There's something fishy going on here: a critique of the National Marine Fisheries Service's definition of species under the Endangered Species Act.

I. INTRODUCTION

Even after all of the controversy and litigation spawned by the federal Endangered Species Act of 1973 (ESA),(1) it is ironic that over twenty years after its enactment agencies which administer the Act lack a clear sense of what "species" they should protect. The National Marine Fisheries Service (NMFS), which together with the U.S. Fish and Wildlife Service (FWS) is charged with implementing the statute,(2) recently attempted to define populations which qualify for listing.(3) In a recent law review article,(4) attorneys with NMFS and the National Oceanic and Atmospheric Administration (NOAA) explained the agency's rationale for its definition of a "distinct population segment" which qualifies for listing under the ESA's definition of species.(5) After providing relevant background, this article argues against NMFS' approach to this issue, and criticizes the reasoning which Gleaves employs to justify the agency's notion of what constitutes a species eligible for ESA protection. It also explores how FWS and the federal courts have wrestled with the same question and examines a draft policy issued by FWS in 1992. The article concludes that FWS' suggested approach better comports with the current state of biological sciences and with the policy objectives Congress sought to further through the ESA

II. BACKGROUND

Although characterized as the "pit bull of federal environmental statutes,"(6) the ESA is as meek as a kitten unless an imperiled creature appears on the statute's lists of threatened and endangered species. The law offers virtually no substantive protections to organisms not listed,(7) a fact which focuses tremendous interest and controversy on the ESA's listing process.(8) In order to qualify for listing, FWS or NMFS must find that a group of organisms constitutes a species, and determine that the species is endangered or threatened. The ESA defines a species to include subspecies and "distinct population segments" of vertebrate fish and wildlife.(9) Neither the Act nor its implementing regulations define the terms subspecies or distinct population segment. The statute defines an endangered species as a species in danger of extinction throughout all or a significant portion of its range.(10) A threatened species is one likely to become endangered in the foreseeable future.(11)

For much of the ESA's history, providing protections to groupings below the species or subspecies levels was relatively uncontroversial.(12) The current list of protected species includes many populations of species which as a whole do not necessarily face extinction. These populations are usually defined by geographical occurrence.(13) For example, grizzly bears are listed as a threatened species in the contiguous United States, whereas grizzlies in Canada and Alaska are not listed.(14) However, controversy over ESA protection for runs of Pacific salmon in the Pacific Northwest prompted NMFS and FWS to reassess their authority to list distinct population segments. Historically, fisheries managers differentiated salmon populations based on run timing and geographical occurrence.(15) In 1990, Indian tribes and environmental organizations filed several petitions requesting listings of various salmon populations based on these historic management classifications.(16) These petitions, which almost overnight raised the issue of salmon conservation to a level of controversy approaching that surrounding efforts to protect northern spotted owls, prompted NMFS to closely examine its authority under the ESA to list groups of organisms other than entire subspecies and species.

NMFS' deliberations resulted in the agency formulating a salmon-specific definition of the phrase "distinct population segment" within the ESA's definition of species. According to NMFS, only an "evolutionarily significant unit" (ESU) of a broader salmonid species qualifies as a distinct population segment eligible for protection.(17) A population constitutes an ESU if it meets two criteria. First, a population "must be substantially reproductively isolated from other conspecific population units."(18) This criterion does not require that reproductive isolation be absolute, but the population's isolation must be "strong enough to permit evolutionarily important differences to accrue in different population units."(19) Evidence of such isolation, according to NMFS, can be gleaned from movements of tagged fish, recolonization rates of other populations, evaluations of the efficacy of natural barriers, and measurements of genetic differences between populations.(20) Second, a population must "represent an important component in the evolutionary legacy of the species."(21) NMF's characterizes a species' evolutionary legacy as "the genetic variability that is a product of past evolutionary events and which represents the reservoir upon which future evolutionary potential depends."(22) Thus, in order to satisfy the second element of the ESU test, a population must "contribute[] substantially to the ecological/genetic diversity of the species as a whole."(23) Indicators of such a contribution include genetically-driven physical characteristics and data from scientific tests which analyze genetic make-up.(24)

NMFS' ESU concept has significantly influenced the agency's listing decisions. While NMFS has found some salmon runs to qualify as ESUs, the agency has used this concept to reject ESA protection for other runs. For example, petitioners asked the agency to protect separately spring and summer runs of chinook salmon in the Snake River.(25) However, citing "the possibility of substantial levels of gene flow" between the two runs, NMFS lumped these stocks together as constituting a single ESU.(26) Finding that this ESU was experiencing an alarming decline, NMFS listed Snake River "spring/summer chinook" as a threatened species.(27) The agency also used its ESU concept to deny ESA protection to a winter steelhead run in Oregon's Illinois River. Citing insufficient data to demonstrate that these fish were genetically distinct from other nearby steelhead populations, NMFS threw out a petition to list the Illinois winter run.(28) More recently, utilities opposed to a pending petition to list mid-Columbia summer chinook acknowledged that numbers of summer chinook have declined, but argued strenuously against protecting this stock because of its alleged genetic similarities to the more numerous fall chinook run.(29)

Although NMFS published its definition of distinct population segment as a "policy" specific to Pacific salmon rather than as a regulation formally interpreting the ESA, the agency's ESU concept clearly carries potentially broad implications for all listings under the statute. It seems unlikely that NMFS would abandon its ESU policy when the agency makes listing decisions involving organisms other than salmon; NMFS drafted the policy and its technical justification in terms sufficiently broad to readily apply to vertebrates other than salmonids. NMFS' policy could also influence FWS, which itself has struggled with the scope of its listing authority in recent years.(30) Adding to the uncertainty over this issue, FWS in 1992 issued a draft policy interpreting distinct population segment in a manner dramatically different from NMFS' ESU approach, but days later made a significant listing decision on the basis of ESU concepts.(31) NMFS and FWS are currently attempting to formulate a joint definition of groupings eligible for ESA protection.(32) This issue has also generated tremendous interest in legal and scientific circles.(33) Ultimately, a decision on the meaning of distinct population segment may fall to Congress; resolution of this issue is widely mentioned as an integral part of the Act's pending reauthorization.

III. INTEGRATING POLICY AND BIOLOGY

Formulating and implementing policy involving biodiversity conservation presents Congress and administrative agencies with tremendous challenges. First, policymakers must consider and accurately employ biological concepts in order to devise sensible regulatory schemes. On the other hand, however, administrative agencies with a great deal of technical expertise must implement statutory directives in a manner which carries out lawmakers' policy aims. Further, lack of scientific certainty often substantially complicates both of these processes. The ESA's protection for distinct population segments serves as an excellent example of the steps--and difficulties--involved in integrating law and biology.

In giving NMFS and FWS authority to list groupings below the subspecies level, Congress was responding to both scientific and policy considerations. When they created the ESA, lawmakers stressed the importance of maintaining genetic diversity, and astutely recognized that populations can play a crucial role in the persistence and evolution of a species as a whole.(34) However, Congress later chose to afford agencies authority to protect only vertebrate populations.(35) Since no biological reasons exist to distinguish between vertebrates and invertebrates with respect to the role genetic variability and healthy populations play in maintaining an entire species, Congress clearly excluded invertebrates from the definition of distinct population segment as a matter of policy.(36)

Congress makes these types of policy choices as a matter of course, but faces significant problems in dealing with scientific issues. Humans' limited understanding of the planet's biosphere makes it devilishly difficult to integrate law and biology. Part of this difficulty stems from the differences in the processes of making laws and gaining scientific knowledge. While biologists' understanding of other species and ecosystems continually evolves hopefully increasing), federal law remains static until Congress finally changes it. Mindful of this problem, as well as of their limited scientific expertise, legislators have devised mechanisms which attempt to build into law a flexibility to respond to scientific advances. The ESA itself contains two of the most oft-used of these mechanisms. First, when faced with scientific uncertainty, Congress often employs broad terms whose meaning must be defined by agencies with expertise, and which can evolve with scientific understanding. The term "distinct population segment" arguably falls into this category. Another way federal lawmakers deal with uncertainty is by simply instructing the bureaucrats charged with carrying out policy to do so in light of the most up-to-date scientific understanding. In the ESA, Congress instructed NMFS and FWS to make listing decisions based on the "best scientific and commercial data available."(37) While this directive clearly applies to the Secretary's listing determinations in consideration of the five factors set forth in the statute,(38) it also arguably mandates that the Secretary define ambiguous terms affecting the listing process using the most up-to-date scientific knowledge.

Administrative agencies charged with implementing federal statutes must themselves walk a difficult path. While the staffs of agencies such as NMFS and FWS possess a great deal of expertise which allows these agencies to apply state-of-the-art science in interpreting statutory directives, they must not lose sight of the legislature's policy goals. Therefore, expert agencies must modify their implementation of a statute to keep up with scientific advances, but must remain focused on the same policies. In an ESA context, NMFS and FWS bear primary responsibility for carrying out Congress' policies concerning protection of endangered species in a manner which reflects current biological knowledge. To do so, these agencies must have a clear sense of the policy directions provided to them by Congress, as well as a complete and up-to-date understanding of the biological processes which affect species extinctions.

Given the challenges of integrating law and biology, an analysis of whether NMFS or FWS is properly discharging its duties under the ESA must evaluate whether the agency has accurately identified congressional policy, and whether the agency is carrying out that policy consistent with current biological knowledge. More specifically, the issues surrounding protection of populations break down into two fundamental questions. First, has NMFS accurately identified and accounted for lawmakers' policy objectives in authorizing protection of "distinct population segments" of vertebrate species? Second, does NMFS' interpretation of this term within the necessary policy framework reflect the current state of biological knowledge?

A. Congressional Policy Goals in Authorizing Listing of Populations

In defense of NMFS' ESU policy, Gleaves provides only a cursory answer to the first question above. While acknowledging that NMFS' interpretation of the meaning of distinct population segment "should reflect the purposes, policies, and intentions of Congress," Gleaves contends that NMFS has wide latitude to define this term because Congress "has not directly addressed or resolved this precise question," and the Act's legislative history "provides no specific guidance" and is of problematic" relevance.(39) Hence, they fall back upon Chevron deference to an administrative agency, arguing that NMFS' ESU policy has a rational basis and "appears consistent with the broad values and goals enunciated by Congress."(40)

Contrary to the position adopted by Gleaves, the ESA as well as its legislative history provide fairly clear indications of what Congress intended when it gave NMF's and FWS authority to list distinct population segments. Although Gleaves discusses many of the relevant statutory provisions and passages from the ESA's legislative history, that analysis fails to identify the common threads running through these sources which indicate the policies lawmakers sought to further by formulating a legal definition of "species" to include groupings below the subspecies level. These policies fall into four categories, each of which is discussed below.

1. Protection for Genetic Resources

Congress clearly intended the ESA to protect genetic resources. The House Report accompanying the original version of the Act in 1973 analogized these resources to books in a library which require protection from indiscriminate destruction.(41) The report also justified conservation of this resource on economic grounds, arguing that "[t]he value of this genetic heritage is, quite literary, incalculable."(42) Events over the past two decades have proved Congress' emphasis on preserving genetic resources to be wise. Rare species directly provide extensive benefits to humans. Biotechnology, for which genes are the raw materials, is a multibillion dollar industry with no limits in sight. Further, humans have thus far assessed only a tiny fraction of known species for their potential benefits.(43)

2. Protection of Ecosystem Health

The ESA seeks not only to protect listed species, but to also provide a means whereby the ecosystems upon which threatened and endangered species depend may be conserved. . . ."(44) The statute itself emphasizes the connection between protecting species and conserving ecosystems, noting that threatened and endangered species are of ecological value.(45) The ESA's legislative history further emphasizes Congress' desire to protect imperiled species because such species may play important roles within ecosystems. The Senate Report accompanying the ESA when it was passed in 1973 points out that "many of these animals perform vital biological services to maintain a |balance of nature' within their environments."(46) Similarly, the House of Representatives cited "the critical nature of the interrelationship of plants and animals between themselves and with their environment" as a reason to conserve threatened and endangered species.(47) In 1978, Congress again referred to "return[ing] the species to the point where they are viable components of their ecosystems" as a purpose of the ESA.(48) These pronouncements clearly indicate that Congress, through the ESA, sought to limit adverse impacts on the earth's ecosystems caused by human-induced elimination of important components of those ecosystems.

Current events once again have demonstrated Congress' foresight in establishing this policy direction through the ESA. The term "ecosystem management" has become almost a mantra for federal land managers and policy makers. At its fundamental level, this concept simply indicates a recognition that the well-being of a biotic community depends largely on the health and security of its members. Lawmakers integrated this idea into the ESA 20 years ago by providing protection to imperiled species, subspecies, and populations--in part to conserve the ecosystems which they inhabit.

3. Protection of Domestic Populations

Through the ESA, lawmakers also authorized a domestic orientation to species conservation.(49) Congress noted that the ESA marked a "significant shift" away from previous federal statutes, which protected only species facing "worldwide extinction."(50) In discussing the requirements for an effective bill, the Senate Report accompanying the ESA in 1973 asserted that "the bill must provide protection throughout the nation for animals which are either endangered or threatened. . . "(51) In its language reciting Congressional findings, the ESA itself emphasizes "the Nation's heritage in fish, wildlife, and plants," and notes that various species of fish, wildlife, and plants in the United States have been rendered extinct as a consequence of economic growth untempered by adequate concern and conservation. . . "(52) When passed in 1973, the ESA included new provisions designed to protect endangered species within the United States.(53) Congress also expressly pointed out that the ESA affords the Secretary authority to list the domestic population of a species whose principal range is in another country.(54) Lawmakers in 1979 reiterated their intent to allow NMFS and FWS to specifically conserve domestic populations, asserting that the U.S. population of an animal should not necessarily be permitted to become extinct simply because the animal is more abundant elsewhere in the world."(55)

In addition to allowing U.S. citizens, without crossing international borders, to observe, study, and enjoy species which otherwise may be wiped out in the U.S.,(56) Congress' policy objective of maintaining U.S. populations could have substantial economic benefits. Negotiations involving the international Biodiversity Convention signed in Rio in 1992 largely centered on access to, and compensation for, countries' genetic resources.(57) The Convention may require a country or company which seeks to make economic use of genetic material to first obtain permission from the country of origin to obtain the material, and also pay compensation for its use.(58) Accordingly, by protecting its domestic populations of species which may be more abundant elsewhere in the world, the U.S. maintains its ability to use these species' genetic resources without the need to gain permission from or compensate another country.

4. Management flexibility and Uncertainty

As Gleaves notes, the 1979 ESA amendments provide important clues to the meaning of a distinct population segment because Congress expressly considered this issue during the amendment process.(59) Earlier in 1979, the General Accounting Office (GAO) had released a report criticizing several aspects of the federal endangered species program. The report recommended that Congress eliminate agencies' ability to list populations, contending that this authority could lead to absurd results such as listing squirrels in a specific city park even though an abundance of squirrels lived elsewhere.(60) NMFS and FWS, on the other hand, strongly opposed this change. Citing the example of bald eagles--not listed in Alaska or Canada but protected as endangered in the lower forty-eight states--the agencies argued that GAO's recommendation, if adopted, would limit their ability to require the appropriate level of protection for a species based on its actual biological status."(61) Congress went along with NMF's and FWS, although it expressed a note of caution:

One of the weaknesses of the 1969 Act which was corrected in the

1973 amendments was the inability of the FWS to adopt different

management practices for healthy, threatened or endangered populations.

The committee agrees that there may be instances in which

FWS should provide for different levels of protection for populations

of the same species .... [L]isting of populations may be necessary

when the preponderance of evidence indicates that a species

faces a widespread threat, but conclusive data is available with

regard to only certain populations. Nonetheless, the committee is

aware of the great potential for abuse of this authority and expects

the FWS to use the ability to list populations sparingly and only

when the biological evidence indicates that such action is

warranted.(62)

Two important and related policy directions emerge from this discussion. First, Congress clearly intended to provide NMFS and FWS flexibility to manage populations of a biological species differently according to different degrees of threat faced by these populations. The Senate Report makes this particularly clear by linking the statutory definition of species to the ESA's threatened and endangered categories. The statute defines the term 'endangered species" to include 'any species in danger of becoming extinct throughout all or a significant part of its range," and includes similar language in the definition of threatened species.(63) These definitions, coupled with the 1979 Senate Report's reference to management tailored to a grouping's listing category, demonstrate that Congress intended to allow NMFS and FWS the flexibility to differentiate for listing purposes based on geographical occurrence and degree of threat, rather than requiring a finding of some sort of biological or genetic differences between populations.(64) The animals cited in the Senate Report, bald eagles, provide an example to support this assertion. There is no doubt that the lower forty-eight states make up a significant portion of eagles' range, and that the birds face the threat of extinction in that area. However, there is no indication that bald eagles in the Lower 48 are reproductively isolated or genetically or biologically different from eagles in Canada and Alaska.(65) Therefore, if Congress intended to require a finding of reproductive isolation coupled with some sort of physical or genetic differences in order for a population to qualify as a "species" for listing purposes, FWS would have to list all eagles (including those in Canada and Alaska) as threatened and endangered. However, FWS listed bald eagles only in the portion of their range in which they face threats of extinction, a regulatory approach which Congress sanctioned in 1979. Indeed, Congress expressly refused to adopt GAO's recommendation to eliminate authority to list populations in large part due to NMFS' and FWS' protests that such action would eliminate their ability to extend different degrees of protection to a species according to the threat faced by the species in different areas.

In addition to affording implementing agencies authority to adopt flexible management schemes based on degree of threat, Congress also intended to give NMFS and FWS flexibility to avoid problems of scientific uncertainty in conserving species. The 1979 Senate Report indicates Congress recognized that assessments of threats faced by species are fraught with uncertainty. Lawmakers viewed the Secretaries' authority to list portions of an entire species as a means to deal with this uncertainty, as well as a way to enhance agencies' regulatory flexibility. The 1979 Senate Report anticipates a situation likely to become increasingly common: biological evidence indicates that an entire species faces threats to its existence, but the evidence is not sufficient to support a decision to list the species. However, available information clearly indicates that certain identifiable populations are likely to become extinct. In such cases the Senate Report expressly provides for listing these populations. This action would be consistent with congressional intent for two reasons. First, it would give the species the benefit of the doubt by preventing scientific uncertainty surrounding the overall status of the species from standing in the way of protections for populations of the species known to face extinction.(66) Additionally, it affords FWS and NMFS flexibility to manage healthier portions of a species differently from populations known to be in trouble--the strategy expressly cited with approval by the 1979 Senate Report.(67)

Finally, while Congress clearly afforded NMFS and FWS flexibility to list populations by taking into account varying degrees of threat and scientific uncertainty, lawmakers perceived a "great potential for abuse" of this authority and ordered the agencies to use it "sparingly."(68) It is difficult to pin down precisely what limits Congress intended to place on agencies' flexibility to list populations. Lawmakers would likely not condone ESA protection for one of numerous small populations of a wide ranging species, since such a listing may resemble the GAO's city park/squirrel hypothetical. On the other hand, Congress clearly approved of listings based on geopolitical boundaries when the geographically-defined population inhabits a significant portion of the entire species' or subspecies' range, as in the case of bald eagles. The only guidance for agencies attempting to find the line between these two situations is the 1979 Senate Report's admonition that NMFS and FWS list populations "only when the biological evidence indicates that such action is warranted."(69)

B. Biological Factors Relevant to Protecting Populations

Although Gleaves sets forward a portion of the statutory language and legislative history relevant to determining congressional policy goals inherent in agencies' authority to list populations, his analysis falls to provide any basic background in current biological understandings of species, populations, interactions between these groupings, and these groupings' roles in ecosystems. Such information is critical in order to determine whether a given interpretation of distinct population segment, when put into effect on the ground through management decisions, will result in implementing Congress' policy directives. Gleaves falls into a trap that snares most lawyers by virtually ignoring the environmental side of the environmental law coin. Gleaves cites a book published in 1970 for a biological definition of the terms "species" and "subspecies,"(70) but provides little information on further biological developments in this area. This discussion of scientific concepts is analogous to explaining American politics in the 1990s by describing opposition to the war in Vietnam and the Watergate scandal. In fact, our understanding of biological concepts has increased dramatically since the ESA was enacted over two decades ago. Agencies charged with implementing Congress' policy goals through the ESA must take this evolution of knowledge into account when interpreting their authority under the Act. The following discussion briefly summarizes important scientific developments which should affect the manner in which agencies interpret the ESA.

While taxonomists of the past plugged organisms into neat hierarchical classification schemes, modem biologists question even the most fundamental taxonomic categories. One scientist recently observed that "there is no agreement on what species are, how they should be delimited, or what they represent."(71) Given such uncertainties about concepts as fundamental as species, it does not come as a surprise that there is little scientific consensus about how to define groupings of organisms such as subspecies and populations.

While becoming less certain of the boundaries of traditional taxonomic categories, biologists in recent years have learned a great deal about dispersal and reproductive interactions among populations, and the importance of these interactions to the survival of species. Conservation biologists label these interactions "metapopulation dynamics."(72) A metapopulation is a series of populations of the same species connected by dispersal of individuals from one population to another. Spacial organization and dispersal patterns among a metapopulation's constituent populations may exert a crucial influence over persistence of the species. For example, some species inhabit patches of habitat that "move" over the landscape in response to natural processes such as fire and plant succession.(73) The interacting populations of these species play a crucial role in the species' ability to persist and maintain genetic diversity in the face of common localized extinctions. Other species, including Pacific salmon, have spatially extensive metapopulation structures consisting of a network of locally-adapted populations.(74) When regional disturbances such as earthquakes or volcanism wipe out a population, surviving populations nearby rapidly recolonize the vacant habitat. This distribution and interaction of salmon populations have enabled the various salmon species to persist for thousands of years in a very dynamic habitat prone to catastrophic changes.(75)

Genetic exchange among populations may also play an important role in securing a species' future. Genetic diversity is important to both a species' short-term survival and long-term capability to respond to environmental changes.(76) A positive correlation exists between individuals' genetic diversity and the number of individuals which can exchange genes through reproduction. Even low levels of genetic exchange among populations can help maintain the genetic diversity of individual members of a species. Unfortunately, however, human-caused habitat fragmentation has led to geographical fragmentation of many species, breaking the bonds of genetic exchange among populations and decreasing the genetic diversity of individuals within those populations. Conservation efforts for many species thus may require attempts to reestablish the genetic connectedness between populations.(77)

In addition to learning a great deal about dynamics within species, biologists have gained new insights into the roles species play in maintaining the overall health of a biotic community. Aldo Leopold was aware of such relationships when he wrote that "[t]o keep every cog and wheel is the first precaution of intelligent tinkering."(78) Subsequent research has borne out the wisdom of this advice. So-called "keystone species" provide the most dramatic example of individual species contributing to ecosystem stability. A keystone species exerts influence over its ecological community out of proportion with its abundance. For example, less than one percent of the plant diversity in Manu National Park in Peru--those plants and trees which produce fruit during the dry season--sustains nearly the entire fruit-eating animal community for three months of the year.(79) Eliminating these plant species could therefore decimate the park ecosystem. Like dry season fruitbearers, sea otters play a critical role in maintaining balance in their environments. Comparing two similar islands and their near shore environments, researchers discovered that without sea otters to prey on urchins, the exploding urchin population destroys much of the kelp beds offshore, greatly decreasing available food and habitat for many other species and even causing accelerated coastal erosion as kelp no longer buffers the shore from ocean waves.(80)

As is usual in science, however, biologists in recent years have also learned that there is much they do not know. In addition to uncertainty about the relevance of traditional taxonomic classifications, biologists have identified areas of uncertainty in virtually all aspects of how organisms interact and behave. Scientists also readily admit that their abilities to observe, measure, and evaluate the role of diversity on all levels--genetic, species, and ecosystems--are often crude and fraught with uncertainty.

IV. EVALUATION OF NMFS' ESU POLICY

Without attempting to identify lawmakers' policy aims in providing NMFS and FWS discretion to list populations, nor attempting to evaluate whether NMFS' ESU policy is consistent with recent scientific advances, Gleaves simply defends NMFS' characterization of the term "distinct population segment" based on deference to an administrative agency.(81) However, as discussed below, NMFS' ESU concept fails to advance several of the policy objectives Congress sought to further through the ESA. Moreover, recent scientific developments in the field of species conservation suggest that NMFS failed to interpret its listing duties under the ESA in light of the best scientific information available. This shortcoming contributes to the ESU policy's inadequate implementation of congressional policy directives.

A. Protection for Genetic Resources

NMFS clearly based its ESU policy almost exclusively on one of the policy goals Congress sought to further through the ESA, that of conserving genetic diversity. The NMFE' technical paper setting forth the ESU concept asserts that "[a] review of legislative history indicates that a major motivating factor behind the Act was the desire to preserve genetic variability, both between and within species."(82) The paper does not mention any other policy objectives which are either explicitly identified or inherent in the statute.(83)

The ESU policy only partially succeeds in carrying out Congress' desire to protect genetic variability. This interpretation classifies as a distinct population segment eligible for listing groupings within a species which are genetically distinct from other groupings. The ESU policy thus allows NMFS to protect populations based on their unique genetic makeup, thus conserving a resource potentially valuable to humans--a resource Congress likened to unread books in a library. However, the genetic diversity within populations eligible for listing under the ESU policy will not contribute to the persistence of a species or subspecies as a whole. NFMS' ESU criteria demand that populations exhibiting unique genetic characteristics also be deemed reproductively isolated from other populations of the same species. Accordingly, the genetic variability within an ESU does not, by definition, contribute to the genetic diversity of individuals throughout a species, and thus does not contribute to the fitness or evolutionary capacity of the entire species. Given enough time, all isolated populations will eventually evolve into new species, assuming of course that the population does not first become extinct or reestablish genetic interaction with other populations of the same species.(84) The ESU policy allows NMFS to protect this long-term process of speciation, but does so at the expense of preserving genetically connected populations which can contribute to the genetic diversity and hence persistence of an entire species.(85)

NMFS' emphasis on reproductive isolation and genetic uniqueness also inhibits its ability to protect species which exist as metapopulations. Interacting populations of many species are not reproductively isolated and may not possess unique genetic traits, but nevertheless contribute substantially to success of the metapopulation as a whole. For example, as discussed earlier, seemingly redundant salmon stocks likely form the population structure which allows salmon species to persist in a changing environment.(86) NMFS' exclusive focus on reproductive isolation and genetic differentiation thus may cause the agency to exclude from consideration for ESA protection populations which play key roles in the survival mechanisms of many species.

NMFS' ESU policy thus creates a sort of regulatory catch-22 involving two important congressional goals. As part of its efforts to protect species' evolutionary legacy, NMFS extends protections to genetically unique populations--but only if such populations are reproductively isolated and thus incapable of playing a role in the entire species' survival and evolution. While NMFS' definition of distinct population segment allows the agency to account for lawmakers' concern over conserving unique genetic attributes, the agency's insistence upon reproductive isolation as a listing prerequisite undercuts conservation of entire species. This is inconsistent with one of Congress' fundamental purposes in enacting the ESA--halting the increase in species extinctions.

The above discussions lead to important generalizations about different methods to protect biodiversity. Through its ESU policy, NMFS essentially advocates one of two broad listing strategies under the ESA. One approach, which one could label a "preventative" policy, involves protecting populations whose existence contributes to the security of entire subspecies or species. Since these populations' importance to the greater whole may stem from their role as a source of dispersing colonists or simply from the geographical area they inhabit, a lack of reproductive isolation or unique genetic makeup would not disqualify these groups from protected status. Listing efforts would thus focus on a species' or subspecies' constituent populations which individually faced threats to their existence--even if the species or subspecies as a whole was relatively healthy--based on a policy that saving these populations would prevent eventual risks to the well-being of the subspecies or species itself. The second type of listing strategy, basically reflected in NMFS' ESU policy, calls for protecting only those populations which themselves are in the process of becoming new subspecies or species. These populations thus must be reproductively isolated and genetically distinct from others of the same subspecies or species. Under this approach, agencies would not direct their listing efforts at groupings other than ESUs until an entire subspecies or species faced extinction, at which point it would qualify for protection as a whole.(87)

While one could perhaps construct an argument based on fine points of genetics and an extremely narrow reading of the ESA that the latter strategy is valid--as indeed NMFS has attempted to do--the preventative approach to listing under the ESA is vastly superior both biologically and as a matter of public policy. NMFS' ESU policy results directly in the sort of "management to the brink" for which the ESA has been repeatedly criticized.(88) In other words, NMFS' approach essentially demands that the agency do nothing in the face of population-by-population erosion of a species or subspecies until the entire species or subspecies faces extinction. At this point, last ditch efforts to protect the species or subspecies may entail much higher costs, cause greater social tensions, and may ultimately prove unsuccessful. Further, species and subspecies bear the risks of scientific uncertainty under this approach. If scientists suddenly gain new information about conservation requirements for a species or subspecies, it could be too late to stem a decline fueled by progressive losses of constituent populations. In contrast, a listing strategy which attempts to prevent broad declines in species and subspecies by extending protections to constituent populations suffers from none of these flaws. It allows agencies to focus their conservation on discreet problems, possibly minimizing some of the costs and economic and social upheavals caused by listing an entire species or subspecies. Moreover, by getting ahead of the extinction curve, managers have more flexibility to adjust protection measures based on new scientific data. Accordingly, this approach best carries out lawmakers' wish to protect genetic diversity.

B. Protection of Ecosystem Health

NMFS, through its ESU policy, also ignores the ESA's ecosystem protections. In addition to explicitly listing ecosystem conservation as a purpose of the Act, lawmakers cited the importance of threatened and endangered species in maintaining ecosystem functions. However, NMFS' criteria for defining distinct population segments focus on populations in isolation from their biotic communities.

Defending its position, NMFS asserted that "[t]here may be a number of good reasons for maintaining populations of keystone' species in ecosystems where they play a key role in fostering diversity, but unless such populations can be shown to be |distinct,' such efforts must be accomplished outside the purview of the ESA as presently written."(89) The agency clearly implies that "distinct" within the phrase distinct population segments refers only to populations which are genetically distinct.(90) While a geneticist may reach such a conclusion, an ecologist would likely disagree.(91) It is possible to distinguish between some populations based on their contributions to an ecosystem. For instance, both sockeye and kokanee salmon inhabit Redfish Lake in Idaho.(92) The former variety migrates downriver as juveniles 900 miles to the ocean, where these fish spend their adult lives before making the return journey to spawn; kokanee salmon, on the other hand, remain in the lake or its inlet for their entire lives.(93) NMFS listed sockeye as endangered only after determining that these fish were probably reproductively isolated and genetically different from kokanee.(94) Had NMFS not found that sockeye met both of these criteria, the agency would have lumped them into an ESU with kokanee, and would not have listed the ESU had it found the kokanee population secure. In this case, sockeye could even become extinct without triggering ESA concerns since the ESA would not prohibit darn construction or other activities down-stream from the lake so long as the lake-dwelling portion of the ESU remained healthy. However, elimination of sockeye would have an adverse impact on the riparian and ocean ecosystems which they inhabit most of their lives. In such a case, even human members of the ecosystem would suffer adverse impacts, losing the opportunity to harvest, study, appreciate, and fish for sockeye.

The above hypothetical demonstrates that a population can play a vastly different ecological role from other genetically similar or reproductively connected species, refuting NMFS' contention that "distinct" in the statutory phrase "distinct population segment" refers exclusively to genetically distinct populations. Population which can be distinguished from others based on their ecological importance should thus be eligible for protection under the ESA regardless of their genetic makeups or reproductive interactions.

C. Protection of Domestic Populations

NMFS' interpretation of distinct population segment completely ignores lawmakers' clear intent that NMFS and FWS should have authority to protect the U.S. population of a species even if that species was abundant elsewhere in the world. Under NMFS' ESU policy, the agency would not list a dwindling domestic population unless that population was reproductively isolated and genetically distinct from other populations elsewhere in the world. One might argue that this policy does not really harm conservation of domestic populations since the U.S. population of a species not reproductively isolated from others would quickly reestablish itself through colonization in the event it became extinct. However, habitat loss and fragmentation now cause most extinctions. Without habitat protections provided under the ESA, the habitats of U.S. populations could be wiped out, eliminating the possibility of recolonization in the U.S. by individuals from adjacent populations outside the country.

In its response to comments on its ESU policy, NMFS dismissed the above position on the grounds that "the population concept used in the ESA is a biological one, and that political boundaries alone should not be used to define populations."(95) This reasoning serves as an excellent example of the agency's failure to recognize that Congress establishes policy even in defining technical terms.(96)

D. Management Flexibility and Uncertainty

NMFS' characterization of its listing authority expressed in the ESU policy also overlooks specific expressions of intent by Congress in 1979 when it debated whether to continue to allow listings of populations. hi fact, the listings of eagle populations, cited favorably in the Senate report as an example of the ESA's flexibility to allow managers to protect populations based on the degree of threat they face,(97) Would not be possible under NMFS' ESU policy. This policy effectively eliminates from listing decisions considerations of management flexibility to respond to degrees of threat faced by constituent populations of a species. In addition to conflicting with clear legislative intent, this policy makes little practical sense. For example, application of the ESU concept to bald eagles may result in a listing of eagles throughout North America as threatened or endangered.(98) Such a listing, among other things, would require federal agencies proposing to take actions in Alaska to engage in section 7 consultation with FWS if their actions affected eagles.(99) Since bald eagles exist in plentiful numbers in Alaska, such consultations would not make the best use of chronically scarce dollars for federal species conservation efforts.

In its discussions of its ESU policy, NMFS defends its refusal to extend ESA protections to "artificial units without a biological basis."(100) The agency repeatedly emphasizes that its interpretation of the term distinct population segment, as well as its future applications of the ESU concept, rest on scientific determinations.(101) Thus, according to NMFS, the ESU policy will permit a consistent approach to identifying distinct population segments.(102) Gleaves echos this refrain, arguing that "an agency's expertise concerning quasi-technical matters may be entitled to considerable deference."(103)

Much like the Wizard of Oz employed smoke and mirrors to lend him an air of might and wisdom, NMFS invokes science in an effort to portray its definitions of distinct populations eligible for listing as beyond question. When one looks behind this curtain, however, it becomes apparent that NMFS' ESU policy merely trades the discretion inherent in historic approaches to listing populations for a more technically complex but equally discretionary scheme. The tremendous discretion inherent in NMFS' ESU policy stems from two related sources: scientific uncertainty and extremely imprecise definitions of the two ESU criteria: reproductive isolation and whether a population represents an important component in the evolutionary legacy of a species.

Scientists know that genetic diversity plays a crucial role in species' and populations' survival and evolution, but measuring this diversity and determining which genes make contributions to evolutionary changes present formidable hurdles which for the most part remain unconquered. Geneticists employ two basic methods to attempt to discern genetic differences between organisms of groups of organisms.(104) The first relies upon observations of physical, behavioral, and life history differences and similarities to infer genetic differences and similarities. The second type consists of methods which use sophisticated technology to directly measure organisms' genetic makeup. These have come into use much more recently, and include various types of electrophoresis and DNA analysis, which detect the presence of different genes at specific places along a strand of mitochondrial DNA, allowing scientists to peer directly into an organism's genetic structure.(105) However, both of these types of methods are fraught with uncertainties. It is often extremely difficult or even impossible to separate environmental influences from genetically controlled traits when observing physical characteristics and life history.(106) For example, a salmon run characterized by large size may be genetically different from populations of smaller individuals which spawn in nearby streams, or the environmental conditions in the first stream (e.g., high turbulence or obstacles such as waterfalls) may simply make it impossible for smaller individuals to survive. Thus, physical variations such as size differences may or may not indicate genetic differences. Even when scientists employing sophisticated techniques detect specific genetic differences, they often have no way to tell whether these differences are "neutral" or are the result of natural selection and important to the survival and adaptability of the organisms.(107) Genetic differences also do not necessarily cause any physiological differences or differences in morphological functions. Random processes not linked to survival, long-term evolution, or even physical differences thus may account for the bulk of measurable genetic diversity between individuals and groups. Accordingly, even when modem instruments detect genetic differences, such differences do not necessarily indicate anything more than ubiquitous random genetic combinations. Moreover, since genetic analysis techniques can feasibly sample only a minute portion of organisms' genetic makeup, a lack of detectable genetic variation does not necessarily prove an absence of genetic differences.

Biologists are also often uncertain about how populations within a species interact. While researchers can sometimes determine conclusively that given populations are completely isolated from each other, in other situations experts disagree or are unable to say for certain whether populations are reproductively isolated. Similarly, the extent of dispersal or reproductive interactions among populations may be very difficult to determine with confidence.

Due to these scientific uncertainties, NMFS has only loosely defined the two criteria a population must meet to qualify as an ESU. NMFS qualities its first ESU requirement by providing that a population need only be "substantially" reproductively isolated from others of the same species.(108) The agency mentions various techniques for determining a population's degree of isolation, but recognizes that "each of [the] methods has its limitations."(109) According to NMFS, a population's reproductive isolation "must be strong enough to permit evolutionarily important differences to accrue in different population units,"(110) but the agency does not expand further on how to make such a determination. NMFS couches its definition of the second ESU criterion, that a population represent an important component in the evolutionary legacy of the species, in similarly qualitative terms. The agency defines a species' "evolutionary legacy" as its genetic diversity.(111) However, NMFS provides only vague guidance as to what constitutes an "important component" of a species' genetic diversity, noting that populations which "contribute substantially" to a species' overall genetic diversity, or whose loss would "significantly" diminish this diversity, meet the second criterion for ESU status.(112)

The above definitions serve as excellent examples of a key concept which virtually all scientists recognize, but which many other people do not: that science, and especially biology, is not the bastion of truisms, ironclad answers based on precise measurements, and experts with calculators who produce the one "right" answer. Rather, biologists formulate and reject hypotheses in attempts to find patterns in data sets, reach conclusions often based in part on their values, beliefs, or interpretations of uncertainty, and commonly conclude that they do not know the answer. Biologists frequently reach opposite conclusions, even when working from the same set of data. In other words, just because an agency calls a determination "scientific" rather than based on "policy" does not mean that the decision was void of discretion and value judgments. Indeed, NMFS admits as much with respect to its ESU policy by characterizing it as "no simple yardstick" for making determinations of distinct populations.(113)

Herein lies the real danger of NMFS' ESU policy. Judges for the most part have the unfortunate tendency of most lawyers to avoid scientific issues altogether. Accordingly, as Gleaves suggests, courts are generally most deferential to agency decisions when those decisions involve technical or scientific determinations.(114) Therefore, by converting the question of what constitutes a "distinct population segment" under the ESA from a mix of science and policy to one of wholly genetics, NMFS has attempted to insulate its listing decisions from judicial review. The agency realizes that courts will take an active role in reviewing its "policy" judgment calls for consistency with statutory language and congressional intent, but will simply defer to "scientific" judgment calls so long as NMFS scientists can set forth merely rational reasoning. Therefore, NMFS' ESU policy establishes the agencies' geneticists as the ultimate arbiters of which populations are eligible for ESA protections. This raises the disturbing possibility that NMFS may make what are really policy choices under the guise of science.

NMFS' approach to scientific uncertainty provides an example of the manner in which the agency can employ its ESU policy in an attempt to put a shield of science around what are fundamentally policy decisions. The technical document setting forth the ESU concept provides that ESUs should correspond to "more comprehensive" groups of organisms unless "there is clear evidence that evolutionarily important differences exist between smaller population segments."(115) In other words, scientific uncertainty will always cut against defining a given population as an ESU; unless a listing proponent carries the burden of proving that a population qualifies as an ESU, NMFS will not consider listing it as a distinct population segment. NMFS has thus made a choice about how to proceed in the face of scientific uncertainty in the context of defining the "scientific" concept of evolutionarily significant units.

In fact, however, deciding what to do given uncertainty is not a scientific or technical issue--it involves making a value judgment of how to proceed under the risk of making the wrong choice. Not only is this a policy decision, Congress has already spoken on the matter with respect to protecting populations. The 1979 Senate report clearly provides that NMFS and FWS should exercise their authority to list distinct populations when evidence indicates that a population faces widespread threats, but conclusive data demonstrating threatened or endangered status exists only for certain populations.(116) This language specifically applies lawmakers' intention to give species the benefit of the doubt to listing decisions.(117) NMFS' listing policy disfavoring ESA protections for populations when uncertainty exists, disguised as "science" through the ESU policy, is clearly at odds with this congressional directive.

NMFS has cited lack of information as a reason to deny listing of a population of Pacific salmon. In 1992, several environmental and fishing organizations petitioned NMFS to list winter steelhead in the Illinois River.(118) Although it recognized some potential differences between these fish and other steelhead runs, NMFS cited a lack of information which conclusively demonstrated reproductive isolation and evolutionary significance to support its finding that Illinois winter steelhead did not constitute an ESU and hence were not eligible for listing as a distinct population segment.(119) The agency did not discuss the petitioners' evidence of the Illinois winter steelhead's declining numbers. However, the agency noted a general decline in steelhead populations and initiated a West Coast-wide status review of the species.(120)

This is precisely the type of listing determination Congress sought to avoid. NMFS recognized that steelhead faced widespread threats, but did not at the time have sufficient information to list steelhead coastwide. However, concerned groups presented the agency with much more specific information documenting the decline of a specific population. Yet NMFS declined to even consider information on threats to the existence of Illinois winter steelhead because the petitioners could not prove that these fish met the agency's ESU criteria. This finding almost directly conflicts with the Senate's example of appropriate use of the Secretary's authority to list distinct population segments. The agency thus employed its "scientific" ESU identification process to ignore lawmakers' (and Aldo Leopold's) admonition that uncertainty over a part's usefulness should not cause us to discard it.

E. Incongruity Between Evolutionary Significance and Listing Thresholds

Finally, NMFS' requirement that a population qualify as an ESU to be eligible for listing, coupled with the manner in which the agency apparently interprets the point at which a species becomes endangered or threatened, leads to a situation which fails to protect the very characteristics which NMFS employs to define populations which it deems worthy of ESA protections. Understanding this interaction requires a brief examination of NMFS' interpretation of the ESA's vague definitions of the terms "threatened" and "endangered."

NMFS examined the issue of the ESA's listing threshold--the point at which a species is considered endangered or threatened--in a draft technical paper.(121) To its credit, the agency acknowledged that such thresholds must be expressed in terms of a species' likelihood of persistence over time.(122) NMFS proposed that an "endangered" species be defined as a species which has less than a 95 percent chance of continuing to exist for 100 years.(123) While the method employed to reach this definition was arguably flawed,(124) NMFS' suggestion of listing thresholds is significant in that it suggests a planning horizon for species conservation efforts 100 years into the future.

Given its definition of what constitutes an "endangered" species, NMFS' suggested thresholds policy undermines statutory protections for the very qualities which the agency uses to identify populations worthy of protection. In order to qualify as a distinct population segment eligible for listing, an isolated population must exhibit "evolutionarily significant" genetic differences. However, in order to actually be listed under NMFS' threshold criteria, a species must face serious threats to its existence over the next century--a minuscule time frame from an evolutionary standpoint. Put another way, NMFS places determinative weight on evolutionary significance in order for a population to qualify as a legal "species," but will refuse to actually extend the ESA's protections to that species until its very existence is in peril over the short term. In effect, therefore, NMFS greatly values a population's evolutionary past, but will not direct its efforts toward protecting an ESU over a time frame that will allow this evolutionary process to continue.

F. Summation 

In sum, NMFS' ESU policy falls significantly short of interpreting its listing authority in a manner which carries out policies established by Congress, does not account for important biological factors which influence species' persistence, and works against protective efforts in instances of scientific uncertainty.(125) Courts and other interested reviewers should not hesitate to reject NMFS' contention that it determines what constitutes distinct populations segments solely on "scientific" grounds. Congress unambiguously included several policy aspects in its legal definition of "species," policies which the ESU approach largely does not implement. Additionally, the ESU policy's technical shortcomings suggest that NMFS has not fulfilled its obligations to conserve species and to construe its listing authority in light of the best scientific data available.

V. FISH AND WILDLIFE SERVICE'S APPROACH TO DISTINCT POPULATION SEGMENTS

Given NMFS' flawed interpretation of its authority to list distinct populations under the ESA, the obvious next analytical step is to examine how its sister agency has dealt with this issue. Such an analysis reveals that FWS has also struggled in its efforts to determine which populations deserve the statute's protection. The following discussion traces FWS' application of its authority to list distinct populations from the agency's free-wheeling listings during the ESA's early years, through the failed hybrid policy, the agency's subsequent roller coaster approach to listing populations, and finally to a draft policy formulated by FWS in 1992 and recent developments.

A. Early Years

Early FWS listings of populations made virtually no attempts to ascertain

whether biological or genetic differences existed between groups listed as threatened or endangered and other groups within the same species which either received no ESA protection, or were protected under the alternate listing category. Rather, FWS listings differentiated between populations based on degree of threat faced by groups of organisms within geopolitically-defined boundaries, and on administrative concerns.

In its initial years of implementing the ESA, FWS listed groups below the subspecies level based primarily upon abundance of, or degree of threat faced by, a portion of a species delineated on the basis of geopolitical boundaries. For example, in 1978 FWS listed bald eagles in the coterminous United States as endangered with the exception of eagles in five Northwest and Great Lake states.(126) The agency declared eagles in those states threatened because they were relatively more abundant and enjoyed greater reproductive success.(127) FWS carried its practice of using geopolitical boundaries to differentiate more abundant populations to an even finer focus in the case of American alligators, which the agency divided into different listing categories defined by parish boundaries within the state of Louisiana.(128) In the case of sea turtles, the agency acknowledged that three species of turtles faced threats to their existence throughout their range, but had sufficient information to identify only specific areas in Florida and Mexico as places where portions of two species faced particularly serious perils.(129) Accordingly, FWS listed the three entire species as threatened, and designated groups of two species in Florida and Mexico as endangered.(130)

In making distinctions based on geopolitical boundaries, FWS put particular emphasis on the United States' border. For example, in both its listings of bald eagles and grizzly bears, the agency considered only the populations located in the coterminous forty-eight states.(131) Although FWS placed importance on abundance by geographic area, the agency made virtually no effort to consider whether these species had suffered declines in portions of Canada. In the case of grizzlies, the agency appears to have applied some sort of implicit assumption that it should focus on the animals in the United States (outside Alaska), going so far as to define grizzlies' entire range as the lower forty-eight states even though it was obviously well aware that grizzlies existed in Canada and Alaska.(132)

Issues related to administration of protection for populations also played a role in FWS' early listing decisions. The agency justified its split listing of alligators on the basis that "[s]ome populations ... are now at the point where the species could be best served by more flexible management programs."(133) An environmental group and scientific organizations suggested that FWS list bald eagles as endangered in southern portions of three states in which FWS proposed threatened status, presumably because of relatively lower abundance in these areas. FWS gently rejected this suggestion, asserting that delineation based on state boundaries was "most appropriate."(134) Given the conspicuous absence of any discussion by FWS of the status of eagle numbers in these areas, the agency appears to have differentiated eagle groups by state boundaries simply for administrative convenience.

FWS was well aware that no genetic or biological differences existed between conspecific groups it had distinguished for listing purposes. In its alligator listing decision, the agency noted that the animals listed as threatened were "taxonomically and morphologically identical" to those listed as endangered.(135) Similarly, FWS noted that bald eagles in Canada are "indistinguishable" from their listed American cousins.(136) The agency in fact expressly rejected an argument from an environmental organization that it was required to treat all eagles within the same subspecies as one unit for listing purposes.(137) While this pronouncement concerned subspecies rather than distinct populations, by basing its criteria for distinguishing eagles on geopolitical boundaries and degree of threat, FWS clearly rejected biological or genetic uniqueness as a prerequisite for a population qualifying as a distinct population segment.

B. Hybrid Policy

FWS began in the late 1970s to wrestle with whether to differentiate between organisms based on their genetic makeup. Similar to NMFS' ESU policy, FWS' debates centered on how an organism's genes affect its eligibility for protections under the ESA. However, FWS' hybrid policy focused on whether certain progeny of listed individuals were themselves protected by the statute rather than attempted to define the term "distinct population segment."(138)

FWS' hybrid policy arose from a question posed by the agency's law enforcement division: does the ESA protect hybrid offspring when only one parent is listed under the statute? A 1977 legal opinion from the Interior Department's Office of the Solicitor initially reached an affirmative conclusion, but the Solicitor reversed this decision a few months later.(139) FWS subsequently held fast to its position that the ESA does not protect hybrids, even when implementing this policy essentially led to the extinction of a listed species.(140) The agency went so far as to conclude that progeny of parents of different species who were themselves both listed still did not qualify for protection, even when the hybridization resulted from matings which took place in the wild.(141) FWS justified this policy on grounds similar to those employed by NMFS to defend its ESU policy, arguing that protection of hybrids would not serve to conserve the genetic heritage of listed species.(142)

In 1990, FWS again reversed its position on hybrids, expressly rejecting its previous reasoning and actions. This later Solicitor's opinion reasoned that "[n]ew biological information casts doubt on the validity of the absolute pronouncements relied upon in the [1977 opinion rejecting legal protections for hybrids] since in some cases genetic introgression between taxa is not only a naturally-occurring phenomenon but may also be a necessary recovery measure to avoid the loss of a seriously depleted tax-on."(143) The Solicitor also called for a new policy on "this complex issue."(144)

C. Uncertainty

Listing decisions published by FWS after the hybrid policy's demise reveal that the agency was struggling to define its authority to list distinct population segments. Like NMFS, FWS usually discussed reproductive isolation and genetic distinctiveness in analyzing whether specified groupings within a species qualify for listing; however, the agency gave seemingly conflicting indications of the importance of these factors. Further, unlike its sister agency, FWS provided almost no discussion of how it defined the concepts of reproductive isolation and genetic distinctiveness. Finally, FWS appeared to have pursued a course which--with one notable exception--gave the benefit of the doubt to the population under consideration for protection in cases of scientific uncertainty.

FWS' listing decisions after 1990 gave mixed signals as to whether the agency believed that a population must be reproductively isolated from other populations of the same species or subspecies in order to be eligible for listing. The agency cited reproductive isolation between the Mojave and Sonoran populations of desert tortoises as support for its decision to list the Mojave population as threatened.(145) Similarly, FWS supported its refusal to consider protection for northern goshawks in the United States west of the 100th meridian by raising the possibility that this grouping was not completely isolated from goshawks in the eastern portion of the country.(146) The agency also pointed to evidence of negligible intermixing between interior and coastal Western snowy plover populations in its decision to list the Pacific coast population,(147) and cited the issue of whether populations were "disjunct" in one of its decisions involving marbled murrelets.(148) On the other hand, in a decision notable for its complete repudiation of the former hybrid policy, FWS downplayed evidence of reproductive interaction among three canid species. In denying an industry group's petition to delist red wolves on grounds that the animals were actually wolf/coyote hybrids, FWS acknowledged that genetic evidence supported a hypothesis that past hybridizations had occurred, but that this merely indicated "evolutionary stages" of red wolves.(149) Using parallel reasoning, the agency defended its decision to list Louisiana black bears despite their potential reproductive interactions with other subspecies black bears. Acknowledging the "opportunity for free movement of black bear from adjoining states into the range of the Louisiana black bear," FWS argued that "interbreeding between subspecies is a normal and expected occurrence" and that protection of Louisiana black bears was the only means to protect remaining genetic attributes of the subspecies.(150)

FWS has likewise waffled on the importance of genetic makeup in distinguishing between distinct population segments. Predictably, the agency cited the presence or absence of genetic distinctiveness in instances in which it found reproductive isolation to be important, and downplayed genetics in cases where it had made listing decision despite a lack of such isolation.(151)

FWS' listing decisions demonstrate that in addition to strictly biological factors, the agency continues to consider geopolitical boundaries in delineating distinct population segments. The agency continues to focus on the lower forty-eight states as an important geographical unit. In its listings decision for marbled murrelets, for example, FWS listed only the birds within California, Oregon, and Washington even though the population extends into Canada and Alaska and nothing indicated that the US-Canada border serves as a barrier to migration.(152) While the listing decisions noted that the birds faced serious threats in these three states, FWS made little effort to assess their status in Canada and Alaska. Interestingly, FWS apparently balks at continuing its old practice of using state boundaries or borders of smaller political subdivisions to differentiate distinct population segments within the lower forty-eight states. It rejected a petition to protect yellow-billed cuckoos in five Western states because birds "in the petitioned states cannot be regarded as a population separate from adjoining states that were not included in the petition."(153)

The marbled murrelet listing decision is also notable in that it links the concept of distinct population segment with the ESA's definitions of endangered and threatened species. FWS justified its finding that murrelets in California, Oregon, and Washington constitute a distinct population segment not due to isolation or biological differences with more northern birds--which the agency acknowledged were likely absent--but because it found that the three-state area comprises a significant portion of the range of the eastern subspecies of murrelets.(154) This decision is significant in that it adopts an entirely different approach to defining distinct population segments by suggesting that the ESA!S reference to a "significant portion" of its range in the definitions of endangered and threatened species provides a mechanism to delineate distinct populations eligible for protection, one which does not necessarily require any evidence of reproductive isolation or genetic or other biological differences.(155) Whether this approach will influence FWS' future listing decisions is somewhat unclear, however, because a federal district court virtually mandated that the agency protect murrelets.(156)

Finally, recent FWS listing decisions in which the agency has discussed reproductive isolation and genetic factors have given conflicting signals as to how the agency proceeds in the face of scientific uncertainty. In rejecting petitions to list northern goshawks, FWS bluntly asserted that "the burden of proof for genetic isolation rests with the petitioner ....:(157) Although conceding that "gene flow among goshawks across large geographic areas may be low," the agency refused to protect goshawks in the western U.S. based on speculation that genetic exchange with eastern goshawks could occur via birds in Canada, and on supposition that such genetic exchange could eliminate any genetic uniqueness in western goshawks.(158) In a remarkably similar decision reaching the opposite conclusion, FWS recognized the possibility of low levels of genetic interchange between fisher populations in the Rocky Mountains and Pacific states via Canada.(159) However, the agency cited a lack of data suggesting genetic interchange, as well as the improbability of significant genetic exchanges between populations, to support its finding that Pacific fishers qualified as a distinct population segment.(160) In another decision suggesting that FWS construes scientific uncertainty in the light most favorable to protection of populations facing clear threats, the agency supported its decision not to delist red wolves even in light of some hybridization with other canids by asserting that the ESA demands a "conservative approach" to delisting protected species.(161) Finally, although opponents of listing Louisiana black bears argued that lack of proof of genetic differences between these bears and similar subspecies rendered Louisiana black bears' taxonomic distinction invalid, FWS refused to pursue this line of reasoning in listing the bears.(162)

D. Draft Policy and Beyond

Prompted by NMFS publication of its ESU policy and growing controversy over the issue, the FWS Deputy Director in 1992 issued a draft policy on how the agency would interpret its authority to list distinct population segments.(163) The cover memorandum instructed field offices to implement the policy for a trial period of six months and promised publication of the policy in the Federal Register with a request for public comment.(164) However, FWS never published the draft policy, and eventually published a notice withdrawing its intent to do so.(165) FWS has yet to publish updated guidance on this issue. The draft policy thus provides an indication of how the agency may ultimately interpret its listing authority.

The foundation of FWS' draft policy is a list of four factors. In order to qualify for listing consideration as a distinct population segment, a grouping must meet one or more of the following criteria:

1. It is significantly isolated from other members of the same species

or subspecies (i.e., it rarely interbreeds with other populations);

2. It occupies an ecosystem that is in danger of destruction

throughout all or a significant portion of its historical distribution,

and the species or subspecies is not present in surrounding ecosystems;

3. It is the only occurrence of a species or subspecies within United

States jurisdiction; or

4. It can be defined by geopolitical boundaries that delineate an

area (representing a significant portion of the species or subspecies

range) where existing legal protection is inadequate to ensure its

survival.(166) The policy provided further substantive guidance in applying these factors:

Biological significance, ecological characteristics, and geographical

distribution are the primary factors considered when evaluating the

appropriateness of listing a population segment. Political boundaries,

regulatory mechanisms, and management are also considered.

Genetic or morphological distinctness may or may not be present or

known; these factors may be considered but are not essential in

making a determination. In evaluating these attributes, the Service

will rely upon the best scientific data available.(167) The draft policy placed particular emphasis on a population's "biological significance." FWS found this consideration important due to Congress' admonition in 1979 that FWS and NMFS should use sparingly their authority to list population segments.(168) Although it is not listed among the policy's four basic factors, FWS noted that the agency would consider a population's biological significance to a species or subspecies as a whole in all determinations of listing eligibility. According to the draft policy, a population is biologically significant if it is important in maintaining the viability of a species or subspecies, or is "important in preventing further decline (or in assisting recovery, if listed) of the species or subspecies."(169)

While neither the cover memorandum nor draft policy itself included a complete discussion of FWS' rationale, the draft policy contained some very interesting reasons supporting the agency's approach. In a clear technical swipe at NMFS' ESU policy, FWS explained that its approach does not require a population to exhibit significant genetic or morphological differences to qualify for listing as a distinct population segment because if such differences were present, "the population would likely be judged a separate subspecies or species."(170) Turning to policy arguments, FWS maintained that its approach, while likely resulting in additional listing proposals and petitions, would allow the agency to address "local issues" rather than emphasize wide scale efforts to protect entire species and subspecies.(171) The agency reasoned that this approach will result in a "more effective program" by protecting a population segment before the entire species or subspecies declines to the point that it requires protection.(172) FWS also argued that such "smaller scale" listings would prove less costly than implementing broad protections for entire species and subspecies.(173)

While no further statement of FWS' policy on interpreting distinct population have apparently been formalized, the agency has continued to wrestle with this issue.(174) FWS, prompted to some extent by NMFS' continued application of its ESU concept, has begun to move away from some of the ideas expressed in its 1992 draft policy.(175) As of early 1994, FWS generally required that in order to qualify for listing, a population must be both "discreet" and "significant."(176) The first criterion requires that a population either be reproductively isolated from others of the same species or subspecies or be defined by international borders.(177) The second criterion characterizes "significant" populations as genetically, physically, or behaviorally unique, or as those which occupy a "gap" in the broader species' range.(178)

Recent listing decisions provide an indication of how FWS has applied these ideas. The agency denied a petition to protect lynx in Washington's North Cascades, finding that the identified population "does not constitute the entire coterminous United States population of the species," as well as citing indications that the lynx in Washington could readily disperse back and forth across the Canadian border.(179) FWS also denied a petition to list the southermost population of longfin smelt.(180) Although the agency determined that the Sacramento-San Joaquin delta population was probably reproductively isolated from more northern and more abundant populations, FWS noted that isolation did not necessarily mean that this population was "significant" to the wide-ranging species.(181) FWS also cited the population's apparent lack of genetic distinctiveness.(182) Finally, although the agency noted that longfin smelt were formerly the fourth most abundant fish in the Sacramento-San Joaquin delta ecosystem, the agency rejected the population's former contribution to this ecosystem as an indicator of significance because "the role of this declining species in the estuary today is unknown."(183)

VI. JUDICIAL INTERPRETATIONS OF SPECIES ELIGIBLE FOR LISTING

Gleaves examined the sparse and largely unenlightening case law relating to the question of which populations are eligible for ESA protections.(184) Subsequent to Gleaves' discussion, a federal district court directly addressed this issue in an unreported opinion.(185) The court's conclusions carry potentially significant implications for ultimate resolution of the meaning of distinct populations segment. Another recent federal district court opinion(186) raises important procedural issues regarding efforts to define groupings eligible for listing. This section summarizes and analyzes these opinions.

In 1988, a coalition of environmental organizations filed a petition to list marbled murrelets in California, Oregon, and Washington as a threatened species.(187) After putting off a final decision on the petition several times, FWS published a notice announcing an additional six-month delay to resolve disagreements over the issue of whether the birds in these three states qualified as a distinct population segment.(188) Petitioners sought an order from a federal district court declaring the delay illegal. The court ruled in plaintiffs' favor, holding that murrelets in the tri-state area qualified for protection under the ESA under two lines of argument.

As its first rationale for finding that the identified murrelet population was eligible for listing, the court cited the ESA's definition of threatened species. Since that definition refers to a significant portion of a species' range, the court accepted plaintiffs' argument that "in order to support listing [the tri-state population of] the marbled murrelet as threatened under the ESA, the Secretary need only find that the North American subspecies of marbled murrelet is threatened throughout a significant portion of its range."(189) The court noted FWS had acknowledged that California, Oregon, and Washington constituted a significant portion of the birds' range, and pointed out the agency also admitted that murrelets were threatened by habitat loss in those states. Accordingly, the court held that the population qualified for listing under the ESA's definition of threatened species and thus "there is no need to consider the alternative basis of whether the tri-state population is a distinct population segment which might qualify for protection under the ESA."(190)

Despite this holding, the court went on to examine the issue of whether FWS had acted arbitrarily in finding existence of a dispute over whether the tri-state murrelet population was a distinct population segment. After determining that public comments did not raise such a dispute, the court examined whether controversy over murrelets' status as a distinct population segment existed within FWS. In answering this question negatively, the court prominently cited a draft final rule sent to Washington, D.C. by the FWS Portland, Oregon Field Office. The court thus ruled that FWS had acted arbitrarily in delaying its listing decision due to a purported controversy over murrelets' status as a distinct population segment.(191)

This decision is significant in that it suggests an entirely different source of statutory authority upon which FWS and NMFS could base listings of populations. Rather than focusing exclusively on the ESA's definition of species, the court in essence found an additional definition of the term "species" within the statutory definitions of "threatened species" and "endangered species." Rather than the murky phrase "distinct population segment," these definitions simply refer to a "significant portion" of a species' range.(192) To the district court, inclusion of this phrase within the ESA's definition of threatened species meant that Congress intended to allow FWS and NMFS to list the population within a significant portion of a species' or subspecies' entire range.(193)

This interpretation defensibly construes a rather confusing jumble of statutory definitions. The ESA in fact does contain two definitions of the term "species," one defining the term standing alone and another defining the term as modified by the terms "threatened" and "endangered." The former definition contains the ambiguous term "distinct population segment." This term makes no sense when plugged into the definitions of threatened species and endangered species.(194) Finally, the ESA's definitions of threatened species and endangered species fail to specify whether an agency must list the entire species when it finds a species to be in trouble in a significant portion of its range, or whether it may list only the portion of the species in trouble. The court's finding that the definition of threatened species allows an agency to adopt that latter course provides a way to harmonize the ESA's various references to species, as well as make sense of the term "distinct population segment." In effect, the court's ruling brings together FWS' 1992 draft policy interpreting distinct population segment and the statute's definitions of threatened species and endangered species. The former allows FWS to list populations inhabiting a significant portion of a species' entire range;... the court interpreted the latter to provide this authority as well.

Finally, the upshot of the Marbled Murrelet decision for NMFS' ESU policy is this: much of the broad listing approach which NMFS rejected in devising its ESU concept may be ingrained in the ESA itself through the definitions of threatened species and endangered species. Accordingly, NMFS' listing decisions which rely exclusively on the agency's narrow ESU policy may be vulnerable to legal challenges.

The second recent case dealing with agencies' listing authority involved subspecies of California gnatcatchers, birds found along the west coast of California and the Baja peninsula. Dr. Jonathan Atwood, an ornithologist studying gnatcatchers, published a paper in the late 1980s placing the boundary between the two gnatcatcher subspecies at 25 degrees north latitude in southern Baja California. After other ornithologists criticized this conclusion, Atwood reanalyzed his raw data and modified his original determination. Atwood published a second paper in 1990 placing the boundary between the gnatcatcher subspecies considerably farther north, at 30 degrees north latitude.

Several entities, including Dr. Atwood, subsequently petitioned FWS to protect the northern gnatcatcher subspecies under the ESA. In considering these requests, FWS assigned two of the agency's taxonomists to review the birds' taxonomy, in particular Atwood's conclusions about the boundary between the gnatcatcher subspecies. However, FWS did not acquire or examine Atwood's raw data. After extending the public comment period for six months to consider additional comments on gnatcatcher taxonomy, FWS ultimately listed the northern gnatcatcher subspecies as threatened.(196)

This decision sparked a great deal of controversy because gnatcatchers inhabit areas with tremendous real estate value near the coast in southern California. The area's construction industry hired ornithologists who criticized Atwood's 1990 conclusion regarding the boundary between gnatcatcher subspecies. This was a key issue because if the boundary was further south, as Atwood had originally concluded, the population of the northern subspecies would likely be too large to qualify for listing. The industry ornithologists sought Atwood's raw data from Atwood himself, who refused to provide it. However, he offered the data to FWS, but the agency relied only on Atwood's 1990 paper and its own taxonomists' review of Atwood's findings in reaching its decision to list the northern subspecies. The industry ornithologists also sought Atwood's raw data from FWS, but the agency denied their request because it had not reviewed the data in making its listing determination. Ultimately, the construction industry challenged the gnatcatcher listing in federal court.

The District of Columbia District Court found in favor of this challenge, and remanded the decision to list the northern gnatcatcher subspecies back to FWS.(197) Although the court recited its duty to defer to an administrative agency's findings on technical matters, the court reasoned that "[i]t is the disputed nature of [Atwood's 19901 report from other cases where a scientific report alone has been considered sufficient for ESA purposes."(198) Accordingly, the court overturned the gnatcatcher listing, concluding that when "the underlying data from such a critical and disputed report is readily available to the Secretary, even though he chose not to review the data, it was error for the Secretary not to make the data available to those interested parties from whom the Secretary sought comment."(199)

This ruling holds potential to vastly complicate listings of populations pursuant to the ESU policy or a similar approach to defining population segments. While the gnatcatcher controversy demonstrates that scientists are often at odds over the declineation of subspecies, the process of identifying evolutionarily significant units offers even more potential for scientific disputes due to the amorphous definition of this term.(200) Moreover, studies examining populations' reproductive isolation and genetic makeup will likely involve large quantities of raw data. If NMFS (and FWS, should it adopt an ESU-type approach to listing) has an obligation to obtain and distribute(201) raw data whenever it makes a controversial decision as to whether a population constitutes an ESU, it could be swamped by this new administrative burden. Such a load on listing agencies, in addition to consuming scarce resources, could delay listings themselves. Further, both listing proponents and opponents could employ the underlying raw data to engage listing agencies in protracted technical debates over data interpretation, and could also use the information to mount legal challenges to listing decisions.(202)

The gnatcatcher case thus illustrates a major drawback to NMFS' ESU policy. Rather than providing a consistent approach to identifying distinct population segments, the ESU approach sets forth a recipe for endless technical bickering. While scientists debate populations' allele frequencies, the creatures themselves will edge ever closer to extinction.

VII. WHERE TO FROM HERE?

Events will undoubtedly push toward a clearer, unified definition of distinct population segment in the not-too-distant future. The two agencies charged with implementing the ESA have reached, at least on the basis of their most recent written policies in the public arena, dramatically different interpretations of their listing authority. As one might expect, NMFS and FWS are currently involved in discussions aimed at issuing a joint regulatory definition of distinct population segment.(203) Other activities could substantially influence this process. Several environmental organizations recently filed the requisite sixty-day notice of intent to file a lawsuit challenging FWS' decision to reject the petition to list the western U.S. population of northern goshawks.(204) This could prompt FWS to explicitly abandon the reasoning it employed to deny the goshawk petition, reasoning very similar to that found in NMFS' ESU policy. Alternatively, if FWS adheres to its decision, a federal court would examine whether ESU considerations should in fact determine populations' eligibility for listing. The legislative branch may also join in the fray when Congress eventually votes on ESA reauthorization. At that point, of course, lawmakers could choose to have the final say on which, if any, groupings below the species level are deserving of protection.

Whichever branch of government ultimately decides this issue, a scheme similar to FWS' 1992 draft policy should govern determinations of what constitutes a distinct population segment. FWS' 1992 proposed policy eliminates many of the shortcomings of NMFS' ESU approach. It largely dispenses with NMFS' ironclad requirements that distinct population segments be reproductively isolated and genetically for sound technical as well as policy reasons. As a technical matter, FWS observed in its 1992 draft policy, with much justification, that populations meeting these criteria would likely be considered separate subspecies or even species and thus be eligible for listing anyway. Elimination of these criteria as the sole determinants of a distinct population segment would also do away with NMFS' fuzzy definitions of what constitutes a sufficient degree of reproductive isolation and an "important" component of a species' evolutionary legacy, as well as quash NMFS' attempt to curtail judicial review by passing off its listing decisions as purely "scientific" determinations. Legally, FWS' proposed approach is consistent with the Marbled Murrelet court decision. On policy grounds, FWS' 1992 draft policy's repudiation of the ESU concept allowed the agency to formulate an approach which could extend ESA protections to U.S. populations, considers a population in context of its place in a particular ecosystem, and permits regulatory flexibility to formulate a listing which allows the Secretary to direct protection at only the portion of a species or subspecies which actually faces threats to its existence. All of these results carry out lawmakers' intent in providing agencies with the authority to list distinct population segments.

FWS' 1992 proposed approach also represents a very pragmatic effort to deal with species conservation. As the agency successfully argued in 1979, and again asserted in its 1992 proposed policy, it may not make sense to list an entire species, subspecies, or even ESU when an agency could instead direct resources toward solving what FWS terms a more "local" problem. Such a listing approach, which targets problems before an entire species faces extinction, could help to quiet criticism characterizing the ESA as simply an "emergency room" statute which does nothing until a species teeters on the brink of extinction. Further, there may be substantial merit in FWS' argument that conservation efforts targeted at the population level could prove less costly than waiting until larger units merit listing, both in terms of lower administrative costs and less economic disruption to society in general.

FWS also structured its 1992 draft policy in a manner likely to put on solid legal ground listing decision made pursuant to this approach. Unlike the ESU concept, an approach similar to FWS' draft policy would give agencies the ability to tip the benefit of the doubt in favor of protection in cases of scientific uncertainty. This comports with expressions of congressional intent. Moreover, linking the definition of distinct population segment to the significant portion of species' range language in the statutory definitions of threatened and endangered species makes a great deal of sense from a statutory interpretation standpoint. Lastly, the emphasis on a population's "biological significance" should accommodate lawmakers' concern that agencies not abuse their authority to list distinct population segments.

Finally, FWS' 1992 draft policy succeeds because it recognizes the definition of distinct population segment to be, in the words of one ESA commentator, "a mixture of both art and science."(205) FWS has credibly combined the policies expressed by Congress through the ESA with biological principles, in stark contrast to NMFS' genetically-driven ESU approach. While agencies would have fairly broad listing discretion under FWS' proposed policy, their listing decisions would be subject to meaningful judicial review. This approach thus provides for science-based listing decisions within a framework defined by important policy considerations, and it enables judicial scrutiny to guard against potential abuses.

Unfortunately, FWS appears to be moving away from ideas set forth in its 1992 draft policy and toward incorporating ESU-based concepts advocated by NMFS.(206) Accordingly, FWS risks incorporating into its own listing process the significant shortcomings of NMFS' approach. FWS should resist pressure to move its interpretation of its listing authority to NMFS' ESU policy. Although the Clinton Administration has sensibly worked to ensure that the federal government speaks with one voice, FWS should not compromise with its sister agency when it has developed a definition of distinct population segment both legally and biologically superior to that created by NMFS.

VIII. CONCLUSION

Unfortunately, controversy has unfortunately dogged the ESA throughout much of its twenty-year history. One of the most significant recent controversies to arise under the statute implicates one of its most basic concepts--the definition of "species" eligible for protection. NMFS developed its ESU policy to delineate distinct population segments of Pacific salmon, but this policy obviously raises the question whether both NMFS and FWS should use the ESU concept in all of their listing decisions.

Contrary to Gleaves' contention that Congress left the definition of distinct population segment to agency discretion, the ESA itself as well as its legislative history reveal several policy goals lawmakers sought to advance in affording NMFS and FWS discretion to list groups of organisms below the subspecies level. NMFS' ESU policy largely fails to advance these aims.

NMFS' application of the ESU concept to listing decisions also suffers from biological shortcomings. In addition to problems such as limiting protections for metapopulations, scientific uncertainty coupled with the loose manner in which NMFS has defined key ESU criteria give the agency wide latitude to deny ESA protection to populations. However, since the ESU policy purports to establish purely "scientific" grounds for making listing decisions, courts could be reluctant to subject NMFS' determinations to scrutiny.

While a review of FWS listing decisions shows that the agency has struggled with the issue of what constitutes a distinct population segment, FWS' 1992 draft policy on the subject demonstrates a thoughtful and well-reasoned effort to integrate the ESA's policy goals with biological concepts. It is consistent with a federal district court's approach to populations eligible for listing. Current administrative efforts to produce a joint FWS/NMFS policy governing interpretation of the term "distinct population segment" should adopt a similar approach, a position Congress should ratify when it finally reauthorizes the ESA.

The ESA stands as one of the United States strongest and most important environmental statutes. It is a bulwark against erosion of biological diversity, a worsening problem which looms as one of humankind's major challenges. Abandoning the ESU concept will allow NMFS and FWS to pursue targeted, cost-effective species conservation programs rather than spend substantial time and resources engaging in arcane and potentially arbitrary genetic wild goose chases which ultimately result in excluding from listing populations which merit the statute's protection. The last word on this issue, not surprisingly, should go to Aldo Leopold: "We console ourselves with the comfortable fallacy that a single museum-piece will do, ignoring the clear dictum of history that a species must be saved in many places if it is to be saved at all."(207) (1.) 16 U.S.C. [subsection] 1531-1543 (1988). (2.) The Secretary of Commerce, who heads the department which includes NMFS, has jurisdiction over marine species. The Secretary of the Interior, and thus FWS, has jurisdiction over all other species. (3.) NMFS announced its definition of species in the context of its policy on applying ESA protections to Pacific salmon. See 56 Fed. Reg. 58,612 (Nov. 20, 1991). The agency expanded on this policy in an earlier technical publication. See Robin Waples, "Definition of Species Under the Endangered Species Act: Application to Pacific Salmon," NOAA Technical Memorandum NMFS F/NWC-194 (1991). (4.) Karl Gleaves, Michele Kuruc, Patricia Montanio, 7he Meaning of "Species" Under the Endangered Species Act, 13 Pub. Land L. Rev. 25 (1992) [hereinafter "Gleaves"]. (5.) The ESA defines the term "species" to include "any subspecies of fish or wildlife or plants, and any distinct population segment of any subspecies of vertebrate fish or wildlife which interbreeds when mature." 16 U.S.C. [section] 1532(16). (6.) See Gleaves, supra note 4, at 25 (citation omitted). (7.) See Daniel Rohlf, The Endangered Species Act: A Guide to Its Protections and Implementation 37 (1989). (8.) Congress amended the ESA in 1982 to emphasize that the Secretaries must make listing decisions "solely" on the basis of the best scientific data available. See 16 U.S.C. [section] 1533(b)(1)(A). However, the listing agencies nevertheless occasionally come under political pressure to consider non-scientific factors in the listing process, as clearly occurred when FWS refused to list spotted owls. See Northern Spotted Owl v. Hodel, 716 F. Supp. 479 (W.D. Wash. 1988). For a revealing discussion of some of the political battles fought over early listings, see S. Yaffe, Prohibitive Policy: Implementing the Federal Endangered Species Act (1982). (9.) See supra note 5. (10.) ESA [section] 3(6), 16 U.S.C. [section] 1532(6). (11.) Id. [section] 1532(20). (12.) However, a 1979 GAO report criticized the Secretaries' authority to list populations, prompting Congress to address the issue when it amended the ESA in 1979. General Accounting Office, Endangered Species--A Controversial Issue Needing Resolution 1, 52 (1979) [hereinafter 1979 GAO Report]. See Gleaves supra note 4, at n.29-34 and accompanying text. See infra notes 60-69 and accompanying text for further discussion of the congressional response to GAO's criticisms on this issue. (13.) See Rohlf, supra note 7, at 40. (14.) Id. (15.) Populations differentiated in this manner are termed "stocks." For example, Mid-Columbia summer chinook salmon spawn during the summer in tributaries to the middle portion of the Columbia River in east-central Washington. In addition to widespread use of the stock concept by resource managers, many biologists employ the term, including the authors of a seminal article documenting the decline of salmonids in the Pacific Northwest. See Willa Nelson, Jack Williams & James Lichatowich, Pacific Salmon at the Crossroads: Stocks at Risk from California, Oregon, Idaho, and Washington, 16 Fisheries 4 (1991). However, some biologists have challenged the stock concept. See Andrew Dizon et al., Rethinking the Stock Concept: A Phylogeographic Approach, 6 Conservation Biology 24 (1992). (16.) The Shoshone-Bannock Tribes asked NMFS to list Snake River sockeye on April 2, 1990. 55 Fed. Reg. 22,942, 22,943 (1990). A coalition of environmental and scientific organizations requested protection for four additional Columbia Basin stocks on June 7, 1990. 55 Fed. Reg. 37,342, 37,342 (1990). (17.) 56 Fed. Reg. at 58,618 (1991); see also Waples, supra note 3. (18.) 56 Fed. Reg. at 58,618 (1991). (19.) Id, (20.) Id. (21.) Id. (22.) Id. (23.) 56 Fed. Reg. at 58,618 (1991). (24.) Id. (25.) 57 Fed. Reg. 14,654 (1992). (26.) Id. at 14,658. (27.) Id. at 14,662. (28.) 58 Fed. Reg. 29,390 (1993). (29.) Letter from Mid-Columbia Public Utility Districts to NMFS (July 13, 1993) (on file with author). (30.) See infra part V. (31.) See infra notes 146, 163 and accompanying text. (32.) See infra note 203 and accompanying text. (33.) For legal commentary in addition to Gleaves, supra note 4, see Kevin Grierson, 7he Concept of Species and the Endangered Species Act, 11 Va. Envtl. L.J. 463 (1992); Kevin Hill, The Endangered Species Act: What Do We Mean by Species?, 20 Envtl. Aff. 239 (1993). This issue has caused a much greater proliferation of scientific debate. In addition to Dizon et al., supra note 15; see, e.g., Thomas Backman & Laura Berg, Managing Molecules or Saving Salmon? The Evolutionarily Significant Unit, 2 Wana Chinook Tymoo 8 (1992) (a publication of the Columbia River Inter-tribal Fish Commission); Stephen O'Brian & Ernst Mayr, Bureaucratic Mischief. Recognizing Endangered Species and Subspecies, 251 Science 1187 (1991); Martha Rojas, The Species Problem and Conservation: What Are We Protecting?, 6 Conservation Biology 170 (1992). An entire conference on this issue sponsored by the American Fisheries Society (AFS), entitled "Evolution and the Aquatic Ecosystem: Defining Unique Units in Population Conservation," took place on May 23-25 in Monterey, CA. AFS plans to publish a book containing papers given at the conference. (34.) See infra notes 4142 and accompanying text. (35.) See supra note 5. This definition, added when Congress amended the ESA in 1978, replaced an earlier definition of species which permitted listings of "any other group [in addition to subspecies) of fish or wildlife of the same species or smaller taxa in common spatial arrangement that interbreed when mature." Gleaves, supra note 4, at 28. (36.) There is virtually nothing in the ESA's legislative history that explains the policy reasons why Congress chose to exclude invertebrate populations from the term of distinct population segment when it amended the statute in 1978. Lawmakers could have conceivably believed protection of invertebrates to be less important than conserving vertebrates, but Congress' later admonition to NMFS and FWS not to favor so-called "higher" taxa such as vertebrates over others in establishing a listing priority scheme undercuts this explanation. See Rohlf, supra note 7, at 46. On the other hand, lawmakers may have felt that protecting populations entails additional management and social costs, and draw the line against protecting invertebrate populations because these creatures are almost inevitably less popular and thus the general public would be less willing to bear additional conservation costs for invertebrates. However, FWS has asserted that protecting populations may ultimately cost less than waiting until entire species or subspecies face extinction. See infra notes 171-173. Perhaps the best explanation of the current legal divide between vertebrates and invertebrates comes from a marine biologist who reviewed a draft of this article and commented that "this [distinction between vertebrate and invertebrate populations] is daft." (37.) ESA [sections] 4(b)(1)(A), 16 U.S.C. [sections] 1533(b)(1)(A). (38.) The ESA sets forth five factors which NMFS and FWS must consider in deciding whether a species is threatened or endangered, including habitat destruction, overutilization, disease or predation, inadequacy of existing regulatory protections, and other factors affecting the species' continued existence. Id. [sections] 1533(a)(1). (39.) Gleaves, supra note 4, at 37-38, 48. (40.) Id. at 45, 48. (41.) H.R. Rep. No. 412, 93d Cong., 1st Sess. 4-5 (1973). (42.) Id. (43.) For example, alkaloid chemicals found in plants have medicinal importance, including effective cancer-fighting ability. Yet humans have thus far examined fewer than three percent of the world's flowering plants for presence of alkaloids. E.O. Wilson, The Diversity of Life 283-85 (1992). (44.) ESA [sections] 2(b), 16 U.S.C. [sections] 1531(b). (45.) ESA [sections] 2(a)(3), 16 U.S.C. [sections] 1531(a)(3). (46.) S. Rep. No. 307, 93d Cong., 1st Sess. 2 (1973). (47.) H.R. Rep. No. 4126, 93d Cong., 1st Sess. (1973). (48.) H.R. Rep. No. 1625, 95th Cong., 2d Sess. 5 (1978). (49.) While Congress authorized FWS and NMFS to protect U.S. population of species also found elsewhere, the statute authorizes listing of species which do not occur in the United States. In fact, foreign species comprise more than one-third of species currently listed as threatened and endangered. See 50 C.F.R. [sections] 17.11 (1993). (50.) H.R. Rep. No. 412, 93d Cong., 1st Sess. 10 1973). (51.) S. Rep. No. 307, 93d Cong., 1st Sess. 3 (1973). (52.) ESA [sections] 2(a)(1), (5), 16 U.S.C. [sections] 1531(a)(1), (5) [emphasis added]. (53.) The ESA gave federal agencies authority to use existing funding mechanisms and land acquisition power to acquire land as part of their conservation programs for listed species. ESA [sections] 5, 16 U.S.C. [sections] 1534. The statute also establish a cooperative program through which the federal government encourages and provides funding to state programs aimed at protecting listed species. ESA [sections] 6, 16 U.S.C. [sections] 1535. (54.) H.R. Rep. No. 412, 93d Cong., 1st Sess. 10 (1973). (55.) S. Rep. No. 151, 96th Cong., 1st Sess. 7 (1979). (56.) Aldo Leopold provided perhaps the most succinct and pragmatic explanation for protecting U.S. populations, and even populations in the lower forth-eight states: Mere seems to be a tacit assumption that if grizzlies survive in Canada and Alaska, that is good enough. It is not good enough for me . . . . Relegating grizzlies to Alaska is about like relegating happiness to heaven; one may never get there." Aldo Leopold, A Sand County Almanac 277 (1966). (57.) Convention on Biological Diversity, opened for signature at United Nations Conference on Environment and Development June 5, 1992, 31 I.L.M. 818. For additional discussion on the issue of compensation for use of a country's biological resources--and indigenous people's knowledge about such resources--see S. Rubin & S. Fish, Biodiversity Prospecting: Using Innovative Contractual Provisions to Foster Ethnobotanical Knowledge, Technology, and Conservation, 5 Colo. J. Int'l Envt'l. L. & Pol'y 23 (1994). (58.) Convention on Biological Diversity, supra note 57, at Article 15. (59.) Gleaves, supra note 4, at 31-33. (60.) 1979 GAO Report, supra note 12; see also Gleaves, supra note 4, at 31-32 n.30. (61.) S. Rep. No. 151, 96th Cong., 1st Sess. 7 (1979). (62.) Id. (63.) ESA [sections] 3(6), (20), 16 U.S.C. [sections] 1532(6), (20). (64.) Legislative history from 1973 bolsters this interpretation. In the report accompanying its original version of the ESA, the Senate noted that "[f]lexibility in regulation is enhanced by a provision which allows for listing if the animal is endangered over a |substantial portion of its range.'" S. Rep. No. 307, 93d Cong., 1st Sess. 3 (1973). (65.) On the contrary, FWS expressly recognized a lack of any such differences when it listed bald eagles. See infra note 136 and accompanying text. (66.) Congress has consistently stressed that the ESA embodies a policy which favors species in cases of uncertainty. In its report accompanying the ESA in 1973, the House termed the statute's approach "the institutionalization of caution." H.R. Rep. No. 412, 93d Cong., 1st Sess. 5 (1973). The Conference Report accompanying the 1979 amendments noted that the statute "continues to give the benefit of the doubt to species, and it would continue to place the burden on the action agency" to demonstrate compliance with the ESA's protections. I.LR. Conf. Rep. No. 697, 96th Cong., 1st Sess. 12 (1979). While the above passages refer to the interagency consultation process pursuant to [sections] 7, they indicate an intent by Congress that agencies should favor species protection given cases of scientific uncertainty. (67.) See supra note 62 and accompanying text. (68.) S. Rep. No. 1512 at 7. (69.) Id. (70.) See, eg., Gleaves, supra note 4, at 26 nn.8, 10. (71.) Rojas, supra note 33, at 170. For further discussion about historic and recent changes in taxonomy, see Hill, supra note 33, at 247-53. (72.) See I. Hanski & Michael Gilpin, Metapopulation Dynamics. Brief History and Conceptual Domain, 42 Biological J. of the Linnean Soc'y 3 (1991); S. Harrison, Local Extinction in a Metapopulation Context: An Empirical Evaluation, 42 Biological J. of the Linnean Soc'y 73 (1991); Michael Gilpin, Spatial Structure and Population Vulnerability, in Viable Populations for Conservation 125-39 (Soule, ed., 1987). (73.) See S. Pickett & John Thompson, Patch Dynamics and the Design of Nature Reserves, 13 Biological Conservation 27 (1978). (74.) Christopher Frissell, Topology of Extinction and Endangerment of Native Fishes in the Pacific Nortwest and California, 7 Conservation Biology 342, 350 (1992). (75.) Daniel Rohlf & Christopher Frissell, Protecting Intraspecific Biodiversity Under the Endangered Species Act: An Evolutionary Past, the Ecological Present, and the Geography of an Uncertain Future 7-8 (unpublished manuscript, on file with author). (76.) See generally Robert Vrijenhoek, Population Genetics and Conservation, in Conservation for the Twenty-First Century 89-98 (D. Western & M. Pearl eds., 1989); Christine Schonewald-Cox et al, Genetics and Conservation (1983). (77.) Vrijenhoek, supra note 76, at 97. (78.) Leopold, supra note 56, at 190. (79.) John Terborgh, Keystone Plant Resources in the Tropical Forest, in Conservation Biology: The Science of Scarcity and Diversity 33044 (M. Soule ed., 1986). (80.) James Estes & John Palmisano, Sea Otters: Their Role in Structuring Nearshore Communities, 185 Science 1058 (1974). (81.) Gleaves, supra note 4, at 45-49. (82.) Waples, supra note 3, at 2. (83.) It is curious that NMFS' only published document which discusses (albeit briefly) the legal and policy elements within the ESA's definition of species was written by a geneticist. (84.) E.O. Wilson, The Current State of Biological Diversity, in Biodiversity (E.O. Wilson, ed. 1988). (85.) See supra note 76 and accompanying text. (86.) See supra notes 74-75 and accompanying text. (87.) NMFS admitted that its ESU policy embodies this approach in response to criticism that the ESU concept would allow habitat fragmentation to proceed to the point at which an entire biological species faced extinction. The agency argued that the "underlying concern should be whether important genetic resources of the biological species are at risk because of fragmentation. If so, then the appropriate action would be to protect the biological species as a whole rather than the individual fragments." 56 Fed. Reg. at 58,618 (1991). The problem with this approach, in addition to the fact that no protections attach until an entire species faces extinction, is that NMFS and FWS have historically based their decisions whether a species is threatened or endangered on a comparison to the species' past abundance. Under this method of making listing threshold determinations, an agency may not list a species at risk due to fragmentation because the sum of individuals within these populations may compare favorably with the species' original numbers. In other words, agencies may substantially underprotect species and their constituent populations to the extent that they place a great deal of importance on genetics in defining populations eligible for listing consideration, but do not also incorporate as an integral part of their listing threshold determinations genetic risks arising from habitat fragmentation. For an example of how such fragmentation has contributed to the decline of amphibians in two national parks, see David Bradford, Parinaz Tabatabai & David Graber, Isolation of Remaining Populations of the Native Prog, Rana muscosa, by Introduced Fishes in Sequoia and Kings Canyon National Parks, California, 7 Conservation Biology 882 (1993). (88.) See, eg., Richard Hutto, Susan Reel & Peter Landres, A Critical Evaluation of the Species Approach to Biological Conservation, 4 Endangered Species Update 1-3-, Daniel Rohlf, Six Biological Reasons Why the Endangered Species Act Doesn't Work--And What To Do About It, 5 Conservation Biology 273 (1991). (89.) 56 Fed. Reg. at 58,614 (1991). (90.) In its ESU policy statement, NMFS notes that it will consider an isolated population to be an important component in the evolutionary legacy of a species if "the population contributed substantially to the ecological/genetic diversity of the species as a whole." Id. at 58,618. However, NMFS' focus on "ecological diversity" looks at a population in isolation, not in reference to its contribution to a biotic community. In other words, NMFS will use information about a population's adaptation to a particular ecosystem only to infer its potential genetic differentiation; the agency believes "population characteristics that are evoluntionarily significant must have a genetic basis." Id. at 58,616. (91.) See, e.g., Michael Soule & Daniel Simberloff, What Do Genetics And Ecology Tells Us About the Design of Nature Reserves?, 35 BIOLOGICAL CONSERVATION 19, 35 (1986) ("[T]he ecological literature is rife with examples of recondite relationships among species that become apparent only when the decrease or disappearance of one has cascading detrimental effects on the others ...."). (92.) 56 Fed. Reg. at 58,621 (1991). (93.) Id. at 58,621-22. (94.) Id. at 58,621. (95.) Id. at 58,613 (1991). (96.) The term "distinct population segment" itself serves as evidence that the ESA's population concept is not solely biological. In addition to its biologically inexplicable application to vertebrates only, supra note 36, this term is not used within the scientific community. Biologists commonly employ the term population,' but the term "distinct population segment" is conspicuously absent in scientific literature. (97.) See supra note 61. (98.) Lack of reproductive isolation would likely lead to definition of all bald eagles in North America as one ESU. Assuming threats to eagles in the lower forty-eight states satisfied the ESA's listing threshold, the entire ESU would be listed. (99.) See ESA [sections] 7(a)2, 16 U.S.C. [sections] 1536(a)(2). Such a listing would also require the Secretary to develop a recovery plan for bald eagles throughout North America, even though the birds are not in trouble in vast areas. See id. [sections] 1533(f). (100.) 56 Fed. Reg. at 58,617 (1991). (101.) See, e.g., id. at 58,613 ("Expert scientific judgment is required in determining what should be considered distinct populations."). (102.) Waples, supra note 3, at v. (103.) Gleaves, supra note 4, at 45. (104.) See Waples, supra note 3, at 10-11. (105.) See Hill, supra note 33, at 253. (106.) See Waples, supra note 4, at 11. (107.) See 56 Fed. Reg. at 58,615 (1991) ("this issue has been debated by evolutionary biologists for over two decades"). (108.) Id. at 58,618. (109.) Id. (110.) Id. This definition simply describes to the reproductive isolation criterion in terms of the second ESU criterion, which is also vaguely defined. (111.) Id.; see also supra note 22. (112). 56 Fed. Reg. at 58,618 (1991). (113.) See Waples, supra note 3, at 12, 26 ("Although a variety of approaches may prove useful in making [a] determination, none will provide a completely objective assessment of evolutionary significance."). (114.) See Gleaves, supra note 4, at 45-46. (115.) See Waples, supra note 3, at 21. (116.) See supra note 62 and accompanying text. (117.) See supra note 66 and accompanying text. (118.) 58 Fed. Reg. 29,390 (1993). (119.) Id. at 29,391. (120.) Id. at 29,392. (121.) G. Thompson, Determining Minimum Viable Populations Under the Endangered Species Act (1991) (draft manuscript, on file with author). (122.) Id. at 25-26. (123.) Id. at 28. It suggested that a "threatened" species is one which has at least a 50 percent chance of becoming endangered within 10 years. Id. (124.) Thompson, after reviewing various methods employed by scientists to determine what constitutes a viable population, suggested that "conventional wisdom" established the parameters he used to define an endangered species. Id. However, a prominent conservation biologist has pointed out that these parameters are "open questions" which depend on how much risk of extinction one is willing to accept. Mark Shaffer, Minimum Population Sizes for Species Conservation, 31 Bioscience 131 (1981). Determining what constitutes an "acceptable" risk of extinction involves making decisions about tradeoffs between the benefits of protecting species and the costs of protection--a policy rather than scientific choice. Hence, policymakers must first specify a standard of security in order for biologists to then determine whether specific species or populations fall above or below that standard. Congress largely failed to do this in the ESA, which contains extremely vague definitions of 'endangered' and "threatened." See ESA [sections] 3(6), (2), 16 U.S.C. [sections] 1532(6), (2). See also Rohlf, supra note 88, at 275-76; M. Soule, Viable Populations for Conservation (M. Soule ed., 1987). (125.) An additional criticism of the ESU policy not discussed here relates to the question of how the policy treats individuals or groups of organisms produced through artificial propagation. The NMFS' technical paper applying the ESU concept to Pacific salmon asserts that "only naturally-spawning populations should be considered in determining whether a population is distinct for purposes of the Act," except when hatchery populations "represent the only remaining component of a native gene pool." See Waples, supra note 3, at 18. Native American tribes in the Pacific Northwest, which view hatcheries and supplementation' programs (i.e., planting hatchery-reared juveniles in streams in order to restore runs of naturally-spawning fish) as playing an important role in protecting and enhancing the region's dwindling salmon runs, have vociferously criticized this position. See Columbia River Inter-tribal Fish Commission, Briefing Paper on Species Definition (1991) (on file with author); See Backman & Berg, supra note 33. On the other hand, others take the position that hatchery practices have contributed to salmon decline rather than ameliorating it. See Michael Goodman, Preserving the Genetic Diversity of Salmonid Stocks: A Call for Federal Regulation of Hatchery Programs, 20 Envtl. L. ill (1990).

In addition to its discussion of the issue in Waples, supra note 3, NMFS has issued a technical paper on the role of artificial propagation of salmonids. See Jeffry Hard, Robert Jones, Michael Delarm & Robin Waples, Pacific Salmon and Artificial Propagation Under the Endangered Species Act, NOAA Technical Memorandum NMFS-NWFSC-2 (1992). Like the ESU concept itself, NMFS' position could influence policy related to other species recovery efforts which rely on artificial propagation. Under what circumstances organisms produced by artificial means should qualify for protection under the ESA is a sort of 1990s version of debates over whether the statute protects hybrids. See infra notes 138-144 and accompanying text for a discussion of FWS' hybrid policy. (126.) 43 Fed. Reg. 6230 (1978). (127.) Id. at 6231. (128.) 40 Fed. Reg. 44,412 (1975). (129.) 43 Fed. Reg. at 32,803 (1978). (130.) Id. at 32,801. (131.) See supra note 126; 40 Fed. Reg. at 31,734 (1975). (132.) The notice discusses only problems facing bears in the Lower 48 states, concluding that "this species" could become endangered in the foreseeable future "throughout its range." 40 Fed. Reg. at 31,734 (1975). (133.) 40 Fed. Reg. at 44,412. (134.) 43 Fed. Reg. at 6232 (1978). (135.) 40 Fed. Reg. at 44,413-14. (136.) 43 Fed. Reg. at 6231. (137.) Id. at 6232. (138.) Since another commentator has already provided an excellent discussion of the hybrid policy, this analysis will summarize it only briefly. See Hill, supra note 33. (139.) Hill, supra note 33, at 243-44. (140.) Dusky Seaside Sparrows were a subspecies of Seaside Sparrows, found in marshes along the Atlantic and Gulf coasts of the United states. By 1981, only five males remained. Biologists hoped to save the subspecies by breeding the males with a morphologically similar subspecies, then "back-crossing" female offspring to Dusky males. Even though sixth-generation offspring would have been 98.4% Dusky, FWS withdrew its support from the program due to its hybrid policy, and the subspecies essentially became extinct by 1989. See id. at 257-59. (141.) FWS determined that the ESA did not protect hybrids between gray wolves and red wolves. See id. at 245-46. (142.) Id. (143.) Memorandum from Assistant Solicitor, Fish and Wildlife, to Director, U.S. Fish and Wildlife Service 2 (Dec. 14, 1990) (on Me with author). (144.) Id. at 1. (145.) 55 Fed. Reg. 12,178 (1990). (146.) 57 Fed. Reg. 28,474, 28,476 (1992). (147.) 58 Fed. Reg. 12,864, 12,868 (1993). (148.) 57 Fed. Reg. 33,478 (1992). FWS later in 1992 listed murrelets in northern California, Oregon, and Washington as threatened after environmental groups obtained a court order forcing the agency to make an immediate listing decision. See 57 Fed. Reg. 45,328 (1992). (149.) 57 Fed. Reg. 1246-1247 (1992). (150.) 57 Fed. Reg. 588, 592 (1992). (151.) For example, FWS cited evidence of "variations in genetic composition" to support its distinction between the Mojave and Sonoran populations of desert tortoises. 55 Fed. Reg. 12,178 (1990). On the other hand, FWS brushed aside criticism of its listing of Louisiana black bears which argued that the genetic characteristics of black bear subspecies were not conclusively different. 57 Fed. Reg. at 589 (1992). (152.) In contrast to this decision, FWS listed the entire Pacific coast population of western snowy plovers--including the portion found in Mexico. The agency did not discuss the international border as a means of dividing the population. See 58 Fed. Reg. 12,864 (1993). (153.) 53 Fed. Reg. 52,747, 52748 (1988). (154.) 57 Fed. Reg. 45,328, 45,330 (1992). (155.) This approach is consistent with the reasoning in the 1979 Senate Report. See supra notes 62-64 and accompanying text. (156.) See infra section IV. (157.) 57 Fed. Reg. 546, 547 (1992). (158.) 57 Fed. Reg. 28,474, 28,476 (1992). (159.) 56 Fed. Reg. 1159, 1160 (1991). (160.) Id. (161.) 57 Fed. Reg. 1246, 1249 (1992). (162.) 57 Fed. Reg. 588, 589 (1992). (163.) Memorandum (plus attachment) from FWS Deputy Director to Regional Directors (June 15, 1992) (on file with author) [hereinafter Draft Policy]. (164.) Id. at 1. (165.) 57 Fed. Reg. 51,472 (1992). (166.) Draft Policy, supra note 163, at 4. (167.) Id. at 5. (168.) Id. at 4. (169.) Id. (170.) Id. at 3. (171.) Draft Policy, supra note 163, at 6. (172.) Id. (173.) Id. (174.) Telephone Interview with Jim Bartel, Chief, Division of listing and Recovery, FWS, Portland Field Office (Apr. 12, 1994). (175.) Id. (176.) Id. (177.) Id. (178.) Id. Even if a population is "significant" because it occupies a gap in the broader species' range, it apparently still must meet the reproductive isolation requirement. (179.) 58 Fed. Reg. 36,924 (1993). As a result of a court challenge, FWS agreed to complete a rangewide status review of the species. 59 Fed. Reg. 4887 (1994). (180.) 59 Fed. Reg. 869 (1994). (181.) Id. at 870. (182.) Id. (183.) Id. (184.) See Gleaves, supra note 4, at 38-40. (185.) Marbled Murrelet v. Lujan, No. C91-522R (W.D. Wash. Sept. 17, 1992). (186.) Endangered Species Committee of the Building Industry Association of Southern California v. Babbitt, No. Civ. 92-2610 (D.D.C. May 2, 1994) [hereinafter Endangered Species Committee]. (187.) Marbled Murrelet, slip op. at 4. (188.) Id. at 7. (189.) Id, at 12. (190.) Id. (191.) Id. at 16. This language almost directly incorporates text from the agency's 1992 draft policy. See supra note 163 and accompanying text. (192.) Id. at 22. (193.) See supra note 63. (194.) If one substitutes distinct population segment for the term "species" in the definitions of threatened species and endangered species, the result is nonsensical. In this context, the statue would refer to a distinct population segment "in danger of extinction throughout all or a significant portion of its range." See 16 U.S.C. [sections] 1532(6). What would constitute a significant portion of the range of a distinct population segment? (195.) See supra 163 and accompanying text. (196.) 58 Fed. Reg. 16,742 (1993). (197.) Endangered Species Committee, supra note 186, slip op. (198.) Id. at 6. (199.) Id. (200.) See supra notes 108-112 and accompanying text. (201.) The district court in the gnatcatcher case did not indicate the circumstances under which raw data would be considered "readily available" to the listing agency, nor did it indicate precisely which "interested parties" to a listing decision had the right to demand that the agency provide them with raw data. While these holdings could conceivably be limited to the distinctive facts of the gnatcatcher case, the court's broad language will likely lead to similar litigation regarding access to data, (202.) Direct substantive challenges to agencies' listing decisions probably stand relatively little chance of success. See supra note 114 and accompanying text. Nevertheless, such litigation could further tie up agency resources and possibly delay or interfere with species conservation efforts. (203.) Discussions on this issue between NMFS and FWS have actually been ongoing for several years. The agencies held a workshop on the status of populations under the ESA in June 1990. See Waples, supra note 3, at 26. Talks between NMFS and FWS remain ongoing as of this writing, although NMFS appears comfortable with its policy and may not have as a top priority accommodating the views of FWS. See supra note 174. (204.) Letter to Interior Secretary Bruce Babbitt (May 2, 1994) (on file with author). (205.) Yaffee, supra note 8, at 135. (206.) See supra notes 174-183 and accompanying text. (207.) Leopold, Supra note 56, at 194 (emphasis in original).

Daniel J. Rohlf, Senior Advocacy Fellow and Adjunct Professor, Northwestern School of Law of Lewis and Clark College. J.D. Stanford University, B.A Colorado College. The author wishes to thank the staff of the Columbia River Inter-tribal Fish Commission and Professor Deborah Brosnan of Lewis and Clark College for their very helpful input. The author also acknowledges the support of the Bullitt Foundation, which made this work possible.