Abstract

The Bliss Rapids Snail (Taylorconcha serpenticola) is a threatened species that ranges along a short reach of the middle Snake River in southern Idaho. Additional Taylorconcha populations of uncertain taxonomic status have recently been discovered in other portions of the Snake River basin (Owyhee River, lower Snake River). We investigated the phylogenetic relationships and population structure of these snails, together with two outgroups, using cytochrome c oxidase subunit I (COI) of mitochondrial DNA and the first internal transcribed spacer region between the 18S and 5.8S ribosomal DNA. These data show no sharing of haplotypes or genotypes among T. serpenticola and the Owyhee-Lower Snake populations, with both depicted as monophyletic units within the Taylorconcha clade. Both of these datasets and morphological evidence suggest that the Owyhee-Lower Snake populations are a distinct species, which we describe herein (T. insperata new species). Application of an available COI molecular clock suggests that Taylorconcha arose in the late Miocene, when ancestral Snake River drainage was impounded in an extensive lacustrine system (‘Lake Idaho’) in western Idaho. The shallow population structuring of T. insperata suggests that the lower Snake River was only recently colonized subsequent to incision of Hells Canyon, draining of Lake Idaho, and development of a through-going river in the late Neogene. The absence of significant genetic structure in T. serpenticola, which is attributed to the unstable course and flow regime of the middle Snake River during the Quaternary, suggests that this species can be treated as a single management unit.

INTRODUCTION

Taylorconcha serpenticolaHershler et al., 1994, belonging to a monotypic genus in the family Hydrobiidae and commonly known as the Bliss Rapids Snail, is one of five aquatic gastropods from the Snake River which have been placed on the (United States) Federal List of Threatened and Endangered Wildlife and Plants (USFWS, 1992). This small, gill-breathing species ranges within a short reach (about 138 river km) of the middle Snake River in south-central Idaho (Fig. 1), where it lives on the undersides of gravel, cobbles and boulders in free-flowing segments of the river and in tributary springs that emerge from basalt bluffs that form the Snake River canyon (USFWS, 2005). The USFWS (1992) assigned a threatened status to the species based on impacts to its habitat from hydroelectric development, water use and withdrawal, and competition from the introduced New Zealand Mudsnail, Potamopyrgus antipodarum (Gray, 1853).

Whereas the recovery plan for this and the other federally listed Snake River snails (USFWS, 1995) focused on key habitat and ecology issues, it did not address the need to describe the genetic and evolutionary structure of these species, which is also considered an integral component of enlightened management of species at risk (e.g. O'Brien, 1994; Soltis & Gitzendanner, 1999; Moyle et al., 2003). The Bliss Rapids Snail has not received any genetic or phylogenetic study to date. In a frequently cited, unpublished report, Taylor (1982) stated that the species is a ‘relict survivor of ancient Lake Idaho in southwestern Idaho, dated at about 3.5 million years’ and that ‘prior to dam construction there was probably a single population throughout this [its] range, and plausibly upstream as well.’ If this supposition is correct, then the Taylorconcha lineage may be expected to have shallow terminal structure consistent with population fragmentation in the last century and thus can be managed as a single conservation unit. However, the possibility that even within its narrow geographic range the Bliss Rapids Snail has become subdivided into well-differentiated subsets of populations, which individually may need protection in order to conserve overall diversity, is suggested by its putative Pliocene age and the occurrence of two phenotypic (color) morphs in this species (Frest & Johannes, 1992; Hershler et al., 1994). The recent discovery of Taylorconcha populations of uncertain taxonomic status in other parts of the Snake River basin well outside the previously documented range of the Bliss Rapids Snail (Pentec Environmental Inc., 1991; Frest, 2003; Myers & Foster, 2003; Richards et al., 2005), further underscores the need for genetic and evolutionary studies pertinent to recovery and management. Are these broadly disjunct populations conspecific with the Bliss Rapids Snail or do they represent one or more additional congeners? The discovery of Taylorconcha in Hells Canyon (lower Snake River) (Myers & Foster, 2003), which was not part of the Neogene Lake Idaho basin (Wheeler & Cook, 1954; Fig. 1), is additionally significant because it suggests a need to revise the (brief) biogeographic scenario of the genus outlined by Taylor (1982). Does the occurrence of Taylorconcha in Hells Canyon reflect dispersal following the cutting of this deep gorge and inception of a through-going Snake River in the mid-Pliocene (Wood & Clemens, 2004), orcan it be attributed to an ancestral distribution which pre-dated Lake Idaho?

In this paper we assess the phylogenetic relationships and population structuring of Taylorconcha based on DNA sequence data in order to address the questions posed above and to supplement studies being done in conjunction with the Snake River aquatic species recovery plan (USFWS, 1995). In order to provide a robust analysis, we sequenced both a mitochondrial gene, the first subunit of cytochrome c oxidase (COI), and anuclear marker, consisting of the complete first internal transcribed spacer region (ITS-1) between the 18S and 5.8S ribosomal DNA genes.

MATERIAL AND METHODS

Sampling

We sampled twelve populations of T. serpenticola which span the geographic range of and types of habitats occupied by this species in the middle Snake River (Fig. 1). These samples included representatives of both (red, white) colour morphs of this species. We also analysed samples from two other (broadly disjunct) parts of the Snake River basin (well downflow from the range of T. serpenticola) in which Taylorconcha was recently discovered (Owyhee River, lower Snake River in Hells Canyon). We were unable to sample a population which was recently discovered along the Snake River well upflow of the previously documented range of T. serpenticola (Pentec Environmental Inc., 1991). Multiple specimens (up to five) were sequenced from each sample.

Although the phylogenetic relationships of Taylorconcha have yet to be well established, a preliminary molecular analysis which included numerous other species of the superfamily Rissooidea indicated that the genus is most closely related to the family Amnicolidae (HPL & RH, unpublished). Thus, for this study, trees were rooted with two North American amnicolids, Amnicola limosa (Say, 1817) and Colligyrus greggi (Pilsbry, 1935).

All Taylorconcha specimens utilized in our molecular study were recently collected and preserved in 90% ethanol. Locality details, voucher information and GenBank accession numbers for the Taylorconcha sequences are listed in Table 1. COI sequences for the outgroup species were obtained from GenBank (Amnicola limosa, AF354768; Colligyrus greggi, AY196172).

DNA extraction, PCR amplification and DNA sequencing

Genomic DNA was isolated from individual snails using a CTAB protocol (Bucklin, 1992). For the COI gene, COIL1490 and COIH2198 (Folmer et al., 1994; COIL1490 5′GGTCAAC AAATCATAAAGATATTGG3′ and COIH2198 5′TAAACTT CAGGGTGACCAAAAAATCA3′) were used to amplify a 710 base pair (bp) fragment via polymerase chain reaction (PCR). For the ITS-1 region (including partial 18S, complete ITS-1, and partial 5.8S), MUSSEL18S and WHITE5.8 (Hershler & Liu, 2004; MUSSEL18S 5′TCCCTGCCCTTTGTACACA CCG3′; White, McPheron & Stauffer, 1994; WHITE5.8 5′AGCTRGCTGCGTTCTTCATCGA3′) were used to amplify an approximately 580 bp fragment by PCR. The amplification conditions described in Hershler & Liu (2004) were used. The amplified PCR product was incubated at 37°C for 30 min and then at 85°C for another 15 min with five units of Exonuclease I (ExoI, Amersham) and 0.5 unit Shrimp Alkaline Phosphatase (SAP, Amersham) to cleave nucleotides one at a time from an end of excess primers and to inactivate single nucleotides. Approximately 10–30 ng of cleaned PCR product was used as a template in a cycle-sequencing reaction using the CEQ DTCS Quick Start Kit (Beckman Coulter, Inc.). The following cycling conditions were used: 96°C for 2 min, then 30 cycles of 96°C for 20 s, 45°C for COI and 60°C for ITS-1 for 20 s, and 60°C for 4 min. The cycle-sequenced product was cleaned using the Beckman Coulter protocol. Fluorescent dye-labelled DNA was combined with 4 µl stop solution (equal volume of 100 mM EDTA and 3 M NaOAc pH 5.2), 1 µl glycogen (20 mg/ml), and 10 µl milli-Q H2O, mixed well, and precipitated with 60 µl cold 95% (v/v) ethanol/water. Fluorescentdye-labelled DNA was recovered by centrifuging at 13,000 rpm for 20 min at 4°C. Pellets were washed with 100 µl 70% (v/v) ethanol/water, air dried and resuspended in 30 µl of dimethylformamide. Resuspended samples were added to the appropriate wells of the CEQ sample plate, overlayed with mineral oil, and run on the Beckman Coulter CEQ8000. Sequences were determined for both strands and were edited and aligned using Sequencher™.

DNA sequence analysis

Phylogenetic trees based on distance, parsimony, and maximum-likelihood (ML) methods were generated using PAUP* 4.0b10 (Swofford, 2002). Modeltest 3.06 (Posada & Crandall, 1998) was used to obtain an appropriate substitution model (using the Akaike Information Criterion) and parameter values for the distance and maximum-likelihood analyses. Appropriate genetic distance was used to generate neighbor-joining (NJ) trees for both datasets based on the clustering method of Saitou & Nei (1987). Additional trees based on the ITS-1 dataset were generated using an unweighted pair groups method with arithmetic mean (UPGMA). Bootstrapping with 10 000 replications (as implemented in PAUP) was used to evaluate nodal support. Maximum-parsimony (MP) analyses were conducted with equal weighting, using the heuristic search option with tree bisection reconnection branch-swapping and 100 random additions. Bootstrapping with 10 000 replications (as implemented in PAUP) was used to evaluate nodal support. The appropriate model was applied for the maximum likelihood analyses (ML). A neighbour joining tree with appropriate genetic distance was used as the initial topology for branch-swapping. Nodal support was evaluated by 100 bootstrap pseudoreplicates. Nucleotide divergences (uncorrected p) were calculated using MEGA3 (Kumar, Tamura & Nei, 2004) with standard errors estimated by 500 bootstrap replications. Structuring of genetic variance among subunits of Taylorconcha delineated by the phylogenetic analysis and among the colour morphs observed in the middle Snake River populations was assessed using analysis of molecular variation (AMOVA; Schneider, Roessli & Excoffier, 2000). In the latter analysis, populations were divided into three (colour) groups: white, red and a mixture of both morphs. A molecular clock hypothesis (for the COI dataset) was tested using the likelihood-ratio test (Felsenstein, 1981) based on the ML topology under the best model selected with and without the constraint of a molecular clock. Haplotype networks were generated using the statistical software TCS version 1.18 (Clement et al., 2000) and ARLEQUIN 2.0 (Schneider et al., 2000), which constructsthe most probable minimum spanning network using uncorrected pdistances.

ITS-1 dataset

Sequences were usually verified by forward and reverse comparisons, except for some individuals, which were difficult to sequence through. The ITS-1 region showed intra-individual variation and the difficulties experienced in sequencing suggest that small indels (7 bp) probably occurred. Gaps in the aligned ITS-1 matrix were weighted using three different methods, as suggested by Vogler & DeSalle (1994). First, gaps were treated as missing data and excluded from the analysis. Second, gaps were treated as a ‘fifth’ nucleotide base. Third, each insertion/deletion event was treated as a single character, regardless of the length of gaps. Each method was used to generate the most parsimonious trees, and the resulting topologies were compared and found to be congruent. In the final parsimony analysis, gaps were treated as a ‘fifth’ nucleotide base. Gaps in the aligned ITS-1 matrix were treated asmissing in the UPGMA and ML analyses. Since we were unable unambiguously to align the outgroup taxa in the ITS-1 dataset, the phylogenetic trees based upon this dataset were unrooted and the COI and ITS-1 datasets were analysed separately.

RESULTS

The alignment of 78 sequences from the partial COI gene yielded 658 bp, of which 145 sites were variable (22.0%) and 45 were parsimony informative (6.8%). Nineteen haplotypes were identified from these 78 specimens (see Supplementary Data). Modeltest selected the General Time Reversible Model, with some sites assumed to be invariable and with variable sites assumed to follow a discrete gamma distribution (e.g. GTR+I+G) (Rodriguez et al., 1990), as the best fit for the dataset using the Akaike Information Criterion. The optimized parameter values were base frequencies of A=0.2961, C=0.1941, G=0.1726, T=0.3372; Rmat={0.1697 9.1703 6.4790 0.0000 28.4488}; shape of gamma distribution= 0.7090; and proportion of invariant sites=0.4960. GTR distance was used to generate an NJ tree and GTR+I+G model was used for the ML analyses. Likelihood ratio tests could not reject clock-like behaviour of COI sequences (P=0.99).

Sequence divergences between Taylorconcha and its inferred sister, Colligyrus greggi, ranged from 12.10 to 12.87%. Mean genetic distances (based on p-distance with 500 bootstraps for standard error) between T. serpenticola and the Owyhee-lower Snake populations were 1.31±0.39%, while distances were 0.15±0.05% within the former and 0.18%±0.09% within the latter. The haplotype networks generated using statistical parsimony and minimum-spanning network methods were congruent for both the COI and ITS-1 datasets. Only minimum-spanning networks are figured (below).

In all COI analyses, Taylorconcha haplotypes formed a well supported (99–100%) monophyletic group that was resolved into two shallowly structured subclades (Fig. 2, an NJ tree) composed of T. serpenticola and Owyhee-lower Snake populations. The MP and ML topologies (not figured) were congruent with the NJ tree and differed only in the positioning of terminal branches. The T. serpenticola subclade was not supported by bootstrap analyses, while monophyly of the Owyhee-lower Snake haplotypes received strong support (87–90%). Within the clade composed of T. serpenticola haplotypes the only (weakly) supported (55–61%) subgroup consisted of four (of the five sequenced) specimens from Niagara Springs, the furthest upflow locality that was sampled in this study. The geographically separated Owyhee River and lower Snake River populations were not delineated as phylogenetically distinct subunits in any of the resulting topologies.

Taylorconcha serpenticola and Owyhee-lower Snake populations were also clearly distinguished in the haplotype network (Fig. 3), differing by 6–12 bp. Taylorconcha serpenticola was composed of acommon haplotype observed in 10 of 12 samples (III) and 13 other haplotypes (I, II, IV–XIV) which differ from III by 1–3 bp. Common haplotypes of the Owyhee River (XV) and lower Snake River (XIX) samples differed by a single bp, and three other observed haplotypes (XVI–XVIII) differed from these by 1–3 bp.

Over 85% of the variation in COI occurred between T. serpenticola and Owyhee-lower Snake populations, with small but nonetheless significant amounts also occurring among samples within these groups and within samples (Table 2). AMOVA did not reveal significant variation among colour morphs of T. serpenticola (P=0.81623); most of the variation in this analysis (64%) occurred within samples (Table 2).

Homologous nucleotide sequences from the partial 18S rDNA gene, the complete ITS-1 region, and the partial 5.8S gene were obtained from 61 specimens. The 3′ end of the 18S and the 5′ end of the 5.8S ribosomal gene are highly conserved and therefore easily recognizable. Only one of the 166 nucleotides of the18S was variable. Of the 33 nucleotides of the 5.8S rRNAgene, none was variable. The length of ITS-1 varied from 302–308 bp in specimens from the Owyhee River to 317–335 bp in specimens of T. serpenticola. The total aligned data matrix of the ITS-1 region including indels (insertion/deletions) was 350 bp, of which 67 sites were variable (19.1%) and 60 were parsimony informative (17.1%). Eighteen genotypes were observed among the 61 specimens (see Supplementary Data). Jukes-Cantor (JC) (Juke & Cantor, 1969) was the best fit model selected; JC distance was used to generate an UPGMA tree and was also used for the ML analyses.

Mean genetic distances between T. serpenticola and the Owyhee-lower Snake populations were 2.01±0.59%. Distances were 0.15±0.05% within T. serpenticola and 0.00%±0.00% within the Owyhee-lower Snake populations.

The NJ and UPGMA trees resolved Taylorconcha genotypes into two shallowly structured subclades (Fig. 4) consisting of T. serpenticola and the Owyhee-lower Snake populations. The MP analysis yielded 430 equally parsimonious trees (TL=90, CI=0.80) which also resolved genotypes into these two subclades, and the maximum-likelihood tree was also entirely congruent with this result. Within the T. serpenticola subclade, the only consistently supported group (58–62%) consisted of four–five specimens from the Thousand Springs North (TSN) sample. Neither the Owyhee River nor the lower Snake River populations were segregated in any of the analyses.

Taylorconcha serpenticola and Owyhee-lower Snake populations were also well delineated in the genotype network (Fig. 5), differing by 11 mutations. The most common genotype of T. serpenticola (II) was observed in nine of 12 samples while genotype I, which differed from II by one mutation, was observed in six samples. Other genotypes observed in T. serpenticola (III–XV) differed from the common genotype by one to four mutations. The four genotypes observed in the Owyhee-lower Snake populations differed from one another by one to three mutations. One of these genotypes (XVI) was observed in samples from both broadly disjunct areas (HCH, OWS).

Almost 90% of variation in ITS-1 occurred between T. serpenticola and the Owyhee-lower Snake populations, with a significant amount also occurring within samples (Table 2). AMOVA did not reveal significant variation among colour morphs of T. serpenticola (P=0.13) (Table 2).

DISCUSSION

Taxonomic and biogeographic implications

Our findings indicate that the recently discovered Taylorconcha populations from the Owyhee River and lower Snake River form an evolutionarily distinct (monophyletic) and genetically divergent unit relative to the middle Snake River populations which comprise T. serpenticola. The COI sequence divergencebetween these two snails (1.31±0.39%) is within the range observed among congeners of other North American freshwater rissooidean gastropods (e.g. Tryonia, 1.3–14.8%, Hershler, Liu & Mulvey, 1999; Pyrgulopsis, 1.1–13.1%, Liu & Hershler, 2005). Morphological investigations conducted in conjunction with this study showed that these two snails have diagnostic differences in size, shape of shell, convexity and shouldering of shell whorls, thickness of the adapical portion of the shell aperture, and five anatomical features (see below). The molecular evidence of evolutionary independence and divergence of the Owyhee-lower Snake populations relative to T. serpenticola, together with the morphological differences, provides ample justification for recognizing the former as a new species, which we describe below (T. insperata new species).

Based on a COI molecular clock rate of 1.83±0.21% per million years (Ma) derived for European species of the related rissooidean gastropod family Hydrobiidae (Wilke, 2003), a maximum age of 7.94–5.93 Ma (late Miocene) may be inferred for the Taylorconcha lineage. (Note that the ‘universal’ clock rate [2.0%; Brown et al., 1979] commonly applied to animal mitochondrial studies yields a similar result.) Although obviously tentative given the vagaries of the molecular clock and the non-local calibration thereof, these dates are consistent with a previous assertion that Taylorconcha is a relict survivor of Lake Idaho (Taylor, 1982), a complex lacustrine system that occupied the western Snake River Plain (extending westward into the Owyhee River basin of Oregon) from the late Miocene through the late Pliocene (Othberg, 1994; Wood & Clemens, 2004). Divergence of T. serpenticola and T. insperata is estimated to have occurred 1.05–0.45 mya based on the COI clock discussed above. Although we are not aware of any potential barrier that developed during this time interval which could have split ancestral Taylorconcha (which presumably dispersed only within its aquatic habitat; fideTaylor & Bright, 1987) into these two subunits, we note that Link et al. (2004; Fig. 5) proposed a somewhat earlier subdivision of the Lake Idaho basin into eastern and western segments (separated by a medial high slightly east of Boise) on the basis of differences in depositional facies in these two areas.

The shallow structuring of both subclades may be related to well established aspects of Snake River history. The scant differentiation of Owyhee and lower Snake River populations (which comprise T. insperata) is not consistent with an ancestral distribution that spanned these areas prior to their integration, but instead suggests that the latter may have been founded by downflow dispersal following the incision of Hells Canyon and establishment of the lower course of the Snake River during the late Neogene (Wood & Clemens, 2004). Divergence of T. serpenticola populations along the middle Snake River may have been limited by the repeated impoundment and shifting of the course of the river by local basalt flows during the Quaternary (Stearns, 1936; Malde, 1972, 1982). Isolation of populations may have also been constrained by the catastrophic late Pleistocene Bonneville Flood, which has left abundant evidence of devastation and habitat alteration along the middle Snake River (Malde, 1968; O'Connor, 1993).

Management considerations

Our results provide no basis for recognizing separate, genetically distinctive units within the narrow geographic range of T. serpenticola in the middle Snake River and are consistent with Taylor's (1982) hypothesis that this species formed a single continuous deme prior to fragmentation resulting from recent dam construction. Although our findings suggest that federally listed T.serpenticola can be managed as a single conservation unit, additional studies using more rapidly evolving markers (e.g. microsatellites) will be needed to determine whether this issue may be complicated by possible loss of genetic diversity in those populations which became isolated following construction of middle Snake River dams during the last century. Although our results suggest minimal genetic divergence of the two groups of populations which comprise T. insperata, diagnostic (one bp) variation of COI suggests that these should perhaps nonetheless be treated as separate conservation units (fide King et al., 1999). In addition, the broadly disjunct distributions suggest that genetic exchange between these two groups of populations is unlikely at present.

SYSTEMATIC DESCRIPTION

Genus Taylorconcha Hershler et al., 1994

Taylorconcha Hershler et al., 1994: 233 (type species: Taylorconcha serpenticola Hershler et al., 1994, by original designation).

Taylorconcha insperata new species

Types: Holotype (Fig. 6B) USNM 1018182, Owyhee River at the Cave, 1.6 river km below mouth of Big Antelope Creek at river km 269.4, Malheur County, Oregon (N 4703230, E 481740, elevation 1244 m), 9/25/2002, coll. TF, EJ, CT. Paratypes (from same lot) USNM 1071513.

Etymology: From New Latin insperatus, meaning unexpected, unlooked for, or unanticipated, and referring to our surprise in finding a second species of Taylorconcha.

Referred material: OREGON. Malheur County: USNM 1071510, Owyhee River just west of Oregon State Line at a bar, river km 294.4 (N 4685660, E 497840, elevation 1278 m), 9/25/2002, coll. TF, EJ, CT; USNM 1027147, 1071511, Owyhee River at upper end of Lower Deary Pasture, river km 240.5 (N 4724580, E 477860, elevation 1138 m), 9/24/2002, coll. TF, EJ, CT; USNM 1071512, Owyhee River just downstream of mouth of Sharon Creek at a bar, river km 284.0 (N 4693220, E 490880, elevation 1240 m), 9/25/2002, coll. TF, EJ, CT; USNM 1027075, Owyhee River upstream from South Cross Canyon, (N 4700430, E 482355). 8/20/2004, coll. BB, MS; Wallowa County: USNM 1027076, Snake River, Hells Canyon, above High Bar Rapids (N 5039189, E 535240), 8/26/2004, coll. RH, WC, AF, MS; USNM 1068974, Snake River, Hells Canyon, just below Davis Creek rapids, river km 341.8 (N 5056241, E 539008), 10/2002, coll. DR, MF.

Diagnosis: A species of Taylorconcha readily distinguished from its congener (Fig. 6A) by its larger, broader shell (Table 3); more convex and strongly shouldered teleoconch whorls having impressed sutures; more rounded and thinner adapical portion of aperture; and tan periostracum. Also differs from its congener in having a broader lateral radular tooth face; lighter body pigment; more posteriorly elongate ctenidium; and larger, subterminally positioned female genital aperture.

Description: Shell (Figs 6B, 7A–C) fairly thickened, broadly conical with intact spire; height, 2.42–3.40 mm; whorls 3.25–3.75. Protoconch 1.25 whorls, diameter about 560 µm (Fig. 7D). Teleoconch whorls medium convex, having impressed sutures and rather broad shoulders. Aperture ovate, sometimes weakly angled above. Parietal lip complete, adnate, often thickened. Columellar lip medium width. Outer lip slightly thickened, prosocline. Umbilicus usually perforate, sometimes narrow or rimate. Shell clear-white; periostracum tan. Outer side of operculum smooth or having last 0.5 whorl frilled (Fig. 7E); inner side having attachment region slightly roughened (Fig. 7F). Portion of radular ribbon shown in Figure 7G. Central radular tooth trapezoidal, with short neck; width about 27 µm, cutting edge straight or slightly concave, lateral cusps 5; central cusp long, dagger-like; basal cusps 1–2; basal tongue V- or U-shaped, a little shorter than lateral margin (Fig. 7H). Lateral tooth face rectangular, angled; central cusp parallel-sided, pointed, lateral cusps 2–3 (inner), 3–4 (outer); outer wing broad, flexed, about 150% length of cutting edge (Fig. 7I). Inner marginal teeth having 16–21 cusps (Fig. 7J). Outer marginal teeth having 17–28 cusps (Fig. 7K). Head-foot often entirely pale (except for black eyespots); occasional specimens having black granules scattered on snout and side of foot and black streaks lining edges of cephalic tentacles proximally. Pallial roof, visceral coil lightly to moderately pigmented with scattered black granules; pigment cover heavier on gonads. Cerebral, pleural, and pedal ganglia pale. Penis pale or having light grey internal pigment core. Ctenidium filling most of length of pallial cavity, posterior end positioned a little in front of pericardium; ctenidial filaments about 17, well developed, broadly triangular and fairly tall, without pleats. Osphradium elongate, usually curved at both ends, positioned opposite middle of ctenidium. Hypobranchial gland not evident in dissection; kidney having very short pallial section. Anterior vas deferens having a few small undulations on floor of pallial cavity. Penis somewhat flattened, inner edge usually smooth, sometimes having a few folds in incompletely relaxed specimens (Fig. 8A). Penial duct (not figured) narrow, positioned near outer edge, straight except for a few weak medial undulations. Glandular oviduct and associated structures shown in Figure 8B. Rectum without arch, forming deep furrow (Rf) on glandular oviduct. Seminal receptacle (Sr) superficial on albumen gland (Ag), terminating a little in front of posterior edge of gland. Posterior section of capsule gland (Cg) composed of two orthree distinct white lobes. Genital aperture (Ga) a large subterminal pore.

Distribution: Middle Owyhee River, lower Snake River in Hells Canyon (Fig. 1).

Remarks: Shell measurements of T. insperata and T. serpenticola are summarized in Table 3. Note that these two species differ significantly (P<0.02) in all parameters except relative height of the body whorl.

ACKNOWLEDGEMENTS

Barry Bean, Aaron Foster, Michael Stephenson (Idaho Power Company), and Daniel Gustafson and David Richards (Montana State University) provided assistance in obtaining specimens. Karolyn Darrow and Molly K. Ryan (Smithsonian Institution), and Michael Radko (Idaho Power Company) helped prepare artwork. Paul Link (Idaho State University) provided useful information regarding Snake River history. Tom Quinn (University of Denver) and Sara Oyler-McCance (United States Geological Survey) generously shared bench space and equipment in the Rocky Mountain Center for Conservation Genetics and Systematics. Cindy Church (Metropolitan State College of Denver) provided useful comments on an early draft of this paper, and preparation of the final text benefited from the input of three reviewers. This project was supported, in part, by a contract awarded (to RH) by the Idaho Power Company, and by contracts awarded to Deixis Consultants by the Bureau of Land Management (Vale District) and United States Fish and Wildlife Service (Snake River Basin Office).

Figure 1. Map showing Taylorconcha localities sampled for this study. Locality codes are from Table 1 (note that TS represents three closely proximal sites). Thickened lines along the middle Snake River (call-out) represent dams within the geographic range of T. serpenticola.

Figure 1. Map showing Taylorconcha localities sampled for this study. Locality codes are from Table 1 (note that TS represents three closely proximal sites). Thickened lines along the middle Snake River (call-out) represent dams within the geographic range of T. serpenticola.

Figure 2. Neighbour-joining phylogram for the COI dataset based on GTR distance.

Figure 2. Neighbour-joining phylogram for the COI dataset based on GTR distance.

Figure 3. Minimum-spanning network for COI sequences, showing bp changes between the 19 observed haplotypes (I–XIX). Circles are sized in proportion to haplotype frequency. The filled circle refers to an unobserved haplotype inferred to have existed in Taylorconcha.

Figure 3. Minimum-spanning network for COI sequences, showing bp changes between the 19 observed haplotypes (I–XIX). Circles are sized in proportion to haplotype frequency. The filled circle refers to an unobserved haplotype inferred to have existed in Taylorconcha.

Figure 4. UPGMA phylogram for the ITS-1 dataset based on JC distance.

Figure 4. UPGMA phylogram for the ITS-1 dataset based on JC distance.

Figure 5. Minimum spanning network for ITS-1 sequences, showing mutational changes between the 18 observed genotypes (I–XVIII). Circles are sized in proportion to genotype frequency. The filled circle refers to an unobserved genotype inferred to have existed in Taylorconcha.

Figure 5. Minimum spanning network for ITS-1 sequences, showing mutational changes between the 18 observed genotypes (I–XVIII). Circles are sized in proportion to genotype frequency. The filled circle refers to an unobserved genotype inferred to have existed in Taylorconcha.

Figure 6.Taylorconcha shells. A.T. serpenticola, holotype, USNM 860583. B.T. insperata, holotype, USNM 1018182. Scale bar=1.0 mm.

Figure 6.Taylorconcha shells. A.T. serpenticola, holotype, USNM 860583. B.T. insperata, holotype, USNM 1018182. Scale bar=1.0 mm.

Figure 7. Scanning electron micrographs of shells, opercula and radula of Taylorconcha insperata (USNM 107153). A–C. Shells. D. Shell apex. E. Operculum, outer side. F.Operculum, inner side. G. Portion of radular ribbon. H. Central radular teeth. I. Lateral and inner marginal teeth. J. Inner marginal tooth. K. Outer marginal tooth. Scale bars A–C=1.0 mm; D–F=100 µm; G–K=10 µm.

Figure 7. Scanning electron micrographs of shells, opercula and radula of Taylorconcha insperata (USNM 107153). A–C. Shells. D. Shell apex. E. Operculum, outer side. F.Operculum, inner side. G. Portion of radular ribbon. H. Central radular teeth. I. Lateral and inner marginal teeth. J. Inner marginal tooth. K. Outer marginal tooth. Scale bars A–C=1.0 mm; D–F=100 µm; G–K=10 µm.

Figure 8. Reproductive anatomy of Taylorconcha insperata (USNM 107153). A. Outline of head showing penis (stippled). B. Female glandular oviduct and associated structures (viewed from the left side). Abbreviations: Ag, albumen gland; Cov, coiled oviduct; Cg, capsule gland; Ga, genital aperture; Pw, posterior wall of pallial cavity; Rf, rectal furrow; Sr, seminal receptacle; Vc, ventral channel. Scale bars=500 µm.

Figure 8. Reproductive anatomy of Taylorconcha insperata (USNM 107153). A. Outline of head showing penis (stippled). B. Female glandular oviduct and associated structures (viewed from the left side). Abbreviations: Ag, albumen gland; Cov, coiled oviduct; Cg, capsule gland; Ga, genital aperture; Pw, posterior wall of pallial cavity; Rf, rectal furrow; Sr, seminal receptacle; Vc, ventral channel. Scale bars=500 µm.

Table 1.

Taylorconcha samples used for molecular analysis, with codes (used in Figs 2 and 4), locality details, USNM voucher numbers,* and sample sizes and Genbank accession numbers for the two genes.

Code Locality Voucher COI ITS-1 
Taylorconcha serpenticola 
 NI Niagara Springs, outflow at baseof falls, Gooding Co., ID (N 4725863, E 690263). 1027053 DQ075952–DQ075956 DQ076044–DQ076045 
 BA Banbury Springs outlets, Gooding Co., ID (N 4728344, E 676533). 1020768 DQ076019–DQ076022 DQ076085–DQ076088 
 SA Sand Spring, ca. 30 m below source, Gooding Co., ID (N 4732282, E 678735). 1027070 DQ075961–DQ075965 DQ076048–DQ076052 
 TSN Spring pool of Thousand Springs north outlet, Gooding Co., ID (N 4734582, E 676698). 1027064 DQ075996–DQ076000 DQ076067–DQ076069 
 TSM Thousand Springs, springs just south of the Nature Conservancy's water pipeline, south of north inlet, Gooding Co., ID (N 4734545, E 676699). 1027067 DQ076015–DQ076018 DQ076080–DQ076084 
 TSS Thousand Springs, southern-most and largest outflow of the Minnie Miller Springs complex, Gooding Co., ID (N 4734783, E 676558). 1027065 DQ076001–DQ076005 DQ076070–DQ076074 
 BI Billingsley Creek, at spring source, Gooding Co., ID (N 4738102, E 0676107). 1027069 DQ075957–DQ075960 DQ076046–DQ076047 
 CO Cove Creek, just above diversion to Malad River, Gooding Co., ID (N 4748021, E 674124). 1027071 DQ075966–DQ075970 DQ076053–DQ076056 
 MA Snake River, just below Malad Power Plant outfall, Gooding Co., ID (N 4748010, E 670972). 1027066 DQ076006–DQ076010 DQ076075–DQ076079 
 ZI Snake River, above Bliss Reservoir, Gooding Co., ID (N 4750322, E 668698). 1027072 DQ075971–DQ075975 DQ076057–DQ076059 
 BC Snake River, just below Bancroft Springs, Elmore Co., ID (N 4754963, E 650517). 1027074 DQ075981–DQ075985 DQ076064–DQ076066 
 CL Snake River, just above Clover Creek confluence, Elmore Co., ID (N 4762093, E 648284). 1027073 DQ075976–DQ075980 DQ076060–DQ076063 
Owyhee-lower Snake populations 
 OWS Owyhee River, upstream from South Cross Canyon, Malheur Co., OR (N 4700430, E 482355). 1027075 DQ075986–DQ075990 DQ076028–DQ076031 
 OWD Owyhee River, at Lower Deary Pasture, Malheur Co., OR (N 4724580, E 477860). 1027147 DQ076011–DQ076014 DQ076035–DQ076038 
 HCH Snake River, above High Bar Rapids, Wallowa Co., OR (N 5038990, E 534586). 1027076 DQ075991–DQ075995 DQ076032–DQ076034 
 HCD Snake River, just below Davis Creek rapids, Wallowa Co., OR (N 5056241, E 539008). 1068974 DQ076023–DQ076027 DQ076039–DQ076043 
Code Locality Voucher COI ITS-1 
Taylorconcha serpenticola 
 NI Niagara Springs, outflow at baseof falls, Gooding Co., ID (N 4725863, E 690263). 1027053 DQ075952–DQ075956 DQ076044–DQ076045 
 BA Banbury Springs outlets, Gooding Co., ID (N 4728344, E 676533). 1020768 DQ076019–DQ076022 DQ076085–DQ076088 
 SA Sand Spring, ca. 30 m below source, Gooding Co., ID (N 4732282, E 678735). 1027070 DQ075961–DQ075965 DQ076048–DQ076052 
 TSN Spring pool of Thousand Springs north outlet, Gooding Co., ID (N 4734582, E 676698). 1027064 DQ075996–DQ076000 DQ076067–DQ076069 
 TSM Thousand Springs, springs just south of the Nature Conservancy's water pipeline, south of north inlet, Gooding Co., ID (N 4734545, E 676699). 1027067 DQ076015–DQ076018 DQ076080–DQ076084 
 TSS Thousand Springs, southern-most and largest outflow of the Minnie Miller Springs complex, Gooding Co., ID (N 4734783, E 676558). 1027065 DQ076001–DQ076005 DQ076070–DQ076074 
 BI Billingsley Creek, at spring source, Gooding Co., ID (N 4738102, E 0676107). 1027069 DQ075957–DQ075960 DQ076046–DQ076047 
 CO Cove Creek, just above diversion to Malad River, Gooding Co., ID (N 4748021, E 674124). 1027071 DQ075966–DQ075970 DQ076053–DQ076056 
 MA Snake River, just below Malad Power Plant outfall, Gooding Co., ID (N 4748010, E 670972). 1027066 DQ076006–DQ076010 DQ076075–DQ076079 
 ZI Snake River, above Bliss Reservoir, Gooding Co., ID (N 4750322, E 668698). 1027072 DQ075971–DQ075975 DQ076057–DQ076059 
 BC Snake River, just below Bancroft Springs, Elmore Co., ID (N 4754963, E 650517). 1027074 DQ075981–DQ075985 DQ076064–DQ076066 
 CL Snake River, just above Clover Creek confluence, Elmore Co., ID (N 4762093, E 648284). 1027073 DQ075976–DQ075980 DQ076060–DQ076063 
Owyhee-lower Snake populations 
 OWS Owyhee River, upstream from South Cross Canyon, Malheur Co., OR (N 4700430, E 482355). 1027075 DQ075986–DQ075990 DQ076028–DQ076031 
 OWD Owyhee River, at Lower Deary Pasture, Malheur Co., OR (N 4724580, E 477860). 1027147 DQ076011–DQ076014 DQ076035–DQ076038 
 HCH Snake River, above High Bar Rapids, Wallowa Co., OR (N 5038990, E 534586). 1027076 DQ075991–DQ075995 DQ076032–DQ076034 
 HCD Snake River, just below Davis Creek rapids, Wallowa Co., OR (N 5056241, E 539008). 1068974 DQ076023–DQ076027 DQ076039–DQ076043 

*Additional vouchers are deposited in collections of the Orma J. Smith Museum of Natural History, and Deixis Consultants. Locality information is followed by UTM coordinates (in parentheses), all from Zone 11.

Table 2.

Results of analysis of molecular variance (AMOVA) among Taylorconcha populations.

Source COI ITS-1 
 Variance % of total F Variance % of total F 
Among T. serpenticola and Owyhee-lower Snake populations 3.77 87.77 FCT=0.88* 22.80 89.4 FCT=0.89* 
Among populations within subclades 0.22 5.07 FSC=0.41* 0.07 0.29 FSC=0.03 
Within populations 0.31 7.17 FST=0.93* 2.63 10.32 FST=0.89* 
Among morphs −0.04 −7.49 FCT=−0.07 0.13 3.90 FCT=0.04 
Among populations within morphs 0.21 43.46 FSC=0.40* −0.16 −5.02 FSC=−0.05 
Within populations 0.31 64.03 FST=0.36* 3.28 101.12 FST=−0.01 
Source COI ITS-1 
 Variance % of total F Variance % of total F 
Among T. serpenticola and Owyhee-lower Snake populations 3.77 87.77 FCT=0.88* 22.80 89.4 FCT=0.89* 
Among populations within subclades 0.22 5.07 FSC=0.41* 0.07 0.29 FSC=0.03 
Within populations 0.31 7.17 FST=0.93* 2.63 10.32 FST=0.89* 
Among morphs −0.04 −7.49 FCT=−0.07 0.13 3.90 FCT=0.04 
Among populations within morphs 0.21 43.46 FSC=0.40* −0.16 −5.02 FSC=−0.05 
Within populations 0.31 64.03 FST=0.36* 3.28 101.12 FST=−0.01 

Taylorconcha serpenticola is composed of samples BA, BC, BI, CL, CO, MA, NI, SA, TSN, TSM, TSS and ZI; Owyhee-lower Snake samples are HCD, HCH, OWD and OWS. Colour morphs are pale (BA, BC, CL, MA, NI, TSM, ZI), red (CO, SA) and mixed (BI, TSN, TSS). Significance tests were based on 1023 permutations. Asterisked F values are highly significant (P<0.001).

Table 3.

Shell parameters for Taylorconcha insperata and T. serpenticola and results of t-tests (separate variances) comparing paratype samples.

 WH SH SW HBW WBW AH AW SH/SW SH/HBW SH/AH 
T. insperata           
 holotype 3.75 3.40 2.76 2.84 2.28 1.84 1.68 1.23 1.18 1.85 
 paratypes (10) 
 mean 3.45 2.70 2.35 2.30 1.87 1.51 1.37 1.15 1.18 1.79 
 range 3.25–3.75 2.44–2.93 2.20–2.51 2.14–2.47 1.75–1.98 1.39–1.67 1.30–1.48 1.08–1.21 1.14–1.21 1.68–1.98 
 S.D. 0.158 0.158 0.095 0.107 0.083 0.085 0.057 0.041 0.021 0.089 
T. serpenticola           
 paratypes (30)           
 mean 3.61 2.55 1.93 2.15 1.65 1.33 1.17 1.32 1.19 1.92 
 range 3.50–4.00 2.26–2.78 1.81–2.12 1.93–2.32 1.48–1.82 1.20–1.45 1.08–1.30 1.23–1.43 1.15–1.25 1.78–2.11 
 S.D. 0.142 0.120 0.071 0.089 0.087 0.065 0.062 0.048 0.024 0.087 
T 2.811 2.735 12.791 3.964 7.283 6.025 9.699 11.022 1.548 3.874 
 d.f. 14.2 12.7 12.5 13.4 16.1 12.7 16.6 17.7 17.3 15.2 
P 0.014* 0.017* 0.000* 0.002* 0.000* 0.000* 0.000* 0.000* 0.14 0.001* 
 WH SH SW HBW WBW AH AW SH/SW SH/HBW SH/AH 
T. insperata           
 holotype 3.75 3.40 2.76 2.84 2.28 1.84 1.68 1.23 1.18 1.85 
 paratypes (10) 
 mean 3.45 2.70 2.35 2.30 1.87 1.51 1.37 1.15 1.18 1.79 
 range 3.25–3.75 2.44–2.93 2.20–2.51 2.14–2.47 1.75–1.98 1.39–1.67 1.30–1.48 1.08–1.21 1.14–1.21 1.68–1.98 
 S.D. 0.158 0.158 0.095 0.107 0.083 0.085 0.057 0.041 0.021 0.089 
T. serpenticola           
 paratypes (30)           
 mean 3.61 2.55 1.93 2.15 1.65 1.33 1.17 1.32 1.19 1.92 
 range 3.50–4.00 2.26–2.78 1.81–2.12 1.93–2.32 1.48–1.82 1.20–1.45 1.08–1.30 1.23–1.43 1.15–1.25 1.78–2.11 
 S.D. 0.142 0.120 0.071 0.089 0.087 0.065 0.062 0.048 0.024 0.087 
T 2.811 2.735 12.791 3.964 7.283 6.025 9.699 11.022 1.548 3.874 
 d.f. 14.2 12.7 12.5 13.4 16.1 12.7 16.6 17.7 17.3 15.2 
P 0.014* 0.017* 0.000* 0.002* 0.000* 0.000* 0.000* 0.000* 0.14 0.001* 

Abbreviations: WH, total shell whorls; SH, shell height; SW, shell width; HBW, height of body whorl; WBW, width of body whorl; AH, aperture height; AW, aperture width; T=t value; d.f., degrees of freedom; *, P⩽0.05.

References

BROWN, W.M., GEORGE, M. & WILSON, A.C.
1979
. Rapid evolution of animal mitochondrial DNA.
Proceedings of the National Academy of Sciences
 ,
76
:
1967
–1971.
BUCKLIN, A.
1992
. Use of formalin-preserved samples for molecular analysis.
Newsletter of Crustacean Molecular Techniques
 ,
2
:
3
.
CLEMENT, M., POSADA, D. & CRANDALL, K.
2000
. TCS: a computer program to estimate gene genealogies.
Molecular Ecology
 ,
9
:
1657
–1660.
FELSENSTEIN, J.
1981
. Evolutionary trees from DNA sequences: a maximum likelihood approach.
Journal of Molecular Evolution
 ,
17
:
368
–376.
FOLMER, O., BLACK, M., HOEH, W., LUTZ, R. & VRIJENHOEK, R.
1994
. DNA primers for amplication of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates.
Molecular Marine Biology and Biotechnology
 ,
3
:
294
–299.
FREST, T.J.
2003
. Progress report, Owyhee River mollusk inventory, July 2003. [Unpublished] Report to Bureau of Land Management, Vale District Office, OR.
FREST, T.J. & JOHANNES, E.J.
1992
. Distribution and ecology of the endemic and relict mollusk fauna of Idaho TNC's Thousand Springs Preserve. Report to The Nature Conservancy of Idaho, Boise, ID.
HERSHLER, R., FREST, T.J., JOHANNES, E.J., BOWLER, P.A. & THOMPSON, F.G.
1994
. Two new genera of hydrobiid snails (Prosobranchia: Rissooidea) from the northwestern United States.
Veliger
 ,
37
:
221
–243.
HERSHLER, R., & LIU, H.-P.
2004
. Taxonomic reappraisal of species assigned to the North American freshwater gastropod subgenus Natricola (Rissodiea: Hydrobiidae).
Yeliyer
 ,
47
:
66
–81.
HERSHLER, R., LIU, H.-P. & MULVEY, M.
1999
. Phylogenetic relationships within the aquatic snail genus Tryonia: implications for biogeography of the North American Southwest.
Molecular Phylogenetics and Evolution
 ,
13
:
377
–391.
JUKES, T.H. & CANTOR, C.R.
1969
. Evolution of protein molecules. In:
Mammalian protein metabolism
  (H.N. Monro, ed.),
21
–132. Academic Press, New York.
KING, T.L., EACKLES, M.S., GJETVAJ, B. & HOEH, W.R.
1999
. Intraspecific phylogeography of Lasmigona subviridis (Bivalvia: Unionidae): conservation implications of range discontinuity.
Molecular Ecology
 ,
8
:
S65
–S78.
KUMAR, S., TAMURA, K. & NEI, M.
2004
. MEGA3: Integrated software for Molecular Evolutionary Genetic Analysis and sequence alignment.
Briefings in Bioinformatics
 ,
5
:
2
.
LINK, P.K., MCDONALD, H.G., FANNING, C.M. & GODFREY, A.E.
2004
. Detrital zircon evidence for Pleistocene drainage reversal at Hagerman Fossil Beds National Monument, central Snake River Plain, Idaho. In:
Tectonic and magmatic evolution of the Snake River Plain Volcanic Province
  (B. Bonnichsen, C.M. White & M. McCurry, eds),
105
–119. Idaho Geological Survey, Moscow, ID. [Idaho Geological Survey Bulletin 30]
LIU, H.-P. & HERSHLER, R.
2005
. Molecular systematics and radiation of western North American nymphophiline gastropods.
Molecular Phylogenetics and Evolution
 ,
34
:
284
–298.
MALDE, H.E.
1968
. The catastrophic late Pleistocene Bonneville Flood in the Snake River Plain, Idaho.
United States Geological Survey Professional Paper
 ,
596
:
1
–52.
MALDE, H.E.
1972
. History of Snake River Canyon indicated by revised stratigraphy of Snake River Group near Hagerman and King Hill, Idaho.
United States Geological Survey Professional Paper
 ,
644-F
:
1
–21.
MALDE, H.E.
1982
. The Yahoo Clay, a lacustrine unit impounded by the McKinney Basalt in the Snake River canyon near Bliss, Idaho. In: Cenozoic geology of Idaho (B. Bonnichsen & R.M. Breckenridge, eds),
617
–628. Idaho Geological Survey, Moscow, ID. [
Idaho Bureau of Mines and Geology Bulletin
  26]
MOYLE, L.C., STINCHCOMBE, J.R., HUDGENS, B.R. & MORRIS W.F.
2003
. Conservation genetics in the recovery of an endangered animal species: a review of US endangered species recovery plans (1977–1998).
Animal Biodiversity and Conservation
 ,
26
:
85
–95.
MYERS, R. & FOSTER, A.
2003
. Benthic macroinvertebrates of Hells Canyon. Hells Canyon Complex FERC No. 1971, Technical Report Appendix E.3.1-8, Idaho Power Company, Boise, ID. Available from http://www.idahopower.com/riversrec/relicensing/hellscanyon/hellspdfs/techappendices/Aquatic/e31_08.pdf [Accessed 6/1/2005].
O'BRIEN, S.J.
1994
. A role for molecular genetics in biological conservation.
Proceedings of the National Academy of Sciences
 ,
91
:
5748
–5755.
O'CONNOR, J.E.
1993
. Hydrology, hydraulics, and geomorphology of the Bonneville Flood.
Geological Society of America Special Paper
 ,
274
:
1
–83.
OTHBERG, K.
1994
. Geology and geomorphology of the Boise Valley and adjoining areas, western Snake River Plain, Idaho.
Idaho Geological Survey Bulletin
 ,
29
:
1
–54.
PENTEC ENVIRONMENTAL INC.
1991
. Distribution survey of five species of molluscs, proposed for endangered status, in the Snake River, Idaho during March 1991. Report to Idaho Farm Bureau, Boise, ID.
POSADA, D.K. & CRANDALL, K.A.
1998
. Modeltest: testing the model of DNA substitution.
Bioinformatics
 ,
14
:
817
–818.
RICHARDS, D.C., FALTER, C.M., LESTER, G.T. & MYERS, R.
2005
. Responses to FERC Additional Information Request AR-2. Listed Mollusks. Final Report. Hells Canyon Project FERC No. P-1971-079. Idaho Power Company, Boise, ID. Available from http://www.ipchydro.org/pdfs/hellscanyon/aiipc/ipc20050204_1.pdf [Accessed 6/1/2005].
RODRIGUEZ, F., OLIVER, J.F., MARÍN, A. & MEDINA, J.R.
1990
. The general stochastic model of nucleotide substitution.
Journal of Theoretical Biology
 ,
142
:
485
–501.
SAITOU, N. & NEI, M.
1987
. The neighbour-joining method: a new method for reconstructing phylogenetic trees.
Molecular Biology and Evolution
 ,
4
:
406
–425.
SCHNEIDER, S., ROESSLI, D. & EXCOFFIER, L.
2000
. ARLEQUIN ver. 2.00: A software for population genetics data analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland. Available from http://lgb.unige.ch/arlequin/software/ [Accessed 6/1/2005].
SOLTIS, P.S. & GITZENDANNER, M.A.
1999
. Molecular systematics and the conservation of rare species.
Conservation Biology
 ,
13
:
471
–483.
STEARNS, H.T.
1936
. Origin of the large springs and their alcoves along the Snake River in southern Idaho.
Journal of Geology
 ,
44
:
429
–450.
SWOFFORD, D.L.
2002
.
PAUP*: Phylogenetic Analysis Using Parsimony (and other methods)
 . Version 4.0b10. Sinauer Associates, Sunderland, MA.
TAYLOR, D.W.
1982
. Status report on Bliss Rapids Snail. Report to the United States Fish and Wildlife Service, Portland, OR.
TAYLOR, D.W. & BRIGHT, R.C.
1987
. Drainage history of the Bonneville Basin. In:
Cenozoic Geology of Western Utah. Sites for Precious Metal and Hydrocarbon Accumulations
  (R.S. Kopp & R.E. Cohenour, eds). Utah Geological Survey, Salt Lake City, UT. [Utah Geological Association Publication 16]
USFWS (UNITED STATES FISH AND WILDLIFE SERVICE).
1992
. Endangered and threatened wildlife and plants; determination of endangered or threatened status for five aquatic snails in south central Idaho.
Federal Register
 ,
57
:
59242
–59257. Available from http://ecos.fws.gov/docs/frdocs/1992/92-30174.pdf [Accessed 6/1/2005].
USFWS.
1995
. Snake River aquatic species recovery plan. Snake River Basin Office, Boise, ID. Available from http://idahoes.fws.gov/BOR/Final.pdf [Accessed 6/1/2005].
USFWS.
2005
. Biological opinion for Bureau of Reclamation operations and maintenance in the Snake River basin above Brownlee Reservoir. Snake River Fish and Wildlife Office, Boise, ID. Available from http://idahoes.fws.gov/BOR/Final.pdf [Accessed 6/1/2005].
VOGLER, A.P. & DESALLE, R.
1994
. Evolution and phylogenetic information content of the ITS-1 region in the Tiger Beetle Cicindela dorsalis.
Molecular Biology and Evolution
 ,
11
:
393
–405.
WHEELER, H.E. & COOK, E.F.
1954
. Structural and stratigraphic significance of the Snake River capture, Idaho-Oregon.
Journal of Geology
 ,
62
:
525
–536.
WHITE, L.R., McPHERON, B.A. & STAUFFER, J.R.
1994
. Identification of freshwater mussel glochidia on host fishes using restriction fragment length polymorphisms.
Molecular Ecology
 ,
3
:
183
–185.
WILKE, T.
2003
. Salenthydrobia gen. nov. (Rissooidea: Hydrobiidae): apotential relict of the Messinian salinity crisis.
Zoological Journal of the Linnean Society
 ,
137
:
319
–336.
WOOD, S.H. & CLEMENS, D.M.
2004
. Geologic and tectonic history of the western Snake River Plain, Idaho and Oregon. In:
Tectonic and magmatic evolution of the Snake River Plain Volcanic Province
  (B.Bonnichsen, C.M. White & M. McCurry, eds),
69
–103. Idaho Geological Survey, Moscow, ID. [Idaho Geological Survey Bulletin 30].