard, A., Pigeon, D., and Bernatchez, L. 1996. Lack of specialization in trophic 1773 morphology between genetically differentiated limnetic and benthic forms of Lake

47 1774 Whitefish (Coregonus clupeaformis Mitchill) in Lac de l'Est, Québec. Canadian Journal of 1775 Zoology 74: 1989 – 1998.

1776

1777 Conway, K.W., Barrie, J.V., and Hebda, R.J. 2001. Evidence for a Late Quaternary outburst 1778 flood event in the Georgia Basin, British Columbia Edited by Geological Society of 1779 Canada.

1780

1781 COSEWIC. 2005a. COSEWIC assessment and update status report on the Lake Whitefish 1782 (Lake Simcoe population) Coregonus clupeaformis in Canada. Edited by Committee on the 1783 Status of Endangered Wildlife in Canada, Ottawa. p. vii + 36. Available on request from the 1784 COSEWIC Secretariat (COSEWIC/[email protected])

1785

1786 COSEWIC. 2005b. Guidelines for recognizing designatable units below the species level. 1787 Edited by Committee on the Status of Endangered Wildlife in Canada, Ottawa. Available 1788 on request from the COSEWIC Secretariat (COSEWIC/[email protected])

1789

1790 Crandall, K., Bininda-Emonds, O., Mace, G., and Wayne, R. 2000. Considering 1791 evolutionary processes in conservation biology. Trends in Ecology and Evolution 15: 290 – 1792 295.

1793

1794 Crossman, E.J., and D.E. McAllister, 1986. Zoogeography of freshwater fishes of the 1795 Hudson Bay drainage, Ungava Bay and the Arctic Archipelago, In The zoogeography of 1796 North American freshwater fishes, C.H. Hocutt and E.O. Wiley (editors.), John Wiley and 1797 Sons, New York, p. 53 – 104.

1798 Cucin, D., and Faber, D.J. 1985. Early life studies of Lake Whitefish (Coregonus 1799 clupeaformis), cisco (Coregonus artedii) and yellow perch (Perca flavescens) in lake 1800 Opeongo, Ontario. Ontario fisheries technical report series no. 16: iii-28.

1801

1802 Curry, R.A. 2007. Late glacial impacts on dispersal and colonization of Atlantic Canada 1803 and Maine by freshwater fishes. Quaternary Research 67: 225 – 233.

1804

48 1805 Davis, B.M., and Todd, T.N. 1998. Competition between larval lake herring (Coregonus 1806 artedi) and Lake Whitefish (Coregonus clupeaformis) for zooplankton. 55: 1140 – 1148.

1807

1808 Dawson, A.G. 2002. Ice Age Earth. Routledge Press, London.

1809

1810 de Vernal, A., and Hillaire-Marcel, C. 2000. Sea-ice cover, sea-surface salinity and 1811 halo/thermocline structure of the northwest North Atlantic: Modern versus full glacial 1812 conditions. Quaternary Science Reviews 19: 65 – 85.

1813

1814 Denton, G.H., and Hughes, T.J. 1981. The last great ice sheets. John Wiley and Sons, New 1815 York.

1816

1817 Derome, N., and Bernatchez, L. 2006. The transcriptomics of ecological convergence 1818 between 2 limnetic coregonine fishes (Salmonidae). Molecular Biology and Evolution 23: 1819 2370 – 2378.

1820

1821 Derome, N., Duchesne, P., and Bernatchez, L. 2006. Parallelism in gene transcription 1822 among sympatric Lake Whitefish (Coregonus clupeaformis Mitchill) ecotypes. Molecular 1823 Ecology 15: 1239 – 1249.

1824

1825 Department of Fisheries and Oceans, Canada. 2005. Ontario-Great Lakes Area Fish Fact 1826 Sheets. Available from http://www.dfo-mpo.gc.ca/regions/central/pub/fact-fait-ogla- 1827 rglo/lakewhitefish_e.htm.

1828

1829 Douglas, M.R., Brunner, P.C., and Douglas, M.E. 2005. Evolutionary homoplasy among 1830 species flocks of central alpine Coregonus (Teleostei : Salmoniformes). Copeia 2: 347 – 1831 358.

1832

1833 Drouin, M.A., Kidd, R.B., and Hynes, J.D. 1986. Intensive Culture of Lake Whitefish 1834 (Coregonus clupeaformis Mitchill) Using Artemia and Artificial Feed. Aquaculture 59: 107 1835 – 118.

49 1836

1837 Eckmann, R., Appenzeller, A., and Rösch, R. 1996. Biology and Management of 1838 Coregonine Fishes – Proceedings of the Sixth International Symposium on the Biology and 1839 Management of Coregonine Fishes, Konstanz, Germany.

1840

1841 Edge, T.A., McAllister, D.E., and Qadri, S.U. 1991. Meristic and Morphometric Variation 1842 between the Endangered Acadian Whitefish, Coregonus huntsmani, and the Lake 1843 Whitefish, Coregonus clupeaformis, in the Canadian Maritime Provinces and the State of 1844 Maine, USA. 48: 2140 – 2151.

1845

1846 Edsal, T.A. 1960. Age and growth of the whitefish, Coregonus clupeaformis, of the 1847 Munising Bay, Lake Superior. Transactions of the American Fisheries Society 89: 323 – 1848 332.

1849

1850 Fenderson. 1964. Evidence of subpopulations of Lake Whitefish, Coregonus clupeaformis, 1851 involving a dwarfed form. Transactions of the American Fisheries Society 93: 77 – 94.

1852

1853 Flüchter, J. 1980. Review of the present knowledge of rearing whitefish (Coregonineae) 1854 larvae. Aquaculture 19: 191 – 208.

1855

1856 Foote, C.J. 1979. A biochemical genetic study of zoogeography of Lake Whitefish, 1857 Coregonus clupeaformis, in western Canada in relation to their possible survival in a 1858 Nahanni glacial refugium, University of Manitoba, Winnipeg.

1859

1860 Foote, C.J., Clayton, J.W., Lindsey, C.C., and Bodaly, R.A. 1992. Evolution of Lake 1861 Whitefish (Coregonus clupeaformis) in North-America During the Pleistocene – Evidence 1862 for a Nahanni Glacial Refuge Race in the Northern Cordillera Region. 49: 760 – 768.

1863

1864 Ford, D.C. 1974. Evidences of multiple glaciation in South Nahanni National Park, 1865 Mackenzie Mountains, North West Territories, Canada. Canadian Journal of Earth Science 1866 13: 1433 – 1445.

50 1867

1868 Fortin, R., and Gendron, M. 1990. Comparative Reproduction, Growth and Morphology of 1869 Limnetic and benthic Whitefish [Coregonus clupeaformis] in the Outardes-2 Reservoir 1870 (Quebec). Canadian Journal of Zoology 68: 17 – 25.

1871

1872 Franzin, W.G., and Clayton, J.W. 1977. Biochemical Genetic Study of Zoogeography of 1873 Lake Whitefish (Coregonus clupeaformis) in Western Canada. Journal of the Fisheries 1874 Research Board of Canada 34: 617 – 625.

1875

1876 Fraser, D.J., and Bernatchez, L. 2001. Adaptive evolutionary conservation: towards a 1877 unified concept for defining conservation units. Molecular Ecology 10: 2741 – 2752.

1878

1879 Fraser, D.J., and Bernatchez, L. 2005. Allopatric origins of sympatric brook charr 1880 populations: colonization history and admixture. Molecular Ecology 14: 1497 – 1509.

1881

1882 Fraser, D.J., Hansen, M.M., Ostergaard, S., Tessier, N., Legault, M., and Bernatchez, L. 1883 2007. Comparative estimation of effective population sizes and temporal gene flow in two 1884 contrasting population systems. Molecular Ecology 16: 3866 – 3889.

1885

1886 Gmelin, J.F. 1789. Systema Naturae. Ed. 13. (3). Amphibia et Pisces. Leipzig.

1887

1888 Goodchild, C.D. 1999. COSEWIC status report on the Lake Whitefish Coregonus 1889 clupeaformis Mira River population in Canada. Edited by Committee on the Status of 1890 Endangered Wildlife in Canada. p. 29

1891

1892 Hamada, M., Kido, Y., Himberg, M., Reist, J.D., Ying, C., Hasegawa, M., and Okada, N. 1893 1997. A newly isolated family of short interspersed repetitive elements (SINEs) in 1894 Coregonine fishes (whitefish) with sequences that are almost identical to those of the SmaI 1895 family of repeats: Possible evidence for the horizontal transfer of SINEs. Genetics 146: 355 1896 – 367.

1897

51 1898 Harris, K.C., and Hulsman, P.F. 1991. Intensive Culture of Lake Whitefish (Coregonus 1899 clupeaformis) from Larvae to Yearling Size Using Dry Feeds. Aquaculture 96: 255 – 268.

1900

1901 Hasselman, D.J. 2003. Discrimination of adult and early life history stage Coregonine 1902 fishes in Maritime Canada, Acadia University, Wolfville, Nova Scotia, Canada.

1903

1904 Heuts, B. 1999. Lateralization of trunk muscle volume, and lateralization of swimming 1905 turns of fish responding to external stimuli. Behavioral Process 47: 113 – 124.

1906

1907 Hewitt, G.M. 1996. Some genetic consequences of ice ages, and their role in divergence 1908 and speciation. Biological Journal Of The Linnean Society 58: 247 – 276.

1909

1910 Ihssen, P.E., Evans, D.O., Christie, W.J., Reckahn, J.A., and Desjardine, R.L. 1981. Life- 1911 History, Morphology, and Electrophoretic Characteristics of 5 Allopatric Stocks of Lake 1912 Whitefish (Coregonus clupeaformis) in the Great-Lakes Region. 38: 1790 – 1807.

1913

1914 Imhof, M., Leary, R., and Booke, H.E. 1980. Population or Stock Structure of Lake 1915 Whitefish, Coregonus clupeaformis, in Northern Lake-Michigan as Assessed by Isoenzyme 1916 Electrophoresis. 37: 783 – 793.

1917

1918 Kennedy, W.A. 1943. The whitefish Coregonus clupeaformis (Mitchill), of Lake Opeongo, 1919 Algonquin Park, Ontario. Univ. Toronto Stud. Biol. Ser. No. 51 (Publ. Ontario Fish. Res. 1920 Lab. No. 62): 22 – 43.

1921

1922 Kirkpatrick, M., and Selander, R.K. 1979. Genetics of speciation in Lake Whitefishes in the 1923 Allegash bain. Evolution 33: 478 – 485.

1924

1925 Koelz, W. 1929. Coregonine fishes of the Great Lakes. Bulletin of the United States Bureau 1926 of Fisheries 43: 297 – 643.

1927

52 1928 Lande, R. 1992. Neutral theory of quantitative genetic variance in an island model with 1929 local extinction and colonization. Evolution 46: 381 – 389.

1930

1931 Landry, L., Vincent, W.F., and Bernatchez, L. 2007. Parallel evolution of Lake Whitefish 1932 limnetic ecotypes in association with limnological features of their adaptive landscape. 1933 Journal of Evolutionary Biology 20: 971 – 984.

1934

1935 Lasenby, T.A., Kerr, S.J., and Hooper, G.W. 2001. Lake Whitefish Culture and Stocking: 1936 An Annotated Bibliography and Literature Review, Peterborough, Ontario.

1937

1938 Lindsey, C.C. 1962. Distinctions between the broad whitefish, Coregonus nasus, and other 1939 North American whitefishes. Journal of Fisheries Research Board Canada 19: 687 – 714.

1940

1941 Lindsey, C.C. 1963a. Status of the whitefish species, Coregonus nelsonii, and the 1942 designation of a new type. Copeia 1: 173 – 174.

1943

1944 Lindsey, C.C. 1963b. Sympatric occurence of two species of humpback whitefish in 1945 Squanga Lake, Yukon Territory. Journal of Fisheries Research Board Canada 20: 749 – 1946 767.

1947

1948 Lindsey, C.C. 1975. Proglacial lakes and fish dispersal in southwestern Yukon territory. 1949 Verh. Internat. Verein. Limnol. 19: 2364 – 2370.

1950

1951 Lindsey, C.C. 1981. Stocks are chameleons: plasticity in gill rakers of Coregonine fishes. 1952 38: 1497 – 1506.

1953

1954 Lindsey, C.C., and McPhail, J.D. 1986. Zoogeography of fishes of the Yukon and 1955 Mackenzie basins. In . Edited by: New York. pp. 639–673. In The zoogeography of North 1956 American freshwater fishes. Edited by C.H. Hocutt and E.O. Wiley. John Wiley and Sons, 1957 New York.

53 1958

1959 Lindsey, C.C. 1988. The relevance of systematics and nomenclature to Coregonine 1960 management. Finnish Fisheries Research 9: 1 – 10.

1961

1962 Lindsey, C.C., Clayton, J.W., and Franzin, W.G. 1970. Zoogeographic problems and 1963 protein variation in the Coregonus clupeaformis whitefish species complex. In Biology of 1964 Coregonine Fishes. Edited by C.C. Lindsey and C.S. Woods. University of Manitoba Press, 1965 Winnipeg, Canada. pp. 127 – 146.

1966

1967 Lindsey, C.C., and McPhail, J.D. 1986. Zoogeography of fishes of the Yukon and 1968 Mackenzie basins. In The zoogeography of North American freshwater fishes. Edited by 1969 C.H. Hocutt and E.O. Wiley. John Wiley and Sons, New York. pp. 639 – 673.

1970

1971 Lindsey, C.C., and Woods, C.S. 1970. Biology of Coregonine Fishes. University of 1972 Manitoba Press, Winnipeg.

1973

1974 Linnæus, C. 1758. Systema Naturae, Ed. X. (Systema naturae per regna tria naturae, 1975 secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, 1976 locis. Tomus I. Editio decima, reformata.) Holmiae. Systema Nat. ed. 10 v. 1

1977

1978 Loch, J.S. 1974. Phenotypic variation in the Lake Whitefish, Coregonus clupeaformis, 1979 induced by introduction into a new environment. Journal of Fisheries Research Board 1980 Canada 31: 55 – 62.

1981

1982 Lu, G., Basley, D.J., and Bernatchez, L. 2001. Contrasting patterns of mitochondrial DNA 1983 and microsatellite introgressive hybridization between lineages of Lake Whitefish 1984 (Coregonus clupeaformis); relevance for speciation. Molecular Ecology 10: 965 – 985.

1985

1986 Lu, G., and Bernatchez, L. 1998. Experimental evidence for reduced hybrid viability 1987 between limnetic and benthic ecotypes of Lake Whitefish (Coregonus clupeaformis

54 1988 Mitchill). Proceedings of the Royal Society of London B – Biological Sciences 265: 1025 – 1989 1030.

1990

1991 Lu, G., and Bernatchez, L. 1999. Correlated trophic specialization and genetic divergence 1992 in sympatric Lake Whitefish ecotypes (Coregonus clupeaformis): Support for the ecological 1993 speciation hypothesis. Evolution 53: 1491 – 1505.

1994

1995 Lu, G.a.B., Louis. 1998. Experimental evidence for reduced hybrid viability between 1996 limnetic and benthic ecotypes of Lake Whitefish (Coregonus clupeaformis Mitchill). Proc. 1997 R. Soc. Lond. B: 1025 – 1030.

1998

1999 Luczynski, M., Rösch, R., Vuorinen, J.A., and Brzunan, P. 1995. Biochemical genetic study 2000 of sympatric Lake Constance whitefish (Coregonus lavaretus) populations: “Blaufelchen” 2001 and “Gangfish”. Canadian Journal of Fisheries and Aquatic Sciences 57: 136 – 143.

2002

2003 Luikart, G., England, P.R., Tallman, D., Jordan, S., and Taberlet, P. 2003. The power and 2004 promise of population genomics: from genotyping to genome typing. Nature Reviews 2005 Genetics 4: 981 – 993.

2006

2007 Lynch, M., Pfrender, M., Spitze, K., Lehman, N., Hicks, J., Allen, D., Latta, L., Ottene, M., 2008 Bogue, F., and Colbourne, J. 1999. The quantitative and molecular genetic architecture of a 2009 subdivided species. Evolution 53: 100 – 110.

2010

2011 MacCrimmon, H.R., and Skobe, E. 1970. The fisheries of Lake Simcoe. Edited by Ontario 2012 Ministry of Natural Resources, Ontario. p. 140.

2013

2014 Mandrak, N. 2003. Biogeographic zones of Canada In COSEWIC (2005). Guidelines for 2015 recognizing designatable units below the species level. Edited by Committee on the Status 2016 of Endangered Wildlife in Canada, Ottawa. Available on request from the COSEWIC 2017 Secretariat (COSEWIC/[email protected])

2018

55 2019 Mayr, E. 1963. Animal Species and Evolution. Harvard University Press, Cambridge MA.

2020

2021 McDermid, J.L., Reist, J.D., and Bodaly, J.A. 2005. Phylogeography and postglacial 2022 dispersal of whitefish (Coregonus clupeaformis complex) in Northwestern North America. 2023 Arch. Hydrobiol. Special Issues Advances in Limnology 60: 91 – 109.

2024

2025 McLaughlin, R.L., A., G.J.W., and L., K.D. 1994. Foraging movements in relation to 2026 morphology, water-column use, and diet for recently emerged brook trout (Salvelinus 2027 fontinalis) in still-water pools. Canadian Journal of Fisheries and Aquatic Sciences 51: 268 2028 – 279.

2029

2030 McLeod, C., O'Neil, J., Hildebrand, L., and Clayton, T. 1979. An examination of fish 2031 migrations in the Liard River, British Columbia, relative to proposed hydroelectric 2032 development at site A,R,L,and L., Edmonton, Alberta. Environmental Services, Ltd. 2033 Edmonton, Alberta. 199 pages.

2034

2035 McPhail, J.D. 2007. The Freshwater Fishes of British Columbia. University of Alberta 2036 Press.

2037

2038 McPhail, J.D., and Lindsey, C.C. 1970. Freshwater Fishes of Northwestern Canada and 2039 Alaska. Fisheries Research Board of Canada, Ottawa.

2040

2041 Mecklenberg, C.W., Mecklenberg, T.A., and Thorsteinson, L.K. 2002. Fishes of Alaska. 2042 American Fisheries Society, Bethesda, Maryland.

2043

2044 Merila, J., and Crnokrak, P. 2001. Comparison of genetic differentiation at marker loci and 2045 quantitative traits. Journal of Evolutionary Biology 14: 892 – 903.

2046

2047 Murray, K.D. 2005. Population genetic assessment of the endangered Atlantic whitefish, 2048 Coregonus huntsmani, and the Lake Whitefish, C. clupeaformis in Atlantic Canada, 2049 Dalhousie University, Halifax, Nova Scotia, Canada.

56 2050

2051 Nalepa, T.F., Mohr, L.C., Henderson, B.A., Madenjian, C.P., and Schneeberge, P.J. 2005. 2052 Lake Whitefish and Diporeia spp. in the Great Lakes: An Overview. In Proceedings of a 2053 workshop on the dynamics of Lake Whitefish (Coregonus clupeaformis) and the amphipod 2054 Diporeia spp. in the Great Lakes. Ann Arbor, MI. Edited by Great Lakes Fisheries 2055 Commission. pp. 3 – 19.

2056

2057 Nelson, J.S., and Paetz, M.J. 1992. The Fishes of Alberta. University of Alberta 2058 Press/University of Calgary Press, Edmonton/Calgary.

2059

2060 Nikolsky, G.V., and Reshetnikov, J.S. 1970. Systematics of Coregonine fishes in the USSR; 2061 Intraspecific variability and difficulties in taxonomy. In Biology of Coregonine Fishes. 2062 Edited by C.C. Lindsey and C.S. Woods. University of Manitoba Press, Winnipeg.

2063

2064 Ostbye, K., Amundsen, P.A., Bernatchez, L., Klemetsen, A., Knudsen, R., Kristoffersen, 2065 R., Naesje, T.F., and Hindar, K. 2006. Parallel evolution of ecomorphological traits in the 2066 European whitefish Coregonus lavaretus (L.) species complex during postglacial times. 2067 Molecular Ecology 15: 3983 – 4001.

2068

2069 Ostbye, K., Naesje, T.F., Bernatchez, L., Sandlund, O.T., and Hindar, K. 2005. 2070 Morphological divergence and origin of sympatric populations of European whitefish 2071 (Coregonus lavaretus L.) in Lake Femund, Norway. Journal of Evolutionary Biology 18: 2072 683 – 702.

2073

2074 Page, L.M., and Burr, B.M. 1991. A field guide to freshwater fishes of North America north 2075 of Mexico. Houghton Mifflin Co., Boston.

2076

2077 Pielou, E.C. 1991. After the ice age. The University of Chicago Press, Chicago.

2078

57 2079 Pigeon, D., Chouinard, A., and Bernatchez, L. 1997. Multiple modes of speciation involved 2080 in the parallel evolution of sympatric morphotypes of Lake Whitefish (Coregonus 2081 clupeaformis, Salmonidae). Evolution 51: 196 – 205.

2082

2083 Politov, D.V., Gordon, N.Y., and Makhrov, A.A. 1999. Genetic identification and 2084 taxonomic relationships of six Siberian species of Coregonus. In Biology and Management 2085 of Coregonine Fishes – Proceedings of the Seventh International Symposium on the 2086 Biology and Management of Coregonine Fishes.

2087

2088 Politov, D., Gordon, N., Afanasiev, K., Altukhov, Y., and Bickham, J. 2000. Identification 2089 of palearctic Coregonine fish species using mtDNA and allozyme genetic markers. Journal 2090 Of Fish Biology 57: 51 – 71.

2091

2092 Power, G. 2002. Charrs, glaciations and seasonal ice. Environmental Biology of Fishes 64: 2093 17 – 35.

2094

2095 Prest, V.K. 1970. Quaternary geology of Canada. In Geology and Economic Minerals of 2096 Canada. Edited by R.J.W. Douglass. Geological Survey of Canada, Ottawa. pp. 676 – 764.

2097

2098 Rawson, D.S. 1947. Great Slave Lake. In North West Canadian Fisheries Surveys in 1944- 2099 1945. Edited by Fisheries Research Board of Canada. p. 94.

2100

2101 Reist, J.D., Lianne, D., Maiers, R.A., Bodaly, J.A., Vuorinen, J., and Carmichae, T.J.I. 2102 1998. The phylogeny of new – and old-world coregonine fishes as revealed by sequence 2103 variation in a portion of the d-loop of mitochondrial DNA. In Biology and Management of 2104 Coregonine Fishes. Edited by R. Eckmann, A. Appenzeller and R. Rosch. pp. 323 – 339.

2105

2106 Reist, J.D., Vuorinen, J., and Bodaly, R.A. 1992. Genetic and morphological identification 2107 of Coregonine hybrid fishes from Arctic Canada. Polish Archives of Hydrobiology 39: 551 2108 – 561.

2109

58 2110 Rempel, L., and Smith, D. 1998. Postglacial fish dispersal from the Mississippi refuge to 2111 the Mackenzie River basin. 55: 893 – 899.

2112

2113 Reshetnikov, J.S. 1988. Coregonine fishes in recent conditions. Finnish Fisheries Research 2114 9: 11 – 16.

2115

2116 Robinson, B.W., and Schluter, D. 2000. Natural selection and the evolution of adaptive 2117 genetic variation in northern freshwater fishes. In Adaptive Genetic Variation in the Wild. 2118 Edited by T. Mousseau, B. Sinvero and J.A. Endler. Oxford University Press, Oxford.

2119

2120 Robinson, B.W., and Wilson, D.S. 1994. Character release and displacement in fishes: A 2121 neglected literature. The American Naturalist 144: 596 – 627.

2122

2123 Rogers, S.M., Gagnon, V., and Bernatchez, L. 2002. Genetically based phenotype- 2124 environment association for swimming behavior in Lake Whitefish ecotypes (Coregonus 2125 clupeaformis Mitchill). Evolution 56: 2322 – 2329.

2126

2127 Rogers, S.M., and Bernatchez, L. 2005. Integrating QTL mapping and genome scans 2128 towards the characterization of candidate loci under parallel selection in the Lake Whitefish 2129 (Coregonus clupeaformis). Molecular Ecology 14: 351 – 361.

2130

2131 Rogers, S.M., and Bernatchez, L. 2006. The genetic basis of intrinsic and extrinsic 2132 postzygotic reproductive isolation jointly promoting speciation in the Lake Whitefish 2133 species complex (Coregonus clupeaformis). Journal of Evolutionary Biology 19: 1979 – 2134 1994

2135 .

2136 Rogers, S.M., and Bernatchez, L. 2007. The genetic architecture of ecological speciation 2137 and the association with signatures of selection in natural Lake Whitefish (Coregonus Sp. 2138 Salmonidae) species pairs. Molecular Biology and Evolution 24: 1423 – 1438.

2139

59 2140 Rogers, S.M., Campbell, D, Baird, S. J. E., Danzmann, R.G., and Bernatchez, L. 2001. 2141 Combining the analyses of introgressive hybridisation and linkage mapping to investigate 2142 the genetic architecture of population divergence in the Lake Whitefish (Coregonus 2143 clupeaformis, Mitchill). Genetica 111: 25 – 41.

2144

2145 Sajdak, S.L., and Phillips, R.B. 1997. Phylogenetic relationships among Coregonus species 2146 inferred from the DNA sequence of the first internal transcribed spacer (ITS1) of ribosomal 2147 DNA. 54: 1494-1503.

2148

2149 Sanderson, S.L., Cech, J.J., and Patterson, M.R. 1991. Fluid Dynamics in Suspension- 2150 Feeding Blackfish. Science 251: 1346 – 1348.

2151

2152 Sanderson, S.L., Cheer, A.Y., Goodrich, J.S., Graziano, J.D., and Callan, W.T. 2001. 2153 Crossflow filtration in suspension-feeding fishes. Nature 412: 439 – 441.

2154

2155 Sandlund, O.T., Næsje, T.F., and Jonsson, B. 1992. Ontogenetic changes in habitat use by 2156 whitefish, Coregonus lavaretus. Environmental Biology of Fishes 33: 341 – 349.

2157

2158 Sandlund, O.T., Næsje, T.F., and Saksgård, R. 1995. Ecological diversity in whitefish 2159 Coregonus lavaretus: ontogenetic niche shifts and polymorphism. Arch. Hydrobiol. Spec. 2160 Issues Advanc. Limnol. 46: 49 – 59.

2161

2162 Schluter, D. 1996. Ecological speciation in postglacial fishes. Philosophical Transactions of 2163 the Royal Society of London Series B-Biological Sciences 351: 807 – 814.

2164

2165 Schluter, D. 2000. The Ecology of Adaptive Radiation. Oxford University Press, New 2166 York.

2167

2168 Schluter, D., and McPhail, J.D. 1993. Character displacement and replicate adaptive 2169 radiation. Trends in Ecology and Evolution 8: 197 – 200.

60 2170

2171 Schluter, D., and Nagel, L.M. 1995. Parallel Speciation by Natural-Selection. American 2172 Naturalist 146: 292 – 301.

2173

2174 Schmidt, R.E. 1986. Zoogeography of the Northern Appalachians. In Zoogeography of 2175 North American Freshwater Fishes. Edited by C.H. Hocutt and E. Wiley. John Wiley and 2176 Sons, New York. pp. 137 – 159.

2177

2178 Scott, W.B., and Crossman, E.J. 1964. Fishes occurring in the freshwaters of insular 2179 Newfoundland. Edited by Ontario Department of Fisheries, Ontario, Canada. p. 124.

2180

2181 Scott, W.B., and Crossman, E.J. 1973. Freshwater Fishes of Canada. Bulletin of the 2182 Fisheries Research Board of Canada, Ottawa 184. 966 pages.

2183

2184 Sendek, D.S. 1999. Electrophoretic studies of Coregonine fishes from across Russia. In 2185 Biology and Management of Coregonine Fishes – Proceedings of the Seventh International 2186 Symposium on the Biology and Management of Coregonine Fishes. Todd, T. Fleischer, G., 2187 Ann Arbor, Michigan, U.S.A. p. 689.

2188

2189 Smith, G.R. 1992. Introgression in Fishes: Significance for Paleontology, Cladistics, and 2190 Evolutionary Rates. Taylor and Francis, Ltd. for the Society of Systematic Biologists. pp. 2191 41 – 57.

2192

2193 Smith, G.R., and Todd, T.N. 1984. Evolution of species flocks of fishes in North Temperate 2194 Lakes. In Evolution of Fish Species Flocks. Edited by A.A. Echelle and I. Kornfield. 2195 University of Maine at Orono Press, Orono, Maine. pp. 45 – 68.

2196

2197 Spitze, K. 1993. Population Structure in Daphnia obtusa: Quantitative Genetic and 2198 Allozymic Variation. Genetics 135: 367 – 374.

2199

61 2200 Stamford, M.D., and Taylor, E.B. 2004. Phylogeographical lineages of Arctic grayling 2201 (Thymallus arcticus) in North America: divergence, origins and affinities with Eurasian 2202 Thymallus. Molecular Ecology 13: 1533-1549.

2203

2204 Stinchcombe, J.R., and Hoekstra, H.E. 2007. Population genomics and quantitative 2205 genetics: combined approaches for finding the genes underlying ecologically important 2206 traits. Heredity 100: 158-170.

2207

2208 Storz, J.F. 2005. Using genome scans of DNA polymorphism to infer adaptive population 2209 divergence. Molecular Ecology 14: 671-688.

2210

2211 Stott, W., Todd, T.N., and Kallemeyn, L. 2004. Genetic variability among Lake Whitefish 2212 from isle Royale and the Upper Great Lakes. Annales Zoologici Fennici 41: 51-59.

2213

2214 Svaerdson, G. 1998. Postglacial Dispersal and Reticulate Evolution of Nordic Coregonines. 2215 Nordic Journal of Freshwater Research 74: 3-32.

2216

2217 Svärdson, G. 1957. The Coregonine problem. VI. The palearctic species and their 2218 intergrades. Report of the Institut of Freshwater Research, Drottningholm 38: 261-356.

2219

2220 Svärdson, G. 1965. The Coregonine Problem VII. The isolating mechanisms in sympatric 2221 species. Report of the Institut of Freshwater Research, Drottningholm 46: 95-123.

2222

2223 Taylor, E. 1999. Species pairs of north temperate freshwater fishes: Evolution, taxonomy, 2224 and conservation. Reviews In Fish Biology And Fisheries 9: 299 – 324.

2225

2226 Taylor, E., and McPhail, J. 2000. Historical contingency and ecological determinism 2227 interact to prime speciation in sticklebacks, Gasterosteus. Proceedings Of The Royal 2228 Society Of London Series B-Biological Sciences 267: 2375 – 2384.

2229

62 2230 Taylor, E.B. 2006. Guidelines for Recognizing Designatable Units Below the Species 2231 Level. Edited by Freshwater Fisheries Specialist Subcommittee. Available on request from 2232 the COSEWIC Secretariat (COSEWIC/[email protected]).

2233

2234 Taylor, E.B., and Bentzen, P. 1993. Evidence for multiple origins and sympatric divergence 2235 of trophic ecotypes of smelt (Osmerus) in Northeastern North America. Evolution 47: 813 – 2236 832.

2237

2238 Taylor, E.B., Boughman, J.W., Groenenboom, M., Sniatynski, M., Schluter, D., and Gow, 2239 J.L. 2006. Speciation in reverse: morphological and genetic evidence of the collapse of a 2240 threespined stickleback (Gasterosteus aculeatus) species pair Molecular Ecology 15: 343 – 2241 355.

2242

2243 Trudel, M., Tremblay, A., Schetagne, R., and Rasmussen, J.B. 2001. Why are limnetic fish 2244 so small? An energetic analysis of polymorphism in Lake Whitefish (Coregonus 2245 clupeaformis). 58: 394 – 405.

2246

2247 Tsuyuki, H., Uthe, J.F., Roberts, E., and Clarke, L.W. 1966. Comparative 2248 Electropherograms of Coregonus Clupeaformis Salvelinus Namaycush S Alpinus S Malma 2249 and S Fontinalis from Family Salmonidae. Journal of the Fisheries Research Board of 2250 Canada 23: 1599

2251 .

2252 Turgeon, J., Estoup, A., and Bernatchez, L. 1999. Species flock in the North American 2253 Great Lakes: Molecular ecology of Lake Nipigon Ciscoes (Teleostei : Coregonineae : 2254 Coregonus). Evolution 53: 1857 – 1871.

2255

2256 Underhill, J.C. 1986. The fish fauna of the Laurentian Great Lakes, the St. Lawrence 2257 lowlands, Newfoundland and Labrador. In Zoogeography of North American Freshwater 2258 Fishes. Edited by C.H. Hocutt and E. Wiley. John Wiley and Sons, New York.

2259

63 2260 Vamosi, S., and Schluter, D. 2002. Impacts of trout predation on fitness of sympatric 2261 sticklebacks and their hybrids. Proceedings Of The Royal Society Of London Series B- 2262 Biological Sciences 269: 923 – 930.

2263

2264 Vetter, W., Bartha, R., Stern, G., and Tomy, G. 1999. Environ. Toxicol. Chem. 18: 2775 – 2265 2781.

2266

2267 Vuorinen, J.A., Bodaly, R.A., Reist, J.D., Bernatchez, L., and Dodson, J.J. 1993. Genetic 2268 and Morphological-Differentiation between Limnetic and benthic Size Forms of Lake 2269 Whitefish (Coregonus clupeaformis) in Como Lake, Ontario. Canadian Journal of Fisheries 2270 and Aquatic Sciences 50: 210 – 216.

2271

2272 Vuorinen, J.A., Bodaly, R.A., Reist, J.D., and Luczynski, M. 1998. Phylogeny of five 2273 Prosopium species with comparisons with other Coregonine fishes based on isozyme 2274 electrophoresis. Journal of Fish Biology 53: 917 – 927.

2275

2276 Walters, V. 1955. Fishes of western arctic America and eastern arctic Siberia. Taxonomy 2277 and zoogeography. Bull. Am. Mus. Nat. Hist. 106: 259 – 368.

2278

2279 Webb, P.W. 1983. Speed, acceleration and manoeuvrability of two teleost fishes. J. Exp. 2280 Biol. 102: 115 – 122.

2281

2282 Webb, P.W. 1984. Body form, locomotion and foraging in aquatic vertebrates. American 2283 Zoologist 24: 107 – 120.

2284

2285 Wedekind, C., Muller, R., and Spicher, H. 2001. Potential genetic benefits of mate selection 2286 in whitefish. Journal of Evolutionary Biology 14: 980 – 986.

2287

2288 Weihs, D., and Webb, P.W. 1984. Optimal avoidance and evasion tactics in predator-prey 2289 interactions. Journal of Theoretical Biology 106: 189 – 206.

64 2290

2291 Wilson, C., and Hebert, P. 1998. Phylogeography and postglacial dispersal of lake trout 2292 (Salvelinus namaycush) in North America. 55: 1010 – 1024.

2293

2294 Woodger, C.D. 1976. Morphological variations as induced by environment in Coregonines. 2295 Environmental Biology of Fishes 1: 101 – 105.

2296

2297 Zitzow, R.E., and Millard, J.L. 1988. Survival and Growth of Lake Whitefish (Coregonus 2298 clupeaformis) Larvae Fed Only Formulated Dry Diets. Aquaculture 69: 105 – 113.

2299

2300

2301

2302

2303

2304

2305

2306

2307

2308

2309

2310

2311

2312

2313

2314

2315

65 2316 Table 1. Lake Whitefish populations sampled in Canada and relevant regions of the United 2317 States of America. ID: locations (see Figures 2 and 5). Site: river or lake with 2318 provincial/state abbreviations. Taxonomic entity: presence of the C. lavaretus DU from 2319 Section 1 of the key (otherwise C. clupeaformis). SP: presence of a species pair from 2320 Section 1.1 of the key. PG: major phylogeographic groupings; B=Beringian, E=Eurasian, 2321 N=Nahanni, At=Atlantic, Ac=Acadian, and NA=not applicable. BZ: National Freshwater 2322 Biogeographic Zones from Section 5 of the key, (see Figure 5 for details). PDU: Putative 2323 Designatable Unit identification, see Figure 2 for complete list. Ref: References for the 2324 source of the populations sampled. 2325

ID Site Taxonomic entity SP PG BZ PDU Ref

1 Yukon R. AK C. lavaretus B, E 6 1,23 5, 6 2 Minnesota L. AK C. lavaretus B, E n/a 5, 6 3 Chatanika R. AK C. lavaretus B, E n/a 5, 6 4 Davis L. YT B 6 23 7 5 Hanson L. YT NA 6 23 12 6 Tatchun L. YT B 6 23 3 7 Squanga L. YT C. lavaretus (limnetic) Y B, E 6 1,2-3,23 3, 5-7 8 Little Teslin L. YT C. lavaretus (lim + ben) Y B, E 6 1,4-5,23 5, 6 9 Dezadeash L. YT C. lavaretus (limnetic) Y B, E 6 1,6-7,23 7, 3 10 McClintock L. YT B 13 22 7 11 Aishihik L. YT B 6 23 5 – 7 12 Kluane L. YT B 6 23 7, 3 13 Margaret L. YT B 13 22 7 14 Dease L. YT B 13 22 7 15 Finlayson L. YT B 13 22 7 16 Frances L. YT B 13 22 3, 7 17 Simpson L. YT B 13 22 7 18 Watson L. YT B 13 22 3, 7 19 Wheeler L. YT B 13 22 7 20 Toobally L. YT B,N 13 22 7 21 Crooked L. BC B,N 13 22 7 22 Liard R. BC C. lavaretus (upper) B,N,M 13 22 7

23 Fisherman's L. BC N 13 22 7

66 24 Bovie L. BC N 13 22 7 25 Seaplane L. BC N 13 22 7 26 Divide L. BC N 13 22 7 27 Little Doctor L. NT N 13 22 7 28 Crooked R. BC B or N 13 22 5, 6 29 Quesnel L. BC NA 11 24 2 30 Fraser L. BC N 11 24 7 31 Aleza L. BC N 11 24 7 32 Lac la Hache BC N 11 24 3, 7 33 Williams L. BC N 11 24 3, 7 34 Summit L. BC N 11 24 3, 7 35 McLeod L. YT N 11 24 3, 7 36 Moberly L. BC N 13 22 3, 7 37 Utikuma L. AB N 13 22 7 38 Talbot L. AB N 4 25 3, 7 39 Lesser Slave L. AB M 13 22 7 40 Athabasca R. SK M 13 22 7 41 Athabasca L. AB M 13 22 7 42 Great Slave L. NT M 13 22 3, 5 – 7 43 Wabamum L. AB N, M 4 25 3, 5 – 7 44 Waterton L. AB M 7 26 7, 3 45 Fort Simpson NT B,N,M 13 22 7 46 Fort Good Hope NT B,N,M 13 22 7 47 East Channel NT B,N,M 13 22 7 47 Arctic Red R. NT B, M 6 23 5, 6 48 MacKenzie Delta YK B,N,M 13 22 7 48 Fort McPherson NT M 13 22 5, 6 49 Cox L. NT B,N,M 13 22 7 49 McEvoy L. YT C. lavaretus B 13 1,22 5, 6 50 Jack Fish L. SK. M 4 25 5, 6 51 South Indian L. MB M 5 27 5, 6 52 Lake Superior ON Y M 10 28 5, 6

67 53 Lake Michigan MI M 10 28 5, 6 54 Lake Michigan MI M 10 28 5, 6 55 Lake Huron MI M 10 28 5, 6 56 Lake Ontario ON M 10 28 5, 6 57 Como Lake ON Y M 10 16-17,28 5, 6 5, 6 58 Res. Kipawa QC M 10 28 59 Rupert R. QC M 3 29 5, 6 60 Eastmain R. QC M 3 29 5, 6 61 La Grande R. QC M 3 29 5, 6 62 Great Whale R. QC M 3 29 5, 6 63 Inukjuak R. QC M 2 30 5, 6 64 Povungnituk R. QC M 2 30 5, 6 65 Koksoak R. QC M 2 30 5, 6 66 Squaw L. QC M 2 30 5, 6 67 Altikamagen L. QC M 2 30 5, 6 68 Res. ManicV QC M 2 30 5, 6 69 Caniapiscau QC M 2 8-9,30 4, 8 70 Manicouagan QC M 9 10-11,31 4, 8 71 Outardes II QC M 9 12-13,31 4, 8 72 St – Lawrence R. QC M 10 28 5, 6 73 L. Champlain QC M 9 31 5, 6 74 L. St-Francois QC M 9 31 5, 6 75 East L. QC Y Ac 9 20-21,31 8 – 10 76 L. Témiscouata QC Y At, Ac 1 18-19,32 4 – 6, 8-11 77 Spider L. ME At, Ac n/a 5, 6 78 Musquacook L. ME At, Ac n/a 5, 6 79 Cliff L. ME Y At, Ac n/a 4 – 6, 8 – 10 80 Grand L. NB Ac 1 32 5, 6 81 Mira River NS Ac 1 32 5, 6 82 Opeongo Lake ON Y M 10 14-15, 28 5, 6

2326 1 Kennedy 1943, 2 McPhail and Lindsey 1970, 3 Franzin and Clayton 1977, 4 Bernatchez and 2327 Dodson 1990, 5 Bernatchez and Dodson 1991, 6 Bodaly et al. 1991, 7 Foote et al. 1992, 8 Pigeon et 2328 al. 1997, 9 Lu et al. 1999, 10 Lu et al. 2001, 11 Rogers et al. 2001, 12 Scott and Crossman 1973

68 2329 List of Figures

2330 2331 Figure 1. (a) Lake Whitefish (Scott and Crossman 1973) (b) the head of a cisco (left) 2332 versus the head of a whitefish (right) where the head of the jaw projects beyond, or is equal 2333 to, the upper jaw. In C. clupeaformis, the profile of the upper lip slopes backwards in line 2334 with the forehead. (c) Convex brow of a broad whitefish (C. nasus on left) versus the 2335 concave head profile of Lake Whitefish (C. clupeaformis, on right). (d) C. nasus gill raker 2336 (on left) with the longest gill raker measuring one-fifth the inter-orbital width and C. 2337 clupeaformis gill raker (on right) with the longest gill raker measuring more than one fifth 2338 the inter-orbital length. (Key and images modified from McPhail and Lindsey, Journal of 2339 the Fisheries Research Board of Canada, 1970) 2340 2341 Figure 2. A Lake Whitefish species complex DU decision chart. The numbers on the top, 2342 from left to right, reflect the steps used in the key used to identify putative designatable 2343 units (DUs). The boxes show DUs identified for each of these steps. Lines connecting DUs 2344 reflect different stages of the decision process. For example, DU2-3 (Squanga Lake species 2345 pair) is connected to the Beringia DU as well as the Yukon NFEA DU. Dashed boxes and 2346 lines indicate provisional DU status. Species Pair DUs of Lake Whitefish are shown in 2347 yellow because the initial DU status designation occurred in stage 1 (see section 1.1), but 2348 was further supported in stage 3 of the key (local adaptation). 2349 2350 Figure 3. Distribution and phylogeography of Lake Whitefish in Canada. The grey line 2351 shows the overall distribution of the Lake Whitefish species complex while the blue line 2352 represents the maximum extent of glacial ice during the ice age (modified from Bernatchez 2353 and Dodson 1991). Locations of Lake Whitefish populations are colour coded according to 2354 their major phylogenetic groupings representative of their ancestral glacial refugia (see 2355 Section 2 of the key). Blue = Beringian, Green = Nahanni, Yellow = Mississippi, Red = 2356 Atlantic, Brown = Acadian. Locations of species pairs are labelled as diamonds. See Table 2357 1 for details about the samples and Figure 2 for their status as putative DUs. 2358 2359 Figure 4. Limnetic and benthic Lake Whitefish species pairs; (a) Benthic (top) and 2360 Limnetic (bottom) from Indian Lake, Maine, and (b) Limnetic (high gill raker, top) and 2361 Benthic (low gill raker, bottom) from Squanga Lake, Yukon. 2362 2363 Figure 5. Distribution of samples with respect to National Freshwater Biogeographic 2364 Zones. See Table 1 for more information. 2365

2366

2367

2368

2369

69 2370

2371

2372

2373

2374

2375

2376

2377

2378

2379 Figure 1. 2380

2381

2382

2383

2384

2385

2386

70 Lake whitefish DU decision chart

1 2 3 4 5 Dist inct Major Locally Major taxonomic phylogenetic adapt ed range NFEA? entity? grouping? populations? disjunction?

DU22: Western DU2-3: Arctic Beringian Squanga L. DU1: “C. lavaretus” DU4-5Little DU6-7: DU23: DU24: Teslin L. Dezadeash L. Yukon Pacific Nahanni DU8-9: DU25: DU29: Caniapiscau L. C. clupeaformis DU14-15: L. Saskatchewan James Bay species complex DU10-11: Opeongo DU26: DU30: Mississippi Manicouagan V DU16-17: Missouri Eastern Arctic DU12-13: Como L. DU27: Western DU2-32: Lac Outardes II Atlantic Hudson Bay C. clupeaformis DU18-19: L DU28: Temiscout a Great Lakes Acadian

DU20-21: DU31:Eastern DU32: East L. St. Lawrence Maritimes Figure 2 2387 71 2388

2389

2390

2391

2392

2393 Figure 3 2394

2395

2396

2397

2398

72 2399

2400

2401

2402

2403

2404

2405 Figure 4 2406

2407

73 2408

2409

2410

47 49 50 4 52 48

5 13 51 12 46 6 9 11 8 7 10 27 26 45 25 42 23 22 15 18 20 24 67 68 14 16 19 21 41 66 28 40 37 54 69 70 35 36 72 29 65 34 39 71 30 31 38 64 73 32 74 63 33 53 43 62 84 79 41 83 44 78 80 61 75 55 81 60 85 77 82 76 58 56 57 59

2411

2412 Figure 5 2413

2414

2415

2416

2417

74