4
Genetics of Wild Maine Salmon Populations

INTRODUCTION

We now turn to a review and analysis of the genetic evidence. The commit-tee’s charge is to describe the genetic makeup of wild populations of Atlantic salmon in Maine with a focus on their distinctiveness. In other words, the committee is asked to assess the extent to which Maine populations of Atlantic salmon diverge genetically from other Atlantic salmon populations and among themselves. The question of distinctiveness applies at several levels. First, are North American salmon genetically distinct from European salmon? Second, are U.S. Atlantic salmon (i.e., Maine salmon) distinct from other North American (Canadian) salmon? Finally, to what degree are salmon populations in the eight DPS rivers in Maine distinct from each other? The answers to these questions are relevant to questions about the most effective methods for protecting and restoring wild populations of Atlantic salmon in Maine (e.g., Hedrick 2001), but we defer those issues until the committee’s final report. Atlantic salmon currently spawn in North American streams, ranging from Labrador south to Maine. Quebec has a local population in eastern Hudson Bay, and Newfoundland has its own populations, which might have been established by salmon from Europe as well as North America. West Greenland has at least one natural population and Iceland, has several, which are normally considered to be European populations. For purposes of this report, the major question concerns the relationship between Maine populations and neighboring populations in Canada (New Brunswick and Nova



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine 4 Genetics of Wild Maine Salmon Populations INTRODUCTION We now turn to a review and analysis of the genetic evidence. The commit-tee’s charge is to describe the genetic makeup of wild populations of Atlantic salmon in Maine with a focus on their distinctiveness. In other words, the committee is asked to assess the extent to which Maine populations of Atlantic salmon diverge genetically from other Atlantic salmon populations and among themselves. The question of distinctiveness applies at several levels. First, are North American salmon genetically distinct from European salmon? Second, are U.S. Atlantic salmon (i.e., Maine salmon) distinct from other North American (Canadian) salmon? Finally, to what degree are salmon populations in the eight DPS rivers in Maine distinct from each other? The answers to these questions are relevant to questions about the most effective methods for protecting and restoring wild populations of Atlantic salmon in Maine (e.g., Hedrick 2001), but we defer those issues until the committee’s final report. Atlantic salmon currently spawn in North American streams, ranging from Labrador south to Maine. Quebec has a local population in eastern Hudson Bay, and Newfoundland has its own populations, which might have been established by salmon from Europe as well as North America. West Greenland has at least one natural population and Iceland, has several, which are normally considered to be European populations. For purposes of this report, the major question concerns the relationship between Maine populations and neighboring populations in Canada (New Brunswick and Nova

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine Scotia), because of the extended history of stocking Maine hatcheries with fish from Canadian streams. During the early European colonization of North America, Atlantic salmon were found as far south as the Connecticut and Hudson rivers, but continuous attempts to reintroduce the species to the Connecticut have failed to establish self-sustaining wild populations, and the southern limit of wild Atlantic salmon is now the Sheepscot River in Maine (Figure 1). The “at least eight [Maine] rivers” listed by NMFS and FWS (DOI and DOC 2000) as containing wild salmon populations are (from west to east) Sheepscot, Ducktrap, Penobscot, Narraguagus, Pleasant, Machias, East Machias, and Dennys. The Saint John River, whose mouth is in New Brunswick, drains areas in Maine’s northeastern highlands, but its salmon are considered to be among New Brunswick’s populations. The Penobscot and its tributaries harbor the largest populations in Maine, and those from the Narraguagus, Machias, and Dennys are substantial. Those from the other watersheds are smaller (Maine 2000). Occasionally, Atlantic salmon have been seen in the Androscoggin, Kennebec, Union, and several smaller rivers, but they probably include strays or aquaculture escapees. Those additional rivers might well figure in recovery plans, but with the exception of the lower Kennebec and its tributaries, they are not thought to support wild populations. The Evidence Much of the evidence on genetic distinctiveness of Atlantic salmon populations is based on laboratory analyses of the variations in gene products (i.e., proteins) taken from samples in the field. In some cases, the genetic material itself (DNA) is analyzed for variation. In either case, the variation observed is compared among populations, and a variety of tests are performed to decide whether the populations are statistically distinct from each other and how accurately an individual salmon can be identified correctly with its source population. Although there might be some relationship between the genetic variation detected in these analyses and adaptively significant differences among the various populations, none can be inferred from these analyses. The differences could be due to random processes (sampling effect or genetic drift), and the markers themselves are thought to be adaptively neutral. In other words, most of the laboratory analyses of genetic variation discussed below cannot provide information on the degree to which different populations have adapted to different local conditions. They can provide information on the degree of isolation of populations—and isolation of populations is a prerequisite for the development of genetically based adaptive differences in

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine them. Analyses can also describe patterns of variation that can be compared with other sets of populations about which more is known. Stronger inferences about adaptive differences among populations can be obtained. The simple observation that populations differ in phenotypic traits that can affect fitness (such as fecundity, time of spawning, body size or shape, and growth rates) is not sufficient to infer adaptive genetic differences among them, because those traits can also be affected by environmental conditions. However, if salmon from different populations continue to show differences when reared in similar environments (“common-garden” experiments) or if the differences can be shown to segregate genetically in breeding experiments, then the observed differences are more likely to reflect genetic adaptation. This kind of information is not yet available for Atlantic salmon in Maine. Several earlier genetic data sets provide us with information on the geographic structure of genetic variation in Atlantic salmon, and they are useful in setting the stage for more in-depth (and more recent) analyses using microsatellite1 DNA markers of the North American populations. A brief overview of those earlier studies is given below. Allozymes Allozymes (short for allelic enzymes) are protein gene products and can be used as indicators of genetic variation. In many cases, different forms of genes produce enzyme variants that can be detected in the laboratory. Those variants seldom have known adaptive significance, but like the other markers discussed below, they might be physically linked on the same chromosome to gene variants that are adaptively important, and they do provide information that can be used to infer differences among individuals and populations. The allozyme work reported to date on Atlantic salmon over their natural range (Ståhl 1987; reviewed in Davidson et al. 1989) and on Downeast Maine salmon populations (May et al. 1994 and references therein) shows that North American and European continental populations are divergent and that populations from the Baltic diverge from those in the eastern North Atlantic (Ståhl 1   Short repeated sequences of two to four bases (the building blocks of DNA) that provide multiple alleles for each of many genetic loci. They provide ample genetic variation to characterize genetic differences between individuals, both within and among populations. Currently, microsatellite data are the best available for population screening for genetic variation.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine 1987). Analysis of populations within regional zones, defined along national lines in Europe and along provincial and state lines in North America, indicates that there is substantial genetic subdivision within continental collections. The spatial scale represented by national regions in Europe is greater than that represented by provincial and state counterparts in North America. There is genetic divergence from watershed to watershed, within regions, and where people have looked, even among tributaries within a watershed. There is also divergence among populations that spawn at different times. This typical pattern has been attributed to precise homing of salmon to their natal streams (Ståhl 1987, Allendorf and Waples 1996, Nielsen 1998). Mitochondrial DNA (mtDNA) Mitochondria are cellular organelles that contain genetic material (DNA). MtDNA is maternally inherited as a unit. The various markers of the mtDNA genome are free of the recombination that characterizes all the other types of variation under discussion. It conveys strong information about female lineages, but nothing about male lineages. Like allozymes, variants in that DNA material can provide information on population divergence. The mitochondrial genomes of European and North American populations can be effectively discriminated by the presence or absence of a single RFLP2 marker of mtDNA (Bermingham et al. 1991). North American populations are dominated by two haplotypes3 and European populations by one, although the European haplotype has been recovered in Newfoundland (King et al. 1999). There are less-categorical genetic markers that we can use to distinguish statistically among regional and subregional populations within continents (Hovey et al. 1989, Palva et al. 1989, McVeigh et al. 1991, King et al. 1993, Palsson and Arnason 1994, O’Connell et al. 1995, Tessier et al. 1995, Nilsson et al. 2001). The essential pattern of these studies is described by King et al. 2   Restriction fragment length polymorphism. Bacteria have enzymes that cut DNA strands containing particular sequences—useful in degrading invasive viral DNA. These (restriction) enzymes can be used to assay DNA samples. Genetic variation in the population ensures that some individuals contain DNA sequences recognized by the enzymes, while some do not. The DNA is cut into fragments wherever the restriction enzyme encounters its specific recognition sequence. The position of the resulting DNA bands on a gel indicates the individual’s genotype. 3   Combinations of genetic markers that are linked closely enough along the mtDNA molecule to be inherited as a single unit.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine (2000), who partitioned the mtDNA haplotypic variation among continents, regions, and watershed populations with the use of molecular analysis of variance. (Amova was used to partition the molecular genetic variation into components; see Excoffier et al. 1992.) The majority of the variation (68%) is attributable to continental divergence, almost enough to be invariably diagnostic; 8% is attributable to population differences within continents (both regional and interwatershed variation), and 24% is attributable to variation among individuals within single populations. If we examine just the variation within the North American collection, 2% of the variation was attributable to Maine versus Canadian divergence, and about 26% to population divergence within either Maine or Canada. The remaining 72% of the variation was found within single populations.4 Within Maine populations, only 2% of the variation was attributable to divergence among watersheds, 12% was attributable to variation among tributaries in the same watershed, and 86% was attributable to variation within a tributary. Within European populations, virtually none of the variation was attributable to regional (national) divergence, 19% was attributable to divergence among watersheds within regions, and 81 % was attributable to variation within single watersheds. Other Genetic Markers A diagnostic genetic distinction between European and North American populations of Atlantic salmon was also established by an early study of ribosomal RNA variation using RFLPs (Cutler et al. 1991). A later study of single-locus minisatellite5 DNA markers showed virtual separation of European and North American populations but with a low frequency of European alleles in Newfoundland (Taggart et al. 1995). An early analysis (Schill and Walker 1994) using RAPD6 markers of samples from the Sheepscot, Ducktrap, Penobscot, Narraguagus, Pleasant, Machias, East Machias, Dennys, and Saint John rivers showed that genetic divergence among watershed popula- 4   In salmon, the highest percentage of genetic variation is normally among individuals in a population. 5   Long sequences of DNA repeated sequentially. Some individuals have more copies of the repeat element than others, leading to DNA fragments that move differently on an electrophoretic gel. Comparisons of minisatellite banding patterns can be used as “genetic fingerprints” and to measure genetic differences. 6   Randomly Amplified Polymorphic DNA. Small segments of DNA replicated (amplified) to large quantities. As with other similar markers, variation is detected by the presence or absences of bands on a gel.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine tions represented 5.7% to 8.0% of the total variation, depending on which rivers were included in the test. This result is comparable to corresponding allozyme partition of variation. Interestingly, the Saint John River sample was differentiable from the others but not as different from the populations in nearby Maine watersheds as from the population in the more distant Sheepscot River. No European versus North American contrast was tested, the distinction having been firmly established by others as described above. Recent Microsatellite Analyses Microsatellite DNA markers, like the other markers described above, are assumed to be adaptively neutral. Recently, the North American populations have been sampled more intensively, and genetic assays have benefited from the higher resolution available from microsatellite markers. King et al. (2001) measured the relative amounts of genetic variation found within single populations, among watersheds, among North American provincial and state populations, and between European and North American collections. Such measurements have been made in various ways, the simplest of which is Amova (Excoffier et al. 1992). The results are summarized in Table 1. The results are highly significant,7 with p < 0.0001. European and North American populations are divergent enough (22% of the variation) to make it virtually impossible to mistake a North American genotype for a European genotype or vice versa. As mentioned above, European and North American fish can be artificially hybridized, but hybridization would not occur naturally because of the vast separation of spawning habitats. If it were to happen in North American waters because of escape and interbreeding of aquacultural stock, the F1 (first- 7   The variation percentages extracted from an Amova table are the average per-locus figures. It is quite usual (for any single locus) to have the majority of polymorphic variation within populations. In comparing populations, however, it is important to remember that these are multi-gene breeding collections. The among-population divergence in allele frequencies, when multiplied over many loci, translates into “gene pools” that are substantially (or sometimes) completely non-overlapping in their genetic constitution (Smouse et al. 1982; Smouse and Chevillon 1998). The allele-frequency differences between European and North American populations are large enough that a salmon sampled from North America would almost never have a multi-gene combination that would occur in any European population, and vice versa. The differences between Canadian and Maine populations are smaller, and those among populations in Maine streams (e.g., Sheepscot and Narraguagus) are still smaller. The degree of gene-pool non-overlap increases with the size of the interpopulation percentage of the variation.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine generation) hybrids would be fairly obvious genetically, as are artificial F1s (King et al. 1999). European salmon are genetically so different from North American salmon that it should be easy to detect their presence (as aquacultural escapees) among wild spawning adults. Even F1 hybrids between North American and European salmon should be evident. Detecting the presence of farm fish of Penobscot or St. John origin will be more problematic, because the genetic differences from the wild fish will be smaller (King et al. 1999). The presence of advanced generation hybrids will be difficult to identify with certainty. Further genetic characterization of aquacultural stocks used in North America is needed. From Table 1a, we discover that FCT (the fraction of total variation among individuals that is accounted for by continental average differences in allele frequencies)=0.219, while FPC (the fraction of the within-continent variation that separates populations)=0.07. The fraction of the total variation that is within populations is 1-FPT=0.726. From Table 1b, confined to North American variation, we discover that FCT (the fraction of North American variation that is due to mean regional (provincial or state) differences in allele frequencies)=0.032, while FPR (the fraction of the within-region variation that separates populations)=0.032. The fraction of total North American variation that is within populations/drainages is 1-FPR=0.939. The corresponding values for European variation are FCT (the fraction of European variation that is due to mean national differences in allele frequencies) = 0.061, while FPR (the fraction of the within-nation variation that separates populations)=0.060. The fraction of total European variation that is within populations/drainages is 1-FPR=0.886. These F-values are analogous to F-statistics (Excoffier et al. 1992), and they are all highly significant, as determined by permutational testing. Within the North American collection, the differences among provincial and state collections of populations are smaller, representing about 3% of the total variation. Divergence among watersheds within a province or state accounts for another 3% of the variation. That raises the question of whether drawing political lines around collections from different watersheds is the most sensible way to delineate regional collections. Within-watershed variation represents the remaining 94% of the variation. The corresponding numbers for European populations are 6%, 5%, and 89%, respectively, but the European (national) regions are physically more separated than North American (provincial and state) regions. The North American and European popu-

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine TABLE 1 Molecular Analysis of Variance (Amova) for Microsatellite Genotypes Continental and Infracontinental Analysisa (1-FPT)=(1-FPC)×(1-FCT) Source of Variation Variance Components Percentage of Variation F-Statistic Estimates Among continents 1.12 21.88 FCT=0.219 Among populations 0.28 5.49 FPC=0.070 Within populations 3.72 72.63 FPT=0.274 North American Analysisb (1-FPC)=(1-FPR)×(1-FRC) Source of Variation Variance Components Percentage of Variation F-Statistic Estimates Among provinces/states 0.12 3.15 FRC=0.032 Among populations 0.12 2.99 FPR=0.032 Within populations 3.56 93.86 FPC=0.061 European Analysisc (l-FPC)=(l-FPR)×(l-FRC) Variation Source Variance Components Percentage of Variation F-Statistic Estimates Among countries 0.29 6.13 FRC=0.061 Among populations 0.25 5.28 FPR=0.060 Within populations 4.15 88.59 FPC=0.114 aContinental and infracontinental component. bNorth American provincial variation partitioned into components for populations (streams) and variation within populations. cEuropean provincial variation partitioned into components for populations (streams) and variation within populations. Source: Adapted from King et al. 2001 . lations are much more divergent than the regional collections within a continent, and watershed-specific populations within a province or state are less divergent still. This pattern is fairly typical for salmon and their relatives, including various Pacific salmon species (Ryman 1983, Allendorf and Waples

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine 1996). Today’s pattern is typical of anadromous fish species in general and salmon in particular (Ståhl 1987). It is useful to compare the recent microsatellite results with older allozyme and mtDNA results. The nuclear (allozyme and microsatellite) markers tell a consistent story, again with p<0.0001. The mtDNA markers show more continental divergence and about the same watershed-to-watershed divergence, but they show less variation within watersheds than do the allozyme studies. The RAPD studies indicate that 5.7% to 8.0% of the variation in populations from the Saint John River and southwestward is accounted for by watershed-to-watershed differences, the rest being accounted for within populations. Assignment Analyses Another approach to determine the level of genetic divergence among populations, particularly when separation is less than complete, is to determine the likelihood that one fish can be identified correctly as to its population of origin based on the genetic information. The degree of confidence with which one can assign a particular fish is a measure of the nonoverlap of the gene pools (Smouse and Chevillon 1998). Early work of this type used genetic-distance methods (Spielman and Smouse 1976, Smouse et al. 1982), but more recently, likelihood procedures have been used more often (Paetkau et al. 1997). Regional Groupings We consider first the regional groupings, nested within North America and European collections. On the basis of the genetic markers, individual fish can be assigned to the regional population to which they are most similar (e.g., Brown et al. 1996, Paetkau et al. 1997, Smouse and Chevillon 1998). If there were no genetic divergence among the 12 collections, one would expect a successful assignment rate of 1/12 (8%, ignoring continental divergence) or 1/6 (17%, comparing only within a continent). There is no overlap between North American and European populations (Table 2), an observation that is in keeping with the large among-continent

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine variance components in Table 1.8 More interesting here is that the observed correct assignment rates within a continent range from 62% to 100%, a telling demonstration that the provincial and state collections (in North America) and the national collections (in Europe) are strongly separable among themselves (Table 2). Although the European national collections (not the focus of this report) are more strongly separated than provincial and state collections in North America, the latter are markedly distinct from each other, despite an extended history of wide movements of Canadian genetic material for restocking purposes in Maine. Watershed-Specific Populations Assignments to watershed for large numbers of Maine salmon are provided in Table 3. The degree of correct assignment varies from 10% for the Pleasant to 84% for the Penobscot rivers, but the average correct assignment rate (49% overall) is consistent with expectations based on experience with other populations and is less than the regional assignment rate. Published sample sizes are smaller than ideal here, and it would be useful to extend this sort of interwatershed assignment analysis to the populations of the Canadian provinces. Despite voluminous and homogeneous stocking for most of these watersheds until 1990, substantial subregional divergence remains. This situation indicates either that a previously existing structure has persisted because of intense selection against hatchery fish or that watershed-specific stocking since 1990, coupled with homing, has allowed reemergence of a subregional structure. Whether the current structure is due to adaptation of distinct populations to divergent selective pressures or due to genetic drift, the degree of regional substructure is startling, particularly in view of the stocking and run-size history of the last several decades. Interestingly, a study of DNA from old scale samples of an endangered Danish Atlantic salmon population (Nielsen et al. 1997) showed that ancestral genotypes can persist despite years’ of extensive stocking. Similarly, Guffey (2000) reported that brook trout (Salvelinus fontinalis) populations in the southern Appalachians were able to retain much of their ancestral southern genotypes despite 35 years of 8   Newfoundland salmon are grouped with North American salmon, despite some probable contribution to their gene pools of European origin. They would not be confused with European salmon.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine TABLE 2 Numbers of Genetic Assignments of Individual Atlantic Salmon to Regional Populations   ME NB NS QB NF LB IC NO FN SC IR SP ME 571 36 9 1 12 5 0 0 0 0 0 0 NB 8 76 7 23 5 3 0 0 0 0 0 0 NS 5 8 88 6 2 0 0 0 0 0 0 0 QB 0 25 5 89 1 2 0 0 0 0 0 0 NF 5 7 5 4 71 1 0 0 0 0 0 0 LB 0 6 1 2 2 32 0 0 0 0 0 0 IC 0 0 0 0 0 0 95 1 0 0 0 0 NO 0 0 0 0 0 0 0 99 0 2 0 0 FN 0 0 0 0 0 0 0 0 61 0 0 0 SC 0 0 0 0 0 0 0 0 0 38 8 7 IR 0 0 0 0 0 0 1 0 0 11 48 4 SO 0 0 0 0 0 0 0 7 0 5 4 68 Percentages Correctly Assigned   89 62 80 73 76 71 99 98 100 72 75 81 Abbreviations: ME, Maine; NB, New Brunswick; NS, Nova Scotia; QB, Quebec; NF, Newfoundland; LB, Labrador; IC, Iceland; NO, Norway; FN, Finland; SC, Scotland; IR, Ireland; SP, Spain. Source: Adapted from King et al. 2001.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine heavy stocking of northern genotypes. These results lend credence to the Atlantic salmon results in Maine. This same assignment can be taken to examine temporal and subwatershed spatial variation in genetic frequencies. If temporal and spatial variations are large enough within a watershed, then the divergence among watersheds becomes less relevant. The results in Table 3 might be due only to uncertainties associated with the small samples within different watersheds. The available data on subwatershed divergence are sparse, but there is a recent study of temporal variation within and among tributaries (subwatersheds) of the Penobscot River (Spidle et al. 2001), and the data are instructive, as far as they go. The basic idea is to treat each collection separately, assign individuals to collections (based on genotype), and then ask where the misclassified individuals should be placed. If substantial numbers of the misclassified individuals are placed in the wrong tributary, that argues against the reality of genetic differences among them and in favor of sampling effects. If they are placed in the wrong cohort but the correct tributary, that argues in favor of the sort of cohort-to-cohort temporal variation expected of an organism with overlapping generations. The results are shown in Table 4. Table 4 shows that a substructure exists even within the Penobscot River,9 bothamong tributaries and among cohorts; genetic differences among populations in different tributaries are a reality. On the other hand, the misclassifications are placed preferentially into other cohorts in the same tributary. That is exactly what one would expect, given the homing behavior of salmon. The yearly cohort samples from a given tributary vary somewhat, but they show more in common with other cohorts from the same tributary than they do with salmon from other tributaries. This striking result has been seen in other anadromous species (Brown et al. 1996), and it argues strongly for thegenetic cohesion of these populations, notwithstanding predictable variation among cohorts (Figure 4) (Jorde and Ryman 1995, Ryman 1997). Evidently, the temporal differences within a tributary are smaller than the average differences among tributaries. The correct assignment rates in Tables 2, 3 and 4 are 9   Note that for the Penobscot that we have 141= (22 +119) assigned to the correct cohort, and 63=(23+40) to the incorrect cohort, an indication of partial separation, but also an indication of considerable overlap. The departure from the 102:102 null expectation is significant, but the gene pools represented by cohorts of the same population are not as different as tributary to tributary differences. The same is true of the Kenduskeag, where the assignment numbers are 41=(16+25) correct and 26 = (12+14) incorrect. The Cove Brook cohort collections are more strikingly different.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine TABLE 3 Numbers of Genetic Assignments of Individual Atlantic Salmon to Watershed-specific Populations   KB SH DT PB NA PL MA EM DE KB 140 1 5 8 2 6 5 14 4 SH 7 9 0 2 0 4 2 6 0 DT 4 1 17 0 4 0 1 3 0 PB 13 1 0 124 1 3 0 5 0 NA 20 4 3 4 38 16 11 19 1 PL 13 2 1 1 2 30 3 7 2 MA 7 2 0 2 5 4 3 3 0 EM 24 4 1 5 7 6 5 41 4 DE 6 0 0 2 0 1 1 3 17 Percentages Correctly Assigned Watershed 74 30 57 84 32 56 10 41 57 Abbreviations: KB, Kennebec; SH, Sheepscot; DT, Duck Trap; PB, Penobscot; NA, Narraguagus; PL, Pleasant; MA, Machias; EM, East Machias; DE, Dennys Source: Adapted from King et al. 1999, Table 18. collectively much larger than random expectation (p < 0.01) by formal chi-square tests.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine TABLE 4 Numbers of Genetic Assignments of Individual Atlantic Salmon to Cohort and Tributary of the Penobscot River   Pb-95 Pb-96 Cv-93 Cv-94 Cv-95 Cv-98 Kd-96 Kd-97 Pb-95 22 23 0 0 1 0 0 3 Pb-96 40 119 0 0 1 0 1 2 Cv-93 0 0 10 2 1 0 0 0 Cv-94 0 0 3 24 3 0 0 0 Cv-95 0 1 0 3 14 0 0 0 Cv-98 0 1 0 2 1 34 0 0 Kd-96 0 3 0 0 0 0 16 12 Kd-97 1 0 0 0 0 0 14 25 Percentages Correctly Assigned Cohort 44 69 67 80 74 90 52 60 Tributary 92 95 92 Note: Numbers assigned to the wrong cohort but within the correct tributary are shown in italics. Abbreviations: Pb, Penobscot mainstem; Cv, Cove Brook; Kd, Kenduskeag Stream. Source: Adapted from Spidle et al. 2001. OVERALL SUMMARY OF GENETIC STUDIES The overall pattern that emerges from all of these studies is that European and North American populations are substantially different. Newfoundland salmon are a partial exception; they appear to be of North American ancestry for the most part, but they show some evidence of European genetic contribution, in keeping with their post-glacial colonization history.

OCR for page 22
Genetic Status of Atlantic Salmon in Maine: Interim Report from the Committee on Atlantic Salmon in Maine The microsatellite DNA data are the most extensive and therefore our inferences are based most strongly on them. The other data are consistent with the microsatellite DNA data, however, which increases our confidence. For allozyme and microsatellite markers, there is broad regional divergence among European (national) populations and somewhat less among North American (provincial or state) populations, the latter having less spatial separation. The RAPD results parallel those for the allozymes and microsatellites to the extent that the sampling frame was comparable. For mtDNA markers, the patterns within Europe and within North America are reversed with respect to the allozyme and microsatellite results, more variation occurring among North American regions than among European regions. The mtDNA markers show about the same watershed-to-watershed divergence as the nuclear markers, but they show less variation within watersheds. Within either continent, genetic similarities are slightly higher in populations from different tributaries within major watersheds than in populations from different major watersheds. Collectively, the results show the same pattern of hierarchical genetic divergence as that shown by the previous extensive allozyme data (May et al. 1994 and references therein) and that shown by the less-extensive protein and DNA analyses of Atlantic salmon from North America (e.g., Verspoor 1986, Ståhl 1987, Bermingham et al. 1991, King et al. 1993, Schill and Walker 1994, Kornfield 1994, Taggart et al. 1995, McConnell et al. 1997, King et al. 2000). The results show large divergence between continental (North American and European) populations; broad regional divergence on both sides of the Atlantic, European national populations being about twice as divergent as North American provincial and state populations; and comparable interwatershed divergence within Maine and within the Canadian provinces. In addition, intertributary divergence in a major watershed (e.g., the Penobscot River) is sometimes substantial, with predictable temporal variation within a given sampling locality. This is the typical pattern seen in salmon and their relatives (Ryman 1983, Ståhl 1987, Allendorf and Waples 1996).