Você está na página 1de 0

Taphonomy, pathology, and paleoecology of the

terminal Pleistocene Marmes Rockshelter

(45FR50) big elk (Cervus elaphus), southeastern
Washington State, USA
R. Lee Lyman
Abstract: In 1968, remains of what were reported to be a larger-than-modern elk (Cervus elaphus) were recovered from
terminal Pleistocene sediments associated with the Marmes Rockshelter archaeological site in southeastern Washington
State. Originally thought to have been butchered by humans, it is associated with radiocarbon dates suggesting an age of
about 9800
C years B.P. Taphonomic analysis in 2009 indicates the elk likely died of natural causes during winter
months; it was lightly scavenged by carnivores prior to burial from silt-rich spring runoff. The elk suffered from two
pathological conditions: one resulting in fusion of the fourth and fifth cervical vertebrae, and the other resulting in exces-
sive bone tissue on the proximal ends of the first ribs, seventh cervical, and first and second thoracic vertebrae. The
Marmes elk is larger than modern Rocky Mountain elk (C. e. nelsoni) and is on the large end of the size range of modern
Roosevelt elk (C. e. roosevelti). It is also larger than the similarly aged elk skeleton from Three Hills, Alberta. A single
elk bone from the Sentinel Gap archaeological site in central Washington State, dated to about 10 200
C years BP and
located 130 km west of Marmes Rockshelter, is the same size as the same bone of the Marmes elk. Terminal Pleistocene
elk in eastern Washington likely grew to exceptionally large size as a result of abundant grass at the time, forage that de-
creased in abundance as Holocene climatic conditions developed.
Resume : En 1968, des restes alors attribues a` un wapiti plus grand que le wapiti moderne (Cervus elaphus) ont ete prele-
ves dans des sediments du Pleistoce`ne terminal associes au site archeologique de Marmes Rockshelter, du sud-est de lE

de Washington. Interpretes alors comme ayant ete depeces par des humains, ces restes sont associes a` des dates radiocar-
bone qui sugge`rent un age denviron 9 800 ans
C BP. Une analyse taphonomique realisee en 2009 indique que le wapiti
est probablement mort de causes naturelles durant les mois dhiver et que sa carcasse a ete entamee par des carnivores
avant detre enfouie sous du limon transporte par le ruissellement printanier. Lanimal souffrait de deux pathologies, dont
une a entra ne la fusion des quatrie`me et cinquie`me verte`bres cervicales et lautre, un exce`s de tissus osseux sur les extre-
mites proximales des premie`res cotes, de la septie`me verte`bre cervicale et des premie`re et deuxie`me verte`bres thoraciques.
Le wapiti de Marmes est plus imposant que le wapiti des montagnes Rocheuses moderne (C. e. nelsoni) et occupe la partie
superieure du spectre de dimensions du wapiti de Roosevelt moderne (C. e. roosevelti). Il est egalement plus grand que le
squelette de wapiti du meme age de Three Hills (Alberta). Un os isole ayant donne une date radiocarbone denviron
10 200 ans
C BP provenant du site archeologique de Sentinel Gap, dans le centre de lE

tat de Washington, a` 130 km a`

louest du site de Marmes Rockshelter, est de la meme taille que le meme os du wapiti de Marmes. Les wapitis du Pleisto-
ce`ne terminal de lest de lE

tat de Washington atteignaient vraisemblablement des tailles imposantes en raison de labon-

dance de lherbe a` cette epoque, abondance qui allait diminuer par la suite en raison des conditions climatiques
caracteristiques de lHoloce`ne.
[Traduit par la Redaction]
Initially excavated in 196264, the archaeological site
known as Marmes Rockshelter (official state site number:
45FR50) in southeastern Washington State came to national
attention in 1968 when some of the then oldest known hu-
man remains in North America were found in a stratigraphic
trench excavated from the mouth of the shelter out onto the
Palouse River floodplain in front the shelter (Fig. 1; Fryxell
and Keel 1969). Excavations became an emergency salvage
operation when it was realized that a coffer dam built
around the site would fail to keep reservoir waters behind
the recently completed Lower Monumental Dam on the
Snake River (to which the Palouse River is a tributary)
from flooding the rockshelter and floodplain sediments con-
taining archaeological materials and ancient human remains
(Hicks 2004). The human remains received a great deal of
publicity in the popular press (Anonymous 1968a; Grosso
1967; Kirk 1968, 1969, 1970; Young 1969) and among pro-
fessional archaeologists (Fryxell et al. 1968a, 1968b; Krantz
Received 10 February 2010. Accepted 14 June 2010. Published
on the NRC Research Press Web site at cjes.nrc.ca on 8 October
Paper handled by Associate Editor T. Fisher.
R.L. Lyman. Department of Anthropology, University of
Missouri, Columbia, MO 65211, USA
Can. J. Earth Sci. 47: 13671382 (2010) doi:10.1139/E10-059 Published by NRC Research Press
1979). The attention was because the human bones were, at
the time, among the (if not the) oldest human remains
known in the Americas, and initial interpretations were that
the several represented individuals had been the subject of
cannibalism (Anonymous 1968b).
The partial skeleton of a North American elk or wapiti
(Cervus elaphus) associated with the human remains re-
ceived less attention and publicity, and apparently raised
only a bit of interest from the profession at the time. It was,
however, noted almost immediately on discovery, in both
the popular literature (Kirk 1968) and the professional liter-
ature, that the elk was larger than those of living represen-
tatives in the area today (Fryxell et al. 1968a, p. 514) and,
hence, it came to be known in field notes as the big elk.
At the time it was reported that this animal may have been
from 10% to 20% taller at the shoulder than the modern
elk (Kirk 1968, p. 57). Later, it was noted in an unpub-
lished report that the Marmes [big elk] bones averaged
some 4% to 5% greater in length and nearly 8% greater in
diameter [than a large modern bull elk...]. The Marmes wa-
piti apparently was about 15% larger than this very large
modern specimen (Gustafson 1983, p. 24). No comparative
metric data have ever been presented to substantiate state-
ments about the size of the Marmes big elk.
The elk bones were also originally reported as having
been clearly butchered by the terminal Pleistocene human
occupants of the site (Fryxell et al. 1968b, p. 177). Subse-
quently, Gustafson and Wegener (2004, p. 314) reported
that the elk bones did not appear to represent butchered re-
mains because they differ from the [local] typical pat-
tern in terms of fragmentation, plus carnivore activity is
clearly indicated. None of the published and unpublished
reports included detailed descriptions of carnivore damage,
fragmentation, or morphometry, making it difficult to evalu-
ate any of the interpretations previously offered or to per-
form comparative analyses with other, similarly aged elk
remains from North America (Shackleton and Hills 1977).
Further, the remains display pathologies that have not previ-
ously been noticed and hence have not been described in the
literature. In this paper, the Marmes big elk remains are de-
scribed in terms of their taphonomic, pathologic, and metric
attributes and some of these observations are compared with
those made on other like-aged elk remains. Together, attrib-
utes of the Marmes-site big elk suggest aspects of local pa-
leoecology at the end of the Pleistocene.
Geological context and age of the elk
The majority of the big elk remains lay on a fluvially de-
Fig. 1. Location of archaeological sites and palynological sites in eastern Washington State mentioned in the text.
1368 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
posited silt layer approximately 4 m beneath the modern
ground surface. This silt layer was interpreted by project
geoarchaeologist Roald Fryxell to represent an incipient A
pedogenic horizon (Hicks 2004). Some of the elk remains
were found in sediment excavated by a bulldozer or backhoe
such that their precise stratigraphic context is unclear; others
were found lying relatively close together, but maps and
photographs of the bones in situ indicate only that two
thoracic vertebrae were in anatomical (articulated) position
relative to one another. The author has determined that these
articulated vertebrae were T6 and T7. Some of the speci-
mens from the backhoe trench that lack provenience data
clearly articulate with others for which precise information
on recovery location is available, suggesting that many of
the specimens were stratigraphically associated. The author
assumes this was indeed the case for all remains of large
elk recovered from the floodplain, even though several re-
mains were recorded as coming from (within?) the incipient
A horizon upon which many of the other bones of the big
elk reportedly lay. Supposedly , a direct accelerator mass
spectrometry radiocarbon date on an astragalus of the big
elk was obtained (Gustafson and Wegener 2004, p. 313),
but close study of laboratory notes and records indicates
that none of the reported radiocarbon dates from the excava-
tions (Hicks 2004, p. 390) can be unequivocally attributed to
that astragalus. Catalog, inventory, and specimen numbers
either do not align across multiple printed records, or prove-
nience information cannot be aligned across multiple written
Three radiocarbon dates, two on bone and one on fresh-
water pearl mussel shell (Margaritifera falcata), are avail-
able for the incipient A horizon upon which many of the
remains of the large elk lay. These dates are: 9710 40
(Beta-156699, bone); 9820 300 (W-2209, shell); and
9870 50 (Beta-120802, bone). The simple average of these
values is 9800
C years BP. Given that many of the bones
that were found in place are readily attributed to a single in-
dividual, and that many of the bones have the same strati-
graphic provenience, it is likely that all of the bones are of
the same age. The author presumes that an accurate estimate
of the chronometric age of the big elk is 9800
C years BP
or a few years younger, given that many (but not all) of the
elk remains lay just above, and directly on, the dated stra-
The remains of the big elk do not represent an entire skel-
eton despite the excavation of a surface area of approxi-
mately 9 m 27 m. Within this area, many of the remains
were concentrated in two clusters about 4.5 m apart from
one another. One cluster was made up mostly of cervical
and thoracic vertebrae and ribs; the other consisted of hind-
limb elements. No remains of the skull, no teeth, and no
parts of the pelvic girdle were recovered (Table 1). The ver-
tebrae clearly articulate with one another in anatomical se-
quence (Fig. 2). Pathology (see later in the text) of the left
and right first ribs matches that on the first and second
thoracic vertebrae. The right tibia articulates with the lateral
malleolus, naviculo cuboid, one of the calcaneii, and the
right metatarsal. According to field notes, the right astraga-
lus also articulated with these elements, but it was appa-
rently sacrificed for a radiocarbon date; thus, whether or not
it articulates with the remainder of the ankle cannot be de-
termined (these remains are incorrectly described as left
elements in Gustafson and Wegener 2004, p. 313). The re-
maining right astragalus articulates with the other right cal-
caneum and represents a small individual relative to the
more complete skeleton referred to as the big elk in field
notes and publications. These small elk remains are insuffi-
ciently well preserved to be metrically compared with the
big elk and their state of preservation is a bit different than
that of the big elk. They are not considered in detail here,
except when they shed light on aspects of the big elk.
Gustafson and Wegener (2004, p. 312) indicate that the
local typical pattern of fragmentation of artiodactyl long
bones resulting from human butchering results in intact
joint ends where the shafts had been broken off directly
above and below the joint. Carpals and tarsals often are
complete and show no signs of cuts, scratches or other alter-
ations. . . . Shattered shaft fragments, often exhibiting spiral
fractures, and split or broken phalanges suggest further proc-
essing, perhaps for marrow extraction. It was because of
superficial resemblances of the Marmes big elk remains
with this fragmentation pattern that it was originally thought
the elk had been butchered by humans (Fryxell et al. 1968a;
Fryxell and Keel 1969). Gustafson and Wegener (2004, p.
314) later concluded that the big elk had perhaps not been
butchered by early human hunters because most of the ribs
are whole, rather than broken into several pieces and the
vertebrae are intact. Most bones of the hind limb are broken
near mid-shaft [and the] distal end of the right metatarsal
and a first phalanx has been pierced from both sides by car-
nivore canines. Apparently, differences in fragmentation
between the remains of the Marmes big elk and the remains
of ungulates clearly butchered by prehistoric people
prompted the new interpretation.
My examination of the big elk remains confirms that most
vertebrae and ribs are indeed intact and nearly complete,
though a few transverse processes and dorsal spines of ver-
tebrae are damaged or missing (see the following section,
Carnivore damage) as are some distal rib ends. All old frac-
ture surfaces on limb bones indicate fracture occurred when
Table 1. Inventory of Marmes Rockshelter big elk skeletal ele-
Axial skeleton Appendicular skeleton
C1C2, C4C7 L femur, proximal and distal ends
T1T11 R tibia, proximal and distal ends
10 R ribs (includes R1R6) L tibia, distal end
8 L ribs (includes R1R6) R and L astragali
Various rib fragments R lateral malleolus (distal fibula)
R calcaneii
R naviculo cuboid
R metatarsal (proximal half)
L metatarsal (distal half)
2 first phalanges, 1 second phalanx
Note: R, right; L, left
Lyman 1369
Published by NRC Research Press
the bones were fresh rather than dry (Johnson 1985; Lyman
1994). For example, the right and left tibiae both display
spiral fractures, as do the distal femur and left metatarsal.
Several ribs lack the proximal end and resemble archaeolog-
ical specimens apparently butchered by prehistoric humans
in the area (Lyman 1978, 1995), but the majority of ribs are
not broken in this manner (Fig. 2). The degree and nature of
fragmentation is similar to that displayed by large artiodac-
tyl carcasses that have been lightly scavenged by carnivores
(Haynes 1982). The carnivore-generated damage to the
bones also suggests that light scavenging modified the skel-
Carnivore damage
Of 31 skeletal specimens representing vertebrae and limb
bones, 23 display damage from gnawing carnivores (ribs not
included in tallies). The damage is typical of that inflicted
on artiodactyl carcasses consumed by North American Qua-
ternary carnivores such as wolves (Canis lupus), coyotes
(Canis latrans), bears (Ursus spp.), and cougars (Felis con-
color) (Binford 1981; Burgett 1990; Haynes 1980a, 1980b,
1982, 1983). In particular, dorsal and transverse spinous
processes of most vertebrae and the distal ends of several
ribs have been chewed, but none of them have been exten-
sively or intensively chewed. Several vertebral neural spines
display classic U-shaped ends diagnostic of carnivore gnaw-
ing, and the neural spine of one thoracic vertebra has several
classic punctures (Fig. 3). The proximal femur and the distal
femur both have punctures, the former around the neck on
both anterior and posterior surfaces just distal to the femur
head, and the latter on the medial surface of the medial pa-
tellar ridge. The anterior crest of the proximal right tibia has
been bitten off. The distal metatarsal has punctures and
tooth furrows on the anterior and posterior surfaces. One
first phalanx also has furrows, and a hole has been bitten
through the proximal end of its postero-ventral surface.
The right calcaneum of the big elk is complete and un-
damaged; a smaller right calcaneum of a different elk has
been heavily chewed by carnivores and lacks the entire
tuber calcis. The bones of the ankle that articulate with the
large calcaneum display no damage from gnawing carni-
vores. The small calcaneum articulates with a small astraga-
lus that is also gnawed. Given the high degree of gnawing
damage to the small calcaneum and astragalus (the bones
are approximately the size of a modern Rocky Mountain
elk) and the light degree of gnawing damage to all speci-
mens representing the big elk, the author suspects the former
represents a different depositional episode independent of
the deposition of the large elk. It is unclear whether the
gnawed small elk ankle was deposited later than or earlier
than the big elk carcass.
The low degree of carnivore-inflicted damage to the
bones of the big elk suggests that there were very few hun-
gry carnivores in the area, that there was little consumable
soft tissue on the bones and minimal attractive grease and
marrow in the bones when carnivores found the carcass, or
that the carcass was buried shortly after deposition. The for-
mer possibility seems unlikely given that several carnivores
are represented in the faunal remains from approximately
contemporaneous strata at the site. These carnivores are coy-
Fig. 2. Axial skeleton of the Marmes Rockshelter big elk. Ruler is 31 cm long; a light card marks the missing C3.
1370 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
ote (Canis latrans), noble marten (Martes nobilis (extinct)),
long-tailed weasel (Mustela frenata), probable mink (Mus-
tela cf. vison), badger (Taxidea taxus), arctic fox (Alopex la-
gopus), red fox (Vulpes fulva), and bobcat (Lynx rufus)
(Gustafson and Wegener 2004; Lyman, unpublished data,).
This list is substantial given that only about 20% of the col-
lection has been identified and it will likely grow longer
once all remains are studied. For example, it would not be
unexpected to find remains of wolf in the collection.
Whether or not wolf remains are found, it is well docu-
mented that the smaller coyote can damage by gnawing as
well as transport bones of elk (Burgett 1990).
Actualistic research indicates that during warm months in
temperate latitudes ungulate carcasses the size of elk may be
devoid of soft tissue sought by scavengers within four to six
weeks of death (Miller 1975; Weigelt 1989). That the bones
of the Marmes big elk had little consumable tissue of inter-
est to scavenging carnviores is possible, but this cannot be
determined with any confidence. The author suspects the an-
imal died in late winter, froze shortly after death and light
scavenging, and was subsequently buried by the fluvially de-
posited silts in which its remains were found. There are sev-
eral lines of evidence that make this scenario probable. First,
animals that are scavenged by wolves in winter tend not to
be completely consumed but instead display levels of inten-
sity and extent of gnawing damage and types and frequen-
cies of surviving skeletal parts much like that displayed by
the Marmes big elk (Haynes 1980a, 1980b, 1982). Some
limb elements of the big elk, especially proximal elements,
may have been carried away by scavengers to secluded areas
where they would not be bothered by competitors (Burgett
1990). Second, the Marmes elk was a skeletally mature ani-
mal that was not at the peak of skeletal health (see the fol-
lowing section, Pathology), which may have placed the
individual at a level of stress greater than that experienced
by a healthy individual, especially if the winter was more
severe than normal. When it died, the animal was in a rela-
tively warm part of the region; the Snake River canyon bot-
tom tends to be 58 8C warmer than the 250300 m higher
surrounding plateau areas throughout the year, making the
canyon bottom unsavory in summer but a microclimatic ha-
ven in winter. Warmer temperatures translate to less meta-
bolic cost of maintaining body temperature and more non-
snow-covered forage. The elk was also near water in the
Palouse River, a topographic feature often sought by win-
ter-stressed male artiodactyls (Clutton-Brock et al. 1982).
As noted later in the text, the large size of the big elk re-
mains suggests that the individual was a male. Third, given
that some elk migrate seasonally, occupying topographically
lower (warmer, snow-free) areas in winter, it is not unrea-
sonable to suppose that the Marmes elk died where it did in
the winter. Cervid winter mortality tends to be greater
among males than females (Coughenour and Singer 1996;
Wisdom and Cook 2000). Males enter the winter with less
fat reserves than females, having drained those resources
during the rut. Males thus move to lower elevations where
forage is likely to be more accessible, yet they tend to have
higher winterkill mortality than females. Deeper snow re-
sults in higher mortality regardless of other variables (Cook
2002). If the animal died during a cold winter, its tissues
would have frozen and, though accessible to scavengers
(Guthrie 1990; Haynes 1982), bones would have been
gnawed only with difficulty. If the animal died in late winter
or early spring, its consumable tissues likely became quickly
inaccessible thereby precluding extensive and intensive
With the spring thaw, the Palouse River overflowed its
bank and quickly covered much of the carcass with the fine
silts in which the bones lay, leaving little exposed to attract
scavengers. The orientation of the long axis of 15 ribs and
seven thoracic vertebrae, plotted in a rose diagram (Fig. 4)
using 308 bins because of the small sample size, is sugges-
tive of a preferred orientation that implicates fluvial action,
but the distribution of specimens across the orientation
categories is not statistically different from a random pattern
= 4.68, p > 0.25). If some of the carcass was not imme-
diately buried, loss of exposed soft tissue would have been
rapid, much like the loss of soft tissue from carcasses of do-
mestic cattle (Miller 1975) and elk on the landscape today
(Lyman 1989). Exposed bones would have dried out quickly
and begun to weather; most specimens display weathering
stage 1 or 2, indicating minimal exposure duration (Behren-
smeyer 1978; Lyman and Fox 1989). One rib displays dam-
age attributable to rodent gnawing and a second rib displays
Fig. 3. Gnawing damage on the dorsal spines of T5 (on left, with puncture) and T3 (on right).
Lyman 1371
Published by NRC Research Press
damage that likely is the result of rodent gnawing. Rodents
do not gnaw greasy bone, but instead gnaw dry bone
(Lyman 1994), something relatively unattractive to scaveng-
ing carnivores because of the lack of nutritional value. Min-
imal rodent gnawing damage, plus minimal carnivore
damage and minimal weathering damage, indicates the elk
remains were quickly buried by spring runoff and the depo-
sition of overbank silts. The pathologically fused C4 and C5
(see later in the text) were collected with a block of sedi-
ment, indicating they were at least partially covered by over-
bank silts.
Taphonomic history
There is no evidence of human involvement with the car-
cass despite the apparent occupation of Marmes Rockshelter
by humans at the time the elk carcass was deposited (Hicks
2004). There are no butchering marks in the form of ham-
merstone scars on long bones meant to gain access to mar-
row, and there are no cutmarks or striae made by stone tools
during skinning, dismembering, and filleting (Blumenschine
et al. 1996). The femur, tibiae, and metatarsal were likely all
fractured by wolves or perhaps ursids. The inventory of
skeletal parts and gnawing damage indicates that the big elk
likely died of natural causes and was subsequently lightly
scavenged. The low degree of carnivore damage in conjunc-
tion with rodent gnawing suggests a winter death and early
burial from overbank sediments deposited by spring melt-
water. For any of several reasons, the skull and many appen-
dicular elements simply were not recovered by the
excavation and may remain buried in nearby unexcavated
All long bone epiphyses of the big elk are fully fused, in-
dicating that a skeletally mature individual is represented.
Features of the vertebral column and ribs indicate that the
big elk suffered two afflictions. First, C4 and C5 are fused
together and immobile relative to one another; the ventral
surfaces of the centra display an abundance of abnormal
bone tissue that effectively fuses the two vertebrae together
(Fig. 5). Probing with a dental pick suggests the centra
themselves and the zygapophyses are not fused together,
and though this could not be established with certainty, the
apparent lack of fusion suggests the affliction is not spondy-
loarthropathy (Rothschild and Martin 1993). The anterior
transverse foramena of C5, especially the one on the left
side, are partially blocked by extra bone tissue. C6 displays
no unusual features; C3 was not recovered. The manner and
location in which C4 and C5 are fused and the flowing
(Olivieri et al. 2009) appearance of the extra bone tissue in-
dicates diffuse idiopathic skeletal hyperostosis (DISH)
(Rothschild and Martin 1993). The extra bone tissue seems
to be the result of ligamentous ossification (Olivieri et al.
2009). DISH-like osteological features bony thickening
of the ventral longitudinal ligament on two thoracic verte-
brae (Greer et al. 1977, p. 41) have been reported
among modern elk.
The second affliction is perhaps mechanically related to
the first. The first and second thoracic vertebrae have an
abundance of extra bone tissue around and ventral to the
centra and also on and around the rib demi facets (Fig. 6).
These osteophytes a primary criterion of osteoarthritis in
human knees and hips (Rothschild 1997) are only around
the vertebral centra and not on the zygapophyses, indicating
that although the pathology is similar to osteoarthritis, the
affliction is best referred to spondylosis deformans (Roths-
child and Martin 1993, p. 83) or spinal osteoarthritis (Ortner
2003, p. 549). There is also an abundance of osteophyte tis-
sue around the heads of the left and right first ribs (Fig. 7).
As well, C7 shows osteophytes around the postero-ventral
portion of the centrum, but not to the extent of T1 and T2.
The centra of T1 and T2 display unusual porosity, an un-
likely sign of osteoarthritis (Rothschild 1997). There is no
clear evidence of eburnation, a marker of arthritis severity
(Rothschild 1997). Finally, there are shallow dorso-ventral
oriented erosive grooves in the conjoining centra of T1 and
T2, signifying total loss of joint cartilage, a sign of osteoar-
thritis (Rothschild and Martin 1993).
Osteoarthritis is relatively uncommon in modern elk and
tends to mostly affect limb joints (Thorne et al. 2002). It oc-
curs in modern elk where populations are dense and unsani-
tary conditions occur on feed grounds where several bacteria
have been identified as causal and can be introduced
through wounds in feet (Thorne et al. 2002). There is no
evidence of antemortem injury among any of the remains of
the Marmes big elk. Osteoarthritis can result secondarily
from brucellosis (Brucella abortus) infection among modern
elk (Thorne et al. 2002). It is unlikely that unsanitary condi-
tions were the cause of the spinal osteoarthritis in the
Marmes big elk. Age is often strongly correlated with the
onset and severity of osteoarthritis in humans (Weiss and Ju-
rmain 2007). None of the other thoracic vertebrae (T3T11)
of the Marmes big elk displays any pathological features.
The spinal osteoarthritis of C7T2 is likely related to the
DISH of C4C5 since the two conditions commonly are
found together in humans (Ortner 2003, p. 549). Spinal os-
teoarthristis occurs among non-elk cervid taxa and increases
in frequency and intensity with increased age of individual
animals (Peterson 1988; Peterson et al. 1982; Wobeser and
Runge 1975). It also occurs with relatively higher frequency
Fig. 4. Rose diagram of the orientation of the long axis of 15 ribs
and seven thoracic vertebrae in 308 bins. The pattern is not statisti-
cally significantly different than a random pattern of orientation (c
4.68, p > 0.25). The scale denotes the tally of specimens per bin.
1372 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
among South American camelids that were used as beasts of
burden than among those not used as pack animals
(deFrance 2010). Some evidence suggests that weight bur-
dens might exacerbate osteoarthritis symptoms (Weiss and
Jurmain 2007). If the Marmes big elk was in fact a male,
perhaps large antlers placed load-bearing stress on the neck
and anterior-most back, thereby exacerbating susceptibility
to degenerative joint disease (Sokoloff 1969) of the spine
such as osteoarthritis (see also Wobeser and Runge 1975).
The skulls of male elk are disproportionately larger than
those of females, presumably because of the necessity of
bearing the weight of antlers (Schonewald 1994). The DISH
could have also resulted from heavy antlers contributing to
ossification of neck ligaments. Whatever the cause, addi-
tional documentation of prehistoric occurrences of arthritis
and other pathologies among North American elk may even-
tually reveal much about disease history in this large cervid.
As noted earlier in the text, the elk remains from the
floodplain in front of Marmes Rockshelter were originally
reported to be larger than modern elk in the Pacific North-
west of North America (Fryxell et al. 1968a). This observa-
tion became part of the received wisdom regarding the
PleistoceneHolocene transition-era mammalian fauna of
the Pacific Northwest, and it was reported in the literature
many times that local late Pleistocene elk were large relative
to modern elk (Chatters 2001; Kirk and Daugherty 2007;
Leonhardy and Rice 1970; Lyman 2004a). Descriptive data
on the size of the bones of the Marmes big elk have, how-
ever, never been reported, so comparison with size data for
modern elk or other prehistoric elk has been impossible. The
author measured several dimensions of the skeletal elements
of the Marmes big elk (von den Driesch 1976). The same
dimensions were measured in samples of the two subspecies
of elk that today are found in Washington State. The Rocky
Mountain elk (C. e. nelsoni) occurs in eastern Washington,
and is said to be smaller than the Roosevelt elk (C. e. roose-
velti) that is found today in western Washington (Bryant and
Maser 1982; OGara 2002). This size differences between
the two subspecies was observed among the modern skele-
tons that were measured (Lyman 2006).
The Marmes big elk is indeed larger in several dimen-
sions than modern Rocky Mountain elk, the subspecies that
occurs in the eastern Washington today (Figs. 8, 9). It is a
bit larger than the modern size range of Roosevelt elk in
three dimensions (astragalus distal width, distal metatarsal
width of condyles, distal metatarsal condyle posterior
breadth), and in three other dimensions the Marmes big elk
is within the range of modern Roosevelt elk, but on the
large end of that range (astragalus lateral length, distal meta-
podial condyle antero-posterior diameter, proximal first pha-
lanx width). This should not, however, be taken to mean that
the Marmes big elk was of the Roosevelt subspecies because
it has not been robustly established that size of bones can be
used to distinguish prehistoric remains of the two taxa
(Lyman 2006).
Murie (1951, p.17) indicated the Pleistocene elk of Alaska
were larger than their modern conspecifics, and on that basis
Geist (1998, p. 214) stated that North American elk have
declined in body size postglacially. Neither Guthrie (1966)
nor Kurten and Anderson (1980) comment on the size of
late Pleistocene elk relative to modern elk, although post-
cranial remains of late Pleistocene elk from the Yukon are
reported to be larger than modern specimens (Harington
1977, in OGara and Dundas 2002, p. 82). Some records of
late Pleistocene to earliest Holocene elk provide mixed indi-
cations of the size of these animals. Burns (1986) reports
that a 9500
C year-old elk from northern Alberta was of
Fig. 5. Fused C4 and C5 vertebrae; ventral view. Anterior is to the right.
Lyman 1373
Published by NRC Research Press
comparable size to modern individuals, but provides no
metric data. Elk remains from the Dry Creek site in south-
eastern Alaska are 900010 000
C years old, and are said
to represent individuals of larger than modern body size;
available metric data are from teeth, precluding direct com-
parison with the Marmes big elk (Guthrie 1983; Hoffecker
et al. 1996). Shackleton and Hills (1977) report that elk re-
mains what will be termed the Three Hills elk from
central Alberta dating to about 9650
C years BP are
slightly larger in some dimensions than the bones of modern
elk. Data they present bear this out; a comparison of their
data with those for dimensions measured indicates that the
Three Hills elk is indeed larger than an average Rocky
Mountain elk and is about average in size compared with a
Roosevelt elk (Fig. 9). Interestingly, the Marmes big elk is
larger in most dimensions than the similarly aged Three
Hills elk (Fig. 10).
Unpublished laboratory notes indicate that the Marmes
big elk was thought to be exceptionally large because its
bones are larger than a modern bull elk collected in the
1960s from the National Bison Range in northwestern Mon-
tana. The modern elk was said by rangers at the National
Bison Range to be the largest they had seen (Gustafson
1983). The skeleton of this individual is curated in Washing-
ton State University (WSU) Connor Zoological Museum
under accession number 65-68. Laboratorynotes list the size
of several bone dimensions for both the Marmes big elk and
WSU 65-68, all of which were likely recorded by C. E. Gus-
Fig. 6. (A) Anterior view of T1 (left) and T2 (right). (B) posterior view of T1 (left) and T2 (right).
1374 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
Fig. 7. Anterior view of right (top) and left (bottom) first ribs.
Fig. 8. Metrics of four dimensions of Marmes big elk remains compared with modern Rocky Mountain elk (C. e. nelsoni) and modern
Roosevelt elk (C. e. roosevelti) bones.
Lyman 1375
Published by NRC Research Press
Fig. 9. Metrics of selected dimensions of early Holocene Three Hills Alberta elk (from Shackleton and Hills 1977) compared with mod-
ern Rocky Mountain elk (C. e. nelsoni) and modern Roosevelt elk (C. e. roosevelti) bones.
Fig. 10. Ratio diagram comparing the size of several dimensions of the early Holocene Three Hills Alberta elk (from Shackleton and Hills
1977) with the Marmes big elk and the modern WSU 65-68 elk. Points to the right of the Three Hills vertical line represent larger di-
mensions than the Three Hills elk.
1376 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
tafson in 1968 or 1969. Those notes indicate that to deter-
mine the percent difference in size, Gustafson calculated the
difference between the size of dimensions of bones of WSU
65-68 and dimensions of the bones of the Marmes big elk,
and divided those differences by the size of the particular
dimension in WSU 65-68. But whether WSU 65-68 is ex-
ceptionally large among elk, or just a large elk was not
documented. Data the author has collected indicate that
WSU 65-68 is indeed large, and it appears to be larger than
the Three Hills, Alberta, elk (Fig. 10). Importantly, and as
Gustafson observed, the Marmes big elk is considerably lar-
ger than WSU 65-68 (Fig. 10).
Several species of late Pleistocene mammal from North
America consist of animals that were larger than their mod-
ern conspecifics (Edwards 1967; Kurten and Anderson
1980). Such gigantism has been well documented for bison
(Bison spp.) (Hill et al. 2008), and though some individuals
are larger than modern conspecifics, available evidence is
less clear for late Pleistocene bighorn sheep (Ovis canaden-
sis) (Lawler 1996; Wang 1988) and pronghorn (Antilocapra
americana) (Adams et al. 1999; Chorn et al. 1988). The au-
thor knows of no evidence that deer (Odocoileus virginianus
and O. hemionus) were larger during the terminal Pleisto-
cene than today, but there is evidence that middle Holocene
white-tailed deer in the midwestern U.S. were smaller than
modern deer, apparently as a result of differences in avail-
ability of high-quality forage (Purdue 1989, 1991). Given
the reports cited earlier in this section, there is little evi-
dence that all terminal Pleistocene elk were larger than their
modern conspecifics. Is the Marmes Rockshelter big elk
an anomaly, a singular exceptionally large individual that is
not of more or less average size?
The author knows of but a single elk bone from eastern
Washington that is similar in age to the Marmes big elk.
This bone came from the archaeological site known as Sen-
tinel Gap (45KT1362), located in southcentral Washington
130 km west of Marmes Rockshelter (Fig. 1). Five radiocar-
bon dates average about 10 200
C years BP (Galm and
Gough 2001, 2008; Gough and Galm 2003). This small site
(*82 m
) is in a small drainage west of the Columbia River
and represents a single, likely short-term occupation. The
author identified all mammal remains recovered from Senti-
nel Gap, including bison (Bison spp.), elk, mountain sheep
(Ovis canadensis), and badger (Taxidea taxus). A single left
metatarsal of an elk, missing the distal condyles apparently
as a result of carnivore gnawing, was sufficiently complete
to provide size data. The author measured the lateral-medial
width and anterior-posterior depth of the proximal end of the
Sentinel Gap specimen, and also that of the right metatarsal
of the Marmes big elk (anterior-posterior depth had to be es-
timated because of fragmentation). Relative to the Three
Hills elk, two modern Rocky Mountain elk reported by
Shackleton and Hills (1977) for which only lateral-medial
width data are available, six modern Rocky Mountain elk
from Wyoming, Washington, and British Columbia, and
three modern Roosevelt elk from British Columbia that the
author measured, the Marmes big elk and the Sentinel Gap
elk are indeed large (Fig. 11).
It appears, in light of the Sentinel Gap elk, that the
Marmes elk is not an exceptionally large individual given
its calendric age. In light of the evidence for other North
American ungulates bison in eastern Washington State
(Lyman 2004b) follow the pattern of temporal diminution
apparent in the Great Plains of North America (Hill et al.
2008) it seems reasonable to hypothesize that elk of ter-
minal Pleistocene age were indeed larger than their modern
conspecifics, and that this species will display the general
Holocene-diminution pattern once remains of appropriate
age are measured. Why might this hypothesis turn out to be
A probable cause of Holocene diminution involves cli-
matically driven decreases in quality and quantity of nutri-
tious forage (Guthrie 1984a, 1984b; Hill et al. 2008; Purdue
1989, 1991). The argument can be summarized as follows.
Adult body size (of ungulates in this case) is a function of
food availability because the latter influences ontogenetic
growth rate (Geist 1987a, 1987b, 1998; McNab 2010).
Thus, net primary productivity (amount of plant biomass
produced per unit of time and area) in general, and espe-
cially primary productivity that is ecologically relevant or
available during the ungulate growing season (Huston and
Wolverton 2009), influences final adult body size (Festa-
Bianchet et al. 2004; Klein 1964; Wolverton et al. 2009), in-
cluding the skeleton and excluding seasonal fluctuation in
body mass. Climate, of course, influences plant productivity
(Blackburn and Hawkins 2004; Huston and Wolverton
2009). The critical question with respect to the Marmes
big elk therefore iswas ecologically relevant primary
productivity greater 9800
C years ago in southeastern
Washington than it is today? Fortunately, data suggesting
an answer to this question are available.
Palynological and other evidence indicates that grasslands
spread and forests retreated upslope as the continental ice
sheet shrank (Blinnikov et al. 2002; Takeuchi et al. 2009).
Palynological data from two lakes in eastern Washington
Fig. 11. Bivariate scatterplot of lateral-medial width and anterior-
posterior depth of proximal metatarsals of the Marmes big elk, the
Sentinel Gap elk, the Three Hills prehistoric elk (3 Hills), two
modern Rocky Mountain elk (SH1, SH2) reported by Shackleton
and Hills (1977), and six modern Rocky Mountain elk (RKM) and
three modern Roosevelt elk (ROS) measured by Lyman. Anterior-
posterior depth is unknown for 3 Hills, SH1, and SH2.
Lyman 1377
Published by NRC Research Press
are of particular relevance here. Wildcat Lake is 13 km
north of Marmes Rockshelter; Williams Lake Fen is
100 km north-northeast of Marmes Rockshelter (Fig. 1).
Data from these lakes indicate that prior to about 10 200
C BP, conifer pollen dominated over grass pollen. Relative
abundances reverse abruptly after that time; then grass pol-
len makes up as much as 60% of the total pollen (Johnson
et al. 1994; Mehringer 1996). This change in pollen rain at
Williams Lake Fen has been said to be perhaps the most
striking change recorded in this regions history of postgla-
cial vegetation (Mehringer 1996, p. 31). Climate is drying
out and warming up, suggesting that at the time the Marmes
big elk and the Sentinel Gap elk were deposited, ecologi-
cally relevant primary productivity in the form of grass was
relatively high. Not only was there more forage available, it
would have been available for a longer period of time each
year as the glacial anticyclone weakened and the growing
season increased in duration (Takeuchi et al. 2009). Given
that elk tend to graze more than browse when habitats allow
(Cook 2002), it appears that the Marmes and Sentinel Gap
elk were big because they had access to abundant and nutri-
tious forage in the form of grass. Continued warming and
drying would eventually see reduced grasslands as sagebrush
increased in abundance (Johnson et al. 1994; Mehringer
1996). Local elk likely became smaller for want of nutrition;
bison seem to also have shrunk in size, plus they seem to
have been unable to survive in the Columbia Basin of east-
ern Washington in any number during the warm dry middle
Holocene (Lyman 2004b).
Two other possible causes for large late Pleistocene elk
should be considered. First, Geist (1987b, 1998, 1999) pro-
posed that individual ungulates in colonizing populations
would grow to large sizes as a result of minimal intra-
specific competition and relatively nutritiously rich habitats.
Elk were present south of the Pleistocene continental ice
sheets at least 40 000 years ago but do not seem to have be-
come abundant there until the terminal Pleistocene (Graham
and Lundelius 1994; Guthrie 2006; OGara and Dundas
2002). It has been suggested that the similarities in morphol-
ogy, behavior, and genetics of modern North American elk
and related Asian subspecies suggest the U.S. were colon-
ized mostly by elk from more northern latitudes (OGara
and Dundas 2002). If elk were indeed rare in low latitudes
prior to the end of the Pleistocene, they likely would have
been larger bodied; males would have had larger antlers at
younger ages, and females would have reproduced at
younger ages, much like recent colonizing elk populations
(McCorquodale et al. 1988, 1989). Given the possibility
that the record of radiocarbon-dated elk remains may be a
function of specifically choosing what seem to be terminal
Pleistocene and archaeological elk remains (Guthrie 2006;
Meltzer and Mead 1983), whether or not the colonization
process accounts for the size of the Marmes (and other) big
elk remains to be robustly demonstrated. And even if coloni-
zation of low latitudes is determined to be responsible for
the large size of terminal Pleistocene elk, they were large
because of an abundant food supply.
The second possible cause for late Pleistocene elk being
large relative to their modern congeners and conspecifics is
Bergmanns rule; that is, for reasons of thermoregulation
(Millien et al. 2006). However, mounting evidence indicates
that the correlation between body size and ambient temper-
ature (typically measured as latitude) is imperfect and not
universal, prompting the suggestion that body-size trends
that seem to be the result of Bergmanns rule may in fact
be the result of other environmental variables, either singly
or in combination (for a sampling of pertinent literature, see
Allen et al. 2006; Blackburn et al. 1999; McNab 2010;
Meiri et al. 2007; Millien et al. 2006; Watt et al. 2010). It
is also becoming clear that populations seem to respond in-
dividualistically rather than taxonomically (Millien et al.
2006). Robust determination of the specific environmental
variable(s) responsible for a particular, apparently Bergman-
nian response within species will require fine-resolution data
for chronometric age of multiple specimens representing a
temporal series, for paleoecological settings and changes
therein, and numerous morphometric data points (Hill et al.
2008). These same data will provide an indication of the
contribution of colonization variables to the large size of ter-
minal Pleistocene elk (Geist 1987a).
Modern North American elk exhibit geographic size
clines in both body mass and skeletal dimensions (Schone-
wald 1994). Further, the closely related European red deer
(Cervus elaphus) is sufficiently phenotypically plastic that
improved forage conditions can result in larger skeletons
within few generations (Geist 1998; Langvatn and Albon
1986; Post et al. 1999). Unfortunately, we have no data
points from 10 000 to 15 000
C years BP, or from 9500
to 500
C years BP from eastern Washington to plot against
the Marmes big elk. Until such time as the steep data re-
quirements identified above are met, the author prefers the
hypothesis of abundant, highly nutritious forage that was
available for long portions of each year.
Discussion and conclusions
When first discovered, the Marmes big elk seemed to
confirm the then typical notion that the early Americans rep-
resented by human remains from the site were big-game
hunters. Perhaps more remarkable at the time was the size
of the bones that indicated the represented animal had been
quite large by modern standards. It is now clear there is no
evidence the Marmes big elk was preyed upon or even scav-
enged by early American hunters. Rather, the most parsimo-
nious explanation is that the individual likely represents a
winter-killed animal that underwent minimal scavenging by
carnivores before being buried by silt carried by spring
floodwaters of the Palouse River.
Although the possible implication of the Marmes big elk
as signifying early big-game hunters has faded from local
archaeological awareness with time, the observation that
this elk was big has not. Thus archaeologist James Chatters
(2001, p. 150) recently wrote that about 9500 years ago
oversized elk still roamed the grasslands of the Columbia
Basin. Not only were there no metric data at the time to
back up statements of this sort, the paleoecological signifi-
cance of what might be larger than modern elk has not pre-
viously been considered. The data on ancient elk bone size
available in the literature, in conjunction with data on the
size of modern elk bones summarized here, indicate that in-
deed the Marmes elk was quite large relative to modern in-
dividuals in the same area. And the Marmes elk was not
1378 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
alone; at least one other elk of similar age that roamed east-
ern Washington, the Sentinel Gap elk, was also considerably
larger than modern elk and apparently the same size as the
Marmes big elk. Thus the Marmes big elk is not an anom-
aly. The paleoecological significance of these terminal Pleis-
tocene elk is that they align with the greater abundance of
grass as forage in the soon to be arid and shrub-steppe cov-
ered Columbia Basin indicated by the paleobotanical record.
The terminal Pleistocene relative abundance of grass likely
also contributed to the presence of larger than modern indi-
vidual bison in eastern Washington (Lyman 2004b).
Finally, although skeletally large, the Marmes big elk was
not skeletally healthy when it died. It suffered from two,
perhaps related, afflictions. The skeletal disease status of
late Pleistocene ungulates in North America is a seldom re-
ported phenomenon (see Rothschild and Martin (2003) for
an exception). The Marmes elk, afflicted with what likely
represent diffuse idiopathic skeletal hyperostosis and spinal
osteoarthritis, provides an initial glimpse of such among
this large member of the deer family. For that alone the
skeleton holds value as a data point in studies of disease his-
Study of the Marmes faunal remains was courtesy of the
U.S. Army Corps of Engineers, Walla Walla District and
Washington State University Museum of Anthropology,
Pullman, Washington State. Dan Temple assisted with diag-
nosis of the pathologies. This research was funded in part by
National Science Foundation grant BCS-0912851. Mary
Collins and Diane Curewitz, both of the Washington State
University Museum of Anthropology, facilitated my access
to the Marmes Rockshelter faunal collection. Stan Gough
provided the Sentinel Gap elk metatarsal on short notice.
Becky Saleeby and Greg Dixon of the U.S. Department of
Interior National Park Service provided a copy of the Dry
Creek faunal report. Access to the modern elk skeletons cu-
rated by the Royal British Columbia Museum was provided
by L. Kennes and N. Panter. Helpful comments on an early
draft were provided by Tony Barnosky, Timothy Fisher, R.
White, and Steve Wolverton.
Adams, R., Davis, D.P., and Miller, M.E. 1999. Osteometric com-
parisons of early Plains Archaic, protohistoric, and modern
pronghorn: acase for Holocene dwarfing. In The Trappers Point
site (48SU1006): Early Archaic adaptations in the Upper Green
River basin, Wyoming. Edited by M.E. Miller, P.H. Sanders, and
J.E. Francis. Office of the Wyoming State Archaeologist, Lara-
mie, Wyoming, Cultural Resources Series No. 1, pp. 279289.
Allen, C.R., Garmestani, A.S., Havlicek, T.D., Marquet, P.A., Pe-
terson, G.D., Restrepo, C., et al. 2006. Patterns in body mass
distributions: sifting among alternative hypotheses. Ecology Let-
ters, 9: 630643. doi:10.1111/j.1461-0248.2006.00902.x. PMID:
Anonymous. 1968a. Date for New World man. Science News, 93:
Anonymous. 1968b. The man they ate for dinner. Time, 91(19): 98.
Behrensmeyer, A.K. 1978. Taphonomic and ecologic information
from bone weathering. Paleobiology, 4: 150162.
Binford, L.R. 1981. Bones: ancient men and modern myths. Aca-
demic Press, New York, N.Y.
Blackburn, T.M., and Hawkins, B.A. 2004. Bergmanns rule and
the mammal fauna of northern North America. Ecology, 27:
Blackburn, T.M., Gaston, K.J., and Loder, N. 1999. Geographic
gradients in body size: a clarification of Bergmanns rule. Diver-
sity and Distributions, 5: 165174. doi:10.1046/j.1472-4642.
Blinnikov, M., Busacca, A., and Whitlock, C. 2002. Reconstruction
of the Late Pleistocene grassland of the Columbia Basin, Wa-
shington, USA, based on phytolith records in loess. Palaeogeo-
graphy, Palaeoclimatology, Palaeoecology, 177: 77101. doi:10.
Blumenschine, R.J., Marean, C.W., and Capaldo, S.D. 1996. Blind
tests of inter-analyst correspondence and accuracy in the identi-
fication of cut marks, percussion marks, and carnivore tooth
marks on bone surfaces. Journal of Archaeological Science, 23:
493507. doi:10.1006/jasc.1996.0047.
Bryant, L.D., and Maser, C. 1982. Classification and distribution.
In Elk of North America: ecology and management. Edited by
J.W. Thomas and D.E. Toweill. Stackpole Books, Harrisburg,
Pa., pp. 159.
Burgett, G.R. 1990. The bones of the beast: resolving questions of
faunal assemblage formation processes through actualistic re-
search. Ph.D. dissertation, Department of Anthropology, Univer-
sity of New Mexico, Albuquerque, N.M.
Burns, J.A. 1986. A 9000-year old wapiti (Cervus elaphus) skele-
ton from northern Alberta, and its implications for the early Ho-
locene environment. Geographie Physique et Quaternaire, 40:
Chatters, J.C. 2001. Ancient Encounters: Kennewick Man and the
first Americans. Simon and Schuster, New York, N.Y.
Chorn, J., Frase, B.A., and Frailey, C.D. 1988. Late Pleistocene
pronghorn, Antilocapra americana, from Natural Trap Cave,
Wyoming. Transactions of the Nebraska Academy of Sciences,
16: 127139.
Clutton-Brock, T.H., Guinness, F.E., and Albon, S.D. 1982. Red
deer: behavior and ecology of two sexes. University of Chicago
Press, Chicago, Ill.
Cook, J.G. 2002. Nutrition and food. In North American elk: ecol-
ogy and management. Edited by D.E. Toweill and J.W. Thomas.
Smithsonian Institution Press, Washington, DC, pp. 259349.
Coughenour, M.B., and Singer, F.J. 1996. Elk population processes
in Yellowstone National Park under the policy of natural regula-
tion. Ecological Applications, 6: 573593. doi:10.2307/2269393.
deFrance, S.D. 2010. Paleopathology and health of native and in-
troduced animals on southern Peruvian and Bolivian Spanish co-
lonial sites. International Journal of Osteoarchaeology, 20: 508
Edwards, W.E. 1967. The late-Pleistocene extinction and diminution
in size of many mammalian species. In Pleistocene extinctions:
the search for a cause. Edited by P.S. Martin and H.E. Wright,
Jr. Yale University Press, New Haven, Conn., pp. 141154.
Festa-Bianchet, M., Coltman, D.W., Turelli, L., and Jorgenson, J.T.
2004. Relative allocation to horn and body growth in bighorn
rams varies with resource availability. Behavioral Ecology, 15:
Fryxell, R., and Keel, B.C. 1969. Emergency salvage excavations
for the recovery of early human remains and related scientific
materials from the Marmes Rockshelter archaeological site,
southeastern Washington. Report to the U.S. Army Corps of En-
gineers, Walla Walla District.Washington State University De-
partment of Anthropology, Pullman, Wash.
Lyman 1379
Published by NRC Research Press
Fryxell, R., Bielicki, T., Daugherty, R.D., Gustafson, C.E., Irwin,
H.T., and Keel, B.C. 1968a. A human skelton from sediments
of mid-Pinedale age in southeastern Washington. American
Antiquity, 33: 511515. doi:10.2307/278610.
Fryxell, R., Bielicki, T., Daugherty, R., Gustafson, C., Irwin, H.,
Keel, B., and Krantz, G. 1968b. Human skeletal material and ar-
tifacts from sediments of Pinedale (Wisconsin) glacial age in
southeastern Washington, United States. Proceedings of the 8th
International Congress of Anthropology and Ethnological
Sciences, September 1968. Tokyo. Vol. 3, Ethnology and Ar-
chaeology, pp. 176181.
Galm, J.R., and Gough, S. 2001. Site 45KT1362, a c. 10,000 yr
B.P. occupation in central Washington. Current Research in the
Pleistocene, 17: 2931.
Galm, J.R., and Gough, S. 2008. The projectile point/knife sample
from the Sentinel Gap site. In Projectile point sequences in
northwestern North America. Edited by R.L. Carlson and
M.P.R. Magne. Simon Fraser University Department of Archae-
ology, Burnaby, B.C., Publication No. 35, pp. 209220.
Geist, V. 1987a. Bergmanns rule is invalid. Canadian Journal of
Zoology, 65: 10351038. doi:10.1139/z87-164.
Geist, V. 1987b. On speciation in Ice Age mammals, with special
reference to cervids and caprids. Canadian Journal of Zoology,
65: 10671084. doi:10.1139/z87-171.
Geist, V. 1998. Deer of the world: their evolution, behavior, and
ecology. Stackpole Books, Mechanicsburg, Pa.
Geist, V. 1999. Periglacial ecology, large mammals, and their sig-
nificance to human biology. In Ice Age peoples of North Amer-
ica: environments, origins, and adaptations of the First
Americans. Edited by R. Bonnichsen and K.L. Turnmire. Ore-
gon State University Press, Corvallis, Ore., pp. 7894.
Gough, S., and Galm, J.R. 2003. Bone technology at the Sentinel
Gap site. Current Research in the Pleistocene, 19: 2729.
Graham, R. W., and Lundelius, E. L., Jr. 1994. Faunmap: a data-
base documenting late Quaternary distributions of mammal spe-
cies in the United States. Illinois State Museum Scientific
Papers, Vol. 25, Springfield, Ill.
Greer, M., Greer, J.K., and Gillingham, J. 1977. Osteoarthritis in
selected wild animals. Proceedings of the Oklahoma Academy
of Science, 57: 3943.
Grosso, G.H. 1967. Cave life on the Palouse. Natural History,
67(2): 3843.
Gustafson, C.E. 1983. Wapiti populations in and adjacent to Mount
Rainier National Park: archaeological and ethnographic evi-
dence. Unpublished report on file with Mount Rainier National
Park, Ashford, Wash.
Gustafson, C.E., and Wegener, R.M. 2004. Faunal remains. In
Marmes Rockshelter: a final report on 11,000 years of cultural
use. Edited by B.A. Hicks. Washington State University Press,
Pullman, Wash., pp. 253317.
Guthrie, R.D. 1966. The extinct wapiti of Alaska and Yukon Terri-
tory. Canadian Journal of Zoology, 44: 4757. doi:10.1139/z66-
Guthrie, R.D. 1983. Paleoecology of the site and its implications
for early hunters. In Dry Creek: archaeology and paleoecology
of a late Pleistocene Alaskan hunting camp. Edited by W.R.
Powers, R.D. Guthrie and J.F. Hoffecker. Report on file with
the National Park Service, Washington, D.C., pp. 209287.
Guthrie, R.D. 1984a. Alaskan megabucks, megabulls, and mega-
grams: the issue of Pleistocene gigantism. In Contributions in
Quaternary vertebrate paleontology: a volume in memorial to
John E. Guilday. Edited by H.H. Genoways and M.R. Dawson.
Carnegie Museum of Natural History, Pittsburgh, Pa., Special
Publication No. 8, pp. 482510.
Guthrie, R.D. 1984b. Mosaics, allelochemics, and nutrients: an eco-
logical theory of late Pleistocene megafaunal extinctions. In
Quaternary extinctions: a prehistoric revolution. Edited by P. S.
Martin and R. G. Klein. University of Arizona Press, Tucson,
Ariz., pp. 259298.
Guthrie, R.D. 1990. Frozen fauna of the mammoth steppe: the story
of Blue Babe. University of Chicago Press, Chicago, Ill.
Guthrie, R.D. 2006. New carbon dates link climatic change with
human colonization and Pleistocene extinctions. Nature, 441:
207209. doi:10.1038/nature04604. PMID:16688174.
Harington, C.R. 1977. Pleistocene mammals of the Yukon Terri-
tory. Unpublished Ph.D. dissertation, University of Alberta, Ed-
monton, Alta.
Haynes, G. 1980a. Evidence of carnivore gnawing on Pleistocene
and Recent mammalian bones. Paleobiology, 6: 341351.
Haynes, G. 1980b. Prey bones and predators: potential ecologic in-
formation from analysis of bone sites. Ossa, 7: 7597.
Haynes, G. 1982. Utilization and skeletal disturbances of North
American prey carcasses. Arctic, 35: 266281.
Haynes, G. 1983. A guide for differentiating mammalian carnivore
taxa responsible for gnaw damage to herbivore limb bones. Pa-
leobiology, 9: 164172.
Hicks, B.A. (Editor). 2004. Marmes Rockshelter: a final report on
11,000 years of cultural use. Washington State University Press,
Pullman, Wash.
Hill, M.E., Jr., Hill, M.G., and Widga, C.C. 2008. Late Quaternary
Bison diminution on the Great Plains of North America: evaluat-
ing the role of human hunting versus climate change. Quatern-
ary Science Reviews, 27: 17521771. doi:10.1016/j.quascirev.
Hoffecker, J.F., Powers, W.R., and Bigelow, N.H. 1996. Dry
Creek. In American beginnings: the prehistory and palaeoecol-
ogy of Beringia. Edited by F.H. West. University of Chicago
Press, Chicago, Ill., pp. 343352.
Huston, M.A., and Wolverton, S. 2009. The global distribution of
net primary productivity: Resolving the paradox. Ecological
Monographs, 79: 343377. doi:10.1890/08-0588.1.
Johnson, E. 1985. Current developments in bone technology. In
Advances in archaeological method and theory. Vol. 8. Edited
by M.B. Schiffer. Academic Press, Orlando, Fla., pp. 157235.
Johnson, C.G., Jr., Clausnitzer, R.R., Mehringer, P.J., and Oliver,
C.D. 1994. Biotic and abiotic processes of eastside ecosystems:
the effects of management on plant and community ecology, and
on stand and landscape vegetation dynamics. USDA Forest Ser-
vice, Pacific Northwest Region, Portland, Ore., General Techni-
cal Report 322.
Kirk, R. 1968. The search for Marmes Man. Natural History,
77(10): 5659.
Kirk, R. 1969. The rescue of Marmes Man. Pacific Discovery, 22:
Kirk, R. 1970. The oldest man in America. Harcourt Brace Jovano-
vich, New York. N.Y.
Kirk, R., and Daugherty, R.D. 2007. Archaeology in Washington.
University of Washington Press, Seattle, Wash.
Klein, D.R. 1964. Range-related differences in growth of deer re-
flected in skeletal ratios. Journal of Mammalogy, 45: 226235.
Krantz, G.S. 1979. Oldest human remains from the Marmes site.
Northwest Anthropological Research Notes, 13: 159174.
Kurten, B., and Anderson, E. 1980. Pleistocene mammals of North
America. Columbia University Press, New York, N.Y.
Langvatn, R., and Albon, S.D. 1986. Geographic clines in body
weight of Norwegian red deer: a novel explanation of Berg-
manns rule? Holarctic Ecology, 9: 285293.
1380 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press
Lawler, M.C. 1996. Postcranial morphology and systematics of big-
horn sheep (Ovis canadensis) from the late Pleistocene of
Wyoming. Paludicola. Scientific Contributions of the Rochester
Institute of Vertebrate Paleontology, 1: 515.
Leonhardy, F.C., and Rice, D.G. 1970. A proposed cultural typol-
ogy for the Lower Snake River Region, southeastern Washing-
ton. Northwest Anthropological Research Notes, 4: 129.
Lyman, R.L. 1978. Prehistoric butchering techniques in the Lower
Granite Reservoir, southeastern Washington. Tebiwa, 13: 125.
Lyman, R.L. 1989. Taphonomy of cervids killed by the 18 May
1980 volcanic eruption of Mount St. Helens, Washington,
U.S.A. In Bone modification. Edited by R. Bonnichsen and M.
Sorg. University of Maine Center for the Study of Early Man,
Orono, Me., pp. 149167.
Lyman, R.L. 1994. Vertebrate Taphonomy. Cambridge University
Press, Cambridge, U.K.
Lyman, R.L. 1995. A study of variation in the prehistoric butchery
of large artiodactyls. In Ancient peoples and landscapes. Edited
by E. Johnson. Museum of Texas Tech University, Lubbock,
Tex., pp. 233253.
Lyman, R.L. 2004a. Prehistoric biogeography, abundance, and phe-
notypic plasticity of elk (Cervus elaphus) in Washington state.
In Zooarchaeology and conservation biology. Edited by R.L. Ly-
man and K.P. Cannon. University of Utah Press, Salt Lake City,
Utah, pp. 136163.
Lyman, R.L. 2004b. Late-Quaternary diminution and abundance of
prehistoric bison (Bison sp.) in eastern Washington state, U.S.A.
Quaternary Research, 62: 7685. doi:10.1016/j.yqres.2004.04.
Lyman, R.L. 2006. Archaeological evidence of anthropogenically
induced twentieth-century diminution of North American wapiti
(Cervus elaphus). American Midland Naturalist, 156: 8898.
Lyman, R.L., and Fox, G.L. 1989. A critical evaluation of bone
weathering data as an indication of bone assemblage formation.
Journal of Archaeological Science, 16: 293317. doi:10.1016/
McCorquodale, S.M., Eberhardt, L.L., and Eberhardt, L.E. 1988.
Dynamics of a colonizing elk population. The Journal of Wild-
life Management, 52: 309313. doi:10.2307/3801240.
McCorquodale, S.M., Eberhardt, L.E., and Sargeant, G.A. 1989.
Antler characteristics in a colonizing elk population. The Journal
of Wildlife Management, 53: 618621. doi:10.2307/3809185.
McNab, B.K. 2010. Geographic and temporal correlations of mam-
malian size reconsidered: a resource rule. Oecologia, 164:
13?23.. doi:10.1007/s00442-010-1621-5. PMID:20364270.
Mehringer, P.J. 1996. Columbia River Basin ecosystems: late Quatern-
ary environments. Report on file, USDA Forest Service, Interior Co-
lumbia Basin Ecosystem Management Project, Walla Walla, Wash.
Meiri, S., Yom-Tov, Y., and Geffen, E. 2007. What determines
conformity to Bergmanns rule? Global Ecology and Biogeogra-
phy, 16: 788794. doi:10.1111/j.1466-8238.2007.00330.x.
Meltzer, D.J., and Mead, J.I. 1983. The timing of late Pleistocene
mammalian extinctions in North America. Quaternary Research,
19: 130135. doi:10.1016/0033-5894(83)90032-7.
Miller, G.J. 1975. A study of cuts, grooves, and other marks on re-
cent and fossil bone: II, Weathering cracks, fractures, splinters,
and other similar natural phenomena. In Lithic technology: mak-
ing and using stone tools. Edited by E. Swanson. Mouton, The
Hague, the Netherlands, pp. 211226.
Millien, V., Lyons, S.K., Olson, L., Smith, F.A., Wilson, A.B., and
Yom-Tov, Y. 2006. Ecotypic variation in the context of global
climate change: revisiting the rules. Ecology Letters, 9: 853
869. doi:10.1111/j.1461-0248.2006.00928.x. PMID:16796576.
Murie, O.J. 1951. The elk of North America. Stackpole, Harris-
burg, Pa.
OGara, B.W. 2002. Classification. In North American elk: ecology
and management. Edited by D.E. Toweill and J.W. Thomas.
Smithsonian Institution Press, Washington, D.C., pp. 165.
OGara, B.W., and Dundas, R.G. 2002. Distribution: past and pre-
sent. In North American elk: ecology and management. Edited
by D.E. Toweill and J.W. Thomas. Smithsonian Institution
Press, Washington, D.C., pp. 67119.
Olivieri, I., DAngelo, S., Palazzi, C., Padula, A., Mader, R., and
Khan, M. 2009. Diffuse idiopathic skeletal hyperostosis: differen-
tiation from ankylosing sponylitis. Current Rheumatology Reports,
11: 321328. doi:10.1007/s11926-009-0046-9. PMID:19772826.
Ortner, D.J. 2003. Identification of pathological conditions in hu-
man skeletal remains. 2nd ed. Academic Press, Amsterdam, the
Peterson, R.O. 1988. Increased osteoarthritis in moose from Isle
Royale. Journal of Wildlife Diseases, 24: 461466. PMID:
Peterson, R.O., Scheidler, J.M., and Stephens, P.W. 1982. Selected
skeletal morphology and pathology of moose from the Kenai Pe-
ninsula, Alaska, and Isle Royale, Michigan. Canadian Journal of
Zoology, 60: 28122817. doi:10.1139/z82-360.
Post, E., Langvatn, R., Forchhammer, M.C., and Stenseth, N.C.
1999. Environmental variation shapes sexual dimorphism in red
deer. Proceedings of the National Academy of Sciences of the
United States of America, 96: 44674471. doi:10.1073/pnas.96.
8.4467. PMID:10200285.
Purdue, J.R. 1989. Changes during the Holocene in the size of white-
tailed deer (Odocoileus virginianus) from central Illinois. Quatern-
ary Research, 32: 307316. doi:10.1016/0033-5894(89)90097-5.
Purdue, J.R. 1991. Dynamism in the body size of white-tailed deer
(Odocoileus virginianus) from southern Illinois. In Beamers,
bobwhites, and blue-points: tributes to the career of Paul W.
Parmalee. Edited by J.R. Purdue, W.E. Klippel, and B.W. Styles.
Illinois State Museum Scientific Papers, Vol. 23, Springfield,
Ill., pp. 277283.
Rothschild, B.M. 1997. Porosity: a curiosity without diagnostic signif-
icance. American Journal of Physical Anthropology, 104: 529533.
0.CO;2-M. PMID:9453700.
Rothschild, B.M., and Martin, L.D. 1993. Paleopathology. CRC
Press, Boca Raton, Fla.
Rothschild, B.M., and Martin, L.D. 2003. Frequency of pathology
in a large natural sample from Natural Trap Cave with special
remarks on erosive disease in the Pleistocene. Reumatismo, 55:
5865. PMID:12649703.
Schonewald, C. 1994. Cervus canadensis and C. elaphus: North
American subspecies and evaluation of clinal extremes. Acta
Theriologica (Warszawa), 39: 431452.
Shackleton, D.M., and Hills, L.V. 1977. Post-glacial ungulates
(Cervus and Bison) from Three Hills, Alberta. Canadian Journal
of Earth Sciences, 14: 963986.
Sokoloff, L. 1969. The biology of degenerative joint disease. Uni-
versity of Chicago Press, Chicago, Ill.
Takeuchi, A., Goodwin, A.J., Moravec, B.G., Larson, P.B., and
Keller, C.K. 2009. Isotopic evidence for temporal variation in
proportion of seasonal precipitation since the last glacial time in
the Inland Pacific Northwest of the USA. Quaternary Research,
72: 198206. doi:10.1016/j.yqres.2009.06.001.
Thorne, E.T., Williams, E.S., Samuel, W.M., and Kistner, T.P.
2002. Diseases and parasites. In North American elk: ecology
and management. Edited by D.E. Toweill and J.W. Thomas.
Smithsonian Institution Press, Washington, D.C., pp. 351387.
Lyman 1381
Published by NRC Research Press
von den Driesch, A. 1976. A guide to the measurement of animal
bones from archaeological sites. Peabody Museum Bulletin
No. 1. Peabody Museum of Archaeology and Ethnology, Har-
vard University, Cambridge, Mass.
Wang, X. 1988. Systematics and population ecology of Late Pleis-
tocene bighorn sheep (Ovis canadensis) of Natural Trap Cave,
Wyoming. Transactions of the Nebraska Academy of Sciences,
16: 173183.
Watt, C., Mitchell, S., and Salewski, V. 2010. Bergmanns rule; a
concept cluster? Oikos, 119: 89100. doi:10.1111/j.1600-0706.
Weigelt, J. 1989 [original 1927]. Recent vertebrate carcasses and
their paleobiological significance. University of Chicago Press,
Chicago, Ill.
Weiss, E., and Jurmain, R. 2007. Osteoarthritis revisited: a contem-
porary review of aetiology. International Journal of Osteoarch-
aeology, 17: 437450. doi:10.1002/oa.889.
Wisdom, M.J., and Cook, J.G. 2000. North American elk. In Ecol-
ogy and management of large mammals in North America. Edi-
ted by S. Demarais and P.R. Krausman. Prentice Hall, Upper
Saddle River, N.J., pp. 694735.
Wobeser, G., and Runge, W. 1975. Arthropathy in white-tailed
deer and a moose. Journal of Wildlife Diseases, 11: 116121.
Wolverton, S., Huston, M.A., Kennedy, J.H., Cagle, K., and Corne-
lius, J.D. 2009. Conformation to Bergmanns rule in white-tailed
deer can be explained by food availability. American Midland
Naturalist, 162: 403417. doi:10.1674/0003-0031-162.2.403.
Young, M.W. 1969. The story of Marmes man. Gems and Miner-
als, 1969(September): 2223.
1382 Can. J. Earth Sci. Vol. 47, 2010
Published by NRC Research Press