Content uploaded by M. D. Elson
Author content
All content in this area was uploaded by M. D. Elson on Apr 17, 2017
Content may be subject to copyright.
Society for American Archaeology
Lava, Corn, and Ritual in the Northern Southwest
Author(s): Mark D. Elson, Michael H. Ort, S. Jerome Hesse, Wendell A. Duffield
Source:
American Antiquity,
Vol. 67, No. 1 (Jan., 2002), pp. 119-135
Published by: Society for American Archaeology
Stable URL: http://www.jstor.org/stable/2694881
Accessed: 13/09/2010 18:16
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=sam.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to
American Antiquity.
http://www.jstor.org
LAVA, CORN, AND RITUAL IN THE NORTHERN SOUTHWEST
Mark D. Elson, Michael H. Ort, S. Jerome Hesse, and Wendell
A. Duffield
Fifty-five pieces of lava with impressions of prehistoric corn have recently been recoveredfrom
NA 860, a small habitation site
near
Sunset
Crater Volcano in northern
Arizona. Archaeological, geological,
and botanical information suggest that husked
ears of corn were deliberately placed
in the lava's
path when
the
volcano erupted in
the
mid-to-late
eleventh century A.D. Over
40 kg of basalt lava containing the hardened corn casts were then taken to NA 860 located
4 km away from the lava flow. At
the
site,
the rocks underwent lithic reduction
to
expose
the casts. We
suggest
that these "corn rocks" are indicative
of
ritual
practices, perhaps serving
as an
offering
made to
appease
the
forces responsible for
the
eruption.
Although both prehistoric
and modern
offerings
are
commonly
associated with volcanoes in other
parts of
the
world,
this
is
the first evidence from
the
Southwest United States
of possible
ritual behavior related to volcanism.
Cincuenta y
cinco
pedazos
de
lava con
impresiones
de
mafz prehistoricofueron
descubiertos
recientemente
en NA
860,
un
pequeno
sitio
de
habitacion cerca
del
volcan Sunset
Crater en el noroeste de
Arizona. Informacion arqueol6gica,
geol6gica, y botanica
sugieren que
las
mazorcas
de
mafzfueron
intencionalmente
colocadas en
la
trayectoria
de
la
lava cuando el
volcan hizo erupcion
en
la
segunda mitad del siglo
XI
d.
C. Mds de 40
kg.
de lava basdltica conteniendo
impresiones
de mafz se recuperaron en
el
sitio
NA
860,
a 4 km
de
distancia
de
la lava.
En
este
sitio,
las
rocasfueron trabajadas para exponer
las impresiones. Sugerimos que
estas 'rocas
de
maiz'
indican
practicas rituales, y que quizas
estas rocas sirvieron como
ofrendas
para apaciguar
a las
fuerzas
responsables por
la erupci6n.
Aunque las ofrendas asociadas con volcanes son comunes
en otras partes del mundo, tanto en la
prehistoria
como en
tiempos modernos,
esta es la
primera
evidencia de
practicas
rituales asociadas con vulcanismo
que
se ha
recuperado en el suroeste de los
Estados
Unidos.
T he eruption of the Sunset Crater
Volcano in
the mid-to-late eleventh century A.D.
has long
played a prominent role in archaeological
models of northern Arizona prehistory. Starting
with
the
work
of Dr. Harold
S. Colton
and his
colleagues
at the Museum of Northern Arizona (MNA)
in the
early 1930s (Colton 1932a) and continuing
into the
modern period (Anderson 1990; Downum
1988;
Pilles 1979, 1987; Sullivan 1984), the
eruption has
played
a
significant part
in
interpretations
of prehis-
toric settlement and subsistence
in
the
Flagstaff
region. For example, to explain
the
increase
in site
density
at the end of the eleventh century, Colton
(1932a, 1946) proposed
that
thin
layers
of
deposited
cinders acted as
a water-retaining
mulch, allowing
previously infertile areas
to be
farmed. In
Colton's
(1949:24) view,
the
opening up
of new lands resulted
in large-scale migrations
(a prehistoric
"land rush")
that dramatically
changed the nature
of the groups
living in the Flagstaff
area. Colton's
theory has been
adopted with some
modifications,
by more recent
researchers, who
propose that
the cinder mulch was
an important factor
in the initial settlement
and sub-
sequent population
growth
evident in
Wupatki
National Monument
(Downum and Sullivan
1990:87-88).
Although other
models suggest
that it was not the
cinder mulch that allowed new
lands to
be farmed,
but changing climatic conditions (Pilles
1979:479-480;
1996:65),
all researchers
agree
that
the
eruption
of Sunset Crater
had
an
enormous
impact on
the
prehistoric
inhabitants
of
the
Flagstaff
area, and probably
the greater northern
Southwest.
Some 900
years
later, Hopi
accounts
of the
eruption
Mark D. Elson
*
Desert Archaeology, Inc., 3975
N. Tucson Blvd., Tucson,
AZ 85716
Michael H. Ort
*
Departments
of Environmental
Sciences and Geology,
P.O. Box 4099, Northern
Arizona University,
Flagstaff, AZ 86011
S. Jerome
Hesse * SWCA
Inc.,
343 S. Scott Ave., Tucson,
AZ 85701
Wendell A. Duffield
*
Department
of
Geology,
P.O.
Box
4099,
Northern Arizona
University,
Flagstaff, AZ 86011
American
Antiquity, 67(1), 2002, pp.
119-135
Copyright(
2002 by
the
Society for
American
Archaeology
119
120 AMERICAN
ANTIQUITY [Vol.
67, No. 1, 2002]
/ ~~WUPATKI
NATIONAL MONUMENT k
. . . . . . . . . ... ..i.
U.S.
89
Project
Area
. ) I (<OpLek~ ~ ~ ~~~~ry ,
NA
860 Bonito Kana'a
Flow
I if7 \ ~~~~~~SUNSET
CRATER
SAN FRANCISCO
PEAKS
<<t . r \ . Qtez
~~~~~~~~~Flag
./ Miles
0 5
Kilometers
Flagstaff / -
10 N
WALNUT CANYON
* .NAT. MON.
Figure
1.
Location
of the
U.S.
89
project
area,
Sunset
Crater,
and the Corn Rock
site (NA 860). The San Francisco
volcanic
field encompasses
all
of
this
map
between
Interstate
40
and
Wupatki
National Monument
and extends approximately
45
km
west
of
the
map.
Map drafted
by
Susan
Hall.
are still
passed from generation
to generation
as a
part of traditional
clan
knowledge,
underscoring
the
significance
of this event
(Colton 1932b;
Ferguson
and Lomaomvaya 2000; Malotki and
Lomatuway'ma
1987).
This
paper reports
analyses of 55
pieces of Sun-
set Crater
basalt
lava
containing
impressions
of pre-
historic
corn cobs ("corn
rocks"). These
rocks were
recovered from the surface
of an archaeological
site,
NA 860,
located approximately
4 km west of
the
closest
lava
flow
(Figure
1).1
Another
948
pieces of
Sunset
Crater basalt
without corn impressions
("unimpressed
Sunset
clasts")
were also
recovered
from this
site. Physical
and chemical
evidence
indi-
cates that
over 40 kg of corn
rocks and unimpressed
Sunset
clasts were removed
from
Sunset
Crater and
transported
to
NA 860. We
suggest that
the corn
rocks are
indicative
of ritual behavior,
with
the corn
perhaps
serving
as
an
offering
to appease
the forces
responsible
for the
eruption.
The removal
of
the corn
REPORTS 121
. .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~m
:. .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~O
.~~ ~~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1
.:
*~~~~~~~~~~~~~~~~
:. : ..::. ..:. . .... .
* :|: .. .. ::. . .~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~..
.. . ..
*. o~ ~~ ~~ ~~ ~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~5
*:::: e . : ... IS
..~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 7 :.. ls:-
* | - | ->S > .. . | -~~~~~~~~~~~~~~~~~~~~~~~~~"
rocks to
a site
away from
the
flow
suggests
that
they
too
contained
special qualities.
Although
both
pre-
historic
and modem
offerings
are
commonly
asso-
ciated
with volcanoes in
other
parts
of
the world
(Luhr
and
Simkin
1993;
Plunket
and
Urufnuela
1998a,
1998b;
Scarth
1999;
Sigurdsson
1999),
this
is the
first
evidence from
the
Southwest
United
States
of
possible
ritual
behavior related to
volcanism.
Geological
Setting
The
southern
margin
of the
Colorado Plateau is
marked
by late
Cenozoic
basaltic
volcanic
fields,
including the
Grand
Canyon, San Francisco,
Springerville, and Zuni-Bandera volcanic fields. The
San Francisco volcanic field covers an area of
nearly
5000
kM2 in
north-central
Arizona.
Its
600 cinder
cones are
broadly basaltic
in
composition;
the
field
also contains five
large, intermediate-to-silicic cen-
ters (Amos
1996:1;
Holm
1986:27-28).
The
period
of
volcanism ranges from the late Miocene to late
Holocene, with the younger
vents, like Sunset Crater,
in
the northeast (Colton
1967). Recurrence intervals
for basaltic
eruptions
were about
3,000 years,
although
that
rate
appears
to have slowed
in the
past
250,000 years (Tanaka et
al. 1986).
Sunset
Crater,
located about 25
km
northeast of
122 AMERICAN ANTIQUITY [Vol. 67, No. 1, 2002]
Flagstaff (Figures
1
and 2), produced the only erup-
tion in the Southwest United States indisputably wit-
nessed by surrounding prehistoric populations
(Colton 1932a, 1937).
Two
other Southwest
erup-
tions are now dated to
the
period
of human
occupa-
tion: the McCartys flow near Grants, New Mexico,
radiocarbon dated ca. 3160-3200
B.P.
(Laughlin
et
al. 1993:12; Mabery et al. 1999:41), and the Little
Springs
flow near the north
rim of
the
Grand Canyon,
recently dated using cosmogenic Helium at 1.3
?
0.5
ka (Fenton
et
al. 2001). Although
it
is
possible
that
both
of
these areas
were
occupied
at the time of the
eruptions, the impact
on local
populations
is cur-
rently unknown.
Sunset Crater can
be
characterized
as a cinder-
cone
volcano,
which
unlike
stratovolcanoes such
as
Mount
St.
Helens, generally
do not
produce large
explosions. Instead, they
start
as a small fault or crack
in
the earth and grow
into
cone-shaped
features
through fountaining
lava that throws
spatter, cinders,
and ash hundreds to thousands
of meters into the air.
Slow-moving
lava
emanates from vents near the base
of the
cone.
The visibility
of
the eruption
was calculated
using
digital elevational modeling
based
on the
estimated
minimum and maximum
heights (as derived
by
Amos
1986)
of
the ash plume (8
km and 16
km)
and
lava fire fountain (260
m
and 660 m). Using only the
minimum
values
as a conservative estimate,
on
a
clear
day
the ash
plume
could have been seen
from
high points near
Palm
Springs (California),
Las
Vegas
(Nevada), Durango (Colorado),
west-central New
Mexico,
and
along
the
Arizona-Mexico border.
The
fire
fountain,
which
particularly
at
night
would have
been
spectacular,
was visible across a much smaller
area,
but
still
could
have
been seen from
points
in
southern
Utah,
and
east, west,
and central
Arizona,
including
the
top
of the Bradshaw Mountains some
75
km
north of Phoenix.
Areas
of
significant pre-
historic
population,
such as Chaco
Canyon
and the
Phoenix
Basin,
were within
sight
of
the
minimum
ash
plume. Although
it is
not
known how the
erup-
tion affected
people living
outside the
immediate
Flagstaff area,
most
groups living
in
the Southwest
in the
mid-to-late eleventh century
A.D. would have
been well aware that
something very
unusual was
occurring.
The
significance
of
the
eruption
is
underscored
by
the fact
that, although
the San Francisco volcanic
field
contains over 600 cinder
cones,
the
last
major
eruption in the
Flagstaff area prior
to Sunset Crater
probably occurred
before humans
first entered the
North American
continent. Furthermore,
the erup-
tion was likely
foreshadowed by ominous
signs. Evi-
dence from modern
cinder-cone volcanoes
indicates
that earthquakes
occur for several
weeks or months
prior to an
eruption, often increasing
in
frequency
and
magnitude
through time. For
example, at
a cin-
der-cone volcano
very similar to
Sunset Crater that
erupted
in 1943 near the town
of Paricutin
in
Michoacain,
Mexico, the first noticeable
earthquake
occurred 45 days prior to the eruption.2
A
week later,
daily earthquakes
were occurring,
increasing to 25
to 30 a day the
week before the eruption.
Finally, on
the
day
before the eruption, 300
earth tremors were
felt
(Luhr
and Simkin 1993:3;
Yokoyama and De la
Cruz-Reyna
1990).
During
an
eruption, the cloud
of ash and steam
creates its own weather
system.
Thunder claps and
lightning around
the volcano are
common occur-
rences. At
Paricutin,
which
was
surrounded by
sev-
eral
small
villages
of
Tarascan farmers, nobody
was
killed by the lava
or cinder
fall
itself,
but three peo-
ple and a number
of cattle and horses
were killed by
lightning strikes
(Luhr and Simkin
1993:7; Nolan
1979:328).
Additionally,
the
heavy
ash and cinder
fall from Sunset
Crater, along
with the
smoke from
accompanying
forest
fires,
must
have
darkened the
daytime sky,
while
at
night
the horizon would have
glowed a fiery
red. Such volcanoes
can grow at an
amazingly fast
rate;
at
Paricutin,
the cinder
cone
grew
to a height
of 167 m in six days (Rees
1979:251). By
the end of the first
year, Paricutin
had
reached
336
m
(Luhr
and Simkin
1993:8),
or
80
per-
cent of its
final
height, suggesting
that within
a sim-
ilar
time
frame, Sunset Crater
may
have been close
to 300
m
high
and 2
km2
in area.
There is no doubt
that to
the
people
living
in
the
Flagstaff area,
the
erup-
tion
of Sunset Crater was an
awe-inspiring sight.
Sunset Crater
lava flowed from
two
separate pri-
mary vents,
covering an
area of ca. 8
km2
to
depths
ranging
between
2 and 30
m
(Figure
1).
The Bonito
flow extended
approximately
2
km
west
of Sunset
Crater. Given
the
high density
of
prehistoric
settle-
ment
in
areas
just
outside the flow,
it is
likely
that
numerous sites
were buried
(Downum
and Gumer-
man
1998:8;
Pilles
1978, 1979).
In
addition,
the
erup-
tion
spewed
ash and cinders over
an area of about
2,300
km2. Ash and cinder fall were heaviest
in
areas
closest to
the
volcano; deposits
as
thick as
40
cm were
REPORTS 123
found in
the
fill
of a pithouse
at a site
5.5 km south-
west of the
volcano (Swartz
and Elson
2001), and
deposits
over 1.5
m in depth have
been recorded
closer to the cone (Amos
1986:Figure
12). Signifi-
cant accumulations
occurred
at distances
as great
as
20 km,
such as at Wupatki
National
Monument,
where Sunset
Crater cinder
deposits
between 5
and
10 cm thick
have
been recorded (Hooten
and
Ort
2000;
Hooten et al.
2001:15).
The cinder and ash
fall, along with
the forest
fires,
must have
disrupted prehistoric
life
to
a
significant
degree.
At
Panrcutin,
for
example,
an
estimated
4,500
cattle
and
550
horses died
from
breathing
the fine
ash emitted early
in the
eruption (Rees
1979:259).
At Sunset
Crater, it is
likely that the
ash fall detri-
mentally
affected
human
health, particularly
of the
very young
and elderly,
and the continual
smoke
from forest fires would
have added
to air pollution
problems.
Sunset
Crater lies
in
a
relatively arid, pine-
forested
region,
receiving
less than 50 cm
of
pre-
cipitation
per year.
Forest
fires
would have spread
far
beyond
the area
directly impacted
by
the
lava,
burning
homes
and
agricultural
fields. The
fires
would
have caused at least temporary
displacement
of both human
and animal
populations.
Data from modem
eruptions
indicate that 10 cm
of
ash
is
enough, particularly
when
wet,
to collapse
modem roofs
(Blong
1984).
The
Paricutin data
(Rees
1979:280-286),
along with agricultural
experiments
undertaken
by
Mahle
(1963),
Colton
(1965),
and
recently
by Waring (2001),
further
show that corn
germination
rates are
significantly
reduced with as
little as 15 cm of ash
and
cinder cover,
and that
corn
seeds
will not
germinate
in ash
depths greater
than
25 cm
(see
also
Sheets
1980: 10) (silica-rich
ash does
not contain
enough
nutrients, particularly
nitrogen
and
phosphorus,
to allow
for successful
plant prop-
agation.)
A map
of
Sunset Crater
ash and
cinder
thickness indicates
that areas
significantly
impacted
by
these
deposits (greater
than 25 cm
in
depth)
extended over
a minimum
of
450
km2
(Hooten
and
Ort
2000).
It is almost
certain that the inhabitants
of
these
areas moved elsewhere
following
the
eruption.
The
eruption
has
been dated
by
dendrochronol-
ogy, paleomagnetic
secular
variation,
and archaeo-
logical
association.
Although
a date of
A.D. 1064
for
the initial
eruption
has become
extremely
entrenched
in
both the
archaeological
and
geological
literature,
the
dating
is
not conclusive
(Boston
1995:22).
The
most
commonly
accepted
date comes
from
Smiley
(1958),
who tree-ring
dated
several
beams
used in
constructing
the
100-room
pueblo
of Wupatki, 20
km
northeast
of Sunset
Crater.
A few of Smiley's
(1958:190)
Douglas-fir
and
ponderosa
pine
samples
showed
suppressed
rings
after A.D. 1065,
which he
interpreted
as dating
the eruption
to
the winter
of A.D.
1064-65.
However,
many
factors other
than
volcanic
eruptions
can affect
tree-ring
growth,
including
local-
ized drought,
proximity
to
a forest
fire, extreme
cold,
and even insect
infestation (Boston
1995:35-37).
In
addition, the provenance
of the
analyzed
trees is not
known;
Wupatki,
at around
1,495
m asl, is
too low
to
support
the
growth
of
pine
or
fir,
both today and
in the
prehistoric
past
(Sullivan
and
Downum
1991:272).
Therefore,
the trees
could have come
from
anywhere
in the
nearby higher-elevation
areas,
possibly
from the pine-covered
plateau
that was not
heavily
affected
by
ash and cinder
fall from the
erup-
tion. Perhaps
most
significantly,
in the hundreds
of
tree-ring samples
examined
since
Smiley's
work that
span
this period,
none
has
contained
this
unique sig-
nature
(Boston
1995:37;
Jeffrey
S. Dean,
personal
communication 2001).
However,
a comparison
of
dated archaeological
sites
with and without Sunset
Crater ash
does seem to constrain
the initial
eruption
to sometime
in the mid-to-late eleventh
centuryA.D.
(Breternitz
1967;
Colton
1945; see
reviews
in
Boston
1995:22-54
and Downum
1988:222-249).
The overall
length
of the eruption
is also contro-
versial. Paleomagnetic
results allow
an
eruptive
period
of
perhaps
100
years
and
maybe
as
long
as
200
years
(Champion
1980;
Pilles
1979;
Shoemaker
and Champion
1977),
but more volcanic deposits
must
be
sampled
to reduce
error
ranges
and confm
this.
Furthermore,
while definitive
archaeological
evidence
for
cinder
deposits
in
the
A.D. 1050-1150
period
is
relatively
common,
similar evidence
from
the later
part
of the twelfth
century
is
ambiguous
and
difficult to
interpret.
Although
cinder-cone
eruptions
do not
typically
last
for more than
a few decades
(Panrcutin,
for
example,
lasted
just
9 years),
there are
several
well-documented
historic-period
eruptions
lasting
over 100
years.
Archaeological
Setting
The corn-rock
pieces
were found
on the surface
of
a small habitation
site located
approximately
6
km
west
of the
Sunset
Crater Volcano
and 4 km west
of
the Bonito lava
flow,
the flow
closest to the
site
(Fig-
ure
1).
The
site, designated
by
MNA
as NA 860,
124 AMERICAN ANTIQUITY [Vol. 67, No. 1,
2002]
covers an area ca. 12,265
m2
and
sits on top of
an
old Brunhes-age (ca. 600-200 thousand years old)
cinder cone (Moore and Wolfe 1987). Cut into
the
red scoria of the cinder cone are four deep pithouse
depressions and six shallow depressions that are also
most
likely pithouses;
a
single
small
masonry
room
is also present (Figure 3, Feature
1). Most corn-
impressed material was found
in
a 64
m2
concentra-
tion
in
the north-central
portion
of
the site.
NA 860
was
originally
recorded
by
Colton
in
1928 (Colton 1932c).
In
1930,
MNA archaeologist
Lyndon Hargrave fully
excavated one of the
deep
pit-
house depressions (Feature
11
[Pithouse A]),
and
tested three others
(Feature
2
[Pithouse D],
Feature
5 [Pithouse C], and Feature 9 [Pithouse B]). Neither
Colton
nor
Hargrave
noted the
presence
of corn rocks
or
unimpressed pieces
of Sunset
Crater basalt;
the
only remaining
records
from
these
investigations
are
10
pages
of
typed
and
handwritten
notes and a few
sketch maps (Hargrave 1930).
Colton
(1946:79-81) later used Hargrave's
notes to write a brief site description. Hargrave
found
that Feature
11
(Pithouse A),
the
fully
exca-
vated
structure,
had been
remodeled,
based on the
presence
of two ventilator shafts
along
with sev-
eral post supports that
were sealed underneath a
later clay floor.
An
intrusive
flexed
burial
was also
found
in
the structure
fill.
Significantly, Hargrave
encountered
pockets
of
fine black
ash and cinders
in
the fill extending down to a depth of 30
cm
above
the
house
floor, suggesting
that
the house had
already
been abandoned and
partially
filled at the
time of the
eruption.
Cinders were also found
in
a
few of the
postholes, possibly
the result
of
post
decay
or disturbance
factors, although
no
cinders
were
present
on
the structure
floor itself. The
great-
est concentration
of
cinders
was found
in
associa-
tion with the intrusive
burial,
"as
though
cinders
lying
on the surface were
mixed with earth at
[the]
filling
of
[the] grave" (Hargrave 1930:2),
indicat-
ing
that
the
site area continued
to be used follow-
ing
the
eruption.
Hargrave
does not comment
on
whether
cinders
were found
in
the test units of the other houses.
How-
ever,
at the
time
of the
investigations,
the
significance
of
the
cinders and
their
relation to the Sunset Crater
eruption
were
not
known
(Colton 1946:80),
so it
is
possible
that these details were omitted from the
report.
No
tree-ring
dates were
recovered,
but based
on the
ceramic
assemblage,
Colton
(1946:81)
assigned the
site to the period
between A.D.
900 and
1050.
In 1998,
NA 860 was reinvestigated
by Desert
Archaeology,
Inc. personnel,
who conducted
surface
collection and
detailed mapping;
no subsurface
exca-
vation
was
undertaken.3
The area of concentrated
corn rocks
in
the north-central
portion of the
site was
mapped
and collected in 25
2-x-2-m
units
(100 mi2),
although the
concentration
itself was confined
to an
area of
approximately
8
m
by 8
m
within
this.
All
Sunset Crater lava (with and
without
corn impres-
sions) was
collected from the
concentration,
result-
ing
in
the recovery
of 988
items. The remainder
of
the site area was
surface collected
in
20-x-20-m units.
Within each
20-x-20-m
unit,
all ceramic and
lithic
artifacts
were
recovered
from four 2-m-wide
tran-
sects spaced
5
m
apart,
resulting
in
a 40 percent
sam-
ple. Upon completion
of the
transects, the remainder
of the unit
was searched for
diagnostic
ceramics,
lithic tools,
and pieces of Sunset
Crater basalt,
result-
ing
in
the recovery
of
100
percent of these
artifact
classes. Because
of
the very
intensive search
for all
Sunset
Crater
basalt
rocks,
15
additional
specimens
were
found in the site area
away
from the concen-
tration. The locations
of
these
artifacts were
indi-
vidually
mapped (Figure
3).
The surface ceramic
assemblage
recovered by
DesertArchaeology contained
just
over
1,200
sherds;
diagnostic
ceramics
were
dominated by
the
San Juan
Redware
type
Deadmans Black-on-red
(A.D.
750-1075)
and the
Tusayan
Whiteware
type
Black
Mesa Black-on-white
(A.D. 1025-1150).
Based on
these
data,
Colton's
dating
of NA 860 can be
refined
to A.D.
1025-1075, placing
the
later
end
of
the occu-
pation
contemporaneous
with the mid-to-late
eleventh century
date for the
Sunset
Crater
eruption.
Additional
support
for this
dating
comes from
the
nearby site
of
Medicine
Fort
(NA 862),
which has a
similar ceramic
assemblage
and
tree-ring
dates
span-
ning
the A.D. 1040s to
1060s,
with
a likely
con-
struction date
between
A.D.
1058
and 1063
(Colton
1946:81-84;
Robinson et
al. 1975:59).
Hargrave's
excavation
of a burial intermixed
with cinders
in
the
fill
of
a
pithouse
at NA
860, along
with
evidence
that
the house had been
abandoned
prior
to the
eruption,
clearly
indicates both
pre-
and
post-eruption
occu-
pation,
although
it is not known if the
site was con-
tinuously
used
throughout
this time.
Occupation
of
the site
following
at least
one
eruptive period
is fur-
ther
supported
by
our
recovery
of a corn rock embed-
REPORTS 125
i ... X F 392 contour Tntefwol - 10
feet~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
........
1350 0101
9 ,A
)corn
rock
concentrot
onll fal lF
1
2 )
131
Possible possibee
.~~ ~ ~ .~~/l~eoture feature-
Fl 0~~~~
COCONINO
COUN,. ARIZONA
CORN ROCK SITE
F3 ~~~~~~~~~contour
intervol 10 feet
0 50 m
Mopping
ond
digitol
cartography by
GEO-MAP,
Inc.
2001
EXPLANATION
\ ?~~~~~ .~~ point
located corn rock
site boundary IA Hargrove's
designotion
(1930)
--- extent of wall faill
Qdepression
Figure
3. Plan
map
of the Corn Rock
site,
NA 860.
Computer cartography
by
Geo-Map,
Inc.
ded
in
the wall of a
masonry structure, as discussed
below.
The
Corn-Rock Assemblage
The corn-rock
assemblage consists of 1,003 pieces
of Sunset Crater basalt
weighing 40,132.5 g,
recov-
ered from the
surface
of
NA
860. The fresh,
dark
gray
Sunset
Crater
basalt was
easily distinguished
from
the red scoria
of
the
underlying Brunhes-age
cinder
cone.
Of
the
1,003
pieces,
55
(5.5 percent)
have corn
impressions. An
unknown number of other rocks
with corn
impressions
have been removed
from the
site
by collectors.
The great
majority
of the Sunset Crater basalt
(988
or 98.5
percent)
was
concentrated
in a 64
m2
area on the northern
edge
of the site
(Figure 3).
Four-
teen additional
pieces
of
Sunset Crater
basalt were
collected from
the surface
of the
general
site
area.
A
single piece, which
contains the best-formed corn
impressions (Figure
4),
was recovered
from
the north
wall of a masonry structure
(Feature 1) located
approximately
15
m south of
the
corn-rock concen-
tration. This piece was slightly
embedded in the
ground and covered by vegetation,
which probably
spared it from collectors.
The recovered pieces of
Sunset Crater basalt have
dark reddish-black outer
surfaces
commonly with
dark gray broken interior
surfaces. They are irregu-
lar and angular in shape,
measuring approximately
1
cm to about 20 cm in length.
All
rocks collected
are aphanitic, having no
visible phenocrysts upon
inspection
with the naked eye. Spherical to slightly
ellipsoidal vesicles compose
as much as 30 percent
of the volume
of each rock. Most individual vesicles
range from less than
1
mm
to approximately 10 mm
in diameter;
some
irregularly
shaped vesicles
are
20-30
mm
in
length
and are the
result
of
conjoining
vesicles.
In
total, the Sunset
Crater rocks weighed
an
average
of 40.0
g,
with a range
of
less than
1
g
to
1,744.9 g (Table 1). There
is a distinct
difference,
however,
in
weight
and therefore size of the
rocks
with corn
impressions (the
"corn
rocks")
as com-
pared
to
those without
corn
impressions (the
"unim-
pressed Sunset clasts").
The
938 clasts
without
126 AMERICAN
ANTIQUITY [Vol.67,No.1,2002]
4L .40~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Z?:<eL.e<<
4L~p-r {
'^
b _.
...,,\-
OP~iNt,.D '-f
*_f_ j~~~~~~~~~~~~~~~~~~~~~~~~~P
-_, *1 4
{t_~~A4
I St 'ff' *t'' '
*~~~~~~~~~~~~~~~ppp
REPORTS 127
Table 1.
Sunset Crater Basalt
Recovered From NA
860.
Mean Range 1
Standard
Rock
Type No. Weight (g) Weight (g) Deviation
(g)
Unimpressed Clasts 938 29.3 1.0-645.8 35.2
Corn Rocks 55 194.4 2.9-1,744.9 287.1
Indeterminate
Impressions 10 193.6 3.0-702.6 217.3
Total 1,003 40.0 1.0-1,744.9 87.6
impressions were much
smaller, averaging 29.3 g.
In
contrast, the 55 rocks
containing corn impressions
were more than
six
times
larger, averaging 194.4 g.
An
additional
10
rocks
contained only indeterminate
impressions, but these were
very similar
in
size to
the
corn
rocks.
The
55
corn
rocks
contained a total of
110
corn
impressions. The impressions
are located on con-
cave,
relatively vesicle-free
surfaces and
are
of vari-
able
quality, showing
a wide
range
of
detail.
The
number
of
corn impressions on
each
rock
ranged
between 1 and
14; although
most contained only
1
or 2
impressions. Corn
impressions
were recorded
as one of three
types:
positive, negative,
or
husk.
Positive
impressions were
formed from husked corn
with intact
kernels, negative
impressions
were
formed from
husked
corn
with
the
kernels either
deliberately
removed
or
unformed,
while husk
impressions
were from the
surrounding sheaf. Inde-
terminate
impressions
were also
recorded,
but
they
were
not
included as corn
impressions
in the
totals.
These
impressions
were not clear
enough
to deter-
mine their
origin
with
confidence; they could be
from
corn,
from other
vegetal
material,
or
could have
formed
through
other
processes.
Positive
impres-
sions were the most common
type:
79 positive
impressions
were recorded on 47
rocks,
with a mean
of 1.4
(s = 1.6) impressions
per rock;
28
negative
impressions
were found on 24
rocks,
with
a mean
of
.5
(s = .6) per rock;
and
3 husk
impressions
were
found,
all on
the same rock. Most rocks
(63.6 per-
cent)
contained
only
a
single
impression type,
while
the others contained
various
combinations of
posi-
tive,
negative,
and indeterminate
impressions.
Corn
cob,
kernel,
and husk casts occur within and
on the outer surface
of the rock. The rock
fragment
recovered from
the wall of the
masonry
structure
contains the most
impressions
and shows the
great-
est detail
(Figure 4): 10
positive,
1
negative,
and 3
husk
molds are
clearly
present, along
with 2 inde-
terminate
molds.
In
this
specimen,
lava
penetrated
the
spaces between individual
kernels, preserving
the texture of a husked cob with remarkable detail.
Some cobs are cracked, and the lava penetrated along
these cracks, which implies a very fluid, hot lava. In
the kernel and cob casts, a few vesicles break the sur-
face,
whereas in the husk
casts, the impressions are
generally unbroken. One husk cast contains a single,
large (1-cm diameter) vesicle. We interpret this to
indicate a sudden lava quenching by the corn, pre-
serving
the texture of the
lava
and not
allowing
it to
anneal. No organic material was found
in
the casts,
indicating that the corn was fully incinerated after
the impression was set. Similar textures are seen in
recent flows at Kilauea Volcano near Kalapana,
Hawai'i, where large fruits
from the
pandamus tree
dropped
into and
quickly quenched
the
lava, leaving
a
clearly
defined mold but no
organic
remains
(Fig-
ure 5). Many of the other corn-impressed rocks have
much less detailed impressions. The most common
positive impression
is that of an inverted
bumpy
sur-
face that was
probably
formed
by
lava
solidifying
on
the surface of
a cob
but not
penetrating the space
between
individual kernels.
The rock
fragments with
more detailed
impressions
are
larger, denser, and less
vesicular
and
with more broken surfaces than those
containing less detailed impressions.
The
irregular shape
and
relatively
small size of
the corn rocks and
unimpressed
clasts
suggest
that
they
are most
likely
clots
of solidified
spatter. Spat-
ter bombs are
fluid
pyroclasts of basaltic lava
that
are
ejected
out of cinder cone and
spatter
cone vol-
canoes and
smaller,
flow-fed hornitos
(Figure 6).
In
spatter
cones
and
hornitos,
lava flies a short distance
through
the air
and
then
lands, forming
a 1-to-40
m
high cone-shaped spatter rampart. Many
of the recov-
ered clasts and
corn rocks have
a
smooth "sheen" to
them, commonly
seen where hot lava
rock has been
exposed
to radiant heat from
flowing lava,
such
as
around hornitos.
At
least one side of some of the
unimpressed
Sun-
set clasts has
.5-
to 1-cm diameter cinders fused
to
it. This surface is
typically flattened,
as if the
lava
had conformed to a
flat,
cinder-covered surface. Sun-
128 AMERICAN
ANTIQUITY [ol. 67, No. 1, 2002]
=-?v--t>tsP- e v8s ;g;-X ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ .. ....... .. . ...........
1.,~~~~~~~~~~V
Figur 5. andaus fuit nd cst i lav fro Kiluea olcan nea Kalpana Hawi'i.The andaus fuit s aproxi
maey1 mi ent.Poorp y edl ufed
ESG~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/;t~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.
Figure 6. Hornto
at KilaueaVolcano, Hamaii. The hornit is approximatly 1.5-2 m inheight. Photogaph b \Vendel
DufEleld.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4
REPORTS 129
set Crater is known to have had a significant cinder-
production phase prior to the eruption of lava (Amos
1986). Cinders commonly weld to the bottom of lava
as it comes into contact with the previously deposited
cinder
apron. Importantly, the cinder inclusions indi-
cate that the corn rocks were formed at some point
following the initial eruption.
The corn-impressed rocks and the local red sco-
ria of the Brunhes-age cinder cone underlying NA
860 were
analyzed geochemically, using x-ray
flu-
orescence and
inductively coupled plasma
mass
spectrometry
to determine
major
and trace elements.
The overall patterns of the two sets of rocks are sim-
ilar and parallel, which is expected for two basalts
within a few
km of
each other
from
the same vol-
canic
field,
but the
Brunhes-age
values are
slightly
higher,
especially for the light rare earth elements.
These differences are outside analytical
error
and
are
petrologically significant (Elson
and
Ort 1999).
The corn-impressed rocks are therefore not "native"
to the site where they were found. Someone must
have carried these rocks from the Sunset Crater flow
at least
4 km
to their
present
location.
Discussion
A number of
hypotheses
can be advanced for
how
the
corn
rocks were created. One
possibility
is
that
the rocks formed
in
a corn- and cinder-filled
pit,
such
as a storage
pit,
in
a super-heated
fire
(Christian
Downum, personal
communication
1998).
In
this
case,
the
fire would be so
hot that it melted the cin-
ders,
but did this without burning the corn prior to
its
casting.
This seems
unlikely,
especially
as tem-
peratures
in
excess
of
11000
C are
required
to melt
the
cinders,
whereas corn
would combust
at
a
much
lower
temperature. Furthermore,
no
ceramic kilns or
open pit
fires
capable
of
reaching
1
1000 C are known
from the
Flagstaff
area
during
this time
period; pre-
historic
Southwest ceramics
were
generally
fired at
temperatures
between 7000 and
9000
C (Rice
1987:82).
Even if
1
100?
C was
somehow
achieved,
the
melting
of over 40
kg
of cinders seems
implau-
sible.
Additionally,
the
presence
of a
thin,
but cohe-
sive, layer
of
cinders fused
only
to what
appears
to
be the
flattened
bottom of
some
of the rocks further
negates this occurrence because
a more
mixed
and
jumbled assemblage
of melted and unmelted cinders
would be
expected
from a
pit
fire.
A
lightning
strike
into
a pile
of cinder-covered corn
is
also
unlikely.
Lightning
can melt rock and fuse
cinders, forming
fulgerites,
but it does not typically form
large melted
chunks. Instead, it tends to form narrow
tubes or thin
sheets (Bates and Jackson 1984:198).
Another possible scenario is that
piles of corn
were overrun by a lava flow. This event
could be pur-
poseful,
such as an offering of corn
placed in front
of a moving flow, or accidental, as the
moving lava
perhaps overran a just-harvested agricultural
field
with piles of corn waiting to
be collected. Two basic
lava types are present at Sunset Crater:
a'a and pahoe-
hoe (Holm and Moore 1987:395-396).
A'a flows,
which move at rates
of
meters per hour,
have irreg-
ular, blocky surfaces, and
tend
to
burn vegetation as
they approach.
Pahoehoe
flows,
on the other hand,
have smooth surfaces and typically
move at fairly
high velocities through lava
tubes,
which are nearly
isothermal
(Bullard 1979:15-16).
Pahoehoe lava
issues from
the
ends of
these tubes
at high temper-
atures as a
red or
orange liquid.
Although
the lava
is
actually
hotter than
in
most a'a flows,
the flow front
is
approachable because
it
is
a
small
feature and not
a huge
wall as
in
the a'a flow. Much of the visible
Sunset Crater lava
is of the
a'a type,
but
pahoehoe
flows commonly
occur
early
in
an
eruption
and are
present beneath the
a'a flow and in a number of sur-
face
exposures (Amos 1986;
Holm and
Moore
1987).
An a'a flow
probably
would burn a pile
of corn
before it
reached
it,
and even
if
it did
not,
the flow
front is broken
pieces
of solid rock being bulldozed
along,
and no cast of the corn could
be
made.
There-
fore,
a
pahoehoe
flow would have been
necessary
to
form the corn
rocks.
On
a recent visit to Kilauea
Volcano,
Hawai'i,
we
tested the
possibility
of
forming
a corn rock
using
a
pahoehoe
flow lobe
overrunning
corn
(Figure 7).
We
were unable to make more than an elongate depres-
sion around the
corn,
and
could not
produce
a cast
with individual kernel
imprints
because
the lava was
too cold
in
the
thin
lobes at
the front of the flow.
In
addition,
the rock lacked the cavities
found in the
corn-impressed
lava.
In
hotter,
but faster
moving,
flows,
we were unable to retrieve
the
cast
once the
flow covered
the corn cob. Based on these
results,
it
appears
that
the
simple overrunning
of a
pile
of corn
by
lava is an
unlikely hypothesis.
Paleoethnobotanist Charles
Miksicek
(personal
communication
1998)
examined
latex casts of the
corn from
the corn rock
recovered
in
the wall
of
the
masoniy
structure
(Feature 1)
at
NA
860.
Five cobs
were detailed
enough
to make
diagnostic
casts,
from
130 AMERICAN
ANTIQUITY [Vol.
67,
No.
1,
2002]
S.
~~~~~~~~~4t r
i . . .....
Figure
7.
Experimental
archaeology
at its
finest;
volcanologist
Wendell
Duffield
attempting
to recreate
onrcsa iae
Volcano,
Hawai'i.
Photograph
by
Mark
Elson.
which
three
types
of
corn
were
identified:
an
8-or-
10
rowed
flint
corn,
an
8-rowed
Pima-Papago
flour
corn,
and
a 14-rowed
Pueblo
flour
corn.
The
pres-
ence
of
three
types
of
corn
further
suggests
that
this
was
not
a harvested
field
pile
because
the
different
corn
varieties
would
have
most
likely
been
planted
in
separate
plots
to
prevent
cross-pollination
(Castet-
terand
Bell
1951:153; Forde
1931:390). In
addition,
as
Miksicek
noted,
the
denting
evident in
the
kernels
of
two
of
the
five
casts
implies
that
the
corn
was
slightly
immature
when
picked,
and
not at
the
field-
dry
stage
typically
selected
for
harvest
and
storage.
Although
the
denting
could
indicate that
immature
corn
was
being
hurriedly
picked
as
the
lava
advanced,
it
does
not
seem
likely
that
under
these
circumstances
the
time
would
also
have
been
taken
to
husk
the
corn
in
the
field.
Therefore,
the
fact that
almost
all
of
the
corn
had
been
husked
prior
to
coming
in
contact
with
the
lava
(of
the
1
10
impressions,
just
over
97 per-
cent
were
husked)
further
suggests that
the
corn
rocks
were
not
formed
by
the
overrunning
of a
picked
pile
in
a
corn
field.
Our
preferred
model
is
as
follows.
Hornitos
are
common
on
both
the
Kana'a
and
Bonito
flows;
these
features
are
approximately
1-15
m
in
height
(Holm
and
Moore
1987:395),
suggesting
that
the
spatter
was
probably
thrown
3-30 m
into
the
air
and
only
affected
a
limited
area.
The
cone-shaped
hornitos
can
be
seen
as
essentially
small
volcanoes
(see
Figure
6)
and
provide
a
"window"
into
the
source
of
the
erup-
tion,
something
that
may
have
had
symbolic
mean-
ing
to
the
corn-rock
makers.
Perhaps as
an
offering,
the
corn
was
placed
either
on
the
hornitos
themselves
or
around
the
hornitos
on
the
cinder-covered
old
flow,
during
a
relatively
qui-
escent
phase.
Because
both
the
archaeological
and
geological
evidence
suggest
that
this
event
occurred
sometime
after
the
initial
eruption, there
was
prob-
ably
some
familiarity
with
hornitos,
lava
flows,
and
spatter,
affording
a
relative
degree
of
safety.
The hor-
nito
continued
to
erupt
after
the
corn
was
placed,
and
covered
the
ears
with
spatter.
The
spatter,
especially
the
larger,
more
fluid
blobs,
formed
corn
impressions,
aided
by
the
overlying
weight
of
the
growing
hor-
nito.
The
repeated
deposition
of
spatter
caused the
development of
irregular
cavities;
the
surface of
the
spatter
was
annealed,
forming
a
sheen,
and
droplets
of
lava
also
formed.
Cinders
were
welded
to
the
bot-
tom
of
the
clasts.
The corn-rock
makers
may
have
watched
the
process
from
a safe
distance,
or
they
may have
left
and
returned
to
find
their
corn
offerings buried
by
the
spatter
and
vaporized.
Once
the
lava
cooled,
they
then
knocked
off
the
parts
of
the
hornitos
that
REPORTS 131
contained the impressions
and carried over 40 kg of
basalt a distance of at least
4 km to NA 860. Here
they pared the basalt down
to reveal the corn casts,
leaving mostly waste material
in
the concentration
in
the north-central portion
of the site (see Figure 3).
It
is
unknown why larger,
heavier rocks were brought
back to the site for reduction,
instead of smaller,
already
reduced
rocks.
Undertaking lithic reduction
at the site itself
may
be
related
to
safety factors,
that
is, the proximity of flowing
or spattering lava, from
the
hornito,
or
possibly
to ritual factors associated
with
the
offering.
At least one
of
the
corn
rocks,
the
most detailed
example
recovered,
was then incor-
porated
into the
masonry
wall of a structure. We sus-
pect
that
other
highly
detailed corn rocks have
been
removed from the site by
collectors;
we were told
by
one collector that the
corn
rocks
were most com-
monly
found
in
the
pithouse
depressions, suggest-
ing the possibility that
other corn rocks were also
associated with structures. However,
a
careful
search
of all
pithouse depressions
recovered
only
two addi-
tional corn rocks.
Conclusion
There
is
little
doubt that the
eruption
of Sunset Crater
was one
of
the most
significant
events
in
the
life of
anybody living
in the
Flagstaff
area
in
the mid-to-
late eleventh century
A.D.
Worldwide ethnographic
and
archaeological
data
indicate that most
groups
residing near an active (and
often
inactive) volcano
incorporate
such
features
into their belief systems,
and volcano
offerings
are common
aspects
of
asso-
ciated rituals
(Kirch
1985;
Nolan
1979;
Plunket
and
Uruniuela 1998a, 1998b;
Scarth
1999; Sigurdsson
1999). Furthermore,
studies
of
catastrophic
events
have shown that
religious
or
cultural mechanisms
for
coping
with a natural
disaster,
such
as a volcanic
eruption,
can be
highly
adaptive, enabling
affected
individuals and
groups
to
accept
the event
more
read-
ily
and
begin
the
recovery
process (Nolan 1979:33 1).
Volcano
eruptions
are
commonly
seen
as
signs
of
spiritual transgressions
and
offerings
are made
in
an
attempt
to
rectify
these "sins"
and
avert the
ongoing
destruction
(Scarth 1999:2).
Hopi
accounts
of the
Sunset Crater
eruption
cite various
offenses,
includ-
ing gambling, immoral
behavior,
and the
cuckold-
ing
of
a katsina
(Colton
1932b; Ferguson
and
Lomaomvaya 2000;
Malotki and
Lomatuway'ma
1987).
At
Paricutin,
the Tarascan
villagers
blamed
the
eruption
on the desecration of a shrine and "the
wrath of God
on a sinful
people" (Nolan
1979:301;
see also Carrasco
Pizana 1946). The
residents of
Paricutin erected
a row of 2-m
high wooden
crosses
in front of the
moving flow
to prevent it,
unsuccess-
fully, from
encroaching
upon their village
(Foshag
and Gonzalez-Reyna
1956; Luhr and Simkin
1993:Fig.
45). Other
activities undertaken at Panr-
cutin to stop
the eruption
involved praying
at the
graves of the
ancestors
when the lava
reached the
cemetery,
and a
procession
by
the women of the vil-
lage, who
crawled across
the town plaza
on their
knees
to pray
at the church
(Nolan 1993:190-193).
Corn
is
a
sacred
plant
to
all
Pueblo
groups
and it
is
likely
that it served a similar purpose
in the pre-
historic
Flagstaff area. Although
corn
is most com-
monly offered
today
in the form of corn
meal, whole
ears are used
in
numerous
ceremonies.
Among
the
Hopi,
for example, corn
ears are placed
on altars, car-
ried and
distributed
by
certain
katsinas,
and their dif-
ferent colors
are
used to represent the
six
cardinal
directions (Stephen
1936). Interestingly,
the
most
well-known
Hopi
account of the Sunset Crater
erup-
tion involves the
Ka'nas
(Qa'nas)
katsinas who
wear
"long strands
of baked sweet corn ... slung
across
their bodies
in
the fashion
of bandoliers"
(Malotki
and
Lomatuway'ma
1987:98).
In this
account,
the
Ka'nas katsina
started
the eruption to punish
the peo-
ple
of
Musangnuvi
for the transgressions
of a few
evil
young
men. The
eruption
was
followed
by
a
severe
drought
and
widespread
famine that resulted
in
migration
from Second Mesa.
The
drought
and
famine
persisted
for
years,
until
finally,
following
the
demise
of
the
evildoers,
the Ka'nas katsina
took
pity
on the remaining villagers.
He
gave
each
family
an
ear of
baked
corn to
put
in
their storerooms.
Overnight
the ear
of
corn
magically
multiplied
end-
ing
the
famine
and bringing
harmony
back to the
peo-
ple (Nequatewa
1932).
As
recorded by
Ferguson and
Lomaomvaya (2000:9),
"In
remembering
a Qa'na
Katsina dance performed
in the
1970s,
LaVern
Siweumptewa
(1998)
recalled
that,
'Many
of
his
motions
and
dance
relate to the time
of
u
'wing pangk
yama,
fire came forth here."'
Siweumptewa
and
sev-
eral other
Hopi
elders told
us
that
ears of corn are
distributed
as
part
of the
ceremony.
Therefore,
we
think
it is
plausible
that
whole ears
of corn were
placed
as an
offering
around the base
of
an
hornito,
or
symbolic
volcano,
perhaps
to
appease
the forces
responsible
for the
eruption. Why
the rocks with corn casts
were then removed
to a
132 AMERICAN ANTIQUITY [Vol. 67, No. 1,
2002]
habitation site 4 km distant is unknown, but it can be
speculated that the corn rocks themselves may have
been seen as a source of supernatural power or, given
that one was found embedded in the wall of a struc-
ture, as protection from the malevolent forces of the
volcano.
Two other corn rocks are known from the South-
west United States. One was collected
in
1974 by an
unnamed
surveyor
for the United
States
Geodetic
Survey.
This
piece
is
currently
on
display
in the
Sun-
set Crater National Monument Visitor Center. Loca-
tional and topographic
information
given by
the
surveyor (on
file
at
the
Monument) place
this
corn
rock
approximately
1
km south of
NA 860.
Although
there
are
a
number
of recorded
prehistoric sites
in
this
general area, whether the Visitor Center corn rock was
recovered from a site is unknown. This corn rock is
visually identical
to those recovered from NA
860.
The
other
corn rock
was collected by noted
archaeologist
J. Walter Fewkes sometime
around
1890.
As
reported
in
the
February 14, 1890
issue of
Scribner's
magazine,
Fewkes
collected
a fragment
of lava "from the
neighborhood
of
Zuni,
New Mex-
ico" that contained "the impress of numerous ears
of corn as
perfectly preserved
as that of
any
of the
perishable
articles in
Pompeii" (Wright 1890:238).
This
piece
has
been loaned to
us
by
the
Peabody
Museum of Harvard University where a portion of
Fewkes's collection resides.
Unfortunately,
there
is
no additional
provenience
information with the
rock;
all we
have
is the somewhat
nonspecific
Zuni loca-
tion
given
in the
Scribner's
article.
However,
the
Fewkes
corn
rock
is
macroscopically
different from
the Sunset Crater
corn
rocks, being lighter
in
weight
and brown to tan in color. Because
this
could
be due
to a number of local or surficial
factors, geochemi-
cal
analyses
are
being
undertaken to determine the
origin
of
this
rock.
Volcano
eruptions
are
some
of the most
power-
ful and destructive occurrences in the natural realm.
They
are rife with
symbolism, dramatically altering
the
landscape by spewing
molten material from the
bowels of the earth.
They
are
fire-breathing,
earth-
shaking creatures, who, accompanied by lightning
and
thunder,
have the
power
to turn
day
into
night
and
night
into
day.
It
is
little wonder that ritual and
volcanism
go
hand
in
hand,
as both the
ethnographic
and
archaeological
record attest. The
recovery
of the
corn rocks
provides
a small window to
begin
inves-
tigating
the nature of this behavior.
Acknowledgments. The U.S. 89 Archaeological Project was
funded by the Arizona Department of Transportation (ADOT)
under permit to the Coconino National Forest (CNF). We
thank ADOT archaeologists Bettina Rosenberg and Bob
Gasser and CNF archaeologists Peter Pilles and Linda
Farnsworth for their support and assistance during this project.
Financial and logistical support from Desert Archaeology,
Inc.
and
Northern
Arizona
University
are also
gratefully acknowl-
edged. Discussions with Terry Samples, Peter Pilles, Christian
Downum, David Wilcox, Nancy Riggs, Deborah Swartz, Kirk
Anderson, Jane Sliva, and William Doelle were most helpful.
Analysis of the corn-rock assemblage was undertaken by
Deborah Swartz, Susan Hall, and Jorge Alzaga. While
researching other things, Catherine Gilman found the blurb
on
the Fewkes corn
rock in
the
1890
Scribner's
magazine;
Steven
LeBlanc and Genevieve Fisher of the Peabody Museum,
Harvard University,
facilitated the loan
of
this
specimen.
All
chemical
analyses
were
undertaken
at the
GeoAnalytical
Laboratory
of
Washington
State
University.
The
digital
eleva-
tional
model used
to
explore
the
Sunset
Crater viewshed was
computed by Kyle
Bohnenstiehl.
We
would also
like
to thank
David Sherrod and Donald Swanson of the USGS Hawaiian
Volcano
Observatory
for their
hospitality, interest,
and
input
into
this project. Expenses
for our Hawaiian
experiments
were
covered by
the
participants and
not
by
ADOT. Comments from
Sarah
Herr,
Joanne
Newcomb,
Alan Sullivan,
and William
Walker on previous drafts
of this
paper
are also gratefully
acknowledged.
The Spanish
translation of the abstract was
provided by M. Nieves Zedenjo.
References
Cited
Amos, R. C.
1986 Sunset Crater, Arizona:
Evidence
for a Large Magni-
tude Strombolian Eruption.
Unpublished Masters's thesis,
Department of Geology,
Arizona State University, Tempe.
Anderson, B. A. (compiler)
1990 The
WupatkiArchaeological
Inventory Survey Project:
Final Report. Professional
Paper No. 35. Southwest Cultural
Resource Center,
Division of Anthropology, National Park
Service, Santa Fe.
Bates, R. L., and J. A. Jackson
(editors)
1984 Dictionary of Geological
Terms, 3rd
ed. The American
Geological
Institute.
Anchor
Press/Doubleday,
Garden
City,
New York.
Blong,
R. J.
1984 Volcanic Hazards:A
Sourcebookon the
Effects
of Erup-
tions.
Academic
Press,
Orlando.
Boston, R.
L.
1995 Electron
Microprobe
Sourcing of
Volcanic Ash
Temper
in Sunset Red Ceramics.
Unpublished Master's thesis,
Department
of
Anthropology,
Northern Arizona
University,
Flagstaff.
Bretermitz,
D.
A.
1967 The
Eruption(s)
of Sunset Crater: Dating
and Effects.
Plateau
40:72-76.
Bullard,
F. M.
1979 Volcanoes and Their
Activity.
In Volcanic
Activity
and
Human Ecology, edited
by
P. D. Sheets
and D. K.
Grayson,
pp. 9-48.
Academic
Press,
New York.
Carrasco
Pizana,
P.
1946 Paricutfn Volcano
in
Tarascan Folklore. El Palacio
LIII:299-306.
Castetter,
E.
F.,
and W. H. Bell
REPORTS 133
1951 Yuman
IndianAgriculture: Primitive Subsistence on the
Lower Colorado and Gila
Rivers. University of New Mex-
ico
Press, Albuquerque.
Champion,
D.
E.
1980 Holocene
Geomagnetic Secular Variation in the West-
ern
United States:
Implicationsfor the Global Geomagnetic
Field.
Open-File Report
80-824. United States Department
of
the
Interior
Geological
Survey, Washington.
Colton,
H. S.
1932a Sunset Crater:
The
Effect of
a
Volcanic Eruption on
the Ancient Pueblo People. Geographical Review
22:582-590.
1932b A
Possible Hopi
Tradition of the Eruption of Sunset
Crater. Museum Notes
5(4):23. Museum of Northern Ari-
zona, Flagstaff.
1932c A Survey of
Prehistoric Sites in the Region of
Flagstaff;
Arizona. Bureau of American
Ethnology Bulletin No. 104.
Smithsonian
Institution,
Washington.
1937 The
Eruption
of
Sunset Crater
as
an
Eye
Witness
Might
Have Observed It. Museum
Notes 10(4):9-12. Museum
of
Northern
Arizona, Flagstaff.
1945 A
Revision of the Date of
the Eruption
of
Sunset
Crater.
Southwestern Journal
of
Anthropology
1:345-355.
1946 The
Sinagua:
A Summary of
the
Archaeology of
the
Region of
Flagstaff;
Arizona.
Bulletin 22. Museum of North-
ern
Arizona, Flagstaff.
1949 The Prehistoric Population of
theFlagstaffArea. Plateau
22:21-25.
1965 Experiments
in
Raising Corn
in the Sunset Crater Ash-
fall Area
East
of
Flagstaff,
Arizona.
Plateau 37:77-79.
1967 The Basaltic
Cinder Cones and Lava Flows
of
the San
Francisco Mountain
Volcanic Field. Bulletin 10, rev. 2nd
ed.
Museum of Northern
Arizona, Flagstaff.
Downum,
C. E.
1988 "One
Grand
History":
A
Critical Review
of
Flagstaff
Archaeology,
1851- 1988. Ph.D.
dissertation, University
of
Arizona,
Tucson.
University Microfilms, Ann Arbor.
Downum,
C.
E.,
and G.
Gumerman
1998 Archaeological
Investigations
atSunset CraterVolcano
National Monument.
Archaeological Report
No. 1159.
Northern
Arizona
University, Flagstaff.
Downum,
C.
E.,
and A. P.
Sullivan, III
1990 Settlement
Patterns.
In The
Wupatki Archeological
Inventory Survey Project:
Final Report, compiled by B.
A.
Anderson, pp. 5.1-5.90. Professional
Paper
No. 35. South-
west Cultural Resource
Center,
Division of
Anthropology,
National Park
Service,
Santa Fe.
Elson,
M. D.
1997 Treatment Plan
for
the
U.S.
89
-
Wupatki
to
Fernwood
Project.
Technical
Report
No. 97-10.
Center for Desert
Archaeology,
Tucson.
Elson,
M.
D.,
and M. H.
Ort
1999 A Prehistoric Volcano
Offering
from
the American
Southwest.
Paper presented
at
the 64th Annual
Meeting
of
the
Society
for
American
Archaeology, Chicago.
Elson,
M.
D.,
D. L. Swartz,
T. L. Samples,
and S. D. Hall
2001 The
U.S.
89
Archaeological Project:
Introduction
and
Site
Descriptions.
Anthropological Papers
No.
30.
Center
for
Desert
Archaeology,
Tucson,
in
press.
Fenton,
C.
R.,
R.
H.
Webb,
P. A.
Pearthree,
T.
E.
Cerling,
and R.
J.
Poreda
2001 Displacement
Rates on the
Toroweep
and Hurricane
Faults:
Implications
for
Quaternary Downcutting
in
Grand
Canyon,
Arizona.
Geology,
in
press.
Ferguson,
T.
J.,
and
M.
Lomaomvaya
2000 Nuvatukya'ovi,
Palatsmo, Niqw Wupatki: Hopi
His-
tory, Culture,
and
Landscape. Manuscript
on
file,
Desert
Archaeology, Inc., Tucson.
Forde, C. D.
1931 Hopi Agriculture and Land
Ownership. The Journal of
the RoyalAnthropological Institute
of GreatBritain and
Ire-
land LXI:357-406.
Foshag, W. F., and J. R. Gonzalez-Reyna
1956 Birth and Development
of Paricutin Volcano, Mexico.
U.S.
Geological Survey
Bulletin 965D:355-489.
Hargrave, L. L.
1930 N.A.860A,B,C,D,E & 861.
Manuscriptonfile, NA Site
Files, Museum of Northern
Arizona, Flagstaff.
Holm, R. F.
1986 Field Guide to
the
Geology
of
the Central San
Fran-
cisco Volcanic Field.
In
Geology
of Central and Northern
Arizona, Geological Society
of America, Rocky Mountain
Section,
Field
Trip
Guidebook,
edited
by J.
D. Nations,
C.
M.
Conway,
and G. A. Swann, pp. 27-43. Geological
Soci-
ety
of
America, Boulder.
Holm, R. F., and R. B. Moore
1987 Holocene Scoria Cone and Lava Flows
at
Sunset Crater,
Northern Arizona. In
Rocky
Mountin Section of the
Geo-
logical Society of America
Centennial Field Guide, edited
by S. Beus, pp. 393-397. Geological
Society of America,
Boulder.
Hooten, J. A., and M. A. Ort
2000 Distribution and
Thickness
of the
Northern Component
of the Sunset Crater
Fall
Deposit.
Manuscript
on
file,
Desert
Archaeology, Inc.,
Tucson.
Hooten,
J.
A.,
M. A.
Ort,
and M. D.
Elson
2001 The Origin of Cinders in
Wupatki National
Monument.
Report prepared
for Southwest
Parks
and Monuments
Asso-
ciation
Grant
FY00-1 1. Technical
Report
No.
2001-2012.
Desert Archaeology, Inc., Tucson.
Kirch, P. V.
1985 Feathered
Gods and Fishhooks: An Introduction
to
Hawaiian
Archaeology
and Prehistory. University
of
Hawai'i
Press,
Honolulu.
Laughlin,
A.
W.,
R. W.
Charles,
K.
Reid,
and C. White
1993 Field
Trip
Guide
to the
Geochronology of
El
Malpais
National Monument
and the Zuni-Bandera Volcanic
Field,
New Mexico. Bulletin 149. New
Mexico
Bureau
of Mines
and
Mineral
Resources,
Socorro,
New Mexico.
Luhr,
J.
F.,
and T. Simkin
(editors)
1993 Parfcutin:
The Volcano
Born in a Mexican
Cornfield.
Geoscience
Press, Inc.,
Phoenix.
Mabery,
M.
V.,
R.
Moore,
and K.
Hon
1999 The Volcanic
Eruptions
of
El
Malpais:
A
Guide
to the
Volcanic
History
and
Formations
of
El Malpais
National
Monument. Ancient
City
Press,
Santa Fe.
Mahle,
S. H.
1963 Corn
Growing
at
Wupatki.
Plateau 36:29-32.
Malotki, E.,
and M.
Lomatuway'ma
1987 Earth
Fire: A
Hopi
Legend of
the
Sunset
Crater
Erup-
tion.
Northland
Press, Flagstaff,
Arizona.
Moore,
R.
B.,
and
E.
W.
Wolfe
1987 Geologic Map
of
theEast
Part of the San FranciscoVol-
canic
Field,
North-central Arizona. U.S.
Geological Survey
Miscellaneous Field
Studies, Map MF-1960, 1:50,000.
Nequatewa,
E.
1932 The
Kana-a Kachinas of Sunset Crater. Museum
Notes
5(4):19-23.
Museum
of
Northern
Arizona, Flagstaff.
Nolan,
M. L.
1979 Impact
of
Parfcutin
on
Five
Communities. In
Volcanic
Activity
and
Human
Ecology,
edited
by
P.
D. Sheets
and D.
K.
Grayson, pp. 293-338.
Academic
Press,
New
York.
1993 Human Communities and Their
Responses.
In
Pari-
cutin: The Volcano
Born
in a
Mexican
Cornfield,
edited
by
134 AMERICAN ANTIQUITY [Vol. 67, No. 1, 2002]
J. F. Luhr and T. Simkin, pp. 189-207. Geoscience
Press,
Inc., Phoenix.
Pilles, P. J., Jr.
1978 The Field House and Sinagua Demography. In Limited
Activity and Occupation Sites: A Collection of Conference
Papers, edited by A. E. Ward, pp. 119-133. Contributions
to Anthropological Studies No. 1. Center for Anthropolog-
ical Studies, Albuquerque.
1979 Sunset Crater and the Sinagua:
A
New Interpretation.
In Volcanic Activity and Human Ecology, edited by P. D.
Sheets and D. K. Grayson, pp. 459-485. Academic Press,
New York.
1987 Hisatsinom: The Ancient People of Sunset Crater. In
Earth
Fire:
A
Hopi Legend of
the
Sunset Crater Eruption,
by E. Malotki
and
M. Lomatuway'ma, pp. 105-119.
North-
land Press, Flagstaff, Arizona.
1996 The Pueblo III Period
along
the
Mogollon
Rim: The
Honanki, Elden, and Turkey Hill Phases of the Sinagua. In
The Prehistoric Pueblo World, A.D. 1150-1350,
edited
by
M.
A.
Adler, pp.
59-72.
University
of Arizona
Press,
Tuc-
son.
Plunket, P., and G. Uruniuela
1998a Appeasing the Volcano Gods. Archaeology
51(4):36-42.
1998b Preclassic Household Patterns Preserved
Under Vol-
canicAsh atTetimpa, Puebla, Mexico. LatinAmericanAntiq-
uity 94:287-309.
Rees,
J. D.
1979 Effects
of
the Eruption of Paricutin Volcano
on Land-
forms, Vegetation, and
Human
Occupancy.
In
VolcanicActiv-
ity
and Human
Ecology,
edited
by
P. D. Sheets and
D. K.
Grayson, pp. 249-292. Academic Press,
New York.
Rice, P.
M.
1987 PotteryAnalysis: A Sourcebook. University
of
Chicago
Press, Chicago.
Robinson, W. J.,
B.
G. Harrill, and
R. L.
Warren
1975 Tree-RingDatesfromArizonaH-I,
FlagstaffArea.
Lab-
oratory
of Tree-
Ring Research, University
of
Arizona,
Tuc-
son.
Scarth,
A.
1999 Vulcan's Fury:
Man
Against
the
Volcano. Yale Univer-
sity Press,
New
Haven,
Connecticut.
Sheets,
P.
D.
1980 Archaeological Studies of Disaster: Their Range and
Value. Natural Hazard Research
Working Paper
38. Institute
of
Behavioral
Science, University
of
Colorado,
Boulder.
Shoemaker,
E.
M.,
and
D. E. Champion
1977 Eruption History
of Sunset
Crater,
Arizona.
Investiga-
tor'sAnnual
Report. Manuscript
on
file,
Area National
Mon-
uments
Headquarters, Wupatki,
Sunset Crater
Volcano,
and
Walnut
Canyon
National
Monuments, Flagstaff,
Arizona.
Sigurdsson,
H.
1999 Melting
the
Earth:
The
History of
Ideas
on
Volcanic
Eruptions.
Oxford
University Press,
New
York.
Smiley,
T.
L.
1958 The
Geology andDating
of
SunsetCrater,
Flagstaff,Ari-
zona.
In
Guidebook
of
the Black Mesa
Basin,
Northeastern
Arizona,
edited
by
R.
Y.
Anderson and
J. W.
Harshbarger,
pp.
186-190.
New
Mexico
Geological Society,
Ninth Field
Conference.
Stephen,
A.
M.
1936 Hopi Journal. Columbia University Contributions
to
Anthropology
23.
Columbia
University Press,
New York.
Sullivan,
A.
P.,
III
1984 Sinagua Agricultural Strategies
and
Sunset
Crater Vol-
canism.
In
Prehistoric Agricultural Strategies
in
the South-
west, edited by
S. K. Fish and P.
R. Fish, pp. 85-100. Anthro-
pological Research
Papers No. 33. Arizona
State University,
Tempe.
Sullivan, A. P., III,
and
C. E.
Downum
1991 Aridity,Activity,
andVolcanicAshAgriculture:A
Study
of Short-Term
Prehistoric Cultural-Ecological
Dynamics.
World Archaeology
22:271-287.
Swartz, D. L., and
M. D. Elson
2001 The Lenox
Park Site, NA 20,700
(AR-03-04-02-3352).
In The U.S.
89 Archaeological Project:
Introduction and
Site
Descriptions,
by
M. D.
Elson,
D. L.
Swartz,
T. L. Sam-
ples, and S. D.
Hall. Anthropological
Papers No. 30. Center
for Desert Archaeology, Tucson,
in
press.
Tanaka, K. L.,
E.
M. Shoemaker, G. E. Ulrich,
and
E.
W. Wolfe
1986 Migration
of
Volcanism
in
the
San Francisco Volcanic
Field,
Arizona. Geological Society
of
America Bulletin
97: 129-141.
United States Geological Survey
1989 The
Severity
of an Earthquake.
General
Interest
Publi-
cations of the
U.S. Geological Survey
1989-288-913. U.S.
Government
Printing Office, Washington,
D.C.
Waring,
G. L.
2001 Hopi
Corn and Sunset Crater
Cinders:
A
Test of the Rela-
tionship
Between
Volcanic Ash and
Agriculture
in
Northern
Arizona. Manuscript
on
file,
DesertArchaeology, Inc.,
Tuc-
son.
Wright,
G. F.
1890 AnArchaeologicalDiscovery
inldaho. Scribner'sMag-
azine 7(2):235-
238.
Yokoyama, I.,
and S.
De
la
Cruz-Reyna
1990 Precursory
Earthquakes
of the
1943 Eruption
of the
Paricutin
Volcano, Michoacan,
M6xico. Journal of
Vol-
canology and
Geothermal
Research 44:265-281.
Notes
1. The corn rocks were
gathered
during
recent
investiga-
tions
at
40
prehistoric
sites
along
a 26.7-km section
of
U.S.
89
just
north of
Flagstaff,
Arizona
(Figure
1). This
project
was undertaken
by
Desert
Archaeology,
Inc.
for the
Arizona
Department
of Transportation. Sites
within
the
project
area
were occupied
between A.D. 500 and 1200, and
included
small
pithouse
and
masonry pueblo
habitations
as well as
special-use
sites
such
as fieldhouses,
agricultural areas,
and
ceremonial locales (Elson 1997;
Elson
et
al.
2001).
The
great
majority
of
sites date
to the
period
between A.D. 1050
and
1150,
the
approximate
time
of the Sunset
Crater eruption.
2. This
earthquake
measured
4.4 on the Mercalli scale
(Luhr
and
Simkin
1993:10).
The Mercalli scale
is a measure
of
intensity
and
is not
directly
comparable
to the
Richter
scale,
which is a measure
of
magnitude.
Richter-scale data
are
not
available
for the Paricutin
eruption.
The Mercalli
scale
is a subjective
ranking
based
on observed
effects. An
earthquake
with
a Mercalli
intensity
of
4.0 would
have
been
"felt indoors
by
many,
outdoors
by
few
during
the
day.
At
night
some awakened.
Dishes,
windows,
doors
disturbed;
walls
make
cracking
sounds. Sensation
like
a heavy
truck
striking building.
Standing
motor cars rocked
noticeably"
(USGS 1989).
3. NA 860 was
outside
the U.S. 89
project
area
and would
not be
subject
to
impacts
from road
construction
(see Figure
1). However,
the Arizona
Department
of
Transportation
and
the Coconino National Forest
agreed
to allow limited
inves-
tigations
because of
the
potential significance
of the corn-
REPORTS 135
rock assemblage. The site was brought to our attention by
physical anthropologist
Marc
Gaede,
who visited the site in
1968 with
Edward
Danson, then
director of MNA.
A local
cowboy had reported the presence
of
corn rocks
on
the site to
Danson,
and
Gaede told
us
that over
the
years
a number of
these rocks were removed
and in private
collections.
Recovery of the
remaining corn rocks
was necessary to pre-
serve the assemblage.
Received February 28,
2001; Accepted June 7,
2001; Revised
August 2, 2001.
Prehistory
of the Carson
Desert and Stillwater Mountains
PRIisT'oRY OF THe Environment,
Mobility,
and
Subsistence
in a Great Basin Wetland
CARSON DESERT AND
STILIWATER MOUNTAINS Robert
L. Kelly
E.in.,mme[,
Wibffity,
-ll
S.I.i,tmct-
i. a
G.Hi^.
lAVd "The
research
on
the
prehistory
of
the
Stillwater
region
of
western
N Nevada would be
incomplete
without
publication
of this
monograph."
A University
of
Utah
Anthropological Paper
123
Paper
$45.00
Apaches
de
Navajo
Seventeenth-Century
Navajos in the Chama Valley of New Mexico
Curtis
F. Schaafsma
"Schaafsma
is
the
right person to pull this material together because
of
his
important
early work in the upper Chama Valley, which defined
the
Piedra
Lumbre
phase, and because of his extensive knowledge
of the
ethnohistorical
(Spanish) literature."
-Robert
Preucel, University Museum, University
of
Pennsylvania
152
illustrations
Cloth
$55.00
_g_g_~~~~~~~~~~~~~~~~~~~~~~~
