B. Synthesize results from previous monthly population and habitat studies at BLNWR, Blue Spring, and Willow Spring.

Data synthesis linking physicochemical, habitat and population data for P. chupaderae was completed during this segment. Similar efforts continue under the current grant segment for P. pecosensis. This will facilitate assessment of patterns of hydrobiid abundance and habitat metrics with population parameters.

C. Conduct routine population and habitat monitoring of macroinvertebrate species listed in Table 1 *.

1. Field monitoring will consist of population abundance estimation, habitat assessment, life history observations, and identification of threats.

a. Estimate population densities by benthic samples, dip nets, and/or artificial substrate samples from all occupied habitat types.

b. Habitat quantification will include measures of water depth and velocity, substrate type, and hydrophytes.

c. Measure physicocheinical parameters (water temperature, salinity, specific conductance, total dissolved solids, dissolved oxygen, and pH) at sample sites.

See Appendix A for details.

B. Expand exploratory surveys for target species listed in Table 1 with a particular einphasis on documenting the status (distribution and abundance) of Pyrgulopsis spp. in the Gila River Basin*.

Pyrgulopsis populations in the Gila River basin were not monitored during this grant segment.

(* Pending successfU1 landowner-agency liaison relations.)

Investigate the taxonomic status of the Sangre de Cristo peaclan by reinspection of field voucher material collected from previous surveys.

The Project Biologist reports no activity under this task, which will require substantial commitment of funding to support morphometric and genetic studies. These tasks are contingent on finding an adequate sample size of living specimens of Pisidium sanguinichristi at Middle Fork Lake (type locality; TL), or from another locality which is currently not known to exist since P. sanguinichristi is only reported from the TL.

Appendix B provides published results of a phylogenetic study of the Assimineapecos species complex of New Mexico, Texas, and Mexico. Submit annual reports summarizing activities during the reporting period. These activities will include preliminary analysis of results, identification of threats, and management recommendations.

Annual reports were submitted under all previous grant segments.

E. Prepare a completion report which summarizes the results of work accomplished under Procedures A-D, including management recommendations for species-specific long-tenn populatioil/habitat monitoring protocols and assessment of factors posing imminent threats to target species.

See Appendix C.

111. Geographic Location

Laboratory: Project headquarters will be located at the New Mexico Department of Game and Fish, Santa Fe, NM.

Field: Populatioill habitat monitoring will be conducted in target species' native habitats. Exploratory field investigations will occur within target species' known or presumed historic range. /

Prepared by: Approved by: Brian K. Lang ~e/nae d, Assistant Chief Project Biologist Conservation Services Divisioil

Approved by: rv(afi Wunder, Chief ~edhralAid Coordinator ~ofiservationServices Division Table 1. State listed and federal Candidate and Species of Concern aquatic mollusks and crustaceans of New Mexico. Species categorized by ecological specialization, geographic reson, and conservation status.

2008 StatusL ~axa' Species County State Federal occurrence3

Spring Snails Pecos assiminea Assiminea pecos Chaves E C BLNWR Koster's tryonia* Juturnia kosteri Chaves E C BLNWR, RCC Roswell pyrg* Pyrgiilopsis roswellensis Chaves E C BLNWR, RCC Pecos pyrg* Pyrgulopsis pecosensis Eddy T SC Blue Spring Chupadera pyrg* Pyrgulopsis chupadei-ae Socorro E C Chupadera Mts. Gila pyrg* Pyrguloysis gilae Grant T C Gila River Basin New Mexico hot spring pyrg* Pyrgulopsis thermalis Grant T C Gila River Basin

Aquatic Snails & Bivalves wrinkled marshsnail Stagnicola caperatus Sandoval VCNP, BLNWR star gyro Gyrnulus crista Colfax Black Lake lake frngernailclam Musculium lacustre Colfax Cieneguilla Creek paper pondshell mussel Utterbnckia imbecillis San Miguel Canadian River swamp fingernailclam Musculium partumeium Union Arkansas River Basin long fingernailclam Musculium transversum Union Arkansas River Basin Texas hornshell mussel Popenaias yopeii Eddy Pecos River Lilljeborg's peaclam Pisiiiiun~lilljeborgi Santa Fe Sangre de Cristo Mts. Sangre de Cristo peaclam* Pisiiiium sanguinichristi Taos Sangre de Cristo Mts.

Crustaceans Noel's amphipod* Ganzmarus desper-atus Chaves E SC BLNWR

Taxonomic authorities: (a) Turgeon, D. D., A. E. Bogan, E.V. Coan, W.K. Emerson, W. G. Lyons, W. L. Pratt, C. F. E. Roper, A. Scheltema, F. G. Thompson, and J. D. Williams. 1988. Common and scientific names of aquatic invertebrates fkom the United States and Canada: mollusks. American Fisheries Society Special Publication 16; @) Hershler, R. and F. C. Thompson. 1987. North American Hydrobiidae (Gastropoda: Rissoacea): Redescription and systematic relationships of Tryonia Stimpson, 1865 and Pvr~ulovsisCall and Pilsbry, 1896. The Nautilus 101 (1):25-32. *2008 Status: (State) E = Endangered, T = Threatened; (Federal) C = Candidate, SC = Species of Concern - USFWS. 1996. Endangered and Threatened Wildlife and Plants; Review of Plant and Animal Taxa That are Candidates for Listing as Endangered or Threatened Species. 50 CFR Part 17 (7595-7613). Acronyms: BLNWR - Bitter Lake National Wildlife Refuge; RCC - Roswell Country Club; VCNP - Valles Caldera National Preserve. * Species endemic to New Mexico. Occurrence is last known presence of taxa when monitoring was conducted or access was granted for survey. Appendix A. Aquatic macroinvertebrate population and habitat monitoring (2006-2007). Aquatic Macroinvertebrate Population and Habitat Monitoring (2006-2007)

During this segment, population and habitat monitoring occurred at Bitter Lake National Wildlife Refuge and at Blue Spring. Lack of private land access prevented monitoring of P. chupaderae at Willow Spring.

Bitter Lake National Wildlife Refuge

Overview: To reduce take of federally listed taxa at Bitter Lake National Wildlife Refuge (BLNWR), macroinvertebrate population monitoring is being coordinated with Drs. David Berg and Makiri Sei, Miami University (MU; Oxford, OH), under a IVSF grant titled, "RUI: Patterns of biodiversity of benthic invertebrates in Chihuahuan Desert springs." Sites on BLNWR targeted for this NSF study are Sago Spring and Bitter Creek (BC) at the "Lost River pool confluence."

The NSF inventory protocol calls for quarterly sampling over a 1- year period in each system and employs quantitative effort per the PLOCH method for pond biodiversity studies (Oertli et al. 2005). Following this method, the number of quantitative samples (benthic and sweep) varies exponentially according to the size of the "pond habitat" (e.g., 4 samples each for a 10 m2 pond, 8 for 100 m2, 16 for 1000 m2 and 32 for 10,000 m2). Furthermore, this protocol is being applied to assess longitudinal species richness commencing at spring sources ("pond habitat") and continuing at exponential intervals downstream from there, i.e., springhead, 5m, 25m, and 125m. Accordingly, a total of 18 samples (1 1 benthic, 7 sweep) is collected quarterly for one year in both BLNWR study sites.

While MU'S methods are different than the biannual monitoring approach used by Lang (1 998), there are some similarities, differences, and advantages. Benthic samples are effected using the same technique and sampler that the project biologist has used to monitoring sites on BLNWR since 1995. Lang did not use sweep sampling techniques. Standard physicochemical data are recorded at up- and downstream reaches of the survey area, whereas Lang recorded these data for & sub-sample. While both approaches inventory the entirety of the Sago Spring system, a shorter reach of BC (1 25m) is currently being monitored than in the past (Lang sampled the entire system, ca. 1.0 mile). Per the NSF "Research at Undergraduate Institutions" funding program, MU has access to students for the arduous task of processing macroinvertebrate collections on a timely basis.

Preliminary Results: All collections from 2007 are sorted and voucher data has been processed. The abundance of aquatic macroinvertebrates in both Bitter Creek and Sago Spring appear commensurate to past levels, except for Gammarus desperatus in the BC system. Lang (2005) reported similar survey results and attributed range reductioil of G. desperatus within BC at Lost River pool to possible post-fire effects from the March 2000 Sandhill Fire. G. desperatus at that time appeared restricted to just two spring vents in the this system: one vent located in "Dragonfly Spring Run" and the other located at the "Lost River pool confluence." The current absence from the latter site simply may be due to survey methods, as Lang (2005) sampled with long-handled dip nets inside the spring vents, whereas sampling in 2007 (NSF survey protocol) was effected by benthic grabs and pelagic sweep samples.

The habitat of Sago Spring remains as observed since 1995-stable. As Lang (2005) reported, post-fire encroachment of the common reed, Phragmites australis, throughout the Bitter Creek study area continues, especially in lotic waters of Bitter Creek proper. This growth results in stein densities and root mats that have rendered benthic sampling at two historic samples sites, "Dragonfly Spring" and "Dragonfly Spring Run", virtually ineffectual.

Spring Feeder Ditch Survey: On 8 August 2007,3 sites were surveyed in the Spring Feeder Ditch along the western limits of Refuge impoundinents for the 4 listed invertebrates. While none of these species were observed, numerous invertebrate taxa (odonates, Hyalella amphipods, pulmonate snails [Physella virgata, Lymnaea obrussa], coelopterans), fishes (greenthroat darter, Pecos pupfish, mosquitofisl~,common carp), inacrophytes (chara, widgeongrass) and anergents (Juncus spp., Eleocharis sp.) have colonized this recently created habitat.

Blue Spring

Blue Spring is also included as a study site for MU'S NSF grant. Three of 4 quarterly inventories were completed during this grant segment. The population and habitat of Pyrgulopsis pecosensis at Blue Spring are currently stable. Literature Cited

Lang, B. K. 1998. Status of aquatic mollusks of New Mexico. New Mexico Department of Game and Fish, Completion Report (E-20-6) submitted to the Division of Federal Aid, U. S. Fish and Wildlife Service, Region 2, Albuquerque, hTM.

Lang, B. K. 2005. Macroinvertebrates of Bitter Lake National Wildlife Rehge. New Mexico Department of Game and Fish, Completion Report, E-56 (1 -3), submitted to the Division of Federal Aid, U. S. Fish and Wildlife Service, Region 2, Albuquerque, NM.

Oertli, B., D. A. Joye, E. Castella, R. Llejuge, A. Lehnann, J.-B. Lachavanne. 2005. PLOCH: a standardized method for sampling and assessing the biodiversity in ponds. Aquatic Conservation: Marine and Freshwater Ecosysteins 15: 665-679. Appendix B. Assiminea pecos phylogenetic study. Hydrobiologia (2007) 579:317-335 DO1 10.10071~10750-006-0473-9

Genetic and morphologic variation of the Pecos assiminea, an endangered mollusk of the Rio Grande region, United States and Mexico (Caenogastropoda: Rissooidea: Assimineidae)

Robert Hershler Hsiu-Ping Liu . B. K. Lang

Received: 23 May 2006 I Revised: 1 November 2006 I Accepted: 11 November 2006 I Published online: 13 January 2007 @ Springer Science+Business Media B.V. 2007

Abstract Assiminea yecos is an endangered of A. pecos, which was consistently depicted as a species of an~phibiousgastropod that occupies monophyletic unit nested within or sister to the four widely separated portions of the Rio Grande shallowly structured group comprised of Ameri- region in the southwestern United States (Pecos can members of this species. Consistent with our River basin) and northeastern Mexico (Cuatro findings, we describe the Mexican population as a Cienegas basin). Our statistical and discriminant new species, which is provisionally placed in the function analyses of shell variation among the large, worldwide genus Assiminea pending fur- disjunct populations of this species indicate that ther study of the phylogentic relationships of the Mexican specimens differ in their morphometry North American assimineids. Our n~oleculardata from those of the United States and can be suggest that the Rio Grande region assimineids, diagnosed by several characters. We also ana- which are among the few inland members of the lyzed variation in the mitochondrial genome by otherwise estuarine subfamily Assimineinae, di- sequencing 658 bp of mitochondrial COI from verged from coastal progenitors in the late Mio- populations of A. yecos, representatives of the cene, with subsequent Pleistocene vicariance of other three North American species of Assiminea, Mexican and American species perhaps associ- and several outgroups. Our results indicated ated with development of the modern, lower substantial divergence of the Mexican population course of the Rio Grande.

Handling editor: K. Martens Keywords Assiminea . Gastropoda . North America . Systematics . Biogeography R. Hershler (m) Conservation Department of Invertebrate Zoology, Smithsonian ~nstitution,NHB W-305, MRC 163, P.O. Box 37012, Washington, DC 20013-7012, USA Introduction e-mail: [email protected]

H.-P. Liu The Assimineidae is a diverse family of small, Department of Biological Sciences, University of amphibious and terrestrial caenogastropods found Denver, Denver, CO 80208, USA in tropical and temperate regions throughout much of the world (Abbott, 1958; Ponder & DeKeyzer, B. K. Lang New Mexico Department of Game and Fish, 1998). The North American fauna consists of four One Wildlife Way, Santa Fe, NM 87507, USA amphibious species currently assigned to the large,

a Springer Hydrobiologia (2007) 579:317-335

cosmopolitan genus Assiminea, and referable to Pecos assiminea, although several additional the informal nitida-complex (sensu Abbott, 1958). populations have been reported from the Roswell Two of these species share the brackish coastal and Ft. Stockton areas, and another isolated habitats typical of other members of the subfamily segment of Pecos River drainage near Balmo- Assimineinae (Fukuda & Ponder, 2003) and rhea, Texas (USFWS, 2002,2005). have extremely broad ranges along the Pacific As currently envisaged, the Pecos assiminea is (A. californica [Tryon]) and Atlantic-Gulf of composed of four groups of populations separated Mexico (A. sz~ccinen[Pfeiffer]) margins (Abbott, from each other by ca. 75-760 km (Fig. 1). Given 1974). The other two live in spring-riparian settings its tight linkage with aquatic habitats, A. pecos is in Death Valley (A. injima Berry) and the Rio likely incapable of dispersing long distances within Grande region (A. pecos Taylor) and are among the highly fragmented drainage of the arid Rio the few representatives of this subfamily associ- Grande region and consequently its broadly dis- ated with inland aquatic habitats (Fukuda & junct populations are probably not exchanging Ponder, 2003). Both of these "land-locked" spe- genes at present. The fossil record of the Pecos cies are thought to have been derived from coastal assiminea consists of a few Holocene collections progenitors that dispersed inland during the late from sites closely proximal to extant colonies Tertiary (Berry, 1947; Taylor, 1985) and conse- (Taylor, 1987) and does not provide insight into quently are of biogeographic interest. They have the timing of population isolation. Taylor (1985), also become a focus of conservation attention however, speculated that this snail (which was then because of threats to their fragile, groundwater- undescribed) evolved from Gulf Coastal progen- reliant habitats - both have critically imperiled itors that dispersed along a hypothesized brackish (GI) global heritage rankings (Natureserve, 2006) paleodrainage in the Rio Grande Valley, with and A. pecos was recently listed as endangered by isolation of inland, salt-adapted populations occur- the USWFS (2005). However, despite the compel- ring upon inception of the modern, freshwater ling and important features of the North American lower Rio Grande during the Pleistocene. If this assimineids, these snails are poorly known taxo- hypothesis is correct, Mexican (Cuatro Cienegas) nomically and additional studies are needed to and American (Pecos River basin) populations better clarify their species limits and assess their separated by the putative Rio Grande barrier have phylogenetic relationships. long been isolated and may represent divergent Assiminea pecos, commonly known as the conspecific subunits, or separate species. Pecos assiminea (Turgeon et al., 1998), was In this paper we survey morphological and described from a small number of sites in the mitochondria1 DNA variation of A. pecos popu- upper Pecos River basin near Roswell, New lations, and assess their evolutionary relationships Mexico (including the type locality); lower seg- using nlolecular evidence. We use our results to ment of this watershed near Fort Stockton, Texas; re-evaluate the current taxonomic treatment of and Cuatro Cienegas basin, Coahuila (Mexico), these populations as a single species and test which drains (in part) to the Rio Grande (Taylor, Taylor's (1985) hypothesis of the biogeographic 1987). Taylor (1987) clearly differentiated A. history of assimineids in the Rio Grande region. pecos from its North American congeners on Additionally, we discuss the implications of our the basis of its strongly rounded shell whorls, findings for ongoing efforts to recover the endan- deep suture, absence of a subsutural thread, and gered Pecos assiminea (NMDGF, 2004). broad umbilicus. However, conspecificity of the widely separated populations assigned to A. pecos was not convincingly demonstrated in this study Materials and methods as the only (shell) n~oi-phologicaldata provided were from topotypes and all but one of the Genetics figured specimens were also from the type locality (Taylor, 1987: Table 1. Fig. 2a-d). There have Samples from seven populations of A. pecos been no subsequent taxonomic studies of the spread across its entire geographic range were

Springer Hydrobiologia (2007) 579:317-335 319

Ag. 1 Map showing the four disjunct areas in the Rio Grande region, United States and Mexico, which comprise the geographic range of Assiminen pecos collected between 2000-2004, preserved in 90% species. Locality and other data for all samples ethanol, and used for sequencing of mitochon- are provided in Table 1. drial DNA. Two to seven animals were analyzed Genomic DNA was isolated from individual for each population, totaling 30 specimens. We snails using a CTAB protocol (Bucklin, 1992). Six also analyzed small samples (three-five speci- hundred fifty-eight base pairs (bp) of mitochon- mens) of each of the other three North American drial cytochrome c oxidase subunit I were ampli- species of Assiminea. Inasmuch as the phyloge- fied and sequenced with primers COIL1490 and netic relationships of the North American assi- COIH2198 (Foln~eret al., 1994) following proto- mineids have never been studied, we used as cols of Liu et al. (2003). Sequences were deter- outgroups two members of the family from mined for both strands and then edited and geographically distant Taiwan-Paludinella tai- aligned using SequencherT" version 3.1 .I. New wanensis Habe, Pseudomyhala latericea (H. & A. sequences were deposited in GenBank (Acces- Adan1s)-and rooted our trees with the latter sion numbers DQ533841-533866).

a Springer Table 1 Samples used for sequencing of COI v, T. Species Code Area Locality and voucher Haplotypes GenBank Accession # Sample size % (if available) Assiminea pecos Roswell Impoundment #7, northwest corner, Bitter Lake NWR, Chaves Co., NM; USNM 1007246, USNM 1069820 Roswell Sago Springs, Bitter Lake NWR, Chaves Co., NM; USNM 1011495 Ft. Stockton Monsanto Spring, Diamond Y 11, I11 Draw, Pecos Co., TX Ft. Stockton Diamond Y Spring, ca. 200- VI, VII, VIII, IX 250 m downflow from source, Diamond Y Draw, Pecos Co., TX; USNM 1069823 Ft. Stockton "Johns Hole," ca. 0.8 krn west of TX Hwy 18, Diamond Y Draw, Pecos Co., TX; USNM 1007143, USNM 1085823 Balmorhea East Sandia Spring, Reeves Co., VI, Lx TX, USNM 1069829 Cuatro Cienegas Spring-marsh complex just west I of Hwy 30. ca. 2.5 krn north of Poza de la Becerra, Coahuila, Mexico, USNM 1085799 Assitninea cal(fornica Point San Pablo Yacht Harbor, San Pablo Bay, Contra Costa Co., CA; USNM 1011440 Assimiizea s~rccitlea Nueces Bay, ca. 1.6 km northwest of Indian Point. west side TX Hwy 181, San Patncio Co., TX; USNM 1011486 Assiminea infirna Badwater, Death Valley, Inyo Co., CA; USNM 1068665 Pseudomphnla latericen Hao Mei Li Natural Ecological Preservation Area, Bu Dai, Chiayi Co., Taiwan; USNM 1087367 Poludinella taiwnnensis Gutter along railway, Chi-Chi, Nantou Co., Taiwan; USNM 1087368, USNM 1087369, USNM 1087370 Hydrobiologia (2007) 579:317-335

Base composition differences were evaluated adult specimens with fully formed inner shell lips using the Chi-square test. Phylogenetic relation- were selected from amongst the largest specimens ships were inferred using maximum parsimony of each sample. The total number of shell whorls (MP), neighbor-joining distance (NJ), maximum was counted (WH) for each specimen; and the likelihood (ML), and Bayesian inference meth- height and width of the entire shell (SH, SW), ods. MP, NJ, and ML analyses were performed body whorl (HBW, WBW), and aperture (AH, using PAUP"4.0b10 software (Swofford. 2002) AW) were measured from camera lucida outline and Bayesian analyses were conducted using drawings using a digitizing pad linked to a MrBayes 3.04 (Huelsenbeck & Ronquist, 2001). personal computer (see Hershler, 1989). In addi- MP analyses employed equal weighting, using the tion, three ratios that estimate aspects of shell heuristic search option with 100 random addi- shape were generated from the raw data (SWISH, tions. Modeltest 3.7 (Posada & Crandall, 1998) HBWISH, AHISH). Descriptive statistics were was used to determine which evolutionary model generated for each sample; and sample heteroge- best fits the data under the Akaike Information neity was examined through analysis of variance Criterion, which was then used to construct NIL (ANOVA), with post-hoc testing of differences and Bayesian trees. The NJ tree was generated among means using the Bonferroni correction for using genetic distances that were also based on multiple comparisons. Discriminant analysis was this model. For the Bayesian analyses, several used to evaluate the extent to which these short runs were first conducted using the default samples could be differentiated on the basis of random tree option to determine when the log the multivariate dataset (excluding ratios). This likelihood sum reached a stable value (by plotting technique has been successfully used to differen- the log-likelihood scores of sample points against tiate closely similar species of other assimineid generation time). Metropolis-coupled Markov genera (Fukuda & Ponder, 2003, 2005). Classifi- chain Monte Carlo simulations were then run cation matrices based on canonical scores were with four chains for 1,000,000 generations, and generated to assess accuracy of assignment of Markov chains were sampled at intervals of 10 individual specimens to their samples, and the generations to obtain 100,000 sample points. The first and second scores were presented as a sampled trees with branch lengths were used to bivariate plot to enable visual assessment of generate a 50% majority rule consensus tree with sample differentiation. A second set of discrimi- the first 5000 trees (equal to 50,000 generations) nant analyses was conducted using a dataset in removed to ensure that the chain sampled on a which the mensural parameters were log trans- stationary portion. Node support was evaluated formed (base 10). All analyses were performed by 10,000 bootstrap pseudo-replicates except for using Systat for Windows 11.00.01 (SSI, 2004). the ML analysis, in which support values were Shells were further studied and photographed based on 100 replications. using scanning electron microscopy (SEM). Var- Sequence divergences (uncorrected p distance) iation in other aspects of snail n~orphology(e.g., within and between phylogenetic lineages were radula, soft part anatomy) was also examined calculated using MEGA3 (Kumar et al., 2004); using standard methods (Hershler, 1998; Hershler standard errors were estimated by 1000 bootstrap et al., 2006a). but was constrained by the quality replications with pairwise deletion of missing and quantity of available material. data.

Morphology Results

Standard shell parameters were compared for Genetics single samples from each of the four disjunct parts of the Pecos assiminea's geographic range. Sam- The alignment of the COI dataset yielded 658 bp, ple sizes were not sufficiently large to enable of which 179 sites were variable (27.3%) and 175 assessment of sexual dimorphism. Instead, 20-30 were parsimony informative (26.6%). Average

Springer 322 Hydrobiologia (2007) 579:317-335

base frequencies were 22.6% A, 41.3% T, 15.6% from this basin were restricted to one (11-V, VII, C, and 20.5% G. Based frequencies were homo- VIII) or two (IX) populations, and differed from geneous across all sites (Chi-square = 35.66, the common haplotype and each other by one- df = 147, P = 1.00). Mean genetic distances be- three bp. All three Cuatro Cienegas specimens tween specimens of A. yecos and other North that were analyzed shared the same haplotype (I), American Assiminea species ranged from 10.93- which was differentiated from the Pecos River 11.71%. while the latter differed from each other basin haplotypes by 13 fixed base pair differences. by 3.31-5.47%. Distances between the four North The mean genetic distance between specimens American species and other assimineids that were from these two geographic areas was analyzed ranged from 16.63 to 17.72%. 2.30 k 0.55% (2.13-2.43%). which was much Nine COI haplotypes (I-IX, Table 2) were larger than the differences observed within the resolved in the 30 specimens of A. yecos that Pecos River basin (0.11 + 0.04%). An unpub- were analyzed. The most common haplotype (VI) lished 16s dataset that we gathered shows was shared by five Pecos River basin populations the same pattern, with substantial sequence spread among all three segments of this drainage divergence between snails of these two areas that are inhabited by this species. Each of the (1.4-1.8%). and little variation within populations other seven haplotypes observed in specimens (0-0.4%).

Table 2 COI haplotypes resolved in Assinzinea pecos Base pair positiotz

Code Haplotype 22 76 100 103 254 263 271 283 295 316 349 370 382 407 433 463 511 517 562 583 640 A3A I ATGCATTCGAATGCACGTTAC A3B I ...... A3C I ...... A5C I1 C.T C TA. GCTTGTAACGT A5D 111 C.T. C A GCTTGTAA. GT A8A IV AT.C TAGGCTTGTAA. GT A8C V C.T CCTA GCTTGTAA. GT A31B VI C.T C TAGCTTGTAA. GT A31C VI C.T. C. TAGCTTGTAA. GT A31D VI C.T C TAGCTTGTAA. GT A31E VI C.T C TA. GCTTGTAA. GT A31F VI C.T C TA. GCTTGTAA. GT AlOA VI C.T C. TAGCTTGTAA. GT AlOB VI C.T. C TA. GCTTGTAA. GT AlOE VI C.T C TAGCTTGTAA. GT A9A VI C.T. C TAGCTTGTAA. GT A9B VI C.T C TAGCTTGTAA. GT A16C VI C.T C. TAGCTTGTAA. GT A16E VI C.T C. TAGCTTGTAA. GT A16F VI C.T C. TAGCTTGTAA. GT A6C VII C.TGC. TA. GCTTGTAA. GT A6D VI C.T. C TAGCTTGTAA. GT A17B VI C.T. C TAGCTTGTAA. GT A17C VI C.T. C. TAGCTTGTAA GT A17D VI C.T C TAGCTTGTAA. GT A17E VIII . T. C. TAGCTTGTAA. GT A17F IX CC. T C. TAGCTTGTAA. GT A18B VI C.T. C TAGCTTGTAA. GT A18E VI C.T. C. TAGCTTGTAA. GT A18F IX CC. T C TAGCTTGTAA. GT Code stands for iildividual specimens. Periods indicate identical nucleotides as the first sequence

a Springer Hydrobiologia (2007) 579:317-335 323

Modeltest selected the General Time Revers- monophyletic subunit of A. pecos (Fig. 2), ible model with variable sites assumed to follow a whereas in the Bayesian and ML topologies they discrete gamma distribution (e.g., GTR + G; formed a paraphyletic group in which the Cuatro Tavare, 1986) with the following parameter val- Cienegas subclade was nested. The resulting trees ues best fitting our data: A = 0.2403, C = 0.1432, otherwise differed only in the positioning of G = 0.1871, T = 0.4294; Rmat = (0.1980 16.4996 terminal branches within well supported clades. 2.7368 0.0000 8.87251; shape of gamma distribu- tion = 0.1556. GTR distance was used to generate Morphology a NJ tree and a GTR + G model was used for the ML and Bayesian analyses. Descriptive statistics and ANOVA results (based All phylogenetic analyses congruently depicted on the raw dataset) are reported in Table 3. A. yecos as a moderate-strongly supported mono- Sample heterogeneity was highly significant for phyletic unit sister to a clade composed of the all parameters except aperture height (AH). Post- other three North American species, with the hoc pairwise testing for each parameter indicated Taiwanese outgroups positioned basally (Fig. 2, a no statistically significant differences among the MP tree). Specimens from Cuatro Cienegas were Pecos River basin samples, which are closely consistently resolved as a well supported subclade similar in all respects (Fig. 3). In contrast, the within A. yecos. In the MP and NJ trees, Pecos Cuatro Cienegas sample (Fig. 4) significantly River basin specimens were also resolved as a differed (P < 0.005) from all three Pecos River

Fig. 2 One of six shortest MP trees (TL = 277, CI = 0.82) depicting phylogeiletic relationships of Pecos assiminea

-populatioils - and divisioil of these into reciprocally monophyletic subunits (Pecos River basin, Cuatro Cienegas). Numbers are bootstrap values for nodal support

Paludinella taiwanensis-A roo Paludinella taiwanensis-D Paludinella taiwanensis-C Paludinella taiwanensis-B Paludinella taiwanensis-A Pseudornphala latericea-D Pseudomphala latericea-C 10

Springer 324 Hydrobiologia (2007) 579:317-335

Table 3 Variation in shell parameters among A. pecos populations Area Parameter Roswella (25) Ft. stocktonb (20) BalmorheaC (20) Cuatro cienegasd (30) *ANOVA

WH *XF = 9.38

SH **F = 24.917

SW "*F = 15.584

HBW **F = 9.677

WBW "*F = 16.683

AH F = 1.563

AW **F = 6.379

SWISH **F = 8.84

AHISH ""F = 30.64

HBWISH "*F = 30.181

Sample sizes are in parentheses. Values are mean & standard deviation, and range a USNM 1086379, Sago Springs USNM 1007143, "John's Hole" " USNM 1086193, East Sandia Spring USNM 1086441, spring-marsh complex north of Poza de la Becerra * Df for all parameters were 3, 91 :'* Highly significant (P 5 0.01) basin samples for all but two parameters (AH, closely similar results to the above in all respects, AW), and for AW significantly differed suggesting that distinctiveness of the Cuatro (P < 0.016) from two (Roswell, Ft. Stockton) Cienegas specimens cannot be solely attributed samples. to their smaller shells. This is also evidenced by The discriminant function analysis revealed the ratio data (Table 3), which indicated that the significant differences among samples (Wilk's Cuatro Cienegas shells are broader and have lambda = 0.258, F = 7.034, df = 21, P < 0.0001). larger apertures and body whorls (relative to shell Most of the variation (90.4%) was explained by height) than specimens of the Pecos River basin. the first canonical function, on which SH, SW, and ~xaminationof available material suggests that WBW had the highest loadings. A plot of the the Cuatro Cienegas shells (Fig. 4) can be further scores for the first two functions (Fig. 5) showed differentiated from those of the Pecos River basin extensive overlap of the Pecos River basin sam- (Fig. 3) by their flatter whorls, narrower whorl ples, and substantial differentiation of Cuatro shoulders, narrower umbilicus, and thinner lip. Cienegas shells. The classification matrix cor- rectly distinguished 90% of the Cuatro Cienegas specimens, with considerably lower values (50- Discussion 64%) for the Pecos River basin samples. Only four of 27 misclassified specimens from the Pecos Our main goal was to describe and analyze River basin were attributed to the Cuatro Ciene- variation within the Pecos assiminea and assess gas sample. The log transformed dataset yielded the evolutionary and taxonomic status of its

Springer Hydrobiologia (2007) 579:317-335 325

Fig. 3 (a-i) Pecos assiminea from the Pecos River basin. Co., TX; (f), USNM 1086193, East Sandia Spring, Reeves (a-c) USNM 1004623, Sago Springs, terminus of spring Co., TX; (g-i), USNM 1007143, "Johns Hole," Diamond Y run, Bitter Lake National Wildlife Refuge, Chaves Co., Draw, Pecos Co., TX. Scale bar = 500 pm NM; (d-e) USNM 1014738, East Sandia Spring, Reeves

disjunct groups of populations. We have shown ulations are distributed among habitat patches that A. yecos populations distributed among the separated by broad expanses of desert, it is likely type locality area and two other hydrographically that they are reproductively isolated from each separate portions of the Pecos River watershed other. If so, it would be desirable to preserve are closely similar morphologically, and have populations in each of these areas in order to limited genetic diversity and extensive sharing maximize the long term genetic potential and of haplotypes. These results confirm the conspec- viability of the Pecos assiminea. On the other ificity of these populations and suggest that they hand, it is possible, although less likely in our could be managed as a single conservation unit. view, that these demes are linked by contempo- However, given that these three groups of pop- rary gene flow (e.g., facilitated by passive trans-

Springer 326 Hydrobiologia (2007) 579:317-335

Fig. 4 (a-i) Pecos assiminea from the Cuatro Cienegas basin. (a-i) USNM 1086441, spring-marsh complex just west of Hwy 30, ca. 2.5 km tlorth of Poza la Becerra, Coahuila, Mexico. Scale bar = 500 pnl port on waterfowl or through periodic flooding), population and those of the Pecos River basin in which case treatment as a single, widespread (2.30 + 0.55%) is within the range observed population is appropriate. Additional studies among congeners of other North American gas- using more rapidly evolving genetic markers tropods belonging to the superfamily Rissooidea should be undertaken to evaluate these alterna- (e.g., Tryonia, 1.3-14.8%, Hershler et al., 1999; tives and help guide plans to conserve and Pyrgulopsis, 1.1-13.1%, Liu & Hershler, 2005; manage the endangered Pecos assiminea. Taylorconcha, 1.31%, Hershler et al., 2006b). Our findings also indicated that the Pecos Although a molecular clock is not available for assiminea population in the Cuatro Cienegas the Assimineidae, application of the COI calibra- basin is morphologically and genetically divergent tion of 1.83 k 0.21% per million years (my) relative to Pecos River basin snails and can be derived for other members of the superfamily diagnosed using either of these criteria. The COI Rissooidea (Wilke, 2003) suggests that the sequence divergence between specimens of this broadly disjunct Mexican and American popula-

@ Springer Hydrobiologia (2007) 579:317-335

this reasoning, we contend that our evidence of substantial divergence and long isolation of the broadly allopatric Mexican and American popu- lations assigned to A. yecos amply justifies treat- ing these as distinct congeners, even though they may not have yet achieved the evolutionary stage of reciprocal rnonophyly. Accordingly we de- scribe the former as a new species (A, cienegensis) below. The large molecular sequence divergence (10.93-11.71%) between assimineids of the Rio Grande region (A. cienegensis, A. pecos) and their North American congeners suggests a need to reassess a prior model for the biogeographic -3 1 I I I I I I I -4 -3 -2 -1 0 1 2 3 4 history of these snails. As discussed above, Taylor SCORE(1) (1985) speculated that these inland snails became Roswell A Balmorhea isolated from Gulf Coastal progenitors as a result Ft. Stockton Cuatro Cienegas of the inception of the modern freshwater lower Rio Grande, which is currently thought to have Fig. 5 Plot of the first two discriminant function scores occurred 1.0-0.75 Ma when the river breached a based on seven shell parameters from four populations of the Pecos assiminea, showing divergence of Cuatro former impoundment in the El Paso region Cienegas specimens. Confidence ellipses have P = 0.6827 (Mack, 1997; Cole et al., 2001). Our sequence divergence data suggest that the clade composed of Rio Grande assimineids instead evolved ca. tions of A. yecos have been separated for 6.39-5.97 Ma (late Miocene), well prior to the 1.13 + 0.27 my to 1.42 + 0.34 my, and thus have assembly of the modern, integrated Rio Grande. had independent evolutionary histories since at Additional evidence of the antiquity of this clade least the mid-Pleistocene. Note that the applica- is provided by our finding that it evolved prior to tion of molecular clock is laden with difficulties the split between Atlantic-Gulf Coastal (A. sz~c- and is constrained in this study by the application ciizea) and Pacific (A. californica, A. infirna) of a non-local clock based on a single calibration congeners (Fig. 2). which can be attributed to point (Wilke, 2003). Given these problems, the the Neogene rise of the Isthmus of Panama. This divergence times estimated above should be finding is also consistent with fossil evidence that considered tentative. assimineid snails have lived along the Gulf Coast Our analysis of mitochondria1 DNA sequences since the late Oligocene or early Miocene (e.g., indicates that divergence of these two groups has Assimiizea aldra Dall, Tampa Member, Arcadia progressed (at least) to the point where they do Formation; age fide Brewster-Wingard et al., not share haplotypes and Cuatro Cienegas spec- 1997). imens form a monophyletic unit nested within the We speculate that the Rio Grande assimineid Pecos River basin group ("simple paraphyly," clade may have originated as a result of stranding Omland et al.. 2006). Although reciprocal mono- of ancestral estuarine snails following one of the phyly is commonly considered a necessary crite- series of marine transgressions that occurred rion for defining species limits (Sites & Marshall, along the Gulf Coast during the Neogene (Hos- 2004), others have argued that recently evolved man, 1996). Alternatively, it is possible that species may have non-monophyletic patterns inland habitats were colonized in association with owing to incon~pletelineage sorting (Baker et the long history of fluvial sedimentation to the al., 2003: Omland et al., 2006) and other corre- northwest Gulf of Mexico that preceded forma- lates of their youthful evolutionary status (Funk tion of the modern Rio Grande (Galloway et al.. - & Omland, 2003: Kondo et al., 2004). Following 2000; Galloway, 2005). Although our data indicate

a Springer Hydrobiologia (2007) 579:317-335

that inception of the lower Rio Grande in the Material examined mid-Pleistocene well postdated the divergence of this inland clade in conflict with Taylor's (1985) Holotype: USNM 1071402, southern portion of biogeographic hypothesis, this major hydro- spring-marsh complex just west of Hwy 30, ca. graphic event nonetheless may have contributed 2.5 km north of Poza de la Becerra, Cuatro to the subsequent 1.25 + 0.30 Ma vicariance of Cienegas basin, Coahuila, Mexico, ca. 26'53' N, Mexican (A. cienegensis) and American (A. 102'8' W. Collected by R. Hershler, June 1,1981. pecos) elements of this fauna. Paratypes: USNM 1086641, from same collection as above. Other material: USNM 1085799, ibid., 26'53'39.4'' N: 102'8'24.3'' W. Collected by R. Taxonomy Hershler & J. J. Landye, August 2-3, 2005. Family Assimineidae H. & A. Adams Subfamily Assimineinae Genus Assiminen Fleming, 1828 Assiminen cienegensis is distinguished from other (type species, A. graynna Fleming) North American assimineids by its smaller, Assiminea cienegensis, n, sp. broader shell. The shells of A. cienegensis also Figs. 4, 6-10 differ from those of its closely similar sister species (A. pecos) in having a lower spire, less Assiminea sp.-Taylor, 1966: 208 (shells from coilvex whorls, narrower whorl shoulders, broad- northernmost pool of Pozos del la Becerra, 14 km er umbilicus, and thinner lip. Assimin,ea cienegen- southwest of Cuatro Cienegas). esis can be further distinguished from A, pecos by Assimiizea pecos.-Taylor, 1987: 8-9 [in part, the weaker pigmentatioil of the head-foot; pres- Mexico: Cuatro Cienegas basin, Coahuila]. ence of dark pigment along the right edge of the osphradium; much larger coiled section of the rectum in the pallial roof; and smaller, weakly differentiated glandular oviduct.

Description

Shell (Figs. 4, 6) very small (SH usu- ally < 2.0 mm), broadly conical to ovate-conic, thin, translucent. Protoconch (Fig. 7a, b) of 1.6 smooth, slightly convex whorls. Teleoconch of 2.5-3.5 weak or moderately convex, narrowly shouldered whorls; suture medium impressed; sculptured with growth lines, and occasional, weak spiral threads on later whorls. Shell color clear-white; periostracum amber. Aperture wide, pyriform; outer lip thin, slightly prosocline, weakly sinuate; peristome complete, very thin, parietal lip curved, broadly adnate; columellar lip thick, curved, slightly reflected, partly covering umbilical region. Umbilicus open. Operculum (Fig. 7c) pyriform, inner edge Fig. 6 Assiminen cienegensis, holotype, USNM 10171402, weakly sinuate; paucispiral, horny, thin, transpar- southern portion of spring-marsh complex just west of Hwy 30, ca. 2.5 km north of Poza de la Becerra, Cuatro ent, slightly concave. Muscle scar elongate, Cienegas basin, Coahuila, Mexico. Scale bar = 1.0 mm extending along most of inner surface (Fig. 7d). a Springer Hydrobiologia (2007) 579:317-335 329

Fig. 7 (a4) Shells and operc ula of A, ciene*gensis, usmrl 1086441. (a, b) Shell apex, showing smool:h protoconch. (4 Outer side of operc ulum. (d) I11.ner side of operc ulum. Scale bars =- 100 fin1

Head-foot usually pale except for black eye- pallial component. Gill consisting of two tiny, spots; occasional specimens having scattered finger-like filaments on posterior section of effer- black granules on snout, neck, and bases of eye ent vein near left edge of pallial cavity (Fig. Sb). stalks (Fig. 8a). Eye stalks short. Small, ciliated Osphradium small, narrow, right edge darkly swelling usually present just behind right eyestalk pigmented (Fig. Sb). Hypobranchial gland not (Figs. 8a, 9a-c). Snout moderately long, bilobed. seen in dissection. Foot large, anterior edge horizontal or slightly Mouth opens between muscular lips; buccal rounded, distal end tapering. Anterior mucus mass occupying most of snout. Radula sac short. gland of two small units that open to middle of Radula taenioglossate (Fig. lOa), consisting of transverse slit. Posterior mucus gland absent. about 40 well developed tooth rows. Ceiltral teeth Omniphoric grooves (Fig. 9c) well developed. (Fig. lob) longer than wide: median cusp long. Pallial cavity large. Pallial roof pale. Kidney pointed, flanked by 1-2 shorter cusps; basal cusps opening having fleshy lips; kidney with little or no 2-3, pointed, obliquely arranged; lateral sides

8 Springer 330 Hydrobiologia (2007) 579:317-335

Fig. 8 (a-d) Anatomy of A. cienegensis, USNM 1085799. duct, Es = eye stalk, Fp = fecal pellet, Ga = genital aper- (a) Head and penis. Scale bar = 1.0 mm. (b) Contents of ture, Gi = gill, Go = glandular oviduct, Me = mantle edge, pallial cavity. Scale bar = 250 pm. (c) Female glandular 0s= osphradium, Pd = penial duct, Pn = penis, oviduct and associated structures. Scale bar = 250 pm. (d) Pw = posterior wall of pallial cavity, Re = rectum, Bursa copulatrix and associated structures. Scale as in (c). Sd = sperm duct, Sn = snout, Sr = seminal receptacle, Bu = bursa copulatriu, Co = coiled oviduct, Dbu = bursa1 Sw = swelling behind right eye stalk, Ve = vestibule nearly vertical; base with large, V-shaped basal middle of testis. Posterior vas deferens opening to tongue. Lateral teeth (Fig. 10c, d) larger than ventral edge of prostate gland just behind pallial central teeth, taller than wide, wit11 long basal wall. Prostate gland large, bean-shaped, with process; median cusp long, pointed, flanked on about 50% of length in pallial roof. Anterior vas both sides by 2-3 small cusps; lateral flange very deferens originates from anterior end of prostate thin, narrow, attached to base of tooth, often gland, straight, narrow. Penis (Figs. 8a, 9d) large, difficult to discern in material prepared for narrow-elongate, coiled clockwise or nearly scanning electron microscopy (Fig. 10d). Inner straight, attached to near middle of head well marginal teeth (Fig. 10e) with curved sides; cut- behind eye stalks. Distal end of penis terminates ting edge wit11 5-8 large, pointed cusps; base in small papilla. Penial duct nearly straight, rounded. Outer marginal teeth (Fig. 10f) with 13- centrally positioned, narrow. 18 small, pointed cusps; edges of teeth nearly Ovary a simple, yellow sac. Female glandular parallel sided, outer edge vely thin. oviduct and associated structures shown in Testis small (ca. 0.5 whorl), pale, consisting of Fig 8C, D. Oviduct runs to posterior wall of 5-6 simple lobes. Seminal vesicle composed of pallial cavity and makes a single posterior, or thick, tightly clustered coils; originating from near posterior-oblique loop (Co). Seminal receptacle

Springer Hydrobiologia (2007) 579:317-335 331

Fig. 9 (a-d) Scanning electroil micrographs of head and omniphoric groove. Scale bar = 10 pm. (d) Penis. Scale penis of A. cienegensis, USNM 1085799. (a) Head, showing bar = 100 pm. Es = eye stalk, Om = omniphoric groove, small, ciliated swelling behind right eye stalk. Scale Pp = penial papilla, Sn = snout, Sw = swelling behind eye bar = 100 pm. (b) Close-up of swelling. Scale bar = 20 pm. stalk (c) Oblique view of head-foot, showing swelling and

(Sr) small, pouch-like, with pink sheen, opens to and eastern (Rio Salado drainage) portions of the oviduct slightly anterior to coiled portion, ante- Cuatro Cienegas basin based on the occasional rior to bursa copulatrix, overlapping posterior occurrence of empty shells in aquatic collections. edge of glandular oviduct, duct about as long as The only area where we found living specimens of pouch. Bursa copulatrix (Bu) large, ovate, largely A. cienegensis is the immediate vicinity of the type or entirely posterior to glandular oviduct, con- locality, which is in the western lobe of the basin taining dark brown material. Bursa1 duct (Dbu) (Fig. 11). Snails were found in small stands of long, narrow, strongly curved, originating from sedges emerging from slightly elevated mounds in right side near anterior edge, sometimes ex- areas where the groundwater was very close to the panded distally. Glandular oviduct short, without land surface. Collections were made by pulling obvious division into albumen and capsule glands, apart the hard, salt-encrusted exterior of these distal portion forming nonglandular vestibule mats to expose the moist, black-colored bases of (Ve). Sperm duct undulates within glandular vegetation on which the snails live. Specimens oviduct. Genital aperture (Ga) at anterior end were less commonly found on riparian vegetation of vestibule, forming a muscular papilla. alongside spring brooks.

Distribution and habitat Remarks

Taylor (1987) suggested that this snail may be Assimineids which have small, thin, brownish widespread in both the western (internal drainage) shells and live in association with aquatic habitats

@ springer 332 Hydrobiologia (2007) 579:317-335

Fig. 10 (a-f) Scanning electron nlicrographs of the radula USNM 1085799. Scale bar = 2 pm. (e) Iilller (to left) and of A. cienegensis. (a) Portion of radula ribbon, USNM outer (right) marginal teeth, USNM 1085799. Scale 1086641. Scale bar = 10 pm. (b) Central and lateral teeth, bar = 10 pm. (f) Outer marginal tooth, USNM 1085799. USNM 1085799. Scale bar = 2 pm. (c) Lateral tooth, Scale bar = 2 pm. F1 = flange of lateral tooth USNM 1086441. Scale bar = 10 pm. (d) Lateral teeth,

have traditionally been placed in the large, by Fukuda & Ponder (2003). Note that A. worldwide genus Assirninen (Abbott, 1958), and cienegensis differs from An.gustassiminen, to the new species described herein is provisionally which Abbott (1974) assigned A. californica and allocated to the genus on this basis. In a recent A, succinea (but see Fukuda & Mitoki, 1996), in review of the generic names of the Assimineidae, lacking spiral sculpture below the suture and in Fukuda & Ponder (2003) noted that species the umbilical region of its shell, and in having a attributed to Assiminen are morphologically shorter female sperm duct (see Fukuda & Mitoki, diverse, and suggested that the genus should be 1996; Fukuda & Ponder, 2003, for descriptions of restricted to its European type species and other Angustassiminea). Although A. cienegensis and snails that share its various distinctive features. the North American assimineids ultimately will Assiminea cienegensis (and the other North have to be transferred to another genus, a American assimineids) does not have any of confident assessment of their generic status must these characters nor does it well conform mor- await a more comprehensive study of phyloge- phologically to other genus-group taxa diagnosed netic relationships than has been provided herein.

a Springer Hydrobiologia (2007) 579:317-335 333

SIERRA LA MADERA

26" 55'

012345 IKilometers Adapted from Carta Topografica Cuatro Cienegas GI3059

102' 10' 1

Fig. 11 Map showing location of type locality of A. cienegemis. Inset shows location of the Cuatro Cienegas basin in Coahuila state (Mexico)

Etymology and equipment in the Rocky Mountain Center for Conservation Genetics and Systematics. The first autl~or'sfieldwork was supported, in part, by contract Named for the Cuatro Cienegas basin. funds provided by the United States Fish and Wildlife Service (Region 2) and the State of New Mexico. Acknowledgements The followiilg individuals collected Permission to collect on Nature Conservancy land in material used in this study and/or assisted with our Texas was authorized by John Karges. This paper fieldwork-N. Allan, L. L. Crisostomo, V. Gervasio, M. benefited from comments provided by Martin Haase and E. Gordon, T. Hyde, M. G. Kellogg, S. Knapp, J. J. an anonymous reviewer. Landye, A. L. Metcalf, L. Manning, F. G. Thon~pson,C.-1. Tsai, and S.-K. Wu. Scanning electron microscopy and shell measurements were performed by Y. Villacampa. K. Darrow and M. K. Ryan assisted wit11 preparation of the References map and specimen drawings, and D. Barrett kindly wrote a Per1 Script that eased data file conversion. Tom Quinn Abbott, R. T., 1958. The gastropod genus Assirninen in the (University of Denver) and Sara Oyler-McCance (United Philippines. Proceedings of the Academy of Natural States Geological Survey) generously shared bench space Sciences of Philadelphia 110: 213-278.

a Springer 334 Hydrobiologia (2007) 579:317-335

Abbott, R. T., 1974. American seashells. The marine nia-Nevada. Proceedings of the Biological Society of Mollusca of the Atlantic and Pacific Coasts of North Washington 102: 176-248. America. 2nd edn. Van Nostrand Reinhold Con~pany, Hershler, R., 1998. A systematic review of the hydrobiid New York. snails (Gastropoda: Rissooidea) of the Great Basin, Baker, J. M., E. L6pez-Medrano, A. G. Navarro-Sigiienza, 0. western United States. Part I. Genus Pyrgzilopsis. R. Rojas-Soto & K. E. Omland, 2003. Recent speciation Veliger 41: 1-132. in the Orchard Oriole group: divergence of Icterzcs Hershler, R., H.-P. Liu & M. Mulvey, 1999. Phylogenetic spurius and Icterlts spurius ftiertesi. Auk 120: 848-859. relationships within the aquatic snail genus Tryonia: Berry, S. S., 1947. A surprising molluscan discovery in implications for biogeography of the North American Death Valley. Leaflets in Malacology 1: 5-8. Southwest. Molecular Phylogenetics and Evolution Brewster-Wingard, G. L., T. M. Scott, L. E. Edwards, 13: 377-391. S. D. Weedman & K. R. Simmons, 1997. Reinterpre- Hershler, R., H.-P. Liu, T. J. Frest 6r E. J. Johannes, 2006a. tation of the peninsular Florida Oligocene: an inte- Extensive diversification of pebblesnails (Lithoglyph- grated stratigraphic approach. Sedimentary Geology idae: Flziminicoln) in the upper Sacramento River 108: 207-228. basin, northwestern United States. Zoological Journal Bucklin, A., 1992. Use of formalin-preserved samples for of the Linnean Society (in press). molecular analysis. Newsletter of Crustacean Molec- Hershler, R., H.-P. Liu, T. J. Frest, E. J. Johannes & ular Techniques 2: 3. W. H. Clark, 2006b. Genetic structure of the western Cole, J. C., B. D. Stone, R. R. Shroba 6r D. P. Dethier, North American aquatic gastropod genus Tnylorcon- 2001. Episodic Pliocene and Pleistocene drainage cho and description of a second species. Journal of integration along the Rio Grailde through New Molluscan Studies 72: 167-177. Mexico. Geological Society of America Abstracts Hosman, R. L., 1996. Regional stratigraphy and subsurface with Programs 33: 357. geology of Cenozoic deposits, Gulf Coastal Plain, Folmer, O., M. Black, W. Hoeh, R. Lutz 6r R. Vrijenhoek, South-Central United States. United States Geologi- 1994. DNA primers for amplication of mitochondrial cal Survey Professional Paper 1416: GlLG35. cytochrome c oxidase subunit I from diverse meta- Huelsenbeck, J. P. & F. Ronquist, 2001. MRBAYES: zoan invertebrates. Molecular Marine Biology and Bayesian inference of phylogeny. Bioinformatics Biotechnology 3: 294-299. 17: 754-755. Fukuda, H. & T. Mitoki, 1996. A revision of the family Kondo, B., J. M. Baker & K. E. Omland, 2004. Recent Assimineidae (Mollusca: Gastropoda: Neotaenioglos- speciation between the Baltimore Oriole and the sa) stored in the Yamaguchi Museum. Part 3, Black-backed Oriole. Condor 106: 674-680. Subfamily Assimineinae (2) Angzlstassinzinen and Kumar, S., K. Tamura 6r M. Nei, 2004. MEGA3: Psezidomphnln. The Yuriyagai 4: 109-137. Integrated software for Molecular Evolutionary Fukuda, H. & W. F. Ponder, 2003. Australian freshwater Genetic Analysis and sequence alignment. Briefings assimineids, with a synopsis of the Recent genus- in Bioinformatics 5: 2. group taxa of the Assimineidae (Mollusca: Caenogas- Liu, H.-P. & R. Hershler, 2005. Molecular systematics and tropoda: Rissooidea). Journal of Natural History radiation of western North American nymphophiline 37: 1977-2032. gastropods. Molecular Phylogenetics and Evolution Fukuda, H. & W. F. Ponder, 2005. A revision of the 34: 284-298. Australian taxa previously attributed to Assiminen Liu, H.-P., R. Hershler & K. Clift, 2003. Mitochondria1 bziccinoides (Quoy & Gaimard) and Assiminea DNA sequences reveal extensive cryptic diversity tnsmnnicn Tenison-Woods (Mollusca: Gastropoda: within a western American springsnail. Molecular Caenogastropoda: Assimineidae). Invertebrate Sys- Ecology 12: 2771-2782. tematics 19: 325-360. Mack, G. H., 1997. Neogene evolution of the Rio Grande, Funk, D. J. & K. E. Omland, 2003. Species-level paraphyly the axial river of the Rio Grande Rift, southwestern and polyphyly: frequency, causes, and consequences, USA. Geological Society of America Abstracts with with insights from animal mitochondrial DNA. Programs 29: 239. Ailnual Reviews of Ecology and Systematics NatureServe, 2006. Natureserve explorer: an online 34: 397423. encyclopedia of life [web application]. Version 4.7. Galloway, W. E., 2005. Gulf of Mexico depositional record NatureServe, Arlington, VA. Available from http:// of Cenozoic North America11 drainage basin evolu- www.natureserve.org/explorer. (Accessed: May 18, tion. International Association of Sedimentology 2006). Special Publication 35: 409423. New Mexico Department of Game and Fish (NMDGF), Galloway, W. E., P. E. Ganey-Curry, X. Li & R. T. Buffler, 2004. Recovery and conservation plan for four inver- 2000. Cenozoic depositional history of the Gulf of tebrate species: Noel's amphipod (Gnmmnrus despe- Mexico basin. American Association of Petroleun~ rntus), Pecos assimiilea (Assiminen pecos), Icoster's Geologists Bulletin 84: 1743-1774. springsnail (Juturnin kosteri), and Roswell springsnail Hershler, R., 1989. Spriilgsnails (Gastropoda: Hydrobii- (Pyrgz~lopsisroswellemis). New Mexico Department dae) of Owens and Amargosa River (exclusive of Ash of Game and Fish, Conservation Services Division, Meadows) drainages, Death Valley system, Califor- Santa Fe, NM.

Springer Hydrobiologia (2007) 579:317-335

Omland, K. E., J. M. Baker & J. L. Peters, 2006. Genetic Clarkia fossil beds of northern Idaho. American signatures of intermediate divergence: population Association for the Advancement of Science, San history of Old and New World Holarctic ravens Francisco, CA: 265-321. (Corvrls corax). Molecular Ecology 15: 795-808. Taylor, D. W., 1987. Fresh-water molluscs from New Ponder, W. F. & R. G. DeKeyzer, 1998. Superfamily Mexico and vicinity. New Mexico Bureau of Mines Rissooidea. In Beesley P. L., G. J. B. Ross, & A. Wells and Mineral Resources Bulletin 116: 1-50. (eds), Mollusca: The Southern Synthesis. Fauna of Turgeon, D. D., J. F. Quinn Jr., A. E. Bogan, E. V. Coan, Australia. Vol. 5. CSIRO Publishing, Melbourne: F. G. Hochberg, W. G. Lyons, P. M. Mikkelsen, 745-766. R. J. Neves, C. F. E. Roper, G. Rosenberg, B. Roth, Posada, D. & K. A. Crandall, 1998. Modeltest: testing the A. Scheltema, F. G. Thompson, M. Vecchione & model of DNA substitution. Bioinformatics 14: 817- J. D. Williams, 1998. Common and scientific names of 818. aquatic invertebrates from the United States and Sites, J. W. Jr & J. C. Marshall, 2004. Operational criteria Canada, 2nd edn. American Fisheries Scientific for delimiting species. Annual Review of Ecology, Publication 26: 1-526. Evolution, and Systematics 35: 199-227. United States Fish, Wildlife Service (USFWS), 2002. Swofford, D. L., 2002. PAUP*: Phylogenetic analysis using Endangered and threatened wildlife and plants; listing parsimony (and other methods). Version 4.0b10. Roswell springsnail, Kosters' tryonia, Pecos Assimi- Sinauer Associates, Sunderland, MA. nea, and Noel's amphipod as endangered with critical Systat Software, Inc. (SSI), 2004. systatB for Windowso. habitat. Federal Register 67: 6459-6479. Richmond (CA). United States Fish, Wildlife Service (USFWS), 2005. Tavare, S., 1986. Some probabilistic and statistical prob- Endangered and threatened wildlife and plants; listing lems in the analysis of DNA sequences. Lectures on Roswell springsnail, Kosters' springsnail, Noel's Mathematics in the Life Sciences 17: 57-86. amphipod, and Pecos Assinlinea as endangered with Taylor, D. W., 1966. A remarkable snail fauna from critical habitat; final rule. Federal Register 70: 46304- Coahuila. MBxico. Veliger 9: 152-228. 46333. Taylor, D. W., 1985. Evolution of freshwater drainages and Wilke, T., 2003. Salenthydrobia gen. nov. (Rissooidea: molluscs in western North America. In C. J. Smiley & Hydrobiidae): a potential relict of the Messinian A. J. Leviton (eds), Late Cenozoic history of the salinity crisis. Zoological Journal of the Linneail Pacific Northwest, interdisciplinary studies on the Society 137: 319-336.

a Springer Appendix C. Management recommendations for aquatic macroinvertebrate species of New Mexico with state and federal conservation status. Management Recommendations for Aquatic Macroinvertebrate Species of New Mexico with State and Federal Conservation Status

The management recommeildations described below pertain to aquatic macroinvertebrates of New Mexico (this report, Table 1) that possess state or federal status (endangered, threatened, species of concern), including other taxa recognized as Species of Greatest Conservation Need (SGCN) under the Department's draft Comprehensive Wildlife Conservation Strategy (NMDGF 2005a).

Invertebrates of Bitter Lake National Wildlife Refuge

Status: Recovery and conservation planning for four macroinvertebrates of BLNWR (Noel's amphipod, Gammarus desperatus; Pecos assiminea, Assiminea pecos; Koster's springsnail, Juturnia kosteri; and Roswell springsnail, Pyrgulopsis roswellensis)" was provided in detail by the NMDGF (2005b). This plan identifies three outcomes ("Objective Parameters") perceived to foster the objectives of conservation (maintaining extant populations in situ), habitat restoration for repatriation management, and protection by establishing refuge populations to prevent extinction of these species.

Statutory mandates of the New Mexico Wildlife Conservation Act ameildmeilts of 1995 stipulate that these activities use existing resources and funding, to the extent possible, to implement the plan. The NNIDGF has little jurisdiction over habitats currently occupied by these species, or historic range where repatriation might be considered. Accordingly, successful implementation of this plan will require collaboration with state, federal, and local government entities, and private landowners. Currently, the NMDGF is working with the U. S. Fish and Wildlife Service (Region 2, Ecological Services State Office of New Mexico and Texas), New Mexico Interstate Stream Commission, Texas Parks and Wildlife Department, and private landowners.

Management Recommendations: Six strategies were identified in the recovery plan (Section 3.5) to address issues related to conservation of these species. Numerous activities stipulated under the Action Plan (Section 3.6) are already in progress that ellcompass these strategies. These activities include:

1. Monitor habitats and populations of these four invertebrates on BLNWR.

2. Coiltinue research on species-specific autecology (population demography, habitat affinities, life history) and amphipod genetics, and phylogenetics of the Assiminea pecos species complex.

3. Explore wetland restoration possibilities of North Spring on the Roswell Country Club with the intent to repatriate these four species at this site. Specific recommendations here would be to consider first lowering of the lake level to induce flow of North Spring, alternative approach, which is considered more costly and physically intrusive, would be to rehabilitate North Spring by constructing a stream channel and fish barrier with man- made materials to create a flowing system, Either option, or other alternative designs, will require consensus among stakeholders, especially the Roswell Country Club. With particular regard to gammarid amphipods, any repatriation effort should strive to reintroduce Gammarus desperatus sensu strict0 (Cole 198 1) to North Spring, as there is further evidence of cryptic species of gammarids on BLWNR (Gervasio et al. 2004, Seidel and Berg 2008, Seidel et al. 2008).

.. Collaborate with personnel of BLWNR to expand the range of these four, macroinvertebrates by introducing them into a recently created stream course along the western limit of refuge units (impoundments) 3, 5-7, and 1 5. Habitat suitability assessment is recommended prior to any introduction. Similar comments per #3 above (last sentence) pertain to species-specific introductions, which may require intra-specific genetic studies among isolated populations of these invertebrates to elect the most appropriate founding populations fi-om all candidate populations on the refuge.

5. Continue exploratory surveys for these species on private land in Chaves County.

6. Efforts to establish captive populations of these species in refugia (Recovery and Conservation Plan Strategy 6) will require substantial logistical support that well-exceeds past and current levels of funding. Such activities require dedicated space and full-time staff to maintain aquaculture facilities and captive populations, while also carrying out research studies to assess the efficacy of controlled propagation (see Federal Register 2000, Lang 2001a, Lang et al. 2006, Shuster et al. 2005).

7. It is recommended that any reintroduction of these species occur only within known historic or current range, as identified in recovery and conservation plan (NMDGF 2005b).

Pyrgulopsis chupaderae

Status: Pyrgulopsis chupaderae is known fi-om two isolated stenothermal springs along the southwest flank of the Chupadera Mountains, Socorro County, New Mexico. Monthly monitoring of the P. chupaderae in the native habitat spanned the period May 1997 to July 1998. Routine population and habitat monitoring ceased in September 1999 since a change in land ownership precluded access. However, a recent transference of ownership (2004) may hold promise for future conservation efforts. The NMDGF is collaborating with the new landowner in this regard.

Primary threats to P, chupaderae include: diversion of spring flows, lowering of the ground water aquifer budget that supports Willow Spring, and over-grazing of the Willow Spring riparian corridor (Lang 2001b, 2002; NMDGF 2006). Management Recommendations:

1. Renegotiate access to continue monitoring populations of P. chupaderae in its native habitat.

2. Synthesize results from previous population and habitat use studies.

3. Recommend development of a habitat management plan in cooperation with the land owner under the "candidate conservation agreement" (Federal Register 1997) that perpetuates the existence of ths species in Willow Spring. This conservation plan must consider balancing water stewardship wisely to meet the needs of historic land use practices, which, heretofore, have allowed for persistence of native habitat critical to the survival of P. chupaderae.

4. Establish a refuge population of P. chupaderae at either the Albuquerque Biological Park or NMDGF headquarters.

Pyrgulopsis gilae

Status: The "metapopulation" of P. gilae consists of 10 widely disjunct populations in the Gila River Basin, New Mexico (Mehlhop 1992, 1993; New Mexico Natural Heritage Program database). Isolated populations occur on U. S. Forest Service (USFS) lands along the Middle Fork Gila River and Gila River inainstem. The East Fork Gila River sub-basin harbors eight isolated populations: two each on private and dual stewardship (private-USFS) lands and four on USFS managed lands.

Populations at the type locality (East Fork Gila River; Taylor 1987) were found stable during ths project. However, recreational bathing at the Alum Spring site may have adverse impacts to this species downstream of the existing bathing pool (Lang 2002), as only empty shells of P. gilae have been found in this reach where it was once more abundant (Taylor 1987).

Benefits potentially afforded this species by signage to eliminate bathing in Alum Spring must be weighed against potential detriment precipitated by malinterpretations of such a posting as limiting recreational use of this site.

Management Recommendations:

1. Initiate population surveys to document the status of all known populations of P. gilae in the Gila River Basin. Expand surveys to unexplored reaches within the basin.

2. Recommend allocation of Section 6 funding to assess genetic divergence of P. gilae within and among geographically isolated populations throughout the Gila River Basin. The genetic affinities of such geographically isolated populations are poorly understood, seldom studied, and must be adequately explored prior to development of proper management options (Weins 1996). Genetic divergence between disjunct populations may warrant taxonomic reevaluation of the species, which in turn could confer specific inanageinent recomineildatioils particular to genetically distinct populations relative to current ownership and land-use practices.

Pyrgulopsis pecosensis

Status: Taylor (1 987) reported two populations of the endemic P. pecosensis from perennial tributaries of the Black River, Eddy County, New Mexico: Blue Spring (type locality) and Castle Spring. Extirpation of the Castle Spring population is attributed to adverse land-use practices in the watershed (Landye 1981, NMDGF 1988, Mehlhop 1992). The habitat and population of P. pecosensis in Blue Spring was monitored monthly at two localities from July 1997 to September 1998. Blue Spring was monitored during all grant segments from 2001 to 2004. Pyrgulopsis pecosensis of Blue Spring appears stable under current grazing pressure and irrigation withdrawals.

Temporary acquisition of Blue Spring surface water rights (NMSA 1995) and the "...lack of oil and gas reserves in the area ..." prompted reclassificatioll of P. pecosensis from a federal Candidate for listing under the ESA to a Species of Concern (Federal Register 1996). Contrary to conclusions possibly drawn from this reclassification, the Black River Valley has experienced repeated problems of ground-water depletion and contamination. Water levels of domestic and agriculturallrai~gewells in the Black River Valley have lowered and even dried-up (residents of Black River Village and ei~virons,pers.corn.). Oil and gas operations are ongoing in the Black River drainage (BLM 1994), and have expanded markedly during the period 2001 to present, particularly in areas around Blue Spring. Threats from such activities are detailed in Lang (200 1b, 2002).

Management Recommendations:

1. Recorninend exploring options for a conservation agreement (Federal Register 1997) that provides a mechanism for species and habitat conservation compatible with past and present land-use practices. Such an agreement should also consider collaboration with the State Oil Conservation Division to insure conformance of oil and gas field operations to regulatioils and practices for installing wells in karst areas.

2. Financial support under Section 6 will facilitate the Project Biologist's efforts to process monthly voucher collections, compile a database, analyze data, and synthesize reports.

3. Coiltinue routine monitoring habitats and population of P. pecosensis in Blue Spring.

Pyrgulopsis thermalis Status: Mehlhop (1 993) documented the persistence of P. thermalis at historic site occurrences (see Taylor 1987) in the East Fork Gila River, and in the main stem of the Gila River (type locality), Grant Coui~ty.The former populations occur in unnamed springs on lands under joint stewardship (private-USFS), while the latter populations occurs in Alum Spring located along Gila River mainstem in the Gila Wilderness.

All ktlowil populations were found stable during this project. However, recreational bathing at the Alum Spring (type locality; Taylor 1987) may have adverse impacts to this species (Lang 2002). Providing that physical habitat disturbances associated with this type of use are limited to the stone pools frequented by bathers, this species appears tolerant of current levels of visitation. However, any efforts to increase discharge from the perched spring source could have adverse impacts to P. thermalis, as this species is restricted to the rheocrene of Alum Spring nearest the springhead discharge. Benefits potentially afforded this species by signage to alert bathers of this fact must be weighed against potential detriment precipitated by malinterpretations of such a postiilg as limiting recreational use of this site.

Management Recommendations:

(1) Initiate populatioil surveys to documeilt the status of all know11 populations of P. thermalis in the Gila River Basin. Expand survey area to malacologically unexplored reaches within the basin.

(2) Recommend allocation of Section 6 funding to assess genetic divergence of P. thermalis within and among geographically isolated populations throughout the Gila River Basin. The genetic affinities of such geographically isolated populations are poorly understood, seldom studied, and must be adequately explored before proper management options can be developed (Weins 1996). Genetic divergence between disjunct population may warrant taxonomic reevaluation of the species, which may in turn confer specific management recommendations particular to genetically distinct populations relative to current ownership and land-use practices.

Pisidium sanguinichristi

Status: Taylor (1987) described P. sanguinichristi as a narrowly restricted peaclam endemic to Middle Fork Lake, Questa Ranger District, Carson National Forest (CNF). In 1995, the NMDGF commenced annual population monitoring of P. sanguinichristi in response to a multi-agency conservation effort initiated by the U. S. Forest Service (1 996). Pisidium sanguinichristi has not beell collected in New Mexico since Taylor's species description. However, two collectioils in September 1999 yielded four specimens of Pisidium sp. from Middle Fork Lake that conchologically resemble the putative Pisidium sanguinichristi. This has stimulated reinspection of voucher material collected from Middle Fork Lake during previous grant segments (1995- 1999) to obtain sufficient sample sizes of this species for genetic and morphologic studies.

Management Recommendations: 1. Continue sphaeriid inventory in high-elevation wetland habitats throughout the Sangre de Cristo Mountains, and expand this effort into the Jeinez Mountains.

2. Conduct conchological morphometric study of P. sanguinichristi and P. milium, if such an investigation merits the effort. While shell meristics may help resolve outstanding taxonomic questioils (NMDGF 1996b), significant ecophenotypic variation in shell morphology and hinge dentition of sphaeriid clams inanifested by local environmental influences could render such an effort futile (Herrington 1962). Moreover, genetic study is contingent of securing an adequate sample size of the putative P. sangzainichristi.

Status: Southern populatioils of this widely distributed pond snail are disjunct in New Mexico, Texas, aid higher elevatioiis of tlie western states (Bequaert and Miller 1973; Taylor 1983, 1985). In New Mexico, Taylor (1 983) first reported populations of S. caperata from the VCNP and in Hunter Marsh, BLNWR. While the former popu.lation is extant, extirpation of the latter populatioil was attributed to extensive wetland habitat loss, alteration, and sewage containination between 1983 - 1985 (Taylor 1985, NMDGF 2006). However, the wrinkled marshsnail was recently (2003-2004) documented from live specimens in Hunter Marsh at BLNWR; in grassland pools on the Valle Grande, Valles Caldera National Preserve; and in high-elevation woodland vernal and roadside pools and glacial cirque lakes in the Costilla River drainage, Vermejo Park Ranch, Taos County.

Management Recommendations:

1. The Department should continue statewide surveys of high-elevation ephemeral pools.

2. There is considerable ecophenotypic variation in shell characters between high-elevation populations (VCNP, Vennejo Park) and. low-elevation populations (BLNWR; Leon Creek, Diamond Y Preserve, Pecos County, TX; B. Lang,pers. obs.). Studies are recommended to assess the validity of this observation. Literature Cited

Bequaert, J. C. and W. B. Miller. 1973. The inollusks of the arid southwest: with an Arizona check list. University of Arizona Press, Tucson, AZ. 27 1 pp.

Bureau of Land Management. 1994. Draft resource management planlenvironmental impact statement for the Roswell Resource Area, Roswell, New Mexico, and draft resource management plan amendinent/environmental impact statement for the Carlsbad Resource Area, Carlsbad, New Mexico. Bureau of Land Management, Roswell, NM.

Cole, G. A. 198 1. Gammarus desperatus, a new species from IVew Mexico (Crustacea: Amphipoda). Hydrobiologia 76:27-32.

Federal Register. 1996. Endangered and threatened wildlife and plants; Review of plant and animal taxa that are candidates for listing as endangered or threatened species. 50 CFR Part 17. 61(40):7596-7613.

Federal Register. 1997. Announcement of draft policy for candidate conservation agreements. 50 CFR Part 17. 62(113):32183-32188.

Federal Register. 2000. Policy of controlled propagation of species listed under the Endangered Species Act. 50 CFR Part 17. 65(183):56916-56922.

Gervasio, V., D. J. Berg, B. K. Lang, N. L. Allan, and S. I. Guttinan. 2004. Genetic diversity in the Gammarus pecos species complex: implications for conservation and regional biogeography in the Chihuahuan Desert. Lirnnology and Oceanography. 49(2):520-531.

Herrington, H. B. 1962. A revision of the Sphaeriidae of North America (Mollusca: Pelecypoda). University of Michigan Museum of Zoology. Miscellaneous Publications No. 18.

Landye, J. J. 1 981. Current status of endangered, threatened, and/or rare mollusks of New Mexico and Arizona. Endangered Species Office, U. S. Fish and Wildlife Service, Final Report 13.

Lang, B. K. 2001 a. New Mexico endangered invertebrates: monitoriilg and managemeilt. New Mexico Department of Game and Fish, Completion Report E-37(1-5) submitted to the Office of Federal Aid, U. S. Fish and Wildlife Service, Albuquerque, NM.

Lang, B. K. 2001b. Status of aquatic lnollusks of New Mexico. New Mexico Department of Game and Fish, Annual Performance Report (E-20-8) submitted to the Office of Federal Aid, U. S. Fish and Wildlife Service, Albuquerque, New Mexico.

Lang, B. K. 2002. Status of aquatic inollusks of New Mexico. Completion Report E-20(6-9) submitted to the Office of Federal Aid, U. S. Fish and Wildlife Service, Albuquerque, New Mexico.

Lang, B. K., D. A. Kelt, and S. M. Shuster. 2006. The role of controlled propagation on an endangered species: demographic effects of habitat heterogeneity among captive and native populations of the Socorro isopod (Crustacea: Flabellifera). Biodiversity and Conservation 15(12):3909-3935.

Mehlhop, P. 1992. Establishment of a rare mollusc inventory and monitoring program for New Mexico. Progress Report. NMGF Professional Services Contract 80-519 -52.

Mehlhop, P. 1993. Establislment of a rare mollusc inventory and monitoring program for New Mexico. Year I1 Progress Report. NMGF Professional Services Contract No. 80- 519 -52-Amendment 1.

New Mexico Department of Game and Fish. 1996a. Threatened and endangered species of New Mexico: Biennial Review and Recommendations. 122 pp.

New Mexico Department of Game and Fish. 1996b. Status of aquatic and terrestrial mollusks of New Mexico. New Mexico Department of Game and Fish, Coinpletion Report E-20(1-4) submitted to the Office of Federal Aid, U. S. Fish and Wildlife Service, Albuquerque, New Mexico.

New Mexico Department of Game and Fish. 1988. Handbook of species endangered in New Mexico. Account: A-299.

New Mexico Department of Game and Fish. 2006. Threatened and endangered species of New Mexico: biennial review and recommendations.

New Mexico Department of Game and Fish. 2005a. Comprehensive Wild.life Conservation Strategy. First draft.

New Mexico Department of Game and Fish. 2005b. Recovery and conservation plan for four invertebrate species: Noel's amphipod (Gammaraus desperatus), Pecos assiminea (Assimineapecos), Koster's springsnail (Juturnia kosteri), and Roswell springsnail (Pyrgulopsis roswellensis). New Mexico Department of Game and Fish, Conservation Services Division, Santa Fe, New Mexico.

New Mexico Statutes Annotated. 1995. Interstate Stream Commission Water Conservation Program: Pecos River Portion. NMSA Supplement 72-5-28.

Seidel, R. A. and D. J. Berg. 2008. Genetic assessment of the Gamrnarus pecos species complex (Crustacea: Amphipoda) of New Mexico. Final Report to New Mexico Department of Game and Fish under Share with Wildlife Contract No. 04-516.0000 -0093.

Seidel, R. A., B. K. Lang, and D. J. Berg. 2008. Phylogeographic analysis reveals multiple cryptic species of amphipods (Crustacea: Amphipoda) in Chihuahuan Desert springs. Molecular Ecology (In Review).

Shuster, S. M., M. P. Miller, B. K. Lang, N. Zorich, L. Huynh, and P. Keim. 2005. The effects of controlled propagation on an endangered species: genetic differentiation and divergence in body size among native and captive populations of the Socorro isopod (Crustacea: Flabellifera). Conservation Genetics 6:355-368.

Taylor, D. W. 1983. Endangered Species: Status investigation of mollusks of New Mexico. Professional Service Contract Nos. 519 -69-01 and 519 -69-01 - A.

Taylor, D. W. 1985. Aquatic molluscs from Pecos County, Texas. Unpublished manuscript submitted to the New Mexico Department of Game and Fish.

Taylor, D. W. 1987. Fresh-water mollusks from New Mexico and vicinity. New Mexico Bureau of Mines & Mineral Resources Bulletin 116.

United States Forest Service. 1996. Habitat conservation assessment for the Sangre de Cristo peaclam (Pisidium sanguinichristi).

Weins, J. A. 1996. Wildlife in patchy environments: metapopulations, mosaics, and management. Pp 53-84, In Metapopulations and Wildlife Conservation (Ed. D. R. McCollough). Island Press.