[Federal Register Volume 82, Number 66 (Friday, April 7, 2017)]
[Proposed Rules]
[Pages 16981-16988]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2017-06995]


=======================================================================
-----------------------------------------------------------------------

DEPARTMENT OF THE INTERIOR

Fish and Wildlife Service

50 CFR Part 17

[Docket No. FWS-R2-ES-2015-0148; 4500030113]
RIN 1018-BA86


Endangered and Threatened Wildlife and Plants; Threatened Species 
Status for the Headwater Chub and Roundtail Chub Distinct Population 
Segment

AGENCY: Fish and Wildlife Service, Interior.

ACTION: Proposed rule; withdrawal.

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

SUMMARY: We, the U.S. Fish and Wildlife Service (Service), withdraw the 
proposed rule to list the headwater chub (Gila nigra) and a distinct 
population segment (DPS) of the roundtail chub (Gila robusta) from the 
lower Colorado River basin as threatened species under the Endangered 
Species Act (Act). This withdrawal is based on a thorough review of the 
best scientific and commercial data available, which indicate that the 
headwater chub and

[[Page 16982]]

the roundtail chub DPS are not discrete taxonomic entities and do not 
meet the definition of a species under the Act. These fish are now 
recognized as a part of a single taxonomic species--the roundtail chub 
(Gila robusta). Because the entities previously proposed for listing 
are no longer recognized as species, as defined by the Act, we have 
determined that they are not listable entities and we are withdrawing 
our proposed rule to add them to the List of Endangered and Threatened 
Wildlife.
    Section 4(b)(6) of the Act and implementing regulations at 50 CFR 
424.17 provide that the Service must, within 1 year of a proposed rule 
to list, delist, or reclassify species, or to designate or revise 
critical habitat, withdraw the proposal if the available evidence does 
not justify the proposed action. The document withdrawing the rule must 
set forth the basis upon which the proposed rule has been found not to 
be supported by available evidence. Once withdrawn, the action may not 
be re-proposed unless sufficient new information is available.

FOR FURTHER INFORMATION CONTACT: Steve Spangle, Field Supervisor, U.S. 
Fish and Wildlife Service, Arizona Ecological Services Office, 9828 
North 31st Ave., #C3, Phoenix, AZ 85051-2517; telephone 602-242-0210. 
Persons who use a telecommunications device for the deaf (TDD) may call 
the Federal Relay Services at 800-877-8339.

SUPPLEMENTARY INFORMATION: 

Previous Federal Action

    On October 7, 2015 (80 FR 60754), we published a proposed rule to 
list the headwater chub and the lower Colorado River basin roundtail 
chub DPS (roundtail chub DPS) as threatened species under the Act (16 
U.S.C. 1531 et seq.). On August 15, 2016 (81 FR 54018), we announced a 
6-month extension on the final listing determination that the Act 
allows when there is substantial disagreement regarding the sufficiency 
or accuracy of the available data, and reopened the comment period on 
the proposed listings for 30 days. During this comment period we 
received new information. On November 1, 2016 (81 FR 75801), we 
reopened the comment period on the proposed listings for an additional 
45 days to provide the public additional time to review and consider 
the proposed rulemakings in light of this new information. As a result 
of the 6-month extension, the deadline to finalize, modify, or withdraw 
the proposed rule is April 7, 2017.
    For a description of additional previous Federal actions concerning 
these species, please refer to the October 7, 2015, proposed listing 
rule (80 FR 60754).

Background

    At the time we published our proposed rule (October 7, 2015; 80 FR 
60754), the Committee on Names of Fishes, a joint committee of the 
American Fisheries Society and American Society of Ichthyologists and 
Herpetologists (the Societies) (Page et al. 2013, p. 71), considered 
headwater chub and roundtail chub to be separate species. As a 
consortium of fisheries scientists, the American Fisheries Society is 
the recognized and accepted scientific authority on fish taxonomy. 
Accordingly, our proposed rule assessed the headwater chub and 
roundtail chub as separate species. However, commenters on our proposed 
rule raised questions during the public comment period regarding the 
taxonomic distinctness of the headwater and roundtail chubs, as related 
to the Gila chub (Gila intermedia). At that time, some scientists 
knowledgeable about the fish contended that the three entities were not 
separate species (Carter et al. 2016 in press; Copus et al. 2016). For 
this reason, the Arizona Game and Fish Department requested that the 
Societies evaluate the most recent literature associated with roundtail 
chub, headwater chub, and Gila chub taxonomy. In their final report to 
the Arizona Game and Fish Department, the Societies panel concluded 
that ``no morphological or genetic data define populations of Gila in 
the lower Colorado River basin (which, as defined by the Service, 
includes the Little Colorado River, Bill Williams River, Gila River, 
Verde River, and Salt River drainages) as members of more than one 
species'' and ``that the data available support recognition of only one 
species of Gila, the roundtail chub, Gila robusta'' (Page et al. 2016, 
p. 1). These three fish are now considered by the Societies to be a 
single species, roundtail chub (Gila robusta) because data do not 
support recognition of three species.

Taxonomy

Introduction
    The taxonomic history of the genus Gila in the Colorado River basin 
has changed over time, especially for the three forms (roundtail, 
headwater, and Gila chub) found in the Gila River basin. These forms 
have been variously classified as full species, assigned as different 
species, subspecies of Gila robusta, or as part of a ``Gila robusta 
complex'' (Miller 1945; Holden 1968; Rinne 1969; Holden and Stalnaker 
1970; Rinne 1976; Smith et al. 1977; DeMarais 1986; Rosenfeld and 
Wilkinson 1989; Dowling and DeMarais 1993; Douglas et al. 1998; 
Minckley and DeMarais 2000; Gerber et al. 2001). As noted by nearly all 
researchers investigating the systematics of Gila spp., the taxonomic 
situation is complicated and problematic (Holden and Stalnaker 1970; 
Minckley 1973; Minckley and DeMarais 2000; Gerber et al. 2001; 
Sch[ouml]nhuth et al. 2014) due to various factors including multiple 
independent hybridization events over time (Rinne 1976; DeMarais 1986; 
Rosenfeld and Wilkinson 1989; DeMarais et al. 1992; Dowling and 
DeMarais 1993; Minckley and DeMarais 2000; Gerber et al. 2001; Schwemm 
2006; Sch[ouml]nhuth et al. 2014; Brandenburg et al. 2015,) potential 
past introgression (the transfer of genetic information from one 
species to another as a result of hybridization between them and 
repeated backcrossing) (DeMarais et al. 1992; Minckley and DeMarais 
2000), recent divergence within the three fish (Schwemm 2006). Further, 
the original assignment to species was based on the assumption that the 
three fish do not overlap geographically (parapatry), which we 
recognize now is not an accurate assumption. Additionally, in some 
instances when the same fish was identified based on morphology 
(physical characteristics) it was identified as one species and when 
identified based on genetic analysis it was identified as a different 
species (Dowling et al. 2015, pp. 14-15). Recent and ongoing genetic 
and morphologic analyses of chubs in the Gila River basin continue to 
yield conflicting results (DeMarais et al. 1992; Schwemm 2006; Dowling 
et al. 2008 and 2015; Sch[ouml]nhuth et al. 2014; Marsh et al. 2016, 
all entire).
History
    Gila robusta (roundtail chub) was first described by Baird and 
Girard (1853, p. 365-369) from specimens collected in 1851 from the 
Zuni River (tributary to Little Colorado River). Gila nigra (headwater 
chub; formerly known as G. robusta graham or G. grahami) was first 
described as a subspecies (G. robusta graham) from Ash Creek in the San 
Carlos River in east-central Arizona in 1874 (Cope and Yarrow 1875, p. 
663), but not returned to full species status (G. robusta) until 
proposed so by Minckley and DeMarais (2000, p. entire). The Societies 
accepted Gila nigra as a full species (Nelson et al. 2004, p. 71), as 
did the New Mexico Department of Game, Fish (Carman 2006, p. 3), 
Arizona Game, and Fish

[[Page 16983]]

Department (AGFD 2006, p. 3) and continued to recognize G. robusta as a 
distinct species. Therefore, based on the best available commercial and 
scientific data the Service accepted both Gila robusta and Gila nigra 
as full species as documented in our 12-month findings (May 3, 2006; 71 
FR 26007 and July 7, 2009; 74 FR 32352). In their 2013 publication of 
Common and Scientific Names of Fishes from the United States, Canada, 
and Mexico, the Societies continued to list both Gila robusta and Gila 
nigra as distinct species (Page et al. 2013, p. 71). A summary of the 
historic and current nomenclature from Rinne (1976, entire), Sublette 
et al. (1990, entire), and Minckley and DeMarais (2000, entire) is 
summarized in Voeltz (2002, pp. 8) and Copus et al. (2016, pp. 1&6). 
The Gila chub (Gila intermedia) is currently listed as an endangered 
species (November 2, 2005; 70 FR 66664).
    These entities were originally classified based on the streams in 
which they were found (Minckley and DeMarais 2000, p. 252), under the 
assumption that G. robusta and G. nigra either did not overlap 
(allopatric, no gene flow) or there was only a narrow overlap 
(parapatric; limited interaction and opportunity for gene flow) 
(Minckley and DeMarais 2000 pp. 252-254). Because hybridization between 
G. robusta and G. intermedia indicates that these fish must co-occur in 
some streams (Minckley and DeMarais 2000, entire), we conclude that 
Minckley and DeMarais's (2000) assumption they did not overlap was 
unfounded. Further, other studies have found that fish designated as G. 
robusta, G. nigra, and G. intermedia overlap geographically or occur 
adjacent to one another (Dowling and Marsh 2009, p. 1; Marsh et al. 
2016, p. 57; Brandenburg et al. 2015, p. 18).
Morphology
    The approach for classifying G. robusta, G. nigra, and G. 
intermedia developed by Minckley and DeMarais (2000, pp. 254-255) 
presumes there is little intraspecific variation (differences within a 
species) in the morphologic and meristic (counting quantitative 
characteristics such as fins) characteristics used to distinguish these 
three taxa. However, the three purported species overlap in physical 
characteristics, and many fish have intermediate physical 
characteristics. Those characteristics that do not overlap are 
separated by very small margins, making species-level identification of 
individual fish problematic, even when the geographic origin of the 
species is known (Brandenburg 2015, entire). Minckley and DeMarais 
(2000, pp. 253-254) indicate that G. nigra is physically different from 
G. intermedia even though they appear physically more similar to one 
another than either is to G. robusta. In addition, Copus et al. (2016, 
p. 13) did not find physical characteristics in the Minckley and 
DeMarais (2000, pp. 254-255) classification key to reliably 
differentiate G. robusta, G. nigra, and G. intermedia from one another. 
Copus et al. (2016 p. 16) concluded that there was no morphological 
basis for taxonomic distinctions within the Gila spp. complex.
Genetics
    Multiple genetic analysis studies have been conducted that reveal 
differences between different chub populations, but have been unable to 
identify differences between G. robusta, G. nigra, and G. intermedia 
(DeMarais et al. 1992, pp. 2748-2749; Schwemm 2006, p. 29; Dowling et 
al. 2008, p. 2, and 2015, p. 13; Copus et al. 2016, pp. 14-15; Marsh et 
al, 2016, p.58). Mitochondrial DNA analysis (Sch[ouml]nhuth et al. 
2014, p. 223) indicates that G. robusta, G. nigra, and G. intermedia 
belong to one clade (a grouping that includes a common ancestor and all 
its descendants, living and extinct, of that ancestor). Sch[ouml]nhuth 
et al. (2014, p. 223) hypothesized that this could reflect 
hybridization or incomplete lineage sorting (when the lineage of a 
specific gene is not the same as the lineage of the species, obscuring 
the true species relationship).
    However, when nuclear DNA (rather than mitochondrial DNA) was 
analyzed, a broader grouping was identified that included G. seminude 
and G. elegans, but when mitochondrial and nuclear DNA results are 
combined G. robusta, G. nigra, and G. intermedia were in one grouping 
(Sch[ouml]nhuth et al. 2014, p. 223). Preliminary studies by Chafin et 
al. (2016) indicate evolutionary independent lineages for G. robusta, 
G. nigra, and G. intermedia, and that the hybrid origin of G. nigra is 
not supported. Studies by Marsh et al. (2016, entire) point to genetic 
variation between populations of G. robusta and G. nigra, and 
demonstrate evidence that distinct ecological differences between some 
populations are now thought to exist. Minckley and DeMarais (2000, 
entire) supported recognition of three species, but acknowledged that 
most genetic variation was within populations for G. robusta, and was 
among populations for G. intermedia and G. nigra. Minckley and DeMarais 
(2000, p. 253) also indicated that these three fishes share genetic 
features (that had been studied so far) while behaving as separate non-
overlapping (allopatric) morphological species. In addition, some 
populations assigned to species based on genetics appeared to conflict 
with the species level-assignment based on morphology (Dowling et al. 
2008, p. 27).
Speciation
    Minckley and DeMarais (2000, p. 253) describe three different 
taxonomic options for chubs in the Gila River basin: a single species 
with many different forms or stages (polymorphic species), a species 
containing multiple subspecies, or three full species. They acknowledge 
that none of these taxonomic options is biologically justified without 
knowing if these fish naturally occur in the same geographic area 
(sympatry, indicating an initial interbreeding population that split), 
or occur immediately adjacent to each other but not significantly 
overlapping (parapatry, indicating there is no barrier to gene flow). 
They further acknowledge that a persistent narrow interaction zone 
(parapatry, indicating there is no barrier to gene flow) of 
morphologically distinguishable G. robusta, G. intermedia, and G. nigra 
has been confirmed, but note that in no instance was any two of the 
three caught at the same locality (allopatric, no gene flow; p. 251). 
However, they also acknowledge that hybridization (between G. robusta 
and G. intermedia, resulting in G. nigra) in the past must have 
occurred in some places and not others, thereby demonstrating 
occurrence in the same geographic area (sympatry) (p. 253). They 
conversely hypothesized that the current minimal overlap in an area 
where species are adjacent (parapatry, indicating there is no barrier 
to gene flow) may thus reflect an ancestral ecological segregation area 
(sympatry, indicating an initial interbreeding population that split 
due to the use of different habitats and resources) that promoted 
persistence in the ever-increasing aridity of the Southwest (p. 253).
    In Fossil Creek, G. nigra and G. robusta appear to be sympatric, 
including hybrids between G. robusta and G. nigra (Marsh et al. 2016, 
p. 57). Brandenburg et al. (2015, p. 18) concluded that the 
morphological assessment of Gila spp. in New Mexico confirmed that the 
three fish were found in the same geographic area (sympatric) in almost 
all cases, contradicting Minckley and DeMarais' results (2000, p. 251) 
as well as other previous literature suggesting that these Gila spp. 
are occurring in separate non-overlapping geographical areas 
(allopatric) through their ranges (Rinne

[[Page 16984]]

1969, p. entire; DeMarais 1986, p. entire; Minckley and DeMarais 2000, 
p. 253). In Fossil Creek, they found that G. nigra and G. robusta are 
locally in the same geographic area (sympatric) and have hybridized 
(Marsh et al. 2016, p. 57). Marsh et al. (2016, p. 58) concluded there 
are two morphologically similar, but genetically distinguishable, chub 
in Fossil Creek, G. robusta and G. nigra.
Conservation Implications
    Dowling et al. (2015, pp. 14-15) reasoned that the lack of 
diagnostic molecular characteristics does not inform the status of 
these three fish, but rather highlights the role that local evolution 
has played in shaping patterns of variation in these taxa and the 
importance of accounting for this variation when managing the complex. 
Most, if not all, scientists agree that conservation actions for these 
chubs must be directed at the population level and must include 
consideration of the complex as a whole (Dowling et al. 2008, pp. 30-
31; Dowling and DeMarais 1993, p. 445; Gerber et al. 2001, p. 2037; 
Schwemm et al. 2006, pp. 32-33). The Arizona Game and Fish Department 
recognizes the importance of conserving the currently recognized 
roundtail chub population rangewide (including the formerly known 
headwater chub and Gila chub) and is committed to the conservation 
agreements and practices that have been in place since 2006 (AGFD 2017, 
entire; AGFD 2006, entire).

Public Comments

    In our October 7, 2015 proposed rule (80 FR 60754), we requested 
that all interested parties submit comments or information concerning 
the proposed listings during a 60 day comment period, ending December 
7, 2015. We particularly sought comments concerning genetics and 
taxonomy. In our August 15, 2016, 6-month extension document (81 FR 
54018), we reopened the comment period on the proposed rule for 30 
days, ending September 14, 2016, and we again requested comments and 
information regarding genetics and morphology that would aid in 
resolving the ongoing taxonomic issues regarding classification of 
these fish. On November 1, 2016 (81 FR 75801, we announced an 
additional 45-day comment period, ending December 16, 2016, on the 
October 7, 2015 proposed rule.
    We provided notification of these publications and their comment 
periods through email, letters, and news releases faxed and/or mailed 
to the appropriate Federal, State, and local agencies; county 
governments; elected officials; media outlets; local jurisdictions; 
scientific organizations; interested groups; and other interested 
parties.
    In accordance with our peer review policy published in the Federal 
Register on July 1, 1994 (59 FR 34270), we solicited independent 
opinions from at least three knowledgeable individuals who have 
expertise with these fish, who possess a current knowledge of the 
geographic region where the fish occurs, and/or are familiar with the 
principles of conservation biology.
    We reviewed all comments received from peer reviewers and the 
public for substantive issues and new information regarding the 
proposed listing of G. nigra and the G. robusta DPS. Substantive 
comments pertaining to the taxonomy of these fish received during the 
comment period are addressed below. We also received several comments 
from both the public and peer reviewers concerning threats to these 
fish; however, because our withdrawal is due to taxonomic revision such 
comments are outside the scope of this withdrawal.

Peer Review Comments

    (1) Comment: One peer reviewed stated that there are no recent 
(since 2000) publications in the peer-reviewed literature that provide 
evidence that Gila intermedia, G. nigra, and G. robusta are other than 
separate and distinct species. The peer reviewer further stated that 
there are articles that study the genetics or morphology of these fish 
without questioning its taxonomy, specifically Sch[ouml]nhuth et al. 
2014, Sch[ouml]nhuth et al. 2012, and Marsh et al. in press.
    Response: Multiple studies since 2000 provide information on the 
genetic analysis for these fish, including Schwemm 2006, Dowling et al. 
2008 and 2015, and Copus et al. 2016. While these studies may not have 
questioned the taxonomic classification, they also have not been able 
to identify genetic markers that have the ability to distinguish among 
G. robusta, G. nigra, and G. intermedia. Sch[ouml]nhuth et al. (2008, 
p. 213; 2014, p. 223), using mitochondrial and nuclear DNA sequencing, 
found that G. robusta, G. nigra, and G. intermedia were well supported 
as having a common ancestor. Using mitochondrial DNA, Sch[ouml]nhuth et 
al. (2008, p. 213; 2014, p. 223) found that G. robusta, G. nigra, and 
G. intermedia were in one grouping that included a common ancestor and 
all the descendants (living and extinct) of that ancestor (clade), and 
this could reflect incomplete lineage sorting or hybridization. 
However, when nuclear DNA was analyzed, a broader grouping was 
identified that included G. seminuda and G. elegans, but when 
mitochondrial and nuclear DNA results were combined, G. robusta, G. 
nigra, and G. intermedia were alone in one grouping. While Marsh et al. 
(2016, entire) concluded there are two similar but genetically 
distinguishable species in the creek they studied, their findings 
differ somewhat from Schwemm (2006) and Dowling et al. (2008 and 2015, 
entire), who were unable to conclusively identify distinct species 
using genetic markers across a much wider range. Further, the Societies 
conducted a review of the literature and found no evidence to support 
three species. The Service has reviewed the best available scientific 
and commercial data and also found a lack of sufficient evidence to 
support more than one species.
    (2) Comment: Recognized authorities on the taxonomy and ecology of 
these fish recognized these fish as separate species based on 
morphological diagnostics.
    Response: Minckley and DeMarais (2000), Miller et al. (2005), and 
Minckley and Marsh (2009) report identification of three species using 
a diagnostic morphological key. However, additional reports were unable 
to reliably identify these three fish to species using the same 
diagnostic key (Carter et al. 2016, p. 2 and 20, in press; Brandenburg 
2015, entire; Copus et al. 2016, p. 13). Further, Minckley and DeMarais 
(2000, pp. 253-254) stated that G. nigra is morphologically separate 
from G. intermedia, but that G. nigra and G. intermedia appear 
morphologically more similar to one another than either is to G. 
robusta. In addition to issues surrounding morphological 
identification, multiple genetic analysis studies have found 
population-level differences, but have been unable to identify genetic 
markers that have the ability to distinguish among G. robusta, G. 
nigra, and G. intermedia (DeMarais 1992, pp. 2748-2749; Schwemm 2006, 
p. 29; Dowling et al. 2008, p. 2, and 2015, p. 13; Copus et al. 2016, 
pp. 14-15). There are also the findings of Sch[ouml]nhuth et al. 
(2014), Sch[ouml]nhuth et al. (2012) as described in Response to 
Comment 1.
    (3) Comment: Conclusions are mainly based on two ``gray 
literature'' reports that have not undergone peer review (Copus et al. 
2016) or were not available for public consideration (Carter et al. 
2016, in press).
    Response: Section 4(b)(1)(A) of the Act requires the Service to 
make listing or delisting decisions based on the best scientific and 
commercial data available. Further, our Policy on Information Standards 
under the Act

[[Page 16985]]

(July 1, 1994; 59 FR 34271), the Information Quality Act (section 515 
of the Treasury and General Government Appropriations Act for Fiscal 
Year 2001 (Pub. L. 106-554; H.R. 5658)), and our associated Information 
Quality Guidelines (https://www.fws.gov/informationquality), provide 
criteria and, guidance, and establish procedures to ensure that our 
decisions are based on the best scientific data available. They require 
us, to the extent consistent with the Act and with the use of the best 
scientific data available, to use primary and original sources of 
information as the basis for our determinations. Primary or original 
information sources are those that are closest to the subject being 
studied, as opposed to those that cite, comment on, or build upon 
primary sources. The Act and our regulations do not require us to use 
only peer-reviewed literature, but instead they require us to use the 
``best scientific and commercial data available.'' We use information 
from many different sources, including articles in peer-reviewed 
journals, scientific status surveys and studies completed by qualified 
individuals, Master's thesis research that has been reviewed but not 
published in a journal, other unpublished governmental and 
nongovernmental reports, reports prepared by industry, personal 
communication about management or other relevant topics, conservation 
plans developed by States and counties, biological assessments, other 
unpublished materials, experts' opinions or personal knowledge, and 
other sources. For these reasons, we think it is appropriate to include 
review of Copus et al. (2016) and Carter et al. (2016, in press), as 
well as other sources, within our review.
    (4) Comment: Several authors presented data and conclusions that 
conflicted with the previously cited Carter et al. (2016, in press) and 
Copus et al. (2016) reports pertaining to morphological identification, 
DNA analysis, and ecological equivalency to a subset of the Joint 
Committee convened in April 2016, to specifically address the taxonomy 
of the roundtail chub complex.
    Response: We were present at the April 2016 Joint Committee 
webinar, and experts beyond Carter and Copus, such as Brandenburg, 
Schwemm, Dowling, O'Neill, and Chafin, also provided information based 
on research they either had previously conducted or are currently 
conducting on Gila. A complete list of references cited may be obtained 
on the Internet at http://www.regulations.gov and upon request from the 
Arizona Ecological Services Office (see FOR FURTHER INFORMATION 
CONTACT). The Service has reviewed the best available scientific and 
commercial data and found a lack of sufficient evidence to support more 
than one species.
    (5) Comment: This taxonomic dispute is not simply an academic 
exercise of whether to lump or split taxa, because the decision has 
enormous implications for the conservation of imperiled species. 
Multiple experts recommended that the roundtail chub complex, however 
it is constituted, be managed as separate populations or managed as a 
complex.
    Response: The Service recognizes that multiple experts agree that 
conservation actions must be directed at the population level and must 
include consideration of the complex as a whole (Dowling et al. 2008, 
pp. 30-31; Dowling and DeMarais 1993, p. 445; Gerber et al. 2001, p. 
2037; Schwemm 2006, pp. 32-33). However, the Service must adhere to the 
Act and its implementing regulations, which define a ``species'' as any 
species or subspecies of fish, wildlife, or plant, and any distinct 
population segment of any vertebrate species which interbreeds when 
mature (16 U.S.C. 1532(16) and 50 CFR 424.02). The best available 
scientific and commercial data as discussed above in the Taxonomy 
section, support recognition of only one species, Gila robusta. The 
Service's withdrawal of our proposed rule to list the headwater and 
roundtail chub based on new taxonomic classification does not diminish 
the conservation efforts of our partners to conserve this species and 
habitat, nor does our decision affect the State's ability to conserve 
this species under its own authority. The Arizona Game and Fish 
Department recognizes the importance of conserving the currently 
recognized roundtail chub population rangewide (including the formerly 
known headwater chub and Gila chub) and is committed to the 
conservation agreements and practices that have been in place since 
2006 (AGFD 2017, entire; AGFD 2006, entire).
    (6) Comment: Multiple commenters raised concerns with Copus et al. 
(2016) methods and conclusions, particularly small sample size, lack of 
key analytical and laboratory steps, the study's DNA sequence data 
filtering and analyses that failed to follow best practices for 
phylogenetic analysis, and specimen shrinkage associated with duration 
of preservation impacting morphological diagnostics.
    Response: The Service did not rely solely on Copus et al. 2016. We 
considered the best available commercial and scientific data; you may 
obtain a complete list of references cited on the Internet at http://www.regulations.gov and upon request from the Arizona Ecological 
Services Office (see FOR FURTHER INFORMATION CONTACT). In regards to 
the mitochondrial DNA and phylogenetic analysis, Copus et al.'s 
findings are consistent with Sch[ouml]nhuth et al.'s (2014) and 
Sch[ouml]nhuth et al.'s (2012) mitochondrial DNA and phylogenetic 
analysis. In addition, multiple genetic analysis studies have been 
conducted that indicate population-level differences, but do not 
identify genetic markers that have the ability to distinguish among G. 
robusta, G. nigra, and G. intermedia (DeMarais 1992, pp. 2748-2749; 
Schwemm 2006, p. 29; Dowling et al. 2008, p. 2, and 2015, p. 13).
    In regards to morphological diagnostic errors due to using 
preserved specimens, Copus et al. (2016) did use preserved specimens. 
However, they also analyzed fresh material and concluded that no single 
diagnostic character can be used for species identification, and with 
considerable overlap among species in every morphological character, no 
suite of characters can distinguish species unambiguously (Copus et al. 
2016, p. 13). Brandenburg et al. (2015, entire) also reported overlap 
in the meristic and morphometric characteristics, records of many 
individual fish with intermediate physical characteristics, and even 
those characters that do not overlap are separated by very small 
margins making species-level identification of individual fish 
problematic, even when the geographic origin of the species is known.

Public Comments

    (7) Comment: Multiple commenters requested various listing 
alternatives under the Act including: List G. robusta as threatened and 
encompass all populations of the chub complex within the Gila basin 
requiring a revision of the recovery plan, list G. robusta and G. nigra 
as threatened and retain the current endangered species status of G. 
intermedia, list G. robusta as threatened and retain the current 
endangered species status of G. intermedia, or other combinations.
    Response: The Service must adhere to the Act and its implementing 
regulations, which define a ``species'' as any species or subspecies of 
fish, wildlife, or plant, and any distinct population segment of any 
vertebrate species which interbreeds when mature (16 U.S.C. 1532(16) 
and 50 CFR 424.02), and based on our review, the best available 
scientific and commercial data

[[Page 16986]]

support recognition of only one species, Gila robusta. As the headwater 
chub and roundtail chub DPS no longer meet the definition of a 
``species'' under the Act, we must withdraw our proposed rule to list 
them as threatened species.
    (8) Comment: Multiple commenters stated that there is a great 
amount of morphological overlap among counts and measures for these 
chub taxa and that this has long been recognized. If a taxonomic key is 
not 100 percent correct, that does not necessarily mean that these are 
not taxa that are biologically distinct at the specific level. A test 
of the key would require the a priori identification of each individual 
to species. Rather than dismiss the species' taxonomic status, 
biologists should be working to make a better key that can be used in 
the field for the effective identification and management of the 
species.
    Response: We recognize that diagnostic keys do not produce correct 
results all the time, whether due to human error or morphological 
similarities among purported species. However, Copus et al. (2016, p. 
13) concluded that, based on genetic analysis, no single diagnostic 
character can be used for species identification, and with considerable 
overlap among species in every morphological character, no suite of 
characters can distinguish species unambiguously. Brandenburg et al. 
(2015, entire) also reported overlap in the meristic and morphometric 
characteristics, and there are many individual fish whose morphology 
resides on an intermediate spectrum, and even those characters that do 
not overlap are separated by very small margins, making species-level 
identification of individual fish problematic, even if the geographic 
origin of the species is known. In regards to a priori identification 
of fish, assignment to species has been based on the stream in which 
the fish occurs (Minckley and DeMarais 2000, p. 252), so the 
identification of the fish that occurs in each stream is assumed to be 
known. Consequently, there exists the ability to compare findings from 
the diagnostic key to the fish within a particular stream. An updated 
key may be prudent; however, the Service must use the best available 
scientific and commercial data available, and we have concluded from 
our review that the data currently support only one species, Gila 
robusta. Further, given the overlap in diagnostic characteristics, the 
development of a valid key seems unlikely.
    (9) Comment: Multiple commenters stated that it has long been 
hypothesized that G. nigra formed as the result of hybridization 
between the other two taxa, so we would expect the greatest 
morphological overlap from that species with the other two taxa. The 
question then becomes, is G. nigra continuing to differentiate from 
ancestral G. robusta? When in sympatry, G. nigra and G. robusta are 
becoming increasingly reproductively isolated from one another (Desert 
Fishes Council meeting, Dowling et al. 2016).
    Response: We recognize that multiple studies have indicated that 
hybridization has occurred among G. intermedia and G. robusta resulting 
in G. nigra and that continuing evolution may occur (Schwemm 2006; 
Dowling et al. 2008, entire). However, there has also been information 
presented showing no evidence of the hybrid origin of G. nigra, and 
that G. intermedia and G. nigra evolved separately in non-overlapping 
areas (parapatry) (Chafin 2016, entire). In addition, past research 
(Dowling et al. 2008, 2015; Schwemm 2006) indicate that there is more 
variation among populations and unique genetics within specific 
populations (streams).
    (10) Comment: If only G. robusta and G. intermedia are evaluated, 
there is no question that they would be considered distinct 
morphological species.
    Response: Carter et al. (2016, in press) found that the physical 
characteristics did not reliably differentiate among G. robusta, G. 
intermedia, and G. nigra. In addition, Brandenburg et al. (2015, pp. 8-
9) found physically similarity of the three species, as numerous 
individuals exhibited intermediate characters along the species 
gradient. The discriminant function analysis (a statistical analysis 
tool to determine which variables discriminate between two or more 
naturally occurring groups) classified only 16 percent (n = 42) of G. 
intermedia (the fewest) while the majority of the samples were 
classified as G. robusta (53.2 percent, n = 140), which indicates that 
the ability to classify these fish correctly to G. intermedia or G. 
robusta based on physical characteristics was low. Due to the complex 
genetic makeup and observable characteristics or traits (i.e., physical 
appearance, behavior, or physiology) of these species, there are some 
stream locations where we do not know where the geographic overlap of 
headwater, roundtail, and, in some cases Gila chub, begins and ends, 
because of the plasticity of observable characteristics or traits of 
these fish within individual streams. Our review of the data does 
indicate that there are differences in observable characteristics or 
traits between the fish in different streams, but the Societies' 
review, as well as the Service review, of the best available scientific 
and commercial data did not result in a species-level differentiation 
between G. robusta and G. intermedia, or among G. robusta, G. 
intermedia, and G. nigra.
    (11) Comment: One commenter recommend that we proceed with an 
amended recovery plan to list the status of this species as threatened 
under the Act. The listing of this species is necessary even if all 
populations of G. intermedia and G. nigra are subsumed into G. robusta.
    Response: An assessment of the entire range of the new taxonomic 
group of roundtail chub is planned. We are initiating a status review 
of the new taxonomic entity in 2 to 4 years. Following that review, we 
will take action as appropriate.

Determinations

    An entity may only be listed under the Act if that entity meets the 
Act's definition of a species. The recent report by the Societies 
indicates that neither the headwater chub nor the roundtail chub can be 
considered species, as defined by the Act. Under section 3 of the Act 
(16 U.S.C. 1532(16)) and associated implementing regulations at 50 CFR 
424.02, a ``species'' is defined to include any species or subspecies 
of fish, wildlife, or plant, and any distinct population segment of any 
vertebrate species which interbreeds when mature. The Act's 
implementing regulations at 50 CFR 424.11(a) and the Service Director's 
November 25, 1992, ``Taxonomy and the Endangered Species Act'' 
Memorandum (Memo) provide additional guidance on how to consider 
taxonomic information when assessing a species for listing under the 
Act. The regulations at 50 CFR 424.11(a) state, ``In determining 
whether a particular taxon or population is a species for the purposes 
of the Act, the Secretary [of the Interior] shall rely on standard 
taxonomic distinction and the biological expertise of the Department 
[of the Interior] and the scientific community concerning the relevant 
taxonomic group.'' The Director's Memo specifies that the Service is 
``required to exercise a degree of scientific judgment regarding the 
acceptance of taxonomic interpretations, particularly when more than 
one possible interpretation is available. The Memo further states, 
``When informed taxonomic opinion is not unanimous, we evaluate 
available published and unpublished information and come to our own 
adequately documented conclusion regarding the validity of taxa.''
    The Act requires that we finalize, modify, or withdraw the proposed 
rule

[[Page 16987]]

within 12 months. The Act provides for one 6-month extension for 
scientific uncertainty, which we have used. As such, we are required to 
make a decision regarding the entities' eligibility for listing at this 
time. In addition, section 4(b)(1)(A) of the Act requires the Service 
to make listing or delisting decisions based on the best scientific and 
commercial data available. Further, our Policy on Information Standards 
under the Act (July 1, 1994; 59 FR 34271), the Information Quality Act 
(section 515 of the Treasury and General Government Appropriations Act 
for Fiscal Year 2001 (Pub. L. 106-554; H.R. 5658)), and our associated 
Information Quality Guidelines (https://www.fws.gov/informationquality), provide criteria, guidance, and establish 
procedures to ensure that our decisions are based on the best 
scientific data available. They require us, to the extent consistent 
with the Act and with the use of the best scientific data available, to 
use primary and original sources of information as the basis for 
recommendations. Primary or original information sources are those that 
are closest to the subject being studied, as opposed to those that 
cite, comment on, or build upon primary sources. The Act and our 
regulations do not require us to use only peer-reviewed literature, but 
instead they require us to use the ``best scientific and commercial 
data available.'' We use information from many different sources, 
including articles in peer-reviewed journals, scientific status surveys 
and studies completed by qualified individuals, Master's thesis 
research that has been reviewed but not published in a journal, other 
unpublished governmental and nongovernmental reports, reports prepared 
by industry, personal communication about management or other relevant 
topics, conservation plans developed by States and counties, biological 
assessments, other unpublished materials, experts' opinions or personal 
knowledge, and other sources.
    We conducted a similar internal review of the information presented 
by and available to the Societies in their review. Our review primarily 
focused on Marsh et al. (2016), Carter et al. (2016, in press), Copus 
et al. (2016), Minckley and DeMarais (2000), and Chafin et al. (2015), 
as well as other literature as discussed above in the Taxonomy section. 
In their most recent publication of Common and Scientific Names of 
Fishes (Page et al. 2013, p. 8), the Societies state the following 
regarding the common process of their naming committee: ``In accepting 
species as valid from various works, we made little or no judgment on 
authors' species concepts. Taxa of uncertain status were dealt with on 
a case-by-case basis.'' Based on the Societies' expertise and their 
internal guidance (stated above) on making such decisions, we conclude 
that the preponderance of evidence before them was clear and decisive 
enough to make a taxonomic change.
    After reviewing the best available scientific and commercial 
information (as described above in the Taxonomy section and summarized 
below) and applying statutory and regulatory guidance, we determined 
that the Societies' report considered the best commercial and 
scientific data available. We agree with the conclusion that available 
data support recognition of only one species, Gila robusta. Our 
determination is based on various factors, including the method of 
original assignment to species, hybridization events, conflicting 
identification of species based on morphology versus genetics, 
evolutionary history, morphological identification limitations, and 
lack of genetic markers to identify species. We lack confidence in the 
initial species assignments to G. robusta, G. nigra, and G. intermedia 
due to the scientific methods used (fish were assigned to a species 
based on the stream in which they occurred, the erroneous assumption 
that these fish did not overlap geographically, and the absence of 
genetic or morphological diagnostic information). Minckley and DeMarais 
(2000, entire) based their diagnostic key on the assumption that none 
of these species occurs in the same locality; however, they acknowledge 
hybridization among G. robusta and G. intermedia. Further, other 
studies have found that fish designated as G. robusta, G. nigra, and G. 
intermedia overlap geographically or occur adjacent to one another 
(Dowling and Marsh 2009, p. 1; Marsh et al. 2016, p. 57; Brandenburg et 
al. 2015, p. 18). In addition, some populations appeared to conflict 
genetically with the species-level assignment based on morphology 
(Dowling et al. 2015, pp. 14-15). Multiple scientists (as described 
above) found Minckley and DeMarais's (2000, entire) key for 
identification of G. robusta, G. nigra, and G. intermedia to not 
reliably differentiate among these three fish. In Fossil Creek, Marsh 
et al. (2016, entire) concluded there are two morphologically similar, 
but genetically distinguishable chub. However, there are several 
genetic analysis studies indicating population-level differences among 
these fish, but the studies were not able to identify genetic markers 
distinguishing between the three fish. Finally, Sch[ouml]nhuth et al. 
(2014, p. 223) found that G. robusta, G. nigra, and G. intermedia were 
in one grouping that included a common ancestor and all the descendants 
(living and extinct) of that ancestor (clade), and hypothesized this 
could reflect incomplete lineage sorting or hybridization, but this was 
not studied.
    For the purposes of our determination, we accept the ``single 
species'' finding by the Societies described above and, consequently, 
withdraw the proposed rule to list the headwater chub (Gila nigra) and 
a DPS of the roundtail chub (Gila robusta) from the lower Colorado 
River basin as threatened species under the Act. This withdrawal is 
based on a thorough review of the best scientific and commercial data 
available, which indicate that the headwater chub and the DPS of the 
roundtail chub are not discrete taxonomic entities and do not meet the 
definition of species under the Act. These fish are now recognized as a 
single taxonomic species--the roundtail chub (Gila robusta). Because 
the entities previously proposed for listing are no longer recognized 
as species, as defined by the Act, we have determined that they are not 
listable entities, and we are withdrawing our proposed rule to list.

Future Actions

    Following the publication of this withdrawal, we intend to 
reevaluate the status of the Gila chub (currently listed as endangered) 
in the near future and initiate a range-wide species status assessment 
(SSA) of the newly-recognized roundtail chub (Gila robusta).

References Cited

    A complete list of references cited in this document is available 
on the Internet at http://www.regulations.gov and upon request from the 
Arizona Ecological Services Office (see FOR FURTHER INFORMATION 
CONTACT).

Authors

    The primary authors of this document are the staff members of the 
Arizona Ecological Services Office.

Authority

    The authority for this action is section 4(b)(6)(B)(ii) of the 
Endangered Species Act of 1973, as amended (16 U.S.C. 1531 et seq.)


[[Page 16988]]


    Dated: March 21, 2017.
James W. Kurth,
Acting Director, U.S. Fish and Wildlife Service.
[FR Doc. 2017-06995 Filed 4-6-17; 8:45 am]
 BILLING CODE 4333-15-P