Browse: Departments   Dates   Agencies  

CFR Citation: 50 CFR Part 17

RIN ID: RIN 1018-AI47

NOTICE: Part II

DOCUMENT ACTION: Final rule.

SUBJECT CATEGORY: Endangered and Threatened Wildlife and Plants; Designation of Critical Habitat for Seven Bexar County, TX, Invertebrate Species

DATES: This rule becomes effective on May 8, 2003.

DOCUMENT SUMMARY: We, the U.S. Fish and Wildlife Service (Service), designate critical habitat for seven endangered invertebrate species found in Bexar County, Texas, pursuant to the Endangered Species Act of 1973, as amended (Act). The critical habitat designation totals approximately 431 hectares (1,063 acres) in 22 units. Section 7 of the Act requires Federal agencies to ensure, in consultation with the Service, that actions they authorize, fund, or carry out are not likely to result in the destruction or adverse modification of critical habitat. Section 4 of the Act requires us to consider economic and other impacts when specifying any particular area as critical habitat. We solicited data and comments from the public on all aspects of the proposed rule, including data on economic and other impacts of the designation. As a result of comments and information received, we are not designating critical habitat as originally proposed for two species that occur entirely on Stateowned lands that are subject to a conservation plan.

SUMMARY: Interior Department, Fish and Wildlife Service,

SUPPLEMENTAL INFORMATION

The seven species for which we are designating critical habitat in this rulemaking inhabit caves or other features known as karst. The term ``karst'' refers to a type of terrain that is formed by the slow dissolution of calcium carbonate from limestone bedrock by mildly acidic groundwater. This process creates numerous cave openings, cracks, fissures, fractures, and sinkholes, and the bedrock resembles a honeycomb.

As a result of climatic changes beginning two million years ago and lasting until ten thousand years ago, invertebrate species colonized caves and other subterranean voids (Barr 1968; Mitchell and Reddell 1971; Elliott and Reddell 1989). Species that dwell exclusively in caves and other subterranean voids are referred to as ``troglobites.'' Through faulting and canyon downcutting, the karst terrain colonized by these species along the Balcones Fault Zone (a zone approximately 25 kilometers (km) in width, extending from the northeast corner of Bexar County to the western edge of the County) became increasingly dissected, creating ``islands'' of karst and barriers to dispersal. These ``islands'' isolated troglobitic populations from each other, probably resulting in further speciation.

The following nine Bexar County, Texas, troglobitic invertebrate species were listed as endangered on December 26, 2000 (65 FR 81419): spider (no common name) (Cicurina venii), Robber Baron Cave harvestman (Texella cokendolpheri), vesper cave spider (Cicurina vespera), Government Canyon cave spider (Neoleptoneta microps), Madla's cave spider (Cicurina madla), Robber Baron cave spider (Cicurina baronia), beetle (no common name) (Rhadine exilis), beetle (no common name) (Rhadine infernalis), and Helotes mold beetle (Batrisodes venyivi). These are karst dwelling species of local distribution in north and northwest Bexar County. They spend their entire lives underground.

Since publication of the listing final rule, the common names for the following six arachnid species have been changed as a result of a meeting of the Committee on Common Names of Arachnids of the American Arachnological Society in 2000. Accordingly, we are changing the common names of the species currently in the list of Endangered and Threatened Wildlife (50 CFR 17.11) as Robber Baron Cave harvestman, Robber Baron cave spider, Madla's cave spider, vesper cave spider, Government Canyon cave spider, and one with no common name (Cicurina venii) to Cokendolpher cave harvestman, Robber Baron Cave meshweaver, Madla Cave meshweaver, Government Canyon Bat Cave meshweaver, Government Canyon Bat Cave spider, and Braken Bat Cave meshweaver, respectively.

Individuals of the listed species are small, ranging in length from 1 millimeter (0.039 inch (in)) to 1 centimeter (0.39 in). They are eyeless, or essentially eyeless, and most lack pigment. Low quantities of food in caves have caused adaptations in these species, including low metabolism, long legs for efficient movement, and loss of eyes, possibly as an energysaving tradeoff (Howarth 1983). Survival may be possible from months to years with little or no food (Howarth 1983). Adult Cicurina spiders have survived in captivity without food for about 4 months (James Cokendolpher, Museum of Texas Tech University, pers. comm. 2002).

Although little is known about the life history of listed Texas troglobitic invertebrates, they are believed to live for longer than 1 year. This belief is based, in part, on the amount of time some juveniles have been kept in captivity without maturing (Veni and Associates 1999; James Reddell, Texas Memorial Museum, pers. comm. 2000). For example, James Cokendolpher (Museum of Texas Tech University, pers. comm. 2002) maintained a juvenile troglobitic Cicurina spider from May 1999 through April 2002. Reproductive rates of troglobites are typically low (Poulson and White 1969; Howarth 1983). According to surveys conducted by Culver (1986), Elliott (1994a), and Hopper (2000), population sizes of troglobitic invertebrates are typically small, with most species known from only a few specimens (Culver et al. 2000).

As described below, the primary habitat requirements of these species include: (1) Subterranean spaces in karst with stable temperatures, high humidities (near saturation), and suitable substrates (for example, spaces between and underneath rocks suitable for foraging and sheltering); and (2) a healthy surface community of native plants and animals that provide nutrient input and, in the case of native plants, act to buffer the karst ecosystem from adverse effects (for example, invasions of nonnative species, contaminants, and fluctuations in temperature and humidity). These karst invertebrates require stable temperatures and constant, high humidity (Barr 1968; Mitchell 1971a) because they are vulnerable to desiccation in drier habitats (Howarth 1983) or cannot detect or cope with more extreme temperatures (Mitchell 1971a). Temperatures in caves typically remain at the average annual surface temperature, with little variation [[Page 17157]]
(Howarth 1983; Dunlap 1995). Relative humidity is typically near 100 percent in caves that support troglobitic invertebrates (Elliott and Reddell 1989). During temperature extremes, the listed species may retreat into small interstitial spaces (humaninaccessible) connected to a cave, where the physical environment provides the required humidity and temperature levels (Howarth 1983). These species may spend the majority of their time in such retreats, only leaving them to forage in the larger cave passages (Howarth 1987).

Since sunlight is absent or present in extremely low levels in caves, most karst ecosystems depend on nutrients derived from the surface either directly (organic material brought in by animals, washed in, or deposited through root masses) or indirectly through feces, eggs, and carcasses of trogloxenes (species that regularly inhabit caves for refuge, but return to the surface to feed) and troglophiles (species that may complete their life cycle in the cave, but may also be found on the surface) (Barr 1968; Poulson and White 1969; Howarth 1983; Culver 1986). Primary sources of nutrients include leaf litter, cave crickets, small mammals, and other vertebrates that defecate or die in the cave.

As described in our final rule to list the nine species (65 FR 81419), the continuing expansion of the human population in karst terrain constitutes the primary threat to the species through: (1) Destruction or deterioration of habitat by construction; (2) filling of caves and karst features and loss of permeable cover; (3) contamination from septic effluent, sewer leaks, runoff, pesticides, and other sources; (4) exotic species, especially nonnative fire ants (Solenopsis invicta); and (5) vandalism.

Karst in Bexar County

The northern portion of Bexar County is located on the Edwards Plateau, a broad, flat expanse of Cretaceous carbonate rock that ranges in elevation from 335.5 meters (m) (1,100 feet (ft)) to 579.5 m (1,900 ft) (Veni 1988; Soil Conservation Service 1962). This portion of the Plateau is dissected by numerous small streams and is drained by Cibolo Creek and Balcones Creek. To the southeast of the Plateau lies the Balcones Fault Zone, a 25kmwide fault zone that extends from the northeast corner of the County to the western County line. The many streams and karst features of this zone recharge the Edwards Aquifer.

The principal, cavecontaining rock units of the Edwards Plateau are the upper Glen Rose Formation, Edwards Limestone, Austin Chalk, and Pecan Gap Chalk (Veni 1988). The Edwards Limestone accounts for one third of the cavernous rock in Bexar County and contains 60 percent of the caves, making it the most cavernous unit in the County. The Austin Chalk outcrop is second to the Edwards in total number of caves. In Bexar County, the outcrop of the upper member of the Glen Rose Formation accounts for approximately onethird of the cavernous rock, but only 12.5 percent of Bexar County caves (Veni 1988). In Bexar County, the Pecan Gap Chalk, while generally not cavernous, has a greater than expected density of caves and passages (Veni 1988).

Veni (1994) delineated six karst areas within Bexar County. The regions were named after places within their boundaries. These karst fauna regions are bounded by geological or geographical features that may represent obstructions to the movement (on a geologic time scale) of troglobites, which has resulted in the presentday distribution of endemic (restricted to a given region) karst invertebrates in the Bexar County area.

These areas have been delineated by Veni (1994) into five zones that reflect the likelihood of finding a karst feature that will provide habitat for the endangered Bexar County invertebrates based on geology, distribution of known caves, distribution of cave fauna, and primary factors that determine the presence, size, shape, and extent of caves with respect to cave development. These five zones are defined as:

Zone 1: Areas known to contain one or more of the nine endangered karst invertebrates;

Zone 2: Areas having a high probability of suitable habitat for the invertebrates;

Zone 3: Areas that probably do not contain the invertebrates;

Zone 4: Areas that require further research but are generally equivalent to zone 3, although they may include sections that could be classified as zone 2 or zone 5; and

Zone 5: Areas that do not contain the invertebrates.

Under contract with the Service, Veni (2002) reevaluated and, where applicable, redrew the boundaries of each karst zone originally delineated in Veni (1994). Revisions were based on current geologic mapping, further studies of cave and karst development, and the most current information available on the distribution of listed and nonlisted caveadapted species (Veni 2002).

Endangered Karst Invertebrate Distribution

As of December 2002, 475 caves were known to occur in Bexar County, some of which have been biologically surveyed for listed species (Veni 2002). At least 97 of the 475 caves were sealed or destroyed before they could be biologically surveyed (Veni 2002). Not all of the remaining caves in Bexar County have been adequately surveyed for invertebrates. It is likely that some of these caves will be found to contain one or more of the listed species. When the species were listed as endangered in December 2000, the Service knew of 57 occupied caves. When critical habitat was proposed in Bexar County in August 2002, we knew of 69 occupied caves. We now know of 74 caves containing one or more of the listed species in Bexar County (Table 1). The following species status descriptions are based on information available to us as of December 23, 2002.

Braken Bat Cave Meshweaver

The Braken Bat Cave meshweaver, Cicurina venii (Araneae: Dictynidae), was first collected on November 22, 1980, by G. Veni and described by Gertsch (1992). Braken Bat Cave remains the only location known to contain this species (Table 1).

Cokendolpher Cave Harvestman

The Cokendolpher cave harvestman, Texella cokendolpheri (Opilionida: Phalangodidae), was collected in 1982 and described by Ubick and Briggs (1992). This species, along with the Robber Baron Cave meshweaver, is only known from Robber Baron Cave (Table 1). Government Canyon Bat Cave Meshweaver

The Government Canyon Bat Cave meshweaver, Cicurina vespera (Araneae: Dictynidae), was first collected on August 11, 1965, by J. Reddell and J. Fish (Reddell 1993), and described by Gertsch (1992). The species is currently known from Government Canyon Bat Cave in Government Canyon State Natural Area and an unnamed cave referred to as ``5 miles northeast of Helotes.'' However, the specimen collected from the latter cave has been tentatively identified as a new species (Cokendolpher, in press).

Government Canyon Bat Cave Spider

The Government Canyon Bat Cave spider, Neoleptoneta microps (Araneae: Leptonetidae), was first collected on August 11, 1965, by J. Reddell and J. Fish (Reddell 1993). The species was originally described by Gertsch (1974)
[[Page 17158]]
as Leptoneta microps and later reassigned to Neoleptoneta following Brignoli (1977) and Platnick (1986). The species is known from 2 caves in Government Canyon State Natural Area (Table 1).

Madla Cave Meshweaver

The Madla Cave meshweaver, Cicurina madla (Araneae: Dictynidae), was first collected on October 4, 1963, by J. Reddell and D. McKenzie (Reddell 1993) and described by Gertsch (1992). The Madla Cave meshweaver has been found in eight caves (Table 1).

The Service is aware of 11 additional caves from which immature, eyeless troglobitic Cicurina spiders have been collected (SWCA 2000). Eight of these are in caves that have other listed species and are either included in critical habitat areas or areas that are not included in the designation due to the provision of adequate special management. The remaining three are in caves where authorization for take of C. madla was granted to La Cantera under a section 10(a)(1)(B) permit. These three caves have been, or will be, heavily impacted and are, therefore, not expected to contribute to the species recovery. Robber Baron Cave Meshweaver

The Robber Baron Cave meshweaver, Cicurina baronia (Araneae: Dictynidae), was first collected in Robber Baron Cave February 28, 1969, by R. Bartholomew (Reddell 1993) and described by Gertsch (1992). The Robber Baron Cave meshweaver (a spider) is only known from Robber Baron Cave (Table 1).

Beetle (No Common Name) Rhadine exilis

The beetle Rhadine exilis (Coleoptera: Carabidae) was first collected in 1959. The species was described by Barr and Lawrence (1960) as Agonum exile and later assigned to the genus Rhadine (Barr 1974). The species is currently known to have been found in 47 caves (Table 1).

Beetle (No Common Name) Rhadine infernalis

Rhadine infernalis (Coleoptera: Carabidae) was first collected in 1959. The species was initially described by Barr and Lawrence (1960) as Agonum infernale, but later assigned to the genus Rhadine (Barr 1974). Scientists have recognized three subspecies (Rhadine infernalis ewersi, Rhadine infernalis infernalis, Rhadine infernalis new subspecies) (Barr 1974; Barr and Lawrence 1960; Reddell 1998), all of which are included as protected under the Federal listing of the full species as endangered. A total of 35 caves are known to contain Rhadine infernalis (Table 1).

Rhadine infernalis ewersi is known from 3 caves. Rhadine infernalis infernalis is known from 19 caves. The unnamed new subspecies (Rhadine infernalis new subspecies) was known from 6 caves at the time of the proposed rule designating critical habitat. During the public comment period, we received confirmation that R. infernalis collected from Obvious Little Cave has been identified as R. infernalis new subspecies. An additional 5 caves were identified in the proposed rule as containing Rhadine infernalis that have not yet been identified at the subspecies level. During the public comment period, we received survey information confirming the presence of R. infernalis in Continental Cave (Table 1). According to Veni (2002), specimens from these caves are probably R. infernalis infernalis, but have either not yet been fully identified or not reported.

Helotes Mold Beetle

The Helotes mold beetle, Batrisodes venyivi (Coleoptera: Pselaphidae), was first collected in 1984 and described by Chandler (1992). The species is currently known from six caves (Table 1). The location of one of the caves, referred to as ``unnamed cave \1/2\ mile north of Helotes,'' is unknown. The original record for this cave is from Barr's (1974) description of Rhadine exilis. Because the number of caves in the general area is large, the location of this cave cannot be positively identified (George Veni, George Veni & Associates, pers. comm. 2002). However, this cave may not be a separate location after all, but may be an existing cave listed by the collector under the alternative name ``5 miles NE of Helotes.'' The cave referred to as ``5 miles NE of Helotes,'' also has an unknown location.
Table 1.Caves Known as of December 23, 2002, To Contain One or More of the Nine Bexar County, Texas, Karst Invertebrates Federally Listed as Endangered
Species ( of caves) Cave name Braken Bat Cave meshweaver (C. venii) Braken Bat Cave.
(1).
Cokendolpher cave harvestman (Texella Robber Baron Cave. cokendolpheri) (1).
Government Canyon Bat Cave meshweaver Government Canyon Bat Cave. (C. vespera) (1).
Government Canyon Bat Cave spider Government Canyon Bat Cave, (Neoleptoneta microps) (2). Surprise Sink.
Madla Cave meshweaver (Cicurina madla) Christmas Cave, Madla's Cave, (8). Madla's Drop Cave, Helotes Blowhole, Headquarters Cave, Hills and Dales Pit, Robber's Cave, Lost Pothole. Robber Baron Cave meshweaver (C. Robber Baron Cave
baronia) (1).
Beetle (no common name) (Rhadine 40 mm Cave, B52 Cave, exilis) (47). Backhole, Black Cat Cave, Boneyard Pit, Bunny Hole, Cross the Creek Cave, Dos Viboras Cave, Eagles Nest Cave, Hairy Tooth Cave, Headquarters Cave, Hilger Hole, Hold Me Back Cave, Hornet's Last Laugh Pit, Isocow Cave, Kick Start Cave, MARS Pit, MARS Shaft, Pain in the Glass Cave, Platypus Pit, Poor Boy Baculum Cave, Ragin' Cajun Cave, Root Canal Cave, Root Toupee Cave, Springtail Crevice, Strange Little Cave, Up the Creek Cave. Christmas Cave, Helotes Blowhole, Helotes Hilltop Cave, Logan's Cave, unnamed cave \1/2\ mile N. of Helotes. Creek Bank Cave, Government Canyon Bat Cave, Lithic Ridge Cave, Pig Cave, San Antonio Ranch Pit, Tight Cave. Hills and Dales Pit, John Wagner Ranch Cave No. 3, Kamikazi Cricket Cave, La Cantera Cave No. 1, La Cantera Cave No. 2, Mastodon Pit, Robber's Cave, Three Fingers Cave, Young Cave No. 1. Beetle (no common name) R. infernalis Canyon Ranch Pit, Continental (6) (subspecies not indicated Cave, Fat Man's Nightmare probably R. infernalis infernalis but Cave, Pig Cave, San Antonio individual specimens are either not Ranch Pit, Scenic Overlook fully identified or reported (Veni Cave.
2002)).
[[Page 17159]]
R. infernalis ewersi (3)............... Flying Buzzworm Cave, Headquarters Cave, Low Priority Cave.
R. infernalis new subspecies (7)....... Caracol Creek Coon Cave, Game Pasture Cave No. 1, Isopit, King Toad Cave, Obvious Little Cave, Stevens Ranch Trash Hole Cave, Wurzbach Bat Cave. R. infernalis infernalis (19).......... Bone Pile Cave, Dancing Rattler Cave, Government Canyon Bat Cave, Hackberry Sink, Lithic Ridge Cave, Surprise Sink, Christmas Cave, Helotes Blowhole, Logan's Cave, Madla's Cave, Madla's Drop Cave, Crownridge Canyon Cave, Genesis Cave, John Wagner Ranch Cave No. 3, Kamikazi Cricket Cave, Mattke Cave, Robber's Cave, Scorpion Cave, Three Fingers Cave. Helotes mold beetle (Batrisodes San Antonio Ranch Pit, Scenic venyivi) (6). Overlook Cave, Christmas Cave, unnamed cave \1/2\ mile N of Helotes, Helotes Hilltop Cave, unnamed cave 5 miles NE of Helotes.
Animal Community

Cave Crickets

Cave crickets are a critical source of nutrient input for karst ecosystems (Barr 1968; Reddell 1993). Cave crickets in the genus Ceuthophilus occur in most caves in Texas (Reddell 1966). Being sensitive to temperature extremes and drying, cave crickets forage on the surface at night and roost in the cave during the day. Cave crickets lay their eggs in the cave, providing food for a variety of karst species (Mitchell 1971b). Some karst species also feed on cave cricket feces (Barr 1968; Poulson et al. 1995) and on adults and nymphs directly (Cokendolpher, in press; Elliott 1994a). Cave crickets are scavengers or detritivores, feeding on dead insects, carrion, and some fruits, but not on foliage (Elliott 1994a).

Elliott (2000) studied the community ecology of three caves in protected areas of varying size in northwest Travis and Williamson Counties, Texas, from 1993 to 1999. The three caves are in areas protected as mitigation for two listed species found in Lakeline Cave during the development of Lakeline Mall. Lakeline Cave is located on a 0.9 hectares (ha) (2.3 acres (ac)) protected area and is surrounded by parking lots and a shopping center. Temples of Thor Cave and Testudo Tube are within much larger tracts of undeveloped land, being located on 42.5 ha (105 ac), and 10.5 ha (26 ac) of protected areas, respectively. During the monitoring study (19931999), the number of cave crickets drastically declined in Lakeline Cave, while they increased slightly or decreased moderately in the other two caves. Elliott (2000) concluded that drought, fire ants, and a decrease in racoon visitation caused the decline of the cave crickets. These results are consistent with reports of declines and extinctions of several invertebrates and small mammals (resulting from lower survivorship, higher emigration, and/or lower immigration) from habitat patches ranging in size from 2 to 7 ha (5 to 17 ac) (Mader 1984; Tscharntke 1992; Keith et al. 1993; Lindenmayer and Possingham 1995; Hill et al. 1996).

Elliott (1994a) evaluated cave cricket foraging within 50 m (164 ft) of cave entrances at his study sites and found crickets to the end of the 50 m sampling distance. On a few occasions he observed cave crickets beyond his sampling sites, and on one occasion he set a trap 60 m (197 ft) from the entrance and found one large adult. Elliott (1994a) concluded that the ``largest adults probably are capable of traveling far beyond 60 m from the entrance,'' but he did not have the data necessary to establish how far they go. During recent cave cricket surveys conducted for an ongoing project in central Texas, an adult cave cricket was found foraging 95 m (311 ft) from the study cave (Steve Taylor, Illinois Natural History Survey, pers. comm. 2002).

As trogloxenes, cave cricket populations are dependent on the patchy distribution of karst voids. Therefore, cave cricket populations may have a metapopulation (subpopulations that interact via the dispersal of individuals from one subpopulation to others) or a source sink population structure, and it may be important to protect multiple karst features that support cave crickets in a karst ecosystem (Helf et al. 1995). Metapopulation dynamics require movement among patches, and persistence requires interacting patches that undergo local extinctions and establishment of new subpopulations in areas previously devoid of individuals (Hanski 1999). ``Source'' populations are those that occur ``in a highquality habitat in which birth rate generally exceeds the death rate and the excess individuals leave as emigrants.'' ``Sink'' populations are those that occur ``in a lowquality habitat in which the birth rate is generally lower than the death rate and population density is maintained by immigrants from source populations (Meffe et al. 1997). Because cave crickets are a key source of nutrient input for karst ecosystems, conserving adequate areas between karst patches in a manner that allows for movement of individuals among cave cricket populations is likely an important factor in longterm maintenance for karst ecosystems.

Subsurface karst areas may also be important to allow movement among cave cricket populations through the subsurface environment associated with continuous limestone blocks. For example, Caccone and Sbordoni (1987) studied nine species of North American cave crickets (genera Eukadenoecus and Hadenoecus) from sites in North Carolina, Ohio, Pennsylvania, Tennessee, Virginia, West Virginia, Kentucky, and Alabama. Seven of the species were obligate cavedwelling species that emerged at night to feed. Through genetic analyses of the cavedwelling species, they found that species or groups of populations inhabiting areas where the limestone is continuous and highly fissured are genetically less differentiated than are populations occurring in regions where the limestone distribution is more fragmented, indicating more exchange of individuals in areas of continuous karst.

Helf et al. (1995) suggested that populations of an eastern species of cave cricket (Hadenoecus subterraneus) may be at risk because they do not recover quickly after events such as drought, floods, and temperature extremes that preclude or diminish foraging opportunities. These cave cricket populations may have sourcesink population dynamics, with some
[[Page 17160]]
karst features acting as sources and the majority of karst features acting as sinks, but Helf et al. (1995) recommends that ``even sink populations should be protected because their emigrants can ``rescue'' source populations that experience local decimation.'' These studies suggest that it is important to protect the geological features that connect caves and maintain habitat corridors among caves.

Other Surface Animals

Many central Texas caves with endangered invertebrate species are frequented by mammals and several species of reptiles and amphibians (Reddell 1967). Although there are no studies establishing the role of mammals in central Texas cave ecology, the presence of a large amount of animal materials (such as scat, nesting materials, and dead bodies) indicates they are probably important. An important source of nutrients for the cave species may be the fungus, microbes, and/or other troglophiles and troglobites that grow or feed on feces (Elliott 1994b; Gounot 1994).

For predatory troglobites (such as the listed Bexar County invertebrates), invertebrates that accidently occur in the caves may also be an important nutrient source (Hopper 2000). Documented accidental species include snails, earthworms, terrestrial isopods (commonly known as pillbugs or potato bugs), scorpions, spiders, mites, collembola (primitive wingless insects that are commonly known as springtails), thysanura (commonly known as bristletails and silverfish), harvestmen (commonly known as daddylonglegs), ants, leafhoppers, thrips, beetles, weevils, moths, and flies (Reddell 1965; 1966; 1999).

Vegetation Community

Surface vegetation is an important element of the karst habitat for several reasons, including its role in providing nutrients from: (1) Direct flow of plant material into the karst with water; (2) habitat and food sources provided for the animal communities that contribute nutrients to the karst ecosystem (such as cave crickets, small mammals, and other vertebrates); and possibly, (3) roots that extend into subsurface areas. Surface vegetation also acts as a buffer for the subsurface environment against drastic changes in the temperature and moisture regime and serves to filter pollutants before they enter the karst system (Biological Advisory Team 1990; Veni 1988). In some cases, healthy native plant communities also help control certain exotic species (such as fire ants) (Porter et al. 1988) that may compete with or prey upon the listed species and other species (such as cave crickets) that are important nutrient contributors (Elliott 1994a; Helf, in litt. 2002).

Tree roots have been found to provide a major energy source in shallow lava tubes and limestone caves in Hawaii (Howarth 1981). Jackson et al. (1999) investigated rooting depth in 21 caves on the Edwards Plateau to assess the belowground vegetational community structure and the functional importance of roots. They observed roots penetrating up to 25 m (82 ft) into the interior of 20 of the caves, with roots of 6 tree species common to the plateau penetrating to below 5 m (16.4 ft).

Along with providing directly and indirectly nutrients to the karst ecosystem, a healthy vegetative community may also help control the spread of exotic species. The red imported fire ant (Solenopsis invicta) is an aggressive predator, which has had a devastating and longlasting impact on native ant populations and other arthropod communities (Vinson and Sorenson 1986; Porter and Savignano 1990) and is a threat to the karst invertebrates (Elliott 1994b; USFWS 1994). Fire ants have been observed building nests both within and near cave entrances, as well as foraging in caves, especially during the summer. Shallow caves inhabited by listed karst invertebrates are especially vulnerable to invasion by fire ants and other exotic species. In addition to preying on cave invertebrate species, including cave crickets, fire ants may compete with cave crickets for food (Elliott 1994a; Helf in litt. 2002). Helf (in litt. 2002) states that competition for food between fire ants and cave crickets (Ceuthophilus secretus) may be a more important interaction than predation. The presence of fire ants in and around karst areas could have a drastic detrimental effect on the karst ecosystem through loss of both surface and subsurface species that are critical links in the food chain.

The invasion of fire ants is known to be aided by ``any disturbance that clears a site of heavy vegetation and disrupts the native ant community'' (Porter et al. 1988). Porter et al. (1991) state that control of fire ants in areas greater than 5 ha (12 ac) may be more effective than in smaller areas since multiple queen fire ant colonies reproduce primarily by ``budding,'' where queens and workers branch off from the main colony and form new sister colonies. Maintaining large, undisturbed areas of native vegetation may also help sustain the native ant communities (Porter et al. 1988; 1991).

Listed species, and their associated prey items, have adapted to native vegetation, with its associated nutrients, surface foliage, and subsurface roots. Before 1860, Bexar County native vegetation consisted of an approximate equal mix of areas with woody and grassland plants (Del Weniger 1988). In more recent times, exotic species have often replaced native plants. The effects on listed invertebrates of replacement of native with exotic vegetation have not been reported. WoodlandGrassland Community

Because of the various roles played by surface vegetation in maintaining the cave and karst ecosystem, including the listed karst invertebrate species that are part of the ecosystem, we examined the best available scientific information to estimate the surface vegetation needed to support ecosystem processes. The woodland grassland mosaic community typical of the Edwards Plateau is a patchy environment composed of many different plant species. Van Auken et al. (1980) studied the woody vegetation of the Edwards and Glen Rose formations in the southern Edwards Plateau in Bexar, Bandera, and Medina counties. They encountered a total of 24 species of plants on the Edwards or Glen Rose geologic formations, two of the principal, cavecontaining rock units of the Edwards Plateau.

To maintain natural vegetation communities over the long term, enough individuals of each plant species must be present for successful reproduction. The number of reproductive individuals necessary to maintain a viable or selfreproducing plant population is influenced by needs for satisfactory germination (Menges 1995), genetic variation (Bazzaz 1983; Menges 1995; Young 1995), and pollination (Groom 1998; Jennersten 1995; Bigger 1999). Pavlik (1996) stated that longlived, selffertilizing, woody plants with high fecundity would be expected to have minimum viable population sizes in the range of 50250 reproductive individuals. Fifty reproductive individuals is a reasonable minimum figure for one of the dominant species of the community (Juniperous ashei) based on reproductive profiles (Van Auken et al. 1979; Van Auken et al. 1980; Van Auken et al. 1981). This figure would likely be an underestimate for other woody species present in central Texas woodlands, however, because these other species are more sensitive to environmental changes and do not meet several of the life history criteria needed for the lowest minimal viable population size. Although these species may require population sizes at
[[Page 17161]]
the higher end of range (that is, nearer 250 individuals) suggested by Pavlik (1996) to be viable, we do not have the data to support that contention. Therefore, on the basis of our review of information available to us, and after soliciting input from a botanist with expertise in the Edwards Plateau (Dr. Kathryn Kennedy, Center for Plant Conservation, pers. comm. 2002), we consider a minimum viable population size for individual plant species composing a typical oak/ juniper woodland found in central Texas to be 80 individuals per species. This estimate is based on a habitat type that, as a whole, is fairly mature, and on knowledge that the species are relatively long lived and reproductively successful.

On the basis of an analysis of recorded densities, corrected for nonreproductive individuals, we then calculated the area needed to support 80 mature reproductive individuals per species for the 24 species reported by Van Auken et al. (1980). Based on our calculations, the four highest area requirements to maintain at least 80 mature individuals were for species that occur at lower densities. These included 80 ha (198 ac) for Condalia hookeri, and approximately 32 ha (79 ac) for each of Ptelea trifoliata, Ungnadia speciosa, and Bumelia lanuginosa. Our calculations indicate that the area needed to maintain the 7 species with the highest average dominance values (Juniperus ashei, Quercus fusiformis, Quercus texana, Acacia greggii, Rhus virens, Berberis trifoliata, and Ulmus crassifolia) is approximately 13 ha (33 ac). This number would maintain 80 reproductive individuals for 15 of the 24 species. Nine of the species are rarer in the community and all have importance values of less than 1.0. The area needed to maintain these nine species ranges from approximately 20 to 80 ha (49 to 198 ac), with 7 of them in the 26 ha to 32 ha (65 to 79 ac) range.

Most literature found for Central Texas native grasslands was descriptive and not quantitative in its treatment of species composition and dispersion. No literature was located that provided grassland species area curves or quantitative species density tables for the Central Texas area. Two papers by Lynch (1962, 1971) examined species on an 8acre tract over time, with 123 species, but a high species turnover. High species turnover can be indicative of a habitat area which is too small; however, pre and postdrought conditions may also have affected this situation. Robertson et al. (1997), in a slightly more mesic grassland habitat, found that a 4 ha (10 ac) site captured most of the species diversity (100 species) present even in much larger patches, although it does not address population sizes and persistence in isolation, and an increase to a 6 ha (14 ac) tract increased species representation to 140. One paper on a grassland in a more westerly and drier location in Central Texas recorded 157 taxa in a 16 ha (40 ac) exclosure studied between 1948 and the mid1970's (Smeins and Merrill 1976).

Primary recruitment of new individuals of grass species in grasslands is from seedling establishment. Many grass species use wind to disperse their seeds and dispersal distances may be small. The process of expansion through rhizomes (underground stems) is slow and clonal, which reduces genetic variability. Seed dispersal, soil texture, and suitable soil moisture profiles at critical times are important factors for maintaining viability (Coffin et al. 1993).

As described above, we have reviewed the available information concerning grasslands and grassland species in Central Texas. The information is of a relatively general nature, and we did not find specific information addressing the role that grasslands or grass species might play in contributing, directly or indirectly, to karst ecosystems. While grassland communities and species may be important to maintaining the karst community, we lack adequate information to credibly estimate surface habitat patch size requirements for grass species in relation to karst ecosystems.

The presence of surface vegetation communities is important for maintaining the humid conditions, stable temperatures, and natural airflow in cave and karst environments. Vegetation also plays an important role in water quality. Since soil depth is shallow over the limestone plateau, water collects as sheet flow on the surface following rain and enters the subsurface environment through cave openings, fractures, and solutionallyenlarged bedding planes. This direct, rapid transport of water through the karst allows for little or no purification (Veni 1988), allowing contaminants and sediments to enter directly into the subsurface environment. As a result, karst features and karst dependent invertebrates are vulnerable to the adverse effects of pollution from contaminated ground and surface water. Maintaining stable environmental conditions and protecting groundwater quality and quantity requires managing a healthy vegetation community to avoid threats from surface and subsurface drainage to the karst environment needed by the karst dependent species. This includes not only the cave entrances accessible to humans, but also sinks, depressions, fractures, and fissures, which may serve as subsurface conduits into caves and other subsurface spaces used by the invertebrates.

Buffer Areas

To maintain a viable vegetative community, including woodland and grassland species, a buffer area is needed to shield the core habitat from impacts associated with edge effects or disturbance from adjacent urban development (Lovejoy et al. 1986; Yahner 1988). In this context, edge effects refer to the adverse changes to natural communities (primarily from increases in invasive species and pollutants, and changes in microclimates) from nearby areas that have been modified for human development.

The changes caused by edge effects can occur rapidly. For example, vegetation 2 m (6.6 ft) from a newly created edge can be altered within days (Lovejoy et al. 1986). Edges may allow invasive plant species to gain a foothold where the native vegetation had previously prevented their spread (Saunders et al. 1990; Kotanen et al. 1998; Suarez et al. 1998; Meiners and Steward 1999). When plant species composition is altered as a result of an edge effect, changes also occur in the surface animal communities (Lovejoy and Oren 1981; Harris 1984; Mader 1984; Thompson 1985; Lovejoy et al. 1986; Yahner 1988; Fajer et al. 1989; Kindvall 1992; Tscharntke 1992; Keith et al. 1993; Hanski 1995; Lindenmayer and Possingham 1995; Bowers et al. 1996; Hill et al. 1996; Kozlov 1996; Kuussaari et al. 1996; Turner 1996; Mankin and Warner 1997; Burke and Nol 1998; Didham 1998; Suarez et al. 1998; Crist and Ahern 1999; Kindvall 1999). Changes in plant and animal species composition as a result of edge effects may unnaturally change the nutrient cycling processes required to support cave and karst ecosystem dynamics. To minimize edge effects, the core area must have a sufficient buffer area.

One recommendation for protecting forested areas from edge effects that are in proximity to clearcut areas is use of the ``three tree height'' approach (Harris 1984) for estimating the width of the buffer area needed. We used this general rule to estimate the width of buffer areas needed to protect the habitat core areas. The average height of native mature trees in the Edwards woodland association in Texas ranges from 3 to 9
[[Page 17162]]
m (10 to 30 ft) (Van Auken et al. 1979). Applying the ``three tree height'' general rule, and using the average value of 6.6 m for tree height, we estimated that a buffer width of at least 20 m (66 ft) is needed around a core habitat area to protect the vegetative community from edge effects. Based on this rule, 7 acres is necessary to protect a 33acre core area. We recognize that the ``three tree height'' approach described by Harris (1984) was based on the distance that effects of storm events (``windthrow'') from a surrounding clearcut ``edge'' will penetrate into an oldgrowth forest stand. Since the effects of edge on woodland/grass land mosaic communities have not been well studied, the ``three tree height'' recommendation is considered to be the best available peerreviewed science to protect woodland areas from edge effects (Dr. Kathryn Kennedy, Center for Plant Conservation, pers. comm. 2003). The Texas Parks and Wildlife Department is also in general agreement about the need for some type of buffer as a means of addressing edge effects, but currently has not specific recommendations on appropriate size for such a buffer ( John Herron, Texas Parks and Wildlife Department, pers. comm. 2003).

Animal communities also should be buffered from impacts associated with edge effects or disturbance from adjacent urban development. Edges can act as a barrier to dispersal of birds and mammals (Yahner 1988; Hansson 1998). Invertebrate species are affected by edges. Mader et al. (1990) found that carabid beetles and lycosid spiders avoided crossing unpaved roads that were even smaller than 3 m (9 ft) wide. Saunders et al. (1990) suggested that as little as 100 m (328 ft) of agricultural fields may be a complete barrier to dispersal for invertebrates and some species of birds. In general, for animal communities, species need buffers of 50 to 100 m (164 to 328 ft) or greater to ameliorate edge effects (Lovejoy et al. 1986; Wilcove et al. 1986; Laurance 1991; Laurance and Yensen 1991; Kapos et al. 1993; Andren 1995; Reed et al. 1996; Burke and Nol 1998; Didham 1998; Suarez et al. 1998).

Nonnative fire ants are known to be harmful to many species of invertebrates and vertebrates. In coastal southern California, Suarez et al. (1998) found that densities of the exotic Argentine ant (Linepithema humile), which has similar life history and ecological requirements to the red imported fire ant (Dr. Richard Patrock, University of Texas at Austin, pers. comm. 2003), are greatest near disturbed areas. Native ant communities tended to be more abundant in native vegetation and less abundant in disturbed areas. Based on the association of the Argentine ant and distance to the nearest edge in urban areas, core areas may only be effective at maintaining natural populations of native ants when there is a buffer area of at least 200 m (656 ft) (Suarez et al. 1998).

Information on the area needed to maintain populations of animal species, including cave crickets, found in Central Texas is lacking. As discussed above, animal communities should be buffered by areas of 50 to 100 m (164 to 328 ft) or greater to ameliorate edge effects, and by areas of 200 m (656 ft) to buffer against the effects of fire ants. From this data, we determined that a buffer of 100 m (328 ft), in addition to the 50 m (164 ft) cave cricket foraging area, would, at a minimum, protect the cave cricket foraging area from the effects of edge and nonnative species invasions.

Fragmentation

Haskell (2000) examined the effect of habitat fragmentation by unpaved roads through otherwise contiguous forest in the southern Appalachian Mountains and found reduced soil macroinvertebrate species abundance up to 100 m (328 ft) from the road and declines in faunal richness up to 15 m (50 ft) from the road. Haskell (2000) pointed out that ``these changes may have additional consequences for the functioning of the forest ecosystem and the biological diversity found within this system. The macroinvertebrate fauna of the leaf litter plays a pivotal role in the ability of the soil to process energy and nutrients.'' Haskell further points out that these changes may in turn affect the distribution and abundance of other organisms, particularly plants. Changes in abundance in litter dwelling macroinvertebrates may also affect groundforaging vertebrate fauna (Haskell 2000).

Invertebrate biomass per unit area has been found to be less in small fragmented habitats, which may result in reduced food available for cave crickets. Burke and Nol (1998), working in southern Ontario, Canada, found a greater biomass of leaf litter invertebrates in large (=20 ha (49 ac)) than in smaller forested areas. Zanette et al. (2000) in New South Wales, Australia, reported that the biomass of ground dwelling invertebrates was 1.6 times greater in large ( 400 ha (988 ac)) than in smaller ([sim]55 ha (136 ac)) forested areas.

Dispersal

The ability of individuals to move between preferred habitat patches is essential for colonization and population viability (Eber and Brandl 1996; Fahrig and Merriam 1994; Hill et al. 1996; Kattan et al. 1994; Kindvall 1999; Kozlov 1996; Kuussaari et al. 1996; Turner 1996). Patch shapes allowing connection with the highest number of neighboring patches increase the likelihood that a neighboring patch will be occupied (Fahrig and Merriam 1994; Kindvall 1999; Kuussaari et al. 1996; Tiebout and Anderson 1997). If movement among populations is restricted and a population is isolated, the habitat patch size must be large enough to ensure that the population can survive (Fahrig and Merriam 1994).

It is likely that many cave systems are connected throughout the subsurface geologic formation even though this may not be readily apparent from surface observations. The extent to which listed species use interstitial spaces and passages is not known. Troglobitic species may retreat into these small interstitial spaces where the physical environment is more stable (Howarth 1983) and may spend the majority of their time in such retreats, only leaving them during temporary forays into the larger cave passages to forage (Howarth 1987). During several karst invertebrate surveys conducted in Bexar County caves, Service biologists have observed that troglobites, including listed species, were not found when temperature and humidity in the cave was low. Upon returning to the same cave once environmental conditions returned to optimal, the listed species and other troglobites were observed.

Small voids (inaccessible to humans) and interstitial spaces can also provide subsurface corridors for movement of listed species and cave crickets between and among caves and karst features. Cores drilled around and between occupied caves have led to discovery of additional void space that was hydrologically, but not physically connected to the humanlyaccessible portion of an occupied cave. Listed species were found in this void space.

Summary

The conservation of the endangered karst invertebrates depends on a selfsustaining karst ecosystem; surface and subsurface drainage basins to maintain adequate levels of moisture; and a viable surface animal and plant community for nutrient input and protection of the subsurface from adverse impacts. The area needed to conserve such an
[[Page 17163]]
ecosystem includes a core area buffered from the impacts associated with fragmentation, isolation, edge effects, and other factors that may threaten ecosystem stability. Depending on the size and shape of these core habitat areas or patches, in order to remain viable, they may also require connections to other habitat patches.

Previous Federal Action

On January 16, 1992, we received a petition submitted by representatives of the Helotes Creek Association, the Balcones Canyonlands Conservation Coalition, the Texas Speleological Association, the Alamo Group of the Sierra Club, and the Texas Cave Management Association to add the nine invertebrates to the List of Threatened and Endangered Wildlife. On December 1, 1993, we announced in the Federal Register (58 FR 63328) a 90day finding that the petition presented substantial information that listing may be warranted.

On November 15, 1994, we added eight of the nine invertebrates to the Animal Notice of Review as category 2 candidate species in the Federal Register (59 FR 58982). We intended to include Rhadine exilis in the notice of review, but an oversight occurred and it did not appear in the published notice. Category 2 candidates, a classification since discontinued, were those taxa for which we had data indicating that listing was possibly appropriate, but for which we lacked substantial data on biological vulnerability and threats to support proposed listing rules.

On December 30, 1998, we published a proposed rule to list the nine Bexar County karst invertebrates as endangered (63 FR 71855). Incorporating comments and new information received during the public comment period on the proposed rule, we published a final rule to list the nine Bexar County karst invertebrate species as endangered in the Federal Register on December 26, 2000 (65 FR 81419).

In the proposed rule for listing these species, we indicated that designation of critical habitat was not prudent for the nine invertebrates because the publication of precise species locations and maps and descriptions of critical habitat in the Federal Register would make the nine species more vulnerable to incidents of vandalism through increased recreational visits to their cave habitat and through purposeful destruction of the caves. We also indicated that designation of critical habitat was not prudent because it would not provide any additional benefits beyond those provided through listing the species as endangered.

Based on recent court decisions (for example, Natural Resources Defense Council v. U.S. Department of the Interior 113 F. 3d 1121 (9th Cir. 1997); Conservation Council for Hawaii v. Babbitt, 2 F. Supp. 2d 1280 (D. Hawaii 1998)) and the standards applied in those judicial opinions, we reexamined the question of whether critical habitat for the nine invertebrates would be prudent. After reexamining the available evidence for the nine invertebrates, we did not find specific evidence of collection or trade of these or any similarly situated species. Consequently, in our final rule listing the species, we found that ``by designating critical habitat in a manner that does not identify specific cave locations, the threat of vandalism by recreational visits to the cave or purposeful destruction by unknown parties should not be increased'' (65 FR 81419). Therefore, our final rule to list the species as endangered also included our determination that critical habitat designation was prudent as we did not find specific evidence of increased vandalism, and we found there may be some educational or informational benefit to designating critical habitat. Thus, we found that the benefits of designating critical habitat for the nine karst invertebrate species outweighed the benefits of not designating critical habitat.

The Final Listing Priority Guidance for FY 2000 (64 FR 57114) stated that we would undertake critical habitat determinations and designations during FY 2000 as allowed by our funding allocation for that year. As explained in detail in the Listing Priority Guidance, our listing budget was insufficient to allow us to immediately complete all of the listing actions required by the Act during FY 2000. We stated that we would propose designation of critical habitat in the future at such time when our available resources and priorities allowed.

On November 1, 2000, the Center for Biological Diversity (Center) filed a complaint against the Service alleging that the Service exceeded its 1year deadline to publish a final rule to list and to designate critical habitat for the nine Bexar County cave
invertebrates. Subsequent to the Service publishing the final rule to list these nine species as endangered on December 26, 2000, the Center agreed to dismiss its claim regarding the listing of the species. Under the terms of a settlement reached between the Center and the Service, the Service agreed to submit to the Federal Register for publication a proposed critical habitat determination on or by June 30, 2002, and a final determination on or by January 25, 2003. Sixtyday extensions on the deadlines to submit both the proposed and final critical habitat determinations to the Federal Register for publication were approved by the court, and the new deadlines became August 31, 2002, and March 26, 2003, for the proposed and final rules, respectively.

On February 28, 2002, we mailed letters to the Texas Parks and Wildlife Department and the Texas Natural Resource Conservation Commission informing them that we were in the process of designating critical habitat for the nine Bexar County karst invertebrates. We requested any additional available information on the listed species, including biology; life history; habitat requirements; distribution, including geologic controls to species distribution; current threats; and management activities, current or in the foreseeable future. The letters contained a current list of Bexar County caves known to contain listed species, a map showing the general distribution of these species within each Karst Fauna Region, and a list of the references pertaining to these species and their distribution as we know it. We requested their review and comments on our current information and asked their assistance in providing any additional available information.

We also mailed approximately 300 preproposal letters to interested parties and cave biologists on March 20, 2002, informing them that we were in the process of designating critical habitat for the 9 listed karst invertebrates. The letters contained a copy of the final rule to list these Bexar County invertebrate species as endangered, a map showing the general distribution of these species, a list of literature about these species and their habitats, and a brief summary with questions and answers on critical habitat. We requested comments on: (1) The reasons why any habitat should or should not be determined to be critical habitat as provided by section 4 of the Act, including whether the benefits of excluding areas will outweigh the benefits of including areas; (2) land use practices and current or planned activities in the subject areas and their possible impacts on possible critical habitat; (3) any foreseeable economic or other impacts resulting from the proposed designation of critical habitat, and particularly any impacts on small entities or families; and (4) economic and other benefits associated with designating critical habitat for the Bexar County karst invertebrates.

On August 27, 2002, we proposed that 25 units encompassing a total of approximately 3,857 ha (9,516 ac) in
[[Page 17164]]
Bexar County, Texas, be designated as critical habitat for the nine karst invertebrates (67 FR 55064). The comment period for the proposed rule was originally scheduled to close on November 25, 2002, but was extended until December 23, 2002 (67 FR 70203), to allow for a 30day comment period on the draft economic analysis. Thus, we accepted comments on the proposed rule and the economic analysis until December 23, 2002.

Summary of Comments and Recommendations

In the August 27, 2002, proposed rule, we requested all interested parties to submit comments or information concerning the designation of critical habitat for the nine endangered Bexar County invertebrates (67 FR 55064). During the comment period, we held a public hearing in San Antonio on October 30, 2002. We published a newspaper notice inviting public comment and announcing the public hearing in the San Antonio ExpressNews. A transcript of the hearing is available for inspection (see ADDRESSES section). The comment period was originally scheduled to close on November 25, 2002.

On November 21, 2002, we announced the availability of the draft economic analysis and requested comments on it and the proposal during an extension of the comment period until December 23, 2002 (67 FR 70203). We contacted all appropriate State and Federal agencies, county governments, scientific organizations, and other interested parties and invited them to comment. We also provided notification of these documents through email, telephone calls, letters, and news releases faxed and/or mailed to affected elected officials, media outlets, local jurisdictions, and interest groups. For the notice of the proposed rule, we mailed over 1,500 letters to interested parties. Later we sent over 1,200 post cards notifying interested parties of the availability of the draft economic analysis and the extension of the comment period. The number of parties on the mailing list fell as we deleted outof date and duplicate addresses. We also published all of the associated documents on the Service's regional Internet site following their release.

We solicited 11 independent experts who are familiar with these species and the karst ecosystem to peerreview the proposed critical habitat designation. Only one of the peer reviewers submitted comments, generally in support of the proposed designation (see ``Peer Review'' section below). We also received a total of 42 written comments, and 3 oral comments at the public hearing. Of those comments indicating a preference, 10 supported the critical habitat designation and 13 indicated opposition to designation. Many commenters did not express opposition to the designation, but did express opposition to specific areas being included. We reviewed all comments received for substantive issues and new data regarding critical habitat and the draft economic analysis. Here, we address all comments on both documents received during the comment periods, as well as public hearing testimony. We have grouped similar comments and addressed them in the following summary.
Issue 1: Biological Justification and Methodology for Size of Critical Habitat Units
(1) Comment: The Service should designate smaller areas for critical habitat units, including: (1) Surface and subsurface drainage areas; (2) cave cricket foraging areas; and (3) dominant and subdominant woody species, rather than uncommon plant species. The Service focused its methodology on surface plant communities, but little information exists relating particular vegetation communities to the subsurface habitat of the listed species.

Our Response: We believe it is well documented that surface flora and fauna communities are an essential energy source for fauna, including the nine endangered invertebrates, in the karst environment. The areas needed to support dominant, subdominant, and ``other woody species'' common to the Edwards Plateau were included in our proposal to incorporate key components of the native vegetative community that contribute directly to nutrient input, and which also support the animal community that is another source of nutrient input to karst areas. We do not have data from vegetation surveys conducted around occupied caves to determine the importance of rarer plant species. Therefore, in this final designation we have reduced the size of all of the critical habitat units based on the amount of area that we believe, based on the best available information, is needed to support at least 15 of 24 species of vegetation on the Edwards Plateau, including the seven species with the highest dominance values, but not the rarer plant species (see ``Criteria Used to Delineate Critical Habitat'' section below for further explanation).
(2) Comment: The Service should designate larger areas for the critical habitat units to: (1) Include all or most of Karst Zone 1; (2) all or portions of Karst Zone 2; (3) reduce fragmentation of habitat; (4) consider subsurface karst voids between known caves that may provide habitat for the species; (5) provide better protection against pollution; and (6) provide dispersal corridors for cave crickets.

Our Response: We agree that it is likely that all of these concerns have the potential to affect the conservation of the endangered karst invertebrates. Much of the biology and ecology of these karstadapted listed species is not well understood. Critical habitat was delineated to encompass areas on which are found those components of the karst ecosystem for which sufficient information exists to determine that they are essential to the conservation of the listed species.

We recognize that areas outside of the boundaries of critical habitat may be important for the karst invertebrates for purposes such as providing habitat in interstitial karst voids (beyond the known caves), additional sources of nutrients, or dispersal corridors. However, we did not have sufficient data when we proposed critical habitat, nor were any data provided during the comment period, that would allow us to adequately assess the importance to occupied caves of other areas of Karst Zones 1 or 2, karst voids between known caves, larger buffers, or areas that are needed for dispersal corridors for cave crickets. For instance, members of the Technical Subcommittee of the Karst Invertebrate Recovery Team, who are experts on the species and the karst ecosystems, agree that it is likely the invertebrates spend considerable time, perhaps the majority of time, in the human inaccessible karst voids (interstitial spaces) associated with the cave (Steve Taylor, Technical Subcommittee chair, pers. comm. 2002). However, the distance that these invertebrates go from the cave into the surrounding karst is unknown. Since protection of the surface and subsurface drainage areas associated with each occupied cave is important to buffer the cave from pollutants, these drainage areas were included, where possible, in the critical habitat designation. Additional scientific discovery may show that larger areas are needed for longterm conservation, and we will continue to incorporate such information into planning and implementing various conservation activities for these species. Given the best available information, we believe the specific areas designated in this rule contain one or more of the physical or biological features that are essential to the conservation of the species and meet the definition of critical habitat as provided in section 3 of the Act.
[[Page 17165]]
(3) Comment: The proposed rule did not show that designating critical habita

FOR FURTHER INFORMATION CONTACT Robert Pine, Supervisor, U.S. Fish and Wildlife Service, Austin Ecological Services Field Office, at the above address (telephone 512/4900057; facsimile 512/4900974).