Endangered and Threatened Wildlife and Plants; Proposed Endangered Status for the Neosho Mucket, Threatened Status for the Rabbitsfoot, and Designation of Critical Habitat for Both Species

Federal RegisterOct 16, 2012

Ask Donna

What actually matters in this document.

Text

DEPARTMENT OF THE INTERIOR

Fish and Wildlife Service

50 CFR Part 17

[Docket No. FWS-ES-R4-2012-0031; 4500030113]

RIN 1018-AX73

Endangered and Threatened Wildlife and Plants; Proposed Endangered Status for the Neosho Mucket, Threatened Status for the Rabbitsfoot, and Designation of Critical Habitat for Both Species

AGENCY:

Fish and Wildlife Service, Interior.

ACTION:

Proposed rule.

SUMMARY:

We, the U.S. Fish and Wildlife Service, propose to list the Neosho mucket (

Lampsilis rafinesqueana

), a freshwater mussel, as endangered and rabbitsfoot (

Quadrula cylindrica cylindrica

), a freshwater mussel, as threatened under the Endangered Species Act; and propose to designate critical habitat for both species. This rule fulfills our obligation under a settlement agreement. The effect of this regulation is to conserve the Neosho mucket and rabbitsfoot and their habitats under the Endangered Species Act.

DATES:

We will accept comments received or postmarked on or before December 17, 2012. Comments submitted electronically using the Federal eRulemaking Portal (see

ADDRESSES

section, below) must be received by 11:59 p.m. Eastern Time on the closing date. We must receive requests for public hearings, in writing, at the address shown in the

ADDRESSES

section by November 30, 2012.

ADDRESSES:

You may submit comments by one of the following methods:

(1)

Electronically:

Go to the Federal eRulemaking Portal:

http://www.regulations.gov.

In the Keyword box, enter Docket No. FWS-R4-ES-2012-0031, which is the docket number for this rulemaking. Then, in the Search panel on the left side of the screen, under the Document Type heading, click on the Proposed Rules link to locate this document. You may submit a comment by clicking on “Send a Comment or Submission.”

(2)

By hard copy:

Submit by U.S. mail or hand-delivery to: Public Comments Processing, Attn: FWS-R4-ES-2012-0031; Division of Policy and Directives Management; U.S. Fish and Wildlife Service; 4401 N. Fairfax Drive, MS 2042-PDM; Arlington, VA 22203.

We request that you send comments only by the methods described above. We will post all comments on

http://www.regulations.gov.

This generally means that we will post any personal information you provide us (see the Public Comments section below for more information).

FOR FURTHER INFORMATION CONTACT:

James F. Boggs, Field Supervisor, U.S. Fish and Wildlife Service, Arkansas Ecological Services Office, 110 South Amity Road, Suite 300, Conway, AR 72032, by telephone 501-513-4470 or by facsimile 501-513-4480. Persons who use a telecommunications device for the deaf (TDD) may call the Federal Information Relay Service (FIRS) at 800-877-8339.

SUPPLEMENTARY INFORMATION:

This document consists of: (1) A proposed rule to list the Neosho mucket (

Lampsilis rafinesqueana

) as endangered and rabbitsfoot (

Quadrula cylindrica cylindrica)

as threatened; and (2) a proposed critical habitat designation for both species.

Executive Summary

Why we need to publish a rule.

Under the Endangered Species Act (Act), a species may warrant protection through listing if it is endangered or threatened throughout all or a significant portion of its range. The Neosho mucket and rabbitsfoot are highly restricted in their ranges and the threats occur throughout their ranges; therefore, the species qualify for listing. We are proposing to list the Neosho mucket as an endangered species and rabbitsfoot as a threatened species. Their protection under the Act can only be done by issuing a rule.

• We estimate the Neosho mucket has been extirpated (no longer in existence) from approximately 62 percent of its historical range with only 9 of the 16 historical populations remaining (extant). This mussel is declining rangewide (eight of the nine extant populations) with only one remaining large viable population.

• We estimate the rabbitsfoot has been extirpated from approximately 64 percent of its historical range. While 51 of the 140 historical populations are extant (remain), only 11 populations (22 percent of extant populations or 8 percent of the historical populations) are viable; 23 populations (45 percent of extant populations) are at risk of extirpation; and 17 populations (33 percent of extant populations) show limited recruitment with little evidence of sustainability. Rabbitsfoot is extirpated from 2 States within its historical range.

• The majority (8 of the 11 or 73 percent) of the viable rabbitsfoot populations live in waters considered impaired under section 303(d) of the Clean Water Act or have numerous tributaries in their watersheds also listed as impaired. Thus, these mussels are subjected to water quality and quantity and sediment quality constraints. These constraints (impairment) are expected to be exacerbated by increased water demand, habitat degradation, and climate change. Therefore, the viability of the majority of rabbitsfoot populations is uncertain.

• The majority of extant rabbitsfoot populations are marginal to small (40 of 51 extant populations (78 percent)) and isolated (41 of 51 extant populations (80 percent)); because of the isolation, it is unlikely that recruitment between populations or establishment of new populations could occur naturally.

• We are proposing to list the Neosho mucket as an endangered species in Arkansas, Kansas, Missouri, and Oklahoma and the rabbitsfoot as a threatened species in Alabama, Arkansas, Georgia, Kansas, Kentucky, Illinois, Indiana, Louisiana, Mississippi, Missouri, Ohio, Oklahoma, Pennsylvania, Tennessee, and West Virginia.

The basis for our action.

Under the Endangered Species Act, a species may be determined to be endangered or threatened based on any of five factors: (1) Destruction, modification, or curtailment of its habitat or range; (2) overuse; (3) disease or predation; (4) inadequate existing regulations; or (5) other natural or manmade factors.

We have determined that both species are threatened by destruction, modification, or curtailment of habitat or range, inadequate existing regulatory mechanisms, and other manmade factors:

This rule designates critical habitat for each species.

• We are proposing to designate critical habitat for the Neosho mucket in Arkansas, Kansas, Missouri, and Oklahoma and for the rabbitsfoot in Alabama, Arkansas, Illinois, Indiana, Kansas, Kentucky, Mississippi, Missouri, Ohio, Oklahoma, Pennsylvania, and Tennessee.

• In total, approximately 779 river kilometers (rkm) (484 river miles (rmi)) in the Cottonwood, Elk, Fall, Illinois, Neosho, Shoal, Spring, North Fork Spring, and Verdigris Rivers are being proposed for designation as critical habitat for the Neosho mucket in Arkansas, Kansas, Missouri, and Oklahoma.

• The proposed critical habitat for the Neosho mucket is located in:

○ Benton and Washington Counties, Arkansas;

○ Allen, Chase, Cherokee, Coffey, Elk, Greenwood, Labette, Montgomery, Neosho, Wilson, and Woodson Counties, Kansas;

○ Jasper, Lawrence, McDonald, and Newton Counties, Missouri; and

○ Adair, Cherokee, and Delaware Counties, Oklahoma.

• In total, approximately 2,662 rkm (1,654 rmi) in the Neosho, Spring (Arkansas River system), Verdigris, Black, Buffalo, Little, Ouachita, Saline, Middle Fork Little Red, Spring (White River system), South Fork Spring, Strawberry, White, St. Francis, Big Sunflower, Big Black, Paint Rock, Duck, Tennessee, Red, Ohio, Allegheny, Green, Tippecanoe, Walhonding, Middle Branch North Fork Vermilion, and North Fork Vermilion Rivers and Bear, French, Muddy, Little Darby and Fish Creeks in Alabama, Arkansas, Kansas, Kentucky, Illinois, Indiana, Mississippi, Missouri, Ohio, Oklahoma, Pennsylvania, and Tennessee are being proposed for designation as critical habitat for the rabbitsfoot.

• The proposed critical habitat for the rabbitsfoot is located in:

○ Colbert, Jackson, Madison, and Marshall Counties, Alabama;

○ Arkansas, Ashley, Bradley, Clark, Cleveland, Dallas, Drew, Fulton, Grant, Hot Spring, Independence, Izard, Jackson, Lawrence, Little River, Marion, Monroe, Montgomery, Newton, Ouachita, Randolph, Saline, Searcy, Sevier, Sharp, Van Buren, White, and Woodruff Counties, Arkansas;

○ Allen and Cherokee Counties, Kansas;

○ Ballard, Green, Hart, Livingston, Logan, Marshall, and McCracken Counties, Kentucky;

○ Massac, Pulaski, and Vermilion Counties, Illinois; Carroll, Pulaski, Tippecanoe, and White Counties, Indiana; Hinds, Sunflower, Tishomingo, and Warren Counties, Mississippi;

○ Jasper, Madison, and Wayne Counties, Missouri;

○ Coshocton, Madison, Union, and Williams Counties, Ohio;

○ McCurtain and Rogers Counties, Oklahoma; Crawford, Erie, Mercer, and Venango Counties, Pennsylvania; and

○ Hardin, Hickman, Marshall, Maury, and Robertson Counties, Tennessee.

Peer review of our methods.

During the public comment period, we will obtain review and opinions from knowledgeable individuals with scientific expertise on our technical assumptions, analysis, adherence to regulations, and whether or not we used the best available information in developing the proposed rule.

Information Requested

We intend that any final action resulting from this proposal will be based on the best scientific and commercial data available and be as accurate and as effective as possible. Therefore, we request comments or information from the public, other concerned governmental agencies, Native American tribes, the scientific community, industry, or any other interested party concerning this proposed rule. We particularly seek comments concerning:

(1) Biological, commercial trade, or other relevant data concerning any threats (or lack thereof) to these species and regulations that may be addressing those threats.

(2) Additional information concerning the historical and current status, range, distribution, and population size of these species, including the locations of any additional populations of these species.

(3) Any information on the biological or ecological requirements of the species and ongoing conservation measures for the species and their habitat.

(4) Any information regarding water quality data that may be helpful in determining the water quality parameters necessary for Neosho mucket and rabbitsfoot.

(5) The reasons why we should or should not designate habitat as “critical habitat” under section 4 of the Act (16 U.S.C. 1531

et seq.

), including whether there are threats to the species from human activity, the degree of which can be expected to increase due to the designation, and whether that increase in threat outweighs the benefit of designation such that the designation of critical habitat is not prudent.

(6) Specific information on:

(a) The amount and distribution of Neosho mucket and rabbitsfoot habitat;

(b) What areas, that were occupied at the time of listing (or are currently occupied) and that contain features essential to the conservation of the species, should be included in the designation and why;

(7) Land use designations and current or planned activities in the areas occupied by the species or proposed to be designated as critical habitat, and possible impacts of these activities on these species and proposed critical habitat.

(8) Information on the projected and reasonably likely impacts of climate change on the Neosho mucket and rabbitsfoot and proposed critical habitat.

(9) Any foreseeable economic, national security, or other relevant impacts that may result from designating any area that may be included in the final designation. We are particularly interested in any impacts on small entities, and the benefits of including or excluding areas from the proposed designation that are subject to these impacts.

(10) Whether our approach to designating critical habitat could be improved or modified in any way to provide for greater public participation and understanding, or to assist us in accommodating public concerns and comments.

(11) The likelihood of adverse social reactions to the designation of critical habitat and how the consequences of such reactions, if likely to occur, would relate to the conservation and regulatory benefits of the proposed critical habitat designation.

Please note that submissions merely stating support for or opposition to the action under consideration without providing supporting information, although noted, will not be considered in making a determination, as section 4(b)(1)(A) of the Act directs that determinations as to whether any species is a threatened or endangered species must be made “solely on the basis of the best scientific and commercial data available.”

You may submit your comments and materials concerning this proposed rule by one of the methods listed in the

ADDRESSES

section. We request that you send comments only by the methods described in the

ADDRESSES

section.

If you submit information via

http://www.regulations.gov,

your entire submission—including any personal identifying information—will be posted on the Web site. If your submission is made via a hardcopy that includes personal identifying information, you may request at the top of your document that we withhold this information from public review. However, we cannot guarantee that we will be able to do so. We will post all hardcopy submissions on

http://www.regulations.gov.

Please include sufficient information with your comments to allow us to verify any scientific or commercial information you include.

Comments and materials we receive, as well as supporting documentation we used in preparing this proposed rule, will be available for public inspection on

http://www.regulations.gov,

or by appointment, during normal business hours, at the U.S. Fish and Wildlife Service, Arkansas Ecological Services Office, Conway, Arkansas (see

FOR FURTHER INFORMATION CONTACT

Previous Federal Actions

Neosho Mucket

The Neosho mucket was first identified as a candidate for protection under the Act in the May 22, 1984,

Federal Register

(49 FR 21664) notice. As a candidate, it was assigned a status Category 2 designation, which was given to those species with some evidence of vulnerability but for which additional biological information was needed to support a proposed rule to list as endangered or threatened. In our Notices of Review dated January 6, 1989 (54 FR 554), November 21, 1991 (56 FR 58804), and November 15, 1994 (59 FR 58982), we retained a status Category 2 designation for this species. We discontinued assigning categories to candidate species in our Notice of Review dated February 28, 1996 (61 FR 7596), and only species for which the Service had sufficient information on biological vulnerability and threats to support issuance of a proposed rule were regarded as candidate species. Thus, Neosho mucket was no longer considered a candidate species.

On October 30, 2001, we identified the Neosho mucket in the

Federal Register

(66 FR 54808) as a candidate species based on available information to support a proposed rule. Candidate species are assigned listing priority numbers (LPNs) based on immediacy and magnitude of threats, as well as taxonomic status. The lower the LPN, the higher priority that species is for us to determine appropriate action using our available resources. We assigned an LPN of 5 to Neosho mucket. In our Notices of Review dated June 13, 2002 (67 FR 40657), and May 4, 2004 (69 FR 24876), we maintained an LPN of 5.

We published a petition finding for the Neosho mucket on May 11, 2005 (70 FR 24870), in response to a petition received on May 11, 2004, stating in the finding that the Neosho mucket would retain an LPN of 5. In our Notices of Review dated September 12, 2006 (71 FR 53756), December 6, 2007 (72 FR 69034), and December 8, 2008 (73 FR 75176), we maintained an LPN of 5, reflecting the nonimminent threats of high magnitude. The LPN was elevated to 2 in our Notice of Review dated November 10, 2010 (75 FR 69222), to reflect the change from nonimminent to imminent threats of high magnitude.

Rabbitsfoot

The rabbitsfoot was first identified as a candidate for protection under the Act in the November 15, 1994,

Federal Register

(59 FR 58982). As a candidate, it was assigned a status Category 2 designation. The category 2 list was eliminated in 1996 (61 FR 7596). On November 9, 2009, we added the rabbitsfoot to our candidate list in the

Federal Register

(74 FR 57804) with an LPN of 9. An LPN of 9 indicates threats of a moderate magnitude; some of the threats are nonimminent, most are ongoing, and the threats are imminent overall. In our Notice of Review dated November 10, 2010 (75 FR 69222), it was again identified as a candidate species with an LPN of 9.

Status Assessment for Neosho Mucket and Rabbitsfoot

Background

It is our intent to discuss below only those topics directly relevant to the listing of the Neosho mucket as endangered and the rabbitsfoot as threatened in this section of the proposed rule.

Introduction

North American freshwater mussel fauna is the richest in the world and historically numbered around 300 species (Williams

et al.

1993, p. 6). Freshwater mussels are in decline, however, and in the past century have become more imperiled than any other group of organisms (Williams

et al.

2008, p. 55). Approximately 66 percent of North America's freshwater mussel species are considered vulnerable to extinction or possibly extinct (Williams

et al.

1993, p. 6). Within North America, the southeastern United States is the hot spot for mussel diversity. Seventy-five percent of southeastern mussel species are in varying degrees of rarity or possibly extinct (Neves

et al.

1997, pp. 47-51). The central reason for the decline of freshwater mussels is the modification and destruction of their habitat, especially from sedimentation, dams, and degraded water quality (Neves

et al.

1997, p. 60). These two mussels, like many other southeastern mussel species, have undergone reductions in total range and population density.

General Biology

Freshwater mussels generally live embedded in the bottom of rivers, streams, and other bodies of water. They siphon water into their shells and across four gills that are specialized for respiration and food collection. Food items include algae, bacteria, detritus (disintegrated organic debris), and microscopic animals (Strayer

et al.

2004, pp. 430-431). It also has been surmised that dissolved organic matter may be a significant source of nutrition (Strayer

et al.

2004, p. 430). Adults are filter feeders and generally orient themselves on or near the substrate surface to take in food and oxygen from the water column. Juveniles typically burrow completely beneath the substrate surface and are pedal (foot) feeders (bringing food particles inside the shell for ingestion that adhere to the foot while it is extended outside the shell) until the structures for filter feeding are more fully developed (Yeager

et al.

1994, pp. 200-221; Gatenby

et al.

1996, p. 604).

Sexes in unionid (refers to taxonomic family Unionidae) mussels, such as the Neosho mucket and rabbitsfoot, are usually separate. Males release sperm into the water column, which are drawn in by females through their siphons during feeding and respiration. Fertilization takes place inside the shell, and success is apparently influenced by mussel density and water flow conditions (Downing

et al.

1993, pp. 153-154). The eggs are retained in the gills of the female until they develop into mature larvae called glochidia. The glochidia of most freshwater mussel species, including the two species addressed in this rule, have a parasitic stage during which they must attach to the gills, fins, or skin of a fish to transform into a juvenile mussel. Depending on the mussel species, females release glochidia either separately, in masses known as conglutinates (gelatinous or jelly-like), or in one large mass known as a super-conglutinate. The duration of the parasitic stage varies by mussel species, water temperature, and perhaps host fish species. When the transformation is complete, the juvenile mussels drop from their fish host and sink to the stream bottom where, given suitable conditions, they grow and mature into adults. Host specificity is discussed in more detail below.

Growth rates for mussels are highly variable among individual mussel species, but overall, mussels tend to grow relatively rapidly for the first few years (Scruggs 1960, pp. 28-30; Negus 1966, pp. 517-518) then slow appreciably (Bruenderman and Neves 1993, p. 88; Hove and Neves 1994, pp. 34-36). This reduction in growth rate is correlated to sexual maturity, probably as a result of energy being diverted from growth to gamete production (Baird 2000, pp. 63-71). Heavy-shelled species, such as Neosho mucket and rabbitsfoot, grow slowly relative to thin-shelled species (Coon

et al.

1977, pp. 19-21; Hove and Neves 1994, p. 38).

Strayer (1999a, pp. 468 and 472) demonstrated that mussels in streams occur chiefly in “flow refuges” (relatively stable areas that displayed little movement of substrate particles during flood events). Other researchers

also concluded that mussel location and density are greatest in areas where shear stress (stream's ability to entrain and transport bed material created by the flow acting on the bed material) is low and sediments remain generally stable during flooding (Layzer and Madison 1995, p. 341; Strayer 1999a, pp. 468 and 472; Hastie

et al.

2001, pp. 111-114). These “flow refuges” conceivably allow relatively immobile mussels, such as the Neosho mucket and rabbitsfoot, to remain in the same general location throughout their life span. However, these areas may be more important for the rabbitsfoot since it typically does not burrow like the Neosho mucket, making it more susceptible to displacement into unsuitable habitat. However, flow refuges are not created equally and other habitat variables are important, but poorly understood (Roberts 2008, pers. comm.).

Taxonomy, Life History, and Distribution

The Neosho mucket and rabbitsfoot are freshwater mussels in the family Unionidae. Both species are currently deemed valid by the Committee on Scientific and Vernacular Names of Mollusks of the Council of Systematic Malacologists and the American Malacological Union (Turgeon

et al.

1998, pp. 35 and 37).

Neosho Mucket

Neosho mucket was originally described as

Lampsilis rafinesqueana

from Indian Creek, McDonald County, Missouri (Frierson 1927, pp. 69-70). There is no synonomy (scientific names previously describing the same species) of the Neosho mucket. Frierson (1927, pp. 69-70) described the Neosho mucket as a dimorphic (male and female shape differs) species; the male is elliptical, rounded before biangulate behind, with dorsal and basal margin equally arched, while the female is ovate with a widely expanded fan-shaped posterior. The shell is up to 9.5 centimeters (cm) (4 inches (in)), compressed, and relatively thin (Oesch 1984, pp. 219-221). The epidermis is olive-yellow to brown, becoming darker brown with age; green rays cover the surface, but are often discontinuous. Oesch (1984, pp. 219-221) describes the left valve as having two stout, divergent, striated, triangular pseudocardinal teeth. The two lateral teeth are short, stout, and slightly curved. The right valve has a single, tall, triangular to columnar, striated pseudocardinal tooth. The nacre (crystalline carbonate shell material of freshwater mussels) is bluish white to white.

Neosho mucket glochidia are an obligate parasite on smallmouth bass (

Micropterus dolomieu

), largemouth bass (

Micropterus salmoides

), and spotted bass (

Micropterus punctulatus

) (Barnhart and Roberts 1997, p. 18; U.S. Fish and Wildlife Service 2005, p. 7). Neosho mucket is unusual among other

Lampsilis

species in the timing of reproduction. Neosho mucket spawns in late April and May, and female brooding occurs May through August. Most other

Lampsilis

spawn in the late summer or fall and brood glochidia throughout the winter months into the following spring or summer. Barnhart (2003, p. 9) reported an average fecundity to be approximately 1.3 million glochidia per female in the Spring River, Kansas. The female Neosho mucket inflates and extends a pair of mantle flaps (actually an extension of the inner lobe of the mantle edge) that, from a side angle, remarkably resembles a small fish. Each mantle flap in addition to its fish-like shape has pigmentation that resembles an eyespot as well as a fish's lateral line. Muscular contractions of the mantle flaps create an undulating or “swimming” motion that suffices to lure fish hosts (Obermeyer 2000, p. 9).

The Neosho mucket is associated with shallow riffles and runs comprising gravel substrate and moderate to swift currents. The species is most often found in areas with swift current, but in Shoal Creek and the Illinois River it prefers near-shore areas or areas out of the main current (Oesch 1984, p. 221; Obermeyer 2000, pp. 15-16). Neosho mucket historically occurred in at least 16 streams within the Illinois, Neosho, and Verdigris River basins covering four states (Arkansas, Kansas, Oklahoma, and Missouri). It is endemic to the Arkansas River system (Gordon 1980, pp. 318 and 347; Harris and Gordon 1987, pp. 53-54; Obermeyer 1996, pp. 3-4; Vaughn 1996, pp. 3-5; Mather 1990, pp. 7-13; Obermeyer

et al.

1997a, pp. 44-47; Harris

et al.

2009, p. 68). The Neosho mucket's known river and creek occurrences and current status are shown in Table 1.

For the purposes of this rule, a population is considered extant if live individuals or fresh dead specimens have been located since 1985. A population is considered viable if it is sizeable, comprised of different age classes, recruiting juveniles, and able to sustain itself over several decades without human intervention (Butler 2005, p. 23). Population trend estimates were generally made with a 20- to 30-year perspective when adequate historical information was available. Populations were deemed to have improving, stable, declining, or unknown status (Table 1).

Table 1—Neosho Mucket River and Creek Occurrences and Current Population Status

River basin

River/Creek

State(s)

Current status

Date of last observation

Neosho River

KS, OK

Declining

2000.

Cottonwood River

Unknown

2011.

South Fork Cottonwood River

Extirpated

Pre-1979.

Spring River

KS, MO, OK

Stable

2010.

North Fork Spring River

Declining

1995.

Center Creek

KS, MO

Extirpated

1995.

Shoal Creek

KS, MO

Declining

2001.

Elk River

MO, OK

Declining

1995.

Indian Creek

Extirpated

Pre-1980.

Little Sugar Creek

Extirpated

Pre-1980.

Illinois River

AR, OK

Declining

2008

Verdigris River

KS, OK

Declining

2010

Otter Creek

Extirpated

Pre-1993.

Fall River

Declining

2004.

Elk River

Extirpated

Pre-1979.

Caney River

KS, OK

Extirpated

Pre-1979.

Neosho River Basin

Neosho River:

The Neosho River drains southeast through Kansas and Oklahoma. Historical data of Neosho mucket densities for the Neosho River are not available prior to the late 1970s (Obermeyer

et al.

1997b, p. 112). Mussel harvest records from the early 1900s provide useful insight on the abundance of mussels in the river. From 1911 through 1912, the Neosho River provided 17 percent or approximately 85 million mussels used in the nation's pearl button industry. Many of the 30 tons of mussel shells processed weekly in 1918 at a shell blank factory in Iola, Kansas, came from the Neosho River near LeRoy, Kansas (Obermeyer

et al.

1997b, p. 112).

Since the 1990s, extant populations have been found downstream of John Redmond Reservoir Dam to near Parsons, Kansas, in Allen, Coffey, Labette, and Neosho Counties, Kansas. In addition, fresh dead or relict (shell shows no sign of recent mortality, such as tissue inside shell or outer shell material (periostracum) is weathered) shells were collected at 11 sites extending to near the Kansas-Oklahoma state line in Cherokee County, Kansas (Obermeyer

et al.

1997a, pp. 44-46; Obermeyer 2000, pp. 8-9). In 1994, Obermeyer

et al.

(1995, p. 24) collected 32 live Neosho mucket specimens (relative abundance = 0.6 percent) at 7 of 19 sites in Kansas. The Neosho mucket is becoming increasingly rare in the Oklahoma segment of the river (Tabor 2011, pers. comm.) with searches yielding no live or recently dead specimens. However, relict Neosho mucket shells confirm the historical presence of the species (Mather 1990, pp. 16-17; Vaughn 1996, p. 3; 1997, pp. 7-9).

Cottonwood River:

The Cottonwood River drains easterly through eastern Kansas. There are few historical records of Neosho mucket from the Cottonwood River prior to the late 1970s. Obemeyer

et al.

(1997a, p. 111) collected 59 live mussels from 6 sites surveyed from 1993 through 1995, but only found weathered dead shells of Neosho mucket. Neosho mucket was considered extirpated from the Cottonwood River until Kansas Department of Wildlife and Parks (KDWP) reintroduced mature male and brooding female Neosho mucket individuals at two sites east of Cottonwood Falls, Chase County, Kansas, in 2011 (Tabor and Barnhart 2012, pers. comm.).

Spring River:

The Spring River drains southwesterly through southwest Missouri, southeast Kansas, and eastern Oklahoma. There are few historical records of Neosho mucket from the Spring River prior to the late 1970s. Miscellaneous records from 1979 to 2010 report 10 localities yielding 119 live Neosho mucket specimens between Missouri Highway 97 near Stott City, Lawrence County, Missouri, and the Missouri and Kansas state line (McMurray 2011, pers. comm.). Cope (1985, pp. 19-20, 26-27, 33-34) collected 424 live Neosho mucket specimens out of 993 live mussels collected in 79 total one-square-meter quadrat samples from three Kansas sites upstream of Empire Lake.

Obermeyer (1996, p. 11) provides the most comprehensive status assessment of Neosho mucket in the Spring River. He collected 1,104 live Neosho mucket specimens from 13 of 20 sites extending from Missouri Highway 97 downstream to near the Turkey Creek confluence in Kansas. The KDWP surveyed a site approximately 0.5 to 0.8 rkm (0.3 to 0.5 rmi) downstream of the Kansas and Missouri state line in 2003 and collected 201 live Neosho mucket specimens (approximately 30 percent of live mussels collected). In 2006, KDWP collected 141 live Neosho mucket specimens (approximately 30 percent of live mussels collected) at a site just upstream of the Kansas and Missouri Highway YY (Miller 2011, pers. comm.). Eight to 10 percent of live Neosho mucket specimens collected at the 2006 site were quantitatively aged at less than 5 years (Tabor 2008, pers. comm.). A 2010 survey, 6 km (4 miles) east of Crestline, Kansas, found 400 live mussel specimens, of which approximately half were Neosho mucket (Tabor 2011, pers. comm.). The Spring River Neosho mucket population represents the only viable population rangewide.

North Fork Spring River:

The North Fork Spring River is a tributary of the Spring River in Missouri. There are no historical records for Neosho mucket in the North Fork Spring River prior to 1980. Neosho mucket distribution is limited to a few sites downstream of the Dry Fork confluence southwest of Jasper, Jasper County, Missouri. Three sites yielded 136 live Neosho mucket specimens in the mid 1990s (Obermeyer

et al.

1997a, p. 45; McMurray 2011, pers. comm.).

Shoal Creek:

Shoal Creek is a southern tributary of the Spring River draining portions of southwest Missouri and southeast Kansas. There are few historical records for Neosho mucket in Shoal Creek prior to 1979. Surveys of Shoal Creek conducted from 1979 to 2001 from Missouri Highway W near Ritchey, Missouri, to Empire Lake, Cherokee County, Kansas, yielded 75 live Neosho mucket specimens from 11 sites (Obermeyer

et al.

1995, p. 45; McMurray 2011, pers. comm.). No specimens were found in the Kansas portion of Shoal Creek.

Elk River:

The Elk River, a tributary of the Spring River, drains southwestern Missouri and northeastern Oklahoma. The Oklahoma reach downstream of Buffalo Creek just west of the Missouri and Oklahoma state line is inundated by Grand Lake O' the Cherokees, resulting in the loss of Neosho mucket habitat. Live Neosho mucket individuals have been collected from two sites in Missouri, eight individuals in 1978 and two individuals in 1995, and the species is rare from Noel, Missouri, to the Kansas and Missouri state line (McMurray 2011, pers. comm.). Brooding Neosho mucket females and juveniles were reported in this reach at two sites in 1992 and 1998 (Barnhart 2008, pers. comm.).

Illinois River Basin

Illinois River:

The Illinois River drains portions of northwest Arkansas and northeast Oklahoma. There are few historical records of Neosho mucket from the Illinois River prior to the late 1970s. In 1978, Gordon

et al.

(1979, pp. 35-36) surveyed 16 sites between Hogeye and Siloam Springs, Arkansas, but only report Neosho mucket as part of the mussel fauna. Eighteen live Neosho mucket specimens were reported from four Arkansas locations in the early 1990s, including the only specimen ever collected from the Muddy Fork Illinois River (Harris 1991, p. 7; Environmental and Gas Consulting, Inc. 1994, pp. field data sheets). Harris (1998) conducted a status survey of the Neosho mucket and found live specimens at 19 of 22 sites in the 48 rkm (30 rmi) reach, Washington and Benton Counties, Arkansas. Neosho mucket was the third most abundant species collected, but there was little evidence of recent recruitment (Harris 1998, p. 5).

In 2005, 92 live Neosho mucket specimens were collected from two Benton County, Arkansas, sites (Robinson Road Bridge and 800 m (2,624 feet) downstream of Chambers Spring Road, Benton County, Arkansas; Posey 2005, pers. comm.). The Arkansas Game and Fish Commission (AGFC) and the Service conducted a comprehensive status survey for Neosho mucket in the Arkansas portion of the Illinois River in 2008. Live specimens of Neosho mucket were collected at 9 of 15 survey sites. There was a 32 and 53 percent decline in number of extant (still in existence) mussel sites and sites inhabited by live Neosho mucket specimens, respectively, versus the Harris (1998) status survey. Sixty-seven percent of the sites with

Neosho mucket present were represented by three or fewer live specimens. Neosho mucket was the fourth most abundant species in this portion of the river, but 3 sites accounted for 85 percent of live Neosho mucket specimens (52 individuals) collected during this survey. Of the 15 survey sites, only 2 appear stable with the rest in decline, indicating imminent extirpation. No mussels were collected at the sites AGFC sampled in 2005 in 2008 further documenting the precipitous decline of mussels in the Arkansas portion of the Illinois River (Davidson 2011, pers. comm.).

Neosho mucket was locally common prior to the late 1990s in approximately 89 rkm (55 rmi) of the Illinois River from the Oklahoma and Arkansas state line downstream to Lake Tenkiller, Cherokee County, Oklahoma (Mather 1990, pp. 7-11). The population within the survey reach was estimated at more than 1,200 individuals in 1990. In 1995, Vaughn (1995, p. 3; 1997, p. 14) estimated the Neosho mucket population in the same reach surveyed by Mather in 1990 at between 500 and 1,000 individuals and locally common at 9 of 52 sites. Although some evidence of reproductive potential was observed during 1990 and 1995 (for example, gravid females displaying mantle lures), there was little evidence of recruitment into the population. Neosho mucket specimens were not found in or downstream of Lake Tenkiller.

Verdigris River Basin

Fall River:

The Fall River is a southern tributary of the Verdigris River in southeast Kansas. There are few historical records from the Fall River prior to the mid 1990s (Obermeyer

et al.

1995, p. 24). In 1994, Obermeyer

et al.

(1995 p. 24) found 34 live specimens (relative abundance = 1.7 percent) from 5 sites in the Fall River, with little evidence of recruitment into the population. In 2004, two sites were resurveyed and Neosho mucket composed 1.0 and 0.5 percent of qualitative and quantitative surveys, respectively (Tabor 2008, pers. comm.). All specimens were found downstream of Fall River Lake in Greenwood, Elk, and Wilson Counties (Obermeyer

et al.

1995, p. 24).

Verdigris River:

The Verdigris River flows through southeast Kansas and northeast Oklahoma until it reaches the Arkansas River in Oklahoma. There are few historical records from the Verdigris River in either State prior to the 1990s. Obermeyer

et al.

(1997a, p. 44; 1997b, p. 111) collected five Neosho mucket specimens from 4 of 14 sites from 1993 to 1995, representing 0.2 percent of the total sample from the Verdigris River between Altoona, Wilson County, Kansas, and Sycamore, Montgomery County, Kansas. The KDWP surveyed eight sites between the Fall and Verdigris River and Elk and Verdigris River confluences in 2003 and 2010. Six live Neosho mucket specimens were collected from two of these sites in 2003 (0.1 percent of the total mussel community) and seven live specimens from four sites in 2010 (0.2 percent of the total mussel community). Overall relative abundance of Neosho mucket in the Verdigris River in Kansas has ranged between 0.1 to 0.3 percent in the years from 1993 to 2010 (Miller 2011, pp. 1-2).

The majority of the Oklahoma reach has been inundated (Oologah Lake) and channelized as part of the McClellan-Kerr Arkansas River Navigation System. In 1996 and 1997, searches in the Verdigris in Oklahoma found no live Neosho mucket specimens at 32 sites. However, relict Neosho mucket shells confirmed the historical presence of the species (Vaughn 1996, p. 3; 1997, pp. 7-9). In 2008, researchers confirmed that the species is still extirpated from the Oklahoma reach (Boeckman 2008, pers. comm.).

Summary of Neosho Mucket Rangewide Population Status

The Neosho mucket is declining rangewide, with the exception of one population. Based on historical and current data, Neosho mucket has been extirpated from approximately 1,342 rkm (834 rmi) of its historical range (62 percent). Most of this extirpation has occurred within the Oklahoma and Kansas portions of its range. The extirpation of this species from numerous streams and stream reaches within its historical range signifies that substantial population losses have occurred. Extant populations are disjunct (not contiguous) in approximately 819 rkm (509 rmi). The Spring River in Missouri supports the only viable population based on the presence of a large number of individuals and evidence of recent recruitment. Given this compilation of current distribution, abundance, and status trend information, the Neosho mucket exhibits range reductions and population declines throughout its range.

Rabbitsfoot

The rabbitsfoot was originally described as

Unio cylindricus

(Say, 1817, no pagination but p. 13 of publication). The type locality is the Wabash River (Parmalee and Bogan 1998, p. 210), probably in the vicinity of New Harmony, Posey County, Indiana, and adjacent Illinois. Parmalee and Bogan (1998, p. 210) summarize the synonomy of the rabbitsfoot. The rabbitsfoot has been considered a member of the genera

Unio, Mya, Margarita, Margaron,

and

Orthonymus

at various times in history. It was first considered a member of the genus

Quadrula

by Lewis (1870, p. 218). The description of

U. cylindricus strigillatus

B.H. Wright, 1898 (=

Q. cylindrica strigillata,

the federally endangered rough rabbitsfoot; Turgeon

et al.

1998, p. 37), rendered the rabbitsfoot,

Q. c. cylindrica,

a subspecies for

Q. cylindrica.

Davis and Fuller (1981, p. 241) and Sproules

et al.

(2006, p. 3) conducted taxonomic and genetic studies on the rough rabbitsfoot (

Q. c. strigillata

) and rabbitsfoot (

Q. c. cylindrica

). Although discussion continues over the correct taxonomic placement of the rabbitsfoot, the designation of the rabbitsfoot as a species would not affect its qualification for listing under the Act as it would qualify as a listable entity whether it was a subspecies or a species.

The rabbitsfoot is a medium to large mussel, elongate and rectangular, reaching 12 cm (6 inches) in length (Oesch 1984, pp. 91-93). Parmalee and Bogan (1998, pp. 210-212) describe the beaks as moderately elevated and raised only slightly above the hinge line. Beak sculpture consists of a few strong ridges or folds continuing onto the newer growth of the umbo (raised or domed part of the dorsal margin of the shell) as small tubercles (small, rounded projection on surface of the shell). Shell sculpture consists of a few large, rounded, low tubercles on the posterior slope, although some individuals will have numerous small, elongated pustules (small raised spots) particularly on the anterior. The periostracum (external shell surface) is generally smooth and yellowish, greenish, or olive in color becoming darker and yellowish-brown with age and usually covered with dark green or nearly black chevrons and triangles pointed ventrally (Say 1817, p. 13). These patterns are absent in some individuals.

Internally, the color of the nacre is white and iridescent, often with a grayish-green tinge in the umbo cavity. Specimens from the southern periphery of its range are occasionally purplish. Soft parts generally have an orange coloration (Oesch 1984, p. 91; Parmalee and Bogan 1998, pp. 211-212). However, Vidrine (1993, p. 55) noted that the rabbitsfoot in the Ouachita River system in Louisiana had black soft parts. Aspects of the soft anatomy are

described by Ortmann (1912, pp. 256-257), Utterback (1915, pp. 148-149), Davis and Fuller (1981, pp. 228-233 and 241), and Oesch (1984, p. 91).

Suitable fish hosts for rabbitsfoot populations west of the Mississippi River include blacktail shiner (

Cyprinella venusta

) from the Black and Little River and cardinal shiner (

Luxilus cardinalis

), red shiner (

C. lutrensis

), spotfin shiner (

C. spiloptera

), and bluntface shiner (

C. camura

) from the Spring River, but host suitability information is lacking for the eastern range (Fobian 2007, p. ii). In addition, rosyface shiner (

Notropis rubellus

), striped shiner (

L. chrysocephalus

), and emerald shiner (

N. atherinoides

) served as hosts for rabbitsfoot, but not in all stream populations tested (Fobian 2007, p. 69).

Rabbitsfoot populations west of the Mississippi River reach sexual maturity between the ages of 4 to 6 years (Fobian 2007, p. 50). Rabbitsfoot exhibit seasonal movement towards shallower water during brooding periods, a strategy to increase host fish exposure but one that also leaves them more vulnerable to predation and fluctuating water levels, especially downstream of dams (Fobian 2007, pp. 48-49; Barnhart 2008, pers. comm.). It is a short-term brooder, with females brooding between May and late August (Fobian 2007, pp. 15-16). Similar to other species of

Quadrula,

the rabbitsfoot uses all four gills as a marsupium (pouch) for its glochidia (Fobian 2007, p. 26). Female rabbitsfoot release glochidia as conglutinates (matrices holding numerous glochidia together and embryos and undeveloped ova), which mimic flatworms or similar fish prey. Fecundity (capacity of abundant production) in river basins west of the Mississippi River ranged from 46,000 to 169,000 larvae per female (Fobian 2007, p. 19).

Rabbitsfoot is primarily an inhabitant of small to medium sized streams and some larger rivers. It usually occurs in shallow water areas along the bank and adjacent runs and shoals with reduced water velocity. Specimens also may occupy deep water runs, having been reported in 2.7 to 3.7 m (9 to 12 feet) of water. Bottom substrates generally include gravel and sand (Parmalee and Bogan 1998, pp. 211-212). This species seldom burrows but lies on its side (Watters 1988, p. 13; Fobian 2007, p. 24).

Rabbitsfoot historically occurred in 140 streams within the lower Great Lakes Subbasin and Mississippi River Basin (Table 2). The historical range included Alabama, Arkansas, Georgia, Illinois, Indiana, Kansas, Kentucky, Louisiana, Mississippi, Missouri, Ohio, Oklahoma, Pennsylvania, Tennessee, and West Virginia. Rabbitsfoot populations are considered to be extant in 51 streams in 13 states (Butler 2005, pp. 18-20; Boeckman 2008, pers. comm.), representing a 64 percent decline (51 extant streams of 140 historical populations). In streams where it remains extant, populations are highly fragmented and restricted to short reaches. Based upon existing habitat use (need for flowing vs. impounded habitats) and fish host (small minnow species with limited individual ranges) data, it is unlikely that recruitment between populations or establishment of new populations could occur naturally.

Although quantitative historical abundance data are rare for rabbitsfoot, relative abundance information can be gathered from museum lots. Historical museum data indicated stable rabbitsfoot populations occurred in the Ohio, Walhonding, Big Sandy, Scioto, Olentangy, Nolin, Wabash, North Fork Vermilion, Obey, Tennessee, White, Black, Spring (White River system), Strawberry, Illinois, Glover and Cossatot Rivers (Butler 2005, p. 20). Call (1895, p. 15) considered the rabbitsfoot “abundant in the St. Francis, Saline, and Ouachita Rivers in Arkansas.”

Table 2—Rabbitsfoot River and Creek Occurrences and Current Population Status

River basin

River/Creek

States

Current status

Date of last observation

Lower Great Lakes

Maumee River

St. Joseph River

IN, OH

Extirpated

1927.

1967.

Fish Creek

IN, OH

Declining

2009.

Feeder Canal

Extirpated

1908.

St. Mary's River

Extirpated

Circa 1920.

Auglaize River

Extirpated

Mid 1900s.

Ohio River

IL, IN, KY, OH, PA, WV

Stable

2005.

Allegheny River

Declining

2007.

French Creek

Stable

2008.

Le Boeuf Creek

Unknown

2006.

Muddy Creek

Declining

2003.

Conneautee Creek

Unknown

2006.

Monongahela River

Extirpated

Circa 1890.

West Fork River

Extirpated

Pre-1913.

Beaver River

Extirpated

1898.

Shenango River

Unknown

2009.

Pymatuning Creek

Extirpated

1909.

Mahoning River

OH, PA

Extirpated

Unknown.

Muskingum River

Declining

2007.

Tuscarawas River

Extirpated

Circa 1990.

Walhonding River

Declining

2009.

Killbuck Creek

Extirpated

Pre-1990.

Mohican River

Extirpated

1977.

Black Fork Mohican River

Extirpated

Pre-1990.

Little Kanawha River

Extirpated

Circa 1900.

Elk River

Extirpated

Unknown.

Big Sandy River

Extirpated

Circa 1800.

Levisa Fork

Extirpated

1909.

Scioto River

Extirpated

1962.

Olentangy River

Extirpated

1962.

Whetstone Creek

Extirpated

Pre-1930.

Big Walnut Creek

Extirpated

1961.

Alum Creek

Extirpated

1961.

Walnut Creek

Extirpated

Pre-1990.

Big Darby Creek

Declining

2002.

Little Darby Creek

Declining

2000.

Deer Creek

Extirpated

Pre-1980.

Ohio Brush Creek

Extirpated

1970.

Little Miami River

Extirpated

Circa 1900.

Licking River

Extirpated

Circa 1990.

South Fork Licking River

Extirpated

Pre-1980.

Kentucky River

Extirpated

Circa 1920.

South Fork Kentucky River

Declining

1998.

Salt River

Extirpated

Pre-1980.

Green River

Improving

2009.

Russell Creek

Extirpated

1908.

Nolin River

Extirpated

1983.

Barren River

Declining

1993.

Drakes Creek

Extirpated

1926.

West Fork Drakes Creek

Extirpated

1927.

Rough River

Declining

1993.

Wabash River

IL, IN

Declining

1988.

Mississinewa River

Extirpated

Pre-1990.

Eel River

Declining

2007.

Tippecanoe River

Stable

2005.

Vermilion River

Extirpated

Pre-1990.

North Fork Vermilion River

Declining

2006.

Middle Branch North Fork Vermilion River

Declining

2002.

Middle Fork Vermilion River

Extirpated

1918.

Salt Fork Vermilion River

Extirpated

Circa 1920.

Sugar Creek

Extirpated

1932.

Embarras River

Extirpated

Circa 1980.

White River

Extirpated

Circa 1960.

East Fork White River

Extirpated

1964.

Driftwood River

Extirpated

Circa 1940s.

Big Blue River

Extirpated

Early 1900s.

Brandywine Creek

Extirpated

Pre-1990.

Sugar Creek

Extirpated

Mid 1990s.

Flatrock River

Extirpated

Mid 1900s.

West Fork White River

Extirpated

Pre-1990.

Black Creek

Extirpated

Unknown.

Cumberland River

KY, TN

Extirpated

1979.

Rockcastle River

Extirpated

1911.

Big South Fork

Extirpated

1911.

Beaver Creek

Extirpated

1949.

Obey River

Extirpated

1939.

East Fork Obey River

Extirpated

Unknown.

Caney Fork

Extirpated

1961.

Stones River

Extirpated

1964.

East Fork Stones River

Declining

2002.

West Fork Stones River

Extirpated

1966.

Harpeth River

Extirpated

Late 1800s.

Red River

KY, TN

Declining

1992.

Whippoorwill Creek

Extirpated

Pre-1980.

Tennessee River

AL, KY, MS, TN

Stable

2009.

Holston River

Extirpated

1915.

French Broad River

Extirpated

Unknown.

Little Pigeon River

Extirpated

Unknown.

Little Tennessee River

Extirpated

Unknown.

Clinch River

Extirpated

1935.

Lookout Creek

Extirpated

1973.

Sequatchie River

Extirpated

Pre-1925.

Paint Rock River

Improving

2007.

Hurricane Creek

Extirpated

1991.

Estill Fork

Extirpated

1970.

Larkin Fork

Extirpated

1966.

Flint River

Extirpated

1955.

Elk River

Declining

2006.

Shoal Creek

AL, TN

Extirpated

Pre-1990.

Bear Creek

AL, MS

Declining

2005.

Duck River

Improving

2009.

Big Rock Creek

Extirpated

Pre-1990.

Buffalo River

Extirpated

1969.

Lower Mississippi River

St. Francis River

AR, MO

Declining

2008.

Big Creek

Extirpated

1976.

Yazoo River

Extirpated

Unknown.

Big Sunflower River

Declining

2004.

Big Black River

Declining

1980.

White River

AR, MO

Stable

2004.

War Eagle Creek

Unknown

2004.

Buffalo River

Declining

1995.

North Fork White River

Extirpated

1914.

Black River

AR, MO

Declining

2005.

Current River

Declining

1983.

Spring River

Declining

2004.

South Fork Spring River

Declining

2002.

Strawberry River

Unknown

2006.

Little Red River

Extirpated

Circa 1970.

Middle Fork Little Red River

Stable

2009.

Reeses Fork Cache River

Extirpated

1980.

Arkansas River

Verdigris River

KS, OK

Unknown

2009.

Fall River

Extirpated

Circa 1900.

Neosho River

KS, OK

Declining

1999.

Cottonwood River

Extirpated

Pre-1990.

Spring River

KS, MO

Declining

2006.

Center Creek

Extirpated

Circa 1920.

Shoal Creek

Extirpated

Pre-1920.

Illinois River

AR, OK

Declining

2008.

Red River

Blue River

Extirpated

Circa 1900.

Little River

AR, OK

Stable

2006.

Glover River

Declining

1996.

Mountain Fork Little River

Extirpated

1968.

Cossatot River

Declining

2007.

Ouachita River

AR, LA

Stable

2007.

Caddo River

Extirpated

Pre-1986.

Little Missouri River

Declining

1996.

Saline River

Declining

2006.

North Fork Saline River

Extirpated

Pre-1986.

Bayou Bartholomew

Declining

2005.

Butler (2005, pp. 89-90) categorized the extant populations of rabbitsfoot into three groups based on population size, general distribution, evidence of recent recruitment, and assessment of current viability. Sizeable populations with evidence of recent recruitment were categorized as viable. Small populations were categorized based on limited levels of recent recruitment, generally highly restricted distribution, or doubtful or limited viability increasing its susceptibility to extirpation in the near future. Marginal populations were considered rare, with no evidence of recent recruitment, of doubtful viability, and possibly on the verge of extirpation in the immediate future.

Many of the small and marginal populations are demonstrably (clearly evident) declining (Table 2). Of 21 streams with marginal populations, 9 streams (43 percent) are represented by a single recent living or fresh dead specimen. Although we have sporadic collections from the last century, trends indicate declining populations in other streams as well (for example, Allegheny River, Walhonding River, Cossatot River, Buffalo River, and Bear Creek). The following is a summary of relative abundance and trends of extant rabbitsfoot populations by river basin.

Lower Great Lakes Subbasin

The Great Lakes Basin represents the most zoogeographically (geographic distribution of an animal) distinct population center for the rabbitsfoot. All known records for the rabbitsfoot in the Great Lakes Basin are from the Maumee River system, a tributary of western Lake Erie. Populations historically occurred in five streams in addition to a canal in this system, but Fish Creek is the only remaining stream population.

Fish Creek:

Fish Creek is a tributary of the St. Joseph River, flowing through Indiana and eastward into Ohio. In 1988, rabbitsfoot comprised 1.2 percent relative abundance of all mussels in the stream (Watters 1988, p. 17). From 1996 to 2005, 17 live specimens were collected during 3 surveys (Watters 1996

Butler 2005, p. 23; Watters 2000

Butler 2005, p. 23; Brady

et al.

2004

Butler 2005, pp. 23-24; Tetzloff 2009, pers. comm.). In 2009, Ahlstedt (2009, p. 3) found one fresh dead rabbitsfoot specimen in Fish Creek. This population is categorized as marginal.

Ohio River Basin

Historically, rabbitsfoot populations were found in 66 streams within the Ohio River basin, the largest eastern tributary of the Mississippi River. Today, rabbitsfoot is extant in 20 streams, a 70 percent decline from historical stream occurrences. Several of the extant populations are represented by single living or fresh dead specimens in recent years (Muskingum, Wabash, Eel, South Fork Kentucky, Barren, and Rough Rivers and Big Darby Creek).

Ohio River:

Historically, about 60 records for rabbitsfoot have been reported over 1,570 rkm (981 rmi) of the main stem (Butler 2005, p. 25). Linear river kilometers of mussel beds in the river declined greater than 20 percent from 1967 to 1982 (Williams and Schuster 1989, pp. 7-10). By 1982, a

1,069-rkm (664-rmi) mussel survey of the Ohio River (Ohio River Mile 317.0 to 981.0) yielded one rabbitsfoot specimen from near the mouth of the Green River, Kentucky (Williams and Schuster 1989, p. 23).

Currently, two extant rabbitsfoot populations exist in the Ohio River. One population is located near Spencer County, Indiana and Hancock County, Kentucky (Clarke 1995, p. 81). The largest Ohio River rabbitsfoot population is located downstream of Lock and Dam 52 and 53. Numerous live or fresh dead rabbitsfoot specimens have been reported over the past 25 years from this reach, mostly downstream of Lock and Dam 52 (approximately Ohio River km 1,511.2 or mile 939) near Paducah, Kentucky (Butler 2005, p. 26). In addition, the rabbitsfoot population downstream of Lock and Dam 52 and 53 includes multiple age or size classes (Butler 2005, p. 26). The Ohio River and lower Tennessee River (downstream of Kentucky Lake Dam) populations may be considered a single meta-population due to the absence of a significant barrier separating them and are considered to be a sizeable population (Butler 2005, p. 26).

Allegheny River:

The Allegheny River begins in northwestern Pennsylvania, flows into New York, and then continues south into Pennsylvania before converging with the Monongahela River near Pittsburgh, Pennsylvania, to form the Ohio River. Historical records from Pennsylvania indicate rabbitsfoot was sporadically known from at least Armstrong County upstream to Warren County, Pennsylvania (Butler 2005, p. 28), but little sampling effort was performed over the past 100 years. Five live rabbitsfoot specimens were found from 1998 to 2001 at three of four intensely sampled sites at Kennerdell, Venango County, Pennsylvania (Villella 2008, pers. comm.). During surveys from 2001 to 2002 (25 sites) and 2007 (63 sites) encompassing 129 rkm (80 rmi), rabbitsfoot was found only at four sites, with very low densities. Three of four sites were downstream of the French Creek confluence (Villella 2008, pers. comm.). A 2006-2007 survey yielded no evidence of rabbitsfoot at five pools within the Allegheny River, approximately 60 rkm (37 rmi) (Smith and Meyer 2010, p. 558). The lower Allegheny River and French Creek likely represent a metapopulation because no barriers exist between the streams, but the Alleghany population is considered marginal (Butler 2005, p. 29).

French Creek:

French Creek is a major tributary of the Allegheny River, with rabbitsfoot known from downstream of Union City Reservoir to approximately 11 rkm (7 rmi) above the Allegheny River confluence, a total of 121 rkm (75 rmi) (Butler 2005, p. 31). Museum records from 1985 to 1994 indicate that rabbitsfoot was known from 12 sites (Butler 2005, p. 30). Intensive quantitative sampling at 4 sites in Venango County from 1998 to 1999 yielded 205 live rabbitsfoot specimens (Butler 2005, p. 30). In 2003 and 2004, timed searches (qualitative) yielded 41 live rabbitsfoot specimens from 12 of 25 sites in Erie, Crawford, Mercer, and Venango Counties, Pennsylvania, while a quantitative survey at 7 of 10 sites yielded 57 live rabbitsfoot specimens (Smith and Crabtree (2010 p. 391-398). Rabbitsfoot abundance at the seven sites was estimated to be from 43 to 372 individuals (standard error = 30 to 123). Evidence of recent recruitment was found at three sites (Smith and Crabtree 2010, p. 400). The French Creek population appears to be healthy and stable, with evidence of recruitment.

LeBoeuf and Conneautee Creeks:

LeBoeuf and Conneautee Creeks are tributaries of French Creek in Pennsylvania. Historical surveys for rabbitsfoot in these creeks are restricted to one relict found in 1991 from LeBoeuf Creek. In 2006, live rabbitsfoot specimens were confirmed near the confluence of each creek with French Creek. Recruitment has not been confirmed in either creek and the populations are considered marginal and likely a single meta-population with French Creek.

Muddy Creek:

Muddy Creek is a tributary of French Creek in Crawford County, Pennsylvania. Dennis (1984 p. 34) first reported the rabbitsfoot from Muddy Creek in the 1970s from a site near its confluence with French Creek. Three live rabbitsfoot specimens were collected at 3 of 20 sites in 2003, a 3-rkm (2 rmi) reach located 6 rkm (4 rmi) upstream of its confluence with French Creek (Butler 2005 p. 32; Mohler

et al.

2006, pp. 574 and 581). The rabbitsfoot population is categorized as small.

Walhonding River:

The Walhonding River converges with the Tuscarawas River to create the Muskingum River near Coshocton, Coshocton County, Ohio. The rabbitsfoot was historically common at some sites in the Walhonding River (Butler 2005, p. 32). While subsequent surveys in the early 1990's collected live mussels, relative abundance of rabbitsfoot was 0.3 percent with limited evidence of recruitment (Hoggarth 1995-1996, pp. 157, 166-174). In 2009, five live rabbitsfoot were collected from four sites located 1,203 m (3,947 ft) to 2,014 m (6,608 ft) upstream of Six Mile Dam. No live or dead rabbitsfoot individuals were collected from Six Mile Dam downstream 2,267 m (7,438 ft) (EnviroScience 2010, Figure 5). The rabbitsfoot population is categorized as small and appears to be in decline (Butler 2005, p. 33).

Shenango River:

The Shenango River is a tributary of the Beaver River in Mercer County, Pennsylvania. Nelson and Villelo (2010, p. 1) surveyed the Shenango River from Pymatuning Reservoir to Shenango River Lake in 2009 and they collected 34 live rabbitsfoot specimens (relative abundance = 1.1 percent) from this reach (Nelson and Villelo 2010, pp. 9-10). Prior to this survey, rabbitsfoot was believed to be extirpated from the Shenango River (Butler 2005, p. 96).

Muskingum River:

The Muskingum River is a major tributary of the Ohio River. Rabbitsfoot was believed to be extirpated circa 1980 until two live specimens were found in 2007 near Dresden, Muskingum County, Ohio (Service 2010, p. 10). This population is categorized as marginal.

Big Darby Creek:

Big Darby Creek is a tributary of the Scioto River in central Ohio. Watters (1994, p. 99) claimed the creek had the highest mussel diversity of any stream its size in North America. Many rabbitsfoot records exist for Big Darby Creek, dating back to the late 1950's (Butler 2005, p. 34). However, only weathered rabbitsfoot specimens were found during two intensive sampling years, 1986 and 1990 (Watters 1990, p. 31; 1994, p. 101). Since 1990, live and fresh dead rabbitsfoot records are limited to five live specimens from two localities (Tetzloff 2008, pers. comm.; Butler 2005, p. 35). Currently, the population is considered marginal.

Little Darby Creek:

Little Darby Creek is the main tributary for Big Darby Creek. Rabbitsfoot were known from Little Darby Creek dating back to circa 1960, primarily in Madison County, Ohio (Butler 2005, p. 35-36). Watters (1994, p. 101) located seven live rabbitsfoot specimens at three sites during a 1990 survey. The population in Little Darby Creek, although categorized as small, appears to be persisting and stable in approximately 32 rkm in Union and Madison Counties, Ohio (20 rmi) (Watters 1994, p. 106; Tetzloff 2008, pers. comm.).

South Fork Kentucky River:

The South Fork Kentucky River is a tributary of the Kentucky River in southeastern Kentucky that essentially converges to form the latter near Beattyville, Lee County, Kentucky. The rabbitsfoot was

first discovered in the river in the late 1990s in Owsley County; a single relict rabbitsfoot specimen was collected in 1996 and a single live specimen was observed in 1998. The population is considered marginal and of questionable viability (Butler 2005, p. 37).

Green River:

The Green River is a major Ohio River tributary, located in west-central Kentucky. Rabbitsfoot occurrences span almost 241 rkm (150 rmi) of the upper Green River (Butler 2005, p. 37). Historical rabbitsfoot records date back to circa 1900 (Butler 2005, p. 38). Periodic sampling from 1984 to 1996 produced live and fresh dead rabbitsfoot specimens from nine Green River sites between Green River Lake Dam and Munfordville, Kentucky (Cicerello 1999, p. 23). Cicerello (1999, Figure 1 and Table 1) sampled 40 sites from 1996 to 1998 over the 153-rkm (95-rmi) reach between Mammoth Cave National Park and Green River Lake Dam and reported the rabbitsfoot to be “uncommon” at 13 sites extending from Green River km 373.0 to 489.1 (mile 231.8 to 303.9; relative abundance of 0.1 percent) upstream of Munfordville, Kentucky. Sampling from 2000 to present has produced high numbers of fresh dead and numerous living specimens in Adair, Green, and Hart Counties (Butler 2005, pp. 38-39). The Green River population is one of a few rabbitsfoot populations that appear to be sizeable and improving, based on evidence of recruitment.

Barren River:

The Barren River is the largest tributary of the Green River and flows in a northwesterly direction towards its confluence with the Green River in west-central Kentucky. Historical records of rabbitsfoot in the Barren River prior to the 1990s are limited to a couple collections in the 1920s and 1940s (Butler 2005, p. 40). Two surveys since the 1990s have yielded one live rabbitsfoot and relicts in small numbers (Gordon and Sherman 1995, Appendix A). If extant, the rabbitsfoot population in the Barren River is marginal and its viability is highly doubtful (Butler 2005, p. 41).

Rough River:

The Rough River is a major Green River tributary flowing westward towards its confluence in western Kentucky. There are no historical rabbitsfoot records from the Rough River prior to the 1990s (Butler 2005, p. 41). A single fresh dead specimen collected in 1993 is the only known record of the rabbitsfoot in the Rough River (Gordon and Sherman 1995, Appendix A). This single specimen suggests a marginal and nonviable population (Butler 2005, p. 41).

Wabash River:

The Wabash River is the largest northern tributary of the Ohio River. It originates in west-central Ohio, flows across Indiana, and then forms the boundary between southwestern Indiana and southeastern Illinois. The rabbitsfoot was once widespread throughout the Wabash River prior to the 1960s (Cummings and Mayer 1997, p. 137). Surveys conducted from the 1960s through 2004 yielded a single live rabbitsfoot specimen and a few relicts (Cummings

et al.

1992, p. 3; Butler 2005, p. 42). Fisher (2006, p. 107) considered the rabbitsfoot “functionally extirpated (in the Wabash River) and restricted to the tributaries.”

Eel River:

The Eel River is a northern tributary of the Wabash River in north-central Indiana. Historical records from the Eel River prior to 1997 are sparse (Henschen 1987

Butler 2005, p. 43), but rabbitsfoot was considered common by Daniels (1903, p. 651). Collections since 1997 are limited to nine live rabbitsfoot specimens found at sites in Miami and Cass Counties, Indiana (Butler 2005, p. 43). The rabbitsfoot is no longer considered common in the Eel River, restricted to less than 32 rkm (20 rmi) of the lower main stem, and is now categorized as marginal (Butler 2005, p. 43).

Tippecanoe River:

The Tippecanoe River flows across north-central Indiana until reaching its confluence with the Wabash River. Daniels (1903, p. 651) considered the rabbitsfoot to be common in the Tippecanoe River. Surveys conducted between 1987 and 2001 yielded numerous live rabbitsfoot specimens at numerous sites (Cummings and Berlocher 1990, pp. 84-87; Ecological Specialists, Inc. 1993, pp. 47-50, 55-67, 84). Survey efforts over the past decade continue to produce similar results (EnviroScience, Inc. 2005, p. 35; Ecological Specialists, Inc. 2003, p. 9-15; Fisher 2008 and 2009, pers. comm.). The rabbitsfoot population is sizable, stable and viable in the Tippecanoe River, but at disjunct localities within the lower two-thirds of the river in Fulton, Pulaski, White, Carroll, and Tippecanoe Counties (Butler 2005, p. 45).

North Fork Vermilion River:

The North Fork Vermilion River flows south out of western Indiana into eastern Illinois until reaching its confluence with the Wabash River. Through 45 years of collection history, four sites in an approximately 10-rkm (6-rmi) reach have produced rabbitsfoot records. Since 1980, researchers have documented 28 live and 6 fresh dead rabbitsfoot specimens (Illinois Natural History Survey (INHS) museum records; Cummings

et al.

1998, p. 99). Cummings

et al.

(1998, p. 92) considered the North Fork to have “perhaps the last reproducing population of the rabbitsfoot in the state [Illinois].” The North Fork Vermilion River is considered a small metapopulation with the Middle Branch North Fork Vermilion River population (Butler 2005, p. 47).

Middle Branch North Fork Vermilion River:

The Middle Branch North Fork Vermilion River is a tributary of the North Fork Vermilion River. Headwaters of the Middle Branch drain northwestern Warren County, Indiana, and northeastern Vermilion County, Illinois. The rabbitsfoot was discovered in the lowermost reach of the Middle Branch North Fork Vermilion River in 1998 (Butler 2005, p. 47). Since that time, a few live and fresh dead rabbitsfoot specimens are known from two sites sampled in 2000 and 2002. The population is very small and apparently contiguous with the rabbitsfoot population occurring in the North Fork Vermilion River (Butler 2005, p. 47).

Cumberland River Basin

The Cumberland River is a large southern tributary of the Ohio River. Historically, the rabbitsfoot was known from the main stem and 12 tributaries. Most records for the species were prior to 1950. Parmalee

et al.

(1980, pp. 93-95) found shells of the rabbitsfoot in shellers cull and stock piles in 1977, 1978, and 1979. Rabbitsfoot was considered rare at the time, comprising less than one percent of 1,000 specimens. No more recent records exist for the main stem. Recent collections suggest populations may still exist in only two tributaries of the Cumberland River, an 85 percent decline of stream populations. The East Fork Stones and Red Rivers are the only tributaries with extant populations, and their continued survival is tenuous.

East Fork Stones River:

The East Fork Stones River is one of two major headwater tributaries, the other being the West Fork Stones River, which converge to form the Stones River. Researchers sampled numerous pre-impoundment sites from 1964 to 1967 on the East Fork Stones River, reporting rabbitsfoot from two sites but never more than three live specimens per site (Butler 2005, p. 49). Schmidt

et al.

(1989, pp. 56-59) sampled 23 East Fork Stones River sites during 1980 to 1981 and reported the rabbitsfoot to be “rare” at two lower sites. Sampling in 2002 at these two sites produced a single fresh dead specimen (Butler 2005, p. 48). The rabbitsfoot in the East Fork Stones River is considered very rare and declining;

thus it is categorized as marginal (Butler 2005, p. 49).

Red River:

The Red River is a large tributary of the lower Cumberland River that drains southwestern Kentucky and northwestern Tennessee. Despite its size, no thorough survey of the stream has ever been attempted, although there are intermittent sampling dates to the 1960s. Records indicate that a small population of the rabbitsfoot existed from a few sites on the main stem in Logan County, Kentucky, and Robertson County, Tennessee. From 1988 to 1990, the rabbitsfoot has been found live and fresh dead at five sites in Kentucky (Butler 2005, p. 49). Subsequent sampling efforts in Kentucky have yielded no additional specimens. In 1990 and 1992, the Aquatic Resources Center (ARC) (1993, p. 1 and Appendix 1) qualitatively surveyed a reach of the Red River in Tennessee and collected a total of four live rabbitsfoot (relative abundance of 2.1 and 1.3 percent, respectively). The Red River rabbitsfoot population is categorized as marginal due to its small size, distribution and doubtful viability (Butler 2005, p. 50).

Tennessee River Basin

The Tennessee River is the largest tributary of the Ohio River. Historically the rabbitsfoot was known from the entire length of the Tennessee River and 17 of its tributaries. Today, it is known only from five streams in the Tennessee River basin, a 71 percent reduction in stream populations. Almost the entire length of the 1,046-rkm (650-rmi) Tennessee River main stem has been impounded beginning in 1925, destroying hundreds of km of riverine habitat for the rabbitsfoot. Extant rabbitsfoot populations persist in the two lowermost tail waters of the Tennessee River, Duck River, Paint Rock River, Elk River, and Bear Creek.

Tennessee River:

The Tennessee River is formed from the confluence of the Holston and French Broad Rivers near Knoxville, Tennessee. Historically, the rabbitsfoot was found throughout the length of the Tennessee River (Ortmann 1925, p. 337). Today, extant populations only occur in the two lowermost tail waters, downstream of Pickwick Landing Dam and Kentucky Dam (Hubbs 2008, pers. comm.).

Over 20 live rabbitsfoot specimens were located along the marginal shelf of the Pickwick Lake tail waters in 1991 (Butler 2005, p. 51). From 1993 to 2000, live and fresh dead rabbitsfoot specimens were found at Tennessee River km 316.7 (mile 196.8, Diamond Island) and km 321.9 (mile 200). Fresh dead rabbitsfoot specimens aged at less than 10 years have been found in this same general reach of river as late as 2003 (Butler 2005, p. 124). This portion of the rabbitsfoot population exhibited recruitment in the 1990s (Hubbs 2010, pers. comm.).

Downstream of Kentucky Lake Dam, the rabbitsfoot has been found live and fresh dead at several sites in low numbers from 1985 to 2005 (Butler 2005, p. 52). In 1999, a 3.0-cm (1.2-inch) fresh dead rabbitsfoot juvenile was found at Tennessee River km 28.2 (mile 17.5) (Butler 2005, p. 52). In 2011, surveyors found greater than 80 live rabbitsfoot from Kentucky Lake Dam to the confluence with the Ohio River. Rabbitsfoot were found to occur most frequently in a narrow band of transitional substrate from clay and silt to sand and gravel along the toe of descending banks. Although not considered common, there were a few locations at which rabbitsfoot occurred in greater numbers (Koch 2012, pers. comm.). This population is likely contiguous with the population in the lower Ohio River, although the rabbitsfoot appears to be concentrated from Tennessee River km 16 to 32 (mile 10 to 20) (Butler 2005, p. 52). The Tennessee River rabbitsfoot population is considered sizable and viable (Butler 2005, pp. 89-90).

Paint Rock River:

The Paint Rock River is a northern Alabama tributary of the Tennessee River. Historically, the three headwater tributaries, Estill and Larkin Forks and Hurricane Creek, of the Paint Rock River had metapopulations of rabbitsfoot. Live rabbitsfoot specimens were collected at three of five Paint Rock River sites in 1965 and 1967 (Isom and Yokley 1973, pp. 444-445). In 1980, only two live rabbitsfoot specimens were found in the middle reaches of the river during the first comprehensive survey (18 sites; Ahlstedt 1991a, p. 168). Ahlstedt (1995-96a, pp. 69-73) sampled 18 sites in 1991 and reported good numbers of rabbitsfoot. He collected 35 live rabbitsfoot specimens at 8 of 18 main stem sites. Seven tributary sites also were sampled, but no rabbitsfoot were found in tributaries.

During more recent sampling efforts in 1995 and 2002, three fresh dead and nine relict shells were found at a main stem site and a single live specimen upstream of the Larkin Fork confluence, respectively (McGregor and Shelton 1995, Appendix A; Godwin 2002, pp. 10-11, 22-23). In 2004, two live and some fresh dead rabbitsfoot specimens were found at a site on the lower main stem (Butler 2005, p. 54). An intensive survey (42 main stem and 5 Estill Fork sites) in 2008 found 218 live and fresh dead rabbitsfoot at 19 sites. Rabbitsfoot was the second most abundant species (Fobian

et al.

2008, pp. 6-37). This population is categorized as sizeable and viable (Butler 2005, pp. 89-90).

Elk River:

The Elk River is a tributary of the Tennessee River draining portions of south-central Tennessee to north-central Alabama. From 1965 to 1967, Isom

et al.

(1973, pp. 438-440) found the rabbitsfoot at three locations on the Elk River. Survey efforts on Elk River tributaries, Sugar and Richland Creek, did not yield any rabbitsfoot. In 1980, Ahlstedt (1983, pp. 44-45) found 10 live rabbitsfoot specimens at 6 of 108 sites in the Elk River, Lincoln County, Tennessee (Ahlstedt 1983, pp. 46-49). Two live rabbitsfoot specimens were found at approximately Elk River km 122 (mile 76) in 1999 (Service 1999, p. 6). Tennessee Valley Authority conducted a survey in 2006 and found three live individuals, one objectively aged at 6 or 7 years (Chance 2008, pers. comm.). This population is categorized as marginal (Butler 2005, pp. 89-90).

Bear Creek:

Bear Creek is a southern tributary of the Tennessee River in northwestern Alabama and northeastern Mississippi. Historical records indicate rabbitsfoot occurred in 72 rkm (45 rmi; Ortmann 1925, p. 337; Butler 2005, pp. 56-57). In 1977, three live rabbitsfoot specimens were found at approximately Bear Creek km 90 (rmi 56) in Alabama (Butler 2005, p. 56). A 1991 record of a single fresh dead specimen is known from approximately Bear Creek km 40 (mile 25) in Colbert County, Alabama. McGregor and Garner (2004, p. 64) conducted the only comprehensive survey of the system from 1996 to 2001 and found rabbitsfoot live or fresh dead at two sites. It occurred on the main stem in the immediate vicinity of the Natchez Trace Parkway of the National Park Service (NPS) system in Colbert County, Alabama (Bear Creek km 39.4 and 40.9; mile 24.5 and 25.4). In Mississippi, one live and eight fresh dead specimens were found in a four-rkm (2.5-rmi) reach in 2002 and 2005 (Jones 2011, pers. comm.). Bear Creek is categorized as a small population (Butler 2005, pp. 89-90).

Duck River:

The Duck River is a large tributary of the lower Tennessee River in central Tennessee. Ortmann (1924, pp. 24-33) documented the presence of rabbitsfoot in the early 1920s, considering it “all over the interior region (and elsewhere).” Surveys conducted between 1965 and 1979 found similar results (Isom and Yokley 1968, p. 36; Ahlstedt 1981, p. 62; Ahlstedt 1991, pp. 142-147).

Using stratified random sampling, Barr

et al.

(1993-94, p. 205) in 1981 estimated that 591 live rabbitsfoot

occurred at Lillards Mill. Twenty rabbitsfoot were collected from Lillards Mill and translocated to a site in Bedford County in 1988 (Layzer and Gordon 1993, pp. 89-91). Resampling the Bedford County site in 2002, evidence of recruitment was noted by Ahlstedt

et al.

(2004, p. 101). Madison

et al.

(1999, Table 1) reported 34 live rabbitsfoot specimens from a Maury County site in 1998.

Ahlstedt

et al.

(2004, p. 101) conducted an extensive mussel survey in the system beginning in 2000. They reported 403 live and fresh dead rabbitsfoot specimens from 31 of 78 sites sampled (a few sites were sampled more than once). An average of 13 live or fresh dead rabbitsfoot specimens was found per site of occurrence. The rabbitsfoot population on the Duck River is primarily located between rkm 209 to 288 (miles 179 to 130), and scattered in the lower river (rkm 60 to 61; rmi 37 to 38; Hickman County) (Hubbs 1995, p. 46; Schilling and Williams 2002, p. 409; Butler 2005, p. 59). The extant rabbitsfoot population extends over at least 274 rkm (170 rmi; approximately Duck River km 60 to 333, mile 37 to 207) and “* * * represents one of the best known populations rangewide” (Ahlstedt

et al.

(2004, p. 101).

Lower Mississippi River Subbasin

The rabbitsfoot is known from five streams within the lower Mississippi River subbasin (excluding the White, Arkansas, and Red River systems). The five streams include St. Francis River, Big Creek, Yazoo River, Big Sunflower River, and Big Black River. Rabbitsfoot is extirpated from Big Creek and the Yazoo River (Butler 2005, p. 61).

St. Francis River:

The St. Francis River is a tributary of the Mississippi River draining portions of southeastern Missouri and northeastern Arkansas. In the 1800s the rabbitsfoot was considered abundant in the St. Francis River (Call 1895, p. 15). Extant rabbitsfoot records are from the upper part of the river in Butler and Wayne Counties, Missouri (Butler 2005, p. 61). Hutson and Barnhart (2004, pp. 84, 109) in 2002 found 16 live rabbitsfoot specimens at 3 sites upstream of Lake Wappapello, Missouri; including 11 at rkm 277.0 (rmi 172.1), 3 at rkm 294.5 (rmi 183.0), and 2 at rkm 306.6 (rmi 190.5). At rkm 277.0 (rmi 172.1), 35 live rabbitsfoot specimens were found in the 1970s, but only 8 and 11 live specimens were found in 2001 and 2002, respectively. In 2005, seven live rabbitsfoot specimens were sampled at a site in the same reach (Butler 2005, p. 62). With the exception of Call's description, no rabbitsfoot have been found in the St. Francis River, Arkansas (Butler 2005, p. 61). The rabbitsfoot is rare in the St. Francis River, may be at risk from extirpation (Hutson and Barnhart 2004, p. 84), and is categorized as a small population (Butler 2005, pp. 89-90).

Big Sunflower River:

A major tributary of the Yazoo River, the Big Sunflower River drains a large portion of the Mississippi Delta in west-central Mississippi. The rabbitsfoot was first reported in 1969 from the lower portion of the river (Florida Museum of Natural History, museum lot # 233299). Currently, rabbitsfoot occurs in a 32-rkm (20-rmi) reach upstream of the Quiver River confluence in Sunflower County. From 2000 to 2010, live and fresh dead rabbitsfoot specimens were collected at Blaine Road west of Blaine, Mississippi, downstream to near the Quiver River confluence (Jones 2011, pers. comm.). Butler (2005, pp. 89-90) categorized this population as small.

Big Black River:

The Big Black River is a tributary to the lower Mississippi draining central and southwestern Mississippi. Hartfield and Rummel (1985, pp. 117-119) sampled the lower three-quarters of this 426-rkm (265-rmi) long river. The rabbitsfoot is restricted to a small portion of the lower river cutting through the Loess Hills physiographic division where mussels were generally found in gravel riffles and runs. At that time, 19 dead rabbitsfoot specimens were recorded at nine sites in Hinds and Warren Counties (Butler 2005, p. 64). The only other record is for a dead specimen located in 2000. Rabbitsfoot is still considered extant in this reach (Jones 2011, pers. comm.), and the population is categorized as small.

White River Basin

Historically, 13 rivers within the White River system harbored rabbitsfoot populations. Extant populations occur in 9 of 13 (69 percent) rivers in the basin. Further, no other major river basin has as many sizeable populations. At one time, the main stem of White River and 11 of its tributaries had a large metapopulation of rabbitsfoot (Butler 2005, p. 65). Three of the streams may still contain a metapopulation (Black, Spring, and Strawberry Rivers). Unfortunately, many of the tributaries appear to have declining populations (Buffalo, Black, Current, Spring, and South Fork Spring Rivers).

White River:

The White River is a large western tributary of the Mississippi River. The rabbitsfoot population once extended throughout most of the 1,110-rkm (690-rmi) length of the White River and site records date back to circa 1910, but now it is restricted to the lower reaches downstream of Batesville, Independence County, Arkansas (Harris

et al.

2009, p. 73). Historical abundance data are scarce. However, records indicate that the population was large (Butler 2005, p. 65).

From the 1980s to late 2000s, numerous live and fresh dead rabbitsfoot specimens have been found at numerous sites in two disjunct reaches of the White River (rkm 319 and 410; rmi 198 and 255 and rkm 92 to 146; rmi 57 to 91) (Bates and Dennis 1983, p. 42; AGFC Mussel Database 2011). In 1992, Christian (1995, pp. 146-197) estimated the total rabbitsfoot population from 13 sites on the lower White River at 928 individuals. The rabbitsfoot population is categorized as sizable, but remains extant in two disjunct reaches separated by approximately 161 rkm (100 rmi). The uppermost reach extends from the Batesville Dam at Batesville, Independence County, Arkansas, downstream to the Little Red River confluence north of Georgetown, White and Woodruff Counties, Arkansas. The lowermost reach extends from U.S. Highway 79 at Clarendon, Monroe County, Arkansas, downstream to Arkansas Highway 1 near St. Charles, Arkansas County, Arkansas (Butler 2005, p. 66; AGFC mussel database 2011).

War Eagle Creek:

War Eagle Creek is a small, eastern White River tributary located in northwest Arkansas. Rabbitsfoot was not documented in War Eagle Creek until 1974. Since 1979, one live specimen was collected in 1981, and two fresh dead were found in 2004 (AGFC mussel database 2011). Little is known about the viability of this population. Therefore, it has been categorized as marginal (Butler 2005, pp. 89-90).

Buffalo River:

The Buffalo River is a western White River tributary in north-central Arkansas. Rabbitsfoot was first documented in the Buffalo River in 1910 by Meek and Clark (1912, pp. 7-20). They reported rabbitsfoot as “common” at 11 of 26 sites; almost all specimens were located within the lower 40 rkm (25 rmi) within Searcy County, Arkansas. Two comprehensive surveys of the Buffalo River mussel fauna in 1995 and 2004 to 2005 found live rabbitsfoot specimens concentrated between Arkansas Highway 7 in Newton County to near the Cedar Creek confluence downstream of Rush, Arkansas (Harris 1996, p. 12; Matthews

et al.

2009, pp. 116 and 122). NPS staff collected four live rabbitsfoot in 2008 from a site near the Cedar Creek

confluence near Rush, Arkansas (Hodges 2011, pers. comm.). During a 2011 survey of this same site, changes in channel geomorphology caused by 2009 and 2011 flooding resulted in the entire site being covered with sand. Few live mussels were encountered, but one live rabbitsfoot was found and relocated to more suitable habitat downstream. While no live rabbitsfoot were encountered at the downstream relocation site, 2 fresh dead and 23 weathered rabbitsfoot shells were found at this site. Two live rabbitsfoot also were collected in 2011 at two sites located between Arkansas Highway 7 and U.S. Highway 65. The Buffalo River population is small and very susceptible to extirpation based on recent surveys (Davidson 2011, pers. comm.).

Black River:

The Black River is the largest White River tributary draining southeastern Missouri and northeastern Arkansas. Based on data from the 1970s and 1980s, the rabbitsfoot was abundant at some Arkansas sites in the lower main stem between the confluences of the Current and Strawberry Rivers (approximately 121 rkm, 75 rmi; Ohio State University Museum of Biological Diversity (OSUM) museum lot #s 47673 and 47933; Miller and Hartfield 1986, pp. 8-9). In 1992, Rust (1993, Appendix 1.1) surveyed 48 sites in the Black River, finding rabbitsfoot live at 4 sites, and a combined population estimate of 1,503 individuals, between rkm 105 to 124 (rmi 65 to 77). A 2000 to 2003 survey at 51 sites in Missouri did not locate any rabbitsfoot (Hutson and Barnhart 2004, pp. 162-169). In 2005, AGFC collected 25 live rabbitsfoot specimens from a site located approximately two rkm (1 rmi) upstream of U.S. Highway 63 at Black Rock, Arkansas (AGFC Mussel Database 2011). The Black River population is considered one of the largest remaining range-wide (Butler 2005, pp. 89-90).

Current River:

The Current River is a Black River tributary draining southeastern Missouri and northeastern Arkansas. The rabbitsfoot is known only from the Arkansas portion of the stream. Few records exist for the species in the Current River, including several live and dead specimens in 1983-1984 and 1994 (AGFC mussel database). The rabbitsfoot population in the Current River is categorized as marginal.

Spring River:

The Spring River is a Black River tributary draining south-central Missouri and northeastern Arkansas. Based on pre-1986 records, the rabbitsfoot was once known from at least 14 sites in the 80-rkm (50-rmi) reach downstream of the South Fork Spring River confluence (Harris

et al.

1997, pp. 80-82). Records from the 1980s also indicate that the rabbitsfoot was “relatively common” (Miller and Hartfield 1986, pp. 9-10; Harris and Gordon 1987, p. 54; ANSP 359907). A survey upstream of the South Fork Spring River confluence in 1985 did not find any rabbitsfoot (Miller and Hartsfield 1986, p. 9). In 1991, Rust (1993, Appendices 1.2 and 1.4) estimated rabbitsfoot relative abundance at 1.9 to 4.0 percent at 5 of 6 sites and total population size at 563 individuals at 3 of these sites. Sixty-eight live rabbitsfoot were collected in the river reach from near Ravenden to Imboden, Arkansas, during 2004 to 2005 (Harris

et al.

2007, p. 16). The rabbitsfoot population appears to be recruiting, but the numbers of individuals are decreasing from the high numbers found in the mid-1980s (Butler 2005, p. 72). For this reason, the Spring River is categorized as a small rabbitsfoot population.

South Fork Spring River:

The South Fork Spring River is a Spring River tributary draining portions of Howell County, Missouri, and Fulton and Sharp Counties, Arkansas. The rabbitsfoot was discovered in the South Fork Spring River in 2002 in central Fulton County, Arkansas (Butler 2005, p. 72). Judging from the number of fresh dead and relict shells found, it appears to have been the dominant species at this site, although no live mussels were located (Butler 2005, pp. 72-73). In 2006, a qualitative survey to assess mussel communities at 35 sites in the South Fork Spring River did not yield any rabbitsfoot (Martin

et al.

2009, pp. 106-107). However, one live rabbitsfoot specimen was located on the river a week later, representing the only live specimen ever collected from the river (AGFC mussel database 2011). Based on limited information collected over the past decade on the rabbitsfoot status in the South Fork Spring River, this population is categorized as small.

Strawberry River:

The Strawberry River is a Black River tributary draining portions of northeastern Arkansas. The most upstream record of live rabbitsfoot in the Strawberry River was collected 2.9 rkm (1.8 rmi) upstream of Hars Creek southeast of Franklin, Arkansas, in 1998 (AGFC Mussel Database 2011). From 1983 to 2006, 84 live rabbitsfoot specimens, including some juveniles, have been collected from 14 sites extending from the most upstream record downstream through Sharp and Lawrence counties (greater than 80 rkm or 50 rmi) (Rust 1993, p. 30; Harris

et al.

2007, pp. 23-27; INHS 27526). The Strawberry River rabbitsfoot population is categorized as sizable.

Middle Fork Little Red River:

The Middle Fork Little Red River is a headwater tributary of the Little Red River in north-central Arkansas. Rabbitsfoot was first discovered in the Middle Fork in 1991 with a single specimen from 26 sites (Harris 1992, p. 64). The Middle Fork Little Red River has been extensively surveyed during the past decade. Winterringer (2003, p. 46 and Appendix F) found 28 live rabbitsfoot specimens, including 2 juveniles, at 2 sites sampled in 2001 downstream of Little Tick Creek. The AGFC and Service collected seven live rabbitsfoot, including one juvenile, from two sites in this same reach in 2009 (Davidson 2011, pers. comm.). The rabbitsfoot population is categorized as small.

Arkansas River Basin

The rabbitsfoot distribution in the Arkansas River system is restricted to tributaries draining the western fringe of the Ozark Plateaus and adjacent Central Lowlands physiographic provinces located to the west. The rabbitsfoot range in the system includes east-central and southeastern Kansas, northeastern Oklahoma, extreme northwestern Arkansas, and extreme southwestern Missouri. Rabbitsfoot was once distributed throughout hundreds of km (miles) of streams in the basin, with populations in the Fall and Cottonwood Rivers and Center and Shoal Creeks now extirpated (50 percent reduction in stream populations). Scammon (1906, pp. 348-349) described rabbitsfoot as “seeming to be nowhere abundant, it is not a rare species in [the Spring, Neosho, and Verdigris Rivers].” Rabbitsfoot is now confined to reduced portions of the Verdigris, Neosho, Spring, and Illinois Rivers.

Neosho River:

The Neosho River is a large northern tributary to the Arkansas River in eastern Kansas and northeastern Oklahoma. Historical evidence indicates rabbitsfoot was present in almost the entire 740-rkm (460-rmi) main stem of the Neosho River (Butler 2005, p. 75). Live rabbitsfoot specimens, including some juveniles, have been collected in a 12.8-rkm (8-rmi) reach from near Iola to Humboldt, Allen County, Kansas, from 1994 to 1999 (Obermeyer

et al.

1995, pp. 31-32; Mulhern

et al.

2002, p. 243; Butler 2005, p. 76). Relict shells were collected at 8 of 21 additional main stem sites from 1993 to 1995 (Obermeyer

et al.

1995, p. 63). The rabbitsfoot is thought to be extirpated from the Oklahoma portion and remaining stretches in Kansas. The extant population in Kansas is categorized as small.

Spring River:

The Spring River is a Neosho River tributary draining portions of southwest Missouri, southeast Kansas, and northeast

Oklahoma. Rabbitsfoot is extant in the Spring River from Missouri Highway 96 in Carthage, Jasper County, Missouri, downstream to the confluence of Turkey Creek north of Empire, Cherokee County, Kansas. Six live rabbitsfoot specimens were collected from four Missouri sites in the early 1990's and 2006 (Obermeyer

et al.

1995, p. 48; Missouri Natural Heritage Database 2011). In 2003, a Kansas site (known as the Pierce Site) located approximately 0.5 to 0.8 rkm (0.3 to 0.5 rmi) yielded 10 live rabbitsfoot, including 7 gravid females (Miller 2011). In 2006, KDWP collected eight live rabbitsfoot specimens from one 30 m

quadrat sample (1.9 percent of live mussels collected) at a site just upstream of Kansas and Missouri Highway YY. This rabbitsfoot population is categorized as small.

Illinois River:

The Illinois River is an Arkansas River tributary draining portions of northwest Arkansas and northeast Oklahoma. Gordon

et al.

(1979, p. 35) surveyed 11 sites in Arkansas in the 1970s and found only a single shell. In 1994, Harris (1998, p. 4) found 34 live rabbitsfoot specimens at 7 of 22 sites in a 48-rkm (30-rmi) reach in Washington and Benton counties, Arkansas. In 1995, Vaughn (1997, pp. 28-30) surveyed 45 sites in Oklahoma and found live rabbitsfoot at 2 sites. A 2008 survey in Benton and Washington Counties found 10 live rabbitsfoot at 2 of 15 sites extending from just upstream of Muddy Fork to the Arkansas Highway 59 Bridge (Davidson 2011, pers. comm.). This population is categorized as marginal.

Verdigris River:

The Verdigris River is an Arkansas River tributary draining portions of Kansas and Oklahoma. Rabbitsfoot is extant in a short reach from Oologah Lake dam north of Claremore, Oklahoma, downstream to Interstate 44 (Will Rogers Turnpike) west of Catoosa, Rogers County, Oklahoma. Numerous live rabbitsfoot specimens were collected at three sites clustered upstream and downstream of Oklahoma Highway 20 west of Claremore, Oklahoma, in 2006 and 2007 (Boeckman 2008, pers. comm.). Rabbitsfoot has been extirpated from reaches of the Verdigris River upstream of Oologah Lake in Kansas and Oklahoma. This population is categorized as marginal due to its restricted distribution.

Red River Basin

Streams within the Red River basin primarily drain the Ouachita Mountains in southeastern Oklahoma and southwestern Arkansas, but extant populations still occur in three stream reaches within the Gulf Coastal Plain ecoregion in southern Arkansas and northern Louisiana. The rabbitsfoot is extant in 7 of 11 historical streams (64 percent) within the Red River basin.

Little River:

The Little River is a Red River tributary draining portions of southeastern Oklahoma and southwestern Arkansas. Isley (1924, p. 57) discovered one specimen in 1910. In 1983, six live individuals were located in Sevier County, Arkansas (AGFC mussel database 2011). Vaughn and Taylor (1999, p. 920) collected live rabbitsfoot specimens at six sites in the Little River located downstream of the Glover River confluence. Its “abundance,” defined as the number of mussels found per hour spent searching, ranged from 0.6 to 8.0 at these sites. In 2002, survey work occurred in the lowermost section, downstream of Millwood Reservoir, and no rabbitsfoot were located at any of the 14 sites surveyed (Farris

et al.

2003, Appendix A). From 2006 to 2008, the AGFC and Service collected 89 live rabbitsfoot specimens from 13 Little River sites extending from near the Arkansas and Oklahoma state line to near U.S. Highway 71 north of Ashdown, Arkansas (AGFC Mussel Database, 2011). The rabbitsfoot population is sizeable and considered viable in this reach of the Little River (Davidson 2011, pers. comm.).

Glover River:

The Glover River is a Little River tributary draining portions of southeastern Oklahoma. Museum records indicate a healthy population of rabbitsfoot once occupied a 48-rkm (30-rmi) reach of the river (Butler 2005, p. 82). An unspecified number of specimens were located in a 1993 to 1995 survey (Vaughn 2000, pp. 229). In 1996, researchers systematically surveyed 22 sites, and rabbitsfoot relative abundance was 0.7 and 3.0 percent at 2 sites (Vaughn 2003, p. 3). The Glover River appears to support a marginal population of rabbitsfoot that is greatly diminished from historical accounts (Vaughn 2003, p. 1).

Cossatot River:

The Cossatot River is a Little River tributary draining portions of southwestern Arkansas. Few mussel collections have been made in the Cossatot River. Rabbitsfoot was first collected in 1970, with evidence of population recruitment (Butler 2005, p. 83). Twelve specimens were found in 1983 at a site in Sevier County, Arkansas (AGFC mussel database 2011). In 2004, four live specimens were found at one site (AGFC mussel database 2011). Viability of the population is doubtful, based on its small size and isolated location, and the population is categorized as marginal. However, no comprehensive survey data for the river exists (Butler 2005, p. 83).

Ouachita River:

The Ouachita River is the largest tributary of the Red River, draining a large portion of southern Arkansas and eastern Louisiana. Wheeler (1918, pp. 122-123) observed rabbitsfoot in the Ouachita River and declared it “in nearly every mussel bed of the river.” Call (1895, p. 15) also considered the rabbitsfoot “abundant.” The rabbitsfoot is extant in a short reach (two sites) of the Ouachita River from Arkansas Highway 379 south of Oden, Montgomery County, Arkansas, downstream to Arkansas Highway 298 east of Pencil Bluff, Montgomery County, Arkansas (AGFC Mussel Database, 2011). Three reservoirs (Lakes Ouachita, Hamilton, and Catherine) separate the headwaters in the Ouachita Mountains from the Gulf Coastal Plain reaches in southern Arkansas and Louisiana.

Researchers collected 38 live specimens from 1992 to 2005 at 8 sites in Clark, Hot Spring, and Ouachita Counties, Arkansas (Posey 1997, Appendix 1.3; Butler 2005, p. 84, Harris 2006, Appendix 1e—1i; AGFC Mussel Database, 2011). Posey (1997, Appendix 1.3) estimated the rabbitsfoot population at 1,456 individuals in the Ouachita River from rkm 547 to 563 (rmi 340 to 350). Rabbitsfoot has not been observed in the Louisiana reach of the Ouachita River in over 100 years (Butler 2005, p. 84). The Ouachita River population is categorized as small due to its greatly diminished distribution and limited evidence of recent recruitment.

Little Missouri River:

The Little Missouri River originates in the Ouachita Mountains and flows southeast to the Ouachita River in southwest Arkansas. The rabbitsfoot is known from a single collection in 1996 in the lower main stem in Clark County, Arkansas (Davidson 1997, pp. 46 and 130). The Little Missouri population likely is a metapopulation with the Ouachita River population and is categorized as marginal (Butler 2005, p. 85).

Saline River:

The Saline River flows southward through south-central Arkansas before converging with the Ouachita River at Felsenthal National Wildlife Refuge (NWR) north of the Arkansas and Louisiana State line. Call (1895, p. 15) considered the rabbitsfoot “abundant” in the Saline River. Two fresh dead and one live specimen were documented in 1993 and 2006, respectively, in Grant County (AGFC Mussel database 2011). Davidson (1997) surveyed the Saline River from the northern boundary of Felsenthal NWR to its confluence with the Ouachita

River and was unable to locate any live rabbitsfoot. Davidson and Clem (2002, p. 17; 2004, p. 16) collected 26 live rabbitsfoot specimens from 13 of 230 sites from near Tull, Arkansas, to the northern boundary of Felsenthal NWR. Rabbitsfoot comprised 0.2 percent of the total mussel community. In 2005, Harris (2006, Appendix 1b-1d) quantitatively sampled three of the sites sampled by Davidson and Clem in 2004. He collected 24 live rabbitsfoot, representing 0.1 to 0.8 percent of the total mussel community per site. These sites were resampled in 2011 and four live rabbitsfoot were collected, representing zero to 0.1 percent of the total mussel community (Davidson 2012, pers. comm.). In 2011, the AGFC and Service collected 33 live rabbitsfoot, representing 0.1 to 0.3 percent of the total mussel community. Numerous dead rabbitsfoot were observed near the shoreline, apparently having succumbed to desiccation caused by severe drought conditions (Davidson 2012, pers. comm.). The rabbitsfoot population is categorized as small due to its “patchy” distribution, but there is evidence of recent recruitment (Davidson and Clem 2004, p. 16; Davidson 2011, pers. comm.).

Bayou Bartholomew:

Bayou Bartholomew originates in southeast Arkansas and flows south into Louisiana before converging with the Ouachita River. The first record of rabbitsfoot in Bayou Bartholomew is from 1992 in Louisiana (Butler 2005, p. 87). One live specimen was found in Louisiana between 2000 and 2001 (Alley 2005, p. 75). From 2004 to 2005, two sites yielded five live and six dead specimens. A 2004 survey at 50 sites in the Arkansas portion of Bayou Bartholomew did not yield any live, dead, or relict rabbitsfoot specimens (Brooks

et al.

2008, pp. 9-10). All records since 2000 are from three sites in Louisiana, two in the middle Louisiana reach and one near the Arkansas state line (Butler 2005, p. 87). This population is categorized as marginal.

Summary of Rabbitsfoot Rangewide Population Status

Based on historical and current data, the rabbitsfoot is declining rangewide. In ten of the 15 States comprising the rabbitsfoot's historical range, the species is considered by State law to be endangered (Illinois, Indiana, Kansas, Mississippi, Ohio, and Pennsylvania); threatened (Kentucky and Tennessee); of special concern (Arkansas); or it is assigned an uncategorized conservation status (Alabama). The American Malacological Union and American Fisheries Society also consider the rabbitsfoot to be threatened (

Butler 2005, p. 21). It is presently extant in 51 of the 140 streams of historical occurrence, a 64 percent decline. Further, in the streams where it is extant, populations with few exceptions are highly fragmented and restricted to short reaches. In addition, the species has been extirpated from West Virginia and Georgia. The extirpation of this species from numerous streams and stream reaches within its historical range signifies that substantial population losses have regularly occurred in each of the past several decades. Seventeen streams (33 percent of extant populations or 12 percent of historical populations) have small populations with limited levels of recruitment and are generally highly restricted in distribution, making their viability unlikely and making them extremely susceptible to extirpation in the near future. In addition, 15 of those 17 streams (88 percent) have populations that are declining. In many of these streams, rabbitsfoot is only known from one or two documented individuals in the past decade. Its viability in these streams is doubtful and additional extirpations may occur if this downward population trend is not eliminated. Eleven populations located in historical streams (22 percent of extant populations or 8 percent of historical populations; Ohio, Green, Tippecanoe, Tennessee, Paint Rock, Duck, White, Black, Strawberry, and Little Rivers and French Creek) are considered viable (Butler 2005, p. 88; Service 2010, p. 16). Given this compilation of current distribution, abundance, and status trend information, the rabbitsfoot exhibits range reductions and population declines throughout its range.

Summary of Factors Affecting the Species

Section 4 of the Act (16 U.S.C. 1533) and its implementing regulations at 50 CFR part 424 set forth the procedures for adding species to the Federal Lists of Endangered and Threatened Wildlife and Plants. Under section 4(a)(1) of the Act, we may list a species based on any of the following five factors: (A) The present or threatened destruction, modification, or curtailment of its habitat or range; (B) overutilization for commercial, recreational, scientific, or educational purposes; (C) disease or predation; (D) the inadequacy of existing regulatory mechanisms; and (E) other natural or manmade factors affecting its continued existence. Listing actions may be warranted based on any of the above threat factors, singly or in combination. Each of these factors is discussed below.

A. The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range

The habitats of freshwater mussels are vulnerable to water quality degradation and habitat modification from a number of activities associated with modern civilization. The decline, extirpation, and extinction of mussel species are often attributed to habitat alteration and destruction (Neves

et al.

1997, pp. 51-52). Bogan (1993, pp. 599-600 and 603-605) linked the decline and extinction of mussels to a wide variety of threats including siltation, industrial and municipal effluents, modification of stream channels, impoundments, pesticides, heavy metals, invasive species, and the loss of host fish. Chief among the causes of decline in distribution and abundance of the Neosho mucket and rabbitsfoot, and in no particular order of ranking, are impoundment, channelization, sedimentation, chemical contaminants, mining, and oil and natural gas development (Mather 1990, pp. 18-19; Obermeyer

et al.

1997b, pp. 113-115; Neves

et al.

1997, pp. 63-72; Davidson 2011, pers. comm.). Neosho mucket and rabbitsfoot are both found within medium to large river drainages exposed to a variety of landscape uses. These threats to mussels in general (and Neosho mucket and rabbitsfoot where specifically known) are individually discussed below.

Impoundments

Dams eliminate and alter river flow within impounded areas, trap silt leading to increased sediment deposition, alter water quality, change hydrology and channel geomorphology, decrease habitat heterogeneity, affect normal flood patterns, and block upstream and downstream movement of mussels and fish (Layzer

et al.

1993, pp. 68-69; Neves

et al.

1997, pp. 63-64; Watters 2000, pp. 261-264). Within impounded waters, decline of mussels has been attributed to direct loss of supporting habitat, sedimentation, decreased dissolved oxygen, temperature levels, and alteration in resident fish populations (Neves

et al.

1997, pp. 63-64; Pringle

et al.

2000, pp. 810-815; Watters 2000, pp. 261-264). Downstream of dams, mussel declines are associated with changes and fluctuation in flow regime, channel scouring and bank erosion, reduced dissolved oxygen levels and water temperatures, and changes in resident fish assemblages (Williams

et al.

1992, p. 7; Layzer

et al.

1993, p. 69; Neves

al.

1997, pp. 63-64; Watters 2000, pp. 265-266; Pringle

et al.

2000, pp. 810-815). Dams that are low to the water surface, or have water passing over them (small low head or mill dams) can have some of these same effects on mussels and their fish hosts, particularly reducing species richness and evenness and blocking fish host movements (Watters 2000, pp. 261-264; Dean

et al.

2002, pp. 235-238). The decline of mussels within the Arkansas, Red, White, Tennessee, Cumberland, Mississippi, and Ohio River basins has been directly attributed to construction of numerous impoundments (Miller

et al.

1984, p. 109; Williams and Schuster 1989, pp. 7-10; Layzer

et al.

1993, pp. 68-69; Neves

et al.

1997, pp. 63-64; Obermeyer

et al.

1997b, pp. 113-115; Watters 2000, pp. 262-263; Sickel

et al.

2007, pp. 71-78; Hanlon

et al.

2009, pp. 11-12; Watters and Flaute 2010, pp. 3-7). Population losses due to impoundments have likely contributed more to the decline of the Neosho mucket and rabbitsfoot than any other factor. River habitat throughout the ranges of the Neosho mucket and rabbitsfoot has been impounded, leaving short, isolated patches of suitable habitat that sometimes lacks suitable fish hosts. Neither Neosho mucket nor rabbitsfoot occur in reservoirs lacking riverine characteristics. They are unable to successfully reproduce and recruit under these conditions (Obermeyer

et al.

1997b, p. 114; Butler 2005, p. 96). On the other hand, rabbitsfoot may persist and even exhibit some level of recruitment in some large rivers with locks and dams where appropriate habitat quality and quantity remain (Ohio and Tennessee Rivers in riverine reaches between a few locks and dams) (Butler 2005, p. 96).

The majority of the main stem Ohio, Cumberland, Tennessee, and White Rivers and many of their largest tributaries are impounded, in many cases resulting in tail water (downstream of dam) conditions unsuitable for rabbitsfoot (Butler 2005, p. 96). There are 36 major dams within the Tennessee River basin (Holston, Little Tennessee, Clinch, Elk, Flint, and Sequatchie Rivers, and Bear Creek) that have resulted in the impoundment of 3,680 rkm (2,300 rmi) of the Tennessee River and its largest tributaries (Butler 2005, p. 95). Only three of these rivers support viable populations—Tennessee, Paint Rock, and Duck Rivers. Ninety percent of the Cumberland River downstream of Cumberland Falls (rkm 866, rmi 550) as well as numerous tributaries are either directly impounded or otherwise adversely affected by cold tail water releases from dams. Rabbitsfoot and its fish hosts are warm water species and the change in temperature to cold water below the dams further reduces suitable habitat for the species and may eliminate fish hosts that cannot adapt to colder water temperatures (see the Temperature section below for more information). Other tributary impoundments that adversely affected rabbitsfoot and its fish hosts within the Ohio River basin include, but are not limited to, the Walhonding, Barren, Rough, and Eel Rivers and two rivers with viable populations, Green and Tippecanoe Rivers. The majority (7 of 11 populations or 64 percent) of viable rabbitsfoot populations (Ohio, Green, Tippecanoe, Tennessee, Duck, White, and Little Rivers) occur downstream of main stem impoundments that make these populations more susceptible to altered habitat quality and quantity associated with the impoundment or dam operation, which may be exacerbated during stochastic events such as droughts and floods.

Navigational improvements on the Ohio River began in 1830, and now include 21 lock and dam structures stretching from Pittsburgh, Pennsylvania, to Olmsted, Illinois, near its confluence with the Mississippi River. Lock and dam structures convert riverine habitat to unsuitable static habitat for the mussel and prevent movement of their fish hosts. Numerous Ohio River tributaries also have been altered by lock and dam structures. For example, a 116-rkm (72-rmi) stretch of the Allegheny River in Pennsylvania has been altered with nine locks and dams from Armstrong County to Pittsburgh. A series of six locks and dams were constructed on the lower half of the Green River decades ago that extend upstream to the western boundary of Mammoth Cave National Park, Kentucky. The declines of rabbitsfoot populations are attributable to navigational locks and dams on the Ohio, Allegheny, Monongahela, Muskingum, Kentucky, Green, Barren, and White Rivers, and are widespread throughout the species range.

Impoundments have eliminated a large portion of the Neosho mucket population and habitat in the Arkansas River basin. For example, mussel habitat in the Neosho River in Kansas has been adversely affected by at least 15 city dams and 2 Federal dams, both with regulated flows. Almost the entire length of the river in Oklahoma is now impounded or adversely affected by tail water releases from three major dams (Matthews

et al.

2005, p. 308). Several reservoirs and numerous small watershed lakes have eliminated suitable mussel habitat in several larger Neosho River tributaries in Kansas and Missouri (Spring, Elk and Cottonwood Rivers and Shoal Creek). The Verdigris River (Kansas and Oklahoma) has two large reservoirs with regulated flows, and the lower section has been channelized as part of the McClellan-Kerr Arkansas River Navigation System. All the major Verdigris River tributaries in Kansas and Oklahoma have been partially inundated by reservoirs with regulated flows and numerous flood control watershed lakes (Obermeyer

et al.

1995, pp. 7-21). Construction of Lake Tenkiller eliminated Neosho mucket populations and habitat in the lower portion of the Illinois River, Oklahoma (Davidson 2011, pers. comm.).

Dam construction has a secondary effect of fragmenting the ranges of mussel species by leaving relict habitats and populations isolated upstream or between structures as well as creating extensive areas of deep uninhabitable, impounded waters. These isolated populations are unable to naturally recolonize suitable habitat downstream and become more prone to further extirpation from stochastic events, such as severe drought, chemical spills, or unauthorized discharges (Layzer

et al.

1993, pp. 68-69; Cope

et al.

1997, pp. 235-237; Neves

et al.

1997, pp. 63-75; Watters 2000, pp. 264-265, 268; Miller and Payne 2001, pp. 14-15; Pringle

et al.

2000, pp. 810-815; Watters and Flaute 2010, pp. 3-7). We conclude that habitat effects due to impoundment are a significant and ongoing threat to the Neosho mucket and rabbitsfoot.

Channelization

Dredging and channelization activities have profoundly altered riverine habitats nationwide. Hartfield (1993, pp. 131-139), Neves

et al.

(1997, pp. 71-72), and Watters (2000, pp. 268-269) reviewed the specific upstream and downstream effects of channelization on freshwater mussels. Channelization affects a stream physically (accelerates erosion, increases sediment bed load, reduces water depth, decreases habitat diversity, creates geomorphic (natural channel dimensions) instability, eliminates riparian canopy) and biologically (decreases fish and mussel diversity, changes species composition and abundance, decreases biomass, and reduces growth rates) (Hartfield 1993, pp. 131-139). Channel modification for navigation has been shown to increase flood heights (Belt 1975, p. 684), partly as a result of an increase in stream bed slope (Hubbard

et al.

1993, p. 137). Flood events are exacerbated, conveying large quantities of sediment, potentially

with adsorbed contaminants, into streams. Channel maintenance often results in increased turbidity and sedimentation that often smothers mussels (Stansbery 1970, p. 10).

Channel maintenance operations for commercial navigation have affected habitat for the rabbitsfoot in many large rivers rangewide. Periodic navigation maintenance activities (such as dredging and snag removal) may continue to adversely affect this species in the lower portions of the Ohio, Tennessee, and White Rivers, which represent 44 percent of the viable rabbitsfoot populations. In the Tennessee River, a plan to deepen the navigation channel has been proposed (Hubbs 2009, pers. comm.). Some rabbitsfoot streams were “straightened” to decrease distances traversed by barge traffic (for example, Verdigris River). Hundreds of miles of many midwestern (Eel, North Fork Vermilion, and Embarras Rivers) and southeastern (Paint Rock and St. Francis Rivers and Bear Creek) streams with rabbitsfoot populations were channelized decades ago to reduce the probability and frequency of flood events. Because mussels are relatively immobile they require a stable substrate to survive and reproduce and are particularly susceptible to channel instability (Neves

et al.

1997, p. 23) and alteration. Channel and bank degradation have led to the loss of stable substrates in numerous rivers with commercial navigation throughout the range of rabbitsfoot. While dredging and channelization have had a greater effect on rabbitsfoot, the Neosho mucket has been affected by these activities in the Verdigris River. We conclude that habitat effects due to channelization are a significant and ongoing threat to the Neosho mucket and rabbitsfoot.

Sedimentation

Excessive sediments are believed to adversely affect riverine mussel populations requiring clean, stable streams (Ellis 1936, pp. 39-40; Brim Box and Mossa 1999, p. 99). Adverse effects resulting from sediments have been noted for many components of aquatic communities. Potential sediment sources within a watershed include virtually all activities that disturb the land surface. Most localities occupied by the Neosho mucket and rabbitsfoot, including viable populations, are currently being affected to varying degrees by sedimentation.

Sedimentation has been implicated in the decline of mussel populations nationwide, and remains a threat to Neosho mucket and rabbitsfoot (Ellis 1936, pp. 39-40; Vannote and Minshall 1982, pp. 4105-4106; Dennis 1984, p. 212; Brim Box and Mosa 1999, p. 99; Fraley and Ahlstedt 2000, pp. 193-194; Poole and Downing 2004, pp. 119-122). Specific biological effects include reduced feeding and respiratory efficiency from clogged gills, disrupted metabolic processes, reduced growth rates, limited burrowing activity, physical smothering, and disrupted host fish attraction mechanisms (Ellis 1936, pp. 39-40; Marking and Bills 1979, p. 210; Vannote and Minshall 1982, pp. 4105-4106; Waters 1995, pp. 173-175; Hartfield and Hartfield 1996, p. 373). In addition, mussels may be indirectly affected if high turbidity levels significantly reduce the amount of light available for photosynthesis, and thus, the production of certain food items (Kanehl and Lyons 1992, p. 7).

Studies tend to indicate that the primary effects of excess sediment levels on mussels are sublethal, with detrimental effects not immediately apparent (Brim Box and Mossa 1999, p. 101). The physical effects of sediment on mussel habitat appear to be multifold, and include changes in suspended and bed material load; bed sediment composition associated with increased sediment production and runoff in the watershed; channel changes in form, position, and degree of stability; changes in depth or the width and depth ratio that affects light penetration and flow regime; actively aggrading (filling) or degrading (scouring) channels; and changes in channel position. These effects to habitat may dislodge, transport downstream, or leave mussels stranded (Vannote and Minshall 1982, p. 4106; Kanehl and Lyons 1992, pp. 4-5; Brim Box and Mossa 1999, pp. 109-112). For example, many Kansas streams (such as Verdigris and Neosho Rivers) supporting mussels have become increasingly silted in over the past century, reducing habitat for the Neosho mucket and rabbitsfoot (Obermeyer

et al.

1997a, pp. 113-114).

Increased sedimentation and siltation may explain in part why Neosho mucket and rabbitsfoot are experiencing recruitment failure in some streams. Interstitial spaces in the substrate provide crucial habitat (shelter and nutrient uptake) for juvenile mussel survival. When interstitial spaces are clogged, interstitial flow rates and spaces are reduced (Brim Box and Mossa 1999, p. 100), and this decreases habitat for juvenile mussels. Furthermore, sediment may act as a vector for delivering contaminants, such as nutrients and pesticides, to streams, and juvenile mussels may ingest contaminants adsorbed to silt particles during normal feeding activities. Neosho mucket and rabbitsfoot reproductive strategies depend on clear water (enables fish hosts to see mussel lures) during critical reproductive periods.

Agricultural activities are responsible for much of the sediment affecting rivers in the United States (Waters 1995, p. 170). Sedimentation associated with agricultural land use is cited as one of the primary threats to 7 of the 11 (64 percent) viable rabbitsfoot populations (French Creek, Tippecanoe, Paint Rock, Duck, White, Black, and Strawberry Rivers; Smith

et al.

2009, Table 1; USACE 2011, pp. 21-22; Indiana Department of Environmental Management (IDEM) 2001, pp. 11-12; EPA 2001, p. 10; Brueggen 2010, pp. 1-2; MDC 2012,

http://mdc.mo.gov/landwater-care/stream-and-watershed-management/;

EPA Water Quality Assessment Tool,

http://ofmpub.epa.gov/tmdl_waters10/attains_nation_cy.control?p_report_type=T

). In addition, numerous stream segments in the Duck, White, Black, Little, and Strawberry River watersheds are listed as impaired waters under section 303(d) of the Clean Water Act (CWA) by EPA due to sedimentation associated with agriculture (USACE 2011, p. 21; EPA Water Quality Assessment Tool,

http://ofmpub.epa.gov/tmdl_waters10/attains_nation_cy.control?p_report_type=T

). An impaired water is a water body (i.e., stream reaches, lakes, water body segments) with chronic or recurring monitored violations of the applicable numeric or narrative water quality criteria. An impaired water cannot support one or more of its designated uses (e.g., swimming, the protection and propagation of aquatic life, drinking, industrial supply, etc.). Once a stream segment is listed as an impaired water, the State must complete a plan to address the issue causing the impairment; this plan is called a Total Maximum Daily Load (TMDL). A TMDL is a calculation of the maximum amount of a pollutant that a water body can receive and still safely meet water quality standards (WQS). Completion of the plan is generally all that is required to remove the stream segment from the 303(d) impaired water list and does not mean that water quality has changed. Once the TMDL is completed, the stream segment may be placed on the 305(b) list of impaired streams with a completed TMDL (

http://water.epa.gov/lawsregs/lawsguidance/cwa/tmdl/intro.cfm

). For example, some stream segments within the White, Barren, Little River Mountain Fork, and Wabash Rivers, and French Creek have completed TMDL plans and have

attained WQS for low dissolved oxygen, pathogens, nutrients, polychlorinated biphenyls (PCBs), and siltation. However, some of these same stream segments still have not attained WQS for lead (Little River Mountain Fork) and mercury (Wabash River).

Impaired streams in the Duck River watershed (approximately 483 rkm (300 rmi)) are losing 5 to 55 percent more soil per year than the natural streams (USACE 2011, pp. 21-22). Unrestricted livestock access occurs on many streams and potentially threatens associated mussel populations (Fraley and Ahlstedt 2000, pp. 193-194). Grazing may reduce water infiltration rates and increase runoff; trampling and vegetation removal increases the probability of erosion (Armour

et al.

1991, pp. 8-10; Brim Box and Mossa 1999, p. 103).

As discussed above, specific impacts on mussels from sediments include reduced feeding and respiratory efficiency, disrupted metabolic processes, reduced growth rates, increased substrata instability, and the physical smothering of mussels. Increased turbidity levels due to siltation can be a limiting factor that impedes the ability of sight-feeding fishes to forage. Turbidity within the rivers and streams during the times that the mussels attempt to attract host fishes may have contributed and may continue to contribute to the decline of the Neosho mucket and rabbitsfoot by reducing their efficiency at attracting the fish hosts necessary for reproduction. In addition, sediment can eliminate or reduce the recruitment of juvenile mussels, interfere with feeding activity, and act as a vector in delivering contaminants to streams. Because the Neosho mucket and rabbitsfoot are filter-feeders and may bury themselves in the substrate, they are exposed to these contaminants contained within suspended particles and deposited in bottom substrates. We conclude that biological and habitat effects due to sedimentation are a significant and ongoing threat to the Neosho mucket and rabbitsfoot.

Chemical Contaminants

Chemical contaminants are ubiquitous in the environment and are considered a major threat in the decline of mussel species (Richter

et al.

1997, p. 1081; Strayer

et al.

2004, p. 436; Wang

et al.

2007a, p. 2029; Cope

et al.

2008, p. 451). Chemicals enter the environment through point and nonpoint discharges including spills, industrial and municipal effluents, and residential and agricultural runoff. These sources contribute organic compounds, heavy metals, nutrients, pesticides, and a wide variety of newly emerging contaminants such as pharmaceuticals to the aquatic environment. As a result, water and sediment quality can be degraded to the extent that results in adverse effects to mussel populations.

Cope

et al.

(2008, p. 451) evaluated the pathways of exposure to environmental pollutants for all four freshwater mollusk life stages (free glochidia, encysted glochidia, juveniles, adults) and found that each life stage has both common and unique characteristics that contribute to observed differences in exposure and sensitivity. Almost nothing is known of the potential mechanisms and consequences of waterborne toxicants on sperm viability. In the female mollusk, the marsupial region of the gill is thought to be physiologically isolated from respiratory functions, and this isolation may provide some level of protection from contaminant interference with a female's ability to achieve fertilization or brood glochidia (Cope

et al.

2008, p. 454). A major exception to this assertion is with chemicals that act directly on the neuroendocrine pathways controlling reproduction (see discussion below). Nutritional and ionic exchange is possible between a brooding female and her glochidia, providing a route for chemicals (accumulated or waterborne) to disrupt biochemical and physiological pathways (such as maternal calcium transport for construction of the glochidial shell). Glochidia can be exposed to waterborne contaminants for up to 36 hours until encystment occurs; between 2 and 36 hours, and then from fish host tissue burdens (for example, atrazine), that last from weeks to months and could affect transformation success of glochidia into juveniles (Ingersoll

et al.

2007, pp. 101-104).

Juvenile mussels typically remain burrowed beneath the sediment surface for 2 to 4 years. Residence beneath the sediment surface necessitates deposit (pedal) feeding and a reliance on interstitial water for dissolved oxygen (Watters 2007, p. 56). The relative importance of exposure of juvenile Neosho mucket and rabbitsfoot to contaminants in overlying surface water, interstitial water, whole sediment, or food has not been adequately assessed. Exposure to contaminants from each of these routes varies with certain periods and environmental conditions (Cope

et al.

2008, pp. 453 and 457).

The primary routes of exposure to contaminants for adult Neosho mucket and rabbitsfoot are surface water, sediment, interstitial (pore) water, and diet; adults can be exposed when either partially or completely burrowed in the substrate (Cope

et al.

2008, p. 453). Adult mussels have the ability to detect toxicants in the water and close their valves to avoid exposure (Van Hassel and Farris 2007, p. 6). Adult mussel toxicity and relative sensitivity (exposure and uptake of toxicants) may be reduced at high rather than at low toxicant concentrations because uptake is affected by the prolonged or periodic toxicant avoidance responses (when the avoidance behavior of keeping their valves closed can no longer be sustained for physiological reasons (respiration and ability to feed) (Cope

et al.

2008, p. 454). Toxicity results based on low-level exposure of adults are similar to estimates for glochidia and juveniles for some toxicants (for example, copper). The duration of any toxicant avoidance response by an adult mussel is likely to vary due to several variables, such as species, age, shell thickness and gape, properties of the toxicant, and water temperature. There is a lack of information on toxicant response(s) for Neosho mucket and rabbitsfoot, but results of tests using glochidia and juveniles may be valuable for protecting adults (Cope

et al.

2008, p. 454).

Mussels are very intolerant of heavy metals (such as lead, zinc, cadmium, and copper) compared to commonly tested aquatic organisms. Metals occur in industrial and wastewater effluents and are often a result of atmospheric deposition from industrial processes and incinerators, but also are associated with mine water runoff (for example, Tri-State Mining Area in southwest Missouri) and have been attributed to mussel declines in streams such as Shoal, Center, and Turkey Creeks and Spring River in the Arkansas River basin (Angelo

et al.

2007, pp. 485-489), which are streams with historical and extant Neosho mucket and rabbitsfoot populations. Heavy metals can cause mortality and affect biological processes, for instance, disrupting enzyme efficiency, altering filtration rates, reducing growth, and changing behavior of freshwater mussels (Keller and Zam 1991, p. 543; Naimo 1995, pp. 351-355; Jacobson

et al.

1997, p. 2390; Valenti

et al.

2005, p. 1244; Wang

et al.

2007b, pp. 2039-2046; Wang

et al.

2007c, pp. 2052-2055; Wang

et al.

2010, p. 2053). Mussel recruitment may be reduced in habitats with low but chronic heavy metal and other toxicant inputs (Yeager

et al.

1994, p. 217; Naimo 1995, pp. 347 and 351-352; Ahlstedt and Tuberville 1997, p. 75). Newly

transformed juveniles (age at 5 days) are more sensitive to acute toxicity than glochidia or older juveniles (age at 2 to 6 months) (Wang

et al.

2010, p. 2062).

Mercury is another heavy metal that has the potential to negatively affect mussel populations. Mercury has been detected throughout aquatic environments as a product of municipal and industrial waste and atmospheric deposition from coal-burning plants. One study on rainbow mussel (

Villosa iris

) concluded that glochidia were more sensitive to mercury than were juvenile mussels, with a median lethal concentration value of 14

g/L for glochidia and 114

g/L for juvenile mussels (Valenti

et al.

2005, p. 1242). The chronic toxicity is a test which usually measures sublethal effects (e.g., reduced growth or reproduction) in addition to lethality. These tests are usually longer in duration or conducted during some sensitive period of an organism's life cycle. For this species, the chronic toxicity test showed that juveniles exposed to mercury greater than or equal to 8

g/L exhibited reduced growth (Valenti

et al.

2005, p. 1245). Mercury also affects oxygen consumption, byssal thread production, and filtration rates (Naimo 1995, Jacobsen

et al.

1997, and Nelson and Calabrese 1988

Valenti

et al.

2005, p. 1245). Effects to mussels from mercury toxicity may be occurring in some streams due to illegal dumping, spills, and permit violations. For example, acute mercury toxicity was determined to be the cause of extirpation of diverse mussel fauna for a 112-rkm (70-rmi) reach of the North Fork Holston River (Brown

et al.

2005, pp. 1455-1457). Of the 11 viable rabbitsfoot populations, 4 populations (French Creek, Duck River, Green River, and Ohio River) currently inhabit river reaches that are impaired by mercury and are listed as impaired waters under section 303(d) of the CWA.

One chemical that is particularly toxic to early life stages of mussels is ammonia. Sources of ammonia include agricultural wastes (animal feedlots and nitrogenous fertilizers), municipal wastewater treatment plants, and industrial waste (Augspurger

et al.

2007, p. 2026) as well as precipitation and natural processes (decomposition of organic nitrogen) (Goudreau

et al.

1993, p. 212; Hickey and Martin 1999, p. 44; Augspurger

et al.

2003, p. 2569; Newton 2003, p. 1243). Therefore, ammonia is considered a limiting factor for survival and recovery of some mussel species due to its ubiquity in aquatic environments and high level of toxicity, and because the highest concentrations typically occur in mussel microhabitats (Augspurger

et al.

2003, p. 2574). In addition, studies have shown that ammonia concentrations increase with increasing temperature, pH, and low flow conditions (Cherry

et al.

2005, p. 378; Cooper

et al.

2005, p. 381; Wang

et al.

2007, p. 2045), which may be exacerbated by the effects of climate change, and may cause ammonia (un-ionized and ionized) to become more problematic for juvenile mussels (Wang

et al.

2007, p. 2045). Sublethal effects include, but may not be limited to, reduced time the valves are held open for respiration and feeding; impaired secretion of the byssal thread (used for substrate attachment), reduced ciliary action impairing feeding, depleted lipid, glycogen, and other carbohydrate stores, and altered metabolism (Goodreau

et al.

1993, pp. 216-227; Augspurger

et al.

2003, pp. 2571-2574; Mummert

et al.

2003, pp. 2548-2552).

Polychlorinated biphenyls (PCBs) are ubiquitous contaminants in the environment due to their widespread use from the 1920s to 1970s as insulating material in electric equipment, such as transformers and capacitors, as well as in heat transfer fluids and in lubricants. PCBs have also been used in a wide range of products, such as plasticizers, surface coatings, inks, adhesives, flame retardants, paints, and carbonless duplicating paper. PCBs were still being introduced into the environment at many sites (such as landfills and incinerators) until the 1990s. The inherent stability and toxicity of PCBs have resulted in them being a persistent environmental problem (Safe 1994

Lehmann

et al.

2007, p. 356). PCBs are lipophilic (affinity to combine with fats or lipids), adsorb easily to soil and sediment, and are present in the sediment and water column in aquatic environments, making them available to bioaccumulate and induce negative effects in living organisms (Livingstone 2001

Lehmann

et al.

2007, p. 356). Studies have demonstrated increased PCB concentrations in native freshwater mussels (Ruessler

et al.

2011, pp. 1, 7), marine bivalves (Krishnakumar

et al.

1994, p. 249), and nonnative, invasive mollusks (zebra mussels and Asian clams) (Gossiaux

et al.

1996, p. 379; Lehmann

et al.

2007, p. 363) in areas with high levels of PCBs. Oxidative stress (imbalance in the normal redox state of cells that causes toxic effects that damage all components of the cell, including proteins, lipids, and DNA) is a direct consequence of exposure to PCBs. Relevant changes, whether directly or indirectly due to oxidative stress, may occur at the organ and organism levels and will likely result in mussel population-wide effects, including reduced fecundity and chronic maladies due to PCB exposure (Lehmann

et al.

2007, p. 363). Two of the 11 viable rabbitsfoot populations (18 percent) inhabit waters listed as impaired due to PCBs under section 303(d) of the CWA.

Agriculture, timber harvest, and lawn management practices utilize nutrients and pesticides. These are two broad categories of chemical contaminants that have the potential to adversely impact mussel species. Nutrients, such as nitrogen and phosphorus, primarily occur in runoff from livestock farms, feedlots, heavily fertilized row crops and pastures (Peterjohn and Correll 1984, p. 1471), post timber management activities, and urban and suburban runoff, including leaking septic tanks, and residential lawns.

Studies have shown that excessive nitrogen concentrations can be lethal to the adult freshwater pearl mussel (

Margaritifera margaritifera

) and reduce the lifespan and size of other mussel species (Bauer 1988, p. 244; Bauer 1992, p. 425). Nutrient enrichment can result in an increase in primary productivity, and the associated algae respiration depletes dissolved oxygen levels. This may be particularly detrimental to juvenile mussels that inhabit the interstitial spaces in the substrate where lower dissolved oxygen concentrations are more likely than on the sediment surface where adults tend to live (Sparks and Strayer 1998, pp. 132-133). For example, Galbraith

et al.

(2008, pp. 48-49) reported a massive die-off of greater than 160 rabbitsfoot specimens at a long-term monitoring site in the Little River, Oklahoma. While the exact cause for the die-off is unknown, the authors speculate that the 2005 Oklahoma drought coupled with high water temperature and extensive blooms of filamentous algae may have resulted in extreme physiological stress. Over-enriched conditions are exacerbated by low flow conditions, such as those experienced during a typical summer season and that may occur with greater frequency and severity as a result of climate change. Three of the 11 viable rabbitsfoot populations (French Creek, Duck River, and Tippecanoe River) are listed as impaired waters under section 303(d) of the CWA due to nutrient enrichment.

Elevated concentrations of pesticide frequently occur in streams due to residential or commercial pesticide runoff, overspray application to row crops, and lack of adequate riparian buffers. Agricultural pesticide applications often coincide with the reproductive and early life stages of

mussels, and effects to mussels may be increased during a critical time period (Bringolf

et al.

2007a, p. 2094). Recent studies tested the toxicity of glyphosate, its formulations, and a surfactant (MON 0818) used in several glyphosate formulations, to early life stages of the fatmucket (

Lampsilis siliquoidea

), a U.S. native freshwater mussel (Bringolf

et al.

2007a, p. 2094). Studies conducted with juvenile mussels and glochidia determined that the surfactant (MON 0818) was the most toxic of the compounds tested and that

L. siliquoidea

glochidia were the most sensitive organism tested to date (Bringolf

et al.

2007a, p. 2094). Roundup®, technical grade glyphosate isopropylamine salt, and isopropylamine were also acutely toxic to juveniles and glochidia (Bringolf

et al.

2007a, p. 2097). The study of other pesticides, including atrazine, chlorpyrifos, and permethrin, on glochidia and juvenile life stages determined that chlorpyrifos was toxic to both

L. siliquoidea

glochidia and juveniles (Bringolf

et al.

2007b, pp. 2101 and 2104). The above results indicate the potential toxicity of commonly applied pesticides and the threat to mussel species as a result of the widespread use of these pesticides.

There are instances where chemical spills have resulted in the loss of high numbers of mussels (Jones

et al.

2001, p. 20; Brown

et al.

2005, p. 1457; Schmerfeld 2006, pp. 12-13), and are considered a serious threat to mussel species. The Neosho mucket and rabbitsfoot are especially threatened by chemical spills because these spills can occur anywhere that highways with tanker trucks, industries, or mines overlap with their distribution.

Other examples of the influence of point and nonpoint-source pollutants on streams throughout the range of the Neosho mucket and rabbitsfoot include two documented mussel kills in Fish Creek (circa 1988) as a result of manure runoff from a hog farm and a diesel spill (Watters 1988, p. 18). Twelve point-source discharges occur on the Green River (Kentucky State Nature Preserves Commission and The Nature Conservancy 1998, pp. 15-19). The Illinois River, a tributary of the Arkansas River, is subject to nonpoint-source organic runoff from poultry farming and municipal wastewater.

Pharmaceutical chemicals used in commonly consumed drugs are increasingly found in surface waters. A recent nationwide study sampling 139 stream sites in 30 States detected the presence of numerous pharmaceuticals, hormones, and other organic wastewater contaminants downstream from urban development and livestock production areas (Kolpin

et al.

2002, pp. 1208-1210). Another study in northwestern Arkansas found pharmaceuticals or other organic wastewater constituents at 16 of 17 sites in seven streams surveyed in 2004 (Galloway

et al.

2005, pp. 4-22). Toxic levels of exposure to chemicals that act directly on the neuroendocrine pathways controlling reproduction can cause premature release of viable or nonviable glochidia. For example, the active ingredient in many human prescription antidepressant drugs belonging to the class of selective serotonin reuptake inhibitors may exert negative reproductive effects on mussels because of the drug's action on serotonin and other neuroendocrine pathways (Cope

et al.

2008, p. 455). Pharmaceuticals or organic wastewater constituents are generally greater downstream of wastewater treatment facilities (Galloway

et al.

2005, p. 28). Pharmaceuticals that alter mussel behavior and influence successful attachment of glochidia on fish hosts may have population-level implications for the Neosho mucket and rabbitsfoot.

The information presented in this section represents some of the threats from chemical contaminants that have been documented both in the laboratory and field and demonstrates that chemical contaminants pose a substantial threat to Neosho mucket and rabbitsfoot. A cursory examination of land use trends, non-point and point source discharges, and the list of impaired waters under section 303(d) of the CWA suggests that all 11 rabbitsfoot populations currently considered viable may be subjected to the subtle, pervasive effects of chronic, low-level contamination that is ubiquitous in these watersheds. For example, 8 of the 11 (73 percent) streams with viable rabbitsfoot populations are listed as impaired waters under section 303(d) of the CWA. Reasons for impairment include mercury, nutrients, organic enrichment and dissolved oxygen depletion, pathogens, turbidity (sediment), and PCBs. Potential effects from contaminant exposure may result in death, reduced growth, altered metabolic processes, or reduced reproduction. We conclude that biological and habitat effects due to chemical contaminants are a significant and ongoing threat contributing to the decline of Neosho mucket and rabbitsfoot populations.

Mining

Gravel, coal, and metal mining are activities negatively affecting water quality in Neosho mucket and rabbitsfoot habitat. Instream and alluvial gravel mining has been implicated in the destruction of mussel populations (Hartfield 1993, pp. 136-138; Brim Box and Mossa 1999, pp. 103-104). Negative effects associated with gravel mining include stream channel modifications (altered habitat, disrupted flow patterns, sediment transport), water quality modifications (increased turbidity, reduced light penetration, increased temperature), macroinvertebrate population changes (elimination), and changes in fish populations, resulting from adverse effects to spawning and nursery habitat and food web disruptions (Kanehl and Lyons 1992, pp. 4-10). Gravel mining activities continue to be a localized threat in several streams with viable rabbitsfoot populations (Ohio, Tennessee, White, Strawberry, and Little Rivers). In the lower Tennessee River, instream mining occurs in 18 reaches totaling 77.1 rkm (47.9 rmi) between the Duck River confluence and Pickwick Landing Dam (Hubbs 2010, pers. comm.).

Coal mining activities, resulting in heavy metal-rich drainage, and associated sedimentation has adversely affected many drainages with rabbitsfoot populations, including portions of the upper Ohio River system in Kentucky, Pennsylvania, and West Virginia; the lower Ohio River system in eastern Illinois; the Rough River drainage in western Kentucky; and the upper Cumberland River system in Kentucky and Tennessee (Ortmann 1909

Butler 2005, p. 102; Gordon 1991, pp. 4 and 5; Layzer and Anderson 1992

Butler 2005, p. 102). Numerous mussel toxicants, such as polycyclic aromatic hydrocarbons and heavy metals (copper, manganese, and zinc) from coal mining contaminate sediments when released into streams (Ahlstedt and Tuberville 1997, p. 75). Low pH commonly associated with mine runoff can reduce glochidial attachment rates on host fish (Huebner and Pynnonen 1990, pp. 2350-2353). Thus, acid mine runoff may have local effects on mussel recruitment and may lead to mortality due to improper shell development or erosion.

Metal mining (lead, cadmium, and zinc) in the Tri-State Mining Area (15,000 km

; 5,800 mi

in Kansas, Missouri, and Oklahoma) has adversely affected Center and Shoal Creeks and the Spring River. It has been implicated in the loss of Neosho mucket and rabbitsfoot from portions of these streams (Obermeyer

et al.

1997b, p. 114). A study by Kansas Department of Health and Environment documented a strong negative correlation between the distribution and abundance of native mussels, including Neosho mucket, and sediment concentrations of lead, zinc

and cadmium in the Spring River system (Angelo

et al.

2007, pp. 477-493). Sediment and water quality samples exceeded EPA 2006 threshold effect concentrations for cadmium, lead, and zinc at numerous sampling locations within the Tri-State Mining Area (Gunter 2007, pers. comm.). These physical habitat threats combined with poor water quality and agricultural nonpoint-source pollution are serious threats to all existing mussel fauna in the basin.

In the St. Francis River basin, past metal mining and smelting (early eighteenth century through the 1940s) have resulted in continuing heavy metal (lead, iron, nickel, copper, cobalt, zinc, cadmium, chromium) contamination of surface waters in the area upstream of the extant rabbitsfoot population. Recent and historical metals mining and smelting produced large volumes of contaminated wastes. Most of these mining wastes are stored behind poorly constructed dams and impoundments (Roberts 2008, pers. comm.). Wappapello Reservoir and the confluence with Big Creek (with habitat degradation primarily from mining activities) may effectively limit the distribution of the rabbitsfoot in the St. Francis River. We conclude that biological and habitat effects due to mining activities are a significant and ongoing threat contributing to declining Neosho mucket and rabbitsfoot populations.

Oil and Natural Gas Development

Oil and natural gas resources are present in some of the watersheds that are known to support rabbitsfoot, including the Allegheny and Middle Fork Little Red Rivers and two watersheds with viable populations (White River and French Creek). Exploration and extraction of these energy resources can result in increased siltation, a changed hydrograph (graph showing changes in the discharge of a river over a period of time), and altered water quantity and quality even at considerable distances from the mine or well field because effects are carried downstream from the original source. Rabbitsfoot habitat in streams can be threatened by the cumulative effects of multiple mines and well fields (adapted from Service 2008, p. 11).

Recently, oil and gas exploration has been able to expand in areas of shale due to new technologies (i.e., hydraulic fracturing and horizontal drilling), making access possible to oil and gas reserves in areas that were previously inaccessible. Extraction of these resources, particularly natural gas, has increased dramatically in recent years in Arkansas, Oklahoma, Pennsylvania, and West Virginia. Although oil and natural gas extraction generally occurs away from the river, extensive road and pipeline networks are required to construct and maintain wells and transport the extracted resources. These road and pipeline networks frequently cross or occur near tributaries, contributing sediment to the receiving waterway. In addition, the construction and operation of wells may result in the discharge of chemical contaminants and subsurface minerals. Several of the viable rabbitsfoot populations occur in active shale basins (areas of shale gas formations) (

http://www.eia.gov/analysis/studies/worldshalegas/

). In 2006, more than 3,700 permits were issued for oil and gas wells by the Pennsylvania Department of Environmental Protection, which also issued 98 citations for permit violations at 54 wells (Hopey 2007; adapted from Service 2008, p. 13). A natural gas pipeline company pled guilty to three violations of the Act in 2011 for unauthorized take of a federally endangered mussel in Arkansas as a result of a large amount of sediment being transported from pipeline right-of-ways to tributary streams in the affected watershed (Department of Justice 2011, pers. comm.). Where oil and natural gas development occurs within the range of extant Neosho mucket and rabbitsfoot populations, we conclude that the resulting biological and habitat effects are a significant and ongoing threat contributing to the decline of both species.

Summary of Factor A

The decline of mussels in the eastern United States is primarily the result of long-lasting direct and secondary effects of habitat alterations such as impoundments, channelization, sedimentation, chemical contaminants, oil and gas development, and mining and it is reasonable to conclude that the changes in the river basins historically and currently occupied by the species are the cause of population level (river basin) effects. Historical population losses due to impoundments have probably contributed more to the decline and range reductions of the Neosho mucket and rabbitsfoot than any other single factor. Seven of the 11 (64 percent) viable rabbitsfoot populations (Ohio, Green, Tippecanoe, Tennessee, Duck, White, and Little Rivers) occur downstream of main stem impoundments that make these populations more susceptible to altered habitat quality and quantity associated with the impoundment and dam operation, which may be exacerbated during stochastic events such as droughts and floods. Sedimentation resulting from a variety of sources such as channelization, agricultural and silvicultural practices, and construction activities has degraded Neosho mucket and rabbitsfoot habitat and altered biological processes essential to their survival. For example, sedimentation associated with agricultural land use is cited as one of the primary threats to 7 of the 11 (64 percent) streams with viable rabbitsfoot populations. Land use conversion, particularly urbanization that increases impervious surfaces in watersheds (impervious surface increases flood intensity and duration), channelization, and instream gravel and sand mining alter natural hydrology and stream geomorphology characteristics that also degrade mussel habitat in streams that support the Neosho mucket and rabbitsfoot. Contaminants associated with industrial and municipal effluents, agricultural practices, and mining degrade water and sediment quality leading to environmental conditions that have lethal and sublethal effects to Neosho mucket and rabbitsfoot, particularly the highly sensitive early life stages. Eight of the 11 (73 percent) streams with viable rabbitsfoot populations are listed as impaired waters under section 303(d) of the CWA, which means that the rabbitsfoot may be subjected to the subtle, pervasive effects of chronic, low-level contamination that is ubiquitous in these watersheds. Chronic contamination can affect the mussels in a variety of ways including sublethal effects (such as suppressed immune systems and effects to reproduction and fecundity from neuroendocrine disrupters) and lethal effects (such as sediment smothers and disruption of other metabolic processes).

In summary, we have determined that impoundments, channelization, sedimentation, chemical contaminants, mining, and oil and natural gas development are significant, ongoing threats to the Neosho mucket and rabbitsfoot that are expected to continue into the future. Although efforts have been made to restore habitat in some areas, these threats are still ongoing, as evidenced by population declines and range reduction. Thus, these changes in the species' historical or current range are not expected to be ameliorated in the future; therefore, we find it reasonably likely that the effects of these threats on both species will continue at current levels or potentially increase.

B. Overutilization for Commercial, Recreational, Scientific, or Educational Purposes

The Neosho mucket was valuable in the pearl button industry (1800s to early 1940s), and historical episodes of overharvest in the Neosho River may have contributed to its decline (Obermeyer

et al.

1997b, p. 115). The rabbitsfoot was never a valuable shell for the commercial pearl button industry (Meek and Clark 1912, p. 15; Murray and Leonard 1962, p. 65), nor the cultured pearl industry (Williams and Schuster 1989, p. 23), and hence these activities were probably not significant factors in its decline. However, it was noted occasionally in commercial harvests as evidenced from mussel cull piles (Isely 1924; Parmalee

et al.

1980, p. 101). Currently, Neosho mucket and rabbitsfoot are not commercially valuable species but may be increasingly sought by collectors as they become rarer. Although scientific collecting is not thought to represent a significant threat, unregulated collecting could adversely affect localized Neosho mucket and rabbitsfoot populations.

Commercial mussel harvest is illegal in some States (for example, Indiana and Ohio), but regulated in others (for example, Arkansas, Alabama, Kentucky, and Tennessee). These species may be inadvertently harvested by inexperienced commercial harvesters unfamiliar with species identification. Although illegal harvest of protected mussel beds occurs (Watters and Dunn 1995, pp. 225 and 247-250), commercial harvest is not known to have a significant effect on the Neosho mucket and rabbitsfoot.

Summary of Factor B

Though it is possible that the intensity of inadvertent or illegal harvest may increase in the future, there is no evidence that this stressor is currently increasing in severity. On the basis of this analysis, we find that overutilization for commercial, recreational, scientific, or educational purposes is not a current threat to the Neosho mucket or rabbitsfoot in any portion of their range at this time nor is likely to become so in the future.

C. Disease or Predation

Little is known about diseases in freshwater mussels (Grizzle and Brunner 2007, p. 6). However, mussel die-offs have been documented in streams inhabited by rabbitsfoot (Neves 1986, pp. 8-11), and some researchers believe that disease may be a factor contributing to the die-offs (Buchanan 1986, p. 53; Neves 1986, p. 11). Mussel parasites include water mites, trematodes, oligochaetes, leeches, copepods, bacteria, and protozoa (Grizzle and Brunner 2007, p. 4). Generally, parasites are not suspected of being a major limiting factor in the species' survival (Oesch 1984, p. 6). However, mite and trematode burdens can affect reproductive output and physiological condition, respectively, in mussels (Gangloff

et al.

2008, pp. 28-30). Stressors that reduce fitness may make mussels more susceptible to parasites (Butler 2007, p. 90). Furthermore, nonnative mussels may carry diseases and parasites that are potentially devastating to the native mussel fauna on an individual or population level basis (river basin), including Neosho mucket and rabbitsfoot (Strayer 1999b, p. 88). However, while individual mussels or beds of mussels historically or currently may have been affected by disease or parasites, we have no evidence that the severity of disease or parasite infestations impact either mussel on a population level (river basin).

The muskrat (

Ondatra zibethicus

) is cited as the most prevalent mussel predator (Kunz 1898, p. 328; Convey

et al.

1989, p. 654-655; Hanson

et al.

1989, pp. 15-16). Muskrat predation may limit the recovery potential of endangered or threatened mussels or contribute to local extirpations of previously stressed populations, according to Neves and Odom (1989, p. 940), who consider it, however, primarily a seasonal or localized threat. Galbraith

et al.

(2008, p. 49) hypothesized that predation may have exacerbated rabbitsfoot mortality in the Little River, Oklahoma, during the 2005 drought. Harris

et al.

(2007, p. 31) reported numerous dead rabbitsfoot from muskrat middens (mound or deposit containing shells) in the Spring River, Arkansas. Other mammals (for example, raccoon, mink, otter, hogs, and rats), turtles, and aquatic birds also occasionally feed on mussels (Kunz 1898, p. 328; Neck 1986, pp. 64-65). Recently, predation of Neosho mucket by reintroduced otters has been documented in a mussel bed also supporting rabbitsfoot in the Spring River, Kansas (Barnhart 2003, pp. 16-17), and likely occurs elsewhere. Muskrat predation has been documented for Neosho mucket and rabbitsfoot, but the overall threat is generally considered insignificant.

Some species of fish feed on mussels (for example, common carp (

Cyprinus carpio),

freshwater drum (

Aplodin

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.