Journal of

Archaeological
SCIENCE
Journal of Archaeological Science 31 (2004) 205–215
http://www.elsevier.com/locate/jas

Biosilicate analysis of residue in Maya dedicatory cache vessels

from Blue Creek, Belize
Steven R. Bozarth a, Thomas H. Guderjan b,*
a
Department of Geography, University of Kansas, Lawrence, KS 66045-7613, USA
b
Department of Sociology, Criminal Justice, and Anthropology, Texas Christian University, Fort Worth, TX 76129, USA
Received 20 March 2003; received in revised form 21 July 2003; accepted 19 August 2003

Abstract

Residues from nine ancient Maya dedicatory vessels were analyzed for biosilicates. In all cases, the analysis was successful in
identifying plant and sponge remains that had been placed in the vessels. This analysis sheds light on ancient Maya plant use as well
as ritual and religious practices.
 2003 Elsevier Ltd. All rights reserved.

Keywords: Maya; Ritual; Caches; Biosilicates; Sponges; Residue

1. Introduction materials, including jade, obsidian, formal chert tools,

Biosilicates were successfully isolated and analyzed
from dedicatory caches at the Maya center of Blue
Creek in northern Belize. These caches date from the
Late Preclassic to the early part of the Late Classic
periods, or approximately AD 200–650. The primary
objective was to expand our understanding of Maya
plant use and ritual behavior by analyzing biosilicates
(opal phytoliths, sponge spicules, algal statospores, and
diatoms) extracted from ceramic vessels placed within
Maya buildings during dedication rituals. Identification
of culturally significant biosilicates will open new
avenues of interpretation of ritual activities.
1.1. Maya dedicatory caches
A common form for dedicatory caches is the place-
ment of a ceramic vessel, commonly a bowl, with a
second vessel, usually identical to the first, inverted on
top of the first. Such placements of two vessels are
commonly termed “lip-to-lip” caches. Within these
vessels, archaeologists have found a wide variety of
* Corresponding author. Tel.: +1-817-257-7470;
fax: +1-817-257-74-77-37
E-mail address: guderjan@tcu.edu (T.H. Guderjan).
0305-4403/04/$ - see front matter  2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jas.2003.08.002
coral, shells such as Pacific spondylus, Caribbean conch
and riverine bivalves, and sting ray spines. Since
preservation of biological materials is poor in the region,
archaeologists have always assumed that only a portion
of the cached materials were being recovered.
During the construction of public monumental build-
ings and secular residential buildings, Preclassic and
Classic Maya peoples commonly placed dedicatory
caches within the construction fill, commonly in the
front of the building along the medial axis. Maya
archaeologists have long understood that such caches
aided in dating construction events due to their primary
context. Further, Maya archaeologists also have long
understood that these caches represent the material
residue of important dedication ceremonies [57]. An
early examination of such caches was William Coe’s
analysis of caches from the site of Piedras Negras [19].
Given the recent expansion of our understanding of
Maya writing and religion, specifically in terms of how
religious and cosmological concepts are embedded in
architecture and site planning [1,44], attention has been
turned to grappling with the meaning of these caches.
Importantly, such caching events must be placed into
the ritual context. The cache is not the event of interest,
rather it is the ritual. The cache is simply our only
existing material remains of the ritual. Consequently, we
206 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
are not attempting to interpret the cache, but the ritual
event that incorporated its placement.
One view of dedicatory caching is that it maintains “a
physical expression to the pipeline between humans and
gods” [46]. In a study of the history of a single building,
Garber et al. argued that buildings themselves were
perceived as going through a sequence of life, death, and
rebirth, and that dedicatory caching is related to the
embodying of life into a building [24]. They further
argued that a cache at the site of Blackman Eddy in
Belize represents a tripartite division (the underworld,
the Earth, and the heavens) of the Maya world in
microcosm. Beyond this, Guderjan argued that Maya
caching intensifies their relationship with the sacred in
ways that were meant to tangibly affect events in the
material world [28]. While dedicatory caches may repre-
sent many things [23], they clearly are related to embed-
ding sacredness into the physical structure and are
symbolic recreations of the landscape of the Creation
myth [29]. However, until now our ability to understand
ritual caching has been limited by the “hard” or non-
perishable materials remaining within cached vessels.
1.2. The Maya view of the cosmos of creation
The Maya view of creation and the cosmos is articu-
lated in a colonial manuscript, the Popul Vuh [56].
Further, studies of contemporary Maya in Chiapas [59]
and elsewhere show us that the Popul Vuh creation myth
exists today and permeates much of contemporary
Maya ritual behavior. Further, the principles of creation
and the cosmos were embedded in the monumental
architecture of the public centers of Maya cities [23].
The Popol Vuh tells the story of the current, fourth,
Creation of the world for the Maya [56]. After the
adventures of a pair of Hero Twins, common elements
in American Indian origin mythology, the Creators once
again set to the task of making the world. Rising out of
a Primordial Sea present at the Creation was the First
Mountain or “Witz Mountain”. On its shores grew all
kinds of food, maize, cacao, and fruits. Then, the
creators molded the first four human beings from the
maize. Above the First Mountain was the pageant of
the sky, which each year plays out the story of Creation
in Maya mythology and is often represented by the
dome shape of a turtle shell. So, the Maya saw three
fundamental components of the cosmos at Creation: the
Primordial Sea, the Witz Mountain, and the shell of
the Sky [23].
We argue in this paper that the components of the
Creation are replayed in the act of placed dedicatory
caches into Maya buildings. In each of these caches, we
see “most favored” elements of each component: the sea,
the land and the sky. Further, this re-enactment is not
restricted to public or elite settings. Instead it is found in
all social strata of the Maya city of Blue Creek and
reflects a shared view of religion and the universe that
bound the strata of Maya communities together.
1.3. Dedicatory caches at Blue Creek
The Maya site of Blue Creek has been the focus
of intensive investigations since 1992 when the Maya
Research Program, under the direction of T.H.
Guderjan, undertook a long-term program of exca-
vation and study of the ancient community [21,33,35].
Since then, the project has yielded significant infor-
mation about this ancient community [30–32,34,36,43].
Large-scale excavations have been conducted in a
number of components of the site: the Core
Area, largely composed of monumental architecture
[20,22,28]; an elite residential area known as Kı́n Tan, or
the Western Group, which includes a complex residence
known as the Structure 37 Plazuela [34,37,38]; and a
non-elite residential group known as Chan Cahal
[18,53]. A decade of research at Blue Creek has created
large databases of several categories of artifacts, such as
jade and other materials, that now enable us to make
“well-grounded” interpretations [30]. Dedicatory caches
are one such facet of the archaeological record.
Nine caches from a variety of contexts were selected
for this study. Residue from most of the caches had been
curated from excavations prior to 1999. Since then, the
samples were collected specifically for the purpose of this
study. There were two selection criteria involved. First,
we selected only caches dating from the Late Preclassic
and Early Classic periods. (One cache was later re-dated
to the early part of the Late Classic period.) Second, we
selected only caches that were lip-to-lip placements.
Each cache consisted of a ceramic vessel in normal
orientation with another vessel, inverted, placed on top
of the lower vessel. Seven of these caches were from
monumental architecture in the Core Area and one
each derived from excavations at Chan Cahal and the
Structure 37 Plazuela. This set of caches represents three
quite different contexts but all derive from a relatively
restricted temporal frame.
Cache 4A, Cache 4B, and Cache 6 were found in
Structure 1 and associated with Tomb 5. Structure 1-III
was an innovative and graceful building that consisted
of a large sub-structure more than 10 meters tall and a
columned super-structure with a perishable roof built in
the Early Classic Period [20]. It was partially razed
about AD 500–550 to accommodate a new super-
structure that was flat-topped, though it may have
incorporated a perishable superstructure. Caches 4A, 4B
and 6 were all placed in the construction of this new
super-structure, Structure 1-IV, as was Tomb 5, a royal
interment that included three ceramic vessels as well as
jade and other artifacts. Contents of all three of these
caches were within pairs of Aguilla Orange ceramic
 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215 207

Table 1 Excavation yielded only a few shell and jade artifacts,

Contents of caches but the matrix inside the vessels was stained red, pre-
Cache Structure/context Objects recovered
sumably by red ochre. Cache 30 was considerably more
complex. Two sets of Aguilla Orange vessels were placed
4A Str.1 public architecture Jade artifacts—10
lip-to-lip, with one pair (Cache 30A) stacked on top of
Shell and shell artifacts—11
Hematite—1 the other (Cache 30B). The upper vessel of Cache 30 was
broken, and overlying matrix had intruded into the
4B Str. 1 public architecture Shell—3
cache; consequently, no sample was taken. The lower
6A Str. 1 public architecture Jade artifacts—6 set, however, was well sealed and contained nearly 90
Shell—3
jade, coral, and shell artifacts. Residue was collected
7 Str. 15 public architecture Shell (bivalve)—1 from Cache 30B.
28 Str. 12 public architecture Shell (bivalve)—1 Cache 43, discovered in the non-elite, residential
30B Str. 12 public architecture Jade artifacts—77
community of Chan Cahal, was the dedicatory cache for
Coral artifacts—4 Structure U17. It consisted of three sets of lip-to-lip
Stingray spine—1 Aguilla Orange vessels with outcurved sides, placed
Shell artifacts lip-to-lip, and one cylinder jar with slightly outcurved-
Spondylus artifacts—2 recurved sides fitted with a small bowl-shaped vessel as
Marine bivalve—1
Possible mosaic inlays:
a lid. These vessels date to the Late Preclassic/Early
Hematite—14 Classic Transition period. Artifacts recovered from the
“White” shell—38 cache included two jade beads, two shell beads, and
45 Str. 3 public architecture Jade artifacts—24 three stone beads (Table 1).
Shell Cache 45 was found under a portico addition to
Spondylus—2 Structure 3, also on the Main Plaza at Blue Creek. This
Zebra—1 cache dates to the early part of the Late Classic
Oyster—3
period, somewhat later than the others in this study (L.
Cockle—2
Other modified shell—2 Kosakowsky, personal communication). The vessel is a
Stingray spines—numerous large lidded anthropomorphic jar depicting a portrait of
Fish vertebrae the Quadripartite God in the guise of the Sun God,
Starfish or sea urchin Kinich Ahau [22]. This vessel contained numerous jade
43 Str. U17 non-elite Jade beads—2 artifacts as well as coral, shell, sting-ray spines, etc.
residence Shell beads—2 (Table 1).
Stone beads—3 Cache 49, found during the 2000 field season under
49 Str. 34 elite residence Jade bead (smashed) Structure 34 in the Structure 37 Plazuela, is contempor-
Jade flakes—25 aneous with Tomb 7 and dates to the Late Preclassic/
Spondylus shell fragments—800+
Early Classic transition. Structure 34 was built sometime
Other marine shells—14
Coral—1 later, during the Early Classic, as a central plaza shrine,
Stingray spine—1 perhaps dedicated to the individual in Tomb 7, believed
Fish bones—3 to be the lineage founder of the residential group. The
Small mammal bone, unknown role of Cache 49, then, is somewhat murky: it may well
species—1
be a dedicatory cache for the entire plazuela group
Red ochre—1
rather than a single building. Cache 49 consists of
material in a lip-to-lip pair of Aguilla Orange bowls.
Recovered from Cache 49 were jade artifacts, marine
shells, coral, fish bones and red ochre (Table 1).
vessels and included a number of jade and shell artifacts
(Table 1).
2. Phytolith analysis
Cache 7 was a dedicatory cache in Structure 15 on the
west side of Plaza B. It consisted of a pair of Aguilla 2.1. Phytolith formation
Orange vessels with only a single bivalve shell inside.
Interestingly, the underside of the upper vessel had a Growing plants typically absorb water containing
mat-weave design inscribed on it, a motif commonly dissolved silica. Microscopic silica bodies are then
associated with Maya royalty. formed by the partial or complete silicification of plant
Caches 28 and 30 were Early Classic dedication cells, cell walls, and intercellular spaces. The resulting
caches in Structure 12, a relatively small, 4.6 meter tall, silica bodies which have characteristic shapes and sizes
public building in an open plaza [26]. Cache 28 consisted are called opal phytoliths [60]. The term phytolith is
of a pair of Aguilla Orange vessels placed lip-to-lip. derived from the Greek words phyton, meaning plant,
208 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
and lithos, meaning stone. Opal is the common name for
amorphous, hydrated silica dioxide.
Phytoliths form in most plants and are produced
in a multitude of shapes and sizes. They are diagnostic
when their shapes and/or sizes are specific to a particular
plant taxon. Fortunately, many phytoliths are resistant
to weathering and are preserved in most soils and
sediments for long periods of time.
3. Taxonomic classification of cultigen phytoliths
3.1. Maize
Much of the early archaeological phytolith research
on cultivated plants in the New World has focused on
maize because of its importance in many Prehispanic
cultures and the abundance of distinctive phytoliths
produced. Cross-shaped phytoliths are commonly pro-
duced in maize, but they are also formed, although less
commonly, in the Panicoid subfamily of grasses and a
few species in the Bamboo subfamily. This type of
phytolith consists of three or four lobes attached to a
central body and is easy to identify [48].
Research by Pearsall [45] on an extensive reference
collection of Panicoid grasses (62 species) native to
southwestern Ecuador demonstrated that only five taxa
produce large (16.0–20.5 µm) cross shaped phytoliths,
and that they were produced in low frequencies. Four
varieties of maize, in contrast, produced relatively abun-
dant large and a few extra large crosses (20.6–25.2 µm).
Measurements were based the on short-axis dimensions
(width). Cross-shaped phytoliths were defined as those
that are not more than 9 µm longer than wide.
A comparative analysis by Piperno [47] on maize and
grasses native to lower Central America and northern
South America demonstrated that a classification system
based on criteria of three-dimensional morphology, in
addition to size, should be used to identify cross-shaped
maize phytoliths in archaeological sites. It was also
shown that Variant 1 (mirror image) extra-large (20.6–
25.2 µm) cross-shaped phytoliths were absent from, and
Variant 1 large cross-shaped phytoliths occurred only
rarely in the many wild grass species studied. Identifying
cross-shaped maize phytoliths should not be applied to
areas where teosinte (Zea mays spp. mexicana), a wild
ancestor of maize, is native because certain races of
teosinte produce maize-like phytoliths [49].
Husks produce cross shapes that are predominately
Variant 6 (irregularly trapezoidal to rectangular). These
Variant 6 cross shapes are larger than those from leaves
and are extremely thick, averaging about 10.8 µm. Husk
phytoliths may be identifiable when isolated from
sediment if recovered in number [48].
Cobs are a particularly important part of the maize
plant to study in archaeological investigations, since ears
of corn would have been taken to Prehispanic villages.
Following removal of the grain, the cobs were either
discarded or used as a fuel source. Therefore, cob
phytoliths should be present in various archaeological
contexts in Mesoamerica.
A comparative study of phytoliths formed in tra-
ditional non-hybrid maize varieties from the Great
Plains and grasses native to the region demonstrated
that cobs produce numerous diagnostic phytoliths and
have been identified at prehistoric sites in the Great
Plains [4,6,10]. Taxonomic classification by Bozarth [8]
in the American Southwest showed that cobs of all five
types of maize (flint, flour, dent, sweet, and pop) pro-
duce diagnostic phytoliths. Moreover, cob phytoliths are
generally well-preserved and have been recovered from
several prehistoric sites in the Southwest [8,9,11,14].
A recent study by Bozarth [13] demonstrates that the
same types of phytoliths are formed in a variety of
popcorn from Mexico (PI 217409); PI #’s denote plant
introduction #’s of the USDA. Piperno et al. [52] report
similar phytoliths in cobs of other maize varieties from
Central America. Archaeological cob phytoliths have
been identified at Nakbé, Guatemala [16,39] and Blue
Creek, Belize [15].
Other taxonomically valuable phytoliths are also
formed in maize. A particular type of ridged bilobate
common in the aerial portion of maize, one with round
and/or pointed ends, is diagnostic of maize in Arizona
[8]. As with cob phytoliths, ridged bilobates are well-
preserved in most sediment and have been identified in
various sites in the American Southwest [8,9,14]. The
same type of ridged bilobate was recently found in husks
from a variety of popcorn from Mexico (PI 217409) and
was identified at Blue Creek [13].
3.2. Squash
The first phytolith study of squash fruit was on rinds
of selected varieties of Cucurbita maxima and C. pepo, in
addition to C. foetidissima (wild buffalo gourd). Analysis
of the rind phytoliths demonstrated that spheroidal
phytoliths with deeply scalloped surfaces of contiguous
concavities are produced in much higher frequencies in
squash than in the wild buffalo gourd and could be used
as indicators of squash in the central United States
[3,10].
Subsequent phytolith research demonstrated that
a variety of C. mixta (Hopi cushaw squash, Native
Seed/SEARCH #182) also produces the distinctively
scalloped spheroidal phytoliths in their rinds. Squash
rind phytoliths have been recovered from sediment
samples collected in several prehistoric sites in the
American Southwest [11,14].
Distinctive scalloped phytoliths are also produced in
squash varieties from northern Belize [13]. Varieties
studied include C. argyrosperma (PI 438546), C. moshata
(PI 438553), and C. pepo (PI 438696). A comparison of
 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
phytoliths formed in the rinds of these domesticated
Cucurbita spp. with those formed in C. lundelliana, the
only wild Cucurbita native to Belize [2], demonstrates
that the later can be differentiated by their asymmetrical
morphology, which consists of a hemisphere and a low
dome. The domesticated Cucurbita rind phytoliths are
more symmetrical, i.e., spherical. Squash phytoliths were
identified in Late Preclassic and Classic contexts at Blue
Creek [12,13,15] and archaeological sites in Panama and
Equador [50,51].
3.3. Gourds
A comparative phytolith analysis of a variety of
bottle gourd (Lagenaria siceraria) from the Southwest
with many reference materials from the region demon-
strated that large, distinctively scalloped, flat to globular
phytoliths are unique to bottle gourd rinds. They differ
from diagnostic squash phytoliths in they are not spheri-
cal and the scalloped areas are larger. Flat grainy plates
with angular-curvilinear edges are also formed in bottle
gourd rinds [8]. Scalloped Lagenaria phytoliths have
been identified at Nakbé, Guatemala [16,39], Blue
Creek, Belize [15], and Equador [51]. Taxonomically
valuable segmented hair cell phytoliths are formed in the
leaves of bottle gourds and squash. However, none are
solidly silicified, indicating that they may not preserve in
soil [48].
3.4. Beans
Cultivated beans (Phaseolus spp.) were a significant
part of Prehispanic diets in many areas. Distinctive
hooked-shaped silicified hairs are produced in pods of
common beans (Phaseolus vulgaris) and lima beans (P.
lunatus). Statistical analysis demonstrated that many of
these phytoliths are significantly wider near the tip than
similar silicified hairs produced in species native to the
Great Plains [5]. Bean phytoliths have been identified in
several archaeological sites in the Great Plains [4,10] and
the American Southwest [9] based on this distinguishing
characteristic.
Similar phytoliths are formed in leaves and pods of
domesticated varieties of P. vulgaris from Mexico (PI
533312) and Guatemala (PI 163584), as well as a wild
variety of P. vulgaris from Guatemala (W 620509) [13].
Phaseolus phytoliths have been identified at Blue Creek
[12,15]. The similarity of hook-shaped phytoliths formed
in domesticated and wild P. vulgaris does not necessarily
present an archaeological problem in distinguishing one
from the other because wild P. vulgaris does not occur in
much of Mesoamerica, including Belize [25].
3.5. Palms
Palms produce abundant distinctive phytoliths that
are readily identifiable in the phytolith record. Two
209
different types of phytoliths are formed in palms, “hat-
shaped” and spinulose spheres. Several palm genera
produce hat-shaped phytoliths in their leaves [58].
Species in those genera that occur in the study area
include Acoelorraphe wrightii, Acrocomia aculeata,
Astrocaryum mexicanum, Bactris gasipaes, B. major,
Chamaedorea oblongata, C. pinnatifons, C. seifrizii, and
C. tepejilote, Desmoncus orthacanthus, and Gaussia maya
[41].
Edible fruit is produced in four palm species
which form hat-shaped phytoliths, including Acrocomia
aculeata, Astrocaryum mexicanum, B. gasipaes, and B.
major [41,42]. Acrocomia fruit can also be used to extract
oil, and its carbohydrate-rich sap can be fermented to
produce a palm wine [41,42]. Astrocaryum mexicanum
produces edible young inflorescences as well as fruits.
Additionally, its leaves are used for thatching. Bactris
gasipaes is not known as a wild plant but is widely
cultivated throughout humid areas of Central America.
Chamaedorea tepejilote is commonly cultivated in
Guatemala for its edible male inflorescences. Desmoncus
orthacanthus stems are used for making baskets. Stems
of G. maya are used in construction [41].
Several other palm genera produce another archaeo-
logically important type of phytolith: the spinulose
sphere [58]. Several species in these genera are native
to the study area, including Attalea butyracea, A.
cohune, Cryosophila stauracantha, Geonoma interrupta,
Reinhardtia gracilis, Roystonea regia, Sabal mauritii-
formis, and S. yapa [41]. None of these palms produce
edible fruit, although oil can be extracted from A.
cohune and R. regia. Other economic uses include thatch
made from fronds of A. butyracea, A. cohune, R. regia,
and S. mauritiiformis [41].
3.6. Bromeliads
Bromeliads produce spinulose spheres much like
those in certain palms, with which there is overlap in
size. A study by Bozarth [7] demonstrated that pineapple
produces spinulose spheres that range up to 10 µm in
diameter.
3.7. Other food producing plants
In addition to palms, there were many other culti-
vated and wild fruits, nuts, leaves, and tubers important
in the prehispanic diet in Central America. Distinctive
phytoliths formed in the fruits and seeds of many
dicotyledon families provide a promising area of
research, as none have been found in leaves and many
appear unique to genera and even species. These phyto-
liths usually have a stippled surface and protuberance
emanating from the center of the bottom of the cell.
Furthermore, production of fruit and seed phytoliths
can be quite high [48]. Similar phytoliths in the form of
210 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
wavy edged round plates with protuberances and
verrucate sculptural elements were found in copal fruits
[15].
Three Neotropical plant species domesticated for
their roots or tubers produce taxonomically significant
phytoliths: achira (Canna edulis), arrowroot (Maranta
arundinaceae), and chufa (Cyperus esculentus). Achira is
usually not considered to be a major root crop, but
circumstantial evidence suggests that it may have been
more important in South American prehistory. Arrow-
root and chufa are minor cultigens in indigenous econ-
omies of Central and South America [49] and occur in
Belize [2].
Taxonomically useful phytoliths are formed in chufa
seeds [49]. Achira leaves produce numerous, finely
rugulate–verrucate, spheroidal phytoliths which have
not been seen in other plant materials. Rugulate–
nodular spheres and rugulate conical phytoliths are
formed in the leaves of arrowroot. The later type may be
unique to arrowroot leaves. More importantly, very
distinctive, and probably diagnostic, large globular
bodies with branching processes are formed in arrow-
root seeds.
3.8. Fiber plants
Fiber plants were also important cultigens in
Mesoamerica. Two species of cotton were domesticated
prehistorically in the New World: upland cotton
(Gossypium hirsutum) in Mesoamerica and North
America and sea island cotton (G. barbadense) in South
America [40]. Non-diagnostic phytoliths are formed in
leaves of G. barbadense. Phytoliths have not been found
in its pods or fiber [48].
Four Agave species were also cultivated for fiber in
Mesoamerica [40], one of which, Agave sisalana, occurs
in Belize [2]. Sisal produces phytoliths similar to those in
A. murpheyi, a species cultivated prehistorically in the
American Southwest for food and fiber that produces a
type of opal phytolith characteristic of the genus [9].
4. Methodology
Ten sediment samples from nine cache vessels were
analyzed for phytoliths and other biosilicates. Seven of
these caches were from monumental architecture in the
Core Area and one each derived from excavations at
Chan Cahal and the Structure 37 Plazuela.
Biosilicates were isolated from 10-g samples based on
a heavy-liquid (zinc-bromide) flotation and centrifuga-
tion procedure. The extraction procedure consists of
seven basic steps: (1) removal of carbonates with dilute
hydrochloric acid; (2) removal of colloidal organics,
clays, and very fine silts by deflocculation with sodium
pyrophosphate, centrifugation, and decantation through
7-µ filter; (3) oxidation of sample to remove organics;
(4) introduction of spike spores for the calculation of
phytolith concentration; (5) heavy-liquid flotations of
phytoliths from the heavier clastic mineral fraction using
zinc bromide concentrated to specific gravities of 2.4 and
2.3; (6) washing and dehydration of phytoliths with
butanol; and (7) dry storage in 1-dram vials.
After thorough mixing, a representative portion of
the isolate was mounted on a microscope slide in immer-
sion oil under 2440 mm cover glasses and sealed with
clear nail lacquer. A minimum of 200 phytoliths were
taxonomically classified at the finest level possible with a
Zeiss microscope at a magnification of 625 in samples
with adequate preservation and concentration (at least
1000 phytoliths/gram of sediment processed). Other
biosilicates (sponge spicules, statospores, and sponge
spicules) were also counted. At least 2000 additional
phytoliths were then scanned for economic species. An
entire slide was studied in samples with low phytolith
concentration. Culturally significant data found in the
scanning beyond that encountered during the classifica-
tion are reported with numbers in the appropriate
graphs.
Phytoliths were taxonomically classified based on
phytolith systematics reported by Piperno [48] and
analysis of the University of Kansas phytolith reference
collection of Central American flora, which was
originally established for an archaeological study in
Costa Rica [7]. Since then it has been augmented with
the addition of leaves/fruits from 39 arboreal species and
nine herbaceous dicots which are native to or cultivated
in the study area. The phytolith reference collection
currently consists of 47 arboreal species from 27 families
and 22 herbaceous species from 12 families for a total of
156 reference slides of leaves, fruits, and tubers.
Calculation of microfossil concentrations were made
using a method based on a known number of spike
spores, in this case Lycopodium, being added to
each sample before flotation. The concentration of
microfossils/gram was computed as follows: microfossil
concentration=no. of microfossils countedtotal no. of
exotics added/no. exotics counted/10. Concentration
permits an evaluation of the phytolith production and
preservation for a given sample interval. Biosilicate
frequencies were generated and graphically illustrated
using Tilia and Tilia graph software [27].
5. Results and discussion
Biosilicates were generally well-preserved and
adequately concentrated for analysis in all samples pro-
cessed. Even though the vessel residues were relatively
small, some of the sediment apparently leaked into
the vessels based on the high frequencies of phyto-
liths typical of the natural vegetation. The following
 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
Fig. 1. Biosilicate freqencies of Maya dedicatory cache vessels from Blue Creek, Belize.
discussion focuses on the biosilicate remains of plants
and sponges evidently placed in the vessels.
5.1. Caches 4A, 4B, and 6A
Analysis of three Early Classic caches from Structure
1 revealed several interesting cultural practices (Fig. 1).
First and foremost, numerous sponge spicules in each of
the three vessels show that sponges were placed in each
vessel, undoubtedly as part of an offering. In addition,
the relatively high frequency of hat-shaped palm phyto-
liths in Cache 4A indicates that palm fruit, leaves,
and/or inflorescences were also placed in that vessel.
Given the ceremonial context, it seems most likely that
edible fruits were placed in the vessel. Of the four species
that produce hat-shaped phytoliths and edible fruit, B.
gasipaes is the most likely species represented as it is the
only one that is cultivated for its fruit [41]. Additional
evidence of cultural activities was found in Cache 6A, in
which Heliconia and Agave opal phytoliths were ident-
ified. The Heliconia phytolith may be from a wrapping
for food because Heliconia produces banana-like leaves
used for that purpose [55]. The agave phytolith is
probably from fiber used to make some type of woven
bag like those used in Chiapas by the modern Maya.
Dicot seeds also may have been placed in Cache 6A,
211
based the similarity of a phytolith found in the vessel
with a type illustrated by Piperno [48].
5.2. Cache 7
A single sample was processed from Structure 15,
Cache 7. Sponges were apparently placed in the vessel
based on culturally elevated frequencies of sponge
spicules (Fig. 1). Their relatively low frequency indi-
cates, however, that fewer sponges were offered in Cache
7 than in Caches 4A, 4B, and 6A. In addition, four opal
phytoliths characteristic of agave were identified in this
sample. This taxonomic classification is supported by
the high frequency of smooth-sided long cells, formed in
both agave and native grass, but without significant
numbers of other types of grass phytoliths. As in Cache
6A, something made from agave fiber was placed in the
vessel.
5.3. Caches 28 and 30
These two Early Classic caches in Structure 12 yielded
high frequencies of sponge spicules demonstrating that
sponges were also placed in these vessels as part of a
dedicatory offering (Fig. 1). Spicules were particularly
numerous in vessel 30. Red ochre was seen in both
212 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215

samples after the carbonates and clays had been

removed. The relative abundance of polyhedral phyto-
liths, as well as silicified dicot stomata and tracheids, in
Cache 28 shows that dicot leaves were also part of the
offering. The high frequency of smooth spheres less than
10 µm in diameter is particularly interesting in that this
phytolith type does not occur in abundance in any other
sample. The spheres are similar to those formed in
Hirtella triandra fruits [48]. Squash was also part of the
offering based on a domesticated Cucurbita phytolith
found in the same vessel. A single fruit phytolith similar
to that reported by Piperno [48] indicates that fruit was
also likely placed in the vessel, but additional research is
necessary to taxonomically identify the species.

5.4. Cache 43

Two samples were processed from a cache vessel at

U-17, a house mound midden at Chan Cahal, a residen-
tial area east of the Bravo Escarpment. Three maize cob
phytoliths were found in each sample, indicating that
ears of corn were placed in the vessel (Fig. 1). Two
squash phytoliths were also discovered in sample 88,
showing that at least part of a squash fruit with the rind
was put in the vessel. A fruit phytolith was also found in
sample 90.

5.5. Cache 45

Two samples were processed from Structure 3, one of

which was from Cache 45 (a Classic period, large lidded
ceremonial vessel). This vessel was surrounded by
organically enriched sediment that resulted from a Late
Classic termination ritual. The vessel isolate was amaz-
ing in that it included tens of thousands of sponge
spicules, with virtually no other biosilicates present (Fig.
1). The abundant spicules demonstrate that many
sponges were placed in the vessel, undoubtedly as part of
an offering. As in Caches 28 and 30B from Structure 1,
red ochre was seen after the carbonates and clays had
been removed. The purity of the isolate demonstrates
that there was no significant leakage of the surrounding
fill into the vessel. A Heliconia phytolith found in the
surrounding sediment indicates that something (perhaps
copal) was wrapped in platanillo leaves before being
burned in the termination ritual.
5.6. Cache 49
A single sample was processed from Structure 34,
Cache 49. The biosilicate assemblage of this sample was
as astonishing as Cache 45 in that it also consists
virtually entirely of large numbers of sponge spicules
(Fig. 1). Moreover, red ochre was observed in the
sample during processing (Fig. 2).
Fig. 2. Red ochre in residue from Cache 39, Structure 34, after removal
of carbonates and clays (photo by S. R. Bozarth).
6. Summary
This biosilicate analysis identified various organic
materials placed in dedicatory cache vessels by the Late
Preclassic and Early Classic Maya of Blue Creek, Belize.
The most unexpected discovery was the offering of
marine sponges in all of the vessels from the core area, as
well as the single vessel from an elite plazuela. More-
over, red ochre was found in the matrices of four of
these vessels, including three in which the biosilicate
assemblages consisted almost entirely of sponge spicules.
Phytolith analysis demonstrated that food was also
placed in many of the vessels. Ears of corn and squash
were placed in the single cache vessel from Chan Cahal,
based on the presence of cob and squash rind phytoliths.
Squash and unclassified fruit and tree leaves were put in
one of the core area vessels (Cache 28). Palm fruits were
evidently part of an offering in another core area vessel
(Cache 4A). Unclassified dicot seeds and agave, possibly
from a woven bag, were placed in Cache 6A in the same
structure. A Heliconia phytolith from the same vessel
 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
may represent a food wrapping. Numerous agave-type
phytoliths were found in Cache 7.
Importantly, the caches we examined derive from
several contexts including the ritual dedications of pub-
lic buildings, residences of some of Blue Creek’s most
elite families, and residences of some of Blue Creek’s
least elite families. With the added dimension of bio-
silicate analysis, we have determined that each cache
included valuable elements of both the land and the sea.
In terms of the Maya Cosmos of Creation, these
elements represent the Primordial Sea and the First
Mountain. The very form of the caches, all with domed
lids, represents the Sky of the Creation. Further, the
data presented in this paper support the concept that
sponges were used to represent the Primordial Sea and
each cache was a symbolic representation of the Maya
cosmos. Not only can we argue that these dedicatory
rituals reflect the Cosmos, but that the Blue Creek Maya
of nearly 2000 years ago shared this perception across all
social strata. In public as well as private rituals, they
re-enacted the Creation itself when they placed caches
within their structures.
The success of the phytolith analysis of residue from
dedicatory caches in public architecture is truly import-
ant. Until now, Maya archaeologists have been very
restricted in their ability to interpret the meaning of such
caches. While their function in building dedication
seems clear, their symbolic purposes are more obscure.
In general, Mayanists argue that such ritual actions
reflect the embedding of sacredness in public architec-
ture. However, they have variously been considered to
be actions that gave “physical expression to the pipeline
between humans and their gods” [46] or reflections of
the cosmos and the act of embedding sacredness to
public space [17,24,54]. Regardless of the future out-
come of such discussion, it is clear that, until now,
archaeologists have labored without access to an
important part of the database. With the findings of this
study, it now becomes possible to more fully understand
the residues found in such dedicatory caches and other
ritual settings.
Acknowledgements
Fieldwork on the Blue Creek project is conducted
under permit by the Government of Belize. We thank
the Department of Archaeology for their support over
the past decade and especially for expediting the export
of samples during the 1998 field season. We also thank
the Foundation for the Advancement of Mesoamerican
Studies, Inc. and Mr Kim Cox for their funding
of the biosilicate study. We are also grateful to Bill
Collins, whose meeting with Dr Bozarth at Nakbé,
Guatemala, brought the authors into contact with each
other.
213
This analysis would not have been possible without a
phytolith reference collection of plants native to the
study area. The senior author is grateful to the following
persons for providing leaves/fruits: Bob Jarret, USDA,
Plant Genetics Resources Unit, Griffin, Georgia; Molly
Welsh, USDA, Western Regional Plant Introduction
Station, Pullman, Washington; Tom Wendt, Curator,
University of Texas Herbaria; James Solomon, Curator
of the Herbarium, Missouri Botanical Garden; and
Patricia Holmgren, Director of the Herbarium, New
York Botanical Garden.
References
[1] W. Ashmore, Site planning principles and concepts of direction-
ality among the ancient Maya, Latin American Antiquity 2 (1991)
199–226.
[2] M.J. Balick, M.H. Nee, D.E. Atha, Checklist of the vascular
plants of Belize with common names and uses, Memoirs of the
New York Botanical Garden 85 (2000).
[3] S.R. Bozarth, Diagnostic opal phytoliths from rinds of selected
cucurbita species, American Antiquity 52 (1987) 607–615.
[4] S.R. Bozarth, Opal phytoliths, in: W. Lees, J. Reynolds, T.J.
Martin, M. Adair, S. Bozarth (Eds), Final Summary Report, 1986
Archaeological Investigations at 14MN328, a Great Bend Aspect
Site Along U.S. Highway 56, Marion, Kansas, Archeology
Department, Kansas State History Society, 1989, pp. 85–90.
[5] S.R. Bozarth, Diagnostic opal phytoliths from pods of selected
varieties of common beans (Phaseolus vulgaris), American
Antiquity 55 (1990) 98–104.
[6] S.R. Bozarth, Maize (Zea mays) cob phytoliths from a central
Kansas Great Bend aspect archaeological site, Plains
Anthropologist 38 (1993) 279–286.
[7] S.R. Bozarth, Opal phytolith analysis of shell middens at
Sites P-309B and P-313N, in: J. Hoopes (Ed.), Prehistoric
Human Ecology on the Golfo Dulce, Southwestern Costa Rica,
Department of Anthropology, University of Kansas, 1993, in
preparation..
[8] S.R. Bozarth, Pollen and phytolith analysis, in: R. Ciolek-
Torrello, J.R. Welch (Eds), The Roosevelt Rural Sites Study,
Changing Land Use in the Tonto Basin, Volume 2: Statistical
Research Technical Series No. 28, 1994, pp. 189–222.
[9] S.R. Bozarth, Pollen and phytolith analysis, in: J. Homburg, R.
Ciolek-Torrello (Eds.), Vanishing River—Landscapes and Lives
of the Lower Verde Valley, the Lower Verde Archaeological
Project, Volume 2: Agriculture, Subsistence, and Environmental
Studies, SRI Press, Tucson, Arizona, 1997, pp. 179–204.
[10] S.R. Bozarth, Opal phytolith evidence of Great Bend aspect
agriculture on the Middle Arkansas River, in: C. Schoen (Ed.),
Arkansas City Archaeological Project, Kansas State Historical
Society, Topeka, 1998, in preparation.
[11] S.R. Bozarth, Microfossil evidence of agriculture and wild plant
utilization, in: C.A. Turnbow (Ed.), A Highway Through Time:
Archaeological Investigations Along NM 90 in Grant and
Hidalgo Counties, New Mexico, New Mexico State Highway
and Transportation Department Technical Series 2000–3,
Albuquerque, 2000, pp. 573–588.
[12] S.R. Bozarth, Biosilicate Evidence of Cultivated Plants and
Ceremonial Activities at Blue Creek, Belize. Abstracts of the 66th
Annual Meeting, Society of American Archaeology, New
Orleans, Louisiana, 2001.
[13] S.R. Bozarth, Pollen and Biosilicate Investigations of Prehispanic
Fields and Biosilicate Analysis of Tombs at Blue Creek, Belize,
214 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
report submitted to Maya Research Program, Texas Christian
University, Fort Worth, 2001.
[14] S.R. Bozarth, Appendix J: pollen and opal phytolith analysis, in:
E.E. Klucas, R. Ciolek-Torrello, R. Vanderpot (Eds), The State
Route 87—Sycamore Creek Archaeological Project. Volume 2:
Analyses of Prehistoric Remains, Technical Series No. 73,
Statistical Research, Tucson, 2003.
[15] S.R. Bozarth, T.H. Guderjan, Evidence of agriculture and
ceremonies based on biosilicate analysis, in: T.H. Guderjan, R.J.
Lichtenstein (Eds.), The Blue Creek Project: Working Papers
from the 1998 and 1999 Field Seasons, Maya Research Program,
Texas Christian University, Fort Worth, 2002, pp. 101–115.
[16] S.R. Bozarth, R.D. Hansen, Estudios Paleo-Botánicos de Nakbé:
Evidencias Preliminares de Ambiente y Cultivos en el Preclásico,
in: J.P. Laporte, A.C. de Suasnavar, B. Arroyo (Eds), XIV
Simposio de Investigaciones Arqueologicas en Guatemala, Museo
National de Arqueologia y Etnologia, Ministerio de Cultura y
Deportes, Insituto de Antropologia e Historia, Asociacion Tikal,
2001, pp. 419–436.
[17] D.Z. Chase, Caches and censorwares: meaning from Maya
pottery, in: C.C. Kolb, L.M. Lackey (Eds.), A Pot for all
Reasons: Ceramic Ecology Revisited, Laboratory of Archae-
ology, University of Philadelphia, Pennsylvania, 1988,
pp. 81–104.
[18] H.L. Clagett, Household archaeology at Chan Cahal: a step
beyond functionalism, in: W.D. Driver, H.L. Clagett, H.R.
Haines (Eds.), The Blue Creek Project: Working Papers from the
1996 Field Season, Maya Research Program, St Mary’s
University, San Antonio, Texas, 1997, pp. 61–70.
[19] W.R. Coe, Piedras Negras archaeology: artifacts, caches,
and burials, Museum Monograph 4, The University Museum,
University of Pennsylvania, 1959.
[20] W.D. Driver, An early classic columned temple at Blue Creek,
Belize, Latin American Antiquity 13 (2002) 63–85.
[21] W.D. Driver, H.R. Haines, T.H. Guderjan (Eds), The Blue Creek
Project: Working Papers from the 1997 Season, Maya Research
Program, St Mary’s University, San Antonio, Texas, 1999.
[22] W.D. Driver, P. Wanyerka, Creation symbolism in the architec-
ture and ritual at Structure 3, Blue Creek, Belize, Mexicon 24
(2002) 6–8.
[23] D.L. Freidel, L. Schele, J. Parker, Maya Cosmos, Morrow and
Co, New York, 1993.
[24] J.F. Garber, W.D. Driver, L. Sullivan, D.M. Glassman, Bloody
bowls and broken pots: the life, death and rebirth of a Maya
house, in: S.B. Mock (Ed.), The Sowing and the Dawning:
Dedication and Termination Ritual Events in the Archaeology
and Ethnology of Mesoamerica, University of New Mexico Press,
Albuquerque, 1998, pp. 125–134.
[25] H.S. Gentry, Origin of the common bean, Phaseolus vulgaris,
Economic Botany 23 (1969) 55–69.
[26] E. Gilgan, Excavations in the structure 13 courtyard group, in:
W.D. Driver, H.L. Clagett, H.R. Haines (Eds.), The Blue Creek
Project: Working Papers from the 1996 Field Season, Maya
Research Program, St Mary’s University, San Antonio, Texas,
1997, pp. 33–38.
[27] E.C. Grimm, Tilia and Tilia Graph: PC Spreadsheet and
Graphics Software for Pollen Data. Illinois State Museum,
Research and Collections Center, Springfield, Illinois, 1992.
[28] T.H. Guderjan, The Blue Creek jade cache: early classic ritual in
northwestern Belize, in: S.B. Mock (Ed.), The Sowing and the
Dawning: Termination, Dedication, and Transformation in
the Archaeological and Ethnographic Record of Mesoamerica,
University of New Mexico Press, San Antonio, Texas, 1998,
pp. 100–111.
[29] T.H. Guderjan, Recreating the Cosmos. Early Classic Dedicatory
Caches at Blue Creek. Acta Mesoamerica, Verlag von Saurwein,
Markt Schwaben, Germany, in press.
[30] T.H. Guderjan, Implementing multiple research strategies: a case
study from the Maya site of Blue Creek, Belize, in: B. Morrison,
J. Mathews (Eds), Lifeways in the Lowlands, University of
Arizona Press, manuscript.
[31] T.H. Guderjan, Patterns of Maya Jade Disposal at Blue Creek,
Latin American Antiquity, Belize. Unpublished manuscript,
Maya Research Program, Texas Christian University, Ft. Worth,
Texas.
[32] T.H. Guderjan, J. Baker, R.L. Lichtenstein, Environmental and
cultural diversity at Blue Creek, in: V. Scarborough, N. Dunning,
F. Valdez (Eds), Heterarchy, Political Economy, and the
Ancient Maya, University of New Mexico Press, Albuquerque, in
press.
[33] T.H. Guderjan, W.D. Driver (Eds), Archaeological research at
Blue Creek, Belize. Progress Report of the Third (1994) Field
Season. Maya Research Program, St Mary’s University, San
Antonio, Texas, 1995.
[34] T.H. Guderjan, C.C. Hanratty, A Thriving Non-Royal Lineage at
Blue Creek. Evidence from a Sequence of Burials, Caches, and
Architecture. Acta Mesoamerica, Verlag von Saurwein, Markt
Schwaben, Germany, in press.
[35] T.H. Guderjan, R. Lichtenstein (Eds), The Blue Creek Project:
Working Papers from the 1998 & 1999 Seasons, Maya Re-
search Program, Texas Christian University, Ft. Worth, Texas,
2002.
[36] T.H. Guderjan, R.J. Lichtenstein, C.C. Hanratty, Elite residences
at Blue Creek, Belize, in: J. Christie (Ed.), Maya Palaces and Elite
Residences, University of Texas Press, Austin, Texas, 2003,
pp. 13–45.
[37] C.C. Hanratty, Late and Terminal Classic Dynamics with-
in the Structure 37 Plazuela, paper presented at the Annual
meeting of the Society for American Archaeology, Seattle,
1999.
[38] C.C. Hanratty, Excavations at the structure 37 Plazuela: Late and
Terminal Classic dynamics, in: T.H. Guderjan, R.L. Lichtenstein
(Eds.), The Blue Creek Project: Working Papers from the 1998
and 1999 Field Seasons, Maya Research Program, Fort Worth,
Texas, 2002, pp. 73–80.
[39] R.D. Hansen, S.R. Bozarth, J. Jacob, D. Wahl, T.
Schreiner, Climatic and environmental variability in the rise of
Maya civilization—a preliminary perspective from Northern
Peten, Ancient Mesoamerica 13 (2002) 273–295.
[40] J.R. Harlan, Crops and Man, third ed, American Society of
Agronomy, Inc. and Crop Science Society of America, Inc,
Madison, Wisconsin, 1992.
[41] A. Henderson, G. Galeano, R. Bernal, Field Guide to the Palms
of the Americas, Princeton University Press, Princeton, New
Jersey, 1995.
[42] D.H. Janzen, Species accounts—Acrocomia vinifera (Coyol), in:
D.H. Janzen (Ed.), Costa Rican Natural History, University of
Chicago Press, Princeton, New Jersey, 1983, pp. 184–185.
[43] R.J. Lichtenstein, Settlement Zone Communities of the Greater
Blue Creek Area, Occasional Paper 2, Maya Research Program,
Fort Worth, Texas, 2000.
[44] J.P. Mathews, J.F. Garber, Models of Cosmic Order: The
Physical Expression of Sacred Space among the Ancient Maya,
Ancient Mesoamerica, in press.
[45] D.M. Pearsall, Phytolith analysis: applications of a new
Paleoethnobotanical technique in archeology, American
Anthropologist 84 (1982) 862–871.
[46] D.M. Pendergast, Intercessions with the gods: caches and their
significance at Altun ha and Lamanai, Belize, in: S.B. Mock (Ed.),
The Sowing and the Dawning: Dedication and Termination
Ritual Events in the Archaeology and Ethnology of
Mesoamerica, University of New Mexico Press, Albuquerque,
1998, pp. 55–64.
 S.R. Bozarth, T.H. Guderjan / Journal of Archaeological Science 31 (2004) 205–215
[47] D.R. Piperno, A comparison and differentiation of phytoliths
from maize and wild grasses: use of morphological criteria,
American Antiquity 49 (1984) 361–383.
[48] D.M. Piperno, Phytolith Analysis—An Archaeological and
Geographical Perspective, Academic Press, San Diego,
California, 1988.
[49] D.M. Piperno, The status of phytolith analysis in the American
tropics, Journal of World Prehistory 5 (1991) 155–191.
[50] D.M. Piperno, D. Pearsall, The Origins of Agriculture in the
Lowland Neotropics, Academic Press, San Diego, 1998.
[51] D.M. Piperno, T.C. Andres, K.E. Stothert, Phytoliths in
Cucurbita and other neotropical Cucurbitaceae and their occur-
rence in early archaeological sites from the lowland American
tropics, Journal of Archaeological Science 27 (2000) 193–208.
[52] D.M. Piperno, I. Holst, A. Ranere, P. Hansell, K. Stothert, The
occurrence of genetically controlled phytoliths from maize cobs
and starch grains from maize kernels on archaeological stone
tools and human teeth, and in archaeological sediments from
southern Central America and northern South America, The
Phytolitharien 13 (2001) 1–7.
215
[53] C. Popson, Prehistoric Political Economies at the Chan
Cahal barrio, Blue Creek, Belize, M.A. Thesis, Department of
Anthropology, SUNY-Albany, 2001.
[54] L. Schele, D.A. Freidel, A Forest of Kings, Morrow and Co, New
York, 1990.
[55] F.G. Stiles, Species accounts—Heliconia latispatha (Platanillo),
Wild Plaintain, in: D. Janzen (Ed.), Costa Rican Natural History,
The University of Chicago Press, New York, 1983, pp. 249–251.
[56] D. Tedlock, Popol Vuh: The Definitive Edition of the Maya Book
of the Dawn of Life and the Glories of God and Kings, Simon
and Schuster, New York, 1985.
[57] J.E.S. Thompson, Excavations at San Jose, British Honduras,
Carnegie Institute of Washington, Washington, DC, 1939.
[58] P.B. Tomlinson, in: C.R. Metcalfe (Ed.), Anatomy of the
Monocotyledons II. Palmae, Clarendon Press, Oxford, 1961.
[59] E.Z. Vogt, Torillas for the Gods: A Symbolic Analysis of
Zincanteco Rituals, Harvard University, Cambridge, MA, 1976.
[60] L.P. Wildling, L.R. Drees, Biogenic opal in Ohio
soils, Proceedings of the Soil Science Society of America. 35
(1971) 1004–1010.