Dietary niche separation of three Late Pleistocene bear species from Vancouver Island, on the Pacific Northwest Coast of North America

Abstract

Competition between taxa related to climate changes has been proposed as a possible factor in Pleistocene megafaunal extinctions, and here we present isotope evidence of the diets of three co‐existing bear species [black bear (Ursus americanus), brown bear (Ursus arctos), and the now extinct short‐faced bear (Arctodus simus)] from a locale in western North America dating to the Late (Terminal) Pleistocene (~14.5–11.7 ka). The three bear species were found at several sites on Vancouver Island, on the western coast of Canada. To examine the chronological overlap and niche partitioning between these species of bear, we used direct radiocarbon dating, stable isotope analysis and ZooMS proteomic identification methods. Here we present new radiocarbon evidence from Terminal Pleistocene U. americanus, U. arctos and A. simus from several sites on the island, along with both bulk collagen and compound‐specific isotope data for these species. Radiocarbon dates confirm the chronological overlap of Arctodus and both Ursus species in the montane regions of the island at the end of the Pleistocene. Stable isotope data reveal niche differentiation between these species, with U. americanus occupying a distinctly lower trophic position than the other two taxa.

Full text

JOURNAL OF QUATERNARY SCIENCE (2022) 1–13 ISSN 0267-8179. DOI: 10.1002/jqs.3451
Dietary niche separation of three Late Pleistocene bear species from
Vancouver Island, on the Pacific Northwest Coast of North America
CARA KUBIAK,
1
* VAUGHAN GRIMES,
2,3
GEERT VAN BIESEN,
4
GRANT KEDDIE,
5
MIKE BUCKLEY,
6
REBA MACDONALD
1
and M. P. RICHARDS
1
1
Department of Archaeology, Simon Fraser University, Burnaby, BC, Canada
2
Department of Archaeology, Memorial University, St. John's, Newfoundland, Canada
3
Department of Earth Sciences, Memorial University, St. John's, Newfoundland, Canada
4
Core Research Equipment and Instrument Training (CREAIT) Network, Stable Isotope Laboratory, Memorial University, St. John's,
Newfoundland, Canada
5
Indigenous Collection and Repatriation Department, Royal British Columbia Museum, Royal British Columbia Museum, Victoria,
Canada
6
Department of Earth and Environmental Sciences, School of Natural Sciences, University of Manchester, Manchester, UK
Received 2 February 2022; Revised 20 May 2022; Accepted 31 May 2022
ABSTRACT: Competition between taxa related to climate changes has been proposed as a possible factor in
Pleistocene megafaunal extinctions, and here we present isotope evidence of the diets of three co‐existing bear
species [black bear (Ursus americanus), brown bear (Ursus arctos), and the now extinct short‐faced bear (Arctodus
simus)] from a locale in western North America dating to the Late (Terminal) Pleistocene (~14.5–11.7 ka). The three
bear species were found at several sites on Vancouver Island, on the western coast of Canada. To examine the
chronological overlap and niche partitioning between these species of bear, we used direct radiocarbon dating,
stable isotope analysis and ZooMS proteomic identification methods. Here we present new radiocarbon evidence
from Terminal Pleistocene U. americanus,U. arctos and A. simus from several sites on the island, along with both
bulk collagen and compound‐specific isotope data for these species. Radiocarbon dates confirm the chronological
overlap of Arctodus and both Ursus species in the montane regions of the island at the end of the Pleistocene. Stable
isotope data reveal niche differentiation between these species, with U. americanus occupying a distinctly lower
trophic position than the other two taxa. ©2022 John Wiley & Sons, Ltd.
KEYWORDS: Pleistocene; bears; Arctodus; amino acid; isotopes
Introduction
At the terminus of the Pleistocene epoch, the final recession of
the Cordilleran Ice Sheet following the Wisconsin glaciation
opened previously ice‐locked land masses to colonization by a
variety of species. The maximum geographical extent of ice
sheet coverage and the exact timing of ice sheet recession from
the northwest coast, and Vancouver Island in particular,
remains an area of study; however, a growing body of
environmental proxy data is providing insight into the specifics
of these changes at local and regional scales (e.g.
Howes, 1983; James et al., 2000; Ward and Thomson, 2004;
Al‐Suwaidi et al., 2006; Cosma et al., 2008; McLaren
et al., 2014; Eamer et al., 2017). Pollen records from several
sites on Vancouver Island indicate that glacial recession and
an initially treeless landscape was followed by the coloniza-
tion of the region by shade‐intolerant lodgepole pine (Pinus
contorta) and later by the establishment of increasingly closed
and shade‐tolerant forests with the appearance of species such
as spruce (Picea), mountain hemlock (Tsuga mertensiana) and
red alder (Alnus rubra) (Hebda, 1983;Al‐Suwaidi et al., 2006;
Lacourse, 2005).
While both Ursus americanus and Ursus arctos were present
elsewhere on the Northwest Coast prior to the Last Glacial
Maximum (LGM) (Heaton and Grady, 2003), these species
have yet to be documented on Vancouver Island prior to the
LGM. Arctodus, however, was known to be present prior to the
LGM on Vancouver Island (Steffen and Harington, 2010). The
recolonization of the Northwest Coast by Ursus following
deglaciation has been previously documented by early post‐
glacial dates from Haida Gwaii (Ramsey et el., 2004) and
Vancouver Island (Steffen and Fulton, 2018). Examples of post‐
glacial Arctodus in the region are currently limited to two
specimens from Vancouver Island (Steffen and Fulton, 2018).
Ursus arctos and Arctodus have been noted to co‐occur in
the paleontological record elsewhere in North America (e.g.
Matheus, 1995; Barnes et al., 2002; Davison et al., 2011).
Additionally, U. americanus and U. arctos were found to co‐
occur to the north on Haida Gwaii during the Younger Dryas
(12 900–11,500 cal a BP) (Fedje et al., 2011) and co‐occur
regionally in several parts of North America to this day,
including part of the northwest coast. Vancouver Island
presents a case of temporal overlap between not only two,
but all three of these bear species in the period shortly
preceding the extinction of the short‐faced bear.
Stable isotope analysis of extinct megafauna
The dietary requirements of now‐extinct Arctodus simus have
been debated within the paleontological literature, with
claims made for everything from an herbivorous to a hyper‐
carnivorous diet based on morphological characteristics,
microwear and paleopathological analyses, and stable isotope
analyses of skeletal remains (Kurtén and Anderson, 1967;
Emslie and Czaplewski, 1985; Bocherens et al., 1995;
Fox‐Dobbs et al., 2008; Figueirido et al., 2010; Donohue
©2022 John Wiley & Sons, Ltd.
*
Correspondence: C. Kubiak, as above. Email: ckubiak@sfu.ca

et al., 2013; Figueirido et al., 2017). Existing stable isotope data
for Arctodus are limited primarily to that of more northernly
specimens from the Yukon and Alaska, with only one specimen
from pre‐LGM Vancouver Island (see Bocherens et al., 1995;
Matheus, 1995;Barneset al., 2002;Fox‐Dobbs et al., 2008;
Steffen and Harington, 2010; Schwartz‐Narbonne et al., 2015),
although recently published data from a single Late Pleistocene
specimen found on the California Channel Islands provide
some insight into the diets of more southernly populations
(Mychajliw et al., 2020).
It has been suggested by Steffen and Fulton (2018) that
U. arctos and Arctodus may have exhibited territorial niche
partitioning on the island, and that competition should be
considered as a factor in extinction. They proposed that future
studies of the trophic interactions of these species may reveal
the influence of competition on resource availability for
Arctodus at the end of the Pleistocene. Additionally, the
influence of competition on Arctodus diet has been suggested
in recent explorations of Arctodus diet that consider regional
variability and the presence of other large carnivore species
(Figueirido et al., 2017).
To better understand the diets of these three co‐existing bear
species in this region, we used both bulk collagen stable
isotope analysis and compound‐specific stable isotope analy-
sis (CSIA) on a subset of specimens from all three species, as
the latter technique offers a high‐resolution dietary analysis. A
variety of contributing factors can influence the bulk collagen
carbon and nitrogen isotope composition of a consumer's
tissues, including diet, environmental and climatic conditions,
and digestive physiology (DeNiro and Epstein, 1978,1981;
Austin and Vitousek, 1998; Craine et al., 2009; McMahon and
McCarthy, 2016). By analyzing the carbon and nitrogen
isotope ratios of individual amino acids within bone collagen,
it is possible to differentiate between some of the factors that
contribute to the resulting isotopic composition of consumer
tissue. This is a relatively novel approach to isotope analysis of
paleontological materials; current challenges in the applica-
tion of CSIA are summarized by Whiteman et al. (2019).
While some amino acids in bone collagen closely reflect the
isotopic composition of an animal's diet, others experience
higher magnitudes of trophic fractionation due to the isotopic
effects of amino acid synthesis and assimilation within the
consumer's body. For the investigation of carbon isotopes,
CSIA allows for the distinction between the isotopic composi-
tions of essential and non‐essential amino acids. Essential
amino acids are those that cannot be synthesized by the
consumer; therefore, their carbon skeletons must be acquired
intact from dietary sources, and the carbon isotope composi-
tion of these amino acids in consumer tissue closely reflects
that of their dietary source. Non‐essential amino acids, which
can be either directly acquired from dietary sources or
synthesized within the body, reflect the carbon isotope
composition of dietary sources to different extents (Whiteman
et al., 2019; however, see Newsome et al., 2011 for a
consideration of the contribution of the gut microbiome to
essential amino acid isotope composition).
Because the nitrogen‐containing amine group of each amino
acid is affected by different processes than its carbon skeleton,
CSIA of nitrogen makes use of the distinction between ‘source’
and ‘trophic’amino acids. Source amino acids are known to
closely reflect the nitrogen isotope composition of a dietary
source, while trophic amino acids undergo significant fractio-
nation through the processes of deamination and transamina-
tion. The distinction between these two groups has been well
established empirically across many food webs (McClelland
and Montoya 2002, Chikaraishi et al., 2009; Naito et al.,
2010,2013) and, more recently, metabolic explanations have
been proposed to explain the fidelity of these groupings
(Chikaraishi et al., 2014; McMahon and McCarthy, 2016;
O'Connel, 2017).
In this study, we aim to contribute to the small but growing
body of individual amino acid stable isotope data for extinct
taxa and provide compound‐specific collagen stable isotope
data for extant North American large mammal species:
U. americanus and U. arctos. In combination with direct
radiocarbon dating and ZooMS proteomic identification, we
use these data to explore the ecological context of these three
bear species on Vancouver Island at the end of the Late
Pleistocene.
Materials
Ten bone samples were analyzed in this study, consisting of
three short‐faced bears (Arctodus simus), four brown bears
(Ursus arctos) and three black bears (Ursus americanus) from
five locations on Vancouver Island, in British Columbia,
Canada. All Terminal Pleistocene material examined in this
paper originates from montane cave sites. Limestone solution
caves can provide shelter for hibernating bears and sometimes
act as traps for large mammals, both situations resulting in the
deposition of bones on cave floors. The specimens examined
in this study are housed at the Royal British Columbia Museum
(RBCM) in Victoria, BC, Canada. The approximate geographi-
cal origins of the samples are given in Figure 1.
Specimens S‐SFU 269 and 270 come from Windy Link Pot
Cave. The cave, with an entrance elevation of ~900 m above
sea level, is part of a large underground system spanning
nearly 10 km. The specimens used in this study were collected
from the cave floor surface at the base of the 78‐m vertical
drop cave entrance. They were identified as U. americanus by
Nagorsen et al. (1995), who report a detailed morphological
description of their identification. S‐SFU 269 was sampled
from a skull, and S‐SFU 270 from a mandible, each
representing a different individual.
Specimens S‐SFU 277–281 come from Pellucidar Cave.
Pellucidar Cave is a large limestone cave system with an
underground stream elevated at ~480 m above sea level
(Steffen and McLaren, 2008); it is described in detail by
Nagorsen and Keddie (2000). The specimens in this study were
collected from the cave floor surface, ~60 m in from the cave
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
Figure 1. Pleistocene bear recovery locations in this study. [Color
figure can be viewed at wileyonlinelibrary.com]
2 JOURNAL OF QUATERNARY SCIENCE

opening. The morphology and radiocarbon dates of these
specimens are described in detail by Steffen and Fulton (2018).
The samples include an immature U. arctos humerus (S‐SFU
277), an A. simus palatine (S‐SFU 278), an A. simus humerus
(S‐SFU 279), an U. arctos femur (S‐SFU 280) and an U. arctos
humerus (S‐SFU 281). Steffen and Fulton (2018) noted that the
A. simus palatine and humerus (S‐SFU 278 and 279) are not
exclusive elements and could potentially represent the same
individual.
S‐SFU 272 was collected from Grey Mountain Cave in
February 1990. The specimen is an U. arctos juvenile femur.
S‐SFU 276 comes from North Vancouver Island Cave (near
Pellucidar cave) and was first reported by Steffen and Fulton
(2018). The specimen sampled is an U. americanus cranium.
Methods
ZooMS
Five samples (S‐SFU‐269, 278, 279, 280, 281) were processed
for ZooMS collagen peptide mass fingerprinting (Buckley
et al., 2009). Approximately 50‐mg bone chunk samples were
prepared and analyzed at the Manchester Institute of
Biotechnology following the procedure described by van der
Sluis et al. (2014). In brief, this involved decalcification of
~25–50 mg bone powder with 0.6 Mhydrochloric acid for 18
h, centrifugation at 12 400 r.p.m. for 1 min and the acid‐
soluble fraction ultrafiltered on a 10‐kDa molecular weight
cut‐off membrane (Vivaspin). Following two washes in 50 mM
ammonium bicarbonate (ABC), 100 µl of retentate was
removed and digested with 0.4 µg sequencing‐grade trypsin
(Promega) for 18 h at 37 °C. Peptide solutions were then
acidified to 0.1% trifluoroacetic acid (TFA; Sigma), and
purified by C18 pipette tips (OMIX) eluting with 50%
acetonitrile (ACN)/0.1% TFA, dried to completion and
resuspended with 0.1% TFA. One‐tenth was then spotted with
an equal volume of 10 mg ml
–1
alpha‐cyano hydroxycinnamic
acid in 50% ACN/0.1% TFA onto a stainless steel matrix
assisted laser desorption ionization (MALDI) target plate and
allowed to air dry. Samples were then analyzed using a Bruker
Autoflex II MALDI time‐of‐flight mass spectrometer with
m/zrange 700–3700 and resultant spectra compared with
reference materials (see Supporting Information Figure S4).
Radiocarbon dating
Approximately 500 mg of bone powder was drilled from
samples S‐SFU‐269, 270, 272 and 276, and sent to the A.E.
Lalonde AMS laboratory at the University of Ottawa for
radiocarbon age determination. Sample preparation included
collagen extraction following a modified version of the Longin
(1971) method utilizing ultrafiltration of collagen through
30‐kDa MWCO filters, the specific laboratory protocol for
which is described as media code BU in Crann et al. (2017).
14
C analysis was carried out by accelerator mass spectrometry
(AMS) on prepared graphite. Calibration of radiocarbon
determination to calendar years was performed using OxCal
4.4 software (Bronk Ramsey 2009) according to the IntCal20
calibration curve (Reimer et al., 2020).
Bulk collagen δ
13
C and δ
15
N stable isotope
analysis
Samples S‐SFU 277–281 were prepared for carbon and
nitrogen bulk stable isotope analysis (BSIA) following the
protocol described in Müldner and Richards (2005) at the
Archaeology Isotope Laboratory, Department of Archaeology,
Simon Fraser University. To summarize, bone chunks (<1g)
were demineralized in 0.5 MHCl, rinsed with distilled water,
then gelatinized at 75 °C in pH 3 HCl and filtered through 30‐
kDa MWCO ultrafilters, with the >30‐kDa fraction lyophilized
for analysis. Duplicate samples were encapsulated in tin
capsules and sent to Iso Analytical (Crewe UK), where δ
13
C
and δ
15
N values were measured in duplicate by elemental
analysis ‐isotope ratio mass spectrometry (EA‐IRMS).
Collagen quality standards required a collagen yield
between 0.5 and 22%, C/N ratio of 2.9–3.6, %C of
15.3–47% and %N of 5.5–17.3% (DeNiro, 1985;Am-
brose, 1990; van Klinken, 1999). The δ
13
C and δ
15
N values
presented here are the average of duplicates, reported as the ‰
difference from the international standards VPBD for carbon
(
13
C/
12
C) and AIR for nitrogen (
15
N/
14
N) (Coplen, 2011).
δ
13
C and δ
15
N values were calibrated to VPDB and AIR
using the standard IA‐RO68 (soy protein, δ
13
C=−25.22 ‰,
δ
15
N=0.99‰). Check standards included IA‐R038 (L‐alanine,
δ
13
C=−24.99‰,δ
15
N=−0.65‰), IA‐R069 (tuna protein,
δ
13
C=−18.88‰,δ
15
N=11.60‰), and a mixture of IAEA‐C7
(oxalic acid, δ
13
C=−14.48‰) and IA‐R046 (ammonium
sulfate, δ
15
N=22.04‰). Average observed values of these
check standards during analysis were δ
13
C=−25.05‰,δ
15
N
=−0.47‰(IA‐R038, n=2); δ
13
C=−18.85‰,δ
15
N=11.72‰
(IA‐R069, n=2); and δ
13
C=−14.57‰,δ
15
N=21.97‰
(IAEA‐C7 and IA‐R046, n=2).
Samples 269, 270, 272 and 276 were prepared and
analyzed at the University of Ottawa G.G. Hatch stable
isotope laboratory, where a subsample of the lyophilized
collagen prepared for radiocarbon dating was analyzed in
duplicate by EA‐IRMS for δ
13
C and δ
15
N values. δ
13
C and
δ
15
N values were calibrated to VPDB and AIR using internal
standards calibrated to the international standards IAEA‐N1
(+0.4‰), IAEA‐N2 (+20.3‰), USGS‐40 (−4.52‰) and USGS‐
41 (47.55‰) for nitrogen and IAEA‐CH‐6(−10.4‰), NBS‐22
(−29.91‰), USGS‐40 (−26.39‰) and USGS‐41 (36.55‰) for
carbon. Analytical precision was determined based on the
results of the internal check standard C‐55 glutamic acid
(expected value: δ
15
N=−3.98, δ
13
C=−28.53; measured
valued =δ
15
N=−4.0, δ
13
C=−28.4).
δ
13
C and δ
15
N analysis of individual amino acids
Extracted collagen was prepared and analyzed for CSIA‐AA at
the Memorial Applied Archaeological Science (MAAS) labora-
tory and Biogeochemistry of Boreal Ecosystems Laboratory at
Memorial University (S‐SFU‐269, 276, 278 and 279) and at the
Archaeology Isotope Laboratory, Department of Archaeology,
Simon Fraser University (S‐SFU 272, 277 and 281), with the
same preparation protocol followed at both labs. For each
sample, 1 mg of lyophilized collagen was hydrolyzed in 6 M
HCl at 110 °C for 20 h, blown down at 60 °C under a gentle
stream of pure nitrogen, and redissolved in 0.1 MHCl. An
internal standard (nor‐leucine) was added to each sample prior
to derivatization. Due to the amide to carboxylic group
conversion during acid hydrolysis, glutamine (Gln) and
asparagine (Asn) are respectively converted to glutamic acid
(Glu) and aspartic acid (Asp), and their δ
13
C and δ
15
N values
reflect the combined contributions; these are hereafter notated
as Glx and Asx (Fountoulakis and Lahm, 1998).
A standard, AAmix, was prepared and derivatized along
with the samples containing each of the analyzed amino acids:
alanine (Ala), glycine (Gly –USGS65), valine (Val –USGS74),
leucine (Leu), threonine (Thr), serine (Ser), proline (Pro),
aspartic acid (Asp), glutamic acid (Glu), hydroxyproline
(Hyp), phenylalanine (Phe) and lysine (Lys). EA‐IRMS was
used to determine the δ
13
C and δ
15
N values of the individual
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
DIET OF LATE PLEISTOCENE BEARS 3

amino acids in the standard mixture prior to their combination,
except for Gly and Val, which had certified values. A quality
control sample, QCmix, contained only Gly, Val, Pro and Glu.
The free amino acids in the standards and samples were
derivatized into N‐acetyl isopropyl (NAIP) esters following the
procedure described in Corr et al. (2007).
Gas chromatography combustion (GC‐C)‐IRMS analysis for
δ
13
C and δ
15
N values of derivatized amino acids was
performed at the CREAIT –Stable Isotope Laboratory at
Memorial University using an Agilent 6890N gas chromato-
graph coupled via a GC Combustion III interface to a Delta
V Plus isotope ratio mass spectrometer. For δ
13
C analysis, a
GC‐PAL autosampler (CTC Analytics) injected samples at 250
°C onto a VF‐23ms GC column (60 m ×0.32 mm ×0.15 µm;
Agilent Technologies) following an oven temperature program
of 70 °C for 0.5 min; to 120 °Cat15°C min
–1
; to 180 °Cat2°C
min
–1
; to 255 °Cat5°C min
–1
; hold for 16 min. Separated
gases then entered a Cu, Ni and Pt wire oxidation reactor held
at 940 °C, followed by a reduction reactor at 640 °C at a flow
rate of 1.5 ml min
–1
. Samples were analyzed in triplicate in a
sequence with an amino acid standard and quality control
(QC) sample. Each sample and standard was then blown down
under nitrogen by a factor of ~5 for δ
15
N analysis. Samples
were injected onto a VF‐23 ms GC column (60 m ×0.32 mm ×
0.50 μm; Agilent Technologies) at 250 °C and subjected to an
oven temperature program of 70 °C for 0.5 min; to 130 °Cat
15 °C min
–1
; to 255 °Cat6°C min
–1
; hold 11 min. A liquid
nitrogen trap was used to capture CO
2
and prevent its entry
into the IRMS.
At the Archaeology Isotope Laboratory, Department of
Archaeology, Simon Fraser University, CSIA‐AA measure-
ments followed a similar protocol, with modifications made
for differences in instrument specifications. Amino acid
derivatives were analyzed on a Trace 1310 gas chromatograph
coupled to a GC Isolink II combustion system and a Delta
V Plus mass spectrometer (GC‐C‐IRMS; all Thermo Scientific).
Separated amino acids passed through a combustion reactor
(1000 °C) consisting of a NiO tube containing CuO, NiO and
Pt wires, which provides quantitative oxidation and reduction
of N
2
. An AS 1310 autosampler (Thermo Scientific) was used
to inject 1.0 µl for both
13
C and
15
N measurements. For δ
13
C
analysis, the same GC column specifications and oven
temperature program as descried above were utilized, with
the exception that the final temperature of 255 °C was held for
21 min. Samples were analyzed in triplicate with the AAmix
and QCmix standards interspersed throughout the sequence.
Following carbon isotope analysis, samples and standards
were concentrated under a gentle flow of nitrogen by a factor
of ~5 for nitrogen isotope analysis. Samples were injected onto
an Agilent DB‐35 GC column (60 m ×0.32 mm ×0.5 μm) at
240 °C with a 3.5‐s pre‐injection dwell time following a short
(12 s) seed oxidation prior to each analysis. Samples S‐SFU
277 and 281 were analyzed with the following GC oven
program with a flow rate of 1.4 ml min
–1
:50°C for 2 min, to
140 °C at a rate of 13 °C min
–1
, to 195 °Cat3°C min
–1
and
held for 7 min, to 245 °Cat8°C min
–1
and held for 11 min, to
280 °Cat15°C min
–1
; this final temperature was held for 8
min. Sample S‐SFU 272 was analyzed with the following GC
oven program with a flow rate of 1.3 ml min
–1
:40°C for 5 min,
to 120 °C at a rate of 15 °C min
–1
, to 180 °Cat3°C min
–1
,to
210 °C at 1.5 °C min
–1
and finally to 280 °Cat5°C min
–1
,
where this final temperature was held for 8.8 min. A liquid
nitrogen trap was used to capture CO
2
and prevent its entry in
to the IRMS.
δ
13
C values were corrected for the carbon added during
derivatization by applying a correction factor following Silfer
et al. (1991) to the measured values. For δ
15
N analysis,
GC‐C‐IRMS‐measured values of the AAmix were plotted against
EA‐IRMS‐measured values for each amino acid, and the linear
relationship was used to correct measured values in samples.
Matrix‐matched QC samples were also derivatized and
analyzed alongside the samples. At Memorial University, this
included a bovine gelatin (BGEL) standard, for which long‐
term data on δ
13
C means and standard deviations are available
(Table S1). Long‐term data on the δ
15
N values of the amino
acids in these QCs are not yet available; standard deviations of
measurements within runs are presented alongside δ
15
N data.
At Simon Fraser University, both seal collagen (SRM‐1) and
deer collagen (SRM‐2) QC samples were used; data on the
δ
13
C means and standard deviations are presented in Table S2,
and the δ
15
N means and standard deviations are presented in
Table S3. These QC standards allow for the evaluation of the
quality of derivatization and GC‐C‐IRMS analysis.
Results
A summary of the results discussed within the text is presented
in Tables 1and 2.
ZooMS Identifications
Due to variations in morphological characteristics through
time and between populations, distinguishing between differ-
ent bear taxa in fragmentary Pleistocene remains based on
skeletal morphology alone can present difficulties (e.g.
Richards et al., 1996; Steffen and Fulton, 2018). The task of
distinguishing U. arctos from U. americanus can present
particular challenges when considering Pleistocene remains.
Although modern grizzly bears are generally larger than
modern black bears, these bears may have been more similar
in size during parts of the Pleistocene (Gordon, 1986;
Wolverton and Lyman, 1998), with U. americanus adults
significantly overlapping in cranial size with immature
U. arctos (Nagorsen et al., 1995). The comparatively large
size of Pleistocene U. americanus has been proposed as an
explanation for why some relatively isolated populations of
U. americanus, such as those found on Pacific Northwest
islands today (including Vancouver Island), are unusually large
(Gordon, 1986), featuring larger teeth and broader skulls.
Vancouver Island's modern black bears feature such distinctly
large skulls that they have been sometimes referred to as
the separate subspecies U. americanus vancouveri (Hall,
1928:231). Molar size criteria have been developed to
distinguish between these species (Gordon, 1977); however,
molars are not always available for Pleistocene specimens
consisting of only one or a few skeletal elements. More
recently, geometric morphometric analyses of Pleistocene
Ursus specimens have sought to reclassify several previously
misidentified specimens (Kantelis, 2017).
Biomolecular identification methods had been previously
attempted on some of the samples included in this study, with
mixed results. One sample in the current study (SFU‐277) was
originally reported as U. americanus for radiocarbon dating
(Steffen and McLaren, 2008) but was later identified using
ancient DNA (aDNA) as U. arctos (Steffen and Fulton, 2018).
Although aDNA analysis of S‐SFU 279 was also attempted by
Steffen and Fulton (2018), no endogenous DNA was recov-
ered, and its identification as Arctodus remained solely
morphological. However, the potential for collagen to survive
longer in a burial environment than DNA (Nielsen‐
Marsh, 2002; Buckley and Collins, 2011) made this sample a
good potential candidate for ZooMS identification, as this
technique allows identification to the genus level and therefore
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
4 JOURNAL OF QUATERNARY SCIENCE

differentiation between Arctodus and Ursus spp. Preserved
collagen peptide sequences from five of our samples con-
firmed S‐SFU‐278 and 279 as Arctodus, and SFU‐269, 280 and
281 as Ursus.
Radiocarbon chronology
Radiocarbon dating was carried out on all specimens of
unknown age and those lacking direct radiocarbon dates. We
report new direct AMS dates on three Late Pleistocene Ursus
individuals (S‐SFU‐269, S‐SFU 270, S‐SFU‐272) along with a
corroborating radiocarbon date for S‐SFU‐276, which is in
close agreement with the radiocarbon date for this specimen
recently reported by Steffen and Fulton (2018) of 11 935 ±40
(UCIAMS 56479). Figure 2shows calibrated radiocarbon date
distributions for the bears analyzed in this study.
Previously, a composite sample of post‐cranial fragments
from an immature individual found on the surface of the cave
floor near S‐SFU 269 and 270 was radiocarbon dated to 9760
±140
14
CaBP (Nagorsen et al., 1995). S‐SFU 269 and 270
represent mature individuals, and this composite date cannot
be directly applied to other materials on the cave floor surface,
as their association is unknown. The new dates given here of
13 160–13,094 cal a BP for S‐SFU 269 (11 202 ±49
14
CaBP)
and 13 161–13 100 cal a BP for S‐SFU 270 (11 219 ±49
14
CaBP)
show that U. americanus was present on Vancouver Island
during the Late Pleistocene at Windy Link Cave, significantly
earlier than the initial date reported in association with these
specimens.
Bulk collagen stable isotope analysis
Carbon and nitrogen stable isotope analysis has been used to
investigate the diet of extinct and extant species of bears in
both modern and paleontological or archeological contexts
(e.g. Hildebrand et al., 1996; Barnes et al., 2002; Richards
et al., 2008; Dykstra, 2015; Hopkins et al., 2017). Hildebrand
et al. (1996) conducted experiments on bears in the modern
Pacific Northwest to determine the relationship between the
stable carbon and nitrogen composition of bear tissues and
their observed dietary inputs. The δ
13
C values of bear bone
collagen have been observed to indicate the amount of marine
versus terrestrial carbon contributed via protein ingestion
(Hildebrand et al., 1996). However, such values can also be
influenced by trophic level fractionation and the consumption
of plants with different pathways of carbon metabolism
(DeNiro and Epstein, 1978). Nitrogen stable isotope composi-
tion can help determine the bear's trophic level (DeNiro and
Epstein, 1981; Hildebrand et al., 1996) due to the significant
enrichment in
15
N that occurs with increasing trophic levels
within a food web. However, δ
15
N values can also be affected
by fluctuations in climate, such as relative aridity (Austin and
Vitousek, 1998; Craine et al., 2009), and the difference in
climatic context between specimens of different ages should
be considered when comparing these bulk tissue stable isotope
results. A summary of bulk collagen carbon and nitrogen stable
isotope data from the bears investigated in this study is
presented in Figure 3.
In the case of bears, anadromous fish such as salmon are a
probable marine source that would elevate δ
13
C values
(Hildebrand et al., 1996). In these Pleistocene Vancouver
Island samples, δ
13
C values appear to increase stepwise with
δ
15
N values, indicating a trophic level diet–tissue offset rather
than a significant contribution of anadromous fish; however,
there is some spread within both the δ
13
C and δ
15
N values of
U. arctos. While the carbon and nitrogen stable isotope ratios
of U. americanus converge around the expected values for a
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
Table 1. ZooMS identification and Radiocarbon results
S‐SFU Taxon Skeletal element Context ZooMS ID
14
C lab designation Radicaron age (
14
CaBP) Calibrated μMedian 68.3% range (cal a BP)
269 Ursus americanus Skull Windy Link Pot Cave Ursus UOC‐4529 11 202 ±49 13 125 13 124 13 160–13 094
270 Ursus americanus Mandible Windy Link Pot Cave N/A UOC‐4530 11 219 ±49 13 135 13134 13 161–13 100
272 Ursus arctos Femur Grey Mtn Cave N/A UOC‐4532 11 830 ±49 13 682 13 684 13 758–13 609
276 Ursus americanus Cranium North V. I. Cave N/A UOC‐4535 11 913 ±49 13 778 13 773 13 981–13 611
277 Ursus arctos Humerus Pellucidar Cave N/A UCIAMS‐41052*11 110 ±30 13 033 13042 13 097–12 998
278 Arctodus simus Palatine Pellucidar Cave Arctodus UCIAMS‐41049*11 615 ±30 13 484 13 480 13 571–13 453
279 Arctodus simus Humerus Pellucidar Cave Arctodus UCIAMS‐41048*11 775 ±30 13 637 13 638 13 741–13 520
280 Ursus arctos Femur Pellucidar Cave Ursus UCIAMS‐41050*12 425 ±3 14 569 14 545 14 830–14 337
281 Ursus arctos Humerus Pellucidar Cave Ursus UCIAMS‐41051** 12 440 ±35 14 599 14 583 14 845–14 366
*
14
C age reported by Steffen and Fulton (2018); all other dates listed are from this study.
DIET OF LATE PLEISTOCENE BEARS 5

terrestrial diet at a low trophic level, the U. arctos values have
a much wider range of values.
Considering that bone collagen provides a signal reflecting
the dietary intake across several years of an animal's life
(Stenhouse and Baxter, 1979), small contributions of anadro-
mous fish to the diets of U. arctos could be obscured in
the bulk collagen isotopic signal by the consumption of
13
C‐depleted resources. Smaller contributions of such resources
to the diet are especially obscured when overall lower quantities
of protein are consumed, causing collagen formation that routes a
higher percentage of carbon from non‐protein (lipid and
carbohydrate) sources (Howland et al., 2003).
Carbon compound‐specific isotope analysis (CSIA)
Due to the low availability of contemporary comparative
paleontological samples from these contexts on Vancouver
Island, it was not possible to employ the CSIA carbon isotope
fingerprinting technique used by other researchers, which uses
different species to reconstruct food‐webs and diets (e.g. Corr
et al., 2007; Larsen et al., 2009,2013; Jarman et al., 2017).
However, several trends in individual amino acid carbon
isotope composition have been observed across a variety of
ecosystems and periods, and syntheses of published CSIA
carbon data (e.g. Corr et al., 2005; Honch et al., 2012; Webb
et al., 2018) allow for broad comparisons with these trends.
The difference between the δ
13
C values of the non‐essential
amino acid Gly and the essential amino acid Phe (Δ
13
C
Gly–Phe
)
has been used as a proxy for aquatic (marine and freshwater)
versus terrestrial protein consumption, with higher Δ
13
C
Gly–Phe
values indicating higher contributions of aquatic resources and
lower Δ
13
C
Gly–Phe
values indicating relatively lower contribu-
tions of such resources (Webb et al., 2018: 6).
The Δ
13
C
Gly–Phe
values do not appear to vary greatly
between the different bear species investigated here, and all
samples fall below the values expected for high aquatic protein
consumers (Figure 4; see Webb et al., 2018: 6). Furthermore,
bulk collagen δ
13
C values and nitrogen CSIA data (discussed
below) further suggest that heavy reliance on aquatic resources
was unlikely for any of these bears.
It is possible that the U. arctos shown here consumed some
amount of anadromous fish; however, it does not appear to
have been a significant protein source. The U. americanus
shown here appear to have consumed a year‐round terrestrial
diet at a significantly lower trophic level. The Arctodus appear
to occupy a middling position between U. americanus and U.
arctos.
Despite popular depictions of U. arctos consuming salmon,
fish are not essential to the survival of U. arctos, and when
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
Table 2. Summary of discussed stable isotope results (‰) with measurement uncertainties for duplicate and triplicate injections. Long‐term
uncertainties are presented in Tables S1–S3.
S‐SFU Taxon
Bulk collagen
δ
13
C
Bulk collagen
δ
15
Nδ
15
N
Glx
δ
15
N
Phe
δ
15
N
Thr
δ
13
C
Phe
δ
13
C
Gly
269 Ursus americanus −19.5 1.7 10.4 (±0.2) 11.9 (±0.3) −11.2 (±1.4) −25.8 (±0.9) −16.7 (±0.1)
270 Ursus americanus −19.9 0.6 –––––
272 Ursus arctos −17.1 10.2 11.0 (±0.4) 7.6 (±0.6) −13.9 (±0.3) −24.6 −18.3
276 Ursus americanus −20.8 1.7 6.4 (±0.1) 7.7 (±0.2) −12.5 (±0.2) −25.8 (±0.2) −14.9 (±0.1)
277 Ursus arctos −19.6 6.2 8.1 (±0.8) –*−13.8
†
−25.9 (±0.1) −19.9 (±0.1)
278 Arctodus simus −19.0 5.3 12.0 (±0.0) 7.1 (±1.1) −25.6 (±0.8) −23.4 (±0.7) −11.3 (±0.2)
279 Arctodus simus −19.0 5.5 11.8 (±0.8) 5.6 (±1.6) −24.9 (±1.6) −25.1 (±0.3) −13.3 (±0.0)
280 Ursus arctos −19.2 7.3 –– – –−−−
281 Ursus arctos −18.3 9.4 13.0 (±0.2) 5.8 (±0.5) −15.2 (±0.8) −26.4 (±0.4) −13.2 (±0.3)
*Excluded due to co‐elution.
†
Measurement uncertainty unknown; first injection caused co‐elution (no duplicate injection).
Figure 2. Calibrated radiocarbon age distributions.
[Color figure can be viewed at wileyonlinelibrary.com]
6 JOURNAL OF QUATERNARY SCIENCE

resources such as salmon are not available the bears rely on
terrestrial vegetation, insects, freshwater fish and mammals
(Davis, 1996). Furthermore, U. arctos can, as observed in some
modern ecosystems, subsist on a completely vegetarian diet;
some populations of U. arctos have been observed to subsist
solely on vegetation across multiple years, despite the
availability of other resources such as fish (Rode et al., 2001;
Mychajliw et al.,2020).
Nitrogen compound‐specific isotope analysis
Nitrogen CSIA was undertaken to further distinguish the
trophic positions of these bears, as such data can provide a
more informative comparison between individuals from
different periods or that consume foods with different
environmental baseline nitrogen isotopic compositions. The
difference in
15
N enrichment between source and trophic
amino acids has been attributed to the amino acids’participa-
tion or lack thereof in deamination or transamination reactions
as the protein from food is incorporated into consumer tissue
(McMahon and McCarthy, 2016; O'Connell, 2017; Whiteman
et al., 2019). Therefore, animals occupying higher trophic
positions display a larger internal spacing between their Phe
(source) and Glx (trophic) amino acid δ
15
N values, regardless
of the level of
15
N enrichment at the base of the food chain.
In determining trophic positions for these samples by
comparing their δ
15
N
Phe
and δ
15
N
Glx
values, some additional
dietary factors should be considered. As a food source is
consumed and its amino acids are incorporated into a
consumer's tissues, the difference between the value by which
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
Figure 3. Bulk collagen carbon versus nitrogen
isotope compositions of the bear specimens. [Color
figure can be viewed at wileyonlinelibrary.com]
Figure 4. Comparison of Δ
13
C
Gly‐Phe
values versus
bulk δ
15
N values of the Vancouver Island bears. [Color
figure can be viewed at wileyonlinelibrary.com]
DIET OF LATE PLEISTOCENE BEARS 7

the source amino acid is enriched and the value by which the
trophic amino acid is enriched is deemed the trophic
discrimination factor (TDF). Therefore, the TDF describes the
expected increase in spacing between source and trophic
amino acid δ
15
N values with each increase in trophic position.
Because the TDF of Glx–Phe has been found to vary with the
quality and quantity of protein in the diet among other factors
(Germain et al., 2013; Chikaraishi et al., 2014; McMahon
et al.,2015; McMahon and McCarthy, 2016; Fuller and
Petzke, 2017), exact values for trophic position are not
presented. Threonine δ
15
N values have be explored as an
indicator of the level of protein consumption (Styring
et al., 2010). High dietary protein concentrations are thought
to cause increased enzymatic activity of threonine ammonia‐
lyase in mammals (Hare et al., 1991; Fuller and Petzke, 2017),
which preferentially removes
15
N over
14
N and thus causes a
decrease in the δ
15
N values of threonine. Therefore, lower
δ
15
N values of threonine have been proposed as a biomarker
for protein consumption. As shown in Table 2, the
U. americanus display moderately low δ
15
N
Thr
values (−11.2
and −12.5‰); U. arctos show similar (albeit slightly lower)
δ
15
N
Thr
values compared to U. americanus (−13.9 and
−15.2‰); and Arctodus display significantly lower δ
15
N
Thr
values than the other two taxa (−24.9 and ‐25.6‰). This may
indicate significantly higher levels of protein consumption by
Arctodus compared to the other two taxa. Previous research
indicates that as dietary protein quality increases, the TDF
between source and trophic amino acids tends to decrease
(McMahon and McCarthy, 2016). Therefore, if higher levels of
protein are being consumed by Arctodus compared to the
other taxa examined here, a lower TDF is probably more
appropriate for determining their trophic positions. If the TDF
value being applied universally to all taxa here is higher than
the appropriate value for the Arctodus samples, the trophic
positions of Arctodus will be underestimated in this com-
parison.
Although a custom TDF could not be calculated for these
ancient samples due to a lack of direct information on protein
quality, the limits of the trophic levels defined using a TDF
Glx‐
Phe
of 7.6‰(Chikaraishi et al.,2014) were considered
consistent enough as used in other studies on ancient bone
collagen (e.g. Naito et al., 2010,2013; Ogawa et al., 2013;
Chikaraishi et al., 2014) to usefully inform broad comparisons
in a visual aid, while keeping in mind the potential under-
estimation of trophic position for the Arctodus samples
(Figure 5). Because the results of both carbon CSIA and bulk
collagen SIA indicate that none of the bears considered in this
study show a high marine protein signal, trophic level can be
considered by employing the βvalue (which represents the
initial estimated spacing between the δ
15
N values of Glu and
Phe at the base of the food chain) for terrestrial ecosystems. For
the sake of comparison, the βvalue as previously determined
for marine ecosystems (Chikaraishi et al.,2014) is also shown
in Fig. 5. The unrealistically low marine trophic position that
the bears would have occupied while displaying these values
(<TP2) is a further indication that terrestrial resources were
instead the main protein source for all taxa examined here.
A comparison of trophic positions using δ
15
N values of Glu
and Phe reveals that U. americanus occupied a distinctly lower
trophic level than the other taxa, while U. arctos and Arctodus
show some overlap in their trophic positions, with a potential
underestimation of Arctodus trophic position as discussed
above (see Figure 5).
An additional noteworthy source of variation in the nitrogen
isotope composition of bear tissue may be sex or size.
Although we were unable to reliably determine the sex of
the specimens in this study based on the morphological
characteristics of the fragmentary remains, Hobson et al.
(2000) observed a significant difference in the δ
15
N values of
modern female U. arctos compared to their male counterparts
in areas where they coexist with U. americanus. This could be
due to the female bears’smaller range, and thus lower access
to higher trophic‐level resources (Hobson et al., 2000).
Additionally, it has been suggested via bite size ecology
studies that smaller U. arctos can sustain their energy on a
vegetarian diet, but very large bears cannot (Rode at al. 2001).
This, considered with the sexual dimorphism of the species,
could explain the sex‐patterned difference in δ
15
N values of
bear collagen, and could indicate size as a constraint on the
level of herbivory possible for short‐faced bears. The Arctodus
specimens analyzed here were noted by Steffen et al. (2018) to
be possibly female, based on their size. If not conclusively
female, they were at least noted to be relatively small in
comparison with other Arctodus specimens. Considering the
small sample size compared here and potential size and sex
influences on bear diet, conclusions drawn regarding trophic
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
Figure 5. Comparison of δ
15
N values for Glx and
Phe. Estimated trophic position delineations for both
marine and terrestrial consumers are defined
following the equation and graphing procedure
outlined by Chikaraishi et al. (2014) following a
trophic discrimination factor of 7.6‰;βvalue of
−3.4‰for marine and βvalue of 8.4‰for
terrestrial. [Color figure can be viewed at
wileyonlinelibrary.com]
8 JOURNAL OF QUATERNARY SCIENCE

position and resource specialization should be taken as
contributors to understanding Arctodus diet in a particular
context, and not as representative of an entire extinct species.
Discussion
Late Pleistocene bear sympatry
As evidenced by the radiocarbon dates presented here,
U.arctos,U.americanusand A. simus were present on Vancouver
Island in the Terminal Late Pleistocene, with all three bear species
present with the approximate range of ~13 800–13 500 cal a BP.
When contemporaneous in the same area, U. americanus have
been observed to occupy ranges of up to 40 km, while U. arctos
have been seen to occupy ranges of up to 111 km (Mowat
et al., 2005); therefore, range overlap between these site locations
is feasible. We surmise that these three bear species could have
lived within overlapping territories during the Terminal Late
Pleistocene on Vancouver Island.
Examples of two bear species occupying the same region
can be found in both modern and past environments, and may
provide insight into the competitive pressures that could have
arisen when the three species examined here were alive.
U. americanus have been present in North America since at
least the Middle Pleistocene and currently overlap geographi-
cally with U. arctos in certain regions, with competition
sometimes causing seasonal displacement of U. americanus
(Herrero, 1972; Belant et al., 2010). It is thought that these
species overlapped south of the Cordilleran and Laurentide ice
sheets as early as 31 000 cal a BP (Davison et al., 2011).
The ranges of these two species overlapped heavily through
the Holocene prior to European arrival in North America
(Herrero, 1972). U. americanus tend to have much lower
population densities in areas where U. arctos are also present
(Miller et al., 1997; Mowat et al., 2005). In locations where
these two species coexist today, U. americanus territorial
ranges are much smaller than the ranges of sympatric U. arctos
(Mowat et al., 2005).
U. arctos and A. simus have been found to co‐occur
elsewhere during the Pleistocene, including their overlap in
eastern Beringia ~45–35 ka (Barnes et al., 2002). Barnes et al.
(2002) noted an inverse correlation between the radiocarbon
dates of Arctodus and Ursus in Eastern Beringia, suggesting
that higher populations of Arctodus in the region from 35 to 21
ka had temporarily excluded large populations of Ursus from
the area. More recently, Steffen and Fulton (2018) proposed a
similar effect at Pellucidar Cave on Vancouver Island, where
an ~1300‐year gap between U. arctos radiocarbon dates is
bridged by two Arctodus dates (here referred to as S‐SFU 278
and 279). However, incorporating new radiocarbon evidence
from other montane sites on the island, it appears that the
presence of A. simus did not act to exclude Ursus from the
region. Despite the limited number of individuals recovered
for any bear species on Vancouver Island from this time, both
U. americanus (S‐SFU 276 at 13 981–13 611 cal BP) and
U. arctos (S‐SFU 272 at 13 758–13 609 cal BP) appear in close
chronological association with each other and with A. simus
(S‐SFU 279 at 13 741–13 520 a BP) (see Figure 2).
Pacific Northwest Coast Late Pleistocene
paleoenvironment
The geographical and temporal specifics of the Cordilleran ice
sheet's extent on Vancouver Island, as well as associated sea‐
level changes, are still being revealed through the accumula-
tion of local paleofauna, paleoflora and sedimentological data,
and the presence of glacial refugia in the region has been
explored. Based on a composite date from an associated
juvenile specimen, Nagorsen et al. (1995) previously reported
that the bear specimens from Windy Link Pot Cave (S‐SFU 269
and 270) were too young to provide insight into the
Pleistocene refugium hypothesis. However, the new direct
dates on these specimens reveal that they are several thousand
years older than originally believed, placing them at the end of
the Pleistocene. These dates add to a growing body of
radiocarbon evidence for ice‐free environments on Vancouver
Island that were able support a variety of large terrestrial
mammals, including these three species of bear, from as early
as 14 000 years ago (e.g. Nagorsen and Keddie, 2000;
Al‐Suwaidi et al., 2006; Harington, 2011; McLaren et al., 2014;
Steffen and Fulton, 2018).
Competition and partitioning
Considering the potential range overlap for some of these
samples, and with their temporal overlap now confirmed through
radiocarbon dating, it appears that inter‐species niche partitioning
within a shared geographical area, rather than long‐term
competitive geographical displacement, occurred on the island.
The partitioning of resources between bear species in shared
geographical areas has been observed in modern case studies.
Observational studies (Belant et al., 2010) and isotopic
investigations (Jacoby et al., 1999) have found that although
U. americanus may consume large amounts of salmon and other
higher trophic‐level resources in environments where U. arctos
are rare or absent, competition with U. arctos will influence the
nature of their resource intake. In environments where these two
bear species overlap, U. arctos are observed to take over the
higher trophic niche, create avoidance at the population level
and seasonally displace the local U. americanus.
Such a pattern of niche partitioning seems apparent in the
isotopic evidence reported by Fedje et al. (2011) from the most
temporally and geographically similar data from U. ameri-
canus and U. arctos –an assemblage from Haida Gwaii, a large
island to the north of Vancouver Island, dated to ~1000–2000
years later than the bears studied here. The bears from the
Younger Dryas interval on Haida Gwaii show very distinct
differentiation in the carbon and nitrogen isotope composition
of U. americanus and U. arctos collagen, with
U. arctos displaying δ
13
C and δ
15
N values up to 5.2 and
17.5‰higher, respectively, than those of U. americanus. This
can probably be attributed to differences in salmon consump-
tion, and the isotopic composition of contextually associated
salmon remains from the site supports this attribution (Fedje
et al., 2011).
Compared to these data, our Vancouver Island Ursus
specimens show much less differentiation. Although the
smaller sample size in the current dataset should not be
ignored, the largest differences between bulk collagen δ
13
C
and δ
15
N values of our samples are 3.7 and 10.0‰,
respectively, showing a much narrower difference between
these species. However, results of CSIA do suggest that
U. americanus occupied a distinctly lower trophic level than
both U. arctos and Arctodus, and the smaller range of isotopic
values could be due to the lack of aquatic resource
consumption by the higher trophic level taxa.
Although these samples show potential range overlap
between species, it is possible that the different taxa were
specialized to different environmental settings, which vary
greatly across small geographical areas on the mountainous
island. It has been noted that U. arctos aremoreopen‐adapted,
while U. americanus are forest‐adapted (Herrero, 1972); however,
this distinction may have been complicated by competition from
A. simus, another potentially open‐adapted species.
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
DIET OF LATE PLEISTOCENE BEARS 9

In a comparison of the bulk collagen carbon and nitrogen
isotopes of Arctodus and U. arctos from sympatric contexts in
Beringia, Matheus (1995) found that Arctodus in the Yukon and
Alaska cluster tightly in terms of their high δ
15
N values, while
U. arctos display a broader range of values. Matheus suggested
that the U. arctos out‐competed Arctodus because they were able
to be more flexible and use vegetation supplements, while
Arctodus could not. Matheus also argued that the data support
Arctodus as being strict carnivores in Beringia, and that
morphological data suggest that they were specialized scavengers
of large mammal carcasses (such as mammoths), while U. arctos
were more generalized consumers. However, Hobson et al.
(2000) caution against interpreting high δ
15
N values in bears as
the consumption of large terrestrial mammal prey, because
modern studies utilizing bulk CN measurements have shown that
bears that feed heavily on insect prey (such as ants) are
isotopically indistinguishable from bears that gain most of their
protein from large ungulates. On Vancouver Island, similar
trophic positions are indicated for U. arctos and Arctodus;
however, the resources exploited by the two taxa probably
differed in protein content (as indicated by the difference in
δ
15
N
Thr
), indicating a differentiation in prey choice within the
same trophic level (i.e. insects versus terrestrial, plant‐consuming
mammals).
Considering the potential effects of competition on the
resource exploitation patterns of all cotemporaneous bear
taxa, rather than simply assigning designations such as
“specialized scavenger”to an extinct taxon such as Arctodus,
may provide a more context‐specific understanding of the
ecological roles of these taxa. For example, Barnes et al. (2002)
report bulk collagen stable isotope data from Arctodus and
U. arctos in Beringia that show an increase in δ
15
N values for
U. arctos after the extirpation of Arctodus from the region,
demonstrating the effects of competition between the taxa on
their diets. In addition, recent isotopic data from Arctodus from
a specimen found on the California Channel Islands points to a
greater degree of omnivory than previously seen in isotopic
data for the taxon, which the authors suggest may have been
caused by inter‐species competition (Mychajliw et al., 2020).
Conclusions
We have measured the isotope values to determine the diet of
three temporally overlapping bear species –U. americanus,
U. arctos and A. simus –from Terminal Pleistocene Canada
(Vancouver Island). The island was sufficiently deglaciated by
~14 500 years ago to have a diverse and productive
environment that could support populations of several large,
omnivorous taxa. According to our bulk collagen and
compound‐specific isotope analysis results, these bear species
appear to occupy distinct ecological niches on the island, with
U. americanus occupying a distinctly lower trophic position,
and U. arctos and A. simus occupying higher trophic positions
with contributions of different specialized resources. As the
body of CSIA data on extinct Pleistocene species grows,
additional high‐resolution dietary comparisons across different
regions and periods will add to our understanding of how
these animals lived and interacted.
Supporting information
Additional supporting information can be found in the online
version of this article.
Tables S1. Carbon isotope results for the Bovine Gelatin QC
sample for the dataset analysed at MUN (S‐SFU 269, 270, 276,
278, 279)
Tables S2. Carbon isotope results for collagen QC samples
(SRM‐1 and SRM‐2) for the dataset measured at SFU (S‐SFU
272, 277 and 281)
Table S3: Nitrogen Isotope results for collagen QC samples
(SRM‐1 and SRM‐2) for the dataset measured at SFU (S‐SFU
272, 277 and 281)
Figure S1. Comparison of δ15N values of Hydroxyproline
and Proline measured within each collagen sample (from both
laboratories).
Figure S2. Typical CO2 gas chromatogram (S‐SFU 227) of
NAIP ester derivatized amino acids from bone collagen.
Ala =alanine, Val =valine, Gly =glycine, Leu =leucine,
Nor =norleucine, Pro =proline, Thr =threonine, Asx =aspar-
tic acid, Ser =serine, Glx =glutamic acid, Phe =phenylala-
nine, Hyp =hydroxyproline, Lys =lysine.
Figure S3. Typical N2 gas chromatogram (S‐SFU 272) of
NAIP ester derivatized amino acids from bone collagen.
Ala =alanine, Gly =glycine, Val =valine, Leu =leucine,
Nor =norleucine, Thr =threonine, Ser =serine, Pro =proline,
Asx =aspartic acid, Glx =glutamic acid, Hyp =hydroxypro-
line, Phe =phenylalanine, Lys =lysine.
Figure S4. MALDI‐ToF mass spectra of collagen digests from
selected specimens for this study.
Acknowledgements. This research was carried out with funding from
SSHRC and NSERC. We thank Megan Wong for assistance with lab
work and Laura Termes for assistance with sample context and
background information.
Abbreviations. LGM, Last Glacial Maxuimum; ACN, acetonitrile; TFA,
trifluoroacetic acid; AMS, accelerator mass spectrometry; BSIA, bulk
stable isotope analysis; CSIA, compound‐specific stable isotope
analysis; EA‐IRMS, elemental analysis ‐isotope ratio mass spectro-
metry; GC‐C, gas chromatography combustion; QC, quality control,
aDNA, ancient DNA; TDF, trophic discrimination factor.
References
Al‐Suwaidi M, Ward BC, Wilson MC, et al. 2006. Late Wisconsinan
Port Eliza Cave deposits and their implications for human coastal
migration, Vancouver Island, Canada. Geoarchaeology 21:
307–332. https://doi.org/10.1002/gea.20106
Austin AT, Vitousek PM. 1998. Nutrient dynamics on a precipitation
gradient in Hawai'i. Oecologia 113(4): 519–529. https://doi.org/10.
1007/s004420050405
Ambrose SH. 1990. Preparation and characterization of bone and
tooth collagen for isotopic analysis. Journal of Archaeological
Science 17: 431–451. https://doi.org/10.1016/0305-4403(90)
90007-R
Barnes I, Matheus P, Shapiro B, Cooper A. 2002. Dynamics of
Pleistocene Population Extinctions in Beringian Brown Bears.
Science 295: 2267–2270. https://doi.org/10.1126/science.1067814
Belant JL, Griffith B, Zhang Y, Follmann EH, Adams LG. 2010.
Population‐level resource selection by sympatric brown and
American black bears in Alaska. Polar Biology 33:31–40. https://
doi.org/10.1007/s00300-009-0682-6
Bennett K, Provan J. 2008. What do we mean by ‘refugia'? Quaternary
Science Reviews 27: 2449–2455. https://doi.org/10.1016/j.
quascirev.2008.08.019
Bocherens H, Emslie SD, Billiou D, Mariotti A. 1995. Stable isotopes
(13C, 15N) and paleodiet of the giant short‐faced bear (Arctodus
simus). Comptes Rendus de l'Acade
mie des Sciences, Se
rie II, Paris
320: 779–784.
Broink Ramsey C. 2009. Bayesian Analysis of Radiocarbon Dates.
Radiocarbon 51: 337–360.
Buckley M, Collins M, Thomas‐Oates J, Wilson JC. 2009. Species
identification by analysis of bone collagen using matrix‐assisted
laser desorption/ionisation time‐of‐flight mass spectrometry. Rapid
Communications in Mass Spectrometry 23: 3843–3854. https://doi.
org/10.1002/rcm.4316
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
10 JOURNAL OF QUATERNARY SCIENCE

Buckley M, Collins MJ. 2011. Collagen survival and its use for species
identification in Holocene‐lower Pleistocene bone fragments from
British archaeological and paleontological sites. Antiqua 1:1.
https://doi.org/10.4081/antiqua.2011.e1
Chikaraishi Y, Ogawa NO, Kashiyama Y, Takano Y, Suga H, Tomitani
A, Miyashita H, Kitazato H, Ohkouchi N. 2009. Determination of
aquatic food‐web structure based on compound‐specific nitrogen
isotopic composition of amino acids. Limnology and Oceanogra-
phy: methods 7: 740–750.
Chikaraishi Y, Steffan SA, Ogawa NO, Ishikawa NF, Sasaki Y, Tsuchiya
M, Ohkouchi N. 2014. High‐resolution food webs based on
nitrogen isotopic composition of amino acids. Ecol Evol 4:
2423–2449. https://doi.org/10.1002/ece3.1103
Coplen TB. 2011. Guidelines and recommended terms for expression
of stable‐isotope‐ratio and gas‐ratio measurement results: Guide-
lines and recommended terms for expressing stable isotope results.
Rapid Communications in Mass Spectrometry 25: 2538–2560.
https://doi.org/10.1002/rcm.5129
Corr LT, Berstan R, Evershed RP. 2007. Development of N‐Acetyl
Methyl Ester Derivatives for the Determination of δ
13
C Values of
Amino Acids Using Gas Chromatography‐Combustion‐Isotope
Ratio Mass Spectrometry. Analytical Chemistry 79: 9082–9090.
https://doi.org/10.1021/ac071223b
Corr LT, Sealy JC, Horton MC, Evershed RP. 2005. A novel marine
dietary indicator utilising compound‐specific bone collagen amino
acid δ
13
C values of ancient humans. Journal of Archaeological
Science 32: 321–330. https://doi.org/10.1016/j.jas.2004.10.002
Cosma TN, Hendy IL, Chang AS. 2008. Chronological constraints on
Cordilleran Ice Sheet glaciomarine sedimentation from core MD02‐
2496 off Vancouver Island (western Canada). Quaternary Science
Reviews 27: 941–955. https://doi.org/10.1016/j.quascirev.2008.
01.013
Craine JM, Elmore AJ, Aidar MPM, Bustamante M, Dawson TE, Hobbie
EA, Kahmen A, Mack MC, McLauchlan KK, Michelsen A, Nardoto
GB, Pardo LH, Peñuelas J, Reich PB, Schuur EAG, Stock WD,
Templer PH, Virginia RA, Welker JM, Wright IJ. 2009. Global
patterns of foliar nitrogen isotopes and their relationships with
climate, mycorrhizal fungi, foliar nutrient concentrations, and
nitrogen availability. New Phytologist 183: 980–992. https://doi.
org/10.1111/j.1469-8137.2009.02917.x
Crann CA, Murseli S, St‐Jean G, Zhao X, Clark ID, Kieser WE. 2017.
First Status Report on Radiocarbon Sample Preparation Techniques
at the A.E. Lalonde AMS Laboratory (Ottawa, Canada). Radiocarbon
59: 695–704. https://doi.org/10.1017/RDC.2016.55
Davis H. 1996. Characteristics and Selection of Winter Dens by Black
Bears in Coastal British Columbia. M.Sc. Thesis, Dept. of Biological
Sciences. Simon Fraser University, British Columbia.
Davison J, Ho SY, Bray SC, Korsten M, Tammeleht E, Hindrikson M,
Østbye K, Østbye E, Lauritzen S, Austin J. 2011. Late‐Quaternary
biogeographic scenarios for the brown bear (Ursus arctos), a wild
mammal model species. Quaternary Science Reviews 30(3):
418–430.
DeNiro MJ. 1985. Postmortem preservation and alteration of in vivo
bone collagen isotope ratios in relation to palaeodietary reconstruc-
tion. Nature 317: 806–809. https://doi.org/10.1038/317806a0
DeNiro MJ, Epstein S. 1978. Influence of diet on the distribution of
carbon isotopes in animals. Geochimica et Cosmochimica Acta 42:
495–506. https://doi.org/10.1016/0016-7037(78)90199-0
DeNiro MJ, Epstein S. 1981. Influence of diet on the distribution of
nitrogen isotopes in animals. Geochimica et Cosmochimica Acta
45: 341–351.
Donohue SL, DeSantis LRG, Schubert BW, Ungar PS. 2013. Was the
giant short‐faced bear a hyper‐scavenger? A new approach to the
dietary study of ursids using dental microwear textures. PLoS ONE 8:
e77531.
Dykstra EA. 2015. Using stable isotope analysis to estimate black bear
(Ursus americanus) diet in Vermont 148. University of Vermont
Scholarworks Graduate College Dissertations and Theses. 388.
https://scholarworks.uvm.edu/graddis/388
Eamer JBR, Shugar DH, Walker IJ, Lian OB, Neudorf CM, Telka AM.
2017. A glacial readvance during retreat of the Cordilleran Ice
Sheet, British Columbia central coast. Quat. res. 87: 468–481.
https://doi.org/10.1017/qua.2017.16
Emslie SD, Czaplewski NJ. 1985. A new record of giant short‐faced
bear, Arctodus simus, from western North America with a re‐
evaluation of its paleobiology. Contributions in science 371:1–12.
https://doi.org/10.5962/p.226835
Fedje D, Mackie Q, Lacourse T, McLaren D. 2011. Younger Dryas
environments and archaeology on the Northwest Coast of North
America. Quaternary International 242: 452–462. https://doi.org/10.
1016/j.quaint.2011.03.042
Figueirido B, Pérez‐Claros JA, Torregrosa V, Martín‐Serra A, Palmqvist P.
2010. Demythologizing Arctodus simus,the‘short‐faced’long‐legged
and predaceous bear that never was. Journal of Vertebrate Paleontol-
ogy 30:262–275. https://doi.org/10.1080/02724630903416027
Figueirido B, Pérez‐Ramos A, Schubert BW, Serrano F, Farrell AB,
Pastor FJ, Neves AA, Romero A. 2017. Dental caries in the fossil
record: a window to the evolution of dietary plasticity in an extinct
bear. Scientific Reports 7(1): 17813. https://doi.org/10.1038/s41598-
017-18116-0
Fountoulakis M, Lahm HW. 1998. Hydrolysis and amino acid
composition analysis of proteins. Journal of Chromatography A
826: 109–134.
Fox Dobbs K, Leonard JA, Koch PL. 2008. Pleistocene megafauna from
eastern Beringia: paleoecological and paleoenvironmental inter-
pretations of stable carbon and nitrogen isotope and radiocarbon
records. Palaeogeogr. Palaeoclimatol. Palaeoecol. 261:30–46.
Fuller BT, Petzke KJ. 2017. The dietary protein paradox and threonine
15
N‐depletion: Pyridoxal‐5′‐phosphate enzyme activity as a me-
chanism for the δ
15
N trophic level effect: Dietary protein paradox
and amino acid specific δ
15
N measurements. Rapid Communica-
tions in Mass Spectrometry 31: 705–718. https://doi.org/10.1002/
rcm.7835
Germain L, Koch P, Harvey J, McCarthy M. 2013. Nitrogen isotope
fractionation in amino acids from harbor seals: implications for
compound‐specific trophic position calculations. Marine Ecology
Progress Series 482: 265–277. https://doi.org/10.3354/meps10257
Gordon KR. 31 May 1977. Molar Measurements as a Taxonomic Tool
in Ursus. Journal of Mammalogy 58(Issue 2): 247–248. https://doi.
org/10.2307/1379593
Gordon KR. 1986. Insular Evolutionary Body Size Trends in Ursus.
Journal of Mammalogy 67: 395–399. https://doi.org/10.2307/
1380895
Hall ER. 1928. A new race of black bear from Vancouver Island, British
Columbia, with remarks on other northwest coast forms of Euarctos.
University of California Publications of Zoology 30: 231–242.
Hare PE, Fogel ML, Stafford TW, Jr., Mitchell AD, Hoering TC. 1991.
The isotopic composition of carbon and nitrogen in individual
amino acids isolated from modern and fossil protein. J. Archaeol.
Sci. 18: 277.
Harington CR. 2011. Quaternary Cave Faunas of Canada: A Review of
the Vertebrate Remains. Journal of Cave and Karst Studies 73:
162–180. https://doi.org/10.4311/jcks2009pa128
Heaton TH, Grady F. 2003. "The Late Wisconsin vertebrate history of
Prince of Wales Island, southeast Alaska". In Ice Age Cave Fauna of
North America, Schubert BW, Mead JI, Graham RW (eds). Indiana
University Press, 17–53 ISBN 978‐0‐253‐34268‐3.
Hebda RJ. 1983. Late‐glacial and post‐glacial vegetation history at
Bear Cove Bog, northeast Vancouver Island, British Columbia.
Canadian Journal of Botany 61(12): 3172–3192. https://doi.org/10.
1139/b83-355
Herrero S. 1972. Aspects of Evolution and Adaptation in American
Black Bears (Ursus americanus Pallas) and Brown and Grizzly Bears
(U. arctos Linné.) of North America. Bears: Their Biology and
Management 2: 221. https://doi.org/10.2307/3872586
Hildebrand GV, Farley SD, Robbins CT, Hanley TA, Titus K, Servheen
C. 1996. Use of stable isotopes to determine diets of living and
extinct bears. Canadian Journal of Zoology 74: 2080–2088. https://
doi.org/10.1139/z96-236
Hobson KA, Mclellan BN, Woods JG. 2000. Using Stable Carbon
(δ
13
c) and Nitrogen (δ
15
N) isotopes to infer trophic relationships
among black and grizzly bears in the Upper Columbia River Basin.
British Columbia.
Honch NV, McCullagh JSO, Hedges REM. 2012. Variation of bone
collagen amino acid δ
13
c values in archaeological humans and
fauna with different dietary regimes: Developing frameworks of
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
DIET OF LATE PLEISTOCENE BEARS 11

dietary discrimination. American Journal of Physical Anthropology
148: 495–511. https://doi.org/10.1002/ajpa.22065
Hopkins JB, Ferguson JM, Tyers DB, Kurle CM. 2017. Selecting the
best stable isotope mixing model to estimate grizzly bear diets in the
Greater Yellowstone Ecosystem. Edited by Jordi Moya‐Larano. PLOS
ONE 12(5): e0174903. https://doi.org/10.1371/journal.pone.
0174903
Howes DE. 1983. Late Quaternary sediments and geomorphic
history of northern Vancouver Island, British Columbia. Canadian
Journal of Earth Sciences 20:57–65. https://doi.org/10.1139/
e83-006
Howland MR, Corr LT, Young SMM, Jones V, Jim S, Van Der Merwe
NJ, Mitchell AD, Evershed RP. 2003. Expression of the dietary
isotope signal in the compound‐specific 13C values of pig bone
lipids and amino acids. International Journal of Osteoarchaeology
13(1–2): 54–65. https://doi.org/10.1002/oa.658
James TS, Clague JJ, Wang K, Hutchinson I. 2000. Postglacial rebound
at the northern Cascadia subduction zone. Quaternary Science
Reviews 19(14–15): 1527–1541. https://doi.org/10.1016/S0277-
3791(00)00076-7
Jarman CL, Larsen T, Hunt T, Lipo C, Solsvik R, Wallsgrove N, Ka'apu‐
Lyons C, Close HG, Popp BN. 2017. Diet of the prehistoric
population of Rapa Nui (Easter Island, Chile) shows environmental
adaptation and resilience: JARMAN et al. American Journal of
Physical Anthropology 164: 343–361. https://doi.org/10.1002/ajpa.
23273
Jacoby ME, Hilderbrand GV, Servheen C, Schwartz CC, Arthur SM,
Hanley TA, Robbins CT, Michener R. 1999. Trophic Relations of
Brown and Black Bears in Several Western North American
Ecosystems. The Journal of Wildlife Management 63: 921. https://
doi.org/10.2307/3802806
Kantelis TM. 2017. Black Bears (Ursus americanus) versus Brown
Bears (U. arctos): Combining Morphometrics and Niche Modeling
to Differentiate Species and Predict Distributions Through Time. A
thesispresented to the faculty of the Department of Geosciences East
Tennessee State University.
Kurtén B, Anderson E. 1974. Association of Ursus arctos and Arctodus
simus (Mammalia: Ursidae) in the late Pleistocene of Wyoming.
Breviora 426:1–6.
Lacourse T. 2005. Late Quaternary dynamics of forest vegetation on
northern Vancouver Island, British Columbia, Canada. Quaternary
Science Reviews 24(1–2): 105–121. https://doi.org/10.1016/j.
quascirev.2004.05.008
Larsen T, Taylor DL, Leigh MB, O'Brien DM. 2009. Stable isotope
fingerprinting: a novel method for identifying plant, fungal, or
bacterial origins of amino acids. Ecology 90: 3526–3535. https://doi.
org/10.1890/08-1695.1
Larsen T, Ventura M, Andersen N, O'Brien DM, Piatkowski U,
McCarthy MD. 2013. Tracing Carbon Sources through Aquatic and
Terrestrial Food Webs Using Amino Acid Stable Isotope Fingerprint-
ing. PLoS ONE 8: e73441. https://doi.org/10.1371/journal.pone.
0073441
Longin R. 1971. New Method of Collagen Extraction for Radiocarbon
Dating. Nature 230: 241–242. https://doi.org/10.1038/230241a0
Matheus PE. 1995. Diet and Co‐ecology of Pleistocene Short‐Faced
Bears and Brown Bears in Eastern Beringia. Quaternary Research 44:
447–453.
McLaren D, Fedje D, Hay MB, Mackie Q, Walker IJ, Shugar DH,
Eamer JBR, Lian OB, Neudorf C. 2014. A post‐glacial sea level hinge
on the central Pacific coast of Canada. Quaternary Science Reviews
97: 148–169. https://doi.org/10.1016/j.quascirev.2014.05.023
McClelland JW, Joseph PM. 2002. Trophic relationships and the
nitrogen isotopic composition of amino acids in plankton. Ecology
83(8): 2173–2180.
McMahon KW, Polito MJ, Abel S, McCarthy MD, Thorrold SR. 2015.
Carbon and nitrogen isotope fractionation of amino acids in an
avian marine predator, the gentoo penguin (Pygoscelis papua).
Ecology and Evolution 5: 1278–1290. https://doi.org/10.1002/
ece3.1437
McMahon KW, McCarthy MD. 2016. Embracing variability in amino
acid δ
15
N fractionation: mechanisms, implications, and applica-
tions for trophic ecology. Ecosphere 7: e01511. https://doi.org/10.
1002/ecs2.1511
Miller SD, White GC, Sellers RA, Reynolds HV, Schoen JW, Titus K,
Barnes VG Jr., Smith RB, Nelson RR, Ballard WB, Schwartz CC.
1997. Brown and black bear density estimation in Alaska using
radiotelemetry and replicated mark‐resight techniques. Wildlife
Monograph 133: 55.
Mowat G, Heard DC, Seip DR, Poole KG, Stenhouse G, Paetkau DW.
2005. Grizzly Ursus arctos and black bear U. americanus densities
in the interior mountains of North America. Wildlife Biology 11:
31–48. https://doi.org/10.2981/0909-6396(2005)11[31:GUAABB]2.
0.CO;2
Müldner G, Richards MP. 2005. Fast or feast: reconstructing diet in
later medieval England by stable isotope analysis. Journal of
Archaeological Science 32:39–48. https://doi.org/10.1016/j.jas.
2004.05.007
Mychajliw AM, Rick TC, Dagtas ND, Erlandson JM, Culleton BJ,
Kennett DJ, Buckley M, Hofman CA. 2020. Biogeographic problem‐
solving reveals the Late Pleistocene translocation of a short‐faced
bear to the California Channel Islands. Scientific Reports 10(1):
15172. https://doi.org/10.1038/s41598-020-71572-z
Nagorsen DW, Keddie G, Hebda RJ. 1995. Early Holocene Black
Bears, Ursus americanus, from Vancouver Island. Canadian Field.
Naturalist 109(1): 11–18.
Nagorsen DW, Keddie G. 2000. Late Pleistocene mountain goats
(oreamnos americanus) from vancouver island: biogeographic
implications. Journal of mammalogy 81: 10.
Naito YI, Chikaraishi Y, Ohkouchi N, Drucker DG, Bocherens H.
2013. Nitrogen isotopic composition of collagen amino acids as an
indicator of aquatic resource consumption: insights from Mesolithic
and Epipalaeolithic archaeological sites in France. World Archae-
ology 45: 338–359. https://doi.org/10.1080/00438243.2013.
820650
Naito YI, Honch NV, Chikaraishi Y, Ohkouchi N, Yoneda M. 2010.
Quantitative evaluation of marine protein contribution in ancient
diets based on nitrogen isotope ratios of individual amino acids in
bone collagen: An investigation at the Kitakogane Jomon site.
American Journal of Physical Anthropology 143:31–40. https://doi.
org/10.1002/ajpa.21287
Newsome SD, Fogel ML, Kelly L, del Rio CM. 2011. Contributions of
direct incorporation from diet and microbial amino acids to protein
synthesis in Nile tilapia. Functional Ecology 25: 1051–1062. https://
doi.org/10.1111/j.1365-2435.2011.01866.x
Nielsen‐Marsh C. 1 June 2002. Biomolecules in fossil remains:
Multidisciplinary approach to endurance. Biochem (Lond) 24(3):
12–14. https://doi.org/10.1042/BIO02403012.
O'Connell TC. 2017. ‘Trophic’and ‘source’amino acids in trophic
estimation: a likely metabolic explanation. Oecologia 184:
317–326. https://doi.org/10.1007/s00442-017-3881-9
Ogawa NO, Chikaraishi Y, Ohkouchi N. 2013. Trophic position
estimates of formalin‐fixed samples with nitrogen isotopic composi-
tions of amino acids: an application to gobiid fish (Isaza) in Lake
Biwa, Japan. Ecol. Res. 28: 697–702.
Ramsey Carolyn, Griffiths Paul, Fedje Daryl, Wigen Rebecca, Mackie
Quentin. 2004. Preliminary investigation of a late Wisconsinan
fauna from K1 cave, Queen Charlotte Islands (Haida Gwaii),
Canada. Quaternary Research 62: 105–109. https://doi.org/10.
1016/j.yqres.2004.05.003
Richards RL, Churcher CS, Turnbull WD. 1996. Distribution and size
variation in North American short‐faced bears Arctodus simus. In
Palaeoecology and palaeoenvironments of late Cenozoic mammals:
tributes to the career of C. S. (Rufus) Churcher, Stewart KM, Seymour
KL (eds). University of Toronto Press: Toronto; 191–246.
Richards MP, Pacher M, Stiller M, Quilès J, Hofreiter M, Constantin S,
Zilhão J, Trinkaus E. 2008. Isotopic evidence for omnivory among
European cave bears: Late Pleistocene Ursus spelaeus from the
Peştera cu Oase, Romania. Proceedings of the National Academy of
Sciences 105, 600–604.
Reimer P, Austin W, Bard E, Bayliss A, Blackwell P, Bronk Ramsey C,
Butzin M, Cheng H, Edwards R, Friedrich M, Grootes P, Guilderson
T, Hajdas I, Heaton T, Hogg A, Hughen K, Kromer B, Manning S,
Muscheler R, Palmer J, Pearson C, van der Plicht J, Reimer R,
Richards D, Scott E, Southon J, Turney C, Wacker L, Adolphi F,
Büntgen U, Capano M, Fahrni S, Fogtmann‐Schulz A, Friedrich R,
Köhler P, Kudsk S, Miyake F, Olsen J, Reinig F, Sakamoto M,
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
12 JOURNAL OF QUATERNARY SCIENCE

Sookdeo A, Talamo S. 2020. The IntCal20 Northern Hemisphere
radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon 62.
Rode KD, Robbins CT, Shipley LA. 2001. Constraints on herbivory by
grizzly bears. Oecologia 128:62–71. https://doi.org/10.1007/
s004420100637
Schwartz‐Narbonne R, Longstaffe FJ, Metcalfe JZ, Zazula G. 2015.
Solving the woolly mammoth conundrum: amino acid
15
N‐enrichment suggests a distinct forage or habitat. Scientific
Reports 5: 9791. https://doi.org/10.1038/srep09791
Silfer JA, Engel MH, Macko SA, Jumeau EL. 1991. Stable Carbon
Isotope Analysis of Amino Acid Enantiomers by Conventional
Isotope Ratio Mass Spectrometry and Combined Gas Chromato-
graphy/Isotope Ratio Mass Spectrometry. Analytical Chemistry 63:
370–374.
Steffen ML, Fulton TL. 2018. On the association of giant short‐faced
bear (Arctodus simus) and brown bear (Ursus arctos) in late
Pleistocene North America. Geobios 51:61–74. https://doi.org/10.
1016/j.geobios.2017.12.001
Steffen ML, Harington CR. 2010. Giant short‐faced bear (Arctodus
simus) from late Wisconsinan deposits at Cowichan Head,
Vancouver Island, British Columbia. Canadian Journal of Earth
Sciences 47: 1029–1036. https://doi.org/10.1139/E10-018
Steffen ML, McLaren D. 2008. Report on a Preliminary Investigation of
Pellucidar II Cave, Northern Vancouver Island, BC. Prepared for:
Archaeology Branch, ‘Namgis First Nation. Royal BC Museum.
Stenhouse MJ, Baxter MS. 1979. The uptake of bomb
14
C in humans.
In Radiocarbon Dating. Berkeley, Berger R, Suess HE (eds).
University of California Press: California, USA; 324–341.
Styring AK, Sealy JC, Evershed RP. 2010. Resolving the bulk δ
15
N
values of ancient human and animal bone collagen via compound‐
specific nitrogen isotope analysis of constituent amino acids.
Geochimica et Cosmochimica Acta 74: 241–251. https://doi.org/
10.1016/j.gca.2009.09.022
van der Sluis LG, Hollund HI, Buckley M, De Louw PGB, Rijsdijk KF,
Kars H. 2014. Combining histology, stable isotope analysis and ZooMS
collagen fingerprinting to investigate the taphonomic history and
dietary behaviour of extinct giant tortoises from the Mare aux Songes
deposit on Mauritius. Palaeogeography, Palaeoclimatology, Palaeoe-
cology 416:80–91. https://doi.org/10.1016/j.palaeo.2014.06.003
van Klinken GJ. 1999. Bone Collagen Quality Indicators for Palaeodietary
and Radiocarbon Measurements. Journal of Archaeological Science
26:687–695. https://doi.org/10.1006/jasc.1998.0385
Ward BC, Thomson B. 2004. Late Pleistocene stratigraphy and
chronology of lower Chehalis River valley, southwestern British
Columbia: evidence for a restricted Coquitlam Stade. Canadian
Journal of Earth Sciences 41: 881–895. https://doi.org/10.1139/
e04-037
Webb EC, Honch NV, Dunn PJH, Linderholm A, Eriksson G,
Lidén K, Evershed RP. 2018. Compound‐specific amino
acid isotopic proxies for distinguishing between terrestrial and
aquatic resource consumption. Archaeological and Anthropolo-
gical Sciences 10:1–18. https://doi.org/10.1007/s12520-015-
0309-5
Whiteman J, Elliott Smith E, Besser A, Newsome S. 2019. A Guide to
Using Compound‐Specific Stable Isotope Analysis to Study the Fates
of Molecules in Organisms and Ecosystems. Diversity 11:8.https://
doi.org/10.3390/d11010008
Wolverton S, Lyman RL. 1998. Measuring Late Quaternary Ursid
Diminution in the Midwest. Quaternary Research 49: 322–329.
https://doi.org/10.1006/qres.1998.1964
©2022 John Wiley & Sons, Ltd. J. Quaternary Sci., 1–13 (2022)
DIET OF LATE PLEISTOCENE BEARS 13