Global climate change is expected to trigger northward shifts in the ranges of natural populations of plants and animals, with subsequent effects on intraspecific genetic diversity. Investigating how genetic diversity is patterned among populations that arose following the last Ice Age is a promising method for understanding the potential future effects of climate change. Theoretical and empirical work has suggested that overall genetic diversity can decrease in colonial populations following rapid expansion into postglacial landscapes, with potential negative effects on the ability of populations to adapt to new environmental regimes. The crucial measure of this genetic variation and a population's overall adaptability is the heritable variation in phenotypic traits, as it is this variation that mediates the rate and direction of a population's multigenerational response to selection. Using two large full-sib quantitative genetic studies (NManitoba = 144; NSouth Dakota = 653) and a smaller phenotypic analysis from Kansas (NKansas = 44), we compared mean levels of pigmentation, genetic variation and heritability in three pigmentation traits among populations of the common garter snake, Thamnophis sirtalis, along a north-south gradient, including a postglacial northern population and a putative southern refuge population. Counter to our expectations, we found that genetic variance and heritability for the three pigmentation traits were the same or higher in the postglacial population than in the southern population.
Citation: Westphal MF, Massie JL, Bronkema JM, Smith BE, Morgan TJ (2011) Heritable Variation in Garter Snake Color Patterns in Postglacial Populations. PLoS ONE 6(9): e24199. https://doi.org/10.1371/journal.pone.0024199
Editor: Brock Fenton, University of Western Ontario, Canada
Received: April 11, 2011; Accepted: August 3, 2011; Published: September 14, 2011
Copyright: © 2011 Westphal et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the U.S. Department of Education through a Graduate Assistance in Areas of National Need GAANN) fellowship to MFW, by the Kansas State Ecological Genomics Institute through a postdoctoral fellowship to MFW and a research seed grant to TJM, and by the KSU NSF Site-based REU program summer award to JMB. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
One of the primary challenges facing contemporary ecological and evolutionary research is predicting the potential effects of global climate change on populations across broad latitudinal ranges. Of recent concern has been the theoretical prediction that populations expanding into northern climates will experience a rapid loss of genetic diversity –, which may in turn restrict the ability of populations to adapt to new selection regimes and novel stressors –. The current global warming event is comparable to past episodes of glacial retreat . Specifically, the rate of air temperature increase in the early Holocene and the concomitant deglaciation are thought to be similar to current trends on the 100-year scale , . Given these similarities, a productive natural experiment for exploring the genetic consequences of rapid colonization after a global warming event is to survey the current genetic makeup of populations that have undergone a rapid expansion following the retreat of glacial ice sheets 10,000 to 12,000 years ago , .
The primary hypothesis for how genetic diversity is patterned across the postglacial landscape states that populations founded after a postglacial expansion should have less genetic diversity as a result of population bottlenecks and/or new selection pressures, resulting in reduced effective population sizes and thus less genetic diversity relative to their historical populations outside the glacial barrier , . Many studies have evaluated this hypothesis by comparing genetic diversity along latitudinal gradients spanning the glacial boundaries for various species using neutral genetic markers, including allozymes , , , , , mitochondrial markers , , , and microsatellite loci , , . Overall, these neutral marker studies have supported the hypothesis that a loss of genetic diversity is associated with the expansion of populations in a postglacial landscape.
Although these studies provide a compelling picture of a population's genetic diversity following postglacial colonization, the true metric of a population's adaptive ability is not neutral genetic diversity per se, but rather heritable genetic variation. This is because heritable genetic variation mediates the rate and direction of population-scale responses to selection –. The few studies that have quantified heritable variation within and among postglacial populations have suggested that heritability of ecologically relevant traits may increase, rather than decrease, following rapid population expansion , . Hypotheses to explain such an increase in heritability following postglacial colonization include: increased effects of dominance in small colonizing populations , ; the admixing of colonists from different refugia populations ; as well as the potential for directional selection to break down epistasis , ,  and expose heritable variation.
To better understand the consequences of postglacial colonization on heritable genetic variation, we quantified levels of genetic variation in pigmentation traits  from two populations of common garter snake, Thamnophis sirtalis, in South Dakota, United States and Manitoba, Canada. The Manitoba population is located in the Interlake near the town of Vogar. The South Dakota population is located at Lake Traverse on the South Dakota-Minnesota border, approximately 600 km due south of the Manitoba site, and represents the southern extremis of the Laurentide ice sheet, which retreated approximately 12,000 years before present . To assess potential clinal variation we analyzed phenotypic data alone from a smaller dataset from a population in Douglas County in eastern Kansas approximately 500 km due south of the South Dakota population. Pathways of garter snake postglacial migration in the region reconstructed from patterns of mitochondrial variation suggest that northern populations likely originated from populations directly to the south .
We focused on pigmentation phenotypes for our study because garter snake color traits are scored easily and are thought to be of adaptive importance in natural populations , . Additionally, molecular phylogenic analysis of T. sirtalis has found a lack of concordance between neutral molecular trees and those based on color pattern, further suggesting a role for selection in pigmentation evolution . Garter snake color patterns are important for predator avoidance, thermoregulation, and other important ecological functions , . In this study, we examined two correlated pigmentation traits and a composite measure of the phenotypes. The first trait is the average area of dorsolateral blotches on individual snakes; the second trait is the pigmentation area of the dorsolateral blotch (the area covered by red pigment), while the composite trait is the ratio of pigmentation area of the blotch to the average area of the dorsolateral blotches.
We chose Thamnophis sirtalis because it is one of only a few reptile species to colonize extreme northern latitudes ; it has been the subject of extensive ecological, physiological, and genetic research , –; and because garter snakes have proven to be tractable organisms in quantitative genetic studies , –. Additionally, prior molecular studies (allozyme and mitochondrial) have documented the expected low genetic diversity in northern populations of T. sirtalis , , , , e.g. that all Manitoba populations have only a single mitochondrial haplotype at the cytochrome b locus .
The primary question we address within this multiple population framework is: Does heritability of adaptive pigmentation traits in the common garter snake decline at northern latitudes? The results of our study suggest that heritability of the three pigment traits is not significantly lower in the postglacial population. The data presented herein further shed light on the effect of rapid colonization on heritable variation in natural populations.
Pearson product-moment correlation coefficients for between-observer scores for this sample were 0.926 (P<0.0001) for blotch area and 0.910 (P<0.0001) for pigment area, both of which are as high, or higher, than repeatability values considered acceptable in previous quantitative genetic studies of garter snakes .
Adult phenotypic data was collected from both male (n = 139) and female (n = 44) snakes at the Manitoba site. At the South Dakota site, females were far more numerous than males on the date of collection and the sample was therefore skewed towards females (51 females, 5 males). The Kansas sample had approximately equal representation of sexes (24 females, 20 males). Blotch area and pigment area were normally distributed for all populations. Distribution of the ratio values tended to be skewed to the right, but were nonetheless amenable to testing given the procedures used, which are generally robust to minor deviations from normality. Two-way ANOVAs with main effects of sex and population revealed no differences between sexes for blotch area (F1,559 = 0.77, P = 0.38), pigment area (F1,561 = 0.18, P = 0.68) or ratio (F1,560 = 2.44, P = 0.12). Furthermore, no sex-by-population effect was detected for any trait (blotch area F1,559 = 0.61, P = 0.43; pigment area F1,561 = 1.79, P = 0.18; ratio F1,560 = 1.72, P = 0.19). Because we saw no difference in trait means between sexes, males and females were pooled in all subsequent analyses. All phenotypic color estimates are in units of blotch-to-scale area, i.e. the proportion of integument that was white or red relative to the immediately adjacent dermal scute.
All trait means for adults were significantly higher in the South Dakota population relative to the Manitoba population (P<0.001, Tukey). However, the Kansas population had higher means for both blotch area and pigment area than the South Dakota sample (P<0.001, Tukey) but the composite ratio of the two pigmentation measures was significantly smaller in the Kansas population (P<0.001, Tukey) relative to the South Dakota population and not significantly different from the Manitoba population (P = 0.97, Tukey).
We generated a second set of phenotypic data from 144 offspring from 25 dams originating at the Manitoba site and 653 offspring from 50 dams from the South Dakota site. For the neonate snakes all pigmentation phenotypes were smaller relative to those measured on the adults (Fig. 1). Again, all means in the South Dakota population were significantly higher relative to the Manitoba population (P<0.001, Tukey). Kansas neonates had significantly higher blotch area (P<0.001, Tukey) and pigment area than the Manitoba population (P<0.02, Tukey) but the blotch area/pigment area ratios were not significantly different (P = 0.72, Tukey). Compared to the South Dakota population, Kansas neonates had significantly higher blotch area (P<0.001, Tukey) but pigment area was not significantly different (P = 0.77, Tukey) nor was the blotch area/pigment area ratio (P = 0.21, Tukey).
MB = Manitoba, SD = South Dakota, KS = Kansas. Left-hand charts show means for blotch area and pigment area. Right-hand charts give means for blotch area/pigment area ratio. The upper panels are for neonates and the bottom panels are for adults. Error bars represent plus or minus one standard error of the mean. All means were significantly different from one another at or below the P = 0.0001 level except for the KS-MB pigment comparison for pigment (P<0.0003) and the KS-MB ratio comparison for adults (P = 0.0997) and neonates (P = 0.83) as well as two SD-KS neonate comparisons: pigment area (P = 0.88) and the blotch area/pigment area ratio (P = 0.75).
Genetic variances and heritabilities
We obtained genetic variance and heritability data from two populations: 144 offspring from 25 dams originating at the Manitoba site, and 653 offspring from 50 dams from the South Dakota site. Because only small numbers of litters were available in the Kansas population, we were unable to generate usable estimates of the genetic variances and heritabilities for this population. Genetic variances and heritabilities were estimated using a full-sib ANOVA on the neonate offspring from each population. The genetic variance for blotch area was 409.62±131.64 s.e. for Manitoba and 360.07±80.96 s.e. for South Dakota; genetic variance for pigment area was 284.44±129.69 s.e. for Manitoba and 162.35±41.64 s.e. for South Dakota; genetic variance for blotch area/pigment area ratio was 0.0466±0.021 s.e. for Manitoba and 0.0067±0.002 s.e. for South Dakota (Fig. 2). Heritabilities for blotch area were 0.72±0.14 s.e. for the Manitoba population and 0.46±0.085 s.e. for the South Dakota population, while heritabilities for pigment area were 0.69±0.26 s.e. for the Manitoba population and 0.41±0.087 s.e. for the South Dakota population (Fig. 2). Heritabilities for the blotch area/pigment area ratio were 0.75±0.28 s.e. for the Manitoba population and 0.25±0.07 s.e. for the South Dakota population.
A. Genetic variances of blotch area and pigment area. B. Genetic variance of blotch area/pigment area ratio. C. Heritabilities of blotch area, pigment area, and blotch area/pigment area ratio. Error bars represent plus or minus one standard error of the variance as estimated by from 10,000 bootstrap replicates.
A common issue in the comparison of genetic variance estimates among populations using a quantitative genetic design is achieving statistical precision that allows the genetic variance and heritability to be bounded away from zero, and also allows those parameters to be bounded away from the population to which it is being compared . Our design is robust in that we have multiple litters with hundreds of individuals from each population, thus for most of the within-population genetic estimates the 95% confidence intervals are easily bounded away from zero. However, we are unable to bound the estimates from each population away from one another based on their 95% confidence intervals (Table 1). Although within-population significance testing of quantitative genetic parameters is feasible in wild populations, comparing the same parameters among populations is difficult given the extremely large numbers of litters required , . The fact that our estimates overlap could therefore be a consequence of a lack of statistical power, or could reflect areal-world equivalence between the populations. That said, given the robustness of the within-population estimates, our data still strongly contradicts a reduction in genetic variance in the recently colonized Manitoba population relative to the South Dakota population (Fig. 2 and Table 1).
In the present study we evaluate levels of heritable variation in pigmentation between two geographically distinct populations of T. sirtalis on either side of the boundary of the Laurentide ice sheet, which retreated approximately 12,000 years ago . We also examine a third population in less depth to gain insight into potential latitudinal trends in trait divergence. We explore the hypothesis that postglacial populations should exhibit reduced levels of genetic variation as a result of population bottlenecks and/or novel selection pressures on these populations relative to the historical populations outside the glacial boundary , . We found two interesting patterns from this among-population study. First, there is no evidence of a decline in heritable variation for the pigmentation traits in question in the northern population. Rather, we found heritabilities to be equal in the northern population relative to the southern population (Fig. 2). Secondly, we detect a significant geographic pattern in size and coloration of the dorsolateral blotches of both neonate and adult snakes moving from southern to northern populations (Fig. 1).
Phenotypes that are adaptive and contribute to responses to environmental stresses that vary with latitude often display clinal patterns in nature . We found that dorsolateral pigmentation traits (i.e., blotch area and pigment area) in T. sirtalis also display patterns of among-population variation consistent with a phenotypic cline (Fig. 1). That is, southern populations express significantly larger mean dorsolateral blotch area compared to populations to the north (i.e., Kansas blotch area>South Dakota blotch area>Manitoba blotch area; Fig. 1). Furthermore, this pattern is robust regardless of age (i.e., neonate or adult), which is consistent with the single population ontogenetic effects demonstrated by . Although our data do not allow us to directly identify the mechanism driving this latitudinal pattern, a likely candidate variable is the difference in the thermal regimes experienced by southern to northern populations. Dorsolateral blotches are thought to function in both thermoregulation and predator startle response behavior , . Reduction in blotch area results in a concomitant increase in the area of the integument covered by melanophores, and would be expected to increase thermoregulatory efficiency , an important adaptive trait in an ectotherm living in the extreme north. Because reduction in blotch area might be maladaptive with regards to predation, and because high expression of red pigment might be maladaptive with respect to thermoregulation, an evolutionary tradeoff may be playing a role in the maintenance of standing variation in pigmentation. The extent to which dorsolateral blotching mediates an evolutionary tradeoff between predator avoidance and thermoregulation deserves further study.
In addition to the pattern observed in the mean phenotypes among populations we also evaluated how heritable genetic variation is patterned amongst the South Dakota population on the southern extremis of the Laurentide ice sheet, and the Manitoba population. We found heritability to be equal in the northern population relative to the southern population. There are multiple hypotheses to explain this pattern of increased variation in postglacial populations, including population admixture or a modification of the genetic architectures underlying the traits after colonization.
In the population admixture hypothesis, genetic variation is increased as a result of genetic variation being introduced from colonists from multiple separate glacial refugia . The populations in question are located in a broad zone of contact between wholly red-pigmented subspecies in the west and wholly white-pigmented in the east. In Manitoba, the zone of contact roughly coincides with the boundaries of glacial Lake Agassiz, which blanketed the northern Midwest during the glacial retreat from 12,000 to 8,000 years before present . Even as the Laurentide ice sheet melted away, Lake Agassiz would have nonetheless precluded colonization of terrestrial sites. It is possible that garter snakes colonized ever northward along the east and west shores of the lake from their respective sides of the contact zone as the ice sheet retreated. Lake Agassiz drained over a very short time period about 8,000 years before present, at which time snakes might have rapidly colonized the new landscape from both the east and west, resulting in a patchwork of new populations representing different frequencies of colonists.
We know of two sources of evidence that argue against the population admixture hypothesis. Rye explored the hypothesis that Manitoba populations were hybrid products of eastern and western clades . Her mitochondrial data suggest that Manitoba snakes are allied with western groups and show no affinity with eastern clades. Moreover, she delineated a mitochondrial contact zone 800 kilometers to the east of the Vogar den, suggesting that admixture with eastern populations was unlikely. Additionally, Westphal and Morgan found a developmental basis for the white/red color variation found at the Manitoba site . Manitoba T. sirtalis were found to express significantly more white blotch area (relative to pigmented blotch area) at birth than adults from the same population, and furthermore were found to increase in red pigmentation during their early development. The South Dakota and Kansas neonates were also found to have reduced pigmentation relative to the adult samples from their respective populations. Although suggestive, neither the Rye  or Westphal and Morgan  results absolutely refute admixture as a root cause of the increase in heritability. Until further molecular work clarifies the historical dynamics of the contact zone between western and eastern clades of T. sirtalis, we cannot completely rule out the admixture hypothesis.
An alternative explanation for increased genetic variation in postglacial populations is the rearrangement of genetic architectures in postglacial populations . Under this scenario, populations that show low heritabilities as a result of stabilizing selection in the source population become subject to drift and founder effects in the colonial population, which in turn reduces epistasis, increases dominance effects, and ultimately exposes variability that was masked in the source population. The founder effects combined with novel selection regimes can also increase genetic variability in the short term by favoring new allelic combinations ,  and thus inflating heritable variation.
Where the strength of selection itself is reduced, heritabilities can be higher than under more stringent selection regimes. For example, documented correlations between quality of environment and heritability suggest that heritabilities are lower in poor environments, e.g., where selection is stronger . We have therefore at least two competing selection-based hypotheses for the increased heritability of the traits in question. 1. Reduced selection. Given the paucity of competing reptile species, it may be that T. sirtalis in Manitoba has experienced a more favorable environment since colonization, with a concomitant decrease in natural selection and increase in heritability. 2. New selection pressures. New selection regimes in a novel environment have exposed hidden variation or favored new allelic combinations.
Although genetic diversity is expected to decrease with distance from the founding population during a stepping-stone colonization event, other factors can rescue genetic diversity. For example, high gene flow between neighboring populations can maintain allelic diversity . In addition, high environmental heterogeneity can reduce genetic diversity in a colonial deme . Because the Manitoba landscape is relatively non-heterogeneous (i.e. has virtually no elevational variation and essentially uniform habitat of aspen parkland), the lack of environmental heterogeneity may have contributed to the high heritability we detected. Finally, new mutations or rare alleles can go to high frequency through a founder event called “surfing,” which occurs at the leading edge of a wave of migration . Due to the rapidity with which the postglacial expansion likely occurred, surfing is a valid hypothesis for the higher heritability of the Manitoba population. Regardless of the specific mechanism, a complex model of colonization with or without selection is likely necessary to explain the results of the present study.
Although at present we are unable to convincingly support or refute any explanation for an increase in genetic variance of three color traits in the Manitoba population, our results suggest that, although genetic diversity at some loci may be reduced in postglacial populations of Thamnophis sirtalis , heritable variation underlying ecologically relevant phenotypic traits may actually be higher than in the founding populations. Our results are consistent with the few previous studies that have examined heritable variation in postglacial populations , . An important focus of future research will be to obtain robust genetic data from both populations to better assess neutral genetic variation in both regions. Analysis of microsatellite loci for T. sirtalis has been performed in other regions  and is a practical next step. Studies of Fst/Qst can address the role of selection over the postglacial landscape and will be forthcoming from the present authors. Finally, forthcoming fine-scale phylogenetic studies are expected to better resolve the issue of historical admixture in the Manitoba population. Because heritable variation can buffer populations against extirpation by allowing them increased capability to respond to selection events , , it is important to understand long-term effects on heritable variation stemming from rapid range expansion if we are to make informed decisions on preserving biodiversity in the face of global climate change.
Materials and Methods
All work was conducted in strict accordance with the US Public Health Service (PHS) Policy on the Humane Care and Use of Laboratory Animals, the US Department of Agriculture's (USDA) Animal Welfare Act & Regulations (9CFR Chapter 1, 2.31), and the United States Government Principles for the Utilization and Care of Vertebrate Animals Used in Research, Teaching and Testing. The work was approved by the respective Institutional Animal Care and Use Committees (IACUCs) at Oregon State University (approval # 3175); Kansas State University (approval #2676); and Black Hills State University (approval #A-008-004).
25 gravid female snakes were collected from a single hibernaculum near Vogar, Manitoba (Latitude: N 50° 56.934, Longitude: W 98° 34.572). A sample of 173 adult snakes (139 male, 44 female) was also collected at the Manitoba site for phenotypic scoring. 50 gravid females were collected from a single hibernaculum on the shore of Lake Traverse near Sisseton, South Dakota (Latitude: N 45° 40.528, Longitude: W 96° 43.747). Two additional gravid females were collected from a single location in Douglas County, Kansas (Latitude: N 38° 56.667, Longitude: W 95° 4.829) (see map, Fig. 3), as well as a sample of 44 adult snakes (20 males, 24 females) for phenotypic scoring. Gravid snakes were placed in breathable cloth sacks and transported to husbandry facilities. The Vogar, Manitoba sample was collected in May 2003; the Sisseton, South Dakota and Douglas, Kansas samples were collected in June 2008. We acknowledge that the temporal gap in the sampling of each population is not ideal. However, multiple years of empirical observations by MFW suggest temporal stability of the phenotypic means and variances at the Vogar, Manitoba site. Collection times at all three sites coincided with the advent of breeding season prior to dispersal to foraging habitat. Moreover, seasonal effects on these traits are likely limited, because unlike other squamates, snakes do not exhibit seasonal changes in coloration . Due to the large numbers of snakes available at the Manitoba sites, specimens there were collected by grab sampling of large groups of snakes that were either basking or in mating aggregations. The South Dakota snakes were collected as they emerged from the den. The Kansas snakes were collected under cover boards within 100 m of each other during a one-day visit.
Female snakes were housed in dedicated facilities at Oregon State University (Manitoba sample) and Black Hills State University (South Dakota sample) and were reared under identical environmental and feeding regimes . The two Kansas litters were obtained from wild-caught dams, which were reared to parturition by a private breeder under the close supervision of MFW. Prior research on the color traits analyzed below demonstrated that individuals born and reared in a common garden laboratory environment express coloration values consistent with samples from the populations of origin . This concordance strongly indicates that the traits in question are stable with regard to the modest environmental variation that might be present among rearing facilities . Therefore, given the similar environmental and feeding regimes at the three facilities, confounding environmental effects were not expected to arise from minor variations in rearing conditions. Nonetheless, it is widely acknowledged that heritability and genetic variance are environmentally dependent . Thus we cannot definitively rule out environmental effects of breeding facility on our population comparisons, but the magnitude of such effects are likely small enough to permit such comparisons. Gravid females were retained until their litters were born. Within three days of parturition, neonates were weighed, measured snout-to-vent, and scored for color traits associated with the distinctive white and red blotches expressed in the dorsolateral region as in .
We used a standardized scoring system, described in . The system quantifies both the average size of dorsolateral blotches and the extent to which red pigment is present in the otherwise white blotch. The final dataset is composed of two variables; blotch area, which measures the total area of blotch (irrespective of pigment type) while the pigment area measures the total area of blotch that was pigmented red. All phenotypic color estimates are in units of blotch-to-scale area, i.e. the proportion of integument that was white or red relative to the immediately adjacent dermal scute (Figure S1). Following  we calculated a composite trait, which was the ratio of pigmentation area to blotch area. We included this trait in the analysis because the two component traits are genetically correlated and the composite trait captures the phenotypic reality of the relative extent to which red pigmentation is expressed across the dorsolateral blotches. That is, for this composite trait individuals with a score of 1 have blotches that are completely red, while individuals with a score of 0 have blotches that are completely white. Color measurements, as well as snout-vent length, mass, and sex, were taken on individual neonates from each litter from all populations as well as representative wild caught adult snakes from each population. The timing of scoring in the present study corresponds to the first scoring period in . Therefore the genetic variance of the traits was not expected to change appreciably within or among population samples across the 3-day scoring window.
Two personnel were involved in scoring the pigmentation traits; therefore we conducted between-observer repeatability tests using 50 adult snakes from the South Dakota population. 50 adult snakes from the South Dakota population were scored independently by MFW and JM. Scorings were separated by a period of several months, but both measurements occurred during the stable portion of the pigmentation ontogeny that occurs when snakes are adults. Pearson correlation coefficients were calculated in SAS v 9.2 using the PROC CORR procedure to assess the repeatability of scoring between observers.
Trait means and standard errors for each population were obtained using PROC MEANS in SAS v 9.2. Trait means among populations were compared by a one-way ANOVA using PROC GLM in SAS v 9.2, followed by multiple comparisons. For the quantitative genetic analysis, we analyzed litter data from only the populations with a large number of litters, that is the Manitoba (25 litters and 144 offspring) and South Dakota populations (50 litters and 653 offspring). Population-specific genetic estimates (i.e. genetic variance and heritabilities) and their associated standard errors were calculated using h2boot . Because neonates express pigment differently than adults  we used an age-specific full-sib ANOVA to estimate the genetic parameters, rather than parent-offspring regression. Estimates of genetic variance and heritability were therefore estimated for only neonates and not adults. We used 10,000 bootstrap replicates to estimate standard errors of the genetic parameters . The full-sib model as implemented in h2boot can be found in . Full-sib analysis is subject to overestimations of heritability due to the potential effects of common environment . As discussed above, a previous common garden study of the same traits from radically disjunct populations (Manitoba, CA and California, USA) found them to exhibit strong environmental stability, suggesting maternal and other environmental effects were relatively minor . Nonetheless we again cannot rule out the possibility that environmental factors contributed to our genetic estimates, but consider such effects unlikely to be so profound as to negate our results.
System for scoring size and pigment saturation of dorsolateral blotches in T. sirtalis. Images on left show among-individual variation in both the size of blotches and the extent to which red pigment saturates the blotches. Images on right focus on a portion of a blotch at one scale row. Top image receives a pigment area score of “0,” middle image receives a score of “0.5,” and bottom image receives a score of “1.” Schematic drawing at bottom depicts how blotch length is measured. Adjacent scale is assigned a length of 10 units, and the blotch is assigned a length relative to the adjacent scale in intervals of 1 unit. In the depicted example, the blotch at that scale row would receive a score of 7. Blotch widths at all scale rows are summed over three adjacent blotches to give individual blotch area score. Pigment area at each scale row is obtained by multiplying the pigment score by the blotch length. Pigment scores at each scale row are then summed over the same three adjacent blotches to give the individual pigment area score.
We would like to thank S. J. Arnold for making his snake rearing facility freely available. R. T. Mason and R. Shine gave invaluable advice. We also thank J. B. Westphal, B. Blake and Dan Krull for field collection and snake rearing assistance.
Conceived and designed the experiments: MFW JLM BES TJM. Performed the experiments: MFW JLM JMB BES. Analyzed the data: MFW JLM BES TJM. Contributed reagents/materials/analysis tools: MFW JLM BES TJM. Wrote the paper: MFW TJM.
- 1. Sage RD, Wolf JO (1986) Pleistocene glaciations, fluctuating ranges, and low genetic variability in a large mammal (Ovis dalli). Evolution 40: 1092–1095.RD SageJO Wolf1986Pleistocene glaciations, fluctuating ranges, and low genetic variability in a large mammal (Ovis dalli).Evolution4010921095
- 2. Taberlet P, Cheddadi R (2002) Quaternary refugia and persistence of biodiversity. Science 297: 2009–2010.P. TaberletR. Cheddadi2002Quaternary refugia and persistence of biodiversity.Science29720092010
- 3. Hewitt GM (1996) Some genetic consequences of ice ages, and their role in divergence and speciation. Biological Journal of the Linnean Society 58: 247–276.GM Hewitt1996Some genetic consequences of ice ages, and their role in divergence and speciation.Biological Journal of the Linnean Society58247276
- 4. Green DM, Sharbel TF, Kearsley J, Kaiser H (1996) Postglacial range fluctuation, genetic subdivision and speciation in the western North American spotted frog complex, Rana pretiosa. Evolution 50: 374–390.DM GreenTF SharbelJ. KearsleyH. Kaiser1996Postglacial range fluctuation, genetic subdivision and speciation in the western North American spotted frog complex, Rana pretiosa.Evolution50374390
- 5. Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature 405: 907–913.GM Hewitt2000The genetic legacy of the Quaternary ice ages.Nature405907913
- 6. Hewitt GM (2004) Genetic consequences of climatic oscillations in the Quaternary. Proceedings of the Royal Society of London Series B, Biological Sciences 359: 183–195.GM Hewitt2004Genetic consequences of climatic oscillations in the Quaternary.Proceedings of the Royal Society of London Series B, Biological Sciences359183195
- 7. Mahy G, Ennos RA, Jacquemart A (1999) Allozyme variation and genetic structure of Calluna vulgaris (heather) populations in Scotland: the effect of postglacial colonization. Heredity 82: 654–660.G. MahyRA EnnosA. Jacquemart1999Allozyme variation and genetic structure of Calluna vulgaris (heather) populations in Scotland: the effect of postglacial colonization.Heredity82654660
- 8. Pamilo P, Savolainen O (1999) Post-glacial colonization, drift, local selection and conservation value of populations: a northern perspective. Hereditas 130: 229–238.P. PamiloO. Savolainen1999Post-glacial colonization, drift, local selection and conservation value of populations: a northern perspective.Hereditas130229238
- 9. Wilson AJ, Pemberton JM, Pilkington JG, Coltman DW, Mifsud DV, et al. (2006) Environmental coupling of selection and heritability limits evolution. PloS Biology 4: 1270–1275.AJ WilsonJM PembertonJG PilkingtonDW ColtmanDV Mifsud2006Environmental coupling of selection and heritability limits evolution.PloS Biology412701275
- 10. Ficetola GF, Garner TWJ, De Bernardi F (2007) Genetic diversity, but not hatching success, is jointly affected by post glacial colonization and isolation in the threatened frog, Rana latastei. Mol Ecol 16: 1787–1797.GF FicetolaTWJ GarnerF. De Bernardi2007Genetic diversity, but not hatching success, is jointly affected by post glacial colonization and isolation in the threatened frog, Rana latastei.Mol Ecol1617871797
- 11. COHMAP (Cooperative Holocene Mapping Project) (1988) Climatic changes of the last 18,000 years: observations and model simulations. Science 241: 1043–1052.COHMAP (Cooperative Holocene Mapping Project)1988Climatic changes of the last 18,000 years: observations and model simulations.Science24110431052
- 12. Bluemle JP, Sabel JM, Karlén W (1999) Rate and magnitude of past global climate changes. Environ Geosci 6: 63–75.JP BluemleJM SabelW. Karlén1999Rate and magnitude of past global climate changes.Environ Geosci66375
- 13. Carlson AE, Legrande AN, Oppo DW, Came RE, Schmidt GA, Anslow FS, Licciardi JM, Obbink EA (2008) Rapid early Holocene deglaciation of the Laurentide ice sheet. Nature Geoscience 1: 620–624.AE CarlsonAN LegrandeDW OppoRE CameGA SchmidtFS AnslowJM LicciardiEA Obbink2008Rapid early Holocene deglaciation of the Laurentide ice sheet.Nature Geoscience1620624
- 14. Pielou EC (1991) After the Ice Age: The Return of Life to Glaciated North America. EC Pielou1991After the Ice Age: The Return of Life to Glaciated North America.University Of Chicago Press, Chicago, IL. University Of Chicago Press, Chicago, IL.
- 15. Lesbarrères D (2009) Post-glacial phylogeography: New insight into an old story: the post-glacial recolonization of European biota. Heredity 102: 213.D. Lesbarrères2009Post-glacial phylogeography: New insight into an old story: the post-glacial recolonization of European biota.Heredity102213
- 16. Ray N, Currat M, Excoffier L (2003) Intra-deme molecular diversity in spatially expanding populations. Mol Biol Evol 20: 76–86.N. RayM. CurratL. Excoffier2003Intra-deme molecular diversity in spatially expanding populations.Mol Biol Evol207686
- 17. Wegmann D, Currat M, Excoffier L (2006) Molecular diversity after a range expansion in heterogeneous environments. Genetics 174: 2009–2020.D. WegmannM. CurratL. Excoffier2006Molecular diversity after a range expansion in heterogeneous environments.Genetics17420092020
- 18. Merilä J, Bjorklund M, Bkaer AJ (1996) Genetic population structure and gradual northward decline of genetic variability in the greenfinch (Carduelis chloris). Evolution 50: 2548–2557.J. MeriläM. BjorklundAJ Bkaer1996Genetic population structure and gradual northward decline of genetic variability in the greenfinch (Carduelis chloris).Evolution5025482557
- 19. Alexandrino J, Froufe E, Arntzen JW, Ferrand N (2000) Genetic subdivision, glacial refugia and postglacial recolonization in the golden-striped salamander, Chioglossa lusitanica (Amphibia: Urodela). Mol Ecol 9: 771–781.J. AlexandrinoE. FroufeJW ArntzenN. Ferrand2000Genetic subdivision, glacial refugia and postglacial recolonization in the golden-striped salamander, Chioglossa lusitanica (Amphibia: Urodela).Mol Ecol9771781
- 20. Comps B, Gömöry D, Letouzey J, Thiébaut B, Petit RJ (2001) Diverging trends between heterozygosity and allelic richness during postglacial colonization in the European beech. Genetics 157: 389–97.B. CompsD. GömöryJ. LetouzeyB. ThiébautRJ Petit2001Diverging trends between heterozygosity and allelic richness during postglacial colonization in the European beech.Genetics15738997
- 21. Tollefsrud MM, Kissling R, Gugerli F, Johnsen Ø, Skrøppa T, et al. (2008) Genetic consequences of glacial survival and postglacial colonization in Norway spruce: combined analysis of mitochondrial DNA and fossil pollen. Mol Ecol 17: 4134–4150.MM TollefsrudR. KisslingF. GugerliØ. JohnsenT. Skrøppa2008Genetic consequences of glacial survival and postglacial colonization in Norway spruce: combined analysis of mitochondrial DNA and fossil pollen.Mol Ecol1741344150
- 22. Knopp T, Merilä J (2008) The postglacial recolonization of Northern Europe by Rana arvalis as revealed by microsatellite and mitochondrial DNA analyses. Heredity 102: 174–181.T. KnoppJ. Merilä2008The postglacial recolonization of Northern Europe by Rana arvalis as revealed by microsatellite and mitochondrial DNA analyses.Heredity102174181
- 23. Stenøien HK, Fenster CB, Tonteri A, Savolainen O (2005) Genetic variability in natural populations of Arabidopsis thaliana from northern Europe. Mol Ecol 14: 137–148.HK StenøienCB FensterA. TonteriO. Savolainen2005Genetic variability in natural populations of Arabidopsis thaliana from northern Europe.Mol Ecol14137148
- 24. Lande R (1979) Quantitative genetic analysis of multivariate evolution, applied to brain:body size allometry. Evolution 33: 402–416.R. Lande1979Quantitative genetic analysis of multivariate evolution, applied to brain:body size allometry.Evolution33402416
- 25. Falconer DS, Mackay TFC (1996) Introduction to Quantitative Genetics, 4th ed. DS FalconerTFC Mackay1996Introduction to Quantitative Genetics, 4th ed.Longman, Harlow, UK. Longman, Harlow, UK.
- 26. Lynch M, Walsh B (1998) Genetics and Analysis of Quantitative Traits. M. LynchB. Walsh1998Genetics and Analysis of Quantitative Traits.Sinauer Associates, Inc. Sunderland, MA. Sinauer Associates, Inc. Sunderland, MA.
- 27. Bradshaw WE, Holzapfel CM (2001) Phenotypic evolution and the genetic architecture underlying photoperiodic time measurement. J Insect Physiol 47: 809–820.WE BradshawCM Holzapfel2001Phenotypic evolution and the genetic architecture underlying photoperiodic time measurement.J Insect Physiol47809820
- 28. van Heerwaarden B, Willi Y, Kristensen TN, Hoffmann AA (2008) Population bottlenecks increase additive genetic variance but do not break a selection limit in rain forest Drosophila. Genetics 179: 2135–46.B. van HeerwaardenY. WilliTN KristensenAA Hoffmann2008Population bottlenecks increase additive genetic variance but do not break a selection limit in rain forest Drosophila.Genetics179213546
- 29. Shine R, Phillips B, Waye H, Lemaster M, Mason RT (2004) Species-isolating mechanisms in a mating system with male mate choice. Can J Zool 82: 1091–1098.R. ShineB. PhillipsH. WayeM. LemasterRT Mason2004Species-isolating mechanisms in a mating system with male mate choice.Can J Zool8210911098
- 30. Andrews JT (1987) The late Wisconsin glaciation and deglaciation of the Laurentide ice sheet. In: Ruddiman WF, Wright HE Jr, editors. The geology of North America. North America and adjacent oceans during the last deglaciation. Geological Society of America, Boulder, CO. pp. 13–37.JT Andrews1987The late Wisconsin glaciation and deglaciation of the Laurentide ice sheet.WF RuddimanHE Wright JrThe geology of North America. North America and adjacent oceans during the last deglaciationGeological Society of America, Boulder, CO1337
- 31. Placyk JS, Burghardt GM, Small RL, King RB, Casper GS, et al. (2007) Post-glacial recolonization of the Great Lakes region by the common gartersnake (Thamnophis sirtalis) inferred from mtDNA sequences. Mol Phylo Evol 2007, 43: 452–467.JS PlacykGM BurghardtRL SmallRB KingGS Casper2007Post-glacial recolonization of the Great Lakes region by the common gartersnake (Thamnophis sirtalis) inferred from mtDNA sequences.Mol Phylo Evol2007, 43452467
- 32. Brodie ED III, Garland T Jr (1993) Quantitative genetics of snake populations. In: Seigel RA, Collins JT, editors. Snakes: ecology and behavior. McGraw-Hill, New York. pp. 315–362.ED Brodie IIIT. Garland Jr1993Quantitative genetics of snake populations.RA SeigelJT CollinsSnakes: ecology and behaviorMcGraw-Hill, New York315362
- 33. McKinnon JS, Pierotti MR (2010) Colour polymorphism and correlated characters: genetic mechanisms and evolution. Mol Ecol. JS McKinnonMR Pierotti2010Colour polymorphism and correlated characters: genetic mechanisms and evolution.Mol Ecol DOI: 10.1111/j.1365-294X.2010.04846.x. DOI: 10.1111/j.1365-294X.2010.04846.x.
- 34. Janzen FJ, Krenz JG, Haselkorn TS, Brodie ED Jr, Brodie ED III (2002) Molecular phylogeny of common garter snakes (Thamnophis sirtalis) in western North America: implications for regional historical forces. Mol Ecol 11: 1739–1751.FJ JanzenJG KrenzTS HaselkornED Brodie JrED Brodie III2002Molecular phylogeny of common garter snakes (Thamnophis sirtalis) in western North America: implications for regional historical forces.Mol Ecol1117391751
- 35. Shine R, Olsson MM, Lemaster MP, Moore IT, Mason RT (2000) Effects of sex, body size, temperature, and location on the antipredator tactics of freeranging gartersnakes (Thamnophis sirtalis, Colubridae). Behav Ecol 11: 239–245.R. ShineMM OlssonMP LemasterIT MooreRT Mason2000Effects of sex, body size, temperature, and location on the antipredator tactics of freeranging gartersnakes (Thamnophis sirtalis, Colubridae).Behav Ecol11239245
- 36. Bittner TD, King RB, Kerfin JM (2002) Effects of body size and melanism on the thermal biology of garter snakes (Thamnophis sirtalis). Copeia 2002: 477–482.TD BittnerRB KingJM Kerfin2002Effects of body size and melanism on the thermal biology of garter snakes (Thamnophis sirtalis).Copeia2002477482
- 37. Conant R (1975) A Field Guide to Reptiles and Amphibians of Eastern and Central North America. R. Conant1975A Field Guide to Reptiles and Amphibians of Eastern and Central North America.Second Edition. Houghton Mifflin Company, Boston, Massachusetts. xvii+429 Second Edition. Houghton Mifflin Company, Boston, Massachusetts.
- 38. Bellemin J, Adest G, Gorman GS, Aleksiuk M (1978) Genetic uniformity in northern populations of Thamnophis sirtalis (Serpentes: Colubridae). Copeia 1978: 150–151.J. BelleminG. AdestGS GormanM. Aleksiuk1978Genetic uniformity in northern populations of Thamnophis sirtalis (Serpentes: Colubridae).Copeia1978150151
- 39. Gregory PT (1977) Life-history parameters of the red-sided garter snake (Thamnophis sirtalis parietalis) in an extreme environment, the Interlake region of Manitoba. Nat Mus Canada Publ Zool 13: 1–44.PT Gregory1977Life-history parameters of the red-sided garter snake (Thamnophis sirtalis parietalis) in an extreme environment, the Interlake region of Manitoba.Nat Mus Canada Publ Zool13144
- 40. Gregory PT, Larsen KW (1993) Geographic variation in reproductive characteristics among Canadian populations of the common garter snake (Thamnophis sirtalis). Copeia 1993: 946–958.PT GregoryKW Larsen1993Geographic variation in reproductive characteristics among Canadian populations of the common garter snake (Thamnophis sirtalis).Copeia1993946958
- 41. Gregory PT, Larsen KW (1996) Are there any meaningful correlates of geographic life-history variation in the garter snake, Thamnophis sirtalis. Copeia 1: 183–189.PT GregoryKW Larsen1996Are there any meaningful correlates of geographic life-history variation in the garter snake, Thamnophis sirtalis.Copeia1183189
- 42. Rossman DA, Ford NB, Seigel RA (1996) The Garter Snakes: Evolution and Ecology. DA RossmanNB FordRA Seigel1996The Garter Snakes: Evolution and Ecology.University of Oklahoma Press, Norman, OK. University of Oklahoma Press, Norman, OK.
- 43. Garland T Jr, Bennett AF (1990) Quantitative genetics of maximal oxygen consumption in a garter snake. Am J Physiol 259(Reg. Integ. Comp. Physiol. 28): R986–R992.T. Garland JrAF Bennett1990Quantitative genetics of maximal oxygen consumption in a garter snake.Am J Physiol259Reg. Integ. Comp. Physiol. 28R986R992
- 44. Brodie ED III (1992) Correlational selection for color pattern and antipredator behavior in the garter snake Thamnophis ordinoides. Evolution 46: 1284–1298.ED Brodie III1992Correlational selection for color pattern and antipredator behavior in the garter snake Thamnophis ordinoides.Evolution4612841298
- 45. Dohm MR, Garland T Jr (1993) Quantitative genetics of scale counts in the garter snake Thamnophis sirtalis. Copeia 4: 987–1002.MR DohmT. Garland Jr1993Quantitative genetics of scale counts in the garter snake Thamnophis sirtalis.Copeia49871002
- 46. Westphal MF, Morgan TJ (2010) Quantitative genetics of pigmentation development in 2 populations of the common garter snake, Thamnophis sirtalis. J Hered 101: 573–580.MF WestphalTJ Morgan2010Quantitative genetics of pigmentation development in 2 populations of the common garter snake, Thamnophis sirtalis.J Hered101573580
- 47. Sattler PW, Guttman SI (1976) An electrophoretic analysis of Thamnophis sirtalis from western Ohio. Copeia 1976: 352–356.PW SattlerSI Guttman1976An electrophoretic analysis of Thamnophis sirtalis from western Ohio.Copeia1976352356
- 48. Rye LA (2000) Analysis of areas of intergradation between described subspecies of the common garter snake, Thamnophis sirtalis, in Canada. LA Rye2000Analysis of areas of intergradation between described subspecies of the common garter snake, Thamnophis sirtalis, in Canada.Unpublished PhD Thesis, University of Guelph, Guelph, ON, Canada. Unpublished PhD Thesis, University of Guelph, Guelph, ON, Canada.
- 49. Morgan TJ, Evans MA, Garland T Jr, Swallow JG, Carter PA (2005) Molecular and quantitative genetic divergence among populations of house mice with known evolutionary histories. Heredity 94: 518–525.TJ MorganMA EvansT. Garland JrJG SwallowPA Carter2005Molecular and quantitative genetic divergence among populations of house mice with known evolutionary histories.Heredity94518525
- 50. Klein TW (1974) Heritability and genetic correlation: Statistical power, population comparisons, and sample size. Behav Gen 4: 171–189.TW Klein1974Heritability and genetic correlation: Statistical power, population comparisons, and sample size.Behav Gen4171189
- 51. Cowley DE, Atchley WR (1992) Comparison of quantitative genetic parameters. Evolution 46: 1965–1967.DE CowleyWR Atchley1992Comparison of quantitative genetic parameters.Evolution4619651967
- 52. Endler JA (1977) Geographic Variation, Speciation, and Clines. Princeton, , NJ: Princeton University Press. JA Endler1977Geographic Variation, Speciation, and ClinesPrinceton, , NJPrinceton University Press
- 53. Boyd M (2007) Early postglacial history of the southeastern Assiniboine Delta, glacial Lake Agassiz basin. J Paleolimnology (Special Issue) 37: 313–329.M. Boyd2007Early postglacial history of the southeastern Assiniboine Delta, glacial Lake Agassiz basin.J Paleolimnology (Special Issue)37313329
- 54. Excoffier L, Foll M, Petit RJ (2009) Genetic consequences of range expansions. Annu Rev Ecol Evol Syst 2009 40: 481–501.L. ExcoffierM. FollRJ Petit2009Genetic consequences of range expansions.Annu Rev Ecol Evol Syst 200940481501
- 55. Manier MK, Arnold SJ (2005) Population genetic analysis identifies source-sink dynamics for two sympatric garter snake species (Thamnophis elegans and T. sirtalis). Mol Ecol 14: 3965–3976.MK ManierSJ Arnold2005Population genetic analysis identifies source-sink dynamics for two sympatric garter snake species (Thamnophis elegans and T. sirtalis).Mol Ecol1439653976
- 56. Lacy RC (1997) The importance of genetic variation to the viability of mammalian populations. J Mammal 78: 320–335.RC Lacy1997The importance of genetic variation to the viability of mammalian populations.J Mammal78320335
- 57. Cooper WE Jr, Greenberg N (1992) Reptile coloration and behavior. In: Gans C, Crews D, editors. Biology of the Reptilia, volume 18. University of Chicago Press. Chicago, Illinois. pp. 298–422.WE Cooper JrN. Greenberg1992Reptile coloration and behavior.C. GansD. CrewsBiology of the Reptilia, volume 18University of Chicago Press. Chicago, Illinois298422
- 58. Phillips PC (1998) H2boot: bootstrap estimates and tests of quantitative genetic data. PC Phillips1998H2boot: bootstrap estimates and tests of quantitative genetic data.Univ. of Oregon. Software available at www.uoregon.edu/~pphil/software.html. Univ. of Oregon. Software available at www.uoregon.edu/~pphil/software.html.