The Bruneau Woodpile: A Miocene Phosphatized Fossil Wood Locality in Southwestern Idaho, USA

Abstract

The Bruneau Woodpile site has long been popular among fossil collectors; however, the deposit has received scant attention from scientists. Our research reveals that the fossilized wood was deposited ca. 6.85 Ma, within the Chalk Hills Formation, and was mineralized with carbonate-fluorapatite. The diverse assemblage of conifers and hardwoods is representative of the warm temperate forests that flourished in southwest Idaho, USA during the late Miocene. Limb and trunk fragments preserved in a single thin sandstone bed appear to represent woody debris that was transported by streams. One possible explanation is that wood, pumice, and sandy volcaniclastic sediment arrived separately as a result of ordinary stream action, and later were combined into a single assemblage during a subsequent high-energy sedimentation event. We favor an alternate hypothesis: a catastrophic event (e.g., a windstorm) damaged trees on slopes bordering the ancient lake. Branches and small trunk fragments were carried by wind and rain into local streams and ponds where they became waterlogged. After a delay that allowed pumice and wood to become saturated, storm water transported these materials, along with finer volcaniclastic sediment, into a lake. The resulting density current produced a fining-upward sedimentary cycle where wood was preserved in the lowest, coarsest stratum.

Full text

geosciences
Article
The Bruneau Woodpile: A Miocene Phosphatized
Fossil Wood Locality in Southwestern Idaho, USA
Mike Viney 1,*, George E. Mustoe 2ID , Thomas A. Dillhoff 3and Paul K. Link 4
1College of Natural Sciences Education and Outreach Center, Colorado State University, Ft. Collins,
CO 80523, USA
2Geology Department, Western Washington University, Bellingham, WA 98225, USA; mustoeg@wwu.edu
3Evolving Earth Foundation, P.O. Box 2090, Issaquah, WA 98027, USA; tdillhoff@evolvingearth.org
4Department of Geosciences, Idaho State University, Pocatello, ID 83209, USA; linkpaul@isu.edu
*Correspondence: mike.viney@colostate.edu; Tel.: +1-970-481-0225
Received: 3 June 2017; Accepted: 5 September 2017; Published: 9 September 2017
Abstract:
The Bruneau Woodpile site has long been popular among fossil collectors; however,
the deposit has received scant attention from scientists. Our research reveals that the fossilized
wood was deposited ca. 6.85 Ma, within the Chalk Hills Formation, and was mineralized with
carbonate-fluorapatite. The diverse assemblage of conifers and hardwoods is representative of the
warm temperate forests that flourished in southwest Idaho, USA during the late Miocene. Limb and
trunk fragments preserved in a single thin sandstone bed appear to represent woody debris that was
transported by streams. One possible explanation is that wood, pumice, and sandy volcaniclastic
sediment arrived separately as a result of ordinary stream action, and later were combined into
a single assemblage during a subsequent high-energy sedimentation event. We favor an alternate
hypothesis: a catastrophic event (e.g., a windstorm) damaged trees on slopes bordering the ancient
lake. Branches and small trunk fragments were carried by wind and rain into local streams and
ponds where they became waterlogged. After a delay that allowed pumice and wood to become
saturated, storm water transported these materials, along with finer volcaniclastic sediment, into
a lake. The resulting density current produced a fining-upward sedimentary cycle where wood was
preserved in the lowest, coarsest stratum.
Keywords:
apatite; carbonate-fluorapatite; Chalk Hills Formation; Chalk Hills Lake; ancient Lake
Idaho; Idaho; Mount Saint Helens; Snake River Plain; phosphatized wood; paleobotany; Bruneau
Woodpile; petrified wood
1. Introduction
The Bruneau Woodpile is an informal name used by petrified wood aficionados to describe a fossil
collecting locality in southwest Idaho, approximately 20 km south of the town of Bruneau (Figure 1[
1
]).
This long-accepted terminology is confusing because the Bruneau Formation, is a younger (Pleistocene)
stratigraphic unit; Bruneau Woodpile occurs in strata of the late Miocene Chalk Hills Formation
(Figure 2[
2
]). Located near the town of Bruneau, Idaho, the site preserves abundant petrified wood in
the form of limbs and driftwood mineralized with carbonate-fluorapatite. The fossil wood assemblage
contains a diverse mix of at least 15 angiosperm and gymnosperm types. Taxonomic affinities of the
Bruneau Woodpile specimens supplement pollen and leaf studies of similar aged deposits that help
establish the vegetative types making up late Miocene ecosystems of southwest Idaho [
3
–
6
]. This is the
first scientific investigation that provides a comprehensive overview of this unusual North American
fossil wood locality. Our research focuses on two puzzling aspects: the occurrence of limb and small
wood fragments in a single stratum within a stratigraphic section that otherwise contains no petrified
wood, and the phosphatic mineralization of wood.
Geosciences 2017,7, 82; doi:10.3390/geosciences7030082 www.mdpi.com/journal/geosciences

Geosciences 2017,7, 82 2 of 25
Geosciences 2017, 7, 82 2 of 25
single stratum within a stratigraphic section that otherwise contains no petrified wood, and the
phosphatic mineralization of wood.
Figure 1. Location map adapted from Lewis et al. [1].
Figure 2. Generalized Stratigraphic Section [2].
Figure 1. Location map adapted from Lewis et al. [1].
Geosciences 2017, 7, 82 2 of 25
single stratum within a stratigraphic section that otherwise contains no petrified wood, and the
phosphatic mineralization of wood.
Figure 1. Location map adapted from Lewis et al. [1].
Figure 2. Generalized Stratigraphic Section [2].
Figure 2. Generalized Stratigraphic Section [2].

Geosciences 2017,7, 82 3 of 25
1.1. Previous Work
In the late 1960s, Lt. Col. Quincy D. Howell and J. V. Root enlisted the help of members of
the Gem Club of Boise to identify fossil wood from Bruneau Woodpile and other sites in southern
Idaho and eastern Oregon. This group published their findings in the October 1971 issue of Gems and
Minerals [
7
]. They concluded that the wood had washed into an ancient lake during the late Pliocene.
In the absence of analytical data, the petrified wood was presumed to consist of calcium sulfate (CaSO
4
)
and calcium carbonate (CaCO
3
). While polished transverse wood sections were used to investigate
taxonomic affinities of the Bruneau Woodpile assemblage, corroboration based on microscopic thin
sections was not attained until now. Individual taxa have been previously described from the Bruneau
Woodpile, including a bracket fungus [
8
] and pine cones [
9
]. Our investigation is a long-delayed
successor to the work begun by the original Bruneau Woodpile study group.
1.2. Geologic Setting
The Bruneau Woodpile is located in the western Snake River Plain, a broad arcuate depression
that extends across southern Idaho, USA. The Neogene volcanic history of southern Idaho was dictated
by the passage of the North American Plate over a stationary mantle plume that is currently situated
beneath Yellowstone National Park, Wyoming [
10
]. The western part of the Snake River Plain is
a NW trending fault-bounded graben, while the eastern part is a structural downwarp created by the
weight of overlying volcanic and sedimentary rocks interacting with the formation of a mid-crustal
sill and thermal contraction after passage over the plume-generated hotspot [
11
]. The western section
contains extensive rhyolitic tuffs and ash flows of the Miocene Idavada Volcanic Group, ranging in age
from 15 to 11 Ma [12,13].
These volcanic materials are overlain by younger mafic volcanic rocks and fluvial and lacustrine
sediments (Chalk Hills and Glenns Ferry formations) that contain volcanic and hyaloclastic interbeds
(Figures 2and 3). Sedimentation began at approximately 9 Ma, when several lakes developed. One of
these was Chalk Hills Lake, in which the Bruneau Woodpile sediments were deposited. These lakes
expanded and retreated in size, eventually coalescing to form a single enormous lake, ancient Lake
Idaho (represented by the Glenns Ferry Formation, Figure 2), which ultimately extended over an area
of several thousand km
2
[
2
,
14
]. During the next 6.5 million years, thick basin-fill sediment accumulated
in the lake basin. This sediment contains material eroded from nearby mountain ranges, as well as
rhyolitic tephra. These sediments locally preserve plant and animal fossils.
Geosciences 2017, 7, 82 3 of 25
1.1. Previous Work
In the late 1960s, Lt. Col. Quincy D. Howell and J. V. Root enlisted the help of members of the
Gem Club of Boise to identify fossil wood from Bruneau Woodpile and other sites in southern Idaho
and eastern Oregon. This group published their findings in the October 1971 issue of Gems and
Minerals [7]. They concluded that the wood had washed into an ancient lake during the late Pliocene.
In the absence of analytical data, the petrified wood was presumed to consist of calcium sulfate
(CaSO4) and calcium carbonate (CaCO3). While polished transverse wood sections were used to
investigate taxonomic affinities of the Bruneau Woodpile assemblage, corroboration based on
microscopic thin sections was not attained until now. Individual taxa have been previously
described from the Bruneau Woodpile, including a bracket fungus [8] and pine cones [9]. Our
investigation is a long-delayed successor to the work begun by the original Bruneau Woodpile study
group.
1.2. Geologic Setting
The Bruneau Woodpile is located in the western Snake River Plain, a broad arcuate depression
that extends across southern Idaho, USA. The Neogene volcanic history of southern Idaho was
dictated by the passage of the North American Plate over a stationary mantle plume that is currently
situated beneath Yellowstone National Park, Wyoming [10]. The western part of the Snake River
Plain is a NW trending fault-bounded graben, while the eastern part is a structural downwarp
created by the weight of overlying volcanic and sedimentary rocks interacting with the formation of
a mid-crustal sill and thermal contraction after passage over the plume-generated hotspot [11]. The
western section contains extensive rhyolitic tuffs and ash flows of the Miocene Idavada Volcanic
Group, ranging in age from 15 to 11 Ma [12,13].
These volcanic materials are overlain by younger mafic volcanic rocks and fluvial and
lacustrine sediments (Chalk Hills and Glenns Ferry formations) that contain volcanic and
hyaloclastic interbeds (Figures 2 and 3). Sedimentation began at approximately 9 Ma, when several
lakes developed. One of these was Chalk Hills Lake, in which the Bruneau Woodpile sediments
were deposited. These lakes expanded and retreated in size, eventually coalescing to form a single
enormous lake, ancient Lake Idaho (represented by the Glenns Ferry Formation, Figure 2), which
ultimately extended over an area of several thousand km2 [2,14]. During the next 6.5 million years,
thick basin-fill sediment accumulated in the lake basin. This sediment contains material eroded from
nearby mountain ranges, as well as rhyolitic tephra. These sediments locally preserve plant and
animal fossils.
The petrified wood assemblage investigated in our study is preserved in lacustrine sediments
of the Chalk Hills Formation. This Miocene-age lake is here referred to as the Chalk Hills Lake to
distinguish it from the Pliocene ancient Lake Idaho. Other important Snake River Plain fossil sites
include the late Miocene Pickett Creek leaf and pollen deposit [6] and the Hagerman Fossil Beds
National Monument, where Pliocene strata preserve a multitude of vertebrate fossils [15].
Figure 3. Modern views of the Snake River Plain, near Salmon Falls (A) and Shoshone Falls (B);
Idaho USA.
Figure 3.
Modern views of the Snake River Plain, near Salmon Falls (
A
) and Shoshone Falls (
B
);
Idaho USA.
The petrified wood assemblage investigated in our study is preserved in lacustrine sediments
of the Chalk Hills Formation. This Miocene-age lake is here referred to as the Chalk Hills Lake to
distinguish it from the Pliocene ancient Lake Idaho. Other important Snake River Plain fossil sites
include the late Miocene Pickett Creek leaf and pollen deposit [
6
] and the Hagerman Fossil Beds
National Monument, where Pliocene strata preserve a multitude of vertebrate fossils [15].

Geosciences 2017,7, 82 4 of 25
1.3. Paleoenvironment
Southern Idaho is now an arid region where extensive lower elevations are sagebrush steppe
flora, bordered by barren mountains (Figure 3). The region typically receives less than 250 mm of rain
each year, with temperatures that range from −34 ◦C in winter to 43 ◦C in summer [5].
Neogene environmental conditions were very different. By the late Miocene lakes occupied
a large area of southern Idaho, eventually expanding in the Pliocene to produce ancient Lake Idaho;
the origin of which was related to formation of the western Snake River Plain graben [
16
]. Topography
of the surrounding watershed was relatively gentle; much of the lakebed sediment consists of small
pebble to clay-sized fractions transported by local stream tributaries as well as larger rivers. Lake
levels fluctuated over time, due to both tectonic and climatic changes. Ultimately, ancient Lake Idaho
drained when the outlet river incised Hells Canyon, breaching the lake basin. In place of the ancient
lake, the Snake River now traverses southern Idaho [17,18].
The late Miocene Chalk Hills Formation, host to the Bruneau Woodpile, represents a time when
Pacific Northwest forests were more uniform and more widespread than today, in large part because
the Cascade Range had not yet reached an elevation sufficient to cause a pronounced rain shadow.
The Chalk Hills Formation paleoflora originated during the late Miocene, when the northwest climate
was transitioning from the warm, humid regime of the middle Miocene to the more arid, summer-dry
climate of today. Low elevation forests bore a resemblance to the modern broadleaved hardwood
forests of north central USA. Annual precipitation was in the range of 1200–1300 mm, with a mean
annual temperature of approximately 13.5
◦
C [
5
]. Analyses of fossil pollen show Pinus as the dominant
conifer, but Abies,Picea, and Tsuga were present. Deciduous trees were abundant and taxonomically
diverse [
3
,
5
]. The Pickett Creek leaf and pollen deposit [
6
] of the Chalk Hills Formation documents
a flora with a high incidence of summer-dry tolerant species, including white oaks, evergreen oaks,
pine and legumes, similar to the wood types found in the Bruneau Woodpile assemblage.
The lake sediments also preserve fish remains that include cold water salmon and trout and
warm water sunfish and catfish, indicative of warm, moist climate with cool summers and mild
winters [
19
,
20
]. Two fish taxa, Mylocheilus inflexus (a minnow) and Paleolox larsonia (a salmon) found in
the Chalk Hills Formation do not occur in the overlying lacustrine sediments of the Pliocene Glenns
Ferry Formation [
2
]. Chalk Hills Formation strata also preserve remains of beaver, rhinoceros, camel,
horse, sloth, and various molluscs [19].
By the mid Pliocene, the development of the Cascade Range rain shadow [
5
] and global climatic
change [
21
] profoundly altered the flora and fauna of the Snake River Plain. Temperate forests
were replaced by pine/oak woodlands with areas of grassland. The mid-Pliocene climate was drier,
with perhaps half the annual rainfall of the Chalk Hills Formation. Lake shore, ponds, open woodlands
and grasslands were inhabited by a diverse vertebrate fauna that included swans, grebes, and other
water birds, rodents, mustelids, saber-tooth tigers, camels, and horses [
14
]. The Pleistocene brought
increased aridity to the Snake River Plain. By 1.3 Ma, annual rainfall had decreased to approximately
250 mm, and the landscape became dominated by sagebrush (Artemisia), saltbush (Sarcobatus) and
grasses, i.e., semi-desert sagebrush/bunchgrass steppe flora that flourishes in modern times [
5
]
(Figure 3).
2. Methods and Materials
SEM/EDS images were obtained for 10 specimens using a Vega Tescan SEM (Tescan, Brno,
Czech Republic) equipped with an EDAX Genesis XRF spectrometer (EDAX, Mahwah, NJ, USA).
X-ray diffraction (XRD) patterns were made for two specimens with a Rigaku Geigerflex diffractometer
using Ni-filtered Cu K-
α
radiation. Tephra geochronology is based on electron microprobe analysis of
Bruneau Woodpile volcanic ash collected by Paul Link, using the elemental data to correlate the tephra
with previously-analyzed specimens from other Snake River Plain strata of known radiometric age.
Glass shards recovered from a vitric ash bed at the Bruneau Woodpile site were analyzed using electron

Geosciences 2017,7, 82 5 of 25
microprobe by B.P. Nash (University of Utah, UT, USA) who interpolated the age by chemostratigraphic
comparison with western North American vitric tuffs spanning 16 to 6 Ma [22].
Fossil wood thin sections were prepared using standard techniques and examined and
photographed using a Nikon Eclipse 50i microscope (Nikon, Tokyo, Japan) coupled with a Nikon
DS-Fi1 camera. Thin sections described in this study were deposited at the University of Washington
Burke Museum in Seattle, WA, USA. Identifications were aided by standard references, including the
Textbook of Wood Technology [23] and the InsideWood image database [24].
3. Results
3.1. Site Description
The wood-bearing beds occur at near-surface depths at several localities, where collecting has
been popular since the 1950s [
7
,
25
]. Our research was conducted at a site accessible via several km of
rough dirt road at coordinates of 42◦45047.1900 N, 115◦53029.4600 W.
Phosphatized wood is preserved in a sandy bed that marks the bottom of a 15 cm thick
fining-upward sequence (Figure 4). Specimens are abundant for those who are willing to first remove
more than a meter of partially-cemented sediment to reach the wood-bearing stratum (Figures 5–7).
The stratum is rich in volcanic material that includes numerous pumice clasts and obsidian granules in
a fine tephra-rich matrix (Figure 7) containing well-rounded, moderately well-sorted quartz grains of
similar size and shape to equally abundant obsidian granules. Disarticulated fish bones are sparsely
distributed within the wood-bearing stratum.
The majority of fossil wood specimens are elongated limbs or irregular fragments of trunk
wood that have characteristics indicative of fluvial transport. Exterior surfaces have been smoothed
by erosion, bark and side branches have been removed, and terminations are rounded (Figure 8).
The wood-bearing bed is overlain by a light gray layer of white fine sand with parallel laminations,
capped by a thin layer of whitish silt. This interval completes the first fining-upward sequence.
A second, 30-cm upward fining succession begins with a bed of cross bedded vitric ash that passes
upward to rippled ash and to massive white argillaceous siltstone. The second upward fining sequence
is overlain by a brown cemented layer and a massive white clay. Above that is a fallout ash layer
consisting of gray fine pumice sand and vitric gray rhyolitic ash. Geochemical analysis and correlation
suggests that this younger ash layer can be assigned a chemostratigraphic age of 6.84
±
0.05 Ma
(late Miocene) [26].
Geosciences 2017, 7, 82 5 of 25
interpolated the age by chemostratigraphic comparison with western North American vitric tuffs
spanning 16 to 6 Ma [22].
Fossil wood thin sections were prepared using standard techniques and examined and
photographed using a Nikon Eclipse 50i microscope (Nikon, Tokyo, Japan) coupled with a Nikon
DS-Fi1 camera. Thin sections described in this study were deposited at the University of Washington
Burke Museum in Seattle, WA, USA. Identifications were aided by standard references, including
the Textbook of Wood Technology [23] and the InsideWood image database [24].
3. Results
3.1. Site Description
The wood-bearing beds occur at near-surface depths at several localities, where collecting has
been popular since the 1950s [7,25]. Our research was conducted at a site accessible via several km of
rough dirt road at coordinates of 42°45′47.19″ N, 115°53′29.46″ W.
Phosphatized wood is preserved in a sandy bed that marks the bottom of a 15 cm thick
fining-upward sequence (Figure 4). Specimens are abundant for those who are willing to first
remove more than a meter of partially-cemented sediment to reach the wood-bearing stratum
(Figures 5–7). The stratum is rich in volcanic material that includes numerous pumice clasts and
obsidian granules in a fine tephra-rich matrix (Figure 7) containing well-rounded, moderately
well-sorted quartz grains of similar size and shape to equally abundant obsidian granules.
Disarticulated fish bones are sparsely distributed within the wood-bearing stratum.
The majority of fossil wood specimens are elongated limbs or irregular fragments of trunk
wood that have characteristics indicative of fluvial transport. Exterior surfaces have been smoothed
by erosion, bark and side branches have been removed, and terminations are rounded (Figure 8).
The wood-bearing bed is overlain by a light gray layer of white fine sand with parallel laminations,
capped by a thin layer of whitish silt. This interval completes the first fining-upward sequence. A
second, 30-cm upward fining succession begins with a bed of cross bedded vitric ash that passes
upward to rippled ash and to massive white argillaceous siltstone. The second upward fining
sequence is overlain by a brown cemented layer and a massive white clay. Above that is a fallout ash
layer consisting of gray fine pumice sand and vitric gray rhyolitic ash. Geochemical analysis and
correlation suggests that this younger ash layer can be assigned a chemostratigraphic age of 6.84 ±
0.05 Ma (late Miocene) [26].
Figure 4. Generalized stratigraphic section of the Chalk Hills Formation beds at Bruneau Woodpile
locality.
Figure 4.
Generalized stratigraphic section of the Chalk Hills Formation beds at Bruneau
Woodpile locality.

Geosciences 2017,7, 82 6 of 25
Geosciences 2017, 7, 82 6 of 25
Figure 5. Bruneau Woodpile site. (A) Revealing the wood-bearing stratum requires excavation of one
meter of cemented sediment; (B) Limb sections and trunk fragments occur in a weakly cemented
sandstone bed that contains large pumice clasts. Some wood specimens have a thin black exterior
coating composed of manganese oxide, as evidenced by SEM/EDS analysis; (C) The wood-bearing
layer is overlain by a ~7 cm white fine sandy ash with parallel laminations. Note the abundance of
pumice clasts in the sandy sediment below the elongate fossil twig.
Figure 6. Phosphatized wood at the Bruneau Woodpile site. Elongate limbs and larger wood
fragments are positioned in different orientations.
Figure 5.
Bruneau Woodpile site. (
A
) Revealing the wood-bearing stratum requires excavation of
one meter of cemented sediment; (
B
) Limb sections and trunk fragments occur in a weakly cemented
sandstone bed that contains large pumice clasts. Some wood specimens have a thin black exterior
coating composed of manganese oxide, as evidenced by SEM/EDS analysis; (
C
) The wood-bearing
layer is overlain by a ~7 cm white fine sandy ash with parallel laminations. Note the abundance of
pumice clasts in the sandy sediment below the elongate fossil twig.
Geosciences 2017, 7, 82 6 of 25
Figure 5. Bruneau Woodpile site. (A) Revealing the wood-bearing stratum requires excavation of one
meter of cemented sediment; (B) Limb sections and trunk fragments occur in a weakly cemented
sandstone bed that contains large pumice clasts. Some wood specimens have a thin black exterior
coating composed of manganese oxide, as evidenced by SEM/EDS analysis; (C) The wood-bearing
layer is overlain by a ~7 cm white fine sandy ash with parallel laminations. Note the abundance of
pumice clasts in the sandy sediment below the elongate fossil twig.
Figure 6. Phosphatized wood at the Bruneau Woodpile site. Elongate limbs and larger wood
fragments are positioned in different orientations.
Figure 6.
Phosphatized wood at the Bruneau Woodpile site. Elongate limbs and larger wood fragments
are positioned in different orientations.

Geosciences 2017,7, 82 7 of 25
Geosciences 2017, 7, 82 7 of 25
Figure 7. Bruneau Woodpile sedimentology. (A,B) are sediments from wood-bearing stratum. (A)
Black obsidian granules and translucent gray quartz grain; (B) Pumice clast; (C) fine-grained ash-rich
bed that overlies wood-bearing stratum; (D) fish vertebra in sandy sediment containing quartz and
obsidian granules.
Figure 8. Bruneau Woodpile specimens have characteristics indicative of fluvial transport. Two
morphotypes are common: elongate limbs (A–C) and irregular fragments of trunk wood (D–F). For
both types, exterior surfaces have been smoothed by erosion, bark and side branches have been
removed, and terminations are rounded. Elongate limbs (A) break into shorter segments during
excavation from the matrix because of the brittleness of the phosphatized wood. These morphologies
are identical to modern fluvial driftwood (G).
Figure 7.
Bruneau Woodpile sedimentology. (
A
,
B
) are sediments from wood-bearing stratum.
(A) Black obsidian
granules and translucent gray quartz grain; (
B
) Pumice clast; (
C
) fine-grained
ash-rich bed that overlies wood-bearing stratum; (
D
) fish vertebra in sandy sediment containing quartz
and obsidian granules.
Geosciences 2017, 7, 82 7 of 25
Figure 7. Bruneau Woodpile sedimentology. (A,B) are sediments from wood-bearing stratum. (A)
Black obsidian granules and translucent gray quartz grain; (B) Pumice clast; (C) fine-grained ash-rich
bed that overlies wood-bearing stratum; (D) fish vertebra in sandy sediment containing quartz and
obsidian granules.
Figure 8. Bruneau Woodpile specimens have characteristics indicative of fluvial transport. Two
morphotypes are common: elongate limbs (A–C) and irregular fragments of trunk wood (D–F). For
both types, exterior surfaces have been smoothed by erosion, bark and side branches have been
removed, and terminations are rounded. Elongate limbs (A) break into shorter segments during
excavation from the matrix because of the brittleness of the phosphatized wood. These morphologies
are identical to modern fluvial driftwood (G).
Figure 8.
Bruneau Woodpile specimens have characteristics indicative of fluvial transport.
Two morphotypes are common: elongate limbs (
A
–
C
) and irregular fragments of trunk wood (
D
–
F
).
For both types, exterior surfaces have been smoothed by erosion, bark and side branches have been
removed, and terminations are rounded. Elongate limbs (
A
) break into shorter segments during
excavation from the matrix because of the brittleness of the phosphatized wood. These morphologies
are identical to modern fluvial driftwood (G).

Geosciences 2017,7, 82 8 of 25
3.2. Age
The Bruneau Woodpile was originally assigned a late Pliocene age [
7
]. Geochemical analysis [
26
]
of rhyolitic glass shards in the tuff bed ~80 cm above the wood bed (stratigraphic location of sample
shown in (Figure 4), matches an unnamed peralkaline ash bed below the Blacktail Creek Tuff [
27
] from
the Heise volcanic field of the eastern Snake River Plain. The Blacktail Creek Tuff is 6.61
±
0.01 Ma
(Ar-Ar age from welded ash of sample PTH4 [
28
]). Our sample of the unnamed ash 80 cm above the
Bruneau Woodpile bed was chemostratigraphically correlated by B.P. Nash (University of Utah) to
an age of 6.84
±
0.05 Ma [
26
]. Based on these dates, the Bruneau Woodpile occurs in the Chalk Hills
Formation (late Miocene) [29].
3.3. Mineralogy and Geochemistry
While the Bruneau Woodpile fossil wood specimens have long been identified as being
mineralized with calcium carbonate or calcium sulfate [
7
], new analytical data reveal that the
mineralization is phosphatic.
X-ray diffraction patterns (Figure 9) show that the only discernable crystalline phase is
a member of the apatite family, comprising members that fall within the general formula of
Ca5(PO4)3(F, OH, Cl) [30,31]
. The three members, fluorapatite, hydroxylapatite, and chlorapatite
are not easily distinguishable from XRD data, but SEM/EDS spectra (Figure 10) show that fluorine
is present at levels of ~1% by weight. Cl is below detection limits, and the inability of EDS to detect
H and poor quantification of O prevent recognition of OH
−
. However, the presence of a significant
fluorine spectral peak suggests that calcium phosphate is present as fluorapatite, which is sometimes
given the mineral name francolite.
Geosciences 2017, 7, 82 8 of 25
3.2. Age
The Bruneau Woodpile was originally assigned a late Pliocene age [7]. Geochemical analysis
[26] of rhyolitic glass shards in the tuff bed ~80 cm above the wood bed (stratigraphic location of
sample shown in (Figure 4), matches an unnamed peralkaline ash bed below the Blacktail Creek Tuff
[27] from the Heise volcanic field of the eastern Snake River Plain. The Blacktail Creek Tuff is 6.61 ±
0.01 Ma (Ar-Ar age from welded ash of sample PTH4 [28]). Our sample of the unnamed ash 80 cm
above the Bruneau Woodpile bed was chemostratigraphically correlated by B.P. Nash (University of
Utah) to an age of 6.84 ± 0.05 Ma [26]. Based on these dates, the Bruneau Woodpile occurs in the
Chalk Hills Formation (late Miocene) [29].
3.3. Mineralogy and Geochemistry
While the Bruneau Woodpile fossil wood specimens have long been identified as being
mineralized with calcium carbonate or calcium sulfate [7], new analytical data reveal that the
mineralization is phosphatic.
X-ray diffraction patterns (Figure 9) show that the only discernable crystalline phase is a
member of the apatite family, comprising members that fall within the general formula of
Ca5(PO4)3(F, OH, Cl) [30,31]. The three members, fluorapatite, hydroxylapatite, and chlorapatite are
not easily distinguishable from XRD data, but SEM/EDS spectra (Figure 10) show that fluorine is
present at levels of ~ 1% by weight. Cl is below detection limits, and the inability of EDS to detect H
and poor quantification of O prevent recognition of OH−. However, the presence of a significant
fluorine spectral peak suggests that calcium phosphate is present as fluorapatite, which is sometimes
given the mineral name francolite.
Figure 9. X-Ray diffraction (XRD) pattern. All peaks are from an apatite group mineral.
The large carbon peak in the SEM/EDS spectrum (Figure 10) suggests that the mineralization is
carbonate-fluorapatite. The 450 °C LOI (loss on ignition) (Table 1) indicates an organic carbon
content of several percent, but the EDS spectrum was obtained from microcrystalline material within
a vessel lumen, where the analyzed material was separated from any possible relict organic matter
in the cell walls. This occurrence is evidence that carbon is present as an inorganic constituent within
the phosphatic structure. Carbonate-fluorapatite results from the partial substitution of CO3−2 in
place of PO4−3, equivalent to a generalized formula of Ca5(PO4, CO3)3(F, OH, Cl) [30].
The amount of carbonate in carbonate-fluorapatite can be estimated based on major element
geochemistry (Table 2). Carbon concentrations cannot be accurately estimated by SEM/EDS
spectrometry because of the element’s low atomic mass that makes it undetectable by
Figure 9. X-ray diffraction (XRD) pattern. All peaks are from an apatite group mineral.
The large carbon peak in the SEM/EDS spectrum (Figure 10) suggests that the mineralization is
carbonate-fluorapatite. The 450
◦
C LOI (loss on ignition) (Table 1) indicates an organic carbon content
of several percent, but the EDS spectrum was obtained from microcrystalline material within a vessel
lumen, where the analyzed material was separated from any possible relict organic matter in the
cell walls. This occurrence is evidence that carbon is present as an inorganic constituent within the
phosphatic structure. Carbonate-fluorapatite results from the partial substitution of CO
3−2
in place of
PO4−3, equivalent to a generalized formula of Ca5(PO4, CO3)3(F, OH, Cl) [30].
The amount of carbonate in carbonate-fluorapatite can be estimated based on major
element geochemistry (Table 2). Carbon concentrations cannot be accurately estimated by
SEM/EDS spectrometry because of the element’s low atomic mass that makes it undetectable by

Geosciences 2017,7, 82 9 of 25
wavelength-dispersive XRF spectrometry. However, when wt. % values are recalculated as atomic %
values, the ratio of Ca:P provides an important compositional indication (Table 2). For pure apatite,
the Ca:P proportion is 1.67. A Ca:P ratio that exceeds this value provides evidence that the mineral is
deficient in P, suggesting that CO
3
is present in an amount sufficient to provide a neutral molecular
structure. Both Bruneau Woodpile samples have ratios that exceed that of ideal apatite, substantiating
the existence of a carbonate component. For this reason, we identify the mineralizing material of the
Bruneau Woodpile fossil wood specimens as carbonate-fluorapatite. The small amount of SiO
2
may
be caused by substitution of silica for phosphorus in the mineral structure; the absence of quartz or
other silicate mineral peaks in the XRD pattern (Figure 9) suggests that Si is not present as a secondary
mineral phase.
Geosciences 2017, 7, 82 9 of 25
wavelength-dispersive XRF spectrometry. However, when wt. % values are recalculated as atomic %
values, the ratio of Ca:P provides an important compositional indication (Table 2). For pure apatite,
the Ca:P proportion is 1.67. A Ca:P ratio that exceeds this value provides evidence that the mineral is
deficient in P, suggesting that CO3 is present in an amount sufficient to provide a neutral molecular
structure. Both Bruneau Woodpile samples have ratios that exceed that of ideal apatite,
substantiating the existence of a carbonate component. For this reason, we identify the mineralizing
material of the Bruneau Woodpile fossil wood specimens as carbonate-fluorapatite. The small amount
of SiO2 may be caused by substitution of silica for phosphorus in the mineral structure; the absence
of quartz or other silicate mineral peaks in the XRD pattern (Figure 9) suggests that Si is not present
as a secondary mineral phase.
Figure 10. SEM/EDS spectrum, 15 KeV. Pd peaks are spectral artifacts from sputter-coating the
specimen with Pd to provide electrical conductivity. Calculated fluorine content = 1.03 wt. % (= 1.11
atomic %); chlorine content = 0.0 wt. % (= 0.0 atomic %).
Table 1. Major element analysis by X-Ray Fluorescence Spectrometry. Elemental results are
calculated as wt. % oxide.
Sample SiO2 Al2O3 TiO2 Fe2O3* MnO MgO CaO K2ONa2OP2O5 LOI **
1 14.53 0.12 0.00 1.88 0.45 0.98 44.36 0.01 0.94 37.79 3.83
2 13.69 0.04 0.00 1.73 0.50 0.55 45.67 0.00 0.00 33.33 5.83
* total iron calculated as Fe2O3. ** wt % loss on ignition after 450 °C heating.
Table 2. Conversion of % oxide values to atomic %, and calculated Ca:P ratios.
Sample Weight % Atomic % Ca:P Ratio
CaO P2O5Ca P
1 44.36 37.79 31.72 16.48 1.92
2 45.67 33.33 32.65 14.53 2.25
Ideal formula: Ca5(PO4)3 Ca:P = 5:3 = 1.67.
High magnification images (Figure 11) show that the carbonate-fluorapatite is present in
crystalline form, consistent with the strong peaks observed in the XRD pattern (Figure 9). No relict
organic matter is visible; anatomical features are preserved as inorganic replacements. Quality of
anatomical preservation is variable. Many specimens show preserved cellular features (Figure 12).
Features of a few specimens include spiral fibril patterns in the middle cell wall (Figure 12B,C),
evidence of trees that have anatomical asymmetry because of oblique gravitational forces, e.g., a tree
growing on a slope, or on branched limbs [23]. One specimen (Figure 12D) shows evidence of
microbes that bored branched galleries in the exterior surface. Possible organisms include
Figure 10. SEM/EDS spectrum, 15 KeV. Pd peaks are spectral artifacts from sputter-coating the specimen
with Pd to provide electrical conductivity. Calculated fluorine content = 1.03 wt. % (= 1.11 atomic %);
chlorine content = 0.0 wt. % (= 0.0 atomic %).
Table 1.
Major element analysis by X-ray Fluorescence Spectrometry. Elemental results are calculated
as wt. % oxide.
Sample SiO2Al2O3TiO2Fe2O3* MnO MgO CaO K2O Na2O P2O5LOI **
1
14.53
0.12 0.00 1.88 0.45 0.98
44.36
0.01 0.94
37.79
3.83
2
13.69
0.04 0.00 1.73 0.50 0.55
45.67
0.00 0.00
33.33
5.83
* Total iron calculated as Fe2O3; ** wt. % loss on ignition after 450 ◦C heating.
Table 2. Conversion of % oxide values to atomic %, and calculated Ca:P ratios.
Sample Weight % Atomic % Ca:P
Ratio
CaO P2O5Ca P
1 44.36 37.79 31.72 16.48 1.92
2 45.67 33.33 32.65 14.53 2.25
Ideal formula: Ca5(PO4)3Ca:P = 5:3 = 1.67.
High magnification images (Figure 11) show that the carbonate-fluorapatite is present in
crystalline form, consistent with the strong peaks observed in the XRD pattern (Figure 9). No relict
organic matter is visible; anatomical features are preserved as inorganic replacements. Quality of
anatomical preservation is variable. Many specimens show preserved cellular features (Figure 12).
Features of a few specimens include spiral fibril patterns in the middle cell wall (Figure 12B,C),
evidence of trees that have anatomical asymmetry because of oblique gravitational forces, e.g., a tree

Geosciences 2017,7, 82 10 of 25
growing on a slope, or on branched limbs [
23
]. One specimen (Figure 12D) shows evidence of microbes
that bored branched galleries in the exterior surface. Possible organisms include cyanobacteria, algae,
and fungi. These microbes are known to penetrate the phosphate-enriched skeletons of corals as
a source of phosphorus, a macronutrient that is often scarce in pristine natural environments [
32
].
Chemolithotrophic bacteria are also a possibility, particularly phosphobacteria, which use phosphorus
reduction as a source of metabolic energy [33].
Geosciences 2017, 7, 82 10 of 25
cyanobacteria, algae, and fungi. These microbes are known to penetrate the phosphate-enriched
skeletons of corals as a source of phosphorus, a macronutrient that is often scarce in pristine natural
environments [32]. Chemolithotrophic bacteria are also a possibility, particularly phosphobacteria,
which use phosphorus reduction as a source of metabolic energy [33].
Figure 11. SEM photomicrographs of Bruneau Woodpile wood show microcrystalline structure of
carbonate-fluorapatite. (A) Longitudinal view of two tracheids showing circular pits; (B) Fractured
cell walls of a tracheid; (C) Interior surface of a tracheid showing numerous pit apertures; (D)
Carbonate-fluorapatite crystals bordering a pit aperture.
Figure 12. SEM images of Bruneau Woodpile specimens. (A) Conifer wood, radial orientation,
showing tracheids with well-preserved bordered pits; (B,C) Radial view of specimens of reaction
wood, where tracheids have spiral architecture in the secondary cell wall; (D) Exterior surface of this
limb shows galleries produced in the mineralized wood by chemolithotrophic microbes (see text for
discussion).
Figure 11.
SEM photomicrographs of Bruneau Woodpile wood show microcrystalline structure of
carbonate-fluorapatite. (
A
) Longitudinal view of two tracheids showing circular pits; (
B
) Fractured
cell walls of a tracheid; (
C
) Interior surface of a tracheid showing numerous pit apertures;
(D) Carbonate-fluorapatite crystals bordering a pit aperture.
Geosciences 2017, 7, 82 10 of 25
cyanobacteria, algae, and fungi. These microbes are known to penetrate the phosphate-enriched
skeletons of corals as a source of phosphorus, a macronutrient that is often scarce in pristine natural
environments [32]. Chemolithotrophic bacteria are also a possibility, particularly phosphobacteria,
which use phosphorus reduction as a source of metabolic energy [33].
Figure 11. SEM photomicrographs of Bruneau Woodpile wood show microcrystalline structure of
carbonate-fluorapatite. (A) Longitudinal view of two tracheids showing circular pits; (B) Fractured
cell walls of a tracheid; (C) Interior surface of a tracheid showing numerous pit apertures; (D)
Carbonate-fluorapatite crystals bordering a pit aperture.
Figure 12. SEM images of Bruneau Woodpile specimens. (A) Conifer wood, radial orientation,
showing tracheids with well-preserved bordered pits; (B,C) Radial view of specimens of reaction
wood, where tracheids have spiral architecture in the secondary cell wall; (D) Exterior surface of this
limb shows galleries produced in the mineralized wood by chemolithotrophic microbes (see text for
discussion).
Figure 12.
SEM images of Bruneau Woodpile specimens. (
A
) Conifer wood, radial orientation,
showing tracheids with well-preserved bordered pits; (
B
,
C
) Radial view of specimens of reaction
wood, where tracheids have spiral architecture in the secondary cell wall; (
D
) Exterior surface of
this limb shows galleries produced in the mineralized wood by chemolithotrophic microbes (see text
for discussion).

Geosciences 2017,7, 82 11 of 25
3.4. Paleobotany
Taxonomic information from fossil wood collected at the Bruneau Woodpile is limited due to poor
anatomical preservation, especially in the radial and tangential sections. Even so, a number of types
can be assigned to family or genus based on available features. The best representatives of each type
were selected for sectioning from more than 200 specimens. Larger images and anatomical descriptions
are provided in Supplement 1. Taxa recognized from the Bruneau Woodpile are as follows (Figure 13):
Conifers
•Cupressaceae
•Pinaceae
Piceoxylon (Spruce/Douglas fir type)
Pinus sp. (Pine)
Angiosperms
•Berberidaceae
cf. Berberis (Barberry/Oregon grape type)
•Fabaceae
cf. Robinia (Black locust type)
•Fagaceae
Quercus sp. (White oak group)
Quercus/Lithocarpus (Live oak group)
•Juglandaceae
Carya sp. (True hickory type)
•cf. Salicaceae
Populus/Salix (Poplar/willow type)
•Sapindaceae
Acer sp. (Soft maple group)
•Ulmaceae
Ulmus sp. (Elm)
•Undetermined angiosperms
Undetermined hardwood 1
Undetermined hardwood 2
Undetermined hardwood 3
Undetermined hardwood 4

Geosciences 2017,7, 82 12 of 25
Geosciences 2017, 7, 82 12 of 25
Figure 13. Fossil wood types from the Bruneau Woodpile (University of Washington Burke Museum
specimen numbers in parentheses). (A) Cupressaceae (UWBM PB98590); (B) Piceoxylon (UWBM
PB98565); (C) Pinus sp. (UWBM PB98568); (D) cf. Berberis (UWBM PB98560) (E) cf. Robinia (UWBM
PB98587); (F) Quercus sp. (UWBM PB98591); (G) Quercus/Lithocarpus (UWBM PB98589); (H) Carya sp.
(UWBM PB98577); (I) cf. Salicaceae (UWBM PB98557); (J) Acer sp. (UWBM PB98558); (K) Ulmus sp.
(UWBM PB98559); (L) Undetermined hardwood #1 (UWBM PB98567); (M) Undetermined hardwood
#2 (UWBM PB98570); (N) Undetermined hardwood #3 (UWBM PB98576); (O) Undetermined
hardwood #4 (UWBM PB98556).
4. Discussion
4.1. Taphonomy and Sedimentology
The origin of the Bruneau Woodpile can be considered using a multiple working hypotheses
approach, considering two alternatives. Hypothesis 1: Fossilized wood originated from normal
processes of fluvial transport and lacustrine deposition. This scenario involves the wood and
sediment being delivered separately to the lake. They were, subsequently reworked into a single
bed. Hypothesis 2: Fossilized wood resulted from an unusual event, when limbs and wood
fragments from catastrophically damaged trees were transported and deposited with pumice and
obsidian in a single bed. These hypotheses are considered with respect to stratigraphic,
sedimentologic, and paleontologic features of the site.
A brief review of the depositional characteristics are in order. At Bruneau Woodpile, fossilized
limbs and small trunk fragments occur in the basal unit of a fining-upward sequence. This mode of
occurrence is very different from a typical lacustrine lahar deposit, where wood occurs in
fine-grained strata at the top of a graded sequence (Figure 14). At Bruneau Woodpile, the fossil
wood comprises limbs and small trunk fragments stripped of their bark—both types having abraded
driftwood-like shapes with tapered ends. This morphology is evidence that the wood had
undergone fluvial transport. Fossil wood is present in only one bed in the stratigraphic sequence at
the study site. Pollen and leaf fossils provide ample regional evidence of the existence of Miocene
forests in the hills bordering Chalk Hills Lake [4,5]. Hypotheses 1 and 2 involve different
assumptions regarding the processes that transported wood into the ancient lake. They also require
different diagenetic conditions that resulted in fossilization.
Figure 13.
Fossil wood types from the Bruneau Woodpile (University of Washington Burke
Museum specimen numbers in parentheses). (
A
)Cupressaceae (UWBM PB98590); (
B
)Piceoxylon
(UWBM PB98565); (
C
)Pinus sp. (UWBM PB98568); (
D
) cf. Berberis (UWBM PB98560) (
E
) cf. Robinia
(UWBM PB98587); (
F
)Quercus sp. (UWBM PB98591); (
G
)Quercus/Lithocarpus (UWBM PB98589);
(
H
)Carya sp. (UWBM PB98577); (
I
) cf. Salicaceae (UWBM PB98557); (
J
)Acer sp. (UWBM PB98558);
(
K
)Ulmus sp. (UWBM PB98559); (
L
) Undetermined hardwood #1 (UWBM PB98567); (
M
) Undetermined
hardwood #2 (UWBM PB98570); (
N
) Undetermined hardwood #3 (UWBM PB98576); (
O
) Undetermined
hardwood #4 (UWBM PB98556).
4. Discussion
4.1. Taphonomy and Sedimentology
The origin of the Bruneau Woodpile can be considered using a multiple working hypotheses
approach, considering two alternatives. Hypothesis 1: Fossilized wood originated from normal
processes of fluvial transport and lacustrine deposition. This scenario involves the wood and
sediment being delivered separately to the lake. They were, subsequently reworked into a single bed.
Hypothesis 2: Fossilized wood resulted from an unusual event, when limbs and wood fragments from
catastrophically damaged trees were transported and deposited with pumice and obsidian in a single
bed. These hypotheses are considered with respect to stratigraphic, sedimentologic, and paleontologic
features of the site.
A brief review of the depositional characteristics are in order. At Bruneau Woodpile, fossilized
limbs and small trunk fragments occur in the basal unit of a fining-upward sequence. This mode of
occurrence is very different from a typical lacustrine lahar deposit, where wood occurs in fine-grained
strata at the top of a graded sequence (Figure 14). At Bruneau Woodpile, the fossil wood comprises
limbs and small trunk fragments stripped of their bark—both types having abraded driftwood-like
shapes with tapered ends. This morphology is evidence that the wood had undergone fluvial transport.
Fossil wood is present in only one bed in the stratigraphic sequence at the study site. Pollen and
leaf fossils provide ample regional evidence of the existence of Miocene forests in the hills bordering
Chalk Hills Lake [
4
,
5
]. Hypotheses 1 and 2 involve different assumptions regarding the processes that
transported wood into the ancient lake. They also require different diagenetic conditions that resulted
in fossilization.

Geosciences 2017,7, 82 13 of 25
Geosciences 2017, 7, 82 13 of 25
Figure 14. Distal lacustrine lahar deposits in the Miocene Wilkes formation in southeast Washington,
USA [34]. (A) Streambank exposure reveals three successive wood mats (arrows). (B) These lahar
deposits contain rough-surfaced fragments of mummified wood in a wide range of sizes. Each wood
mat is preserved at the top of a fining-upward cycle.
Hypothesis 1. “ordinary deposition”
This model assumes that wood was gradually released from watershed forests, transported into
the lake by normal streamflow, and deposited in the lake over a long period of time. Wood is
considered to have been continually introduced into Chalk Hills Lake, rather than by a unique
high-energy sedimentary event. Fossilization of wood in the Bruneau Woodpile stratum is
considered to be the result of unique geochemical conditions; at other stratigraphic levels wood
originally incorporated in the sediments was destroyed during diagenesis.
The natural buoyancy of wood causes it to float until cellular spaces become water saturated,
resulting in a slow settling rate. Similarly, pumice settles only after vesicles become waterlogged,
which is a slow process for large clasts. Experimental evidence reveals that 4–20 mm diameter
pumice clasts require 50-100 hours of water exposure to become saturated [35]. The wood-bearing
layer contains pumice clasts up to 7 cm in diameter, a size that indicates these porous clasts
experienced prolonged moisture contact prior to their transport into the lake. At the Bruneau
Woodpile site, fossil wood occurs only in a single stratum, the lowest bed of a fining-upward
sedimentary cycle, where rhyolitic volcaniclastic material is the dominant constituent.
Although hypothesis 1 is described herein as “ordinary deposition”, a complex sequence of
processes is required to explain the origin of the Bruneau Woodpile bed. The presence of wood and
large pumice clasts in the basal bed requires both materials to have been water-saturated prior to
final deposition. Hypothesis 1 presumes that wood and pumice were independently deposited on
the lake bottom, subsequently intermixed with volcaniclastic sand during an episode of high-energy
sediment transport. To complete the genetic sequence, wood in this stratum became phosphatized;
conditions necessary for phosphatization did not exist in all of the other stratigraphic levels,
consequently the wood decomposed during diagenesis.
Hypothesis 2. “catastrophic deposition”
A very different model is that the wood-bearing stratum originated by an unusual event, when
watershed forests were damaged by a major storm, producing a subaerial reservoir of branches,
single limbs and wood fragments that became waterlogged prior to or during transport. This organic
debris arrived at the lake concurrently with volcaniclastic sediment, producing a density current.
The first materials to settle were coarse clastic grains, waterlogged pumice fragments, and
waterlogged woods. Smaller clastic grains settled at a slower rate, producing a fining-upward
sequence.
Hypothesis 2 involves a scenario where wood is introduced into Chalk Hills Lake as the result
of a catastrophic incident that released large volumes of woody debris. This scenario presumes an
event that resulted in severe structural damage to forests in the local watershed. One possibility is
the onset of botanical disease, akin to the pine bark beetle epidemic currently decimating large forest
area in western North America (Figure 15).
Figure 14.
Distal lacustrine lahar deposits in the Miocene Wilkes formation in southeast Washington,
USA [
34
]. (
A
) Streambank exposure reveals three successive wood mats (arrows); (
B
) These lahar
deposits contain rough-surfaced fragments of mummified wood in a wide range of sizes. Each wood
mat is preserved at the top of a fining-upward cycle.
Hypothesis 1. “ordinary deposition”
This model assumes that wood was gradually released from watershed forests, transported into
the lake by normal streamflow, and deposited in the lake over a long period of time. Wood is considered
to have been continually introduced into Chalk Hills Lake, rather than by a unique high-energy
sedimentary event. Fossilization of wood in the Bruneau Woodpile stratum is considered to be the
result of unique geochemical conditions; at other stratigraphic levels wood originally incorporated in
the sediments was destroyed during diagenesis.
The natural buoyancy of wood causes it to float until cellular spaces become water saturated,
resulting in a slow settling rate. Similarly, pumice settles only after vesicles become waterlogged,
which is a slow process for large clasts. Experimental evidence reveals that 4–20 mm diameter pumice
clasts require 50–100 h of water exposure to become saturated [
35
]. The wood-bearing layer contains
pumice clasts up to 7 cm in diameter, a size that indicates these porous clasts experienced prolonged
moisture contact prior to their transport into the lake. At the Bruneau Woodpile site, fossil wood
occurs only in a single stratum, the lowest bed of a fining-upward sedimentary cycle, where rhyolitic
volcaniclastic material is the dominant constituent.
Although hypothesis 1 is described herein as “ordinary deposition”, a complex sequence of
processes is required to explain the origin of the Bruneau Woodpile bed. The presence of wood and
large pumice clasts in the basal bed requires both materials to have been water-saturated prior to final
deposition. Hypothesis 1 presumes that wood and pumice were independently deposited on the lake
bottom, subsequently intermixed with volcaniclastic sand during an episode of high-energy sediment
transport. To complete the genetic sequence, wood in this stratum became phosphatized; conditions
necessary for phosphatization did not exist in all of the other stratigraphic levels, consequently the
wood decomposed during diagenesis.
Hypothesis 2. “catastrophic deposition”
A very different model is that the wood-bearing stratum originated by an unusual event, when
watershed forests were damaged by a major storm, producing a subaerial reservoir of branches, single
limbs and wood fragments that became waterlogged prior to or during transport. This organic debris
arrived at the lake concurrently with volcaniclastic sediment, producing a density current. The first
materials to settle were coarse clastic grains, waterlogged pumice fragments, and waterlogged woods.
Smaller clastic grains settled at a slower rate, producing a fining-upward sequence.
Hypothesis 2 involves a scenario where wood is introduced into Chalk Hills Lake as the result of
a catastrophic incident that released large volumes of woody debris. This scenario presumes an event

Geosciences 2017,7, 82 14 of 25
that resulted in severe structural damage to forests in the local watershed. One possibility is the onset
of botanical disease, akin to the pine bark beetle epidemic currently decimating large forest area in
western North America (Figure 15).
Geosciences 2017, 7, 82 14 of 25
Figure 15. Lodgepole Pine (Pinus contorta) killed by Mountain Pine Beetle (Dendroctonus), Montana,
USA. This beetle only attacks certain Pinaceae species [36].
Forest destruction caused by disease or insect infestation is inconsistent with the taxonomic
diversity of fossil woods at Bruneau Woodpile; plant pathogens and invading insects are typically
specific to a particular host. In addition, tree death does not directly result in stripping of branches,
which requires a subsequent storm event. However, some Bruneau Woodpile specimens show evidence
of insect and fungal attack (Figure 16); Fossilized bracket fungi have previously been reported [8,25,37];
these fungi are typically found on the trunks of dead or dying trees in modern forests.
Figure 16. Insect galleries are commonly preserved in specimens from Bruneau Woodpile. (A)
Galleries in limb wood just below the bark layer probably made by insect larvae; (B) Galleries
penetrating deeply into a hardwood limb; (C) close up view of insect gallery aligned subparallel to
annual rings.
An alternate possibility to a pest epidemic is that trees were damaged by an intense wind storm,
similar to the forest destruction caused by wind blast from the 1980 eruption of Mount Saint Helens
Figure 15.
Lodgepole Pine (Pinus contorta) killed by Mountain Pine Beetle (Dendroctonus), Montana,
USA. This beetle only attacks certain Pinaceae species [36].
Forest destruction caused by disease or insect infestation is inconsistent with the taxonomic
diversity of fossil woods at Bruneau Woodpile; plant pathogens and invading insects are typically
specific to a particular host. In addition, tree death does not directly result in stripping of branches,
which requires a subsequent storm event. However, some Bruneau Woodpile specimens show evidence
of insect and fungal attack (Figure 16); Fossilized bracket fungi have previously been reported [
8
,
25
,
37
];
these fungi are typically found on the trunks of dead or dying trees in modern forests.
Geosciences 2017, 7, 82 14 of 25
Figure 15. Lodgepole Pine (Pinus contorta) killed by Mountain Pine Beetle (Dendroctonus), Montana,
USA. This beetle only attacks certain Pinaceae species [36].
Forest destruction caused by disease or insect infestation is inconsistent with the taxonomic
diversity of fossil woods at Bruneau Woodpile; plant pathogens and invading insects are typically
specific to a particular host. In addition, tree death does not directly result in stripping of branches,
which requires a subsequent storm event. However, some Bruneau Woodpile specimens show evidence
of insect and fungal attack (Figure 16); Fossilized bracket fungi have previously been reported [8,25,37];
these fungi are typically found on the trunks of dead or dying trees in modern forests.
Figure 16. Insect galleries are commonly preserved in specimens from Bruneau Woodpile. (A)
Galleries in limb wood just below the bark layer probably made by insect larvae; (B) Galleries
penetrating deeply into a hardwood limb; (C) close up view of insect gallery aligned subparallel to
annual rings.
An alternate possibility to a pest epidemic is that trees were damaged by an intense wind storm,
similar to the forest destruction caused by wind blast from the 1980 eruption of Mount Saint Helens
Figure 16.
Insect galleries are commonly preserved in specimens from Bruneau Woodpile. (
A
) Galleries
in limb wood just below the bark layer probably made by insect larvae; (
B
) Galleries penetrating deeply
into a hardwood limb; (C) close up view of insect gallery aligned subparallel to annual rings.

Geosciences 2017,7, 82 15 of 25
An alternate possibility to a pest epidemic is that trees were damaged by an intense wind storm,
similar to the forest destruction caused by wind blast from the 1980 eruption of Mount Saint Helens
(WA, USA). During this explosive eruption, air blasts toppled thousands of trees, stripping away limbs
to leave a carpet of fallen trunks oriented by the wind direction (Figure 17A,B). In other more protected
areas, dead trees were left standing, with many intact branches. Subsequently, the branches and small
trunk fragments were removed, swept downslope by the combined effects of wind and rain, leaving
behind the parent logs (Figure 17C).
Interpretation of past geologic events can seldom be done with certainty, but the principle of
parsimony (“Occam’s razor”) potentially provides a useful tool. This principle states that among
competing hypotheses, the one with the fewest assumptions is preferred [38].
In our study, hypothesis 1 presumes the delivery of wood, pumice, and sandy sediment occurred
as independent events, with subsequent reworking of these materials producing a well-defined stratum
at the base of a fining-upward sedimentary cycle. If wood was routinely delivered to the lake by
streams, this wood underwent decomposition in all beds except for the Bruneau Woodpile stratum,
where the tissues instead became petrified. Compare this complex sequence to the relatively simple
“catastrophic” scenario of hypothesis 2: A major destructive event damages trees on watershed hill
slopes; Wood and pumice clasts become saturated from contact with moist soil or standing water.
A subsequent precipitation event causes streams to carry a pulse of water-saturated wood and pumice
to the lake to produce a density current. Suspended material settles on the lake floor as a fining upward
sediment cycle where wood and pumice are deposited in the basal bed. Fossil wood does not occur in
other strata because the transport of woody debris was not an ongoing process. Instead, this transport
was preceded by an unusual forest destruction event.
Geosciences 2017, 7, 82 15 of 25
(WA, USA). During this explosive eruption, air blasts toppled thousands of trees, stripping away
limbs to leave a carpet of fallen trunks oriented by the wind direction (Figure 17A,B). In other more
protected areas, dead trees were left standing, with many intact branches. Subsequently, the
branches and small trunk fragments were removed, swept downslope by the combined effects of
wind and rain, leaving behind the parent logs (Figure 17C).
Interpretation of past geologic events can seldom be done with certainty, but the principle of
parsimony (“Occam’s razor”) potentially provides a useful tool. This principle states that among
competing hypotheses, the one with the fewest assumptions is preferred [38].
In our study, hypothesis 1 presumes the delivery of wood, pumice, and sandy sediment
occurred as independent events, with subsequent reworking of these materials producing a
well-defined stratum at the base of a fining-upward sedimentary cycle. If wood was routinely
delivered to the lake by streams, this wood underwent decomposition in all beds except for the
Bruneau Woodpile stratum, where the tissues instead became petrified. Compare this complex
sequence to the relatively simple “catastrophic” scenario of hypothesis 2: A major destructive event
damages trees on watershed hill slopes; Wood and pumice clasts become saturated from contact
with moist soil or standing water. A subsequent precipitation event causes streams to carry a pulse
of water-saturated wood and pumice to the lake to produce a density current. Suspended material
settles on the lake floor as a fining upward sediment cycle where wood and pumice are deposited in
the basal bed. Fossil wood does not occur in other strata because the transport of woody debris was
not an ongoing process. Instead, this transport was preceded by an unusual forest destruction event.
Figure 17. Destruction of trees from May 18, 1980 eruption of Mount Saint Helens (WA, USA). (A,B)
“Blast zone” where fallen logs are oriented in the direction of wind created during explosive
eruption. In more protected areas some trees were left standing, and even on fallen trees fractured
limbs remained on the ash-covered ground surface; (C) Streams flowing down blast zone slopes
transported ash, pumice, and broken limbs, but lacked sufficient energy to move fallen logs. Photos
courtesy of United States Forest Service.
“Catastrophic hypothesis 2” is consistent with the Neogene geologic history of the Snake River
Plain. In late Miocene time, a series of rhyolite volcanic eruptions produced periodic influxes of
rhyolitic volcaniclastic debris and tephra. In this geologic setting, the distinction between
“ordinary’” and “catastrophic” processes is blurred; from the perspective of geologic time, “rare”
events may become relatively common. The sequence of late Miocene events that produced the
Figure 17.
Destruction of trees from May 18, 1980 eruption of Mount Saint Helens (WA, USA).
(
A
,
B
) “Blast zone” where fallen logs are oriented in the direction of wind created during explosive
eruption. In more protected areas some trees were left standing, and even on fallen trees fractured limbs
remained on the ash-covered ground surface; (
C
) Streams flowing down blast zone slopes transported
ash, pumice, and broken limbs, but lacked sufficient energy to move fallen logs. Photos courtesy of
United States Forest Service.
“Catastrophic hypothesis 2” is consistent with the Neogene geologic history of the Snake River
Plain. In late Miocene time, a series of rhyolite volcanic eruptions produced periodic influxes of

Geosciences 2017,7, 82 16 of 25
rhyolitic volcaniclastic debris and tephra. In this geologic setting, the distinction between “ordinary’”
and “catastrophic” processes is blurred; from the perspective of geologic time, “rare” events may
become relatively common. The sequence of late Miocene events that produced the unusual Bruneau
Woodpile petrified wood occurrence cannot be reconstructed with certainty, but the principle of
parsimony favors hypothesis 2. A simplified depiction of this scenario is shown in Figure 18.
Geosciences 2017, 7, 82 16 of 25
unusual Bruneau Woodpile petrified wood occurrence cannot be reconstructed with certainty, but
the principle of parsimony favors hypothesis 2. A simplified depiction of this scenario is shown in
Figure 18.
Figure 18. This reconstruction illustrates streams carrying woody debris and sediment to Chalk Hills
Lake in the late Miocene. Original artwork by Jessie Thorenson.
4.2. Mineralogy and Geochemistry
The first detailed description of francolite (fluorapatite) was provided by Sandell et al. [39],
though the mineral name was used as early as 1850. Apatite has previously been reported as a
common mineral in phosphatic sediments and fossils [40].
Studies of lacustrine phosphatic oolite in the Pliocene Glenns Ferry Formation, which overlies
the Chalk Hills Formation have demonstrated the presence of fluorapatite (“francolite”) as an
important constituent [41,42]. SEM/EDS analyses show that isolated fish vertebra in the
wood-bearing stratum at Bruneau Woodpile are composed of carbonate-fluorapatite.
4.3. Source of Phosphorus
Phosphorus has been described as the 10th [43] or 11th [44] most abundant element in the
Earth’s crust, but under pristine environmental conditions this element is typically present only at
low concentration. In the USA., background orthophosphate concentrations from 400 shallow wells
were found to average 0.03 mg/L [45]. Although, in regions affected by anthropogenic contributions:
agricultural fertilizer, leaking septic systems, and animal manure, the levels can be higher [46]. In
typical terrestrial settings phosphorus is a limiting factor for primary producers. Yet, phosphorus is
a macronutrient needed by all living organisms for making membranes, nucleic acids, and ATP-life’s
energy currency.
Phosphorus enters the food chain through primary producers who obtain their phosphorus
from both the residues of organisms and the weathering and erosion of phosphate-containing
minerals and rock, most of which owes its origin to ancient shallow marine environments. All
consumers obtain their phosphorus by eating other organisms. Thus, food webs allow for the
bioaccumulation of phosphorus. Fish obtain their phosphorus from ingesting other organisms;
seabirds that consume these fish produce guano that releases the accumulated phosphorus back into
the environment. Similarly, bats excrete guano that is phosphorus-rich because of their consumption
of insects. The cycling of phosphorus in the natural environment is complex, involving processes
where living organisms interact with and depend upon the rock cycle; the phosphorus cycle is a
topic that has paleontologic relevance. For example, Arena [47] suggested that some Australian
phosphatized wood owed its origin to the close proximity of caves that contained bat guano.
Both organic and inorganic sources of phosphorus are relatively rare in terrestrial environments
making the origin of phosphorus needed for the formation of nonmarime-phosphatized rocks or the
mineralization of phosphatized fossils in many localities enigmatic. Likewise, the origin of
carbonate-fluorapatite mineralized wood at the Bruneau Woodpile poses a mystery. It is tempting to
consider that the Bruneau area may have received phosphorus from regionally extensive bedrock
Figure 18.
This reconstruction illustrates streams carrying woody debris and sediment to Chalk Hills
Lake in the late Miocene. Original artwork by Jessie Thorenson.
4.2. Mineralogy and Geochemistry
The first detailed description of francolite (fluorapatite) was provided by Sandell et al. [
39
],
though the mineral name was used as early as 1850. Apatite has previously been reported as a common
mineral in phosphatic sediments and fossils [40].
Studies of lacustrine phosphatic oolite in the Pliocene Glenns Ferry Formation, which overlies the
Chalk Hills Formation have demonstrated the presence of fluorapatite (“francolite”) as an important
constituent [
41
,
42
]. SEM/EDS analyses show that isolated fish vertebra in the wood-bearing stratum
at Bruneau Woodpile are composed of carbonate-fluorapatite.
4.3. Source of Phosphorus
Phosphorus has been described as the 10th [
43
] or 11th [
44
] most abundant element in the
Earth’s crust, but under pristine environmental conditions this element is typically present only at low
concentration. In the USA., background orthophosphate concentrations from 400 shallow wells were
found to average 0.03 mg/L [
45
]. Although, in regions affected by anthropogenic contributions:
agricultural fertilizer, leaking septic systems, and animal manure, the levels can be higher [
46
].
In typical terrestrial settings phosphorus is a limiting factor for primary producers. Yet, phosphorus is
a macronutrient needed by all living organisms for making membranes, nucleic acids, and ATP-life’s
energy currency.
Phosphorus enters the food chain through primary producers who obtain their phosphorus from
both the residues of organisms and the weathering and erosion of phosphate-containing minerals and
rock, most of which owes its origin to ancient shallow marine environments. All consumers obtain their
phosphorus by eating other organisms. Thus, food webs allow for the bioaccumulation of phosphorus.
Fish obtain their phosphorus from ingesting other organisms; seabirds that consume these fish produce
guano that releases the accumulated phosphorus back into the environment. Similarly, bats excrete
guano that is phosphorus-rich because of their consumption of insects. The cycling of phosphorus
in the natural environment is complex, involving processes where living organisms interact with
and depend upon the rock cycle; the phosphorus cycle is a topic that has paleontologic relevance.
For example, Arena [
47
] suggested that some Australian phosphatized wood owed its origin to the
close proximity of caves that contained bat guano.

Geosciences 2017,7, 82 17 of 25
Both organic and inorganic sources of phosphorus are relatively rare in terrestrial environments
making the origin of phosphorus needed for the formation of nonmarime-phosphatized rocks or
the mineralization of phosphatized fossils in many localities enigmatic. Likewise, the origin of
carbonate-fluorapatite mineralized wood at the Bruneau Woodpile poses a mystery. It is tempting
to consider that the Bruneau area may have received phosphorus from regionally extensive bedrock
source, the Phosphoria Formation, which contains abundant Ca and P. These Permian marine rocks
are part of a regional stratigraphic sequence known as the Western Phosphate Field. The Phosphoria
Formation, which represents shallow subtidal to intertidal deposition, originating at least in part from
intertidal oolitic shoals, has been studied in detail [
48
–
51
]. Because present-day surface outcrops of the
Phosphoria Formation in Idaho are restricted to the eastern part of the state, it is doubtful this could
be the source of phosphorus in the deposits at Bruneau Woodpile. A massive fish kill could also act
as a source of phosphorus and while disarticulated fish remains are mixed with fossil wood at the
Bruneau Woodpile they are sparse.
4.4. Mineralization Process
Phosphatization of wood has received little study, but evidence from other research is useful for
understanding the mineralization processes. These occurrences reveal that phosphate mineralization
can occur in diverse geologic settings (Table 3).
Mineralogic aspects of phosphatized wood have received little previous attention, but more than
one pathway may be involved. This possible variation is reminiscent of wood silicification, which can
result from a variety of sequential processes [
61
,
62
]. SEM images of examples of apatite-mineralized
wood from other North American Cenozoic formations are shown in Figure 19. Briefly previewed here,
these phosphatized specimens are from localities that will be described in detail in a future report.
The specimen from the Virgin Valley Formation (NV, USA) (Figure 19A,B) comes from a late
Miocene lakebed deposit where opalized fossil wood is abundant; this phosphatized wood presumably
represents a highly localized geochemical anomaly. High magnification images show that the apatite
crudely preserves anatomical features of cell walls, but the network of relatively large crystals has
poor fidelity for fine details. This crystallinity suggests that the organic matter provided a favorable
chemical Eh/pH environment for inorganic precipitation of Ca
5
(PO
4
)
3
, but that molecule-by-molecule
organic templating was not involved. An alternate possibility is that original fine-grained apatite was
later subject to diagenetic recrystallization.
Phosphatized wood from Santa Fe River (FL, USA) (Figure 19C,D) has been fluvially transported
from the original locality, so the geologic origin is enigmatic. Like the Nevada wood, tracheid
boundaries and cell surface features are visible, but the coarse crystallinity obscures fine details. In this
Florida wood, apatite is commonly present as botryoidal encrustations made up of radiating crystals.
Table 3. Known phosphatized wood occurrences.
Location Age Setting Reference
Illinois, USA Jurassic marine [52]
Svalbard, Boreal Realm Jurassic marine [53,54]
Swindon, U.K. Jurassic marine [55]
Antarctica Cretaceous continental [56]
New Mexico, USA Cretaceous continental [57]
California, USA Eocene continental This report
Australia Miocene/Oligocene continental [47]
Nevada, USA Miocene continental This report
Idaho, USA Miocene continental This report
Uganda Miocene/Pliocene continental [58]
Florida, USA Neogene continental This report
France Pliocene/Pleistocene continental [59]
Pacific sea floor Holocene marine [60]

Geosciences 2017,7, 82 18 of 25
Geosciences 2017, 7, 82 18 of 25
Figure 19. Phosphatized wood from other Cenozoic localities. (A,B) Radial views of Miocene wood
from Virgin Valley Formation, northern Nevada, USA, mineralized with fluorapatite; (C,D) Neogene
fluorapatite wood from Sante Fe River (FL, USA); (E,F) Apatite crystals on carbonized cell walls,
Eocene, Yuba River, Nevada County (CA, USA). Specimens from G. Mustoe research collection,
apatite compositions determined by SEM/EDS.
A very different form of apatite occurs in Eocene wood from the Yuba River region in central
California (USA) (Figure 19E,F). Samples came from a large carbonized log found in alluvial
sediment. Cell walls consist of relict organic matter that has fused into a single homogenous layer.
Interior surfaces of these cells contain abundant hexagonal microcrystals. SEM/EDS spectra reveal
their composition as pure apatite. The location and morphologies of these crystals are evidence that
they originated by inorganic precipitation of Ca5(PO4)3 from groundwater that permeated the porous
wood long after the cell walls were carbonized. Their origin may have been a result of changes in
groundwater chemistry, such as in increase in phosphorus. Alternately, apatite crystallization may
have resulted in a change in Eh/pH conditions that favored precipitation.
Microbial processes may play a role in phosphatization of faunal remains and phosphatic
mineral deposits [63,64], but possible involvement of microbes during phosphatization of wood has
not been studied. Previous reports provide abundant evidence for the recognition of microbial
associations in aquatic mineral deposits where mineral deposition is facilitated by microbial mats or
biofilms. These include spring sinter [65–69] and clastic lake bed sediment [70–72]. An SEM image of
one Bruneau Woodpile fossil wood specimen exhibits etching on its surface; these traces are inferred
to have been made by a phosphorus-seeking endolith after the wood had already been petrified
(Figure 12D). Other fossil wood specimens in this study show no evidence of microbes.
Degradation of relict organic matter may have been important for controlling the Eh and pH
conditions that favored phosphate precipitation. Phosphatization of organic matter under geologic
conditions has primarily been studied in deposits that contain fossilized invertebrates, particularly
in occurrences were soft tissues are preserved [40,73–76]. These fossils typically were preserved in
marine sedimentary environments, likely explained by the greater abundance of phosphorus in ocean
water and sea floor sediment, compared to the low concentrations typical of terrestrial environments.
Furthermore, animal decay also adds a source of phosphate ions. Rapid burial, a reducing
environment and anaerobic decay are identified as necessary parameters for phosphatization of
animal soft tissue. Once these conditions have been met evidence suggests that a steep chemical
gradient resulting in a pH drop triggers the calcium carbonate—calcium phosphate switch, favoring
apatite precipitation [74]. Plant tissues are less likely to be phosphatized because they are relatively
phosphorus deficient [40]; however, numerous occurrences have been reported and the chemical
Figure 19.
Phosphatized wood from other Cenozoic localities. (
A
,
B
) Radial views of Miocene wood
from Virgin Valley Formation, northern Nevada, USA, mineralized with fluorapatite; (
C,D
) Neogene
fluorapatite wood from Sante Fe River (FL, USA); (
E
,
F
) Apatite crystals on carbonized cell walls,
Eocene, Yuba River, Nevada County (CA, USA). Specimens from G. Mustoe research collection, apatite
compositions determined by SEM/EDS.
A very different form of apatite occurs in Eocene wood from the Yuba River region in central
California (USA) (Figure 19E,F). Samples came from a large carbonized log found in alluvial sediment.
Cell walls consist of relict organic matter that has fused into a single homogenous layer. Interior
surfaces of these cells contain abundant hexagonal microcrystals. SEM/EDS spectra reveal their
composition as pure apatite. The location and morphologies of these crystals are evidence that they
originated by inorganic precipitation of Ca
5
(PO
4
)
3
from groundwater that permeated the porous
wood long after the cell walls were carbonized. Their origin may have been a result of changes in
groundwater chemistry, such as in increase in phosphorus. Alternately, apatite crystallization may
have resulted in a change in Eh/pH conditions that favored precipitation.
Microbial processes may play a role in phosphatization of faunal remains and phosphatic
mineral deposits [
63
,
64
], but possible involvement of microbes during phosphatization of wood
has not been studied. Previous reports provide abundant evidence for the recognition of microbial
associations in aquatic mineral deposits where mineral deposition is facilitated by microbial mats or
biofilms. These include spring sinter [
65
–
69
] and clastic lake bed sediment [
70
–
72
]. An SEM image of
one Bruneau Woodpile fossil wood specimen exhibits etching on its surface; these traces are inferred
to have been made by a phosphorus-seeking endolith after the wood had already been petrified
(Figure 12D). Other fossil wood specimens in this study show no evidence of microbes.
Degradation of relict organic matter may have been important for controlling the Eh and pH
conditions that favored phosphate precipitation. Phosphatization of organic matter under geologic
conditions has primarily been studied in deposits that contain fossilized invertebrates, particularly
in occurrences were soft tissues are preserved [
40
,
73
–
76
]. These fossils typically were preserved in
marine sedimentary environments, likely explained by the greater abundance of phosphorus in ocean
water and sea floor sediment, compared to the low concentrations typical of terrestrial environments.
Furthermore, animal decay also adds a source of phosphate ions. Rapid burial, a reducing environment
and anaerobic decay are identified as necessary parameters for phosphatization of animal soft tissue.
Once these conditions have been met evidence suggests that a steep chemical gradient resulting in a pH
drop triggers the calcium carbonate—calcium phosphate switch, favoring apatite precipitation [
74
].

Geosciences 2017,7, 82 19 of 25
Plant tissues are less likely to be phosphatized because they are relatively phosphorus deficient [
40
];
however, numerous occurrences have been reported and the chemical conditions necessary for their
formation may have important parallels to phosphatization of marine organisms.
One detailed experimental study [
77
] involved shrimp carcasses that were incubated in a marine
medium under aerobic conditions, introducing sulfate-reducing, sulfide-oxidizing and fermentative
bacteria. Levels of introduced sulfate, glucose, microbes and sediment were varied among individual
samples, as was buffering capacity of the culture medium. In most experiments, oxygen was quickly
depleted, pH decreased, and sulfate accumulated. The result was anaerobic decay, accompanied by
steep chemical gradients near the shrimp carcasses. In areas where decay was intense, pH decreased
markedly, and some muscle tissue was replaced by Ca
5
(PO
4
)
3
. In less decayed areas, CaCO
3
crystals
precipitated. Local pH appeared to be the most important factor for determining whether CaCO
3
or
Ca5(PO4)3was deposited.
These experimental results resemble observations of phosphatization of organic matter under
natural conditions. Phosphate-mineralized wood from the Pacific sea floor [
60
] appears to have
resulted from decomposition of organic matter that resulted in decreases in Eh and pH. Because
Eh and pH conditions that control solubility of CaCO
3
and Ca
5
(PO
4
)
3
are similar, high phosphorus
concentration in the water apparently favored apatite deposition. Ca
5
(PO
4
)
3
was deposited when pH
values were around 6.2. At higher pH, mineralization was dominated by CaCO3[60].
At the Bruneau Woodpile, all fossil wood is mineralized with carbonate-fluorapatite,
but phosphate minerals are not detectable in the surrounding sandstone matrix. This mode of
occurrence is evidence that unique geochemical conditions were present within the buried plant tissue.
The hyaloclastic lake sediment would have potentially provided an abundant source of dissolved
silica: pumice and volcanic ash are commonly associated with silicified wood, and experimental
studies have shown obsidian to be a good source of dissolved silica for wood mineralization [
78
].
The presence of diatom fossils in the lakebed sediment is evidence that dissolved silica was available,
and mineralization of the wood with calcium phosphate rather than silica is probable evidence that
that Eh/pH conditions within the buried tissue were not favorable for silica precipitation. This is
consistent with experimental evidence [
78
] demonstrating that when obsidian powder is used as
a silica source, dissolved Si levels were elevated at pH values > 9. For wood that contained absorbed
silica, a decrease in pH caused breakdown of hydrogen bonds between silica complexes and organic
molecules, releasing silica back into aqueous solution. These experiments suggest that the low pH
values typical of calcium phosphate precipitation would not have been favorable for silicification.
The absence of phosphate minerals in the matrix that encloses the fossil wood is an indication that
phospatization involved local geochemical gradients rather than more general diagenetic conditions.
The mode of phosphate mineralization is unclear. One possibility is that at the Bruneau Woodpile
carbonate-fluorapatite was deposited via process of organic templating, where deposition of calcium
phosphate was facilitated by an affinity with cell wall constituents, analogous to the hydrogen
bonds that form between silica and hydroxyl functional groups on organic wood molecules [
56
,
79
].
This organic templating possibility is consistent with SEM views of Bruneau wood, which show that
cell walls contain microcrystalline carbonate-fluorapatite that preserves anatomical details with many
vessel lumen remaining unmineralized (Figure 11).
4.5. Paleoclimate
The Cenozoic climate followed a general cooling trend largely attributable to changes in ocean
circulation patterns, but factors such as volcanic emissions and seabed methane releases also played
roles. The Paleocene-Eocene Thermal Maximum (PETM) and Pleistocene glacial advances and retreats
are notable examples of Cenozoic climatic fluctuations (Figure 20).

Geosciences 2017,7, 82 20 of 25
Geosciences 2017, 7, 82 20 of 25
Figure 20. Cenozoic average global temperatures based on oxygen isotope values in marine
sediments. Adapted from [80].
Neogene paleoclimatic conditions in southern Idaho have been studied using paleontologic
evidence. Plant fossils provide information about regional climatic trends, and fish fossils can be
used to study seasonality.
The taxonomic composition of extant taxa found within a fossil assemblage can be compared to
modern taxon-climate calibrated floras to infer past climate. In the Pacific Northwest, fossil floras
record climatic changes that resulted from the uplift of the Cascade Range in the late Miocene and
early Pliocene [5].
Middle to late Miocene-aged fossil floras between latitudes 41° N and 46° N in the states of
Washington and Idaho are composed of broadleaved hardwoods and conifers with some
assemblages including swamp elements such as Taxodium-type vegetation. These palaeofloras have
strong affinities to modern deciduous forests of eastern and southeastern North America as well as
eastern Asia. These modern analogs suggest that a summer-wet climate existed in the Snake River
Plain during the middle to late Miocene, with annual precipitation of >1000 mm, and warm winters,
and an estimated mean annual temperature (MAT) of 12–13 °C.
Deposition of the late Miocene Chalk Hills Formation, the host sediment for the Bruneau
Woodpile, occurred near the end of this climatic period and the fossil floras of this period indicate a
transition towards a more summer-dry climate regime. Primary evidence for this comes from the
Pickett Creek flora, which is in the lower Chalk Hills Formation and has been dated at
approximately 8.5–10.5 Ma based on geochemical correlation of volcanic ash layers present in the
deposit [6]. The Pickett Creek flora lacks many of the exotic taxa that require high humidity and
summer rainfall found in older middle Miocene floras of the Latah, Sucker Creek, and Mascall
Formations, with increasing dominance of more drought tolerant types such as pines, white oaks,
live oaks, and Robinia. The Bruneau Woodpile is thought to be slightly younger than Pickett Creek
and the wood types found at Bruneau Woodpile closely mirror the Pickett Creek assemblage. Even
though the growth of the Cascade Range had not yet created a significant rain shadow, there is
Figure 20.
Cenozoic average global temperatures based on oxygen isotope values in marine sediments.
Adapted from [80].
Neogene paleoclimatic conditions in southern Idaho have been studied using paleontologic
evidence. Plant fossils provide information about regional climatic trends, and fish fossils can be used
to study seasonality.
The taxonomic composition of extant taxa found within a fossil assemblage can be compared
to modern taxon-climate calibrated floras to infer past climate. In the Pacific Northwest, fossil floras
record climatic changes that resulted from the uplift of the Cascade Range in the late Miocene and
early Pliocene [5].
Middle to late Miocene-aged fossil floras between latitudes 41
◦
N and 46
◦
N in the states of
Washington and Idaho are composed of broadleaved hardwoods and conifers with some assemblages
including swamp elements such as Taxodium-type vegetation. These palaeofloras have strong affinities
to modern deciduous forests of eastern and southeastern North America as well as eastern Asia.
These modern analogs suggest that a summer-wet climate existed in the Snake River Plain during the
middle to late Miocene, with annual precipitation of >1000 mm, and warm winters, and an estimated
mean annual temperature (MAT) of 12–13 ◦C.
Deposition of the late Miocene Chalk Hills Formation, the host sediment for the Bruneau Woodpile,
occurred near the end of this climatic period and the fossil floras of this period indicate a transition
towards a more summer-dry climate regime. Primary evidence for this comes from the Pickett Creek
flora, which is in the lower Chalk Hills Formation and has been dated at approximately 8.5–10.5 Ma
based on geochemical correlation of volcanic ash layers present in the deposit [
6
]. The Pickett Creek
flora lacks many of the exotic taxa that require high humidity and summer rainfall found in older
middle Miocene floras of the Latah, Sucker Creek, and Mascall Formations, with increasing dominance
of more drought tolerant types such as pines, white oaks, live oaks, and Robinia. The Bruneau Woodpile
is thought to be slightly younger than Pickett Creek and the wood types found at Bruneau Woodpile

Geosciences 2017,7, 82 21 of 25
closely mirror the Pickett Creek assemblage. Even though the growth of the Cascade Range had not
yet created a significant rain shadow, there is evidence to suggest that the climate of the northwest
interior was becoming more arid towards the late Miocene [81].
Pliocene-aged fossil floras in the Snake River Plain indicate a continued transition to a more arid
climate, resulting from the rain shadow created by the newly-uplifted Cascade Range. These younger
paleofloras show the decline of hardwoods, and the appearance of abundant grasses and dryland
shrubs, particularly Artemisia (sagebrush). By the early Pleistocene, the transition to modern
sagebrush/bunchgrass steppe flora was well under way [5].
Miocene and Pliocene fish faunas that populated Lake Idaho and its Miocene predecessors can
be used as proxy sources for climatic parameters based upon their distributions and the limiting
temperature for nearest living relatives. During the late Miocene, the cold month mean temperature is
estimated to be 10
◦
C, compared to 0
◦
C for the Pliocene, and a reduction in the number of frost-free
days (<0
◦
C) from 270 to 175. The seasonal range or difference between the mean for coldest and
warmest month increased from 11 ◦C during the late Miocene to 21 ◦C during the Pliocene [20].
In summary, evidence from previous studies of fossil leaf, pollen and fish fossils suggests that
the trees that produced the Bruneau Woodpile came from forests that existed during the climatic
transition that occurred near the close of the Miocene Epoch. This plant community represents
an assemblage that contains more summer-dry adapted taxa than the earlier mid-Miocene floras of the
region; very different from the arid conditions that exist today.
5. Conclusions
The Bruneau Woodpile is an unusual deposit, providing a rare opportunity to collect abundant
specimens of fossil wood mineralized with carbonate-fluorapatite. The taxonomic diversity of the
specimens provides a glimpse of an ancient forest growing in a region that was on the brink of climatic
change. The region would soon transition from warm temperate to semi-arid. One long-standing
enigma of this locality is the abundance of wood in the form of small limbs and trunk fragments,
with no larger specimens. The driftwood-like morphology of the specimens is evidence of fluvial
transport. As described above, two possible hypotheses are considered. Hypothesis 1 assumes
that the wood-bearing stratum resulted from a sequence of ordinary processes, where wood was
continuously introduced to the lake, but only preserved in a single stratum, where it was intermixed
with waterlogged pumice and volcaniclastic sediment. Hypothesis 2 invokes a catastrophic event that
severely damaged local forest, producing a reservoir of branches and wood fragments that gradually
became water saturated from soil moisture or in shallow ponds. This material was later transported to
the lake during a storm event when a high-energy stream flow simultaneous carried pumice, sand,
and silt. This suspended sediment entered Chalk Hills Lake as a turbidity current, the source of the
fining-upward sediment cycle that contains wood in the coarse basal bed. The principle of parsimony
favors hypothesis 2.
Chemical conditions in the deposit caused mineralization of the wood by replacement with
carbonate fluorapatite. While the exact mechanism could not be determined, the process may have
been similar to the silicification of wood in which the wood acts as a template for the mineralization
process. Microbes do not appear to have been directly responsible for phosphatization, but inorganic
phosphate precipitation may have been the result of Eh/pH gradients that were strongly influenced
by the decomposition of relict organic matter.
Supplementary Materials:
The following are available online at www.mdpi.com/2076-3263/7/3/82/s1. Atlas of
wood types from the Bruneau Woodpile, with transverse section images and anatomical descriptions.
Acknowledgments:
Rick Dillhoff and Mary Klass assisted with excavation and specimen collecting.
Elisabeth Wheeler provided helpful advice for interpreting photomicrographs. Nancy and Larry Dillon provided
guidance in how to work the fossil site and provided some samples Richard Rantz contributed specimens collected
in 1997 for initial chemical analyses. Yuba River, California USA specimens (Figure 19) provided by David Lawler,
Farwest Geoscience Foundation Director. Jessie Thoreson provided the (Figure 18) sketch. We thank Barbara Nash
for determining the age of the ash.

Geosciences 2017,7, 82 22 of 25
Author Contributions:
The investigation was conceived and directed by Viney. Link provided stratigraphic
data and age determinations. Link, Viney, Dillhoff participated in field work; taxonomic identifications were
provided by Dillhoff. Mustoe provided graphic art., photomicrographs, SEM/EDS data, XRD and XRF analyses.
The manuscript was written by Mustoe and Viney, with contributions from the other coauthors.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1.
Lewis, R.S.; Link, P.K.; Stanford, L.R.; Long, S.P. Geological Map of Idaho; Idaho Geological Survey Map 9;
scale 1:750,000; Idaho Geological Survey: Moscow, ID, USA, 2012.
2.
Kimmel, P.G. Stratigraphy, age, and tectonic setting of the Miocene-Pliocene lacustrine sediments of the
western Snake River Plain, Oregon and Idaho. In Cenozoic Geology of Idaho; Bonnichsen, B., Breckenridge, R.M.,
Eds.; Idaho Bureau of Mines and Geology Bulletin: Moscow, ID, USA, 1982; Volume 26, pp. 559–578.
3. Leopold, E.B.; Wright, V.C. Pollen profiles of the Plio-Pleistocene transition in the Snake River Plain, Idaho.
In Late Cenozoic History of the Pacific Northwest; Smiley, C.J., Ed.; Pacific Division, American Association for
the Advancement of Science: Washington, DC, USA, 1985; pp. 323–348.
4.
Leopold, E.B.; Denton, M.F. Comparative age of grassland and steppe east and west of the northern Rocky
Mountains. Ann. Mo. Bot. Gard. 1987,74, 841–867. [CrossRef]
5.
Mustoe, G.E.; Leopold, E.B. Paleobotanical evidence for the post-Miocene uplift of the Cascade Range. Can. J.
Earth Sci. 2014,51, 809–824. [CrossRef]
6.
Buechler, W.K.; Dunn, M.T.; Rember, W.C. Late Miocene Pickett Creek flora of Owyhee County, Idaho.
Univ. Mich. Contrib. Mus. Paleontol. 2007,31, 305–362.
7. Root, J.V. Bruneau wood. Gems Miner. 1971,409, 36–37.
8.
Brown, R.W. A bracket fungus from the late Tertiary of southwestern Idaho. J. Wash. Acad. Sci.
1940
,30,
422–424.
9.
Axelrod, D.I. Cenozoic History of Some Western American Pines. Ann. Mo. Bot. Gard.
1986
,73, 565–641.
[CrossRef]
10.
Pierce, K.L.; Morgan, L.A. The track of the Yellowstone hot spot: Volcanism, faulting, and uplift. In Regional
Geology of Eastern Idaho and Western Wyoming; Link, P.K., Kuntz, M.A., Platt, L.B., Eds.; Geological Society of
America Memoir: Boulder, CO, USA, 1992; Volume 179, pp. 1–53.
11.
Rodgers, D.W.; Ore, H.T.; Bobo, R.T.; McQuarrie, N.; Zentner, N. Extension and subsidence of the eastern
Snake River Plain, Idaho. In Tectonic and Magmatic Evolution of the Snake River Plain Volcanic Province;
Idaho Geological Survey Bulletin 30; Bonnichsen, B., White, C.M., McCurry, M., Eds.; Idaho Geological
Survey: Moscow, ID, USA, 2002; pp. 121–155.
12.
Shervais, J.W.; Kauffman, J.D.; Gillerman, V.S.; Othberg, K.L.; Vetter, S.K.; Hobson, V.R.; Zarnetske, M.;
Cooke, M.F.; Matthews, S.H.; Hanan, B.B. Basaltic volcanism of the central and western Snake River Plain:
A guide to field relations between Twin Falls and Mountain Home, Idaho. In Interior Western United States:
Geological Society of America Field Guide 6; Pederson, J., Dehler, C.M., Eds.; Geological Society of America:
Boulder, CO, USA, 2005; Volume 6, pp. 27–52.
13.
Bonnichsen, B.; Godchaux, M.M. Late Miocene, Pliocene, and Pleistocene geology of southwestern Idaho
with emphasis on basalts in the Bruneau-Jarbidge, Twin Falls and Western Snake River Plain regions.
In Tectonic and Magmatic Evolution of the Snake River Plain Volcanic Province; Bonnichsen, B., White, C.M.,
McCurry, M., Eds.; Idaho Geological Survey Bulletin 30; Idaho Geological Survey: Moscow, UD, USA, 2002;
pp. 233–312.
14.
Malde, H.E. Stratigraphy of the Glenns Ferry from Hammett to Hagerman, Idaho; U.S. Geological Survey Bulletin
1331-D; United States Government Printing Office: Washington, DC, USA, 1972; 19p.
15.
Hagerman Fossil Beds National Monument, Idaho. Available online: http://www.nps.gov/hafo/
(accessed on 30 April 2017).
16.
Wood, S.H.; Clemens, D.M. Geologic and tectonic history of the western Snake River Plain, Idaho and Oregon.
In Tectonic and Magmatic Evolution of the Snake River Plain Volcanic Province; Bonnichsen, B., White, C.M.,
McCurry, M., Eds.; Idaho Geological Survey Bulletin 30; Idaho Geological Survey: Moscow, ID, USA, 2002;
pp. 69–103.

Geosciences 2017,7, 82 23 of 25
17.
Jenks, M.D.; Bonnichsen, B.; Godchaux, M.M. Geologic Map of the Grand-View Bruneau Area, Owyhee County,
Idaho; Digitally Revised Version, Idaho Geological Survey Technical Report 98-1; Idaho Geological Survey:
Moscow, ID, USA, 1998.
18.
Malde, H.E.; Powers, H.A. Geologic Map of the Glenns Ferry-Hagerman Area, West-Central Snake River Plane
Idaho; scale 1:48,000; Map I-696; U.S. Geological Survey Miscellaneous Investigations; U.S. Geological Survey:
Reston, VA, USA, 1972.
19.
Smith, G.R.; Swirydczuk, K.; Kimmel, P.G.; Wilkinson, B.H. Fish biostratigraphy of the late Miocene to
Pleistocene sediments of the western Snake River Plain, Idaho. In Cenozoic Geology of Idaho; Bonnichsen, B.,
Breckenridge, R.M., Eds.; Idaho Bureau of Mines Geology Bulletin: Moscow, ID, USA, 1982; Volume 26,
pp. 519–542.
20.
Smith, G.R.; Patterson, W.P. Mio-Pliocene seasonality on the Snake River plain: Comparison of faunal and
oxygen isotopic evidence. Palaeogeogr. Palaeoclimatol. Palaeoecol. 1994,107, 291–302. [CrossRef]
21.
Ruez, D.R., Jr. Middle Pliocene paleoclimate in the Glenns Ferry Formation of Hagerman Fossil Beds
National Monument, Idaho: A baseline for evaluating faunal change. J. Ida. Acad. Sci. 2006,42, 1–16.
22.
Perkins, M.E.; Brown, F.H.; Nash, W.P.; McIntosh, W.; Williams, S.K. Sequence, Age, and Source of Silicic Fallout
Tuffs in Middle to Late Miocene Basins of the Northern Basin and Range Province; Geological Society of America
Bulletin: Boulder, CO, USA, 1998; Volume 110, pp. 344–360.
23.
Panshin, A.J.; DeZeeuw, C. Textbook of Wood Technology, 3rd ed.; McGraw-Hill: New York, NY, USA,
1970; 705p.
24. Wheeler, E.A. InsideWood—A web resource for hardwood anatomy. IAWA J. 2011,32, 199–211.
25. Kirkby, R. Bounty of Bruneau—Fossil treasures in southern Idaho. Gems Miner 1998,28, 52–55.
26. Nash, B.; University of Utah, Salt Lake City, UT, USA. Personal communication, 2015.
27.
Morgan, L.A.; McIntosh, W.C. Timing and Development of the Heise Volcanic Field, Snake River Plain, Idaho,
Western USA; Geological Society of America Bulletin: Boulder, CO, USA, 2005; Volume 117, pp. 288–306.
[CrossRef]
28.
Anders, M.H.; Saltzman, J.; Hemming, S.R. Neogene Tephra Correlations in Eastern Idaho and Wyoming
Yellowstone for Hotspot-Related Volcanism and Tectonic Activity; Geological Society of America Bulletin: Boulder,
CO, USA, 2009; Volume 121, pp. 837–856.
29.
Cohen, K.M.; Finney, S.C.; Gibbard, P.L.; Fan, J.-X. The ICS International Chronostratigraphic Chart. Episodes
2013,36, 199.
30.
Elliot, J.C. Structure and Chemistry of the Apatites and Other Orthophosphates. Studies in Inorganic Chemistry 18;
Elsevier: New York, NY, USA, 1994; 371p.
31.
Hughes, J.M.; Rakovan, J. The crystal structure of apatite Ca
5
(PO
4
)
3
(F, OH, Cl). In Reviews in Mineralogy and
Geochemistry; Kohn, M.J., Rakovan, J., Hughes, J.M., Eds.; Mineralogical Society of America: Washington,
DC, USA, 2002; Volume 48, pp. 1–12.
32.
Godinot, C.; Tribolet, A.; Grover, R.; Ferrier-Pagès, C. Bioerosion by euendoliths decreases in
phosphate-enriched skeletons of living corals. Biogeosciences 2012,9, 2377–2384. [CrossRef]
33.
Golubic, S.; Friedmann, I.; Schneider, J. The lithobiontic ecological niche, with special reference to
microorganisms. J. Sediment. Res. 1981,5, 475–478.
34.
Yancey, T.M.; Mustoe, G.E.; Leopold, E.B.; Heizler, M.T. Mudflow disturbance in latest Miocene forests in
Lewis County, Washington. PALAIOS 2013,28, 343–358. [CrossRef]
35.
White, J.D.; Manville, V.; Wilson, C.J.N.; Houghon, B.F.; Riggs, N.R.; Ort, M. Settling and deposition of AD
181 Taupo pumice in lacustrine and associated environments. In Volcaniclastic Sedimentation in Lacustrine
Settings; White, J.D.L., Riggs, N.R., Eds.; Blackwell Publishing Ltd.: Oxford, UK, 1999; pp. 141–150.
36.
Furniss, R.L.; Carolin, V.M. Western Forest Insects; U.S. Department of Agriculture, Forest Service,
Miscellaneous Publication: Washington, DC, USA, 1977; No. 1339; 346p.
37. Andrews, H.N.; Lenz, L.W. Fossil polypores from Idaho. Ann. Mo. Bot. Gard. 1947,34, 113–114. [CrossRef]
38.
Gauch, H.G. Scientific Method in Practice; Cambridge University Press: Cambridge, UK, 2003; ISBN
0-521-01708-4.
39.
Sandell, E.B.; Hay, M.H.; McConnell, D. The composition of francolite. Mineral. Mag.
1939
,25, 395–401.
[CrossRef]
40.
Lucas, J.; Prévôt, L.E. Phosphates and fossil preservation. In Taphonomy: Releasing the Data Locked in the
Fossil Record; Allison, P.A., Briggs, D.E.G., Eds.; Plenum Press: New York, NY, USA, 1991; pp. 389–409.

Geosciences 2017,7, 82 24 of 25
41.
Swirydczuk, K.; Wilkinson, B.H.; Smith, G.R. The Pliocene Glenns Ferry oolite: Lake-margin carbonate
deposition in the southwestern Snake River plain. J. Sediment. Res. 1979,49, 114–1129.
42.
Swirydczuk, K.; Wilkinson, B.H.; Smith, G.T. Synsedimentary lacustrine phosphorites from the Pliocene
Glenns Ferry Formation of southwestern Idaho. J. Sediment. Res. 1981,51, 1205–1214.
43.
Wedepohl, K.H. The composition of the continental crust. Geochim. Cosmochim. Acta
1995
,59, 1217–1232.
[CrossRef]
44.
Fleischer, M. Recent Estimates of the Abundances of Elements in the Earth’s Crust; U.S. Geological Survey
Circular 285; USGS: Reston, VA, USA, 1953; pp. 1–3.
45.
Dubrowsky, N.M.; Burrow, K.R.; Clark, G.M.; Gronberg, J.M.; Hamilton, P.A.; Hitt, K.J.; Mueller, D.K.;
Munn, M.D.; Nolan, B.T.; Puckett, L.J.; et al. The quality of our Nation’s waters—Nutrients in the Nation’s streams
and groundwater, 1992–2004; U.S. Geological Survey Circular 1350; USGS: Reston, VA, USA, 2010; 174p.
46.
Dunne, T.; Leopold, L.B. Water in Environmental Planning, 1st ed.; W.H. Freeman: New York, NY, USA, 1978.
47.
Arena, D.A. Exceptional preservation of plants and invertebrates by phosphatization, Riversleigh, Australia.
PALAIOS 2008,23, 495–502. [CrossRef]
48.
Sheldon, R.P.; Warner, M.A.; Thompson, M.E.; Pierce, H.W. Stratigraphic Sections of the Phosphoria Formation in
Idaho, 1949. Part 1; U.S. Geological Survey Circular 304; USGS: Reston, VA, USA, 1949.
49.
McKelvey, V.E.; Armstrong, R.A.; Gulbrandsen, R.A.; Campbell, R.M. Stratigraphic sections of the Phosphoria
Formation in Idaho, 1947–48, Part 2; Geological Survey Information Circular 301; USGS: Reston, VA, USA, 1953.
50.
Petrun, R.M. Field guide to the southeast Idaho phosphate district. In Guidebook to the Geology of Eastern Idaho;
Hughes, S.S., Thackray, G.D., Eds.; Idaho Museum of Natural History: Moscow, ID, USA, 1999; pp. 269–280.
51.
Piper, D.Z.; Link, P.K. An Upwelling Model for the Phosphoria Sea: A Permian, Ocean-Margin Sea in the
Northwest United States. Am. Assoc. Petroleum Geol. Bull. 2002,86, 1217–1235.
52.
Gluskoter, H.J.; Pierard, L.H.; Pfefferkorn, H.W. Apatite petrifacations in Pennsylvanian shales of Illinois.
J. Sediment. Res. 1970,40, 1363–1366. [CrossRef]
53.
Krajewski, K.P. Organic geochemistry of a phophorite to black shale transgressive succession: Wilhelmøya
and Janusfjellet Formations (Rhætian-Jurassic) in Central Spitsbergen, Arctic Ocean. Chem. Geol.
1989
,11,
331–334.
54.
Reolid, M.; Philippe, M.; Nagy, J.; Abad, I. Preservation of phosphatic wood remains in marine deposits of the
Brentskardhaugen Bed (Middle Jurassic) from Svalbard (Boreal Realm). Facies
2010
,56, 549–566. [CrossRef]
55.
Scott, A.C.; Collinson, M.E. Non-destructive multiple approaches to interpret the preservation of plant fosils:
Implications for calcium-rich permineralizations. J. Geol. Soc. 2003,160, 857–862. [CrossRef]
56.
Jefferson, T.H. The preservation of conifer wood: Examples from the Lower Cretaceous of Antarctica.
Paleontology 1987,30, 232–249.
57.
Sweeney, I.J.; Chin, K.; Hower, J.C.; Budd, D.A.; Wolfe, D.G. Fossil wood from the middle Cretaceous
Moreno Hill Formation: unique expressions of wood mineralization and implications for the process of
wood preservation. Int. J. Coal Geol. 2009,79, 1–17. [CrossRef]
58.
Pailler, D.; Flicoteaux, R.; Ambrosi, J.P.; Médus, J. Les bois fossils mio-pliocènes de Nkondo (Lac Albert,
Ouganda), composition minéralogique et mode de formation. C R Acad. Sci. Paris Sci. Terre et Planets
2000,331, 279–286.
59.
Koeniguer, J.C. Sur une liane plio-quaternaire du Tchad. Bulletin du Muséum National d’Histoire Naturelle,
Sciences de la Terre 1973,172, 81–91.
60.
Goldberg, E.D.; Parker, R.H. Phosphatized wood from the Pacific sea floor. Bull. Geol. Soc. Am.
1960
,71,
631–632. [CrossRef]
61.
Mustoe, G.E. Late Tertiary petrified wood from Nevada, USA: Evidence of multiple silicification pathways.
Geosciences 2015,5, 286–309. [CrossRef]
62.
Viney, M.; Dietrich, D.; Mustoe, G.; Link, P.; Lampke, T.; Götze, J.; Rößler, R. Multi-stage silicification of
Pliocene wood: Re-examination of an 1895 Discovery from Idaho, USA. Geosciences 2016,6, 21. [CrossRef]
63.
Crosby, C.H.; Bailey, J. The role of microbes in the formation of modern and ancient phosphatic mineral
deposits. Front. Microbiol. 2012,3, 241–247. [CrossRef] [PubMed]
64.
Omelon, S.; Ariganello, M.; Bonucci, E.; Grynpas, M.; Nanci, A. A review of phosphate mineral nucleation in
biology and geobiology. Calcif. Tissue Int. 2013,93, 382–396. [CrossRef] [PubMed]
65.
Jones, B.; Renaut, R.W.; Rosen, M.R. Biogenicity of silica precipitation around geysers and hot-spring vents,
North Island, New Zealans. J. Sediment. Res. 1997,67, 88–104.

Geosciences 2017,7, 82 25 of 25
66.
Walter, M.R.; McKloughlin, S.; Drinnan, A.N.; Farmer, J.D. Paleozoic hot spring sinter in the Drumond Basin,
Queensland, Australia. Econ. Geol. 1998,84, 135–143.
67.
Konhauser, K.O.; Phoenix, V.R.; Bottrell, S.H.; Adams, D.G.; Head, I.M. Microbial-silica interactions in
Icelandic hot spring sinter: Possible analogues for some Precambrian iron stromatolites. Sedimentology
2001,48, 425–433. [CrossRef]
68.
Guidry, S.A.; Chafetz, H.S. Depositional facies and diagentic alteration in a relict siliceous hot-spring
accumulation: Examples from Yellowstone National Park, U.S.A. J. Sediment. Res.
2003
,73, 806–823.
[CrossRef]
69.
Hinman, N.W.; Walter, M.R. Textural preservation in siliceous hot spring deposits during early diagenesis:
Examples from Yellowstone National Park and Nevada, U.S.A. J. Sediment. Res.
2005
,75, 200–215. [CrossRef]
70.
Knoll, A.H. Exceptional preservation of photosynthetic organisms in silicified carbonates and silicified peats.
Philos. Trans. R. Soc. Lond. 1985,B311, 111–122. [CrossRef]
71.
Harding, I.C.; Chant, L.S. Self-cemented diatom mats as agents of exceptional fossil preservation in the
Oligocene Florissant lake beds, Colorado, United States. Geology 2000,28, 195–198. [CrossRef]
72.
O’Brien, N.R. The Role of Biofilm in Fossil Preservation in the Late Eocene Florissant Forest, Colorado; Geological
Society of America Special Paper 435; Geological Society of America: Golden, CO, USA, 2008; pp. 19–31.
73.
Briggs, D.E.K. The role of decay and mineralization in the preservation of soft-bodied fossils. Annu. Rev.
Earth Planet. Sci. 2003,31, 275–301. [CrossRef]
74.
Briggs, D.E.H.; Wilby, P.R. The role of the calcium carbonate-calcium phosphate switch in the mineralization
of soft-bodied fossils. J. Geol. Soc. Lond. 1996,153, 665–668. [CrossRef]
75.
Briggs, D.E.G.; Kear, A.J.; Maretill, D.M.; Wilby, P.R. Phosphatization of soft tissue in experiments and fossils.
J. Geol. Soc. Lond. 1993,150, 1035–1038. [CrossRef]
76.
Wilson, P.; Parry, L.A.; Vinther, J.; Edgecomb, G.D. Unveiling biases in soft-tissue phosphatization: Extensive
preservation of musculature in the Cretaceous (Cenomanian) polychaete Rollinschaeta myoplena (Annelida:
Amphinomidae). Palaeontology 2016,59, 463–479. [CrossRef]
77.
Saggemann, J.; Bale, S.J.; Briggs, D.G.; Parkes, R.J. Controls on formation of authigentic minerals in association
with decaying organic matter: An experimental approach. Geochim. Cosmochim. Acta
1999
,63, 1083–1095.
[CrossRef]
78.
Ballhaus, C.; Gee, C.T.; Bockrath, C.; Greef, K.; Mansekldt, T.; Rhede, D. The silicification of trees in volcanic
ash—An experimental study. Geochim. Cosmochim. Acta 2012,84, 62–74. [CrossRef]
79. Leo, R.F.; Barghoorn, E.S. Silicification of Wood. Bot. Mus. Leafl. Harv. Univ. 1976,25, 1–47.
80.
Zachos, J.C.; Pagani, M.; Sloan, L.; Billups, K.; Thomas, E. Trends, rhythms, and aberrations in global climate
65 Ma to present. Science 2001,292, 686–693. [CrossRef] [PubMed]
81.
Graham, A. Late Cretaceous and Cenozoic History of North American Vegetation; Oxford University Press:
New York, NY, USA, 1999.
©
2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).