Ghosts of glaciers and the disjunct distribution of a threatened California moth (Euproserpinus euterpe)

Biological Conservation 184 (2015) 278–289

Contents lists available at ScienceDirect

Biological Conservation journal homepage: www.elsevier.com/locate/biocon

Ghosts of glaciers and the disjunct distribution of a threatened California moth (Euproserpinus euterpe) Daniel Rubinoff a,⇑, Michael San Jose a, Paul Johnson b, Ralph Wells c, Ken Osborne d, Johannes J Le Roux e a

310 Gilmore Hall, Department of Plant and Environmental Protection Sciences, University of Hawaii, Honolulu, HI 96822, United States Pinnacles National Park, 5000 Hwy 146, Paicines, CA 95043, United States c 303 Hoffman Street, Jackson, CA 95642, United States d Osborne Biological Consulting, 6675 Avenue Juan Diaz, Riverside, CA 92509, United States e Centre for Invasion Biology, Department of Botany and Zoology, Stellenbosch University, Matieland 7602, South Africa b

a r t i c l e

i n f o

Article history: Received 4 October 2014 Received in revised form 12 January 2015 Accepted 13 January 2015

Keywords: Lepidoptera Sphingidae California Phylogeography Endangered species

a b s t r a c t The Kern Primrose Sphinx moth (Euproserpinus euterpe) is a threatened moth twice thought to have gone extinct. It was historically known only from a small basin in the southern Sierra Nevada of California, but a new putative population was recently discovered 115 km to the west. Analysis of both nuclear and mitochondrial DNA suggest discontinuous genetic breaks between the Kern Primrose Sphinx, its closest relative the Phaeton Sphinx (Euproserpinus phaeton), and at least one additional species discovered during the course of this study. Geographic distance is well correlated with genetic distance within species, but not between species. Genetic discontinuities show the influence of past glacial events and suggest recent range expansions, though not always congruent with other phylogeographic studies from the region. Our phylogeographic results demonstrate that past glacial events, the altitudinal suppression of suitable habitat, and isolation may have been more important than population-level ecological factors such as differences in habitat (e.g. sand dunes, oak forest, montane grasslands) in promoting speciation. Effective conservation of the genetic diversity of the widespread Phaeton Sphinx and its two geographically restricted relatives requires genetic data at the population level because relatively few localized populations harbor most of the genetic variation. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction The conservation of rare species and unique populations is usually guided by measures of genetic diversity, as this diversity often correlates with geographic and ecological patterns of isolation (Avise, 1998; Dimmick et al., 1999; Moritz, 1994, 2002). Conservation of rare species whose genetic diversity is decoupled from geographic distance between populations poses an added challenge because discovery of populations of conservation importance requires additional scrutiny, not just finding new localities. Additionally, the causes of genetic discontinuity have broad relevance to our understanding of microevolutionary processes such as speciation and its mechanics. When a species receives legislative protection, its evolutionary independence and, thus, justification for protected status can be called into question (e.g. Ramey et al., 2006; Vignieri et al., 2006). This can be particularly controversial if the protected species is only moderately diverged from adjacent, non-threatened sister taxa. ⇑ Corresponding author. http://dx.doi.org/10.1016/j.biocon.2015.01.023 0006-3207/Ó 2015 Elsevier Ltd. All rights reserved.

The issue of monophyly, evolutionarily significant units (Moritz, 1994), and the controversy in the delimitation and conservation of threatened species and populations has been intensively studied in many North American mammals (Leonard et al., 2005; Ramey et al., 2006; Vignieri et al., 2006) and birds (e.g. Coulon et al., 2008; Delaney and Wayne, 2005), but rarely in endangered insects (e.g. Gompert et al., 2006; Saarinen et al., 2009), despite evidence that insects are at least as threatened by extinction as other groups (McKinney, 1999). The California Floristic Province is a global biodiversity hotspot (Lapointe and Rissler, 2005), making it the focus of many phylogeographic studies, and of high conservation value. Most research has focused on examining the importance of past vicariant events in interpreting congruence, or the lack thereof, in patterns of genetic discontinuity in modern populations across many taxa (Calsbeek et al., 2003; Lapointe and Rissler, 2005), including salamanders (Jockusch and Wake, 2002), newts (Kuchta and Tan, 2005, 2006), frogs (Macey et al., 2001), woodrats (Matocq, 2002), titmice (Cicero, 1996; Lapointe and Rissler, 2005), snakes and lizards (Feldman and Spicer, 2006), beetles (Caterino and Chatzimanolis,

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

2009; Polihronakis and Caterino, 2010a; Polihronakis and Caterino, 2010b; Polihronakis and Caterino, 2012; Polihronakis et al. 2010), spiders (Hedin and Carlson 2011; Hedin et al., 2013), and walking stick insects (Law and Crespi, 2002). Overwhelmingly, these studies have focused on taxa that rely on relatively mesic habitats, which expand and connect during glacial periods and contract into isolation during drier, warmer periods (e.g. Calsbeek et al., 2003; Lapointe and Rissler, 2005; Schoville et al., 2011). The moth genus Euproserpinus (Sphingidae) provides an opportunity to examine the opposite, because a desert species will become isolated when mesic conditions spread (Fig. 1). The genus is also of high conservation relevance because E. euterpe (Edwards, 1888), the Kern Primrose Sphinx moth, is federally listed as a threatened species, and resources are available to protect it and change land management regimes given data on its distribution and species status. Because many previous phylogeographic studies have focused on Central and Southern California as important areas for understanding the impacts of glaciation and geology in shaping genetic discontinuity (e.g. Alexander and Burns, 2006; Chazimanolis and Caterino, 2007; Feldman and Spicer, 2006; Kuchta 2007; Kuchta and Tan, 2006; Law and Crespi, 2002; Macey et al., 2001; Matocq, 2002; Rich et al., 2008; Rodríguez-Robles et al., 2001; Rubinoff and Sperling, 2004; Sandoval et al. 1998; Sgariglia and Burns, 2003; Starrett and Hedin, 2007), Euproserpinus appears to be an ideal study subject for improving our understanding of this phenomenon. Most urgently, because Euproserpinus contains a federally listed species, E. euterpe, understanding the genetic structure of the genus and its underlying determinants across California will be important in identifying the threatened species, managing highly restricted populations, and prioritizing areas for conservation. Recent work has confirmed that insects represent the majority of California Floristic Province’s endemic biodiversity but are rarely considered in conservation planning, perhaps due to a lack of even basic faunistic or genetic data (Caterino, 2007). Even with legislative protection, the genetic status of E. euterpe has remained unstudied and the phylogeography of the genus completely unknown; essential components to circumscribing species, and justifying and effectively directing resources toward those that are endangered. Euproserpinus euterpe was thought to be restricted to the Walker Basin, Kern Co. California, an area of only 15 km2 isolated in the southern Sierra Nevada (Tuskes and Emmel, 1981). Owing to its limited distribution, habitat loss due to farming and overgrazing,

279

and concerns of over-collecting and the ecological impact of an exotic weed, Erodium cicutarium (L.) L’Hér. ex Aiton Geraniaceae, E. euterpe became the first moth species to receive federal protection in the United States (USFWS, 1980). Shortly after being listed as Threatened, researchers failed to find the moth for several years and it was again feared extinct until its rediscovery in the Walker Basin in the 1990s (Osborne, 1999; Shields, 1990) and has persisted since (Osborne, pers obs). In 2006, an additional population of E. euterpe was discovered at Carrizo Plain National Monument, 115 km west of the Walker Basin population (Jump et al., 2006). Whereas E. euterpe, appears to be restricted to the Walker Basin and the vicinity of Carrizo Plain National Monument, the Phaeton Sphinx (E. phaeton) occurs allopatrically from E. euterpe, 27 km east of Walker Basin in Kelso Valley and 53 km away at Walker Pass, with widely-scattered but localized populations distributed northeast to Lassen county and south across the Mojave and Colorado Deserts (Fig. 2). While there are some morphological differences in larval and adult maculation between the two species (Jump et al., 2006), their close proximity at Walker Basin and nearby localities suggests the possibility for gene flow. For E. euterpe to be restricted to two disjunct populations 115 km apart, and yet to be reproductively isolated from E. phaeton just 27 km away, suggests one of two scenarios. Either a taxonomic error should be corrected by lumping a federally listed threatened species with a widespread congener, or else the current taxonomy is valid, being the result of a less parsimonious and remarkably incongruous pattern of speciation and biogeographic isolation. Specifically, we seek to address the following questions: Are E. euterpe and E. phaeton genetically isolated? Is the newly discovered Carrizo Plain population in fact the threatened E. euterpe, as suggested by morphological characters as presented by Jump et al. (2006)? If E. euterpe is truly restricted to just one or two disjunct locations, what are the patterns of phylogeographic relationships among the populations and how do they contrast with E. phaeton, which occurs across thousands of km2 of desert? What are the broader biogeographic and ecological implications of the genetic relationships across the genus? And finally, how does the phylogeography of Euproserpinus elucidate both applied and theoretical conservation planning on a regional scale? The current default hypothesis that has been used to justify conservation action, is that the E. euterpe populations, which occur in the Walker Basin and Carrizo Plain, are more closely related to each other than either is to the many E. phaeton populations surrounding them. If this is true, then E. euterpe populations will cluster under population level analyses, with all E. phaeton populations, even those geographically close to E. euterpe, being more genetically distant, and monophyletic. If the E. euterpe populations do not cluster together, then geographic distance predicts genetic proximity, the two E. euterpe populations will not be each other’s closest relatives, and the validity of the species and its listed status would be called into question. Because we are interested in understanding the probability that different populations may represent distinct species, including newly discovered ones in Carrizo Plain and the Central Coast area, we began with the most conservative approach by assuming that taxa are not distinct at the population level. In this way, the null hypothesis of genetic continuity is tested, rather than presumed.

2. Materials and methods 2.1. Study species, specimen collection, and DNA extraction Fig. 1. Hypothetical change in habitat distribution for Euproserpinus during (A) interglacial and (B) glacial periods. Although the mountains separating Euproserpinus were never glaciated, habitat would have been unsuitable for the moths. Agriculture and habitat alteration in the Central Valley appears to have isolated E. euterpe.in two remaining populations.

The three described species of Euproserpinus (E. phaeton, E. wiesti, and E. euterpe) are small diurnal hawkmoths highly adapted to desert climates. Their pupae may stay underground for years,

280

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

Fig. 2. (A) Map of California indicating the location of study areas with circles representing different Euproserpinus populations. (B) Heat map of pairwise Ust (lower diagonal) and Jost’s D values (upper diagonal) of Euproserpinus populations. Darker colors indicate higher genetic differentiation; lighter colors indicate lower genetic differentiation.

awaiting a winter of adequate rainfall. We attempted to collect samples from throughout the entire known distributions of E. phaeton and E. euterpe, plus E. wiesti from a single locality. Adults emerge and fly for perhaps two weeks in late winter or very early spring when few other insects are active, nighttime temperatures can fall below 0 °C, and their annual Onagraceous host plants have just begun to grow in the short-lived spring. During the frequent droughts in these xeric regions, the moths may hold over for years as pupae, not emerging at all and giving the impression that a population has disappeared. Further, because desert precipitation and temperatures are highly variable, distant populations may be genetically isolated because of temporal isolation due to consistent differences in adult emergence during the course of a season. Thus, for this study, locating and collecting samples of the populations across the deserts of California and Arizona has taken more than a decade (from 1997 to 2013), led to the discovery of many new localities, and may still be incomplete. During the study, one of the authors (P. Johnson) discovered a new population of Euproserpinus at Pinnacles National Park, San Benito Co., the farthest northwest the genus has been found. He subsequently discovered more populations in San Luis Obispo and Monterey counties. These moths appear darker in color than any other Euproserpinus taxa and inhabit open sandy areas adjacent to riparian habitats, oak savannah and openings in chaparral, remarkably mesic locations for a genus typical of scrub and barren desert. In addition, we collected specimens of E. wiesti, the only other member of Euproserpinus, for use as an outgroup taxon in phylogenetic analyses. This species had been considered for federal protection and is restricted to sand dune habitats along the Front Range of the Rocky Mountains. Locality data can be found in Table 1. Total

genomic DNA was extracted using the Qiagen DNeasy extraction kit (Qiagen, USA) from small pieces of abdomen or individual legs. Voucher specimens are held in the University of Hawaii’s Insect Museum’s cryogenic collection and by K.O., P.J, and the Pinnacles National Park Museum collection. 2.2. PCR amplification and DNA sequencing The mitochondrial gene cytochrome oxidase c subunit 1 (COI) was amplified in two parts for all specimens using the universal primers LCO-1490 and HCO-2198 and Jerry-k485 and Pat-k508 (Folmer et al., 1994; Simon et al., 1994). Each 50 lL PCR reaction contained ca. 50 ng of genomic DNA, 200 lM of each dNTP (Bioline, USA), 25 pmol of each primer, 5 U Taq DNA polymerase (Bioline, USA), 1  PCR reaction buffer, 1.5 mM MgCl2. PCR cycles consisted of initial denaturation of 95 °C for 5 min; 35 cycles for denaturation at 94 °C for 30 s, annealing at 53 °C for 60 s, elongation at 72 °C for 90 s; and final extension at 72 °C for 10 min. Due to unexpectedly deep divergences between some clades for the COI data (see results) we also sequenced a second, slower evolving nuclear gene, elongation factor 1 – alpha (EF1a), for representative taxa from each clade. EF1a was amplified using the primers Oscar and Bosie (Haines and Rubinoff, 2012) and PCR cycles consisted of initial denaturation at 95 °C for 5 min followed by 35 cycles at: 94 °C for 30 s, 55 °C for 60 s, and 72 °C for 90 s; and 72 °C for 10 min. All PCR amplifications were done in an MJ PTC-200cycler (MJ Research, USA). Amplified DNA fragments were purified using the QIAquick PCR Purification Kit (Qiagen, USA) and sequenced in both directions using the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction kit and an automated ABI PRISM 377XL DNA sequencer (Applied Biosystems, USA).

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

281

Table 1 Collection and genetic information for Euproserpinus from 30 collection sites. Each site is coded to correspond to population designation for population genetic analysis, number of DNA sequences generated (N), and haplotypes observed at the site. Species

Population

Locality

N

Haplotypes

E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E. E.

Carrizo Plain Carrizo Plain Imperial Co. Lassen Co. Lassen Co. Monterey Co. Monterey Co. New Mexico Pinnacles National Park Riverside Co. Riverside Co. Riverside Co. Riverside Co. Riverside Co. San Diego Co. San Diego Co. San Diego Co. San Diego Co. San Diego Co. San Diego Co. Shell Creek Ventura Co. Ventura Co. Walker Basin Walker Pass Walker Pass Walker Pass Walker Pass Yuma Co. Yuma Co.

CA: San Luis Obispo Co., Carrizo Plain CA: Santa Barbara Co., Cuyama Valley CA: Imperial Co., Imperial Sand Dunes CA: Lassen Co., Doyle CA: Lassen Co., Ft. Sage Mtn. CA: Monterey Co., Stockdale Mtn. CA: Monterey Co., Vineyard Cyn Rd. NM: Quay Co. CA: San Benito Co., Pinnacles National Park CA: Riverside Co., Anza Valley CA: Riverside Co., Black Hills CA: Riverside Co., Gavilan Hills CA: Riverside Co., Temecula Wash CA: Riverside Co., Wilson Valley CA: San Diego Co., Campo CA: San Diego Co., Jacumba CA: San Diego Co., La Posta Road CA: San Diego Co., Mason Valley CA: San Diego Co., Ranchita CA: San Diego Co., Things Valley CA: San Luis Obispo Co., Shell Creek CA: Ventura Co., Lockwood Cr. Rd. CA: Ventura Co., Lockwood Valley CA: Kern Co., Walker Basin CA: Kern Co., Highway 178 CA: Kern Co., Kelso Valley CA: Kern Co., Walker Pass CA: Kern Co., Weldon AZ: Yuma Co., Yuma AZ: Yuma Co., Highway 80

15 2 6 6 6 7 3 6 4 45 2 1 12 3 4 1 23 2 9 8 12 4 6 9 2 6 26 5 6 4

H04, H04 H01, H12, H12, H03, H03 H09, H05, H01, H02, H02 H01, H02, H01, H01 H01, H61, H02, H01, H03, H01, H01, H10, H02 H02, H01, H01, H01, H02,

euterpe euterpe phaeton phaeton phaeton nsp nsp weisti nsp phaeton phaeton phaeton phaeton phaeton phaeton phaeton phaeton phaeton phaeton phaeton nsp phaeton phaeton euterpe phaeton phaeton phaeton phaeton phaeton phaeton

2.3. DNA sequence alignment Sequence contigs were constructed, edited and aligned using BioEdit version 7.0.5.3 (Hall, 1999) or Geneious 6.0.5 (Biomatters). All unique sequences were deposited in GenBank (http://www. ncbi.nlm.nih.gov, Supplementary Materials Table 1). 2.4. Phylogenetics and molecular dating Nuclear and mitochondrial datasets were tested separately for an appropriate nucleotide substitution model using jModelTest v2.1.3 under the Akaike information criterion with correction (AICc) (Darriba et al., 2012). This approach identified the HKY + C and HKY as the most appropriate models of sequence evolution for the COI and EF1a genes respectively (Supplementary Materials Table 2). Each gene was analyzed separately and then concatenated using GARLI 2.0 (Zwickl, 2006) for Maximum Likelihood (ML) and MrBayes 3.2.1 (Ronquist et al., 2012) for Bayesian inference analyses. For all analyses, four independent Bayesian markov chain monte carlo (MCMC) runs were conducted in MrBayes, each with one hot chain and three cold chains. Each run started with a random tree, sampling every one thousand generations for 10 million generations with a relative burn-in of 25% using default priors except that the parameters statefreq, revmat, shape, and pinvar were unlinked between partitions. We used Tracer 1.5 (Rambaut and Drummond, 2009) to determine MCMC convergence for all Bayesian analysis. For the GARLI analysis, we conducted ten ML tree searches with default settings using a random starting tree to find the tree with the best likelihood score. To assess branch support one thousand Maximum Likelihood bootstrap replicates were conducted in GARLI, Maximum likelihood bootstrap trees were summarized in Sumtrees v3.1.0 (Sukumaran and Holder, 2010) with a minimum clade frequency of 50% and branch support was mapped onto the best scoring ML tree. Trees were visualized using FigTree v1.4.0 (Rambaut, 2012) and rooted with E. wiesti.

H08, H56, H63, H64 H18, H23, H57 H27, H28, H35, H58 H20, H28, H39 H38 H48, H49 H41 H02, H07, H11, H13, H26, H31, H53, H54, H55 H52 H06, H11, H32, H33, H34 H14 H19 H07, H62 H40, H21, H05, H36, H16 H29,

H18, H21, H22, H06, H65, H68, H69, H70, H71, H72, H73 H42, H43, H44, H45, H46, H47 H22, H74, H75 H24, H25 H37 H30, H60

H15 H02, H06, H59, H66, H67 H06 H15, H17, H76 H50, H51

We dated the branching events of the different populations of Euproserpinus by first determining if the COI dataset had an equal rate of evolution throughout the tree (i.e. assumptions of a molecular clock) using MEGA 5.2.2 (Tamura et al., 2011). Dating analyses were conducted using BEAST 1.7.5 (Drummond et al., 2012). Our analysis followed Brower’s (1994) approximation of mtDNA pairwise sequence divergence of 2.3% divergence per million years for Lepidoptera, corresponding to 0.0115 substitutions per site on the COI dataset. Because no fossils exist for Euproserpinus and there are no clear biogeographic calibration points to date nodes within our tree, we used a strict molecular clock with a Yule process prior for model of speciation and ran the analysis for 1  108 generations. To guarantee that the MCMC chain had run long enough to get a valid approximations of the parameters, individual log files were analyzed with Tracer v1.5 (Rambaut and Drummond, 2009) to assess convergence and confirm that the combined effective sample sizes for all parameters were larger than 200. All resulting trees were combined using LogCombiner v1.5.3 (Drummond et al., 2012), with a burn-in of 25%. A single maximum clade credibility tree was then drawn using TreeAnnotator v1.7.5 (Drummond et al., 2012) and visualized using Figtree v1.4 (Rambaut, 2012). We realize that our approach using relative rates for divergence estimation is imperfect (Britten, 1986; Lepage et al., 2007; Roger and Hug, 2006). Although it should not be regarded as more than a broad estimate of divergence time, it provides a rough estimate that may be compared with known climatic variation to inform plausible patterns of diversification. 2.5. Population genetics and network analysis The full COI dataset was used to reconstruct networks using statistical parsimony (Templeton et al., 1992) as implemented in TCS version 1.21 (Clement et al., 2000). We decided to analyse the data using a network in addition to traditional phylogenetic tree building approaches, not only because of its higher resolution, but also

4 2 0.5 0.00035 1 0.47871 0.61237 0.061237 0.172 Bold values indicate p-value < 0.05.

Monterey Co. Pinnacles NM Shell Creek

10 2 0.2 0.00014 1 1.34668 1.24341 1.11173 0.339 26 6 0.705 0.00097 6 0.34091 0.27344 0.34181 0.878 17 5 0.66176 0.00095 4 0.34159 0.23149 0.47721 0.55 9 4 0.77778 0.00101 5 0.74755 0.59727 0.91004 0.286 26 9 0.834 0.00216 9 1.01621 0.812113 1.01239 0.907 10 7 0.93333 0.00184 9 1.27406 1.13824 1.1165 2.41 39 7 0.5857 0.00075 7 1.72582 1.70249 0.96861 2.056 10 4 0.64444 0.00053 3 0.96179 0.8049 1.03446 1.466 47 26 0.89362 0.00151 26 3.26492 3.07363 2.14453 26.888 63 16 0.74194 0.00125 17 2.46639 2.38133 1.51513 8.344

The ML tree for the combined EF1a and COI data (Fig. 3) revealed that E. euterpe from the Walker Basin formed a wellsupported (78 BS, 1.0 PP) clade with Euproserpinus populations collected across the Central Valley in the vicinity of the Carrizo plain. In contrast E. phaeton from Walker Pass formed a clade with populations from Lassen Co. in the far north, and all the way south along the eastern flanks of the Sierra Nevada ranges through the Mojave and Colorado Deserts to at least as far south as the Arizona and California borders with Mexico, and west to the Pacific Ocean. Samples from San Benito Co. (Pinnacles National Park), San Luis Obispo Co. (Shell Creek) and Monterey Co. formed a distant and basal (97 BS, 1.0 PP) clade to all other Euproserpinus taxa. The COI ML tree (Fig. 4) suggested that E. phaeton is paraphyletic, E. euterpe is closest to samples from Lassen Co. to the north and Imperial Co. to the south. Samples from San Benito Co., San Luis Obispo Co. and Monterey Co. are basal with respect to all Euproserpinus taxa. The EF1a tree (Fig. 5) showed poor resolution. The lack of resolution for both the COI and EF1a trees indicates relatively recent divergence between E. euterpe and E. phaeton, or repeated periods of genetic exchange through secondary contact and introgression after initial divergence. Tests for molecular clock assumptions did not reject the null hypothesis of equal evolutionary rates throughout the tree (P = 0.95). Our dating analysis of the COI data based on relative rates (Brower, 1994) estimated the MRCA for Euproserpinus within California to 0.67 MYA (range 0.94–0.45 MYA, Fig. 6). The newly discovered species diverged from the rest of Euproserpinus 0.3977

12 7 0.90909 0.00261 10 0.69599 0.63967 0.53551 0.768

3.2. Phylogenetic analysis and molecular dating

Table 2 Genetic variability of COI sequences within species and populations of Euproserpinus (n = 245, 1435 bp).

For the full COI dataset, aligned DNA sequences were 1435 bp long. For the combined COI and EF1a sequence data the aligned sequences were 2197 bp long, requiring no gaps. All unique DNA sequences were deposited in GenBank (accession numbers KP263128-KP263236).

6 4 0.86667 0.00181 7 1.04236 0.99519 0.88407 0.024

3.1. DNA sequence alignment

E. phaeton Imperial Co Lassen Co Riverside Co San Diego Co Ventura Co Walker Pass Yuma AZ E. euterpe Walker Basin Carrizo Plain E. nsp

3. Results

187 58 0.808 0.00163 58 5.00445 5.06986 2.35813 74.9

because networks are better suited, though not perfect, for inferring relationships at the intra-specific level, as these are often not hierarchical as assumed by traditional tree building methods (Posada and Crandall, 2001). Genetic variability of COI sequences was calculated using DNAsp (Librado and Rozas, 2009). Genetic diversity indices calculated included the number of haplotypes (Nh), haplotype diversity (h), nucleotide diveristy (p) and the number of segregating sites (S). Tests of gene neutrality were also calculated using DNAsp for the different species/populations included: Fu and Li’s F⁄, Fu and Li’s D⁄, Tajima’s D and Fu’s F (Table 2) (Tajima, 1989; Fu and Li, 1993). Estimates of population genetic structure were calculated as analysis of molecular variance (AMOVA) and as population pairwise Ust values in Arlequin version 3.5 (Excoffier and Lischer, 2010). Population pairwise Jost’s D (Jost, 2008) values were also calculated in R (R Core Development 2013) using the R packages seqinR (Charif and Lobry, 2007) and APE (Paradis et al., 2004) using an R script from Pennings et al. (2011). A permutation test with 1000 replicates was conducted in R to test significance of Jost D values. Pairwise Ust estimates were used in subsequent Mantel tests and regressed against log-transformed geographical pairwise distances to test for isolation by distance using Arlequin 3.5 (Excoffier and Lischer, 2010). These results were visualized using the R statistical environment (R Core Development, 2013).

12 4 0.77273 0.00116 4 1.26963 1.19483 0.90839 0.497

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

Sample size No. of haplotypes (Nh) Haplotype diversity (h) Nucleotide diversity (l) No. of Segregating sites (S) Fu and Li’s F⁄ Fu and Li’s D⁄ Tajima’s D Fu’s F

282

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

283

Fig. 3. Maximum likelihood tree for the concatenated dataset (COI, EF-1a). Support values above branches are Maximum Likelihood Bootstrap values/Bayesian Posterior Probabilities. Scale bar indicates the number of substitutions per site. Each branch tip represents multiple individuals.

MYA (range 0.56–0.25 MYA). E. euterpe diverged from E. phaeton 0.1543 MYA (range 0.24–0.07 MYA). 3.3. Population genetics and network analysis Various indices were calculated to explore within population genetic variability and test the neutrality of mutations in Euproserpinus (Table 2). Analysis of molecular variance (AMOVA) indicates higher genetic variation between species but lower variance between populations within species (Table 3). Pairwise Ust and Jost’s D values between E. euterpe and E. phaeton populations (Table 4, Fig. 2) are congruent, with higher values in intraspecific population pairs than interspecific pairs. Due to the low number of populations (n < 3 for E. euterpe and E. nsp.) we were only able to conduct the Mantel test with E. phaeton (Fig. 7); it indicated a significant correlation between genetic differentiation and geographic distance for this species. Network analysis indicated significant intraspecific genetic structure, suggesting some isolation between populations, but much greater isolation between each of the species (Fig. 8). 4. Discussion 4.1. Phylogenetics and molecular dating In both the combined analysis and the COI phylogeny (Figs. 3 and 4), E. euterpe is placed within E. phaeton rendering E. phaeton paraphyletic. These are recently diverged species (150 KYA) and due to incomplete lineage sorting, our phylogenetic analysis was not able to differentiate between the two species. In contrast,

population-level analyses such as AMOVA, Ust and Jost’s D values, more appropriate for recently diverged taxa, indicate that both populations of E. euterpe (Walker Basin and Carrizo Plain, Table 4) are significantly different from the other populations of Euproserpinus in California. 4.2. Population genetics Our population level analysis not only confirms the genetic uniqueness of E. euterpe from Walker Basin, but also suggests that the populations in the vicinity of Carrizo Plain represent additional lineages of this species, once thought to have a very restricted distribution; a remarkable result considering the proximity of the Walker Basin Euproserpinus to those of the Walker Pass E. phaeton. Yet these Walker Pass populations are in relatively close genetic contact with E. phaeton that occur hundreds of kilometers to the north in the colder Great Basin sand dune habitats of Lassen Co. (where the moth flies in mid-April), and southern populations from the extreme deserts of Yuma, Arizona (where the moth flies in December and January). In short, there is more genetic diversity in the E. euterpe on opposite sides of the San Joaquin Valley than there is in E. phaeton between far northern California and the Mexican border. The most divergent Euproserpinus in California is the newly discovered species from San Benito, Monterey, and northern San Luis Obispo counties. These appear to have been long-isolated from other Euproserpinus lineages, despite now occurring within 60 km of E. euterpe in San Luis Obispo Co. Our dated phylogeny indicates that this taxon diverged from the rest of Euproserpinus around 400 KYA. Our age estimates should be regarded as approximate, but broadly place the divergences of Euproserpinus in the Ionian Stage

284

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

Fig. 4. Maximum likelihood tree based on the mitochondrial COI dataset. Support values above branches are Maximum Likelihood Bootstrap values/Bayesian Posterior Probabilities. Scale bar indicates the number of substitutions per site.

of the Pleistocene, in which there were cycles of glaciation and warmer interglacial periods, the former of which might have isolated populations of Euproserpinus (Fig. 1). While other recent phylogeographic studies of insects over the same region found evidence of genetic isolation and population structure (Polihronakis and Caterino, 2010a, b; Polihronakis et al., 2010), none of these revealed patterns that were even roughly congruent with Euproserpinus, possibly because these other taxa are montane/ mesic specialists. Perhaps the most dramatic contrast between the phylogeography of Euproserpinus and other taxa from the region is the genetic discontinuity typically seen separating east from west across the Transverse Ranges (e.g. Calsbeek et al., 2003; Chatzimanolis and Caterino, 2007; Caterino and Chatzimanolis, 2009) that is absent in Euproserpinus. Our data showed E. phaeton in genetic contact around the Transverse Ranges, while E. euterpe occurs to the north in the vicinity of the Carrizo Plain. 4.3. Phylogeography Euproserpinus consistently demonstrates inconsistencies between geography and genetic diversification (Figs. 1 and 8), with E. phaeton being genetically connected over long distances across much of California, yet oddly isolated from nearby populations of E. euterpe in Kern and Santa Barbara counties. We suspect that the reasons for the currently incongruous patterns of genetic isolation are the combined artifacts of past climatic and more recent human activity, as discussed below. Genetic isolation of populations in the southern Sierra Nevada has been shown across many taxa (Feldman and Spicer, 2006; Kuchta, 2007; Kuchta and Tan, 2005), including other moths

(Althoff et al., 2006; Rich et al., 2008). While the impacts of glacial events on different organisms have varied in their nature, severity, and spatial extent, genetic isolation caused by glacial cycles appears to be widespread across vertebrates, insects and spiders in the southern Sierra (Alexander and Burns, 2006; Feldman and Spicer, 2006; Kuchta and Tan, 2006; Kuchta, 2007; Law and Crespi, 2002; Macey et al., 2001; Matocq, 2002; Rich et al., 2008; Rodríguez-Robles et al., 2001; Sandoval et al., 1998; Sgariglia and Burns, 2003; Starrett and Hedin, 2007). Because organisms differ in their habitat requirements and generation times, we would not necessarily expect the estimates of isolation events to be congruent across taxa; glacial maxima would have been periods of expansion for cool, mesic dependent species, and retraction for desert taxa such as Euproserpinus. Cold, wet periods greatly restricted desert and scrub habitat during the Pliocene and Pleistocene (Axelrod, 1980). Glaciers would not have directly impacted Euproserpinus, because the areas where they occur were apparently never glaciated, but climatic shifts would have pushed populations to lower elevations where drier, warmer, suitable habitat remained. Populations may have ended up in isolated desert refugia, divided by cooler and more mesic biomes at higher elevations. The Walker Pass area would have sustained a wetter and cooler climate, probably unsuitable for Euproserpinus (Axelrod, 1980). Our molecular dating estimates suggest that divergence between E. euterpe and E. phaeton coincided with the glacial periods about 250,000 years ago. While the glaciers are long gone, they appear to have had lasting impacts on Euproserpinus, promulgating a speciation event across the Southern Sierra. At some point in the glacial cycles, after the glaciers retreated, secondary expansion back to higher elevations would have

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

285

Fig. 5. Maximum likelihood tree based on the mitochondrial EF-1a dataset. Support values above branches are Maximum Likelihood Bootstrap values/Bayesian Posterior Probabilities. Scale bar indicates the number of substitutions per site. Low variability in this gene reduced resolution and branch support.

resulted in the proximal geographic, and remarkably disparate genetic pattern we see between E. euterpe and the Walker Pass E. phaeton today. Interestingly, most other phylogeographic studies in the region find an east/west division (e.g. Polihronakis and Caterino, 2010b), rather than the north/south split evident in Euproserpinus. Again, we suspect this is because montane/mesicdependent taxa have been the focus of most previous studies. The relatively deeper divergence evident in the new species of Euproserpinus from the Central Coast suggests an older origin. In the past 2–5 million years two seaways may have existed, isolating the Coast Ranges at the Monterey peninsula to the north and the southern tip of the Coast Ranges to the south, (Hall, 2002; Yanev, 1980). While the timing of the strong phylogeographic break in San Luis Obispo Co. is more recent than the hypothesized seaways that made the region an island during the Miocene, this appears to be one of the most significant frontiers for Euproserpinus, and at least as important as the southern Sierra. The reasons for the break in this region remain unclear, since all three Euproserpinus species use very similar habitats. Because the molecular clock estimates suggest that this speciation event was only 400,000 years ago, it is possible that Euproserpinus was isolated on the Central Coast islands and secondary, more recent, introgression is obscuring what used to be a deeper genetic division between the groups or, since the methods used to date our phylogeny are not robust, the actual divergence may have occurred much later. The older origin, and genetic distinctiveness of the Central Coast species supports its designation as a new species and it will be

described in a future publication. In contrast to the geologicallymediated events of the more distant past, the unexpected connectivity of E. euterpe in Walker Basin and Carrizo Plain may be due to recent changes to a historic range that was once continuous across the San Joaquin Valley. While it is now highly modified agricultural land or degraded scrub, the southern valley was previously a patchwork of riparian and desert landscapes connecting the Walker Basin with the Carrizo Plain. Thus, the current isolation of E. euterpe into two contemporary refugia of suitable habitat might be the reflection of a few hundred years of anthropogenic activity on what was likely a much more widespread taxon across the San Joaquin Valley. We presume that the genetic discontinuity between E. euterpe in the Walker Basin and Carrizo Plain is due to the relatively recent extinction of genetically and geographically intermediate populations across the Southern San Joaquin valley. Alternately, it may be that the genetic divergence between E. euterpe in the Walker Basin and Carrizo Plain is a robust indicator of long isolated populations because the Sierra Nevada and Transverse ranges are isolating factors for many species of mammals, reptiles and amphibians (Jockusch and Wake, 2002; Kuchta and Tan, 2005, 2006; Leonard et al., 2005; Macey et al., 2001; Ramey et al., 2006; Vignieri et al., 2006). However, Euproserpinus is xerophilic, and the pre-agriculture southern San Joaquin valley is unlikely to have been much different from the current desert of the Carrizo Plain. Thus, Euproserpinus provides a contrasting example to the phylogeography of the mesophilic taxa previously studied in this

286

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

Fig. 6. Bayesian chronogram of the COI dataset of Euproserpinus showing 95% confidence intervals of posterior dates as calculated in BEAST.

Table 3 Analysis of molecular variance. Source of variation

DF Sum of squares

Variance component

Percent variation

Fixation indices FSC = 0.32 Pvalue = 0.00 FST = 0.84

Among species

2 393.44

4.35 Va

77.42

Among populations within species

9

0.40 Vb

7.24

77.62

Within populations

227 195.95

0.86 Vc

Total

238 667.02

5.62

Pvalue = 0.00 FCT = 0.77 Pvalue = 0.00 Fig. 7. The relationship between Ust and log of physical distances between Euproserpinus phaeton populations.

Significance test based on 1023 permutations populations are same as haplotype network.

Table 4 Pairwise Ust values (below diagonal) and Jost’s D values (above diagonal) based on COI sequences (n = 245, 1435 bp). Species

E. phaeton

E. nsp

E. euterpe

Locality

Imperial Co. Lassen Co. Riverside Co. San Diego Co. Ventura Co. Walker Pass Yuma Co. Monterey Co. Pinnacles NP Shell Creek Carrizo Plain Walker Basin

E. phaeton

E. nsp

E. euterpe

Imperial Co.

Lassen Co.

Riverside Co.

San Diego Co.

Ventura Co.

Walker Pass

Yuma Co.

Monterey Co.

Pinnacles NP

Shell Creek

Carrizo Plain

Walker Basin

0 0.36487⁄ 0.23066⁄ 0.12263⁄ 0.16923⁄ 0.24645⁄ 0.22797⁄ 0.90946⁄ 0.87278⁄ 0.83948⁄ 0.78746⁄ 0.80958⁄

1.00000⁄ 0 0.61051⁄ 0.57361⁄ 0.59623⁄ 0.66571⁄ 0.49601⁄ 0.82337⁄ 0.76801⁄ 0.76928⁄ 0.66554⁄ 0.64700⁄

0.26248 1.00000 0 0.03146⁄ 0.14061⁄ 0.0179 0.03547 0.88377⁄ 0.87805⁄ 0.86320⁄ 0.80062⁄ 0.83138⁄

0.07449 1.00000 0.83161 0 0.02415 0.00285 0.09833⁄ 0.86449⁄ 0.86189⁄ 0.84381⁄ 0.78477⁄ 0.81362⁄

0.19588 1.00000⁄ 0.12463 0.15175⁄ 0 0.09846⁄ 0.26087⁄ 0.96178⁄ 0.95293⁄ 0.90287⁄ 0.86825⁄ 0.90028⁄

0.27969 1.00000 0.02672⁄ 0.20475⁄ 0.04368 0 0.14057⁄ 0.92952⁄ 0.92643⁄ 0.90356⁄ 0.8593⁄ 0.88751⁄

0.32468 1.00000⁄ 0.16176 0.1719 0.43157⁄ 0.32628 0 0.89841⁄ 0.85269⁄ 0.84329⁄ 0.78972⁄ 0.79645⁄

1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 0 0.88040⁄ 0.24149⁄ 0.93455⁄ 0.94748⁄

1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 0 0.28421 0.91419⁄ 0.91353⁄

1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 0.55969⁄ 0.10714⁄ 0 0.89153⁄ 0.88700⁄

1.00000 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 0.99999⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 0 0.72257⁄

1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 1.00000⁄ 0

Pairwise Ust were computed with Arlequin. Pairwise genetic Jost’s D values were computed in R with R script from Pennings et al. (2011). bold values indicate significant pvalues 60.05. P-values are in supplementary materials.

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

287

Fig. 8. Statistical parsimony network based on COI (1435 bp) data containing 245 Euproserpinus accessions. Circles represent the 76 unique haplotypes and their sizes are proportional to the number of haplotypes. Colors represent different Euproserpinus populations and pie graphs indicate the proportion of each haplotype represented by different color-coded populations. Black dots indicate missing intermediate haplotypes (mutational steps). The combination of unique haplotypes and geographic isolation support the recognition of the three species designated. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

region, because it would most likely have experienced the opposite temporal patterns of isolation and reconnection. 5. Conclusion For Euproserpinus, the Transverse and Southern Sierra ranges and the Central Coast were the major genetic barriers. South of the Transverse Ranges and east of the Sierras is E. phaeton, north is E. euterpe, and then northwest is the new species. Calsbeek et al. (2003) and Chatzimanolis and Caterino (2007) review many studies that also found the Transverse range to be a major barrier across multiple classes of animal but, again, the divisions were east–west rather than north–south as we found in Euproserpinus. Our study demonstrates that conservation of the Federally listed E. euterpe requires a population-level assessment of genetic diversity. Our data suggests that E. euterpe from Walker Basin and Carrizo Plain should not be considered a single population, because they share no haplotypes in common and have significant genetic differentiation (Ust = 0.72 P-value = 0.00; Jost’s D = 0.99 Pvalue = 0.00); managing them as separate entities dramatically reduces the management areas for the species, particularly the Walker Basin population, which has lost the vast majority of its habitat to anthropogenic disturbance. Thus, although a new population of E. euterpe has been discovered in the Carrizo Plain, each population represents a unique subset of the genetic diversity of the species. The populations in Walker Basin are imminently threatened by agriculture and housing developments, while populations in the Carrizo Plain suffer habitat degradation from sheep grazing and invasive weeds. Both populations require active conservation management and protection. The complex pattern of genetic divergence in Euproserpinus contributes to a finer understanding of the complex biogeographical history of California. Additional markers which are informative at the species/population level may provide important additional resolution regarding the relationships within the genus. Euproserpinus demonstrates the importance of studying taxa whose response to glaciation is likely the inverse of the more

typical mesic-dependent species often studied in the context of Ice Age speciation events. Further research on insects, particularly xerophilic taxa, may provide additional perspectives on the dynamics of speciation both during, and in between past glaciation events. Our study also suggests that more attention be given to insect phylogeography in California. While both E. phaeton and, to a lesser extent, the new species in the Central Coast appear to be relatively secure, patterns of speciation and isolation in Euproserpinus imply the existence of more insect lineages than are currently appreciated, and some may be in need of conservation attention (eg Caterino and Chatzimanolis, 2009). The loss of these species will degrade our ability to understand the phylogeography of one of North America’s most diverse biological provinces. Author contributions D.R. conceived of the study, wrote the paper, collected samples, and conducted some molecular work. M.S. conducted molecular work, ran phylogenetic and population genetic analyses and helped write the paper, P.J. collected samples and assisted with writing, K.O. collected samples and assisted with writing, R.W. collected samples, J. L. R. assisted with molecular work, analysis and writing. Data accessibility DNA sequences: GenBank accessions KP263128-KP263236 Acknowledgements Field assistance, specimens or advice was provided by J. Bundy, J.Smith, L. Crabtree, Peter Jump, P. Opler and D. Bowman. Permits and organizational support were provided by U.S. Fish and Wildlife Service, NPS and BLM. We thank the Kern Primrose Sphinx moth working group, organized by K. Sharum, for long term support and advice. Funding was provided via G. Hinshaw, V. Bloom, and D. Kelly (USFWS). Additional support was provided by A. Kuritsubo

288

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289

and K. Sharum (BLM) in conjunction with T. Longcore (Urban Wildlands group), via agreement L09AC15868-0003-0000 with The Urban Wildlands Group, and the National Park Service (PINN131046) for DNA analysis of specimens from Pinnacles National Park. This work was supported by the USDA National Institute of Food and Agriculture, Hatch project HAW00956-H managed by the College of Tropical Agriculture and Human Resources, University of Hawaii. The authors thank Stellenbosch University’s (South Africa) Department of Botany and Zoology for funding provided through the Ellerman Travel Bursary, which facilitated this collaboration. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biocon.2015.01. 023. References Alexander, M.P., Burns, K.J., 2006. Intraspecific phylogeography and adaptive divergence in the white-headed woodpecker. The Condor 108, 489–508. Althoff, D.M., Segraves, K.A., Leebens-Mack, J., Pellmyr, O., 2006. Patterns of speciation in the yucca moths: parallel species radiations within the Tegeticula yuccasella species complex. System. Biol. 55, 398–410. Avise, J.C., 1998. The history and purview of phylogeography: a personal reflection. Molecular Ecol. 7, 371–379. Axelrod, D.I., 1980. History of the Maritime Closed-cone Pines, Alta and Baja California, Vol. 120. University of California Press, Berkeley. Britten, R.J., 1986. Rates of DNA sequence evolution differ between taxonomic groups. Science 231, 1393–1398. Brower, A.V., 1994. Rapid morphological radiation and convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proc. Nat. Acad. Sci. 91, 6491–6495. Calsbeek, R., Thompson, J.N., Richardson, J.E., 2003. Patterns of molecular evolution and diversification in a biodiversity hotspot: the California Floristic Province. Molecular Ecol. 12, 1021–1029. Caterino, M.S., 2007. Species richness and complementarity of insect faunas in a mediterranean-type biodiversity hotspot. Biodiver. Conservat. 16, 3993–4007. Caterino, M.S., Chatzimanolis, S., 2009. Conservation genetics of three flightless beetle species in Southern California. Conservat. Genet. 10, 203–216. http:// dx.doi.org/10.1007/s10592-008-9548-7. Charif, D., Lobry, J.R., 2007. SeqinR 1.0-2: a contributed package to the R project for statistical computing devoted to biological sequences retrieval and analysis. In: Bastolla, U., Porto, M., Roman, H.E., Vendruscolo, M. (Eds.), Structural Approaches to Sequence Evolution: Molecules, Networks, Populations Biological and Medical Physics, Biomedical Engineering. Springer, New York, pp. 207–232. Chatzimanolis, S., Caterino, M.S., 2007. Toward a better understanding of the ‘‘Transverse Range break’’: lineage diversification in southern California. Evolution 61, 2127–2141. Cicero, C., 1996. Sibling Species of Titmice in the Parus inornatus Complex (Aves: Paridae), vol. 128. University of California Press, Berkeley, pp. 1–217. Clement, M., Posada, D.C.K.A., Crandall, K.A., 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecol. 9 (10), 1657–1659. Coulon, A., Fitzpatrick, J.W., Bowman, R., Stith, B.M., Makarewich, C.A., Stenzler, L.M., Lovette, I.J., 2008. Congruent population structure inferred from dispersal behaviour and intensive genetic surveys of the threatened Florida scrub-jay (Aphelocoma cœrulescens). Molecular Ecol. 17, 1685–1701. Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. JModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 9, 772–772. Delaney, K.S., Wayne, R.K., 2005. Adaptive units for conservation: population distinction and historic extinctions in the Island Scrub-Jay. Conservat. Biol. 19, 523–533. Dimmick, W.W., Ghedotti, M.J., Grose, M.J., Maglia, A.M., Meinhardt, D.J., Pennock, D.S., 1999. The importance of systematic biology in defining units of conservation. Conservat. Biol. 13, 653–660. Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biol. Evolut. 29, 1969–1973. Edwards, H., 1888. Euproserpinus euterpe, a new species of Sphingidae. Entomologica Americana 4, 25–26. Excoffier, L., Lischer, H.E., 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecol. Resourc. 10 (3), 564–567. Feldman, C.R., Spicer, G.S., 2006. Comparative phylogeography of woodland reptiles in California: repeated patterns of cladogenesis and population expansion. Molecular Ecol. 15, 2201–2222. Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R., 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Mar. Biol. Biotechnol. 3, 294–299.

Fu, Y.X., Li, W.H., 1993. Statistical tests of neutrality of mutations. Genetics 133, 693–709. Gompert, Z., Nice, C.C., Fordyce, J.A., Forister, M.L., Shapiro, A.M., 2006. Identifying units for conservation using molecular systematics: the cautionary tale of the Karner blue butterfly. Molecular Ecol. 15, 1759–1768. Haines, W.P., Rubinoff, D., 2012. Molecular phylogenetics of the moth genus Omiodes Guenée (Crambidae: Spilomelinae), and the origins of the Hawaiian lineage. Molecular Phylogenet. Evolut. 65, 305–316. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41, 95–98. Hall, C.A. (Ed.), 2002. Nearshore marine Paleoclimatic Regions, Increasing Zoogeographic Provinciality, Molluscan Extinctions, and Paleoshorelines, California: Late Oligocene (27 Ma) to Late Pliocene (2.5 Ma), Vol. 357. Geological Society of America, USA. Hedin, M., Carlson, D., 2011. A new trapdoor spider species from the southern Coast Ranges of California (Mygalomorphae, Antrodiaetidae, Aliatypus coylei, nov sp), including consideration of mitochondrial phylogeographic structuring. Zootaxa 2963, 55–68. Hedin, M., Starrett, J., Hayashi, C., 2013. Crossing the uncrossable: novel trans-valley biogeographic patterns revealed in the genetic history of low dispersal mygalomorph spiders (Antrodiaetidae, Antrodiaetus) from California. Molecular Ecol. 22, 508–526. Jockusch, E.L., Wake, D.B., 2002. Falling apart and merging: diversification of slender salamanders (Plethodontidae: Batrachoseps) in the American West. Biol. J. Linnean Soc. 76, 361–391. Jost, L., 2008. GST and its relatives do not measure differentiation. Molecular Ecol. 17, 4015–4026. Jump, P.M., Longcore, T., Rich, C., 2006. Ecology and distribution of a newly discovered population of the federally threatened Euproserpinus euterpe (Sphingidae). J. Lepidopterists’ Soc. 60, 41. Kuchta, S.R., 2007. Contact zones and species limits: hybridization between lineages of the California Newt, Taricha torosa, in the southern Sierra Nevada. Herpetologia 63, 332–350. Kuchta, S.R., Tan, A.M., 2005. Isolation by distance and post-glacial range expansion in the rough-skinned newt, Taricha granulosa. Molecular Ecol. 14, 225–244. Kuchta, S.R., Tan, A.M., 2006. Lineage diversification on an evolving landscape: phylogeography of the California newt, Taricha torosa (Caudata: Salamandridae). Biol. J. Linnean Soc. 89, 213–239. Lapointe, F.J., Rissler, L.J., 2005. Congruence, consensus, and the comparative phylogeography of codistributed species in California. Am. Naturalist 166, 290– 299. Law, J.H., Crespi, B.J., 2002. The evolution of geographic parthenogenesis in Timema walking-sticks. Molecular Ecol. 11, 1471–1489. Leonard, J.A., Vila, C., Wayne, R.K., 2005. FAST TRACK: legacy lost: genetic variability and population size of extirpated US grey wolves (Canis lupus). Molecular Ecol. 14, 9–17. Lepage, T., Bryant, D., Philippe, H., Lartillot, N., 2007. A general comparison of relaxed molecular clock models. Molecular Biol. Evolut. 24, 2669–2680. Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25, 1451–1452. Macey, J.R., Strasburg, J.L., Brisson, J.A., Vredenburg, V.T., Jennings, M., Larson, A., 2001. Molecular phylogenetics of western north american frogs of the Rana boylii species group. Molecular Phylogenet. Evolut. 19, 131–143. Matocq, M.D., 2002. Phylogeographical structure and regional history of the duskyfooted woodrat, Neotoma fuscipes. Molecular Ecol. 11, 229–242. McKinney, M.L., 1999. High rates of extinction and threat in poorly studied taxa. Conservat. Biol. 13, 1273–1281. Moritz, C., 1994. Defining ‘evolutionarily significant units’ for conservation. Trends Ecol. Evolut. 9, 373–375. Moritz, C., 2002. Strategies to protect biological diversity and the evolutionary processes that sustain it. Systemat. Biol. 51, 238–254. Osborne, K.H., 1999. Focused survey for Euterpe Sphinx Moth (Euproserpinus euterpe) and assessment of habitats on the 660-acre DeLuca property, Walker Basin. Prepared for U.S. Fish and Wildlife Service, Sacramento, California. Paradis, E., Claude, J., Strimmer, K., 2004. APE: analyses of phylogenetics and evolution in R language. Bioinformatics 20, 289–290. Pennings, P.S., Achenbach, A., Foitzik, S., 2011. Similar evolutionary potentials in an obligate ant parasite and its two host species. J. Evolution. Biol. 24, 871–886. Polihronakis, M., Caterino, M.S., 2010a. Contrasting patterns of phylogeographic relationships in sympatric sister species of ironclad beetles (Phloeodes spp.) in California’s Transverse Ranges. BMC Evolution. Biol. 10, 1–11. Polihronakis, M., Caterino, M.S., 2010b. Multilocus phylogeography of the flightless darkling beetle Nyctoporis carinata (Coleoptera: Tenebrionidae) in the California Floristic Province: deciphering an evolutionary mosaic. Biol. J. Linnean Soc. 99, 424–444. Polihronakis, M., Caterino, M.S., 2012. Asymmetric dispersal and ecological factors mediate geographic effects on phylogeographic structure of the western banded glowworm, Zarhipis integripennis (Coleoptera: Phengodidae) in California’s Transverse Ranges. Annals Entomol. Soc. Am. 105, 241–252. Polihronakis, M., Caterino, M.S., Chatzimanolis, S., 2010. Elucidating the phylogeographic structure among a mosaic of unisexual and bisexual populations of the weevil Geodercodes latipennis (Coleoptera: Curculionidae) in the Transverse Ranges of southern California. Biol. J. Linnaean Soc. 101, 935– 948.

D. Rubinoff et al. / Biological Conservation 184 (2015) 278–289 Posada, D., Crandall, K.A., 2001. Intraspecific gene genealogies: trees grafting into networks. Trends Ecol. Evolut. 16 (1), 37–45. Rambaut, A., 2012. FigTree, version 1.4. University of Edinburgh. . Rambaut, A., Drummond, A.J., 2009. Tracer v1.5. University of Edinburgh. . Ramey, R.R., Wehausen, J.D., Liu, H.P., Epps, C.W., Carpenter, L.M., et al., 2006. Response to Vignieri et al. (2006): should hypothesis testing or selective post hoc interpretation of results guide the allocation of conservation effort? Animal Conservat. 9, 244–247. Rich, K.A., Thompson, J.N., Fernandez, C.C., 2008. Diverse historical processes shape deep phylogeographical divergence in the pollinating seed parasite Greya politella. Molecular Ecol. 17, 2430–2448. Rodríguez-Robles, J.A., Stewart, G.R., Papenfuss, T.J., 2001. Mitochondrial DNAbased phylogeography of north american rubber boas, Charina bottae (Serpentes: Boidae). Molecular Phylogenet. Evolut. 18, 227–237. Roger, A.J., Hug, L.A., 2006. The origin and diversification of eukaryotes: problems with molecular phylogenetics and molecular clock estimation. Philosophical Trans. Royal Soc. B: Biol. Sci. 361, 1039–1054. Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Systemat. Biol. 61, 539–542. Rubinoff, D., Sperling, F.A., 2004. Mitochondrial DNA sequence, morphology and ecology yield contrasting conservation implications for two threatened buckmoths (Hemileuca: Saturniidae). Biol. Conservat. 118, 341–351. Saarinen, E.V., Daniels, J.C., Maruniak, J.E., 2009. Development and characterization of polymorphic microsatellite loci in the endangered Miami blue butterfly (Cyclargus thomasi bethunebakeri). Molecular Ecol. Resourc. 9, 242–244. Sandoval, C., Carmean, D.A., Crespi, B.J., 1998. Molecular phylogenetics of sexual and parthenogenetic Timema walking-sticks. Proc. Royal Soc. London. Series B: Biol. Sci. 265, 589–595. Schoville, S.D., Stuckey, M., Roderick, G.K., 2011. Pleistocene origin and population history of a neoendemic alpine butterfly. Molecular Ecol. 20, 1233–1247. Sgariglia, E.A., Burns, K.J., 2003. Phylogeography of the California thrasher (Toxostoma redivivum) based on nested-clade analysis of mitochondrial-DNA variation. The Auk 120, 346–361.

289

Shields, O., 1990. Field investigation of the Threatened Kern Primrose Sphinx Moth (Euproserpinus euterpe) Habitat. USFWS, Sacramento, (Submitted for publication). Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., Flook, P., 1994. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Annals Entomol. Soc. Am. 87, 651–701. Starrett, J., Hedin, M., 2007. Multilocus genealogies reveal multiple cryptic species and biogeographical complexity in the California turret spider Antrodiaetus riversi (Mygalomorphae, Antrodiaetidae). Molecular Ecol. 16, 583–604. Sukumaran, J., Holder, M.T., 2010. DendroPy: a Python library for phylogenetic computing. Bioinformatics 26, 1569–1571. Tajima, F., 1989. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123, 585–595. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biol. Evolut. 28, 2731–2739. Templeton, A.R., Crandall, K.A., Sing, C.F., 1992. A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics 132, 619–633. Tuskes, P.M., Emmel, J.F., 1981. The life history and behavior of Euproserpinus euterpe (Sphingidae). J. Lepidopterists’ Soc. 35, 27–33. U.S. Fish and Wildlife Service., 1980 Determination that the Kern Primrose Sphinx Moth (Euproserpinus euterpe) is a Threatened Species: U.S. Department of Interior. Federal Register 45, 24088–24090. Vignieri, S.N., Hallerman, E.M., Bergstrom, B.J., Hafner, D.J., Martin, A.P., Devers, P., Grobler, P., Hitt, N., 2006. Mistaken view of taxonomic validity undermines conservation of an evolutionarily distinct mouse: a response to Ramey et al. (2006). Animal Conservat. 9, 237–243. Yanev, K.P., 1980. Biogeography and distribution of three parapatric salamander species in coastal and borderland California. In: Power, D.M. (Ed.), The California islands. Proceedings of a Multidisciplinary Symposium. Santa Barbara Museum of Natural History, Santa Barbara, California, pp. 531–550. Zwickl, D.J., 2006. Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Ph. D. dissertation. The University of Texas at Austin.