Northeastern Naturalist
506
J. Buckner, A.B. Welsh, and K.R. Sime
22001144 NORTHEASTERN NATURALIST 2V1(o4l). :2510,6 N–5o1. 44
Evidence for Population Differentiation in the Bog
Buckmoth of New York State
Janet Buckner1, Amy B. Welsh2, and Karen R. Sime3,*
Abstract - Hemileuca maia (Bog Buckmoth; Saturniidae) is a rare, ecologically unique
variant of the Hemileuca maia complex known only from ten wetlands in the Great Lakes
region of North America. The Bog Buckmoth’s status as a threatened taxon meriting
conservation has been subject to a debate largely centered on its degree of evolutionary isolation
and species status. We studied the genetic variation of two New York Bog Buckmoth
populations using amplified fragment-length polymorphisms (AFLP). Bayesian clustering
analysis identified two genetically distinct population clusters, with membership that did
not coincide consistently with the two sampled populations. There appears to be either
historical or contemporary gene flow between Bog Buckmoth populations, with the results
suggesting either dispersal between the two sampled populations or contributions from a
third unsampled population. Genetic diversity levels were similar. These findings argue for
the utility of population-level analyses of Bog Buckmoth as a tool in conservation practice
as well as in understanding the taxon’s evolutionary history.
Introduction
Hemileuca maia (Drury) (Bog Buckmoth, Lepidoptera: Saturniidae), known
also as Cryan’s Buckmoth, is a rare variant of the Hemileuca maia complex that
was first discovered in 1977 and has been found in ten peatlands in the Great Lakes
region of North America (Tuskes et al. 1996). Six Bog Buckmoth populations have
been identified along the eastern lake plain of Lake Ontario in Oswego County,
NY, and the remaining populations are in eastern Wisconsin and southern Ontario,
Canada (Fig. 1). In 1999, Bog Buckmoth was listed as threatened under the New
York Endangered Species Act largely because the few fens in which it occurs are
disappearing or becoming unsuitable for its larval food-plant, the wetland herb
Menyanthes trifoliata L. (Bog Buckbean, Menyanthaceae) (Stanton 2004). Habitat
succession appears to be accelerating as a result of nutrient enrichment from
runoff and hydrologic variation attributable to lake-level regulation and alteration
of inflows (COSEWIC 2009, Stanton 2004). Also, invasive wetland plants such
as Phragmites sp. (common reed), Lythrum salicaria L. (Purple Loosestrife), and
Frangula alnus Mill. (Glossy Buckthorn) have displaced Bog Buckbean at some
sites (Bonnano 2008, Gratton 2006). Furthermore, extant Bog Buckmoth population
sizes are small and thus may be more vulnerable to extinction than larger
1Department of Ecology and Evolutionary Biology, University of California Los Angeles,
621 Charles E. Young Drive South, Los Angeles, CA 90095. 2Division of Forestry and
Natural Resources, West Virginia University, PO Box 6125, Morgantown, WV 26506. 3Department
of Biological Sciences, Shineman Hall, State University of New York at Oswego,
Oswego, NY 13126. *Corresponding author - karen.sime@oswego.edu.
Manuscript Editor: Adrienne Kovach
Northeastern Naturalist Vol. 21, No. 4
J. Buckner, A.B. Welsh, and K.R. Sime
2014
507
populations. Most of the Oswego County populations have been monitored annually
since the late 1990s, the results of which reveal patterns of population fluctuation
that include dramatic declines and recoveries, and populations in at least three of
the six known sites are apparently extinct (Bonanno 2013).
The degree to which Bog Buckmoth should be prioritized for conservation efforts
has been challenged by disputes over its species status. Hemileuca maia occurs
across the eastern United States and Great Plains regions, and, with the exception of
the Great Lakes Bog Buckmoth populations, feeds on Quercus spp. (oaks ) and Salix
spp. (willows) in wooded habitats (Tuskes et al. 1996). Although H. maia exhibits
clinal color variations across its broad range, Bog Buckmoth populations are not
consistently distinguishable from other H. maia on the basis of any diagnostic set
of color or other morphological characters (Tuskes et al. 1996). However, Legge
et al. (1996) argued that behavioral and ecological differences between the Bog
Buckmoth and other H. maia may warrant the delineation of the Bog Buckmoth as
an evolutionarily significant unit, if not actually a separate species. These disparities
include the occurrence of the Bog Buckmoth in peatland habitats and its use of Bog
Buckbean as the larval food plant. Other populations of H. maia cannot develop on
Bog Buckbean (Legge et al. 1996). Moreover, Bog Buckmoth exhibits numerous
behavioral adaptations to its wetland habitat, including oviposition on plants other
than the larval food plant and resulting peculiarities of the larval foraging strategies
(Pryor 1998, Tuskes et al. 1996).
Genetic studies have attempted to address the question of whether the Bog
Buckmoth merits classification as a species separate from H. maia. To date, neither
allozymes (Legge et al. 1996) nor mitochondrial DNA (mtDNA) (Rubinoff and
Sperling 2004) have differentiated Bog Buckmoth from other H. maia. Analysis of
the mitochondrial COI gene in specimens from Wisconsin and New York showed
that the Bog Buckmoth populations are paraphyletic with respect to other H. maia,
and thus do not comprise a separate species (Rubinoff and Sperling 2004). More
likely, H. maia, like other widespread Saturniidae, exhibits a variety of rapidly
evolved adaptations to local habitats and food plants (Tuskes et al. 1996). However,
we suggest that small sample sizes, limited sampling across the Great Lakes region,
and reliance on a single gene marker render these findings somew hat inconclusive.
Rubinoff and Sperling (2004) recommended analysis at the population level to
better understand the population genetic structure within H. maia, and our study
was directed towards that end. Our specific objective was to conduct a comparative
genetic study of Bog Buckmoth to assess genetic variation between two Oswego
County populations and the levels of genetic diversity within each population.
Although this approach does not resolve the question of species status, knowledge
of population structure can help delineate management units, gauge evolutionary
potential, and assess the extinction risk of local populations in the absence of recolonization
(Crandall et al. 2000, Frankham et al. 2002, Gompe rt et al. 2006).
Like other Bog Buckmoth populations, the study populations occurred in concentrated
areas within fens that span no more than a few hundred square meters. The two
fens we studied are separated by 30 km of wooded or developed habitat unsuitable for
Northeastern Naturalist
508
J. Buckner, A.B. Welsh, and K.R. Sime
2014 Vol. 21, No. 4
the Bog Buckmoth (Fig. 1; Bonnano 2008, Olivero 2001). The topography of the area
includes gently rolling hills and drumlins up to a few hundred meters high. Although
males fly well, female H. maia are weak fliers, do not feed as adults, are short-lived,
and thus not likely to make long-distance dispersive flights (Collins and Tuskes
1979). Females tend to perch on plants where they release pheromones and wait for
males to arrive, then after mating make short, clumsy flights to deposit eggs. Males
make longer flights, but limit their movements to the natural borders of the fen, turning
back when approaching open water or the forest edge. For both sexes, maximum
flying height is about 2 m above the surface of the fen (Pryor 1998; K. Sime, pers.
observ.). With no obvious corridor for migration, travel between the two sampled
sites is probably very infrequent. Our aim was to test the hypothesis that Bog Buckmoth
populations are differentiated, because its low dispersal capability is among
the justifications for listing the Bog Buckmoth as threatened (Stanton 2004). We
hypothesized that we would observe genetic differentiation between the two Oswego
County populations we studied.
Methods
We sampled two Bog Buckmoth populations in Oswego County—Selkirk Fen
and Silver Lake Fen (hereafter, Selkirk and Silver Lake; Fig. 1). Selkirk is on the
eastern shore of Lake Ontario near Port Ontario, NY, approximately 30 km northeast
of Silver Lake, which is an inland site about 10 km south of the lakeshore, near
Minetto, NY. We collected adult males (22 from Silver Lake and 14 from Selkirk) in
late September 2009. Males fly rapidly across wide swaths of the fens in search of
emerging females (Pryor 1998). To ensure a random sample representative of each
population, volunteers scattered throughout both sites took the moths in flight.
Within hours of collection, we killed the insects by freezing, and dissected the
thorax away from the rest of the body. We preserved the samples in 100% ethanol
and stored them at -40 °C until we conducted DNA analysis. We extracted DNA
from the thoracic muscles using the Gentra PureGene Tissue Kit (Qiagen, Valencia,
CA) following the manufacturer’s protocol. Extracts were quantified using a Thermo
Scientific NanoDrop 2000 spectrophotometer (Wilmington, DE) and amplified
fragment-length polymorphisms (AFLPs; Vos et al. 1995) were targeted according
to the protocol provided by Beckman Coulter, Inc. (Pasadena, CA). We used three
selective PCR-primer combinations: EcoACT-MseCAA, EcoAGC-MseCTC, and
EcoACT-MseCAT. Fragments were visualized by capillary electrophoresis on a
Figure 1 (following page). Map of Bog Buckmoth populations in the vicinity of Lake Ontario.
Two localities in Wisconsin approximately 900 km due east of Oswego County, NY
(highlighted area) are not shown. Two Oswego County sites, Selkirk Fen and Silver Lake
Fen, were sampled for the current study. Selkirk Fen is the southernmost of a complex of
five fens (not distinguishable at this resolution) along the eastern shore of the lake, separated
from each other by a few km of forested or developed land, that historically have been
known to harbor Bog Buckmoth. By 2009, when we took our samples, only one fen other
than Selkirk had a significant Bog Buckmoth population, and in 2013 all but the Selkirk
population seemed to have disappeared (Bonnano 2013).
Northeastern Naturalist Vol. 21, No. 4
J. Buckner, A.B. Welsh, and K.R. Sime
2014
509
Beckman Coulter CEQ8000 and scored with the CEQTM 8000 genetic analysis system
software. A single individual evaluated all genotypes.
We used AFLP data to characterize genetic variation between and within the two
populations. Expected heterozygosity (assuming Hardy-Weinberg equilibrium), the
percentage of polymorphic loci, and population differentiation (FST) were calculated
Northeastern Naturalist
510
J. Buckner, A.B. Welsh, and K.R. Sime
2014 Vol. 21, No. 4
with the program AFLP-SURV (Vekemans et al. 2002) using the method described
by Lynch and Milligan (1994). Significance of FST was based on 1000 permutations.
We employed a bayesian approach to determine the most likely number of populations
(K) given the genetic data, using the software STRUCTURE (Pritchard et al.
2000). We used the admixture model (allowing for mixed ancestry of individuals)
and assumed correlated allele frequencies between populations, as recommended
by Falush et al. (2003), to detect subtle differences in population structure. Sampling
location was not used as a prior in the analysis. Values of K ranging from 1
to 4 were tested using 5 replicates for each value of K, a burn-in period of 100,000
iterations, and 100,000 Markov-chain Monte Carlo iterations following the burnin
period. Values of K exceeding the number of sampled locations were tested to
allow for the possibility of multiple populations existing at a single location. Loglikelihood
(LnP[D]) values for the tested K-values were plotted, and the most likely
number of populations was determined based on the K-value with the highest likelihood
and greatest change in K (ΔK), based on the method described by Evanno et
al. (2005), as implemented in the software STRUCTURE HARVESTER (Earl and
von Holdt 2012). Membership coefficients (Q) were calculated for each individual
to estimate the fraction of its genome with ancestry in each of the K clusters.
Results and Discussion
LnP(D) and ΔK both peaked at K = 2, indicating that the most likely number of
populations represented in our sample collection was two (Fig. 2). However, the
Figure 2. Determination of most the likely number of Bog Buckmoth population clusters
(K) in Oswego County, NY, given the genetic data from five replicate runs for each value
of K in STRUCTURE. The most likely number of populations is two, based on the highest
probability (LnP [D]) with lowest variability between runs and greatest change in K (ΔK).
Northeastern Naturalist Vol. 21, No. 4
J. Buckner, A.B. Welsh, and K.R. Sime
2014
511
two populations did not separate based on sample location (Fig. 3). Instead, a single
cluster was predominant in the two locations. At Silver Lake, 82% of the sampled
individuals had a high proportion of membership (Q > 0.70) in this dominant cluster.
At Selkirk, 57% of the sampled individuals belonged to the dominant cluster
(Q > 0.70). One possible explanation is that Silver Lake and Selkirk represent a
single population, with contributions from a third population, i.e., the second cluster
identified in the samples could represent individuals from another population
that we did not sample. These potential migrants (with Q > 0.7 in cluster 2) account
for 9% of the sampled individuals at Silver Lake and 21% of the sampled individuals
at Selkirk. The most likely unsampled source of migrants is the population
nearest Selkirk in South Pond Fen (hereafter South Pond), located approximately
3 km north of Selkirk. Since 2001, the South Pond population has been very small
compared to the Selkirk population, and no Bog Buckmoths have been observed
there since 2012, but historically the South Pond population had probably been
larger (Bonnano 2013).
An alternative explanation is that Silver Lake and Selkirk are genetically distinct,
with gene flow between the two populations. Migration seems to be asymmetrical,
with a higher proportion of migrants going to Selkirk than to Silver Lake, which
may in turn reflect the fact that prevailing winds would tend to move moths in that
direction. Based on the FST value, the Silver Lake and Selkirk populations were genetically
distinct (FST = 0.13, P < 0.001). Some individuals appear to have mixed
ancestry (9% at Silver Lake and 21% at Selkirk; Q < 0.70 in either cluster), which
could have resulted from mating between individuals from the two locations. These
two explanations are not mutually exclusive; a combination of the two scenarios
could be that Silver Lake and Selkirk are genetically distinct, with genetic contributions
from a third, unsampled population.
The use of the three selective PCR-primer pairs resulted in the generation of
203 fragments. The percentage of polymorphic loci (Selkirk: 63%, Silver Lake:
Figure 3. Membership coefficients from program STRUCTURE (y-axis) for each individual
(represented by vertical bars) for Bog Buckmoth individuals sampled from Silver Lake (n =
22) and Selkirk (n = 14) when K = 2. Clusters are represented by light vs. dark gray shading.
The black line separates the two sampling locations.
Northeastern Naturalist
512
J. Buckner, A.B. Welsh, and K.R. Sime
2014 Vol. 21, No. 4
67%) and expected heterozygosity (Selkirk: 0.233, Silver Lake: 0.227) were
similar for the two populations. The heterozygosity values are within the range
expected for Lepidoptera, according to genetic studies that used allozyme data to
produce a range of values from about 0.04 to 0.32 in various taxa (Packer et al.
1998). The similar results are not surprising because population densities at the
two sites have been similar since monitoring at Selkirk began in 2004 and Selkirk
and Silver Lake have consistently been two of the most populous Bog Buckmoth
sites (Bonnano 2013). Additionally, gene flow between the two populations, or
into these populations from a third source, could result in similar levels of genetic
diversity. Our results suggest that either explanation is plausible.
Rubinoff and Sperling (2004) argued that their mtDNA results, which showed
the Bog Buckmoth as a paraphyletic assemblage within H. maia, indicated that
the Bog Buckmoth is not isolated from surrounding populations of H. maia. They
suggested however that AFLP data might shed light on recent isolation events
within the Bog Buckmoth / H. maia complex. Although population differentiation
is apparent among Bog Buckmoth populations, our results indicate that some
between-population gene flow has occurred. Our data are consistent with two possible
scenarios: gene flow between the two sampled fens, or from an additional
unsampled population. Thus, the 30 km of unsuitable habitat between Selkirk and
Silver Lake may not be enough to isolate these populations. It seems most likely
that moths are occasionally delivered by winds from one to the other, with prevailing
winds tending to carry them toward Selkirk. Migration between Selkirk and
South Pond as well as the nearby lakeshore fens is likely as well. These explanations
are not mutually exclusive, and neither can be eliminated by analysis of our
data. Our conclusions are limited by having sampled only two populations, and
further sampling and analysis that includes additional Bog Buckmoth populations
as well as other H. maia would provide a clearer picture of gene-flow patterns and
colonization and isolation events. It would be particularly instructive to obtain
comparable AFLP data from the Wisconsin and Ontario populations, as well as
from non-Bog Buckmoth members of the H. maia complex.
Many Saturniidae are reported to persist at low densities compared to other
Lepidoptera, and cyclic population explosions and crashes appear to be regular demographic
features of the family (Tuskes et al. 1996). Such events can be a cause
for conservation concern when drastic population reductions result in a reduced effective
population size and lowered genetic diversity (Frankham et al. 2002). Bog
Buckmoths may be susceptible to such challenges because recolonization of the
fen by nearby populations may not occur rapidly after a population goes extinct.
However, our finding that there may be some gene flow between populations separated
by as much as 30 km indicates that recolonization is possible. Genetic studies
that focus on the relationships within Bog Buckmoth, rather than considering the
H. maia complex as a whole, may aid management efforts directed at the insect’s
recovery. Studies comparing the genetic structure between all Oswego County localities
as well as more distant populations would help delineate management units
and inform conservation priorities accordingly.
Northeastern Naturalist Vol. 21, No. 4
J. Buckner, A.B. Welsh, and K.R. Sime
2014
513
Acknowledgments
We thank the State University of New York at Oswego for support through Scholarly
and Creative Activities Grants awarded to K.R. Sime and to J. Buckner, and the McNair
Scholars Program for supporting J. Buckner’s undergraduate research. The New York State
Department of Environmental Conservation granted us access to Selkirk Fen and an Endangered
Species License to collect moths, and the Central New York Land Trust gave us access
to Silver Lake. Peter A. Rosenbaum, Eric Hellquist, Sandy Bonnano, John Laundré, Andy
Nelson, and two anonymous reviewers provided helpful guidance at various stages of this
project and in preparing the manuscript.
Literature Cited
Bonanno, S.E. 2008. Eastern Lake Ontario Bog Buckmoth population-monitoring program.
Summary report. New York State Department of Environmental Conservation, Endangered
Species Unit, Albany, NY. 20 pp.
Bonanno, S.E. 2013. Eastern Lake Ontario Bog Buckmoth population-monitoring program.
Summary report. New York State Department of Environmental Conservation, Endangered
Species Unit, Albany, NY. 14 pp.
Collins, M.M., and P.M. Tuskes. 1979. Reproductive isolation in sympatric species of dayflying
moths. Evolution 33:728–733.
Committee on the Status of Endangerd Wildlife in Canada (COSEWIC). 2009. COSEWIC
assessment and status report on the Bogbean Buckmoth, Hemileuca sp., in Canada. Ottawa,
ON, Canada. vii + 29pp. Available online at http://publications.gc.ca/collections/
collection_2011/ec/CW69-14-585-2010-eng.pdf. Accessed 21 April 2014.
Crandall, K.A., O.R.P. Bininda-Emonds, G.M. Mace, and R.K. Wayne. 2000. Considering
evolutionary processes in conservation biology. Trends in Ecology and Evolution
15:290–295.
Earl, D.A., and B.M. von Holdt. 2012. STRUCTURE HARVESTER: A website and program
for visualizing STRUCTURE output and implementing the Evanno method. Conservation
Genetics Resources 4:359–361.
Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals
using the software STRUCTURE: A simulation study. Molecular Ecology
14:2611–2620.
Falush, D., M. Stephens, and J.K. Pritchard. 2003. Inference of population structure: Extensions
to linked loci and correlated allele frequencies. Genetic s 164:1567–1587.
Falush, D., M. Stephens, and J.K. Pritchard. 2007. Inference of population structure using
multilocus genotype data: Dominant markers and null alleles. Molecular Ecology Notes
7:574–578.
Frankham, R., D.A. Briscoe, and J.D. Ballou. 2002. Introduction to Conservation Genetics.
Cambridge University Press, Cambridge UK. 617 pp.
Gompert, Z., C.C. Nice, J.A. Fordyce, M.L. Forister, and A.M. Shapiro. 2006. Identifying
units for conservation using molecular systematics: The cautionary tale of the Karner
Blue Butterfly. Molecular Ecology 15:1759–1768.
Gratton, C. 2006. Interactions between a native silkmoth, Hemileuca sp., and an invasive
wetland plant, Lythrum salicaria. Annals of the Entomological Society of America
99:1182–1190.
Legge, J.T., R. Roush, R. DeSalle, A.P Vogler, and B. May. 1996. Genetic criteria for establishing
evolutionarily significant units in Cryan’s Buckmoth. Conservation Biology
10:85–98.
Northeastern Naturalist
514
J. Buckner, A.B. Welsh, and K.R. Sime
2014 Vol. 21, No. 4
Lynch, M., and B.G. Milligan. 1994. Analysis of population genetic structure with RAPD
markers. Molecular Ecology 3:91–99.
Olivero, A.M. 2001. Classification and mapping of New York’s calcareous fen communities.
New York Natural Heritage Program, Albany, NY. 25 pp. + 9 appendices.
Packer, L., J. Taylor, D. Savignano, C. Bleser, C. Lane, and L. Sommers. 1998. Population
biology of an endangered butterfly, Lycaeides melissa samuelis (Lepidoptera; Lycaenidae):
Genetic variation, gene flow, and taxonomic status. Canadian Journal of Zoology
76:320–329.
Pritchard, J.K. 2000. Inference of population structure using multilocus genotype data.
Genetics 155:945–959.
Pryor, G. 1998. Life history of the Bog Buckmoth in New York State. Journal of the Lepidopterists’
Society 52:125–138.
Rubinoff, D., and F.A.H. Sperling. 2004. Mitochondrial DNA sequence, morphology, and
ecology yield contrasting conservation implications for two threatened buckmoths. Biological
Conservation 118:341–351.
Stanton, E.J. 2004. Development of a population-monitoring program for Bog Buckmoth
(Saturniidae: Hemileuca sp.). The Nature Conservancy, Pulaski, NY. 12 pp.
Tuskes, P.M., J.P. Tuttle, and M.M. Collins. 1996. The Wild Silk Moths of North America.
Cornell University Press, Ithaca, NY. 250 pp.
Vekemans, X., T. Beauwens, M. Lemaire, and I. Roldan-Ruiz. 2002. Data from amplified
fragment length polymorphism (AFLP) markers show indication of size homoplasy and
of a relationship between degree of homoplasy and fragment size. Molecular Ecology
11:139–151.
Vos, P., R. Hoger, M. Bleeker, M. Reijans, T. van de Lee, M. Hornes, A. Frijters, J. Pot, J.
Peleman, M. Kuiper, and M. Zabeau. 1995. AFLP: A new technique for DNA fingerprinting.
Nucleic Acids Research 23:4407–4414.