This final rule downgrades the Mariana fruit bat from endangered to threatened throughout its range in the Mariana archipelago, which is subject to US jurisdiction.  The reason for the down grade is the FWS initially made a mistake in the taxonomy of the Mariana fruit bat.  When the FWS listed the bat as endangered on Guam in 1984, it believed that the bat was a species only endemic to Guam.  Since that time, the FWS has discovered that the bat is endemic to the entire Commonwealth of the Northern Mariana Islands (CNMI) and the Territory of Guam, thus there is actually a larger number of bats with a wider distribution.  Yet, threats to the Mariana fruit bat still remain, so its listing as threatened is still warranted.  Additionally, it is more convenient for the FWS to update the listing of the Mariana fruit bat to threatened in the entire Mariana archipelago, than to keep the species in the Guam as endangered and hold the remainder of the archipelago as threatened.

This final rule downgrades the Mariana fruit bat from endangered to threatened throughout its range in the Mariana archipelago, which is subject to US jurisdiction.  The reason for the down grade is the FWS initially made a mistake in the taxonomy of the Mariana fruit bat.  When the FWS listed the bat as endangered on Guam in 1984, it believed that the bat was a species only endemic to Guam.  Since that time, the FWS has discovered that the bat is endemic to the entire Commonwealth of the Northern Mariana Islands (CNMI) and the Territory of Guam, thus there is actually a larger number of bats with a wider distribution.  Yet, threats to the Mariana fruit bat still remain, so its listing as threatened is still warranted.  Additionally, it is more convenient for the FWS to update the listing of the Mariana fruit bat to threatened in the entire Mariana archipelago, than to keep the species in the Guam as endangered and hold the remainder of the archipelago as threatened.

RULES and REGULATIONS

DEPARTMENT OF THE INTERIOR

Fish and Wildlife Service

50 CFR Part 17

RIN 1018-AH55

Endangered and Threatened Wildlife and Plants; Mariana Fruit Bat (Pteropus

mariannus mariannus %g ): Reclassification From Endangered to Threatened in the

Territory of Guam and Listing as Threatened in the Commonwealth of the Northern

Mariana Islands

Thursday, January 6, 2005

AGENCY: Fish and Wildlife Service, Interior.

*1190 ACTION: Final rule.

SUMMARY: We, the U.S. Fish and Wildlife Service (Service), reclassify from endangered to threatened status the Mariana fruit bat (Pteropus mariannus mariannus) from Guam, *1191 under the authority of the Endangered Species Act of 1973, as amended (Act), and determine the Mariana fruit bat from the Commonwealth of the Northern Mariana Islands (CNMI) to be a threatened species under the authority of the Act. This rule lists the Mariana fruit bat as threatened throughout its range.

The Mariana fruit bat was listed previously as endangered on Guam. The bat populations on the southern islands of the CNMI (Aguiguan, Tinian, and Saipan) were candidates for listing. The best available scientific information indicates that Mariana fruit bats on Guam and throughout the CNMI comprise one subspecies. The protections of the Act, therefore, apply to this subspecies throughout its known range in the Mariana archipelago.

DATES: This final rule is effective February 7, 2005.

ADDRESSES: Comments and materials received, as well as supporting documentation used in the preparation of this final rule, will be available for public inspection, by appointment, during normal business hours at the Pacific Islands Fish and Wildlife Office, U.S. Fish and Wildlife Service, 300 Ala Moana Boulevard, Room 3-122, Box 50088, Honolulu, HI 96850.

FOR FURTHER INFORMATION CONTACT: Gina Shultz, Assistant Field Supervisor, Pacific Islands Fish and Wildlife Office (see ADDRESSES section) (telephone 808/792-9400; facsimile 808/792-9581).

SUPPLEMENTARY INFORMATION:

Background

The Mariana archipelago consists of the 15-island Commonwealth of the Northern Mariana Islands (CNMI) and the Territory of Guam, both within the jurisdiction of the United States. This archipelago extends 470 miles (mi) (750 kilometers (km)) from 13° 14' N, 144° 45' W to 20° 3' N, 144° 54' W and is approximately 900 mi (1,500 km) east of the Philippine Islands (Figure 1). Nine of the 10 northern islands (Anatahan, Sarigan, Guguan, Alamagan, Pagan, Agrihan, Asuncion, Maug, and Uracas) are volcanic in origin, and Farallon de Medinilla and the five southern islands (Guam, Rota, Aguiguan, Tinian, and Saipan) are uplifted limestone plateaus with volcanic outcrops. Mariana fruit bats have historically inhabited all of these islands except Uracas, the northernmost island (Wiles and Glass 1990). Of the largest southern islands (Guam, Rota, Tinian, and Saipan), Guam supports the majority of the human population. The northern islands (north of Saipan) are either unoccupied or support only a few families. The climate is tropical, with daily mean temperatures of 75 to 90° Fahrenheit (24 to 32° Celsius), high humidity, and average annual rainfall of 80 to 100 inches (in) (200 to 260 centimeters (cm)). Typhoons may strike the Mariana Islands during any month of the year, but are most frequent between July and October.

BILLING CODE 4310-55-P

Image 1 (8.5" X 6.5") Available for Offline Print

BILLING CODE 4310-55-C

*1193 Species Description and Biology

The Mariana fruit bat is a medium-sized fruit bat in the family Pteropididae that weighs 0.66 to 1.15 pounds (330 to 577 grams) and has a forearm length ranging from 5.3 to 6.1 in (13.4 to 15.6 cm); males are slightly larger than females. The underside (abdomen) is colored black to brown, with gray hair interspersed, creating a grizzled appearance. The shoulders (mantle) and sides of the neck are usually bright golden brown, but may be paler in some individuals. The head varies from brown to dark brown. The well-formed and rounded ears and large eyes give the face a canine appearance; members of the family Pteropodidae often are referred to as flying foxes.

The Mariana fruit bat is highly colonial, forming colonies of a few to over 800 animals (Wiles 1987a; Pierson and Rainey 1992; Worthington and Taisacan 1995). Bats group themselves into harems (1 male and 2 to 15 females) or bachelor groups (predominantly males), or reside as single males on the edge of the colony (Wiles 1987a). On Guam, the average estimated sex ratio in a single colony varied from 37.5 to 72.7 males per 100 females (Wiles 1982).

Reproduction is believed to occur throughout the year in Pteropus mariannus yapensis on Yap (Falanruw 1988). Mating and the presence of nursing Pteropus mariannus mariannus young have been observed year-round on Guam (Perez 1972; Wiles 1983) with no apparent peak in births (Wiles 1987a). Glass and Taisacan (1988) suggested a similar pattern on Rota, but also indicated that a peak birthing season may occur during May and June, as has been observed in other fruit bats (Pierson and Rainey 1992). Female bats of the family Pteropodidae have one offspring per year (Pierson and Rainey 1992), pups may be born in any month of the year. Observations on Guam between July 1982 and May 1985 found 262 female bats, each with a single young (Service 1990). This reproductive rate, very low for a mammal of this size, results in a low maximum population growth rate, and thus a slow rate of recovery when a population is diminished (Pierson and Rainey 1992). Length of gestation and age of sexual maturity are unknown for the Mariana fruit bat; other related bats have a gestation period of approximately 4.6 to 6.3 months (Pierson and Rainey 1992). Age of sexual maturity is not known for the Mariana fruit bat, but Pteropus species typically do not breed before 18 months of age (Pierson and Rainey 1992).

Taxonomy and Interisland Movements

The fruit bats of the Mariana Islands consistently have been treated as one or more endemic subspecies or species; that is, they occur nowhere outside the archipelago (Andersen 1912; Kuroda 1938; Corbet and Hill 1980, 1986, 1991; Koopman 1982, 1993; Flannery 1995). Following the taxonomic treatments of Kuroda (1938) and Koopman (1993), which are known to be based on examination of numerous specimens, and the most recent treatment by Flannery (1995), Pteropus mariannus is a widely dispersed species occurring north of the equator in portions of Micronesia north to the Japanese Ryukyu Islands. Various authors have attributed different numbers of subspecies to P. mariannus. Kuroda (1938) and Koopman (1982, 1993) recognize seven subspecies; Flannery recognizes three.

Pteropus fruit bats are well known to be strong fliers and traverse long distances (Eby 1991; Palmer and Woinarski 1999; Nelson 2003). Evidence that Mariana fruit bats fly between islands in the archipelago supports consideration of these bats as a single subspecies made up of numerous island populations in the Marianas (Lemke 1986; Service 1990; Wiles and Glass 1990; Worthington and Taisacan 1996). The geography of the archipelago, as well as the flight capability of fruit bats, facilitates interisland exchange. Distances between islands in the Mariana archipelago range from 3 to 62 mi (5 to 100 km). Each island in the chain is visible from neighboring islands (Wiles and Glass 1990).

The August 27, 1984, Federal listing (49 FR 33881) of fruit bats resident on Guam was based on an assumption that these bats were a distinct subspecies isolated from other bat populations in the CNMI. However, current evidence exists that large numbers of bats from Rota have visited Guam for periods of months. Temporary spikes in the Guam fruit bat population were observed in 1992-1993 (from about 350 to 550 bats) and in 1998 (from about 150 to 760 bats) (Anne Brooke, Service, in litt. 2003). These temporary increases lasted for several months. More modest but equally sudden increases in the Guam population were noted 2 and 4 days following Typhoons Chataan and Pongsona, respectively, in 2002 (Dustin Janecke, University of Guam, in litt. 2003). The most likely explanation is a temporary relocation of bats from Rota, which lies 48 mi (77 km) from Guam, is visible from Guam's north shore, and harbors one of the largest fruit bat populations in the archipelago. For example, the 2002 spike on Guam after Typhoon Pongsona was concurrent with an observed dip in fruit bat numbers on Rota (Jake Esselstyn, University of Kansas (formerly CNMI Department of Fish and Wildlife (DFW)), pers. comm. 2004b). Several other instances of apparent immigrations from Rota to Guam documented in the late 1970s and 1980s are described in detail by Wiles and Glass (1990). Although we cannot be certain that "visiting" bats interbreed with resident Guam bats during their months on the island, the fact that Mariana fruit bats breed throughout the year (Wiles 1983, 1987a) leaves this possibility open. The presence of fruit bats on the islands of Tinian and Aguiguan, which are close to one another and to Saipan, is ephemeral (Worthington and Taisacan 1996), indicating that interisland travel likely occurs among these three islands as well.

An example of likely interisland movement in the northern islands of the CNMI comes from Sarigan. Fruit bat surveys on Sarigan documented a roughly stable level of approximately 125-235 bats between 1983 and 2000 (Wiles et al. 1989; Fancy et al. 1999; Wiles and Johnson 2004). In 2001, surveys estimated 300-400 bats (Wiles and Johnson 2004). Recruitment of juvenile bats alone cannot account for this increase, and Wiles and Johnson (2004) posit Anatahan, 23 mi (37 km) to the south, as the likely source for immigrants. Wiles et al. (1989) twice observed individual fruit bats 0.8 mi (2 km) from Guguan, flying south in the direction of Sarigan, which lies 39 mi (63 km) away. Anecdotal observations of likely transits among other northern islands are described in Wiles and Glass (1990) and by other species experts (Worthington and Taisacan 1996; Wiles and Johnson 2004).

Like fruit bats, many other highly mobile vertebrates of Pacific Islands, especially birds, are treated as a single species or subspecies inhabiting multiple islands in an archipelago (Mayr 1945; Pratt et al. 1987; Watling 2001). Immigration rates of perhaps one individual per generation could be necessary for an island population to maintain genetic homogeneity with the populations on other islands (Mills and Allendorf 1996; Wang 2004; Gary McCracken, University of Tennessee, pers. comm. 2004). The chances of witnessing such a low rate of immigration are slight. The evidence described above for interisland movement suggests even greater rates of movement and probable gene flow among the fruit bat populations on various islands in the Mariana *1194 archipelago than the minimum needed to maintain genetic homogeneity.

Preliminary results of a recent study of genetic variation in a similarly gregarious (Pierson and Rainey 1992) and mobile species of fruit bat elsewhere in the Pacific provide further, if circumstantial, support for the existence of a single subspecies of fruit bats in the Marianas. Genetic material collected from the white-collared fruit bat (Pteropus tonganus) in Samoa and Fiji shows a lack of genetic isolation within island groups (Utzurrum et al. 2000; G. McCracken, pers. comm. 2004). Little anecdotal observation of interisland movements exists for P. tonganus, yet apparently it experiences immigration at sufficient intervals to prevent genetic isolation.

Currently, there are two recognized subspecies restricted to the Mariana Islands: the Mariana fruit bat (Pteropus mariannus mariannus) and the Pagan fruit bat (Pteropus mariannus paganensis). Other subspecies are endemic to other archipelagos and do not occur in the Marianas. The taxonomic status of the Pagan fruit bat is questionable. Yamashina (1932) collected three male fruit bats and one female from the islands of Pagan and Alamagan in 1931, and stated: "[t]his species, as compared to the Pteropus mariannus mariannus that inhabit Guam, is distinctly darker in coloration, having brownish wings." He made no further comparisons, and thus the distinction of this taxon is based on a single, equivocal interpretation of the coloration of four specimens. Although future studies may confirm the existence of a distinct taxon of fruit bats in the northern islands, at this time, based on the best available science including peer reviewer comments, we do not consider Pteropus mariannus paganensis as distinct from Pteropus mariannus mariannus to represent a single taxon.

Habitat

Mariana fruit bats forage and roost primarily in native forest and forage occasionally in coconut (Cocos nucifera) groves and strand vegetation (Wiles 1987b; Worthington and Taisacan 1996). Wiles (1987b) described six bat roost sites on Guam, all within native limestone forest. Major roost trees included Ficus spp. and Neisosperma oppositifolia. On Rota, fruit bats used primary and secondary limestone forest for roosting and foraging (Glass and Taisacan 1988). At least nine tree species were used for roosting, including Elaeocarpus sphaericus, Macaranga thompsonii, Guamia mariannae, Hernandia spp., Artocarpus mariannensis, Ficus prolixia, Barringtonia asiatica, Randia cochinchinensis, and the introduced Theobroma cacao (Glass and Taisacan 1988). A small bat colony also was observed roosting in Casuarina equisetifolia on Aguiguan (Worthington and Taisacan 1996). At least 22 plant species are used as food sources by the Mariana fruit bat. Food items include the fruits of 17 species of plants, especially the native Artocarpus mariannensis, Cycas circinalis, Ficus spp., Pandanus tectorius, Terminalia catappa, and the introduced Artocarpus altilis and Carica papaya; the flowers of seven plants, including the native Ceiba pentandra and Erythrina variegata, and the introduced Cocos nucifera; and leaf stems and twig tips of Artocarpus spp. (Wiles 1987a; Service 1990). Although Mariana fruit bats have been observed to feed on and roost in cultivated, introduced food plants, nonnative species make up only a small fraction of the plants they use (Wiles 1987b; Worthington and Taisacan 1996). Fruit bats are important components of tropical forest ecosystems because they disperse plant seeds and thereby help maintain forest diversity and contribute to plant regeneration following typhoons and other catastrophic events (Cox et al. 1992).

CNMI Southern Islands

The relatively large size and moderate topography of the southern islands led to their being, along with Guam, the most heavily populated and intensively cultivated islands in the archipelago. All of the southern Marianas are hypothesized to have been densely forested when first settled by humans some 3,500 years ago (Mueller-Dombois and Fosberg 1998). The loss and alteration of native habitats on these islands began with prehistoric cultivation, accelerated with the 17th century introduction of livestock and mechanized agriculture by Europeans, and likely peaked during the mid-20th century with landscape-scale habitat conversion by commercial agriculture, military infrastructure, and bombardment (Bowers 1950; Fosberg 1960; Stone 1970). This long continuous and intense human disturbance is reflected by the near absence of Mariana fruit bats from Saipan, Tinian, and Guam.

On Saipan and Tinian, agriculture and free-roaming livestock had converted much of the islands' forest to fields and pastures as early as the 18th century (Barrat 1988 in Stinson et al. 1992). Human populations on these islands increased steadily, and virtually all arable land was used to grow cash crops or food (Bowers 1950). Sugar plantations dominated the landscapes of Saipan, Tinian, and Aguiguan prior to World War II (Fosberg 1960). Saipan and Tinian were invaded during World War II, and during and after the war, bombing and extensive military development resulted in the loss of additional fruit bat habitat (Bowers 1950; Fosberg 1960). After the war, Saipan and Tinian were estimated to retain 5 and 2 percent native forest cover, respectively (Bowers 1950), and these proportions apparently were not significantly different in 1982 (Engbring et al. 1986). The introduction of nonnative species such as tangantangan for erosion control has left these islands dominated by alien vegetation that inhibits the growth of native forest (Fosberg 1960; Craig 1993). Feral ungulates are present on both islands, resulting in further degradation and fragmentation. Finally, Saipan is the most heavily populated and industrialized island in the CNMI (CNMI Statistical Yearbook 2001). Aguiguan was not invaded during the war, and has retained a greater proportion of its native forest (20 percent; Bowers 1950).

Similar to Saipan and Tinian, large areas of Rota were converted to sugar plantations in the early part of the 20th century (Fosberg 1960). Rota has more rugged topography, however, and was not invaded during World War II. These two factors are thought to explain the greater amount of native forest cover (25 percent) remaining on Rota following the war (Baker 1946; Bowers 1950). Engbring et al. (1986) estimated that roughly 60 percent of Rota's land area supported native vegetation in 1982. It is not clear whether Engbring's estimate represents some level of native forest recovery since Bowers' (1950) post-war estimate, or is a different interpretation and measurement of forest cover.

Most of Guam's native vegetation has been replaced by land development and invasive species. Guam is the population and commercial center of the archipelago, and commercial and residential development are ongoing. Like the other southern islands, parts of Guam were seeded with tangantangan following World War II to control erosion (Fosberg 1960). Large areas of southern Guam are dominated by savannas; these landscapes are thought to have originated as a result of aboriginal burning (Fosberg 1960). In 1981, northern Guam, which supports the last extensive native forest remaining on the island, was thought to retain no more than 37 percent native forest cover (Engbring and Ramsey 1984). Feral ungulates are abundant and *1195 widespread throughout the island and cause significant damage to all remaining native forest (Fosberg 1960; Stone 1970; A. Brooke, pers. comm. 2004). Lands owned by the U.S. Air Force (Air Force) at Andersen Air Force Base in northern Guam include the largest contiguous forested areas left in northern Guam; the Air Force permits hunting of feral ungulates on parts of the base (U.S. Air Force 2001).

CNMI Northern Islands

Compared with the history of habitat loss in the southern islands, degradation or loss of native forest in the northern islands of the CNMI is a recent phenomenon; therefore, these islands have retained more habitat to support Mariana fruit bats. Some of the northern islands have supported small human settlements, and most of these have been occupied only sporadically. Feral ungulates have been present in the northern islands only since the mid-20th century. For example, Anatahan has had feral goats and pigs for roughly 40 years (Kessler 1997), and forest degradation and erosion were observed to escalate sharply during the 1990s (Marshall et al. 1995; Kessler 2000a; Worthington et al. 2001), possibly because feral ungulate damage was exacerbated by El Nino-related drought in the late 1990s (Kessler 2000a).

Although changes in forest cover were not quantified, evidence from point photo monitoring and other land-based photography conducted on Anatahan in 1983, 1996, and 2000 documented widespread loss of forest, reduced canopy cover in remaining forest, and increased erosion resulting from feral ungulate damage (Marshall et al. 1995; Kessler 1997, 2000a; Worthington et al. 2001). An ungulate eradication project was begun in 2002, but was not completed when Anatahan volcano erupted in 2003. This eruption further compromised the island's forest habitat, and continuing volcanic activity has hindered completion of the ungulate eradication project. A large population of feral pigs still occurs on the island and some goats remain; aerial hunting for goats is ongoing (Curt Kessler, Service, pers. comm. 2004b). Some vegetation recovery has been observed as a result of goat control, but an invasive alien vine, Mikania micrantha, has spread rapidly and may inhibit the growth of native vegetation (C. Kessler, pers. comm. 2004b). This plant is known to smother and displace native vegetation on other Pacific islands (U.S. Department of Agriculture (USDA) 2004).

On Pagan, livestock was maintained in captivity by island residents until the volcanic eruption in 1981, when the human population was evacuated. In the subsequent 23 years, large populations of feral goats, pigs, and cattle have become established on the island and have caused significant damage (Rice and Stinson 1992; Kessler 1997). The degradation and loss of native forest on Pagan is thought to be occurring more rapidly on there than on Anatahan because of the added impact of cattle, which are absent from Anatahan (Kessler 1997). The reductions in fruit bat numbers on Pagan are attributed to feral ungulates causing major damage to the native forest and preventing its regeneration following the 1981 eruption, large areas especially in the northern part of the island being converted to grassland or devegetated and eroded (Kessler 1997), and the spread of the invasive tree Casuarina equisetifolia in monotypic stands (Rice and Stinson 1992; Cruz et al. 2000e). In 1992, Casuarina coverage in the upland areas of the island was estimated at roughly 60 percent (Rice and Stinson 1992). Although this tree is used for roosting by Mariana fruit bats (C. Kessler, pers. comm. 2004b), it does not provide food resources, and it likely displaces native forest, as it has done elsewhere in the Pacific (Cruz et al. 2000e; USDA 2004).

Vegetation surveys in 2000 on Agrihan, the third-largest of the northern islands, documented damage from feral ungulates in the 30 to 40 percent of the island that supports forest habitat (Cruz et al. 2000f). The extremely steep and dissected topography of Agrihan is thought to restrict the distribution of feral ungulates as well as access by humans, and keep goats and pigs geographically separated (Rice et al. 1990; Rice and Stinson 1992), thereby protecting roost sites and sufficient forest habitat to support foraging fruit bats.

Feral goats, pigs, and cattle are present on Alamagan and the extent of native forest remaining on the island is limited to ravines on the south and west slopes and a small plateau in the center of the island (Wiles et al. 1989). Rice (1992) described Alamagan as having "one of the worst feral ungulate problems in the CNMI," and during vegetation surveys in 2000, Cruz et al. (2000b) found the remaining forests to be in decline.

Maug, Asuncion, Guguan, and (since 1998) Sarigan are free of feral ungulates, but the small size of these islands and the limited extent of their forest habitat ultimately limits the number of fruit bats they can support. Maug is only 10 to 14 percent forested (Wiles et al. 1989), and thus supports little habitat for fruit bats. Forest on Asuncion and Guguan is limited to the lower western and southern areas; the northern and steep upper parts of these islands are bare volcanic ash or grassland (Wiles et al. 1989). Roughly 32 percent or 400 acres (ac) (162 hectares (ha)) of Sarigan is forested, but most of this is monotypic coconut forest that provides only minimal forage for fruit bats; only about 72 ac (29 ha) supports relatively diverse native forest that provides both roosting and foraging resources for fruit bats (Wiles and Johnson 2004). Although the eradication of ungulates from Sarigan and initial vegetation recovery may play a role in increased numbers of fruit bats on the island, invasive, alien plants such as tangantangan (Leucaena leucocephala) and Operculina ventricosa also are present on the island and may impede the recovery of native forest over the long term (Kessler 2000b). These plants are known to degrade native vegetation in the Mariana Islands and elsewhere in the Pacific (USDA 2004).

Landownership of Fruit Bat Habitat in the Mariana Islands

Most of the known fruit bat roost sites in the Mariana Islands are located on public lands. On Guam, the single remaining roost and most fruit bat foraging habitat is found on U.S. military lands; some foraging habitat occurs on private lands and lands belonging to the Government of Guam (Wiles 1998). The Air Force controls access to Andersen Air Force Base in northern Guam, and the high security and frequent patrols practiced on base effectively create a refugium for fruit bats (Morton 1996). The remote and relatively pristine area where the roost is located was set aside by the military in 1973 as a research natural area; access to and activities in this area are tightly restricted, but no brown treesnake control currently takes place specifically at the roost site (Air Force 2001). Service and Government of Guam wildlife biologists and authorized researchers are permitted access to the area and to the colony to monitor and conduct research on fruit bats. Similarly, the U.S. Navy (Navy) and the Service restrict access to their lands, which include native forest that provides foraging habitat for the fruit bat.

The remaining roost site is managed as part of the Guam National Wildlife Refuge (Refuge) overlay under a cooperative agreement with the Air Force. The Refuge was created on October 1, 1993, with additional lands (overlay portion) incorporated in 1994 by cooperative agreements between the *1196 Service, the Air Force and the Navy. The establishment and management of the overlay portion of the Refuge on Navy and Air Force lands provides a commitment by the three agencies to develop coordinated programs centered on the protection of endangered and threatened species and other native flora and fauna. Active implementation of such programs by these agencies contributes to the continued survival of the Mariana fruit bat on Guam, as important foraging and roosting habitat is located within the Refuge boundaries. However, the lack of brown treesnake control in the immediate area where the fruit bats roost is a serious deficiency in existing programs to protect endangered species on the overlay refuge.

There is no U.S. Government-owned land in the CNMI, but the Navy leases Farallon de Medinilla and part of Tinian. All other public lands are administered by the CNMI government. Saipan has little public land that is not leased and developed, but a few areas still support native forest that is occasionally used by fruit bats. Tinian has large tracts of public land that contain small stands of native forest suitable for bats, and a large portion of public land on the northern end of the island is under lease to the Navy for military activities (Lusk et al. 1997). All of Aguiguan is owned by the CNMI government. Approximately 60 percent of the land on Rota is publicly owned, although much of this has been leased to private individuals. The primary roosting areas on Rota are on Commonwealth lands, but some private lands still retain native limestone forest that may support fruit bats. The northern islands are mostly public lands, with some land developed as small homestead lots.

Population Surveys and Status

Obtaining accurate estimates of fruit bat populations in Pacific archipelagos depends on regular monitoring, standardized survey methods, and consideration of the unique ecology and physiographic environment of bat populations in various island groups (Utzurrum et al. 2004). The difficult terrain of the Mariana Islands, remote location of the northern islands of the CNMI, and the high costs associated with transits of the island group by sea and aerial surveys of individual islands have hindered the establishment of a standard monitoring program for the archipelago.

No known historical records exist to document the status of the Mariana fruit bat prior to the 20th century. The history of fruit bat surveys and changes in numbers summarized below represent a variety of methods and analyses. Archipelago-wide surveys were conducted in 1983 (Wiles et al. 1989) and 2001 (Johnson 2001).

The relatively isolated northern islands support the majority of the fruit bats in the archipelago, but because of their remote location, these islands have not been surveyed as frequently as the southern islands. Individual surveys have been conducted on several of the southern islands at relatively frequent intervals, and comprehensive surveys of the northern islands were conducted in 1983, 2000, and 2001 (Wiles et al. 1989; Cruz et al. 2000a-f; Johnson 2001). Opportunistic surveys have also occurred sporadically throughout the archipelago. The methods used in the northern islands in 2001 were significantly different from those used in 1983 and 2000; we therefore consider only Wiles et al. (1989) and Cruz et al. (2000a-f) for purposes of comparison (Table 1). A conservative interpretation of this comparison indicates a decline between 1983 and 2000, especially on the two islands that supported the largest numbers of fruit bats in the archipelago 20 years ago (Table 1).

Two of the northern islands are not included in this table: Uracas, the most northerly, where fruit bats are not known to occur; and Farallon de Medinilla, where fruit bats have been observed on only one occasion. See text and Table 2 for information about additional and more recent surveys and observations of fruit bats on the southern islands of the CNMI and Guam, and on Farallon de Medinilla, Anatahan, Sarigan, and Pagan.

   Table 1.--Summary of Mariana Fruit Bat Survey Results: Minimum Estimates    

-------------------------------------------------------------------------------

       Island                 Area              1983 [FN1]        2000 [FN2]   

                         Sq. mi (Sq. km)                                       

-------------------------------------------------------------------------------

Maug ................ 0.8 (2.0) ............ <25 ............. ( [FN3])        

Asuncion ............ 2.9 (7.4) ......................... 400  ( [FN3])        

Agrihan ............. 18.3 (47.4) ..................... 1,000 ........... 1,000

Pagan ............... 18.4 (47.7) ..................... 2,500 ........... 1,500

Alamagan ............ 4.3 (11.0) .......................... 0 ............. 200

Guguan .............. 1.5 (4.0) ......................... 400 ............. 350

Sarigan ............. 1.9 (5.0) ......................... 125 ............. 200

Anatahan ............ 12.5 (32.3) ..................... 3,000 ........... 1,000

Total (Northern                                                                

  Islands) ...........................................  7,450                  

[Total six islands] ........................  [7,025] ................... 4,250

Saipan .............. 47.5 (122.9) ......... <50 ............. ( [FN3])        

Tinian .............. 39.3 (101.8) ......... <25 ............. ( [FN3])        

Aguiguan ............ 2.7 (7.0) ............ <10 ............. 150-200         

Rota ................ 37.0 (95.7) .......... 800-1,000 ....... ( [FN3])        

Guam ................ 212.0 (549.0) ........ 425-500 ......... ( [FN3])        

Total (All Islands) ........................  8,760-9,035 .... N/A             

FN1 Wiles et al. 1989. Dates: August 17-September 10, 1983; 1-4 days/island.   

  Count methods: Evening dispersal counts at colonies; evening station counts  

  of solitary fruit bats.                                                      

FN2 Cruz et al. 2000a-f. Dates: June 4-August 16, 2000; 7-9 days/island. Count 

  methods: Evening dispersal counts at colonies, evening and morning station   

  counts of solitary fruit bats.                                               

FN3 Not surveyed.                                                              

-------------------------------------------------------------------------------

*1197 Status of CNMI Southern Islands

Fruit bats on the southern islands of the CNMI, Tinian, Saipan, Aguiguan, and Rota were not surveyed prior to the 1970s, but historical accounts indicate that fruit bats once were much more common on these islands than they are now. Schnee (1911) reported that bats were commonly seen and heard on Saipan, where they were heavily hunted by local residents. The Navy restricted civilian access to the northern part of Saipan until the early 1970s, effectively providing fruit bats with protected roost sites. The fruit bat population on Saipan was observed to decline rapidly after the Navy turned over the control to the CNMI government and access to the whole island became unrestricted (Wiles et al. 1989). Observations during the 1980s and 1990s suggested that the Saipan population was small; typically fewer than 50 bats were observed (Lemke 1984; Wiles et al. 1989; Wiles 1996; Worthington and Taisacan 1996). Surveys on Saipan in 2001 estimated that roughly 50 bats were present (Johnson 2001).

Fritz (1901) reported a large number of bats on Tinian in 1900 and Fritz (1904) reported that bats were common on all the southern islands. Fruit bats are only occasionally seen on Tinian today (Marshall et al. 1995; Krueger and O'Daniel 1999; Johnson 2001). Observations during the 1990s suggested that the presence of bats on Tinian was intermittent and their numbers were low (Lemke 1984; Wiles 1996; Worthington and Taisacan 1996). Surveys on Tinian conducted in 2001 found no fruit bats (Johnson 2001). In 1995, between 100 and 125 bats were believed present on Aguiguan (Wiles 1996). During a 10-day visit in 2003, however, no fruit bat colonies were observed on Aguiguan despite extensive coverage, and only a few individual fruit bats were seen (J. Esselstyn, pers. comm. 2004a).

The fruit bats on Rota have been surveyed on a regular basis by a large number of workers since 1986, using methods described by Stinson et al. (1992): primarily evening dispersal counts (EDCs), with some station counts of solitary or extracolonial bats and direct counts of colonial roosts (Glass and Taisacan 1988; Stinson et al. 1992; Worthington and Taisacan 1995, 1996; Johnson 2001; J. Esselstyn in litt. 2003, pers. comm. 2004a). This monitoring effort has yielded numbers that vary widely both intra- and interannually (e.g., Glass and Taisacan 1988; Worthington and Taisacan 1995, 1996). Analysis of the census data on Rota is underway (Laura Williams, CNMI DFW, pers. comm. 2004).

Fruit bat numbers declined following Typhoon Roy in 1988 from an estimated 2,400 animals to just under 1,000 (Worthington and Taisacan 1996). Prior to Typhoon Pongsona in 2002, however, the Rota bat population had risen back to approximately 2,500 (J. Esselstyn, in litt. 2003). In the months following the storm, repeated surveys indicated that numbers had again declined sharply to about 600 (J. Esselstyn, pers. comm. 2004b). Continued surveys of Rota's fruit bats indicate that the population was once again rising in 2004; in April it was estimated at roughly 1,500 animals (J. Esselstyn, pers. comm. 2004a, 2004b). The Rota population fluctuates and may be resilient, but severe storms at short intervals could erode this resilience. The most recent available estimate of fruit bat numbers on Rota is 1,100 (C. Kessler, pers. comm. 2004b). This estimate was made in May 2004, prior to Typhoon Chaba. The bats from Rota are believed to move among the southern islands, and this population thus is considered to be important to the long-term stability of fruit bats in the southern islands of the Mariana archipelago (Wiles and Glass 1990), and to the existence of the colony on Guam (Catherine Leberer, Guam Division of Aquatic and Wildlife Resources (DAWR), in litt. 2004).

Status of CNMI Northern Islands

The 1983 survey of the northern islands resulted in an estimate of 7,450 bats for Anatahan, Sarigan, Guguan, Alamagan, Pagan, Agrihan, Asuncion, and Maug (Wiles et al. 1989, Tables 1 and 2). Because field observation of Mariana fruit bats indicate that this species is gregarious and typically roosts in large colonies during the day, this and subsequent surveys focused on locating colonies. Wiles et al. (1989) located colonies by circumnavigating islands by boat, traversing portions of each island on foot, and interviewing residents on islands with human inhabitants. EDCs were conducted at each colony beginning at 1 to 3 hours before nightfall and continuing until complete darkness. These surveys were carried out by observers placed so that fruit bats departing the colony were silhouetted against the sky or the ocean. Rates of fruit bat departure from colonies were observed to be greatest between 10 and 40 minutes after sunset, but because departures continued after darkness when they are difficult to see, EDCs represent minimum counts (Wiles et al. 1989). In addition, evening counts of solitary or extra-colonial bats were made from vantage points determined to overlap least with the apparent dispersal trajectory of colony bats. Islandwide estimates were based on the number of fruit bats recorded, island size, extent of forest cover and abundance and diversity of food-plant species (Wiles et al. 1989).

Surveys of the northern islands undertaken in 2000 (Cruz et al. 2000a-f) employed a combination of the same methods used by Wiles et al. (1989) in 1983 and, on Anatahan, by Worthington et al. (2001) in 1995: land- and sea-based colony searches, EDCs, station-counts of extra-colonial bats, and direct day-time counts at roosts. On each island they visited, Cruz et al. (2000a-f) spent periods conducting fruit bat surveys equal to or greater than periods spent by Wiles et al. (1989) on the same six islands. The individual island-wide estimates of Cruz et al. (2000a-f) thus are comparable to those of Wiles et al. (1989), but owing to logistical and fiscal constraints, Cruz et al. (2000a-f) did not visit Asuncion and Maug. The 2000 surveys yielded an estimate of 4,450 fruit bats for the 6 northern islands they visited (Cruz et al. 2000a-f). The 1983 surveys yielded an estimate of 7,025 fruit bats for the same six islands (Wiles et al. 1989). A conservative interpretation of these data indicates a 37 percent decline in fruit bat numbers between 1983 and 2000 among these six northern islands.

The majority of this decline was recorded on two of the three largest northern islands, Anatahan (12.5 square mi (32.3 square km)) and Pagan (18.4 square mi (47.7 square km)), which together harbored roughly 70 percent of the archipelago's fruit bats in the 1980s (Wiles et al. 1989). These two islands, which were estimated to support a total of 5,500 fruit bats in 1983, were estimated to have only 2,500 fruit bats in 2000; approximately a 45 percent decline since 1983 (Cruz et al. 2000d, 2000e). These declines may be related to severe habitat damage caused by feral ungulates (Cruz et al. 2000d, 2000e; Kessler 2000a; see discussion in Background, Habitat section).

On Anatahan, surveys identified about 3,000 fruit bats in 1983 (Wiles et al. 1989), 1,902-2,136 individuals in 1995 (Marshall et al. 1995; Worthington et al. 2001), and roughly 1,000 in 2000 (Cruz et al. 2000d; Kessler 2000a). In conjunction with the ungulate eradication project, fruit bats on Anatahan have been surveyed frequently since 2002. Aerial (helicopter) surveys were conducted in May 2002; February, March, April, August, October, and December 2003; and January, February, March, July, and September 2004. These surveys are *1198 performed over 2 days, with 4 hours spent over the island each day. Coverage of the island during each survey is complete. Fruit bat colonies are rapidly reconnoitered to verify known roost sites and identify new ones, colonies are counted and mapped, and individual bats in flight also are counted. After the volcanic eruption in May 2003, the island's state of devegetation facilitated accurate location of all colonies (C. Kessler, in litt. 2003, pers. comm. 2004c). In 2002 and early 2003, estimates of the island's bat population ranged from 950 to 1,250 (C. Kessler, in litt. 2003). Following Anatahan's volcanic eruption in May 2003, aerial surveys conducted in August, October, and December of 2003 yielded estimates of 350-700 bats, and in January and February of 2004, bat numbers were estimated at 500-600 and 550- 650, respectively (C. Kessler, in litt. 2003, pers. comm. 2004c). Surveys in March, July, and September of 2004 yielded increased estimates of about 1,000- 1,200 bats (C. Kessler, pers. comm. 2004c). This localized increase in fruit bat numbers over a short period of time (1 to 1.5 years) was concomitant with some vegetation recovery, and indicates that Anatahan's population may have reached its pre-eruption level, whether the source of the additional bats is immigration, recruitment of newly volant (flying) young, or both (see Summary of Factors Affecting the Species section).

On Pagan, fruit bat numbers were estimated at 2,500 in 1983 (Wiles et al. 1983), and at roughly 1,500 in 1999 and 2000 (Cruz et al. 2000e). On the third-largest northern island, Agrihan (18.3 square mi (mi [FN2]) (47.4 square km (km [FN2])), results of surveys in 1983 and 2000 indicate that fruit bat numbers have been stable at about 1,000 individuals (Wiles et al. 1989; Cruz et al. 2000f).

The remaining northern islands with fruit bat populations, Maug, Asuncion, Alamagan, Guguan, and Sarigan, all are less than 5 square mi (13 square km) (Table 1), and harbor from 100 to 500 bats (Cruz et al. 2000a, b, c). Sarigan, the next island north of Anatahan, has been surveyed more frequently in recent years in conjunction with the ungulate eradication there. A 1997 survey of Sarigan estimated the population at 170 fruit bats, and a 1999 survey resulted in an estimate of 150-200 individuals (Wiles 1999). Surveys between 1983 and 2000 on Sarigan estimated populations of approximately 125-235 bats (Wiles et al. 1989; Fancy et al. 1999; Wiles and Johnson 2004). In 2001, surveys estimated 300-400 bats (Wiles and Johnson 2004). The observed increase on Sarigan may reflect a response to the recovery of forest vegetation after the eradication of feral goats and pigs from the island in 1998 (Zoology Unlimited 1998). As described above in the discussion of interislands movements, the increase in 2001 may also reflect immigration to Sarigan from Anatahan, 23 mi (37 km) to the south, as well as recruitment of newly volant young (Wiles and Johnson 2004). The potential for increase in fruit bat numbers on Sarigan is thought to be limited, however, by the island's small size (1.9 mi [FN2] (4.9 km [FN2])), the small extent of forest habitat (as described above, in the Habitat section), and the prevalence of monotypic stands of coconut, which provide only minimal forage habitat for fruit bats (Wiles and Johnson 2004; G. Wiles, Washington Department of Fish and Wildlife (formerly CNMI DFW), pers. comm. 2004).

Guam

On Guam, the sighting of fruit bats was considered to be "not * * * uncommon" in the 1920s (Crampton 1921). However, by 1931, bats were uncommon on Guam, possibly because of the introduction of firearms (Coultas 1931). Woodside (1958) reported that in 1958, the Guam population was estimated to number no more than 3,000, although the method used to make this estimate is not known (Utzurrum et al. 2004). This estimate had dropped by an order of magnitude, to between 200 and 750 animals by 1995, in part because of predation by the introduced brown treesnake (Wiles et al. 1995; Wiles 1996). During 1998, bat populations on Guam varied from an estimated low of 210-245 to a high of 910-980 bats (Wiles 1998), and in 1999, bat numbers ranged from an estimated low of 199-235 to a high of 327-371 (Wiles 1999). The most recent surveys on Guam put the bat population at fewer than 100 individuals (D. Janecke, in litt. 2003; A. Brooke, in litt. 2003). Predation by brown treesnakes on non-volant young probably prevents recruitment of juvenile bats on Guam (Wiles et al. 1995; Wiles 1996; G. Wiles, in litt. 2003).

Previous Federal Action

The Mariana fruit bat (Pteropus mariannus mariannus) was listed as endangered in 1984 on Guam (49 FR 33881). It was listed as a subspecies found only on Guam. More recent research over the years since this subspecies was listed indicates that Pteropus mariannus mariannus is not a subspecies endemic only to Guam but the Guam population is part of a subspecies including populations of bats on other islands that interact with each other (movement between islands). We believe that it is appropriate to list these bat populations in Guam and CNMI as one subspecies (63 FR 14641).

All the bat populations on Guam and in the CNMI are facing a number of threats, with most populations declining. We published a proposed rule on March 26, 1998 to reclassify the Mariana fruit bat on Guam from endangered to threatened and list all the bat populations on Guam and other CNMI islands as one subspecies throughout its range as threatened (63 FR 14641, 69 FR 30277).

We proposed to list the subspecies as threatened because we wanted to: (1) Simplify actions and expenditures. We could affect a downlisting for the population on Guam with little or no additional time and expense in conjunction with proposing to list the subspecies throughout its range, instead of taking a separate action to downlist the population on Guam; and (2) acknowledge a change in taxonomy. When we originally listed the population on Guam, we believed it to be a separate subspecies endemic only to Guam with a declining population and significant threats to it which merited endangered status. However, by including the other populations in the listing, we are evaluating a larger number of bats with a wider distribution, although threats to each population remain. Hence, we proposed threatened status for the entire population, instead of having one population as endangered and the others as threatened.

In that proposed rule, we included a detailed history of Federal actions completed prior to the publication of the proposal. The public comment period closed on May 11, 1998 (63 FR 14641) and was reopened from May 29, 1998, through July 10, 1998 (63 FR 29367) to accommodate requests for public hearings. We designated critical habitat for the Mariana fruit bat on Guam in a final rule published in the Federal Register on October 28, 2004 (68 FR 62944). Pursuant to a settlement agreement approved by the U.S. District Court for the District of Hawaii on August 21, 2002, we must make a final listing decision on the Mariana fruit bat and submit the final rule to the Federal Register by December 31, 2004. See Center for Biological Diversity v. Norton, Civil No. 99-00603 (D. Haw.).

Summary of Comments and Recommendations

In the proposed rule published on March 26, 1998 (63 FR 14641), we requested that all interested parties submit written comments on the *1199 proposal. We also contacted appropriate Federal, Territorial, and Commonwealth agencies, scientific experts and organizations, and other interested parties and invited them to comment on the proposal. Newspaper notices were published in the Marianas Variety (Saipan, CNMI) and Pacific Daily News (Guam), inviting general public comment and attendance at public hearings. We held public hearings on June 24, 1998, on Saipan and June 25, 1998, on Rota.

We reopened the public comment period on May 27, 2004 (69 FR 30277), to permit additional public review. In order to address any additional comments received during the reopened comment period, and meet the court order to submit to the Federal Register a final listing decision for the Mariana fruit bat no later than December 31, 2004, we reopened the comment period for 30 days, until June 28, 2004. The reopened comment period (and associated notifications in local media and via direct mailing) gave interested parties additional time to consider the information in the proposed rule and provide comments and new information.

During the first comment period in 1998, we received 13 written comments, including those submitted at the public hearings. During the reopened comment period in 2004, we received four additional written comments, including one from a Government of Guam agency, and one from a CNMI government agency. Several individuals or groups submitted comments in both the original and the reopened comment periods, or during hearings and later in writing. Of those comments received in 1998, eight opposed listing in the CNMI, one opposed listing in the CNMI and opposed downlisting on Guam, one opposed downlisting on Guam, one opposed downlisting on Guam but was in favor of listing in the CNMI, and one supported listing in the CNMI. In addition to several private citizens, the CNMI Governor, Director of the DFW, Rota DLNR Resident Director, Rota Mayor, and CNMI Senator Thomas P. Villagomez all opposed the proposal. The Air Force supported listing the fruit bat as threatened throughout the archipelago, but also stated that reclassification from endangered to threatened on Guam would be "misleading and confusing to the public," and cited an article in the local press that misrepresented a temporary influx of fruit bats from Rota as an increase in the Guam population (Thomas Churan, Air Force, in litt. 1998; also see Issue 15, below). The Air Force also expressed its belief that the Mariana fruit bat is more susceptible to extirpation on Guam than in the CNMI because of the presence of the brown treesnake there, and recommended that the fruit bat retain its status as endangered on Guam (T. Churan, in litt. 1998). The Mariana Audubon Society supported listing all bats in the Mariana archipelago as endangered rather than threatened. Three of the four parties that submitted comments during the reopened comment period in 2004 supported the listing, including the DAWR. The CNMI DFW opposed the listing.

This final rule has been revised and updated to reflect the pertinent comments and information received during the comment periods. Comments of similar nature are grouped under a single issue. In addition, we considered and incorporated into the final rule all appropriate information obtained through the public comment period.

Peer Review

In 1998, in accordance with our peer review policy published on July 1, 1994 (59 FR 34270), we solicited opinions from four individuals who have expertise with the species and the geographic region where the species occurs, and are familiar with conservation biology principles. We received written comments from two experts and incorporated their information into the final rule. One peer reviewer described the threats posed to the bats on Guam by brown treesnake predation and habitat destruction by feral ungulates. This reviewer did not include any professional judgment about movement of bats between islands, but has published peer-reviewed literature containing information that supports interisland exchange. The other expert expressed agreement and knowledge that there is interisland exchange.

In 2004, we solicited additional scientific peer review of the proposed rule from eight specialists, including one of the two who provided peer review in 1998. Of these, five responded and provided additional factual information, including recent survey results, the impact of typhoons and illegal hunting on fruit bats in the southern islands, and recent genetic studies of other Pteropus species elsewhere in the Pacific. Reviewers also provided citations for literature, corrections on minor factual issues, and input on interpretation of the existing information.

One reviewer provided a synopsis of changes in fruit bat numbers over the past 10-20 years on individual islands in the archipelago and noted declines on Guam, Anatahan, and Pagan. This synopsis was based partly on the reviewer's own research and partly on the work of others. Based on 19 years of fruit bat research, surveys, and personal observations in the Mariana Islands while employed as a Senior Biologist with the Guam Division of Aquatic and Wildlife Resources, this reviewer (who also authored the original recovery plan for the Mariana fruit bat on Guam, agency reports, and numerous peer-reviewed research papers on the Mariana fruit bat (e.g., Wiles and Payne 1986; Wiles 1987a, b; Wiles et al. 1989; Wiles and Glass 1990; Wiles 1992; Wiles et al. 1995; Wiles and Johnson 2004) emphasized three major threats to Mariana fruit bats: illegal hunting (described as "chronic" on Rota), habitat destruction by feral ungulates, and brown treesnake predation. Another reviewer, a biologist who spent two years monitoring fruit bats on Rota and elsewhere in the CNMI for the CNMI DFW, provided specific information about firsthand observations and evidence of illegal hunting of fruit bats on Rota after Typhoon Pongsona, described reports received of numerous other illegal hunting, and provided survey information documenting post-typhoon decline in fruit bats on Rota and subsequent increase in numbers. Three reviewers, two of whom hold doctorates based on research on the biology and ecology of island fruit bats, and one of whom is currently conducting a graduate research project on fruit bats on Guam, expressed their professional opinions that anthropogenic disturbances such as illegal hunting and habitat loss are likely to be significant threats to the Mariana fruit bat, and that these disturbances are periodically exacerbated by severe storms.

Two reviewers cited their own observations and those of other workers that indicated likely interisland movements between Sarigan and Anatahan and between Rota and Guam, and another reviewer cited information collected by others indicating likely interisland movement in the archipelago. Three of the five reviewers provided information and professional opinion that supported our treating all fruit bats occurring in the Mariana archipelago as a single subspecies, Pteropus mariannus mariannus, as described in the proposed rule; the other two expressed concern about the possible occurrence of genetically isolated populations within the range of fruit bats in the Mariana Islands. Two reviewers expressed reservations about treating all fruit bats in the archipelago *1200 as one taxon without empirical data from genetic or radio-telemetry studies. However, one of these reviewers also described unpublished genetic research on fruit bats in Polynesia that indicates a lack of within-archipelago genetic structure in a widespread species that shares social and behavioral traits with the Mariana fruit bat.

Issue 1: The Service lacks adequate data to assess the population status of Mariana fruit bats. Comprehensive surveys are required to determine the status of Mariana fruit bats in the northern islands.

Our Response: In this case, we believe existing data are adequate to assess the overall status of the Mariana fruit bat. Subsequent to listing, two additional multi-island surveys of bats in the Mariana Islands have been conducted. One of these included six of the 10 northern islands (Cruz et al. 2000a-f) and yielded data comparable to those collected in 1983 by Wiles et al. (1989). The other conducted in 2001 (Johnson 2001) included all of the islands in the archipelago but employed methods that precluded direct comparison with other surveys. A conservative interpretation of these data indicate that bat numbers have declined on the two islands, which historically had large numbers of fruit bats in the archipelago.

Issue 2: The Service's evidence of bats moving between islands was inadequate or only anecdotal, and without empirical evidence of interisland movement, a determination that all fruit bats in the Mariana Islands belong to the same subspecies is premature. Fluctuations in bat numbers, particularly on Guam, may be caused by births.

Our Response: Evidence for the movement of bats between islands in the Mariana archipelago is discussed in the Background subsection above. The large fluctuations in the Guam bat population over a short period of time (Wiles 1998; A. Brooke, in litt. 2003) coupled with a low reproductive rate make it unlikely that changes in the Guam population reflect recruitment from births. Predation by brown treesnakes largely precludes the recruitment of young bats into the Guam population (Pierson and Rainey 1992; Wiles 1987a; G. Wiles in litt. 2003).

Issue 3: Long term survey data from Rota indicate natural fluctuations in fruit bat numbers on various timescales. Archipelago-wide surveys and the apparent decline they document may not account for these natural fluctuations.

Our Response: To date, we are aware of no analysis of survey data from Rota that: (1) Demonstrates a correlation between variation in fruit bat numbers and some other natural cycle, or (2) controls for the hunting and other human disturbance.

Issue 4: CNMI government agencies feel the Service overstated the illegal hunting problem, and stated that the CNMI DFW is instituting law enforcement reforms, and the CNMI government is committed to the enforcement of wildlife regulations. In contrast, most peer reviewers identified illegal hunting and lack of enforcement as a significant threat to the Mariana fruit bat, especially in the CNMI, and an official from Guam DAWR expressed concern that recruitment of immigrant bats to Guam is threatened by illegal hunting on Rota.

Our Response: We appreciate the CNMI DFW's commitment to law enforcement. We acknowledge that data on illegal hunting is difficult to obtain and assess, and that most of the information regarding illegal hunting is anecdotal. We have numerous documented observations and reports of illegal hunting incidents in the CNMI (e.g., Arnold Palacios, CNMI DWF, in litt. 1990; T. Eckhardt, Service, in litt. 1998; J. Esselstyn, pers. comm. 2004a; C. Kessler, pers. comm. 2004a). We address the threat to the Mariana fruit bats from illegal hunting in Factor B in the Summary of Factors Affecting the Species section.

Issue 5: The Service was selective in its presentation of the impacts of feral animals on Mariana fruit bats, presenting it in a poor light to justify listing. The Service did not consider the feral animal eradication project on Sarigan, and failed to note that the CNMI DFW has an existing federally funded program addressing feral animal damage (Feral Animal Monitoring and Management (Project No. W-1-R-1-11; Job number 2)).

Our Response: We have incorporated the results of the Sarigan Feral Animal Control Project (Zoology Unlimited 1998) into this final rule and discuss the threats posed to fruit bats by feral animals (see discussion in the Background section, and Factor A in the Summary of Factors Affecting the Species section). Although DFW's Feral Animal Monitoring and Management Program has included survey of feral animals on many of the northern islands and involvement in several other projects, current DFW projections indicate that sufficient funding will not be available to complete the eradication of feral ungulates from Anatahan, and lack of material support will prevent the implementation of plans for feral animal control in the CNMI (L. Williams, pers. comm. 2004).

Issue 6: Present CNMI Coastal Resources Management (CRM) and DLNR land use regulations adequately protect Mariana fruit bat habitat (limestone forest) from development, as exemplified by the modifications required for construction of the Rota Resort and Country Club. Habitat is also being protected through island-wide master planning and through implementation of habitat conservation plans (HCPs) on Saipan and Rota.

Our Response: We support the use of local land use regulations to promote the conservation of the Mariana fruit bat and its habitat. However, the best measure of their past effectiveness in protecting the Mariana fruit bat is the success of these regulations in maintaining the integrity of native limestone forest systems in the CNMI, particularly in the southern islands where development pressures are greatest. Direct and secondary effects of human activity continue to cause alteration of native forest areas despite these protections.

Through the Act's section 10 and HCP planning process, listed species may be lawfully taken and measures implemented to reduce activity impacts on the species and its habitat. Two HCPs are currently under development on CNMI and, if completed and implemented, should contribute to fruit bat conservation. The successful completion of these HCP projects in the CNMI is not sufficiently certain to consider them in making this listing decision. See our Policy for Evaluation of Conservation Efforts When Making Listing Decisions (PECE policy) (68 FR 15100, March 28, 2003).

Issue 7: The Service did not account for actions by the CNMI government to control the brown treesnake, thereby decreasing the threat of this factor to the Mariana fruit bat.

Our Response: We recognize that ongoing actions on Guam, Saipan, Tinian, and Rota are important and reduce the threat of accidental introduction of the brown treesnake. The U.S. Department of the Interior (DOI) Office of Insular Affairs (OIA), U.S. Department of Defense (DOD), USDA Wildlife Services, Service, Government of Guam, CNMI, and State of Hawaii are working together regionally to control brown treesnakes, particularly around transport centers (OIA 1999). The OIA and DOD actively fund research into methods of controlling snakes on Guam, in part to reduce the threat of introduction to other Pacific islands (OIA 1999). Both the CNMI DFW and Guam DAWR conduct brown treesnake public awareness educational campaigns *1201 consisting of school presentations, news releases, workshops, and poster/pamphlet distribution (Perry et al. 1996), and the CNMI maintains a snake reporting hotline (Nate Hawley, CNMI DFW, pers. comm. 2004a). In 1996, the CNMI became a signatory of the Memorandum of Agreement (MOA) between the governments of Hawaii, Guam, and the CNMI, and individual Federal government agencies concerned with brown treesnake eradication and control (DOI et al. 1993; DOI et al. 1996). This MOA commits the CNMI to a proactive brown treesnake program and allows the CNMI to apply for funding from the allotment of money appropriated by the U.S. Congress each year for brown treesnake control and eradication (OIA 1999).

Despite ongoing efforts, evidence exists that treesnakes are present on Saipan. A concrete barrier completed in 2004 at the commercial port on Saipan aids in the prevention of new introductions from Guam, but this barrier does not address the problem of the treesnakes already present on the island. The presence of brown treesnakes on Saipan poses a threat to the recovery of the fruit bat population there until the treesnakes are controlled throughout the island or are eradicated.

On Tinian, brown treesnakes, have been documented and are not thought to be established (Hawley 2002). The upcoming construction of a concrete snake barrier on Tinian will aid in the prevention of treesnake introductions to the island.

On Rota, two dead brown treesnakes were found in a cargo container in 1991, and in another, a live treesnake was sighted (N. Hawley, pers. comm. 2004a). The fence surrounding Rota's port was retrofitted with a snake barrier subsequent to the discovery of the two dead treesnakes, but damage and maintenance difficulties have resulted in deterioration of the barrier, and it was disassembled in 2002 (Gad Perry, U.S. Geological Survey-Biological Resource Division, in litt., 1998; N. Hawley, pers. comm. 2004b). CNMI DFW recommended replacing the fence with a concrete barrier around the cargo area; however, the barrier has not yet been constructed. These efforts were considered in the Summary of Factors Affecting the Species section below.

Issue 8: Existing regulations of the CNMI government are satisfactory for protecting the Mariana fruit bat so Federal listing is not necessary. The Mariana fruit bat is listed as threatened or endangered by the CNMI, and the Service was incorrect in stating that the CNMI lifted the moratorium on hunting of Mariana fruit bats. Therefore, the threat of legalized hunting is non-existent.

Our Response: We acknowledge that the CNMI has regulations protecting the Mariana fruit bat, but we have concluded that these regulations either do not contain sufficient protections or have not been adequately enforced to protect bat populations (see Factor D below).

In the proposed rule, we stated that the moratorium on the taking of Mariana fruit bats on all islands (Public Law 5-21, September 1977) had been lifted. We based this on a memo from the CNMI Assistant Attorney General for DLNR to our Law Enforcement (LE) office on Guam which stated that the hunting moratorium was no longer in effect (Richard Folta, Office of the Governor, Guam, in litt. 1996). In a subsequent letter to the Service, the Assistant Attorney General stated that the previous communication had been in error, and that the moratorium was still in effect (R. Folta, in litt. 1996). This new information has been incorporated into this final rule.

Issue 9: Listing the bat will not improve law enforcement, due in part, to the resource limitations of the Service's Division of Law Enforcement. No Service LE personnel are stationed in the CNMI, so the Service will be unable to enforce Federal regulations associated with the listing.

Our Response: The Service does have a wildlife inspector stationed in the Marianas who provides some enforcement of regulations associated with the Act. Declines in illegal fruit bat imports to Guam and the CNMI have been associated with the presence of LE personnel stationed on Guam and efforts of LE personnel based in Honolulu (Sheeline 1991; George Phocas, Service, pers. comm. 2004). We work in cooperative partnerships with Territorial, Commonwealth, State, local, and Federal agencies to further our interdiction and enforcement efforts. In the Mariana Islands, Service personnel are presently assisted by local customs officers, conservation officers, and quarantine officials in the enforcement of the Act. It is important to note that the Act provides an additional set of enforcement tools for the protection of listed species than are currently available for the fruit bat in the CNMI.

Issue 10: The listing of the Mariana fruit bat in the CNMI may result in severe harassment to the species.

Our Response: There has been no evidence to suggest that harassment of fruit bats is likely to occur as a result of listing. We understand that hunting of fruit bats takes place on a regular basis in the CNMI despite their protection under CNMI law, but all of the information we have received indicates that this hunting is motivated by local tradition, not by malicious intent in response to CNMI laws and regulations. Whatever the motivations for harassment or illegal hunting of Mariana fruit bats, their listing under the Act can provide additional protection through the enforcement of Federal law. In sum, we believe that the protections afforded to Mariana fruit bats by their being listed as threatened throughout their range will aid in their conservation and recovery.

Issue 11: Increased funding to the CNMI for endangered species recovery is unlikely. Listing the bat as threatened instead of endangered has the potential to restrict funding opportunities to conduct research and management because the Service's funding system places higher priority on species designated as endangered as compared to those listed as threatened.

Our Response: Under their cooperative agreement with us, DFW can apply for funding under section 6 of the Act for projects specifically related to Mariana fruit bat conservation. We do not categorically assign higher priority for funding or recovery actions to species that are listed as endangered over those that are listed as threatened.

Issue 12: Protection for the Mariana fruit bat on Farallon de Medinilla should come from the Service through the consultation process under section 7 of the Act. Listing the Mariana fruit bat in the CNMI will provide no additional protection with regard to military activities.

Our Response: Prior to the publication of this final rule, the Mariana fruit bat was not federally listed in the CNMI. Federal agencies, therefore, have not been required to consult on the effects of their actions in the CNMI on the fruit bat. Conversely, 30 days after the publication of this rule, the Mariana fruit bat becomes federally listed as threatened in the CNMI and throughout its range, and Federal agencies will be responsible for consulting with us when their activities may affect the fruit bat on Farallon de Medinilla or other islands in the CNMI.

Issue 13: The Service misinterpreted the data and conclusions of Morton (1996) in stating that military aircraft training activities on Guam cause or create the potential for abandonment of roosting areas.

Our Response: Current air traffic patterns and volume do not pose a threat. There is the potential for roost abandonment if air traffic patterns or volume increase significantly (Morton 1996). Significant changes could *1202 include more frequent departures and arrivals, and larger or noisier aircraft.

Issue 14: The rule is politically motivated, biased, based on assumptions and broad, unsubstantiated statements, speculative observations, and anecdotal evidence.

Our Response: We used the best scientific information available in our determination to list the Mariana fruit bat as threatened in the CNMI and reclassify from endangered to threatened on Guam. Threats to the Mariana fruit bat are documented in the Summary of Factors Affecting the Species section of this final rule. We did not rely solely on anecdotal information in making a decision to list this species as threatened. The rule includes citation to more than 70 published references, more than 40 scientific reports prepared for government agencies and universities, and numerous personal communications from scientists and others knowledgeable about fruit bats and the Mariana Islands and/or closely involved in natural resources management in the archipelago. The anecdotal information we did use is consistent with the body of scientific reports.

Issue 15: Some commenters felt that listing the Mariana fruit bat in the CNMI is justified, but many thought that reclassifying the fruit bat from endangered to threatened on Guam, and listing the fruit bat as threatened rather than endangered in the CNMI, was incorrect. Some of these commenters believe that reclassifying the Mariana fruit bat on Guam has already sent the wrong message to the public because media reports have misinterpreted the proposal as evidence of recovery. Some also expressed concern that reclassification of the fruit bat on Guam could undermine conservation funding. They suggest that the Service either leave the Guam population listed as endangered, or list all bats in the Mariana Islands as endangered rather than threatened.

Our Response: We define an endangered species as one which is in danger of extinction throughout all or a significant portion of its range. Threatened species are defined as those which are likely to become endangered within the foreseeable future throughout all or a significant portion of their range. Because we consider the fruit bats on all individual islands in the Mariana archipelago as part of a single, archipelago-wide subspecies, Pteropus mariannus mariannus, we now are evaluating a larger number of bats with a more widespread distribution than was evaluated for the original listing in 1984, which included only the fruit bat population on Guam. Listing Pteropus mariannus mariannus as threatened throughout its range, including bats in both the CNMI and Guam, retains an appropriate level of protection for this bat on Guam while increasing overall protection to the Mariana fruit bat throughout the Mariana Islands, and it does not undermine potential funding for fruit bat conservation on Guam.

Issue 16: The Service did not properly take into account the cultural importance of the Mariana fruit bat in its listing decision. For example, some commenters suggested that information from the document "Cultural Significance of Pacific Fruit Bats (Pteropus) to the Chamorro People of Guam" (Sheeline 1991) should have been incorporated into the proposed rule.

Our Response: We incorporated information contained in Sheeline (1991) into this final rule in the section Summary of Factors Affecting the Species, subsection B.

Issue 17: If listing occurs, the people of the CNMI deserve the same consideration that the Federal government has given to Native Americans, such as Alaskan natives, through inclusion of a provision to provide for limited take of Mariana fruit bats for cultural use.

Our Response: We recognize the importance of traditional values to native cultures. This is reflected in our close collaboration with agencies in the CNMI to develop HCPs. However, the Act specifically exempts only Alaskan natives from the take prohibitions if such take is primarily for subsistence purposes and meets certain other conditions (16 U.S.C.§ 1539 (e)), but subsistence take by other groups is not exempted by the Act.

Issue 18: One commenter stated that disease is the cause of decline of Mariana fruit bats on Rota.

Our Response: We are unaware of any evidence of disease affecting populations of Mariana fruit bats on Rota or elsewhere in the Mariana Islands.

Issue 19: The Service should clear up taxonomic questions surrounding the Mariana fruit bat and determine exactly how many taxa inhabit the Mariana Islands before listing is considered. Several peer reviewers expressed concern about the taxonomic uncertainties within western Pacific Pteropus, and that there may be more than one taxon endemic to the Marianas.

Our Response: Both the proposed and final rules address taxonomic questions in detail (see the Background subsection under SUPPLEMENTARY INFORMATION). If new information such as results from genetic studies of fruit bats in the Mariana Islands indicate the presence of additional subspecies, we will take appropriate action.

Issue 20: One commenter disagreed with the Service's proposed determination that designation of critical habitat for the Mariana fruit bat would not be prudent because the identification of specific locations as critical habitat would lead to increased illegal hunting, and would thus increase the threats to the species.

Our Response: Since publication of the proposed rule in 1998, several key court decisions have given us new guidance on making our "not prudent" critical habitat determinations. Furthermore, we now have designated critical habitat for the Mariana fruit bat on Guam (69 FR 62944). We have reexamined the prudency of designating critical habitat for the Mariana fruit bat based on these considerations and now determine that such a designation would be prudent. Our reasoning is presented in the Critical Habitat section below.

Issue 21: Why is the Service concerning itself with a listing priority tier 3/4 activity when other species are in greater need of attention? The Service published the proposed rule based on fiscal and timing reasons rather than biological reasons.

Our Response: This final rule was prepared under the terms of a Federal court-approved settlement agreement that stipulated we submit a final listing determination for the Mariana fruit bat to the Federal Register no later than December 31, 2004 (Center for Biological Diversity v. Norton, Civil No. 99- 00603 (D. Haw.)).

Summary of Factors Affecting the Species

Section 4 of the Act and regulations (50 CFR part 424) promulgated to implement the listing provisions of the Act set forth the procedures for adding species to the Federal lists. A species may be determined to be an endangered or threatened species due to one or more of the five factors described in section 4(a)(1). These factors, and their application to the Mariana fruit bat (Pteropus mariannus mariannus) in the Mariana Islands are as follows:

A. The present or threatened destruction, modification, or curtailment of its habitat or range. Mariana fruit bats have been observed to feed on the fruits, flowers, and leaves of at least 22 plants, all but three of which are native to the Mariana Islands; fruit bats also have been documented to establish roosts primarily in mature *1203 native trees within landscapes dominated by native forest (Wiles 1983, 1987a). The Mariana fruit bat depends on native forest trees for food and colonial roost sites where mating, parturition, and other important social and biological functions take place. Although Mariana fruit bats have been observed to feed on cultivated food plants such as Artocarpus altilis and Carica papaya (Wiles 1987a), and have been observed to roost in Theobroma cacao (Glass and Taisacan 1988), nonnative plants make up a very small fraction of the resources used by the subspecies (Wiles 1987b; Worthington and Taisacan 1996) (see Habitat section above). The degradation and loss of native forest, therefore, deprives fruit bats of essential resources for survival and reproduction. The southern islands in the Mariana archipelago have lost most of their original native forest, primarily over several centuries of large-scale agriculture, growing human populations, economic development, and military activities (Bowers 1950; Fosberg 1960; see discussion). Few Mariana fruit bats occur today on Saipan, Tinian, and Guam, the islands that have sustained the greatest human disturbance and habitat loss.

Mariana fruit bats have evolved with, and are dependent for food and shelter on, trees and other plants that occur in native forests in the Mariana Islands. The degradation or loss of these forests is a key threat to the survival of this subspecies. The loss of native forests in the Marianas has various sources. The foraging of feral ungulates such as goats and pigs prevent forest regeneration because they eat ground-layer vegetation and seedlings of understory and canopy species; the rooting and stereotypical path-making of ungulates promote erosion and facilitate the invasion of native forests by alien plants (Marshall et al. 1995; Kessler 1997; Service 1998a,b). These invasive alien plants displace or smother native vegetation and prevent its regeneration (Kessler 2000b). In the southern islands of the CNMI and on Guam, where human influence has the longest continuous history, outright conversion of forests for agriculture or other development, as well as feral ungulates and alien plant species, historically has been a major source of loss of the Mariana fruit bat's forest habitat.

Throughout the archipelago, feral ungulates have caused severe damage to native forest vegetation by browsing directly on plants, causing erosion (Marshall et al. 1995; Kessler 1997; Service 1998a,b), and retarding forest growth and regeneration (Lemke 1992b). The remaining native forest habitat for fruit bats on many of these islands continues to be threatened by the fragmentation and degradation associated with feral ungulates. Mariana fruit bats are dependent on native plants for food and native forest for roost sites. Soil erosion and chronically retarded forest regeneration, the concomitant loss of native forests caused by the browsing and rooting of feral ungulates, and subsequent invasion by nonnative plant species, collectively represent a significant threat to fruit bats. These vegetation and landscape changes deprive the fruit bats of the native plant species on which they depend for food, shelter, and places to conduct their social activities. The diminished quality and extent of native forest thus leads to an associated reduction in the number of fruit bats that the remaining habitat is able to support. The northern islands, for the most part, have escaped the effects of millennia of continuous human settlement, WWII, and post war activities that caused extensive habitat loss and fragmentation of native forest habitat (see Table 2). However, the introduction of feral ungulates to some of these islands as recently as 40 years ago has resulted in rapid degradation and loss of native forest cover, notably on Anatahan and Pagan, two of the largest islands that have supported relatively large numbers of fruit bats (Kessler 1997, 2000a).

Island by Island Summary

Table 2 provides a synopsis of the numbers and status of fruit bats on each island in the archipelago.

     Table 2.--Island Summary of Factors Affecting the Mariana Fruit Bat.      

-------------------------------------------------------------------------------

                        [See text for full discussion]                         

-------------------------------------------------------------------------------

   Island         Area         Historical        Key current    Estimated fruit

                                 factors                              bat      

                Mi [FN2]                           factors        numbers and  

                   (km                                                         

                 [FN2])                                                        

                                                                    status     

-------------------------------------------------------------------------------

Guam ......... 212.0                                                           

                 (549.0) .. Hunting, habitat                                   

                              loss                                             

                              (development,                                    

                              agriculture,                                     

                              feral                                            

                              ungulates),                                      

                              brown                                            

                              treesnakes ..... Brown