Contributed REPORT CR-10-A
Arizona Geological Survey
416 W. Congress St. #100 Tucson, Az
www.azgs.az.gov

Geology of Selected Areas, Sawmill Canyon Fault Zone
Northeast Santa Cruz County, Arizona
Langford G. Brod, Jr.

2010

GEOLOGY OF SELECTED AREAS,
SAWMILL CANYON FAULT ZONE
NORTHEAST SANTACRUZ COUNTY, ARIZONA
BY LANGFORD G. BROD JR.

2005

t ':'.'

/

Table Of Contents
page
i

ii

contents
Table Of Contents
List Of Illustrations

1
2
3
8
12
13
17
21
23
24
26
27
29
30
32
34
35
39
40
41
42

General Introduction
The Geology Of Apache Ridge - Introduction
stratigraphy Of Apache Ridge
structure Of Apache Ridge
The Geology Of Lower Sawmill Canyon - Introduction
Stratigraphy Of Lower Sawmill Canyon
Structures Of Lower Sawmill Canyon
Geology Of The Resurgence Area (Cave Creek)
Geology Of Hilton Ridge - Introduction
Stratigraphy Of Hilton Ridge
Structure Of Hiltpn Ridge
Cryptic outcrops and structure on Hilton Ridge
Geology Of Bells Ridge - Introduction
stratigraphy Of Bells Ridge
Structure Of Bells Ridge
Geology Of Onyx Hill - Introduction
Stratigraphy Of Onyx Hill
Structure Of Onyx Hill
Surficial Deposits
Terraces In The Vicinity Of Onyx Hill
Hydrology Of Onyx Hill

43
44
47
48

References
Appendix 1 - Methodology
Appendix 2 - Survey Markers
Appendix 3 - White Limestone Bed, Eastern Apache Ridge

List Of Illustrations
after
page
Figure
1.'

2.
3.

4.
5.
6.

7.
8.
9.
10.
11.

12.
13.

14.
15.

Title
Location Of study Area . . . . . • . . . . • . . . .
study Areas In This Report . . . . . . .
....
Division Of Apache Ridge Geologic Area .
...•
Resurgence Through Thin Limestone Bed.
•
Sinking stream In Lower Sawmill Canyon •
. • . .
Strike And Dip Across NW Hilton Ridge. . • . •
.
Dip Angles On Traverse, Northwest Hilton Ridge .
Geology Of Enlarged Area On Hilton Ridge . . .
.
Topographic Map Of Enlarged Area, Hilton Ridge .
Structural Domains on Hilton Ridge . . . .
.
Geologic Profile At Hiltons Hole on Hilton :Ridge . .
Erosion Surfaces Along Cave Creek. . . . . . . • . .
stream Hydrology Along Cave Creek, Near Onyx Cave. .
Resurgence Area Along Gardner Creek. • .
....
Stratigraphic Column For Selected areas . . . . . • •

Larger Geologic Maps, Unnumbered, In Separate Envelope
Geology of Apache Ridge
Geology Of Lower Sawmill Canyon
Geology Of Resurgence Area
(This map is a southeastward extension
of the Lower Sawmill Canyon map.
Geology Of Hilton Ridge
Geology Of Bells Ridge, at Cave Of the Bells
Geology Of Onyx Hill, at Onyx Cave

1
1
2
22
22
26
26
27
27
27
28
41
42
42
42

Geology Of Selected Areas, Sawmill Canyon Fault Zone,
Northeastern Santa Cruz county, Arizona
by Langford G. Brod Jr.
General Introduction
In 1991, with the encouragement of the u.S. Forest Service, I
and a number of members of Escabrosa Grotto ( a chapter of the
National Speleological Society) began mapping a large cave, Cave Of
The Bells in northeastern Santa Cruz county, Arizona. While
entering the cave on 20 February 1994, I noted faulting in the
entrance rock. On the same day, I decided to begin a survey of the
surface geology.
The survey was aided py a geologic study
published by the united states Geological Survey: The structural
Geology of the Santa Rita Mountains, Southeast of Tucson, Arizona
(1972), by Harald Drewes.
In sUbsequent years I surveyed a number of areas lying along
the same trend, a section of Sawmill Creek. The next area was on
the north side of Forest Road 92, at the extreme eastern limit of
Paleozoic and Mesozoic rocks, and adjacent to Apache Spring Ranch.
I termed this area Apache Ridge, after the ranch and spring.
The next area to be surveyed was a somewhat larger area lying
between the confluence of Sawmill Creek and the next stream
crossing where Forest Road 4085 crosses Sawmill Creek. I termed the
map area Lower Sawmill Creek. The next area mapped was on the
north-northwest side of FR-4085, Which I named Hilton Ridge after
the small cave near the eastern end of the area.
The next area surveyed was the Onyx Cave area, about 0.8 mile
(1.2 km) south of Cave Of The Bells, which I named Onyx Hill. Thus,
five major map areas were prepared. along with several auxiliary
maps showing related areas of interest related to the primary maps.
All major areas were surveyed at a scale of 50 feet per in inch and
later reduced to 100 and 200 feet per inch; several auxiliary areas
were surveyed at scales of 10 and 20 feet per inch to show details.
After finishing surveying of Onyx Hill additional features were
surveyed in several of the other map areas to clarify certain areas
and fill in blanks. The general location is shown in Figure 1; the
five major areas are shown in Figure 2.
The five major areas are discussed separately, for the most
part, except in those cases where features in one map area may bear
upon features in another map area. All areas were surveyed on foot
to obtain the best accuracy. In a few cases, the position of
certain features or the identification of formations was obscure;
in those cases I used my best jUdgment. To facilitate accuracy,
multiple station lines and triangulation was used in many areas. In
many areas, geographic and man-made features are included to permit
geologic features to be found, if required. On the Apache Ridge map
area, three Forest Service survey markers were found, two along the
road (FR-92) and one not far from the Arizona Trail; details on
these markers are included in an appendix.
PAGE

I

Tucson"

N

10
r-!

..

!-10

.

10
C\I

o

r-!
r-!

!-10

i
I

~!
g

u

jo

Iu

Q)

i
p.i
~

'';

l/)

.r;

.s::::
t>

o

u.

I

I

I
I

I
I

I.
I.

Pima county

----~----------l
Hwy. 82
••

.

SantaCruz Co.

stUd~J·

to .Sierra vista

U

Area

co

Hwy. 83

LOCATION

I a
E!~

a

U

I
I

.~

'§

~ I 8
CIl

I

OF STUDY AREA IN SOUTHEASTERN ARIZONA
FIGURE
\

SEC. 35 \

~EC. 36

R i5 E

R 16 E

T 19S

SEC. 31

T20S

\"SECTlONB-B' OF PLATE 3
PROFESSIONAL PAPER 748

RESURGENCE AREA,
CAVE CREEK

INDEX FOR GEOLOGIC MAPS WITHIN THE SAWMILL CANYON FAULT ZONE, S. ARIZONA

FIGURE

2

L. G. BROD JR. 1994-2003

The Geology Of Apache Ridge - Introduction
Apache Ridge is a name that was chosen by the author to
describe a topographic feature, previously unnamed, lying north of
Apache spring, The authenticity of the name applied to the spring is
unknown, and the possible presence of the Apache people in this area
is not known to the author. The location is shown in Figure 2.
The ridge extends eastward about one mile along the north side
of Gardner Creek valley before terminating; to the west it merges
into a plateau. An unpaved, gravel road established by the Forest
Service, FR-92, runs west along the base of the ridge, along the
northern boundary of Apache Spring Ranch. The core of the ridge is
composed of Paleozoic limestones bounded by Mesozoic siltstones
containing a few thin limestone beds. The limestone core of the
ridge curves to the southeast and essentially terminates at the
gravel road. A survey of the surface geology was begun in 1998 and
essentially ended in 1999, except for small bits surveyed in 2001
and 2004. The primary objective of the survey was an attempt to
determine the sourCe of water emerging at the spring. No definite
source was found, but it was thought the ground water originated on
the plateau area northwest.of the spring and moved southeastward in
the carbonates exposed on the ridge.
The valley south of the forest road is almost flat from the
road to Gardner Creek, a normally dry stream at the base of a steep
slope which rises to an almost flat plateau. The crest of the
plateau stands about 100 meters above the creek and is the remnant
of coalesced alluvial fans resulting from deposition in Pliocene
time. A thin remnant of the very coarse gravel deposited at an
earlier time remains at the crest of Apache Ridge. SUbsequent
Pleistocene to Recent erosion removed most of the gravel from the
ridge and cut into the siltstones at the ranch.
The erosion also exposed several bodies of volcanic rock, Which
are visible in a few places along Cave Creek and Gardner Creek, on
the south side of the ridge. A rhyolite dike is exposed on the
sawmill Canyon map, immediately west of Apache Ridge. Other volcanic
rock is exposed in a low ridge between Sawmill Canyon and the Onyx
Cave area to the south. An igneous batholith is exposed south of
Cave Creek and the Onyx Cave area.
A geological report, Professional Paper 748, was pUblished by
the United States Geological Survey in 1972. The author, Harald
Drewes discussed the structure and stratigraphy of three areas in
the Sonoita area, with the emphasis primarily on the structure of
the three areas, at a scale of 1000 feet per inch. Plate 3, which
was devoted to the Sawmill Canyon Fault zone, proved to be of great
value to this stUdy, although it covered only part of Apache Ridge.
To facilitate description of Apache Ridge, it has been divided
into four blocks, from block 1 on the eastern end to block 4 on the
western end, as shown on Figure,3. The boundaries of the blocks are
based on faUlts or other features, a fence and the Arizona Trail.

PAGE

2

,

BLOCK

I
I
I

4

I
I
I

I

'

/ BLOCK
'

"

! 3
\

\

_C-'~

BLOCK

2

DIVISION OF APACHE RIDGE GEOLOGIC AREA
(TO FACILITATE DESCRIPTION OF DETAILS)

FIGURE

3

stratigraphy of Apache Ridge
Colina Limestone
Prior to surveying Apache Ridge, it was necessary to find a
suitable place to start. For this purpose, a traverse north along
the Arizona Trail and an ascent was made to the top of the ridge,
running in a generally east-west direction. The initial survey was
along the trail from· the southern edge of the limestone to the
northern edge of the limestone, which was almost a width of 200
meters, and which was close to being a central axis.
using this central traverse, the survey was extended E-SE and
W-NW, eventually covering the area in question. Using fossils and
lithologic features, it was possible to determine the stratigraphy
of the various outcrops.
Shortly after beginning the survey, the eastern margin of the
outcrops, was found, which occurred in a small ravine beginning on
the north side of the road near an old ranch house and initially
extending northward. The edge of the outcrops curved around to a
westward direction and the slope diminished toward the crest. In
this area were a number of exposures of dark gray limestone, and in
some were large gastropod fossils, index fossils for the Colina
Limestone. Additional outcrops of colina Limestone were found
farther west along the northern margin of the limestone outcrops.
In one area, a white limestone was found in contact with the
qOlina outcrop. The rock has a fine rained white matrix containing
small particles of calcite. The rock is considered by the author to
be a component of the Colina Limestone.
Epitaph Formation
There are many exposures of light gray limestone which are
devoid of chert and almost completely devoid of fossils. It is
believed that these outcrops are those of the Epitaph Formation,
which is partially a dolomite in some places. Small brachiopods were
found in one exposure, but such occurrences are not common.
Near the north side of the large limestone ridge, on the north
side of a moderate ravine, is a Forest Service signpost which shows
the map of the long pipeline, being followed by the Arizona Trail at
this location. Here, on the west side of the Trail, are three bands
of interesting outcrops. The southernmost one is composed of a
yel,lowish-gray rock about 2 meters wide. The northernmost one, which
is essentially at the northernmost edge of the limestone rock, is a
reddish tinged rock about 10 meters wide. The wider band, when
observed carefully, is seen to consist of fragments tinted red in a
matrix of lighter colored rock. A third band, about 1 meter wide,
actually forms the boundary between the Paleozoic formations and the
Triassic siltstones. This third band consists of thin, light gray to
white beds, broken into fragments which are suspended in a matrix of
darker siltstone or sandstone.

PAGE

3

Similar or identical beds have been observed in the Lower
Sawmill Canyon area, the Hilton Ridge area, and the Bells area, all
of which represent a component of the Epitaph Formation. Both the
relative elevation ,in the formation and the lithology appear to be
essentially the same for all mapped areas. The basal beds appear to
be thick bedded, light gray dolomite. Higher up are beds of broken
and recemented rock. Above these are finely laminated, yellowishgray dolomite. Above these thin beds are thicker beds containing
clasts of various colors in a matrix of light colored limestone. At
the very top are the thin rip-up clasts in a darker matrix.
Scherrer Formation
Immediately west of the Arizona Trail are large areas of a tan
quartzite, which are obviously the Scherrer Formation. The
stratigraphic column of Drewes' Plate 3 provides no hint as to what
member it is, but the description given in Table 1 on page 5 of his
text indicates that it is probably the "Lower quartzite and basal
siltstone" member, which includes "quartzite, pale yellowish gray".
Toward the west, the northern and southern boundaries of the
Epitaph Formation maintain a relatively constant separation, while
the width of the Scherrer quartzite becomes progressively larger.
As a result, the total width of the Epitaph outcrops diminish. The
southern Epitaph exposure is .limited to a relatively narrow band
ranging between zero and 10 meters, while the northern Epitaph
exposure does not exceed 55 meters. Between the two Epitaph beds.
the Scherrer quartzite does not become less that about 135 meters.
Concha Limestone
The ellipsoidal chert nodules of the Colina near the crest of
the ridge change into a black spongiform chert lying above highly
fossiliferous limestone. In addition, two lenses of dark red chert
over one meter long Were found close to the fossiliferous area. Near
the base of the hill near the eastern margin of the limestone are
some limestone bodies which are associated with thick chert.
Finally, west of the Arizona Trail are some medium gray limestones
containing large Dictyoclustusfossils in a silicified condition. It
is thought that all of these features are associated with the Concha
Limestone, the uppermost Permian formation in this survey.
The presumed Concha outcrops west of the Arizona trail have
been traced to the intersection of two faults, about 70 meters west
of the Arizona Trail, where a small streamcourse crosses the concha
and Scherrer contact. It appears that the southern margin of the
Epitaph beds occur a short distance north of the streamcourse and
extends eastward. The contact of the Concha and Epitaph is somewhat
gradational, in that the chert inclusions progressively diminish to
the north. Thus, the position of the Epitaph-Concha contact is not
easily defined. It is thought that the Concha outcrops begin at the
fault offset where the small streamcourse crosses and extends eastsoutheast to the transverse fault between block 2 and block three.
The Concha outcrops cover most of the southern part of Block 3.

PAGE

4

The Concha chert is rather nondescript. It is generally brown
or gray in color. Except for the silicified fossils found near its
western boundary, the form of the chert is generally irregular, as
irregular nodules, blocks, stringers, fragmented masses, or thin
sheets. A few m~sses appear to show thin laminations.
The matrix limestone often occurs as separated masses, though
. larger and more contiguous masses occur. The smaller masses contain
as few as one or no chert masses; larger limestone blocks may
contain several more than a dozen chert masses as large as one's
fist. The interior of the masses are often radically different than
the chert surface, and in a number of cases have reverted to a clear
or colorless silica. Because of the non-soluble nature of the chert,
the masses generally project from the matrix limestone. In one case
however, the silica was confined to a thin layer about 2 em thick
covering a fin-like projection of the interior limestone. the
remainder of the fin projected about 1.5 em to an irregular,
apparently broken surface.
The color of the matrix limestone is generally somewhat darker
than the lower Epitaph beds, with which they are in contact in some
places. Higher Epitaph beds are distinctively colored, so they pose
no problem as to identification.
Some of the outcrops are a darker gray limestone laced with a
network of white fracture fills, probably calcite. This rock
contains little or no chert, in contrast to most of the Concha
outcrops. The boundary between the two rock types is somewhat
irregular, suggesting a set of fault boundaries. The fracture-filled
rock is possibly a component of the Epitaph Formation.
The contact of the cherty Concha with the primarily non-cherty
Epitaph has been traced across the northern part of block 3 to the
barbed-wire fence which is considered to be the western boundary of
block 2. East of the fence, however, the outcrops are somewhat more
difficul t to trace on the northern side of the Tertiary gravel
capping the hill. Most of the outcrops on the north side of the hill
in block 2 appear to be cherty and are thus probably Concha.
. The contact of the Epitaph and Concha, which occurs halfway up
the north hillside, is a problem, as if cannot be readily traced.
The contact has bean tentatively extended eastward to the tentative
extension of the transverse fault which is well exposed on the
southern side of the ridge. It is assumed that the surface outcrop
of the Epitaph is terminated at this fault trace and is not exposed
again until it reappears again at the surface in block 1.
Some apparently non-cherty limestone exposures have been
observed at various places on block 2, but there is no evidence to
suggest a contact with the Epitaph beds at these places. Non-cherty
beds considered to be Epitaph beds are exposed on the south side of
the ridge, below the coarse gravel capping the ridge top. It may be
that these beds extend beneath the gravel to a contact with Concha
beds, in isolated areas exposed at the coarse gravel margin.

PAGE

5

Gardner canyon Formation
The Gardner Canyon Formation h?s been subdivided into two
members, the upper member and the lower member. These two members
have been further subdivided into a number of components based on
their lithologies.Some of these lithologies are not found in the
Apache Ridge area and thus are not considered here. From Plate 3,
the lower member contains two thin limestone beds (one shown as
marker bed a) and b, a sedimentary breccia. The upper member also
contains tw.o marker beds, conglomerate beds, dacite flows, and
volcanic breccia. Of these, the latter three are not present in the
mapped area and are not considered for this study.
Gardner Canyon .Lower Member
The Gardner Canyon lower member consists of a tan siltstone
containing two thin limestone beds. Inasmuch as the main emphasis of
this study is the carbonates, little of the bounding upper and lower
members have been surveyed. The primary area in which the Lower
member has been surveyed is along the northern boundary of the
Epitaph beds, in the western part of that contact. This area
contains no limestone marker beds; thus none are shown.
Gardner Canyon Upper Member
~he Gardner Canyon upper member consists primarily of a red,
fine grained siltstone. The color of the upper member siltstone is
brick red, which aids in its identification. The red siltstone is
found in many places, including a reentrant of moderate size
surrounding the water tanks at the eastern end of the map. In
addition, outcrops of red siltstone are found up the small side
canyon at the eastern end of the map, where several faults occur.

The Gardner Canyon upper member is in contact with about 3/4 of
the primary limestone body shown on the geologic map of this study •.
One marker bed along the southern margin toward the eastern end of
the limestone body has been surveyed and is shown on the map. The
bed is of somewhat variable width, as shown on the map. The bed
consists of a dark gray to black limestone, somewhat similar in
color to the Colina Limestone.
The side canyon at the eastern end of the map eventually turns
toward the northwest. Prior to this turn are some exposures of what
appears to be light gray siltstone, which is probably a variant
color of the upper member. Upper member siltstones farther south
along the Arizona Trail exhibit a light green color.
The upper member also contains several other lithologies, which
are exterior to the central area and not shown. A litharenite occurs
along the Arizona Trail on the south side of the ridge crest and a
bed of conglomerate occurs south of the ridge crest, east of the
Arizona Trail and adjacent to the barbed wire fence. outcrops of
dacite are found farther away, primarily along lower Cave Canyon and
Gardner Canyon.

PAGE

6

Tertiary Terrace Gravel
The eastern end of the ridge slopes upward to a general level
plateau covered with a very coarse/ bouldery gravel of an unknown
depth. The gravel is a remnant of coalesced alluvial fans derived
from the erosion of the Santa Rita Mountains. The gravel eroded from
this and other sites has been redeposited at lower altitudes in the
form of Pleistocene terraces along Cave Creek and Sawmill Creek. The
depth of the redeposited coarse gravel is generally one meter or
greater. The tertiary gravel is denoted on profile views on Plate 3
as the "Pliocene Pediment". More recent .studies refer to this
surface as the Martinez surface.
In most places the exterior edges of the Paleozoic formations
are lying in contact with the lower member of the Gardner Canyon
Formation/ except if the upper member has been emplaced by a fault.
Exception to this general condition occurs where the Arizona Trail
reaches the ridge crest on the south side. Bere a sliver of what
appears to be the upper member penetrates the limestone and extends
some distance north. The sliver is poorly exposed and the exact
boundary is difficult to precisely locate. This inlier is also shown
on Plate 3 as a small triangular area.
The thin rip-up clast beds at the top of the Epitaph are,
without known exception/ in contact with the lower member of the
Gardner Canyon Formation, generally a light tan siltstone. If the
rip-up beds are the top of the Epitaph, it seems unlikely that there
is an unconformity at the base of the lower member. In addition, at
least one member of the Scherrer and at least part of the Concha
Formations have been laid down since the end of Epitaph time.
outcrops Of Uncertain Provenance
When surveying the northern side of the second structural
block, a large outcrop was encountered, which was not limestone. The
outcrop apparently consisted of two closely spaced masses or perhaps
one mass with two projecting above the forest floor. The total width
of the two masses was about two meters; the westernmost mass stood
about one meter above the nominal forest floor. A small sample was
removed and examined; it appeared to be a dark gray quartzite. A
detached sample. possibly derived from the larger mass, was found to
be a reddish-gray quartzite. One side exhibited slickensides/ with
a red overcoating on the grooved surfaces.
The two projections appear to be parts of a single mass. The
mass seems too large to have been transported from alluvial fans
derived from the Santa Rita Mountains to the west. The alluvial
remnants which cap parts of Apache Ridge are generally limited to
clasts no larger than a fraction of one meter in size. However/ the
possibility of stream transport cannot be ruled out.

PAGE

7

Structure Of Aoache Ridge
Faults:
Plate 3 of USGS Professional Paper 748 shows slightly
diverging bands of Paleozoic formations spreading toward the
southeast. The bands of Paleozoic rock are separated and flanked by
Mesozoic rocks, primarily sedimentary but also containing some
minor dacite flows, intrusives, and tuff (Drewes, 1972). The
southernmost of the bands, an outcrop about 215 meters (700 feet)
wide, contains primarily the Rainvalley Formation. North of this
band is a second band about 600 meters (2000 feet) wide. This band
spli ts into two components at about Cave Of The Bells, both of
which trend in an east-southeast direction. The third, or middle
band, extends from Cave Of The Bells along lower Sawmill Canyon and
terminates just prior to the stream crossing of Forest Road 92.
This short outcrop of Paleozoic limestone contains several small
caves and one sink.
The northernmost band of Paleozoic outcrops continues to the
eastern boundary of Plate 3, which is coincident with the eastern
boundary of the Mt. Wrightson 15 minute map, the base map for this
plate. Section B-B' of Plate 3 crosses the outcrops in a northeast
direction. Both section B-B' and plotted strike/dip measurements
show that the limestone beds in this northeastern band are dipping
steeply northeast.
Geologic mapping by this author showed that both the trend of
the outcrops and their attitude continued beyond the eastern border
of Plate 3, up to a short distance past Apache Spring, where the
limestone outcrops terminated. In general, strike/dip measurements
average around 45 degrees dip, with a minor number of dips lower
and higher than this average. These high dip angles confine the
outcrops of the Permian rocks to a relatively narrow band between
the bounding faults. Outside of the faults, the outcrops consist of
the Gardner Canyon Formation, primarily the red siltstone of the
upper member.
The bounding faults appear to be the primary structural
features in the Apache Spring area, exhibiting the greatest
displacement in a vertical direction. Transverse faults appear to
cut across the bounding faults, offsetting the Permian formations.
These secondary faults appear to exhibit only limited vertical
displacement, but the actual displacement, either vertical or
perpendicular to the bedding, is difficult to assess. One of these
transverse faults, near Apache spring, occurs between the first and
second structural blocks, offsetting the first block to the south
with respect to the second. The approximate surface trace of this
fault can be discerned from the pattern of outcrops occurring along
the fault, and from the apparent offset of the bounding faults on
either side of the transverse fault. The apparent offset on the
northern bounding fault is about 200 feet (60 meters); the offset
on the southern side is indeterminate. Small offsets on the eastern
side of Block 1 are well defined by the presence of the Gardner
Canyon red siltstone.
PAGE

8

The first structural block, the southernmost of several,
extends southward to Forest Road FR-92. The southern boundary of
this block is not known. as the south side of the road is grass
covered, but a number of small limestone outcrops occur in the
roadway.
The second structural block has been displaced about 140
meters north of Forest Road 92. The southern boundary of this block
is not well defined because of the lack of outcrops toward its
eastern end, but there are sufficient outcrops on the remainder of
the western end of this bounding fault margin. A small transverse
fault cuts this southern margin about 120 meters west from the
transverse fault forming the eastern boundary, but it apparently
does not extend across the entire block. This second block has a
length of about 300 meters.
Surfaces of the second structural block are mostly covered
with vegetation and talus on its lower slopes and the crest is
covered by very coarse gravel on its summit, so that only the
peripheral outcrop bands are visible. The northern side of Block 2
is covered primarily by the cherty outcrops of Concha Limestone,
with a short remnant of Epitaph near its western boundary.
The second structural block is bordered on its western end by
a third transverse fault, which runs along a barbed wire fence, The
trace of this fault is lost near the crest of the hill at a cattle
guard, and the remainder of the fault is tentative.
The third block is not strictly a structural block, although
two transverse faults extend north along part of its western
margin. The remainder of its margin is defined by the Arizona
Trail. The block, defined in this manner is about 150 meters wide.
A short segment of a transverse fault forms a 15 meter offset in
the limestone along the southern margin of this block. The faulted
limestone forms a small cliff which displays slickensides. the
fault probably continues in a northeast direction, but limited
exposures prevent tracing the fault farther north.
Two subparallel, apparently short faults, occur on the
northern margin of the third structural block. One fault merely
displaces the northern margin a short distance. The second fault,
a short distance east of the first fault, completely truncates the
upper, colored clastic beds of the Epitaph Beds and the thin, upper
rip-up-clast bed. These two beds continue westward from the fault
to the western end .of Apache Ridge. In addition, these upper
Epitaph beds are found on the Lower Sawmill Canyon map and the Cave
Of The Bells map.
The easternmost 60 meters of the colored Epitaph beds and the
thin rip-up-clast bed is separated from the adjacent limestone beds
by an inlier of the Gardner Canyon upper member. The inlier must be
fault bounded, but a lack of adequate outcrops in this area does
not permit the identification of such a situation.

PAGE

9

The fourth, westernmost structural area of this map extends
westward from. the Arizona Trail to slightly west of the Loop in the
Forest Road 4085. The width of the Paleozoic outcrop area does not
diminish appreciably, but the width of insoluble Scherrer Quartzite
increases in a series of steps until it constitutes about 2/3 of
the total band of Paleozoic rock. If the remaining 1/3, a narrow
band of light gray rock. probably lower Epitaph, continues to the
westward border of the map. The northern margin, consisting of the
yellowish gray dolomite bed, the thicker red tinted clastic beds,
and the thin rip-up-clast bed continue almost to the western margin
of the map.
The wider Epitaph bed along the northern margin of the
Paleozoic beds continues westward to what appears to be an inlier,
possibly of the upper member of the Gardner Canyon Formation or as
a remnant of the Tertiary terrace. The inlier is no more than 10
meters wide.
The small dry side canyon at the eastern end of the geologic
map constitutes the approximate eastern end of identifiable
outcrops. A short distance east of this canyon and north of Forest
Road FR-92 is a slab of conglomerate about 2 meters long. The rock
appears to be unconnected with any in-situ beds, but it may be a
remnant. It resembles Cretaceous conglomerates found nearby in Cave
Canyon.
The small structural block at the eastern end of the Apache
Spring map shows that almost the entire area is covered by outcrops
of limestone, except for a narrow band of the Gardner Canyon upper
member siltstone. Two water tanks, one of which supplies water to
Apache Spring Ranch, stand upon this outcrop band. The band extends
from the eastern boundary almost to the western boundary of the
first structural block. Thus, the positions of the insoluble
formations, extending almost completely through the limestone,
would provide only a narrow width of soluble limestone through
which water could flow southward to Apache spring.
other Additional features
The small streamcourse which begins at a low causeway near the
southern side of the limestone near the Arizona Trail, extends to
the northeast, where it intersects a second stream near the
limestone margin. The merged streamcourse turns east-northeast
toward the margin. At about the place where the streamcourse would
intersect the limestone (fault) margin, a small transverse fault
offsets the margin a short distance northward. The stream and the
limestone margin continue almost superimposed for almost 100 feet
(30 meters)
to a second transverse fault,which completely
terminates the northern limestone, the band containing the reddishtinted breccia and the rip-up-clasts.
The limestone bed is separated from a bed of what appears to
be Concha Limestone by an inlier bounded by fau~ts, The inlier
extends back into the carbonates for about 30 meters.
PAGE 10

Its composition is unknown within the inlier, but farther east The
Gardner Canyon upper member is exposed on a steep creek bank. In
the opposite direction, the reentrant appears to terminate at about
the confluence of the two streamcourses.
A second reentrant consists of a window of apparently noncalcareous rock surrounded by limestone, just to the east of, the
Arizona Trail and nearer to the southern ridge crest. This area is
about 70 feet (21 meters) wide by 230 feet (70 meters) long and is
shown on Plate 3 near its eastern margin; however, the difference
in scale reduces the depicted area on Plate 3 to an extremely small
a area. A very small adjacent area is also shown; the color coding
of that area possibly indicates an outcrop of Scherrer quartzite.
Anomalous Features.
A sequence of well exposed, identifiable beds occurs on the
first fault block, from about the crest of the ridge, down toward
the south. Beds at the ridge crest have been identified as Colina,
with a narrow band of Concha just to the south and down the hill.
This narrow band of Concha, only about 60 to 70 feet (20 meters)
wide, is truncated on the south by what appears to be Epitaph
outcrops, in a band even narrower than the Concha. ,On the south
side of'the Epitaph (?) band is an even narrower (15-20 feet) band
of what appears to be Colina. The next lower outcrop, wider but
less well exposed, appears to be Concha. This sequence of
limef;itones is terminated on the south by an outcrop of red
siltstone of the Gardner Canyon Formation upper member, upon which
the water tanks stand.
There is no reasonable explanation for this sequence of
limestones except to assume that they are fault slivers of the
several formations represented. Furthermore, the red siltstone, an
outcrop 80 to 100 feet (25 to 30 meters) wide, is bounded on the
south by additional limestone. The extensive imbrication of the
different formational slices, plus the termination of surface
exposures of the entire Paleozoic column toward the east, implies
a major transverse fault.
In the Tectonic Map Of Southeastern Arizona (1980), Drewes has
projected the Sawmill Canyon Fault Zone southeastward across
Sonoita Creek Valley to the Canelo Hills of eastern Santa Cruz
County. He has not shown a range bounding fault along the eastern
side of the Santa Rita Mountains. Presumably, the Paleozoic and
Mesozoic rocks of the Santa Ritas have been progressively eroded
downward toward Sonoita Creek or Cienega Creek and sUbsequently
buried by alluvium washed down from the mountains. In any event,
the limestone ridge north of Apache spring has been downdropped
east of the spring by a transverse fault, probably extending
between southeastward trending faults of the Sawmill Canyon Fault
Zone.

PAGE

II

Geology Of Lower Sawmill Canyon - Introduction
sawmill Creek is a small, normally dry streamcourse occurring
primarily on the northern border of Santa Cruz county, Arizona. The
upper reaches of the creek extend north into Pima county. The
entire streamcourse and most of its small drainage basin is shown
on the Mt. Wrightson 7.5 minute topographic quadrangle. The total
length of the creek is only about 4.5 miles (about 7 km). The creek
drains the eastern slope of the Santa Rita Mountains and continues
eastward to join Cave Creek, a somewhat larger stream.
From the confluence, Cave Creek flows east less than 0.5 mile
to the eastern margin of the Mt~ wrightson 7.5 minute topographic
quadrangle map. The continuation of the stream is shown on the
western edge of the Sonoita 7.5 minute topographic quadrangle,
where it turns abruptly southward and joins Gardner Creek in about
0.4 mile. On the Sonoita quadrangle, the creek is termed Sawmill
Creek, which appears to be incorrect. Sawmill Creek joins Cave
Creek at a tee Junction, where Sawmill Creek abruptly terminates,
but Cave Creek continues straight through the confluence. The total
distance from Sawmill Creek to the Gardner Canyon confluence is
about 0,9 mile, or slightly less than 1.5 km.
The geology of the northeastern part of the Mt. Wrightson 15
minute topographic quadrangle was surveyed by Harald Drewes and
published as Plate 3 of USGS Professional Paper 748. This plate
also covers the northeastern part of the Mt. Wrightson 7.5 minute
quadrangle, which was published later. Plate 3 shows both the
outcrops and structure of the mapped area as well as 5 sections, at
a scale of 1000 feet per inch: the emphasis is on the structural
relationships.
This author began surveying Lower sawmill canyon in the fall
of 1999 as part of a study of the hydrology of this area. Drewes'
Plate 3 was useful, but not sUfficiently detailed for this study.
Two years were spent in geologic mapping of the Sawmill Canyon
area, extending from the Sawmill Creek-Cave Creek confluence on the
east to the rhyolite dike (shown by Drewes) on the west. Additional
surveying was conducted in 2003 on both the northwest and southwest
parts of the map area to more fUlly depict the geology.
The surveyed area is roughly rectangular in the western twothirds of its area and narrowing to a minimum on the eastern onethird. Sawmill Creek drains the central part of the mapped area
where paleozoic carbonates outcrop; Mesozoic siltstones and marker
beds (thin limestone bands) occupy areas peripheral to the central
beds. In the western part of the mapped area, Sawmill Creek swings
more towards the south, and the carbonate beds lie north of the
creek. On the extreme northwestern side of the map, a complex area
inclUding siltstones, limestone marker beds, separate segments of
a rhyolite dike, and several faults occur. The northern part of the
mapped area is bounded by a Forest Service road, FR-4085, which
heads northeast toward the margin of the Apache Ridge map.

PAGE 12

stratigraphy Of Lower Sawmill Canyon
Lower Sawmill Canyon is formed in part in carbonates of
Permian age, while areas farther upstream and away from Sawmill
Creek are formed in Triassic siltstone. The identity of the lowest
outcrop is uncertain as it occurs over a very small area in the
mouth of a small cave from which a dirt fill was excavated. This
small outcrop, near the dug pit, may be from the Earp Formation, of
late Pennsylvanian and early Permian age.
Colina Limestone
Much of the lower part of Sawmill Canyon is formed in the
Colina Limestone of Lower Permian age. The rock is generally a dark
gray to black in color, containing black chert. Near the top of the
formation, the beds contain both a thin, tabular black chert and a
nodular black chert. At a small cave not far from the lower end of
the canyon, the limestone contains thin fissile beds, but these are
a rarity. certain horizons contain large gastropod fossils of the
genus omphalotrochus. In an adjacent study area, termed Apache
Ridge, white limestone in contact with The Colina is thought to be
a unit of the Colina, but no white limestone has been observed in
this survey area.
Epitaph Formation
A short distance downstream from the small cave mentioned
above are massive blocks of white dolomite which have fallen off
the cliff above the cave. These blocks constitute the basal beds of
the Epitaph. The formation is generally termed the Epitaph Dolomite
but in this area the formation is a mixed dolomite and limestone.
Farther downstream, outcrops reveal both thinner dolomite and
limestone interbedded. Higher on the hill, the formation contains
broken clasts interbedded in a matrix. Above this horizon, at the
level of a small foot trail, a thin series of yellowish gray beds
consist of dolomite. Above the yellowish gray beds is a thicker
sequence of broken clasts contained in a limestone ~atrix. Along
the foot trail are areas where fracturing occurred but the broken
blocks were not displaced; a bluish-gray staining occurs along the
fractures, and some reddish coloration is also present.
At the very top of the formation is a thin (about 1 meter) bed
containing broken thin fragments of white beds distributed in a
darker silty matrix. This thin upper bed is essentially identical
to beds exposed near the Arizona Trail along the northern edge of
the limestone outcrops. The rip-up clasts indicate a shallow water
condition, where beds were disturbed by storm activity.
The distinctive sequence of beds, beginning with massive
dolomite at the base and extending up through a yellowish gray bed
of dolomite to a generally colored, intraclastic sequence and a
final thin rip-up layer is of much help in identifying the Epitaph
beds.
PAGE 13

The distinctive beds were useful in identifying the Epitaph
beds on the north side of Sawmill Creek, at the extreme western end
of limestone adjacent to Sawmill Creek. Here an area of rock about
10 meters wide by 45 meters long was positively identified as
Epitaph, On the eastern end of the mapped area, the bed of Sawmill
creek is covered with large rocks and boulders with the distinctive
red tint diagnostic of the upper Epitaph beds.
No further Permian formations were recognized in the Lower
Sawmill Canyon, including the Scherrer, Concha, and Rainvalley
beds, which are encountered not far away in adjacent areas.
Gardner Canyon Formation (Triassic)
The Gardner Canyon Formation has been subdivided into two
members by Drewes (PP748, page 5). The members have in turn been
subdivided into components, including conglomerate, dacite flows,
and volcanic breccia. In this report, those components will not be
considered, as they do not occur in the study area. The remaining
components which will be considered are the siltstones and the thin
limestone marker beds contained in the siltstones.
Gardner Canyon Lower Member
The Gardner Canyon lower member in the study area consists
almost entirely of a tan siltstone containing at least one thin
limestone bed and some fragmentary limestone which has not been
mapped. In addition, the lower member also contains quartz breccia
(noted as sedimentary breccia) and thin white quartzite (not shown
by Drewes but shown on my map). The lower member siltstone is in
contact with most of the Permian limestones, except where it has
been removed by faulting.
The thin limestone beds, (depicted as thin lines on Drewes'
Plate 3), have been drawn to scale on the study map. These beds aid
in determining the position of faults, which otherwise would be
difficult to define. The thin beds are especially important in the
very complex northwestern area of the map where complex faUlting
apparently has occurred.
The sedimentary breccia of the lower member (denoted by a "b"
on the identifier, is found in small outcrops in several places.
One occurs near the foot trail on the west side of a fault. A
second occurrence occurs on the opposite side of the canyon, near
what is believed to be the same fault. Two of the outcrops stand
out as small pinnacles which are visible from the north side of the
canyon; a third, larger outcrop is low and not readily visible. A
fourth outcrop, not shown on Drewes' Plate 3, occurs a short
distance north of the large elliptical deposit of Tertiary gravel,
near the south margin of the large Epitaph outcrop.
In all of these cases, the rock identified by Drewes as a
sedimentary breccia is a tan color, resembling tan chert. The large
masses indicate that it is generally resistant to erosion.
PAGE 14

The quartzite identified by this author in the south part is
nominally white in color and occurs at scattered locations where
weathering has not reduced it· to rubble. In some cases, the
presence of the quartzite aids in th~ identification of other
features.
Gardner Canyon Upper Member
The Gardner Canyon upper member is generally brick-red in
color but, in contrast with the lower member, can assume almost any
color possible. One example of this color variety occurs at the
confluence of Cave Canyon and Sawmill Canyon. There, thick beds of
brick red siltstone contain a bed of light gray siltstone.
~ smaller side canyon joins sawmill Canyon near the western
end of a long loop in sawmill Creek. The canyon extends north,
cutting through the limestone body to a tee junction. Two smaller
dry streamcourses come to the tee from opposite sides. These two
streams occur approximately at the contact of the upper and lower
members. The upper member in this area is multicolored. In addition
to the characteristic red beds, it exhibits thinner beds ranging
from light gray to black, with beds of various intermediate colors.

A few thin beds exhibit a lithology different from the usual
siltstone but constitute only a minor component of the upper
member. Northwest of the tee junction, near the road, is a fairly
thin bed of quartzite. Whereas the siltstone can be scratched with
a sharp object, the quartzite is generally too hard to scratch. An
orange quartzite occurs farther west, where FR-4085 turns east.
Near the eastern end of the same valley and north of the road, the
siltstone changes to a red sandstone with masses of white sandstone
included inside the' red sandstone. The total extent of these
facies, on the margin of the surveyed area, was not determined.
On the extreme northwest part of the surveyed area, both upper
and lower members and the thin limestone marker bed of the lower
member are all mixed together in a confusing way. It appears that
considerable deformation occurred in this area, so that a sequence
of events will be difficult to unravel.
Tertiary Dike
A dike occurs near the western end of the area. The orlgln of
the dike apparently is a batholith depicted on Drewes' Plate 3, on
the south side of Cave Creek. Material from the batholith has
apparently entered a fault fracture and moved northeast across
sawmill Creek, and ultimately extending to the Epitaph margin.
Drewes shows the dike as a uniformly narrow body along its entire
length, but the narrow width is not uniformly typical. The dike,
mapped by this author, was of variable width and direction. In a
number of places, the dike was intermittent and a number of gaps
were apparent. In a few places, isolated outcrops of dike material
appear to have been displaced by faults, and a few small pieces of
dike material appear some distance away.

PAGE

15

One larger area, off to one side, appeared as if it might have
been a small sill of the dike material exposed by erosion, as its
width was appreciably larger.
In all cases, the dike material exposed from near Sawmill
Creek north to its terminus consisted of a light gray groundmass
wi th small phenocrysts in some cases. The material appears to
weather somewhat more rapidly than limestone and is reduced to
fragments by weathering. The resistance to weathering is somewhat
variable, with some outcrops reduced to ground level and some
protruding from the ground. South, near onyx Hill, some parts of
the dike project more than one meter above ground level.
The age of the dike material is given by Drewes on plate 3 is
26 Ma from the potassium-argon method and 40 Ma by the lead-alpha
method. Further information about the age of the dike may be
available at this time, some 32 years after pUblication of PP748.
Tertiary Gravel
A large diameter but probably not very thick deposit of coarse
Tertiary gravel washed down from the Santa Rita Mountains covers
the crest of the hill in the western part of the map. Much of the
coarse gravel covering the underlying rock was washed away by
Pleistocene erosion; Remnants of the gravel now occur in terraces
bordering Sawmill Creek and Cave Creek. Profiles of the gravel on
Drewes' Plate 3 are termed the Pliocene Pediment, of which most has
been eroded in this area, leaving a hypothetical surface.
Pleistocene Terraces
Lower terraces, formed both by erosion and deposition of the
reworked Pliocene Pediment gravel, are found along Sawmill Creek,
Cave Creek, and Gardner Creek, to which the former two streams are
tributaries. A single terrace, about 10 meters above the present
stream, is the single prominent Pleistocene terrace on this study
area. The four terrace segments along sawmill Creek are younger,
probably of Holocene age.
Mineral Deposits
There are not many mineral deposits occurring in the Lower
Sawmill Canyon area. A few seams of what appear to limonite occur
below the foot trail. A deposit of massive crystalline calcite
ocCurs north near the crest of the ridge above the western end of
the foot trail, possibly resulting from the discharge of calcitebearing water. Some large euhedral crystals of quartz have been
found on the south side of Sawmill Creek, after apparently having
formed in cavities in the lower member siltstone and later exposed
by erosion. The quartz may be the result of hydrothermal solutions
associated with the igneous dike.

PAGE

16

The structure Of Lower Sawmill Canyon
The geologic structure of Lower Sawmill Canyon has been
displayed in both the geologic map of Plate 3 and selected profiles
applicable to cross sections taken ort that map. Those structural
features generally agree with the structural elements defined by
this authors' surveys, but not completely,
One of the primary structural elements shown on Plate 3 is an
anticlinal fold roughly coincident with the axis of Sawmill Creek
in its lower reaches. Adqitional features are three faults, more or
less parallel to each other and oriented in a north-northeast
direction. The first two of Drewes' faults correspond to faults
shown of this authors' map, but a third could not be verified. That
fault is approximately coincident with the side canyon which joins
sawmill Creek near the center of the author's map.
On the other hand, the two faults which correspond to the
faults shown on Plate 3 and additional other faults not shown are
depicted in much greater detail, in part because of this author's
scale factor: 100 feet per inch. Several methods have been used to
delineate the faults. The primary method depends on abrupt offsets
of formational contacts or offsets of marker beds. In two cases,
the actual fault contact is visible. In those cases, denoted with
the letter "f", beds on opposite sides of the fault slope toward
one another and meet at a roughly vertical fault plane.
An interesting structural situation occurred on the north side
of the hill north of the foot trail, near the crest of the hill
where two closely spaced marker beds and two similar beds farther
down the hill occurred. Dip measurements revealed that the
uppermost two beds were dipping at angles slightly less than the
slope of the hill, while the lower two were dipping at angles
greater than the dip of the beds. It was evident that these lower
beds were the same ones observed near the crest of the hill. The
beds emerging at the crest of the hill had been eroded away down to
the place where the uneroded beds dived beneath the hillside.
Fault planes appear to be essentially buried on the west side
of the large Tertiary gravel area. There, the gravel is apparently
thin, allowing occasional exposures of the red siltstone of the
Gardner Canyon upper member. The structure in this area is rather
complex, and the postulated fault is tentative. A second fault is
postulated to intersect the first at a sharp vertex. One of these
faults could continue for a distance past the intersection, but
there is no definite evidence that such a condition exists,
West of the Tertiary gravel is an area where both the Gardner
Canyon lower member marker bed and the Tertiary dike are present.
It appears that the only way in which these two features could be
displaced in a logical manner is for initial displacement of the
marker bed along a fault (shown), injection of the dike, and then
subsequent reversal of the fault movement, displacing both the dike
and the marker bed to cause the final relationship of the two.
PAGE

17

As noted previously, it appears that the dike formed a sill
along one side of the fault carrying the dike material. The sill
apparently formed a cavity 50 meters long at the side of the dike,
but some of this length may be downhill float of weathered dike
material, rather than actual outcrops.
The dike continues, with some gaps, northward through Epitaph
rock and appears to terminate at the contact of the Epitaph and the
Gardner Canyon lower member. No trace of dike material was found
north of the contact. It is possible that the fault injected by the
dike terminated at the contact, or that the pressure in the dike
was relieved by a breakout at the sill.
The dike has been traced, with some interruptions, across
sawmill Creek. The dike is not visible on the terrace north of the
creek, presumably because the terrace was covered by alluvium at
the end of a depositional phase. The trace of the dike can be found
on the north creek bank and south of the creek in several places.
A continuation of the dike was surveyed for the map of onyx Hill.
Slickensides, though not prevalent, do occur in a few places.
Several observed exposures occur at the entrance of the small cave
which opens at the base of a cliff, about 3 meters above the creek
bed. Low angle slickensides, perhaps indicative of thrust faulting,
occur on the exposed upper side of the sedimentary breccia which is
exposed a short distance north of the Tertiary Gravel deposit. An
obvious slickenside is exposed on The red upper member siltstone on
the south side of Sawmill Creek, near the place where the Tertiary
dike crosses sawmill Creek.
The anticlinal fold noted by Drewes on Plate 3 of PP748 has an
axis which is generally south of Sawmill Creek. At the foot trail
on the north side of the creek, the dip is about 20 degrees to the
north; the dip increases toward the north and reaches at least 45
degrees in the small valley north of the creek.
Small faults have been observed on the anticlinal fold, with
dips much steeper than the nominal fold dip and strikes not in
concordance with the fold dip directions. A number of faults have
been noted in the small caves found in this area, all of which
formed prior to solutional development of the caves,
A feature which is the possible result of both lithologic and
structural origin occurs along the foot trail not far from its
origin at the eastern margin of the map, The feature is a sink of
apparent natural origin developed in the insoluble lower member of
Gardner Canyon siltstone. The sink looks as if it were cut down
into the hill, starting at ground level and deepening into the
hillside. There is no evidence of human intervention and the sink
appears to be natural. The sink appears to have a drain at its low
side which leads down into one of the thin marker beds. The water
entering this apparent drain may make its way to a small resurgence
on Cave Creek, a short distance downstream from the confluence of
Cave Creek and SaWmill Creek.
PAGE 18

A similar feature, though not natural, is an apparent
excavation on the south side of Sawmill Creek, a short distance
upstream from the confluence of the small side canyon with Sawmill
Creek. The excavation occurs at the margin where the low terrace
joins the sloping hillside. The hole is about 2 meters deep.
Inliers and Windows
An inlier is a relatively long and narrow outcrop extending
into another different formation. A window is an outcrop of one
formation completely surrounded by another formation. On the south
side of Sawmill Canyon, the contact between the upper and lower
members of the Gardner Canyon Formation extends generally eastward
from a beginning near Sawmill Creek, following a somewhat irregular
path, with the upper member lying south of the lower member. At a
distance of about 200 meters from the western origin , the contact
makes an abrupt turn to the left, at about a right angle to its
original orientation, and extends northwest for about 18 meters.
The contact then turns 90 degrees to the right and extends toward
the northeast for about 11 meters. At this third point, it turns
right again and heads in a southeast direction. At a point near the
first turn, it turns left and heads northeast, in a direction about
the same as its original orientation.
The result of these turns and displacements is that a narrow
extension of the upper member has been intruded into the lower
member. It appears that the displacement has occurred along a set
of closely spaced subparallel faults, with a total displacement of
about 18 meters.
A somewhat similar situation has occurred farther west at
about the western end of the same contact between the upper and
lower members. In this case, however, a narrow outcrop of the lower
member about 30 meters long has been completely surrounded by both
the Colina and Epitaph formations in fault contact.
An apparent second presence of a window occurs toward the
northwest, on the north side of the large Tertiary Gravel deposit.
Here, a somewhat irregular body of Colina Limestone. about 30
meters across, is surrounded on two of its three sides by the
Epitaph Formation. The third side is obscured by hillside alluvium,
but the alluvium probably covers the Epitaph Formation. This rather
small body of rock exhibits several discordant strike and dip
readings, implying some internal deformation.
The Northwest Structural Area
The Northwest Structural Area of the Lower sawmill Canyon is
rather complex. the area is covered by sizable areas of both the
upper and lower members of the Gardner Canyon Formation. There,
identification is in places hampered by vegetation and overlying
rubble. In addition to the Tertiary dike, the area exhibits many
narrow beds of white limestone, believed to be marker beds of the
lower member.
PAGE

19

On the adjacent Apache Ridge geologic area, the marker beds in
the nominally red siltstone of the upper member tend to be dark
gray to black in color, with tinges of dark red in places. None of
the marker beds in the Lower Sawmill Creek area exhibit this dark
coloration. The dike material is also light colored but can be
readily distinguished from the Gardner canyon limestones.
The white limestone marker beds exhibit a confusing
relationship with one another, in that they appear to join in
triple junctions. Simple erosional dissection of a siltstone bed
containing a thin limestone would most likely produce a narrow,
linear limestone outcrop. Triple junctions could seemingly result.
only if thin imbricated beds eroded to produce superimposed marker
beds at the surface.
It is possible that the apparent triple junctions are really
only a mistaken evaluation of the true situation. The limestone
marker beds are generally fragmentary, either as the result of
actual complex faulting or the result of inadequate exposure. The
separated limestone blocks have been shown as a thin contiguous bed
for convenience, but a different interpretation might be possible.
The structure southwest of the confluence of the small side
canyon with Sawmill Creek is somewhat enigmatic because it is
primarily hidden by alluvium on both sides of Sawmill Creek.
Drewes' Plate 3 shows a fault trending west-northwest beneath a
terrace, which probably is the terrace north of Sawmill Creek, but
Details on Plate 3 are insufficient to determine the placement of
the fault.
Mineralization
No mineralization has been noted in the Lower Sawmill Canyon
area, except for the few masses of crystalline calcite and a few
small masses of limonite below the foot trail. The sink-like hole
on the south side of Sawmill Creek appears to be man-made and
probably was a test pit. No trace of any mineralization was
observed at the test pit.
Some fairly large (up to 10 cm long) euhedral quartz crystals
have been found on the Gardner Canyon lower member on the south
side of Sawmill Creek, a short distance south of the limestone. The
crystals appear to have formed in cavities in the siltstone, which
were later breached by weathering. The largest crystal has several
small inclusions of siltstone where the crystal touched the walls
of the cavity in which it formed. The crystals are white, rather
than transparent. This defect is probably the result of surface
weathering during the time when the crystal was exposed.

PAGE 20

Geology Of Resurgance Area On Cave Creek
Introduction
The area studied in this report occurs near the confluence of
cave Creek and Sawmill creek. within the limits of the study, only
two formations are exposed. The lower of the two is the Epitaph
Formation, commonly termed the Epitaph Dolomite, of lower Permian
age. The underlying Colina Limestone, also of Permian age, is not
exposed here, although it outcrops nearby. The second formation is
the Gardner Canyon Formation of Triassic age. It is subdivided into
several members, two of which are considered here. The lower of
the two members, a tan siltstone, contains a thin, nominally white
limestone occurring near the top of the member. The upper member is
nominally a red siltstone. It also contains thin limestone marker
beds, but these beds are not exposed within this study area.
The resurgence area occurs on Cave Creek, only about 100
meters east of the confluence of Sawmill Creek· and Cave Creek.
There are actually two resurgences present, at least in times of
wetter weather. The primary resurgence is essentially a wide seep
at the base of a low ledge; the secondary resurgence is also a seep
about 27 meters upstream. The primary resurgence comes out on top
of a low platform of mud only about 15 em above the normal stream
elevation, and the secondary resurgence is even lower. Both of the
resurgences are obscured when there is even a slight rise of water
in the creek.
On Plate 3 of USGS Professional Paper 748, Drewes shows a
fault extending eastward along Cave Creek at this location, with
the upper member of the Gardner Canyon Formation, a generally red
siltstone, exposed on the south side of the creek. The Gardner
Canyon lower member is shown on the north side of the fault.
The lower member is apparently less resistant to erosion, and
that outcrop is somewhat obscured on the north side of the creek,
which is forested and partially covered with talus. Narrow outcrops
of white limestone, a marker bed in the lower member, are exposed on
the northern side of the streambed and partway up the side of the
sloping canyon wall farther downstream, near the Forest Service
gate. Scattered outcrops of the white limestone occur near the
resurgences.
A short distance downstream from the Forest Service gate, the
incidence of the forested slope on the north side of the canyon
abruptly ends, apparently as the result of a fault, and the red
outcrops of the upper member extend completely across the canyon.
The lower tan siltstone and the marker bed do not occur here; in
fact, those outcrops occur about 75 meters north of their previous
position. At this location, the exposed limestone bed occurs in the
top of the terrace bordering Cave Creek, at least 10 meters above
the bed of the creek. It is obvious that the groundwater has
emerged at a place as low as possible, as far downstream as soluble
beds will allow a resurgence.
PAGE

21

Groundwater in the Colina and Epitaph carbonates occurs in the
downdipping beds, which lie beneath the siltstones exposed at the
resurgences. Apparently these thicker beds are somehow in contact
with the thin limestone bed of the lower member. so that the water
can rise through the limestone to the resurgences. The large sink
opening on the eastern side of Sawmill Canyon is only about 140
meters from the upstream resurgence (Note traverse "A" on map). A
probable profile view is shown in Figure 4,. in which the thin
marker bed is thrust into contact with the Epitaph beds within the
distorted beds along the fault.
Lower Sawmill Creek, developed in the Colina and Epitaph
Formations, was found to be a sinking stream. Observations made
during December 2000 showed that water in Sawmill Creek sank into
the stream bed at places progressively upstream, despite continuous
though diminishing flow upstream. Finally, on December 28, the
lower end of Sawmill Creek (in limestone) was completely dry almost
all the way upstream (Figure 51) to the siltstone, where the creek
was still flowing. The pirated water must obviously resurge not far
away from the end of soluble rock.
Another probable source of groundwater is Cave Of The Bells,
which occurs about 1460 meters upstream and contains pools of water
about 80 meters below the entrance. An evaluation of the various
altitudes nearby indicates that the resurgence for the cave had to
be at least this far away to drain the cave.
At one time it was thought that Onyx Cave might be the source
of the water rising at the Cave Creek resurgences. However, careful
examination of profile B-B' on Plate 3 of PP748 shows that there is
no way that this source is connected with the resurgences. Another
possibility is the lateral movement of groundwater through Colina
Limestone from the north. However, on the basis of the Apache Ridge
survey, it is likely that such groundwater would drain to Apache
Spring.
It is known from surface observations that the groundwater
rising at the Cave Creek resurgences is almost certainly rising
through the thin Triassic limestone of the Gardner Canyon lower
member. This bed is not well exposed but one measurement taken
about 50 meters downstream from the major rise shows a dip of 36
degrees to the south~ Several hypothetical profiles have been
constructed, one of which is shown a connection between the Epitaph
limestone beds and the thin lower member limestone.
The contact between the upper and lower members is also not
well known, despite the fact that these beds are moderately well
exposed on the creek bed. The problem is that there are several
beds near the somewhat obscure fault which seem to be transitiDnal
and which may be components of the lower member. Even the thin bed
of limestone through which the water rises is only sporadically
visible between the confluence and the downstream cross fault.

PAGE

22

-sink In

811ts on8\
t

'NW

SE'

Cave
canyon
are
probably
distorted

beds

~esurgenc.el

RGL

~

RGU

, .,{c::::J
"c=:J

..

RGU

'.',.

.

~

.

. .'1'

.

Gardner Canyon FIll., upper member - siltstone

Triassi~~

-RGLS "Gardner Canyon FIll., lower member? - limestone

.
CJ
'pe~i~n'{D,
' .' CJ

RGL

Gardner canyon FIll., lower member - siltstone

PE,

.Epitaph Dolomite

Pc

Colina Limestone
i

Pc

RESURGENCE THROUGH THIN liMESTONE BED

FIGURE

4

I

SIDE
CANYON

"

\

HIDDEN
CAVE
SAWMILL
CREEK

.

CREEK
SINKS
DEC 22

\

CREEK
SINKS
DEC 15

-CREEK
SINKS
DEC 28

/\

CREEK/
SINKS
DEC 7

CONTINUOUS
FLOW

-r-

STREAM
BED IN
SILTSTONE

o

200

CREEK
FLOWING
~DEC 4

·DOWNSTREAM
END OF
LIMESTONE

400 FEET

CAVE CREEK
LOWER SAWMILL CREEK DURING DECEMBER, 2000

SINKING

STREAM

IN

LOWER

SAWMILL

CANYON

FIGURE

5

Geology Of Hilton Ridge
Introduction
Hilton Ridge occurs in the northeastern part of Santa Cruz
county, Arizona, about 30 miles southeast of Tucson, as shown on
the location map. Figure 1. The local area is shown on the Mt.
Wrightson 7.5 minute topographic quadrangle map. Hilton Ridge is
the informal name for a low ridge lying between Sawmill Creek on
the south and Forest Road 4086 on the north. The southeastern
boundary is on a saddle occupied by Forest Road 4085, and the
northwestern boundary is arbitrary, being essentially defined by
the northwest end of Permian outcrops.
The survey area includes both the central ridge area and some
surrounding areas. The central ridge is bounded on the south and
southwest by Sawmill Creek and a short, generally dry tributary
lying along the southwest flank of the ridge. The eastern end of
the ridge is cut by a small, normally dry streamcourse coming past
the Cave Of The Bells area to the northwest.
The primary geologic feature of Hilton Ridge is a somewhat
ellipsoidal window of Epitaph Dolomite, which is shown on Plate 3
of USGS Professional Paper 748 (Drewes, 1972). The outcrop shown is
about 300 meters long and 60 meters wide, slightly curved. It is
bounded by siltstones of the Gardner Canyon Formation on all sides.
The map provided in this report is slightly more angular and much
more detailed, showing features which were too small to be shown on
Drewes' Plate 3.
A primary feature of interest within the bounds of the study
area is a small cave with the name of Hiltons Hole. The cave has a
single vertical entrance about 1.5 m long at the bottom of a small
sink not much larger than the entrance. A climbable descent of 6
meters allows access to a fairly level flooor. The cave consists
of two rooms with a total combined length of about 25 meters.
After surveying the periphery of the Epitaph outcrop area, it
was found that the area had an essentially triangular shape with
the largest width, about 75 meters, on the southeast. The outcrop
area, after a central section of relatively constant width, tapered
to almost a point on the northwest. A narrower extension on the
southeast end brought the total length of the Epitaph outcrop area
to about 335 meters.
The area is generally well exposed; bedrock outcrops cover
large areas, and areas covered by vegetation are small. Alluvium is
minimal and is primarily limited to a small area on the southeast.
The ridge is lightly forested, primarily on the carbonate outcrops.
The low ridge, of which Hilton Ridge is the southeastern terminus,
continues west-northwest, past the Cave Of The Bells area, which is
part of the same low ridge, but there termed Bells Ridge.

PAGE

23

stratigraphy of Hilton Ridge
Colina Limestone
The primary, central exposure on Hilton Ridge is the Epitaph
Dolomite, as shown by Drewes on Plate 3 of USGS Professional Paper
748. In this general area, the Epitaph is underlain by the Colina
Limestone, which is not exposed on Hilton Ridge, but which is
exposed in surface outcrops not far away. The measured thickness of
the Epitaph in the Lower Sawmill Canyon area is about 100 feet; the
depth of the upper surface of the Colina Limestone is probably no
greater than that measured thickness and may be less.
Epitaph Formation
The Epitaph Formation is generally termed the Epitaph Dolomite
because of the presence of dolomite beds in the formation. In this
specific area, the formation contains much limestone, so the author
has used the term Formation for this specific area. In Professional
Paper 748, page 5, Drewes has subdivided the formation into two
members. The upper member i$ listed as Limestone, moderately thick
bedded, medium gray, slightly cherty, with a nominal thickness of
380 feet (115 meters).
The lower member of the Epitaph Dolomite, as listed by Drewes,
contains light-brownish-gray marlstone, dolomitic siltstone, gypsum
and argillite, and dark to medium gray dolomite, with a nominal
thickness of 620 feet (189 meters). None of these lithologies have
been recognized in any of the geologic areas mapped by this author.
On Plate 3, Drewes also includes a non-specific block designated
only by the symbol Pe.
The Epitaph outcrops surveyed by this author have contained
some dolomite but also contain what appears to be a majority of
limestone, some of which is fractured or brecciated. Some of the
fracture fragments are colored, especially in the Apache Ridge and
the Lower Sawmill Canyon mapped area. Similar fracture fragments
occur in the Hilton Ridge and the Cave Of The Bells mapped areas
but without the more diagnostic coloration. The Epitaph observed on
Hilton Ridge has been termed the Epitaph Formation for this reason.
Scherrer Formation
On Plate 3, Drewes shows that the Scherrer Formation has been
subdivided into four members, an upper Quartzite member, a middle
dolomite member, a lower quartzite member, and a basal siltstone
member. A very small outcrop of what is believed to be the lower
quartzite member has been found on Hilton Ridge, close to the
entrance of the small cave. On page 5 of PP748, Drewes briefly
describes the lower member as "quartzite, pale yellowish gray", a
description Which matches that of the small outcrop near the cave.
No other outcrops of the Scherrer Formation are known to occur
either on Hilton Ridge or on the adjacent Cave Of The Bells area.
However, careful reconnaissance was not conducted far beyond the·
map bounds; other formations may be present.

PAGE

24

Concha Limestone, Rainvalley Formation
Although the Concha Limestone and the Rainvalley Formation
occur in the general vicinity, no trace of them has been found on
Hilton Ridge.
Gardner Canyon Formation Lower Member
The elongated Epitaph outcrop area is surrounded by the upper
and lower members of the Gardner Canyon Formation of Triassic age,
both primarily siltstones. The lower member is adjacent to the
Epi taph outcrop on all sides, except for the extreme southeast,
where the lower member seems to be covered by an alluvial terrace.
within much of' the southwestern exposure of the lower member is a
thin bed of light colored limestone, defined by Drewes as a marker
bed. The bed is not present in the southeastern extremity of the
lower member, an area where the lower member is rather narrow.
A short dist;ance north of the Epitaph margin is a small
outcrop of limestone, either a displaced fragment of the Epitaph or
an inclusion of Triassic limestone. In a few places the white mass
is in contact with a red inclusion. The identity is unknown.
Gardner Canyon Formation Upper Member
The lower member is bounded on the south and southwest by the
upper member of the Gardner Canyon upper member, primarily a deep
red siltstone, grading into a red sandstone in a few places. An
area on the southeast is covered with alluvium, but it is almost
certainly underlain by the upper member. A short distance to the
south is sawmill Creek, which flows in a shallow trench cut in the
upper member.
Pleistocene Terraces
Cut banks along Sawmill Creek and a small dry tributary show
a sectional view of a probably Pleistocene terrace deposit composed
of coarse gravel and overlain by a finer silt. The coarse gravel is
derived from the older Tertiary Martinez surface, the coalesced
alluvial fans derived from weathering of the Santa Rita Mountains.
The surface of this gravel has been termed the Pliocene Pediment by
Drewes and noted on Plate 3 of Professional Paper 748.
other small areas are covered by recent alluvium derived from
the higher parts of the adjacent hill. Other alluvial areas are
found along the small streamcourse flowing southward along the east
margin of the study area, adjacent to and crossing FR-4085.
During surveying, three small areas of massive crystalline
calcite were found and noted on the map. The calcite is thought to
have been deposited by hot water seeps whiqh once surfaced here.
The calcite is massive and shows no crystal faces, although such
surfaces may have exposed prior to weathering. The calcite occurs
near fault traces.
PAGE

25

structure of Hilton Ridge
Drewes shows the southern margin of the Epitaph outcrop to be
bounded by faults, which is partly correct. The southern margin of
the Epitaph outcrop is bounded by the Gardner canyon lower member
along its entire length, although the outcrop width is quite narrow
for a short distance near the southeast end.
The northern margin of the Epitaph body is, however, bounded
by a fault which extends westward for about 100 meters and then
turns toward the southwest and finally almost south, extending
through the Epitaph body and continuing beyond. In addition to
visible outcrops along the fault, it displaces a limestone marker
bed by about 22 meters. Finally, the fault and a nearby parallel
fault segment enclose an inlier of the upper member so that it
projects 15 meters into the lower member. Interestingly, the three
calcite bodies occur near the main fault and a short fault segment
9 meters northwest of the main fault.
The northern margin of the lower member is thought to extend
linearly past the second (inlier) fault, but the tentative margin
is covered with alluvium. The tentative margin is apparently
displaced 10 meters to the north by a short, somewhat curved fault
occurring 45 meters east-southeast of the second (inlier) fault.
The eastern margin of the Epitaph body is a somewhat curved
fault which extends almost to the southern margin. A small offset
to the east is apparently defined by several small fault segments.
Near the northern end of the curved fault, the ·small streamcourse
flowing from the north curves to the west and then makes an abrupt
turn to the south, Past this turn, the stream flows along a low
cliff, which appears to be formed in the lower dolomitic beds of
the formation. Two thrust faults are exposed in this cliff, with
dips of 11 and 21 degree. The northwest end of the Epitaph outcrop
body also appears to be truncated by a fault striking northeast.
The dip, measured on exposed rock adjacent to the fault, is 56
degrees southwest.
On plate 3, Dr€wes shows an anticlinal fold extending almost
the entire length of the Epitaph outcrop. To check this portrayal,
a traverse was surveyed across the entire outcrop in its northwest
part, extending beyond the Epitaph on both sides (Fig. 6). Despite
some minor changes in dip angle, the dip direction always pointed
northeast, as shown in Fig. 7. Thus, the portrayal of an anticlinal
fold appears to be an error.
In the southwest part of this geologic map, the limestone
marker bed in the Gardner Canyon lower member is missing over a
length of about 14 meters; the presence of a fault is suspected.
Toward the southwest, the contact between the upper and lower
members is obscured. Along this same direction, there is a large
offset in the axis of the small dry streamcourse. Northeast along
this same axis, the contact of the lower member with the Epitaph
body is indistinct. These omissions may indicate a hidden fault.
PAGE

26

4~. / x40

'7

/',37
:'<,28

~

",...-

STRIKE

AND

DIP

ACROSS

ACROSS

NW

HILTON

RIDGE

FIGURE

6

EP ITAPH _ - - - o - j
DOLOMITE

1.----

ELE\/.,
FEET
_------, 100

r

THIN

FAULT\

LIMESTONE

50

.

"

:::::::J\..-===:::::::......-C------=----------------~O

DIP

ANGLES

ON TRAVERSE,

FIGURE

NORTHWEST

7

HILTON

RIDGE

cryptic outcrops and structure on Hilton Ridge
While examining the surficial geology in the southeastern part
of the Epitaph outcrop body, a number of anomalous outcrops were
observed. Careful study revealed that these were outcrops of the
Gardner Canyon lower member siltstones. The indistinct outlines
were so obscure and irregular it was necessary to survey them at a
scale of 10 feet per inch (120:1), in contrast to the normally used
scale of 50 feet per inch (600:1), over a rectangular area of 90 by
160 feet (27 by 49 meters). The map thus prepared (Fig. 8) showed
that the siltstone was exposed over an irregular area surrounded by
outcrops of the Epitaph Formation. The opening to Hiltons Hole was
in'the western end of the enlarged map area.
In order to understand the relationship of the siltstone lower
member and the Ep~taph, a topographic map with contour intervals of
two feet was prepared and depicted an elevation differential of
about 25 feet from a low elevation on the eastern end. (Fig. 9) The
map also showed that the siltstone exposures were slightly
depressed with respect to the slightly more resistant carbonate
outcrops. Numerous strike and dip readings taken over the enlarged
map area displayed a tendency to form groups of similar strikes and
dips, which tended to differ markedly from those of adjacent areas.
The groups are termed domains; three domains are shown in Fig. 10.
The anomalous and apparently chaotic structural relationships
depicted in the enlarged map of Figure 10 appear to indicate some
extensive tectonic deformation in this limited area. In addition to
the lower member siltstone, there is what appears to be a small
outcrop of the red upper member, about one meter in diameter. The
presence of the small upper member outcrop and the larger outcrop
of possible Scherrer sandstone may indicate extensive deformation
in that small area.
Ground Water and Cave Development
The small cave indicates movement of ground water which caused
solutional removal of th~ Epitaph. Inasmuch as the cave is bounded
by insoluble siltstone beds, it is probable that the acidic water
came from the subsurface, possibly through buried carbonate beds
and moving southeast toward a lower elevation. Reaching the
impediment of the apparently chaotic siltstone, the water may have
risen to the surface, forming a spring resurgence.
During surveying, three small areas of massive crystalline
calcite were found and noted on the map. The calcite is thought to
have been deposited by hot water seeps at a resurgence which once
existed here. The cave may have formed at a later time as the
result of subsurface dissolution; its entrance formed as the result
of surface erosion. At the present, rain water in the cave drains
internally. Flow marks in floor sediment indicate water movement
toward the west; the water appears to flow under a low ledge. There
are no nearby resurgences where the cave water may rise.

PAGE

27

39

'"
~ ----.,.

jct

DOMAIN

/

A

/

~

/

(NE)

2J-

/OOMAIN

/

15'>-.

/

11'>-.

STRUCTURAL DOMAINS NEAR HILTONS HOLE

FIGURE

10

L. BROD

C
(N) .

SEPT,2001

Hypothetical Cross section At The Cave
The position of the small cave near a major fault was both an
interesting and fortuitous occurrence. For these reasons, a profile
view was prepared extending through the cave in an approximately
north-south direction, as shown in Fig. 11. As shown on the figure,
the hypothetical profile was based on a number of constraints,
including dip measurements and the probable thickness of Epitaph at
this location. The profile was configured so that there was in the
subsurface a continuous path for water flow to eventually emerge at
the ground surface.
The resurgence of ground water at Hiltons Hole occurred at a
time when the future path of Sawmill Creek was still buried under
a thick blanket of sediment. Later erosion along Sawmill Creek
provided a better alternative route for groundwater, so the cave
drained and ceased to be a resurgence. This postulated history, of
course, is highly speculative.

PAGE 28

39

'<.

12
..'-

27

..>-

Dss
c==JIGU

15 -,...

o

I

GEOLOGY

YELLOW 55

UPPER MEMBER

hGL

LOWER MEMBER.

c==JR

L.-.-C..-----------------i
GEOLOGY NEAR HILTONS HOLE
10 20FT.
,
L. BROD 2001

o

LT.

EPITAPH DOL.

I

OF ENLARGED AREA ON HILTON RIDGE

FIGURE

(
I

I
~.J

/

4
20

18

2

a
16

=:::::±'?=:=:::j2P FT.

14

0t::::1

TOPOGRAPHIC MAP OF ENLARGED AREA, HILTON RIDGE
CONTOUR INTERVAL:

FIGURE

9

2 FEET

8

The Geology of Bells Ridge, at Cave of the Bells
Introduction.
Cave of the Bells is a large limestone cavern occurring in a
relatively narrow band of limestone. The limestone is bounded on
both sides by beds of insoluble siltstone and is also terminated on
the southeast end by the same rock. The configuration of the cave
and its hydrology is to a large extent controlled by the geology of
the rock in which it has been formed. An attempt to understand the
cave has provided an impetus to understand the geologic setting in
which the cave formed.
The cave occurs in the northeast corner of the Mt. Wrightson
7.5 minute topographic quadrangle, in the northeastern corner of
Santa Cruz County, Arizona. The geology in this region has been the
subject of a paper by Harald Drewes, the USGS Professional Paper
748, published in 1972. Plate 3 of this paper covers the area in
which the cave is located, at a scale of 1000 feet per inch. The
primary objective of the paper has been the structural geology of
this general area, but the paper also contains much information
about the stratigraphy of the area.
The primary Feature of Plate 3 is the Sawmill Canyon Fault
Zone, a narrow band of extensively faulted rocks. The fault zone is
important because of the fact that the deformation has brought to
the surface an extensive area of Paleozoic and early Mesozoic rocks
which otherwise would lie buried beneath extensive cretaceous
rocks. Two additional areas are' also included in PP748 and depicted
on Plates 2 and 4; Plate 1 is a structural map on which the
individual .outcrops have been omitted, and where the structural
relationships only are shown.
Cave of the Bells is reached by taking Forest Road 92 westward
from Arizona Highway 93 to the vicinity of onyx Cave, then turning
off on Forest Road 4084 to a junction with Forest Road 4086. This
latter road takes one a short distance to a parking area; a short
walk up the valley brings one to the entrance of Cave Of The Bells.
The single Cave entrance opens near the northeastern corner of
a limestone body about 175 metes wide. This limestone body trends
northwest with a nominally constant width for about 200 meters; the
northern margin merges with a similar limestone for a short
distance (about 55 meters) while the south block continues.
The cave occupies the eastern end of the roughly rectangular
limestone block. The cave entrance is near the bottom of a shallow
valley, and the height of the hill increases over passages farther
inside the cave. The cave is complex, with multiple levels forming
a random pattern in places. Two passages descend to water; one on
the eastern end of the cave descends steeply to a depth about 80
meters below the entrance. The cave contains deposits of fine red
clay, resembling that found in many Missouri Caves and thought to
be a deposit formed by extremely slow movement of water.
PAGE

29

stratigraphy Of Bells Ridge, At Cave Of The Bells
Bells Ridge is defined by several steeply dipping formations
caused by faulting toward the eastern part of the Sawmill Canyon
Fault Zone. The geology and structure of the local area is shown on
Plate 3 of USGS Professional Paper 748 (Drewes, 1972). The area of
interest is near the southeastern extremity of the fault zone,
where Permian carbonate rocks are interleaved with Triassic
siltstones.
Colina Limestone - The Colina is the lowest Permian rock exposed in
the local area, It consists of a dark gray limestone containing
some dark chert. Identification was made possible by the presence
of fossils of the large spiral gastropod, genus Omphalotrochus. The
Colina outcrops occur along the southwestern side of the carbonates
on ~ells Ridge. The southeastern margin of the Colina is poorly
defined, but it may lie at a poorly exposed transverse fault.
Epitaph Formation - The Epitaph Formation is usually known as the
Epitaph Dolomite. In this specific area however, the Formation is
a mixed limestone and dolomite, so that in this report the author
has termed it the Epitaph Formation. This formation is a light
gray, medium bedded rock which occupies the larger area between two
siltstone beds, as shown on the geologic map of the area (Figure
14). In this area the Epitaph contains clasts of light gray
limestone in a matrix of gray rock. The color and texture of the
clasts and matrix are similar, and the clast-matrix feature is not
readily evident. In this respect, these outcrops differ from those
exposed on Apache Ridge and in Lower Sawmill Canyon, where the
clasts in the upper Epitaph beds are usually tinted red. Another
component, a thin, yellowish-gray dolomite, was not observed here.
On Plate 3, Drewes shows three aspects of the Epitaph
Dolomite: undifferentiated, an upper member, and a lower member.
The upper member is specified as consisting of limestone and
dolomi te, while the lower member is specified as consisting of
marlstone and dolomitic siltstone. In this author's limited
reconnaissance, no outcrops resembling the lower member were
observed. The area above the cave, defined by faults on Plate 3,
was tentatively identified as the upper member.
Scherrer Formation - The Scherrer is not known to be exposed near
Cave Of The Bells. A very small area, unidentified but color coded
for Scherrer, was shown near the middle of the transverse fault on
Plate 3. This area appears to be near the terminus of the Colina
Limestone and was carefully searched, but no other outcrops were
observed here.
Concha Limestone - The Concha is a cherty, fossiliferous limestone;
in this area it outcrops only in a limited area near the edge of
the Colina and was identified primarily because of the large amount
of chert in the outcrops. No fossils were noted in the limited
exposure here, perhaps because most of the outcrop consisted of
chert which may have hidden fossils.
PAGE

30

Gardner Canyon Formation, lower member - The lower member of the
Gardner Canyon Formation is a tan siltstone, often having a light
reddish tint. Farther east, in the Lower Sawmill Canyon and on
Apache Ridge, the siltstone is almost universally tan. steeply
dipping beds of the siltstone are interleaved with the carbonate
beds. The siltstone is less resistant to weathering than the
carbonates; as a result, streams and valleys tend to occupy the
siltstone outcrops. The lower member contains a limestone marker
bed (Drewes' marker bed A), a short section of which occurs at the
edge of the parking area southeast of the cave entrance. The
limestone bed is only about one meter thick in that place.
Gardner Canyon Formation, upper member - The upper member of this
formation is generally a brick red siltstone. The red siltstone is
found at only two places on this author's map, one in a small area
at the bottom of the small streambed between the cave entrance and
the parking area, and the second in a small ravine at the southwest
extremity of the map where several faults converge (Shown on Plate
3). Farther south, the slope of the ridge diminishes and the flat
part of the valley, not shown on this author's map, is formed in
the red siltstone of the Gardner Canyon upper member. The
occurrence higher on the hill probably results from the presence of
a fault shown in this area on Plate 3.
In addition to the common red siltstone, the upper member also
exhibits outcrops of gray and gray-green siltstone. A number of
small outcrops with these less common colorations occur on the
hillside above the parking area. No additional formations or
members occur in the designated map area~
Younger, Surfaces - No younger features occur in this vicinity, with
the possible exception of alluvial deposits in the shallow valley
south of the surveyed area.
Mineralization
A mine shaft is shown on the Mt. Wrightson 7.5 minute map
immediately north of Cave Of The Bells. People who have visited the
mine site report that there is a nearby talus pile, indicating some
digging has occurred there. The snyder Mine, shown with three
adits, occurs on the opposite side of a small hill about 0.3 mile
away, to the northeast. Both mines are apparently abandoned.
A study of the mineralization in Cave Of The Bells was made by
Kendrick L. Day, and the results were pUblished in a newsletter of
the Escabrosa Grotto, Vol.6, no. 1, March 1976. Escabrosa Grotto is
a chapter of the National Speleological Society and meets in
Tucson. Mr. Day utilized X-ray diffraction to identify the presence
of calcite, aragonite, gypsum, hydromagnesite, amorphous silica,
and huntite. Two of these minerals contain magnesium, indicating
the proximity of magnesium-bearing rocks, such as dolomite. The
source of the magnesium is probably the Epitaph beds now exposed on
the surface.

PAGE

31

structure of Bells Ridge
Bells Ridge is a low lin~ar ridge extending west-northwest,
one of several generally parallel ridges formed by faulting. The
ridge is a westward extension of Hilton Ridge, a short distance
east-southeast of Bells Ridge. The ridge name is an informal name
derived from the Cave of The Bells, which opens on its northern
side. The structure and stratigraphy of Bells Ridge is shown on
Plate 3 of USGS Professional Paper 748 (Drewes, 1972).
The primary outcrop on Bells Ridge is the Epitaph Formation of
Permian age. Southwest of the cave entrance is a linear outcrop of
the underlying Colina Limestone. The linear bands of Epitaph and
Colina are the result of steeply dipping beds in both formations.
The two carbonates are bounded by similarly- linear beds of the
Gardner Canyon Formation lower member, a tan-colored. siltstone.
The cave entrance is formed upon a fault between the Epitaph
Formation on the south and the lower member on the north; the fault
trace is easily observed at the cave entrance. The fault can be
readily traced on the surface for 165 meters to a place where two
faults converge and the contrasting siltstone is terminated.
Drewes shows the entrance fault continuing in the Epitaph
beds, curving somewhat to the south and eventually reaching a
window of Concha Limestone. Surface surveys by this author found
the Concha outcrops, distinguished by large masses of chert, but
the fault as mapped by Drewes could not be followed over the crest
of the hill where the surface outcrops were minimal.
The Epitaph and Colina beds are bounded on the south east, and
north by beds of the Gardner Canyon lower member, On the eastern
end, faulted siltstone beds define an eastern boundary to the
carbonates, forming an insoluble barrier on the surface, between
the siltstones on the north and the south.
On Plate 3, Drewes shows an essentially linear siltstone band
of uniform width which forms the northern boundary of the Epitaph
outcrops forming the ridge. Mapping by this author, as noted,
showed that the faulted boundaries of the siltstone converged, and
that the limestone band was missing for a short distance. At this
point, the Epitaph of The ridge joined the Epitaph on what had been
the far side of. the siltstone. Some distance farther west, the
siltstone bed again reappeared and resumed its previous alignment.
Drewes shows a short anticlinal fold in the vicinity of the
Concha outcrop. A few steep dips have been noted in this area, but
a fold was not observed by this author.
The contact between the siltstone and the carbonate rocks is
often rather irregular, apparently consisting of a series of small
faul ts with limited length which intersect each other at odd
angles. One reentrant of the siltstone into the Epitaph occurs on
the southeast end of the area; several inliers and small windows
occur on the southwestern margin of the limestone.
PAGE 32

One window of lower member siltstone occurs on the fault
passing through the cave entrance. This window is completely
enclosed by the Epitaph Formation. Near the southeast corner, two
windows bounded by small faults are surrounded by siltstone, one of
Epitaph and the second of Colina. An apparent second small Colina
window occurs farther northwest along the larger body of Colina.
strike and dip readings taken over much of the mapped area
shows a number of high dip angles. Such high dips probably should
be expected in this type of setting.
Two small windows of the Epitaph are shown on Drewes' Plate 3,
near the southeast corner of the Cave Of The Bells block. One of
these small window was surveyed by this author. A fault was noted
near the southeast end of this window.
Hydrology
As noted previously, two relatively small pools of water are
found at appreciable depth in the cave. The surface of the eastern
pool, at the time of survey, occurred at a depth of about 270 feet
(82) meters. A careful examination of the Mt. Wrightson topographic
map showed that surface features did not reach a sUfficiently low
altitude until the confluence of Cave Creek and sawmill Creek was
reached. A search of Cave Creek revealed two small resurgences a
short distance downstream from the confluence, as shown on the
geological map of the resurgences in this report.
The geology of the resurgence area was mapped, showing that
the water rose through a thin limestone bed, a marker bed in the
Gardner Canyon lower member. This position of this marker bed was,
over a short distance, as low as it could possibly be. Only a few
meters to the east, the marker bed was displaced about 75 meters to
the north where it was too high to serve as a resurgence.
There are two and sometimes more places where the water flows
out to the surface. The primary resurgence occurs at the base of a
low ledge composed of or covered by soil. The water emerges onto a
wide but low platform, also composed of soil, only about 6 to 8
inches above the normal level of the creek. This resurgence is
generally exposed except during minor floods. The second resurgence
occurs about 27 meters upstream; the stream flows out at stream
level and it is generally difficult to monitor except during dry
weather. A third, tiny resurgence, occurs on top of a limestone
boulder and is essentially a wet-weather seep.
The southern resurgence, in Gardner Canyon, generally fails to
flow during a lengthy spell of dry weather. The resurgence in Cave
Canyon, however, has continued to flow over a period of several
years during which the flow was monitored, at least at monthly
intervals. That flow must be derived from somewhat more extensive
sources. It is believed that the flow comes partially from Cave Of
The Bells, from other smaller caves, and from beneath the bed of
Sawmill Canyon.
PAGE

33

Geology Of Onyx Hill
Introduction
Onyx Hill is a marrow ridge, the southeastern end of a longer
ridge in northeastern Santa Crux County, Arizona. The hill (ridge)
is terminated on its southeast end by erosion. On its northwest end,
the hill (as defined herein) is separated from the northwest
extension of the ridge by a saddle. Onyx Hill is noted primarily for
the presence of Onyx Cave, a cavern about 2 km in length, which was
once mined for its cave onyx about at the beginning of the previous
century.
The central core of Onyx Hill, in which the cave is formed, is
composed of a lower Permian limestone, termed the Rainvalley
Formation. The limestone is somewhat more resistant to erosion than
the surrounding outcrops, thus producing the ridge in which the cave
was later formed. In addition to minimal mining of the cave onyx
from parts of the cave near the entrance. a number of surface sites
were mined. During this author's geologic surface surveys, three
mines of ,varying size were found, all on the northeast side of the
hill near the contact of the limestone with insoluble rock. One such
site was opened into the hillside as a small room. The discarded
overburden and much massive crystalline calcite' was found
in a
large talus platform outside the mined room.
The geology of the area was first mapped in detail by Harald
Drewes of the United States Geological Survey. Three geology maps
and a structure map were published in Professional Paper 748 in
1972, with the emphasis on the structural geology of the area. One
of the three maps, Plate 3, was devoted to the structure and the
geology of the Sawmill Canyon Fault Zone, a southeastward oriented
band of faults extending from the Santa Rita Mountains to the edge
of outcrops on the eastward side of the range.
This author undertook a more detailed geologic mapping effort
in conjunction with a stUdy of the geology, morphology,and
hydrology of the cave. The surface geology was surveyed beginning in
November 2001 and continuing until February 2003. During the latter
part of 2002, the surface morphology along Cave Creek was also
stUdied, with a number of terraces along Cave Creek surveyed.
The surface geology of Onyx Hill was mapped along the contact
of the central li~estone, inclUding narrow peripheral areas. The
southwest part of the hill was difficult because of steep slopes
caused by the proximity of Cave Creek. The northeast side of the
hill was surveyed in more detail. surveying continued from the
southeast terminus of the hill to a saddle on the northwest where
the central limestone outcrop was terminated by quartzite.
The
4084. An
where it
revealed

northeast side of onyx Hill was bordered by Forest Road
igneous dike crossing over onyx Hill extended to the road,
was offset to the northwest. Detailed surveys of the dike
appreciable dimensional errors on Plate 3 of PP748.
PAGE

34

stratigraphy Of Onyx Hill
Detailed surveys of Onyx Hill by this author showed general
correspondence between those surveys and those published on Plate
3 of Professional Paper 748, with the principal differences being
the result of the differences in scale, in most places.
Paleozoic Era -

Colina Limestone

The Colina Limestone is a dark gray to black limestone with
minor amounts of black chert. In addition, fossils of the large
gastropod of the genus Omphalotrochus occur at various places
within the rock. Drewes' Plate 3 shows a very small outcrop area of
Colina Limestone toward the northwest, a short distance beyond the
bounds of the central limestone unit and beyond the limits of the
author's mapped area. However, a small outcrop of Colina Limestone
was found at the southeast end of Onyx Hill. There, a small block
of Colina Limestone containing an omphalotrochus fossil had become
dislodged from the hillside and rolled down onto a low terrace
along Cave Creek. No other outcrops of Colina were found, though
Colina rocks may occur in the subsurface.
Epitaph Formation
The Epitaph Formation in this general area is mixed limestone
and dolomite. Drewes found no trace of the formation on Onyx Hill,
but some rock in the southeastern part of the hill exhibited light
red inclusions similar to the upper, red tinted Epitaph beds found
on Apache Ridge and Lower Sawmill Canyon to the north. At the
present time, this problem has not been resolved.
Scherrer Formation
The Scherrer Formation is a mixed limestone and dolomite, with
a basal siltstone member. Onyx Hill exhibits a good display of
three members which have been observed in this general area. A
quartzite/sandstone bed adjacent to the central limestone probably
is the upper quartzite member. A· dolomi te bed southwest of the
upper quartzite bed is thought to be the middle dolomite member.
Because of a cliff near the base of the hill, a traverse from the
upper part of the hill was not attempted, but a thick quartzite
(generally white in color) observed at creek level appears to be
the basal quartzite member shown on Drewes' stratigraphic column.
The basal siltstone member was not observed here.
A small outcrop of light colored sandstone is found on a low
terrace on Cave Creek, probably a minor remnant of the Scherrer.
Concha Limestone
The Concha Limestone is a very fossiliferous, very cherty
formation of lower Permian age, younger than the Scherrer
Formation. It is not known to occur anywhere near Onyx Hill.

PAGE

35

Rainvalley Formation
The Rainvalley Formation is primarily a limestone, which forms
the central region of Onyx Hill. In addition, a more dolomitic set
of beds to the northeast, in conjunction with sparse beds of
sandstone and siltstone, are reported by Drewes on Plate 3.
Twp additional exposures to the southeast form a generally
linear set of limestone hills. These three hills and a smaller
exposure of the limestone northwest of Onyx Hill constitute the
total exposure of the Rainvalley in and near Onyx Hill. Also, a
tentative exposure of the Rainvalley Formation has been shown on
Plate 3 north of the Apache Ridge, but its presence there has not
been verified in the present study.
The third Rainvalley exposllre occurs almost a mile southeast
of Onyx hill, where the formation intersects Gardner Canyon. Here,
the limestone beds are somewhat shaly and include beds of mudstone.
The mudstone beds are unconsolidated, resembling silty yellow soil,
and are easily eroded. The Rainvalley exposures there appear to be
broken up to a large extent and mixed with outcrops of the Gardner
Canyon upper member siltstone.
Additional Rainvalley outcrops are shown northwest of the main
Rainvalley body containing the cave. This smaller outcrop was not
examined during this survey. A small outcrop, apparently designated
as Colina by a small identifier, occurred only 60 meters beyond the
surveyed body of RainvalleYi a brief examination revealed that it
probably a Colina outcrop, but no careful study was made.
Profile B-B' on Plate 3 shows that the Rairyvalley beds on Onyx
Hill are almost vertical and sloping steeply northeast. Also, the
dolomitic marker horizon is defined by an almost straight line (d)
about at the middle of the Rainvalley outcrop area. Inasmuch as the
horizon d is specified as the base of a more dolomitic sequence,
higher beds in this sequence would occur in the northeastern part
of the Rainvalley outcrop band. No effort was made to determine the
approximate border of the dolomitic sequence.
The Rainvalley Formation contains some fossils but they are
generally sparse and in poor condition. Some apparent fossils are
of questionable origin and unidentifiable. Identifiable fossils are
found inside the cave but their stratigraphic position is unknown.
Numerous fossils occur in the cave 85 meters from the entrance, in
what is called The Garbage Pit, a short lower section under a
descent of about 10 meters.
Some horizons contain profuse chert. outside, along the cave
trail, are rust colored nodules of what appears to be chert. Broken
surfaces of some nodules show that they contain sand in addition to
chert.

PAGE

36

Mesozoic Era - Triassic System
Gardner Canyon Formation, Lower Member
The lower member of the Gardner Canyon Formation is commonly
a tan colored siltstone containing beds of white limestone. On Onyx
Hill, however, the tan siltstone was not observed. Drewes shows a
relatively narrow band of the lower member between the northeastern
Rainvalley margin and the upper member. This band continues along
the Rainvalley outcrop from its eastern end northwest to a mapped
fault, which cuts across the Rainvalley beds. within the noted area
primary outcrops consist of red siltstone and some red sandstone,
unlike typical lower member outcrops. These will be tentatively
considered to be the lower member.
The area within this band contains four narrow beds of white
or nominally light colored limestone, reminiscent of the lower
member marker beds. These beds, however, become sparse and vanish
before reaching the fault. Based entirely on the presence of the
marker beds, the narrow band thus defined appears to be the lower
member of the Gardner Canyon Formation, as shown by Drewes.
The upper member of the Gardner Canyon Formation consists of
a brick-red siltstone. It extends along the entire northeastern
margin on the northwest, and along the probable lower member on the
southeast, On the northwestern end, the RainvalleyLimestone is in
contact with what appears to be the upper member of the Scherrer
Quartzite. The Rainvalley-Scherrer contact continues up the hill
southwestward to the crest of the ridge and down the steeper
southwest side. This contact continues down the southwest side of
Onyx Hill, eventually enclosing the main Rainvalley limestone body.
Tertiary System
Igneous Dike, of Uncertain Age
A narrow igneous dike'of Tertiary age has been depicted on
Plate 3, extending from a larger batholith south of Cave Creek,
across Onyx Hill and terminating north of Sawmill Creek, with some
gaps.The dike is fairly well defined north of Onyx Hill, but its
trace over the crest of Onyx Hill is not shown. The batholith and
dike are composed of rhyolite porphyry, according to Drewes.
A dike segment has been surveyed by this author south of Onyx
Hill, extending northeast. If this orientation were maintained, it
would intersect the cave about 30 meters northwest of the entrance.
However, no such intersection has been observed within the cave.
Followed southwest on the northeast side of the ridge, the dike
ascends the hill and would intersect the hilltop southeast of the
cave. Unfortunately, outcrops of the dike are sparse near the hill
crest. On the basis of very sparse dike fragments, it appears that
the dike crossed Onyx Hill southeast of the present cave entrance.
It was sUbsequently displaced northwest on the south side of the
present cave, thereby missing the present cave entirely.

PAGE

37

Adjacent Igneous Deposition
Although not directly associated with onyx Hill, the dike
extends northeast into a region of igneous rock which may have had
some effect upon the general geology of the area. The region of
interest lies between northeastern Onyx Hill and southwestern
Lower sawmill Canyon. Specifically, the area lies northeas~ of the
Forest Road, FR-4084 and the south side of Sawmill Creek.
There are two components listed by Drewes on Plate 3, both of
which are considered to be members of the Gardner Canyon Formation,
most likely because they are interbedded with conglomerate of that
formation. The first component, age dated at 192 Ma, is listed as
dacite flows, intrusives, and tuff. The second component is listed
as a volcanic breccia in a mudstone matrix, apparently fluvial.
Some reconnaissance mapping was conducted by this author in
order to understand the relationship of these components to the
Onyx Hill and the Lower Sawmill canyon areas . One finding: the
igneous dike cut the older volcanics and thus was younger. One
addi tional observation was that the highest part of the ridge
between FR-4084 and Sawmill Creek appeared to be a deposit of
almost pure quartz. The quartz outcrop appeared to be roughly about
thirty feet (roughly 10 meters) in diameter.
Tertiary Gravels
Onyx Hill is relatively high; the crest stands about 75 meters
above its base. However, there is no trace of the coarse Tertiary
gravel found on other high hills in the vicinity. On profile B-B'
of Plate 3, Drewes shows the Pliocene Pediment intersecting Onyx
Hill below the crest of the hill. If this is the case, residual
gravel would have been eroded from the steeper hillsides. Remnants
of the coarse gravel, primarily sandstone, occur near the base of
Onyx Hill in Pleistocene terraces and along Cave Creek downstream
from the cave. The coarse gravel is generally covered by finer
sediments one to two meters thick.
Pleistocene and Holocene Sediments
Younger deposits consist primarily of fluvial deposits laid
down by streams. Eolian deposits may be present in small amounts
but have not been identified.
Mining
Mention has been made previously of the limited mining of cave
onyx at Onyx Hill. The only other known trace of mining occurs in
the form of a small test pit in the northernmost limestone marker
bed near the base of the hill. There is no sign of mineralization.
Two old concrete platforms occur a short distance north of FR-4084
and an old, circular concrete tank occurs south of the road at the
same locality. These remains may have been associated with mining
of some sort, possibly the onyx mining on onyx Hill.
PAGE

38

structure Of Onyx Hill
The structure of Onyx Hill is fairly simple. On section B-B'
of Plate 3, Drewes has depicted a cross section passing through
Onyx Cave, showing the beds there sloping steeply toward the
northeast. Several faults are shown essentially parallel to the
bedding planes.
The Permian and Triassic rocks associated with Onyx Hill
constitute the southern side of the Sawmill Canyon Fault Zone.
The Fault Zone is an intensely faulted area, spreading from a
vertex on the northwest to a wider area on the southeast, where the
Paleozoic and Mesozoic rocks are buried beneath Cenozoic Gravel.
Across the Fault Zone, from southwest to Northeast, are three
bands of Paleozoic rocks, separated by bands of Triassic rocks. The
first band, to the southwest, is the Onyx Hill geologic area. The
second band contains the Lower sawmill Canyon, Hilton Ridge, and
Cave Of The Bells geologic areas. The third band, to the northeast,
contains the Apache Ridge geologic area. These three areas contain
features of speleological or hydrological interest. The presence of
these features is thought to be highly dependent upon the geologic
structure.
The fault on the southwest side of the Onyx Hill Paleozoic
beds separates the Paleozoic rocks from extensive Cretaceous
outcrops on the southwest side of the sawmill Canyon Fault Zone. In
addition, the batholith feeding the igneous dike crossing Onyx Hill
occurs among the Cretaceous outcrops.
Smaller faults occur along the edges of the primary limestone
outcrop, generally extending along or across the limestone contact.
All appear to be fairly short and do not appear to extend for more
than about 30 meters. Two apparently short faults cut completely
across the limestone at and near its northwestern end. Three (or
perhaps' more) small faults occur along the contact between the
limestone and the Scherrer upper member toward the western margin.
Several short faults apparently occur in the vicinity of the cave
entrance. One relatively short fault offsets a limestone marker bed
near the southeastern end of Onyx Hill.
The Tertiary dike extends down Onyx Hill in a relatively
straight dLrection. However near Forest Road 4084, the dike turns
abruptly northwest and continues in that direction for over 100
meters, crossing the road at a low angle. The dike then turns to a
general north-northeast direction' and continues in that direction.
The dike is believed to follow fault fractures.
The thin limestone marker beds which are components of the
Gardner Canyon lower member extend northwest from the southeastern
base of Onyx Hill. The beds continue northwest for at least half
the length of Onyx Hill. The beds apparently terminate just prior
to a faUlt, leaving what appears to be a blank space. The blank
area may indicate a structurally disturbed area.
PAGE

39

Surficial Deposits
A number of surficial deposits of possibly hydrothermal orlgln
occur on Onyx Hill. Of special interest are the massive calcite
deposits which apparently were dug during mining activity. Some
limi ted mining of cave onyx (calcium carbonate wall and floor
deposits) occurred, apparently around the beginning of the 20th.
century. The cave onyx was used for building facings, but the onyx
was sensitive to acidic rain and the market for this material was
short lived.
At least three small onyx mines occur on the northeast side of
Onyx Hill, a short distance above the contact of the limestone with
the underlying siltstone. The largest mine occurs about 100 meters
northwest of the igneous dike. There, large blocks of crystalline
calcite have been stacked outside the mine face to form a level
platform. Two smaller mine features occur farther northwest near
the contact, the second only a short distance above a short fault.
Other unmined calcite sites occur along the hillside near the
contact. In places, the calcit~ appears to occur where the calcitebearing water seeped out along a bedding plane or fault fracture.
All known exposures of calcite occur on the northeast side of Onyx
Hill, which exposes the upper side of the steeply sloping beds.
In ohe such place, a peculiar appearing rock was found and
later examined. The rock at first view exhibited what appeared to
be numerous small pores, similar to a piece of bone. However, under
more careful examination the rock was seen to contain numerous and
closely spaced, small pyramidal depressions. The rock appeared to
be a cast of a group of very small crystals, about 3 to 4 mm in
length and 1 mm or less diameter. Planar crystal facets occurred on
the interior surfaces. The matrix material was tested and found to
be quartz, Apparently the quartz overcoated the crystals, which
were later dissolved.
Other mineralization occurs near the crest of the ridge, along the
footpath leading to the cave entrance. This mineralization it in
the form of thin coatings, some of contrasting colors, appearing to
be concretionary over an uneven substrate. The coatings apparently
were deposited by water.

PAGE

40

Terraces In The Vicinity Of Onyx Hill
Cave Creek borders Onyx Hill on the southwest side, with a
widening terrace in the upstream direction, which narrows and
disappears as Cave Creek reaches the bottom of Onyx Hill. There,
Cave Creek turns to the northeast and the terrace then again
appears on both sides of Cave Creek. The terrace continues along
Cave Creek and continues on one or both sides of the creek for a
considerable distance. The age of this terrace is unknown to this
author, but it is probably late Pleistocene.
!n places where road cuts and stream erosion has cut into the
terrace,it can be seen that the terrace is depositional, with about
a meter of finer sand and silt overlying an underbed of coarser
cobbles and small boulders probably derived from the coalesced
alluvial fans bordering the eastern flank of the santa Rita
Mountains. The coalesced alluvial fans were deposited in late
Tertiary time and began to be eroded at about the beginning of
Pleistocene time.
This widespread terrace along much of Cave Creek is the oldest
of a sequence of terraces, with three younger terraces being more
limited in extent. The oldest, highest terrace is termed Qt4, as
shown in Figure 12. One small , slightly lower "terrace remnant is
termed Qt3. A lower and somewhat more extensive terraCe occurs
primarily along Cave Creek, downstream from the small side canyon
entering from the south. This third terrace, Qt2, extends eastward
immediately adjacent to Cave Creek and continues northeastward past
the Forest Service Road, FR-4085. A smaller segment of this terrace
about 80 meters long occurs on the south side of cave creek. An
even smaller remnant occurs where the smaller side canyon joins
Cave Creek. The fourth terrace, Qt1, occurs on the south side of
Cave Creek from the small side canyon eastward past FR-4085 and
westward on the north side of Cave Creek for about 140 meters.
The northern occurrence of Qt1 begins at an isolated exposure
of a light gray quartzite, possibly the Permian Scherrer formation,
where the Qt2 narrows and descends about one meter to the level of
Qt1. It appears that the quartzite pedestal shifted the flow of the
creek toward the south, thus preserving remnants of the higher
terrace along Cave Creek.
A remnant of the higher terrace, Qt4, occurs upstream ,a short
distance prior to where the lowest terrace terminates. This terrace
remnant extends only a short distance upstream (beyond the limit of
the map). It terminates where the steep, stream-cut bank meets the
basal slope of Onyx Hill. At this remnant of terrace Qt4, Cave
Creek is rather narrow, confined between a cliff of conglomerate on
the south and the steep bank below the edge of the terrace.
Farther upstream, the canyon widens somewhat. About 150 meters
upstream, a cliff of white quartzite occurs above a lower bed of
conglomerate. Somewhat farther upstream, another terrace occurs,
probably a larger remnant of the higher terrace, Qt4.
PAGE 4\

SW

NE
F2

F3

GEOLOGIC PROFILE AT HILTONS HOLE
CONSTRAINTS ON STRUCTURE NEAR HILTONS HOLE
1.
2.
J.
4.
5.
6.
7.
8.
9.
10.

Epitaph about 100 ft. thick (100 ft. thick at Hidden Cave).
Underlying lower member approximately horizontal NE of F4.
Colina greater than 200 ft. thick NE of Fault F4.
Dip of Fault FJ matches dip of Epitaph at Fault FJ.
Dip of Fault F2 matches dip angle of beds SW of Fault F2.
Dip of lower member SW of Fault F2 matches dip near F2.
Dip of Fault F1 matches dip of beds further SW.
Low angle Fault F1 continues to Fault F4.
Beds between FJ and F4 bent by slight fold, because of dips.
Epitaph on both sides of Fault F4 overlap.

o

50

100

150

200 FEET

1:: 1====±I===::=11====1::1===::::il

FIGURE

II

\ / /'
/<./

'~

' . . , - -li::~.»~</\
...--_-_.. . "'-J
~

\

/_'
//

/;>'-/
\

\/
. . . ./

---

_---

/_
QT4

--

____ - -

c:-----

__

/

QAL
Q,
_,
-- -_
-,
_

/

----Q
-----~
-"' _
. _

__ '

--,

----_/

•

_...

-'---. '

,

-"
, __ -QT2'

.... ,

AL

/___
,Q
'- __
' - ,,'_

I

I

.-::----...-"

I

Q12

/-

(

T2

a

,

,/

Q

Q TI
SYSTEI1

FORMATION
OR SERIES

MEMBER
OR UNIT

! ~.:.

N

w
z
w

QUATERNARY

N/A

ALLUVIUM

U

,

OU
w O
"N

CRETACEOUS

FORT
CRITTENDEN

TRIASSIC

GARDNER
CANYON

PERMIAN

SCHERRER

"'~~
.... 0

<N
0..

BROWN
CONGLOMERATE
LOWER
MEMBER?
LOWER
QUARTZITE

/'

/'

Q SG

ALLUVIUM

QAL

LOWEST
TERRACE

Q TI

MED-LOW
TERRACE

Q T2

MED-HIGH
TERRACE

Q T3

HIGH
TERRACE

QT4

I ~~~"I
[i]

KFB

c=:J'

COM GLOMERATE.
LlTHARENITE
WHITE
LIMESTONE
LIGHT GRAY
QUARTZITE

PsL

NM

SANTA CRUZ COUNTY, ARIZONA
COMPASS AND TAPE SURVEY
BY LANGFORD G. BROD JR,

2002

"':o'~

c:J

/"

EROSION SURF ACES ALONG CAVE CREEK

0

~GL I ;. ~'}I

...--

~'"

N

L:J
LJ
LJ'

STREAM
GRAVEL

"

TI

LITHOLOGY,
LOCATION

U

0

/

'--

,

........

ERA

- -Q
jn

/ ....
...............
/
I/ I
Q" /
) ' Q.... -- - . _ '' , -_/
- / / /-,,--- , - ----

I,,-.\\
'----

_---"

K FB 7

Q

/

, , 0 ) ....
__

'~..L..-::::. I3 -":3

_.\

-~PSL'//
-, , --, / ' --- - ~
-

/

-------~~-::.;-=~~~?z,
----- -~ ~..~ ',------

-- --2'
--

Q

/

AL

,,'_

/

--,

____ -'

.~I':::'
,----' -\

,,/

__,

QT4,~~''''
------- -----------/---- :---

I

Q"
.

,,

. ,.,

Q " "\

I '

/

-

.

,/

-,,,

"

~

/

?,
oI

50

'

100

150

,

200

,

,

250

"

20

40

60

BO

!

!

!

,

SCALES

OF LENGTH

300 FEET

!

100 METERS
!

FIGURE

12

Hydrology
Ground water enters Onyx Cave through cracks and fissures in
the overlying limestone, forming drips and small pools. The water
ultimately drains from the cave. Cave passages containing mUddy
sediments occur in the northwestern part of the cave, in passages
aligned approximately along the strike of the steeply dipping beds.
At the present time incomplete surveys have traced these apparent
drains southeastward along the northeast side of the hill. The
passages have descended about 50 meters below the nominal upper
level of the cave and are continuing downward. The end of surveyed
passage is about 20 meters above the base of Onyx Hill.
The first likely site of a resurgence occurs where Forest Road
4084 crosses Cave Creek. This place is just beyond the base of Onyx
Hill. The Rainvalley bed should cross here but the site is covered
by alluvium. Furthermore, the tentative presence of faults shown on
Plate 3 may offset soluble limestone in the subsurface.
In order to monitor the presence of any water rising into the
gravel, careful examinations of the stream was carried out on a
number of occasions. It was found that the gravel bed terminated
about 280 meters upstream from the road crossing and about 170
meters downstream, where the stream when flowing emerged onto the
insoluble red siltstone of the Gardner Canyon upper member, shown
on Figure 13. When the flow of water ceased at the downstream
gravel terminus, it also ceased at the upstream end, thus implying
no additional efflux of water from the cave.
A reconnaissance along Gardner Creek, where the third outcrop
of Rainvalley limestone occurred, revealed a small resurgence where
water was flowing from a small pool about two meters above the
creek bed. The status of this resurgence was checked at intervals;
The flow from the resurgence ceased on several occasions after long
intervals of meager rainfall. It is considered likely that this
resurgence is fed from Onyx Cave; there is no other likely source.
A map of the resurgence area is shown in Figure 14.
No other possible resurgences occur to the southeast, as the
extension of the Rainvalley bed is lost under Cenozoic gravel.
The approximate flow was measured with a 4-liter pail and a watch
at a time when the flow was relatively strong, and it was found
that the discharge was about 2000 gallons per day (equivalent to
7.57 cubic meters per day).
When the flow ceases the rise ·pool becomes completely dry.
Conversely, when the resurgence is running, the water cascades down
a number of step-like terraces to the creek, and fractures in the
red (upper member) siltstone close to the creek bed exude small
seeps. The presence of the resurgence on Gardner creek is perhaps
facilitated by the presence of two closely spaced faults crossing
the creek at this locality, as shown near the east edge of Plate 3.

page 42

WATER EMERGES

Q AL

l'\ NM

-.-

-,

~_~-

Q
AL
Q- - - - - _ - - - - - - \
_

/
_------

//

/

/

/

I

--

___ ...! _

_

---- -

-- ~

___.

/__
/

Q

TI

-

//

I·

I

,-_----

",.....-----*"'" . / /

\I

--

QT2 .

T

I

EASTERN PART

'---!

AGU

POOLS DOWNSTREAM

WESTERN PART ,~

/

,_~/ FR-4084

/

-_- - - - - -Q- - - - - - - -

/

I

-_...------ ......

....

//

,/

(

I

:
--- ----- - --- ------

/./

I

STREAM
CROSSING

TZ

...

./

--

/

/-----'---"

/

---- - - - - _ / /STREAM

./

CROSSING

FR-4084

I

-j
/

"",-

---- --Q~t.
,

"./

,.."/

(

STREAM SINKS

,,/

----_....:--'.
- - ..... ~,

/' .-.c'
//

0T!

/

------- ___ _

-----------"--

/------

OT!

--

-_

-;..-Q

n

------ /

-7

,//

//

/

/

STREAM HYDROLOGY ALONG CAVE CREEK NEAR ONYX CAVE
,

/

I:

FIGURE

13
¥'~.

N

NM

scattered

.... "'-

main
stream

....

p,
10

path
--~

p

/-"/1/1111/1/....

~

'fI/I",,'l;.

dry
stream

., i

~
Triassic - Gardner Canyon Fm.
upper member

DRGU

) OM

Triassic? - mudstone

Dp
RESURGENCE

AREA

ON

FIGURE

GARDNER

14

CREEK

Io

I

I

It

Permian - Rainvalley Limestone

I

I
5

~IO.

I

I

~e ~EET
10 METERS

STRATIGRAPHIC COLUMN FOR SELECTED' AREAS
NORTHEAST SANTA CRUZ COUNTY, ARIZONA
"

ERA
u

<3

SYSTEM
QUATERNARY

N
0

zw

u

TERTIARY
CRETACEOUS

FORMATION

MEMBER

N/A

N/A

POST PLEISTOCENE TO RECENT ALLUVIUM

N/A

N/A

PLEISTOCENE

N/A

N/A

PRE-PLEISTOCENE ALLUVIAL FANS

N/A

N/A

RHYOLITE DIKE

FT. CRITTENDEN

B.C.

TRIASSIC

UJ

L:

PERMIAN

PERIPHERAL, SOUTHWEST SIDE OF ONYX HILL

LOWER

TAN SILTSTONE, MINOR RED SANDSTONE
THIN LIMESTONE MARKER BEDS, MINOR
SEDIMENTARY BRECCIA, LOWER SAWMILL MAP

GARDNER CANYON

RAINVALLEY

N/A

CONCHA

N/A

PRIMARILY LS, SS 8. SILTSTONE NOT OBSERVED
ON ONYX HILL, DOLOMITE NOT IDENTIFIED
MUDSTONE PRESENT AT RESURGENCE
CHERTY, FOSSILIFEROUS LIMESTONE PRESENT
ON APACHE RIDGE AND BELLS RIDGE

MEMBERS

3 MEMBERS PRESENT ON ONYX HILL
LOWER MEMBER PRESENT ON HILTON RIDGE?
BASAL SILTSTONE MEMBER NOT PRESENT

UPPER

LS, SOME DOLOMITE,IN MOST MAP AREAS

LOWER

LOWER MEMBER NOT PRESENT IN MAP AREAS

COLINA

N/A

LS, MINOR CHERT, PRESENT IN MOST AREAS

EARP?

N/A

PRESENT IN LOWER SAWMILL CANYON?

SCHERRER

0
N
0

4

EPITAPH

--l

<{

c..

PENNSYLVANIAt
PERMIAN
MisSISSIPPIAN
DEVONIAN
SILURIAN
ORDOVICIAN
CAMBRIAN

U

NOT PRESENT IN MAPPED AREAS

CONTINENTAL GRANODIORITE PRESENT NORTH OF SAWMILL CANYON

w
0:
c..

B.C,

;

UPPER

u

w

TERRACES

MAINLY RED SILTSTONE, LESSER SILTSTONE
OF OTHER COLORS, MINOR RED SANDSTONE
CONGLOMERATE, THIN LIMESTONE MARKER BEDS
VOLCANIC COMPONENTS NOT PRESENT

u

<3

N
0

(/)

LITHOLOGY, LOCATION, CHARACTERISTICS, COMMENTS

=

BROWN CONGLOMERATE OF CRETACEOUS SYSTEM, SOUTH OF CAVE CREEK,

FIGURE

15

References
Bryant, Donald L. 1968 Diagnostic Characteristics of the Paleozoic
Formations of southeastern Arizona: part of Southern Arizona
Guidebook III, pp 33-47
Drewes, Harald 1972 Structural Geology of the Santa Rita
Mountains, Southeast of Tucson, Arizona:
USGS Professional Paper 748; 35 pp, 4 maps
1980 Tectonic Map of Southeastern Arizona: USGS
Misc. Investigation Series, Map 1-1109
Moore, Raymond C., Lalicker, Cecil G., and Fischer, Alfred C. 1952
Invertabrate Fossils: 766 pp

PAGE 43

Appendix I - Methodology
The methods described here were developed over several decades
of surveying, initially in cave surveying and later in the
surveying of surface geology. The emphasis is on simple instruments
for single person use, employing hand-held devices.
The primary instruments are a sighting compass, Suunto KB-360,
which utilizes a 0~360 degree, liquid damped drum dial. The compass
contains a magnifying lens to observe the drum at the same time the
user observes the target station over the top of the compass case.
The instrument is fitted with a lanyard which can be worn around
the neck and stored in a shirt pocket. The clinometer is similar to
the compass; it employs a liquid damped drum whicll can be read
through a magnifying lens at the same time as the target station is
seen. The clinometer is also fitted with a lanyard. The instrument
is graduated both in percent grade and degrees, but only the degree
readings are used.
Length measurements are made with a fiberglass tape graduated
in feet, tenths anq hundredths of a foot; only the feet and tenths
are read and recorded. The tape is stored in a light-weight plastic
reel with a wind-Up handle. A steel folding hook installed at the
leading end of the tape was removed and replaced with a V~shaped
stainless steel wire. The apex of the V was made to correspond with
the zero end of the tape. A six-foot length of thin nylon cord was
attached to the V for attachment of the tape to a sighting station.
In use, the leading end of the tape is attached to a starting
station with the nylon cord, and the tape is unreeled until a
target station is reached, at which the tape is draped over the
target so that the tape is taut. Readings of tape length, compass
reading, and clinometer reading are then recorded on the survey
sheet, along with the identifying station numbers.
Recording of the survey data is made on a sheet of 8.5 by 11
inch quadrille paper clipped to a 9 by 12 inch clipboard. Also
attached to the clipboard are a 6-inch rule, graduated in inches
and tenths of inches. a 360 degree protractor, and an automatic
pencil with an attached eraser. All three items are attached with
short lengths of thin nylon cord so that they cannot be lost. The
plastic protractor is a sterling No. 535, 3.. 5 inches in diameter,
fitted with a thin 3.5 inch acrylic backup plate for protection.
Prior to use, the sheet of quadrille paper is clipped to the
clipboard; the geographic location, survey location, date, and page
number are recorded in the upper left corner of the sheet. On the
upper right is the tabulation of the data: origin station, next
station, station line length, compass reading, clinometer reading,
and two initially blank columns. The first column is intended to be
used for the foreshortened station line length; the second is used
to record the plotted length as determined by the scale factor. The
two columns are used by the author when manually plotting the data.
The space between the title data and the tabulated readings can be
use for a brief description of the station.
PAGE 44

The recorded data is plotted on the lower half of the paper,
bearing in mind that the data tabulation will become longer as more
data is recorded. strike and dip readings, if made, can be recorded
in unused places on the recording sheet. Experience has shown that
all data for a single mapped area should be on a single survey data
sheet, at least initially.
When a survey sheet has been sufficiently filled, it can be
removed from the clipboard and placed in a fairly stiff plastic
envelope, which is also where unused quadrille sheets are stored.
In later surveys of the same area, segments of the previous survey
may be included in a subsequent survey sheet, thus providing a
starting area where the next survey can start.
survey stations are selected from various natural objects
which occur on the area to be mapped. Trees are often ideal, as
they can provide stations near eye level and above obstructions on
the ground. The best trees are ones with sizable, non-flexing
branches. Thin limbed bushes and trees are avoided, as well as
trees with many bushy branches, such as some cedars. A branch 3/4
to one inch in diameter, perhaps connected to a larger nearby
branch, is often ideal. other station sites may include isolated
rocks with a pointed upper end, to which the tape can be attached.
Whatever natural item was used, it had to provide a relatively
clearly visible position, sUfficiently high above the ground 'so
that the tape was not caught on low grass, bushes, or rocks. Nonnatural features were sometimes used, provided they were nonmagnetic. Steel fence posts have been used as stations, but it is
necessary to take compass readings some distance from the fence.
In one instance, on the crest of a hill, no natural station
could be found, so it was necessary to erect an artificial.station
using a garden stake. The stake was tethered with three cords tied
to the stake and to small stakes set in the ground at 120 degree
positions. After use, the stakes were removed but a marker was left
at the base of the central stake for possible future recovery.
The angular ~ccuracy of the Suunto hand-held compass is lower
for higher inclined sightings, so station networks were employed in
some cases. In this method parallel survey lines, perhaps 50 feet
apart, were interconnected with intermediate survey lines, £orming
a number of interconnected triangles. This method was used with
some success on the steep southwest side of Onyx Hill.
Geologic Aspects Of The Survey
Despite some small errors in the compass and tape surveys
described in this report, the result was considerably more accurate
than the use of aerial photographs. In some cases, the survey areas
occurred beneath forest cover, thus precluding the use of aerial
photos. Also, the scale factor, 50 feet per inch, provided a much
more accurate illustration of surface geology.

PAGE

45

Tn some cases, the contact, such as that between limestone and
siltstone, was so obvious it was easily plotted. For such cases,
the survey line was placed close to the contact (generally less
than 10 feet) and its position was drawn on the survey. The same
method was used for exposed fault traces. For less clearly defined
contact and faults, the estimated position of separated outcrops
was plotted by measuring along the tape for an outcrop and
estimating the distance of the outcrop from the tape. For larger
outcrop areas, an effort was made to draw the outcrop to scale.
strike and dip readings were made with a magnetically damped
Brunton compass wherever the bedding plane or fault surface was
visible. Inferred faults were drawn with dashed lines when not
readily apparent. In some cases, the apparent displacement of the
thin limestone marker beds in the thicker siltstone beds (as noted
by Drewes) was utilized to define a fault trace.
Small samples of various outcrops were obtained and examined,
in some cases with a 10 power Hastings triplet magnifier. Relative
hardness of samples was tested using a steel scribe. The relative
reaction of samples was tested with dilute hydrochloric acid. A
small magnet was used to test for possible magnetic activity, but
none was found. The color of selected samples and differences of
various samples was examined using the Munsell Soil Color Chart
(handbook), as described in Appendix 2. A monocular microscope was
available but not used.
.
Two minerals were found in readily identifiable form on the
surface. A number of large loose crystals of quartz were identified
by their crystal habit. Some large crystalline calcite was readily
identified by its internal cleavage. A number of small pieces of
hematite were tentatively identified by their appearance; some of
these pieces were found in the vicinity of a test hole near the
Arizona Trail.
A number of fossils were observed, most in a limestone matrix.
The large gastropod, omphalotrochus, was noted in a number of
plabes and was often displayed on a plane perpendicular to its
central axis. The large brachiopod, Dictyoclostus, was observed
silicified with chert in limestone, and were observed inside Onyx
Cave. Bryozons were observed near the crest of Apache Ridge in
Permian limestone. Identification, when possible, was made with the
aid of Invertebrate Fossils (Moore, Lalicker, and Fischer, 1952).

PAGE 46

Appendix 2
Survey Markers On and Near Apache Ridge
Three survey markers were found during surveys of Apache
Ridge. One marker was found within the survey area and two were
found on the periphery of the area. Others may be present but were
not found. The three markers are listed in the sequence of their
discovery.
First survey marker found: Reference - Survey 8, page 1
Location: About 10 feet north of FR-92, about 1 foot west of the
barbed wire fence, east of Arizona Trail parking area
Markings: PE5135 No. 8A SW i/16 86
Setting: Short metal pipe,
Second survey marker found: Reference - Survey 34, page 2
Location: South edge of FR-92 ,about 205 feet west of elbow on
water pipe below steel water tank; a short distance from
Apache spring.
Markings: C S 1/16 S6 PE5145
Third Survey marker found: Reference - Survey 274, page 1
Location: About 90 feet east of straight section of Arizona Trail,
and about 120 feet south of ravine near north side of the
limestone outcrop.
Setting: short metal pipe in rock cairn.

PAGE

47

Appendix 3
White Limestone Bed, Eastern Apache Ridge
An outcrop of white limestone was found near the eastern end
of Apache Ridge, not far below the crest. The bed occurred at the
north edge of Colina outcrops, at the contact with siltstone of the
Gardner Canyon Formation. This bed was in sharp contrast with the
Adjacent Colina outcrops, which are generally black limestone which
weather to a dark gray color.
On closer examination, the white rock is observed to contain
small specks of a darker color. Under low magnification, the specks
are shown to be semi-transparent, appearing to be small masses of
calcite. Treatment with dilute hydrochloric acid causes obvious
effervescence on both the darker inclusions and the white matrix
which comprises the majority -of the rock. It appears that the dark
color of the inclusions is an optical property in which incident
light entering the calcite specks is lost internally.
The specks of calcite range up to about 3 mm in size, though
many are smaller. The position appears to be aligned along a rather
irregular bedding, though this is a somewhat tentative conclusion.
The white rock does not resemble any other outcrops on Apache
Ridge or the other geologic areas nearby. In addition, it does not
resemble any of the beds in the underlying Earp Formation with
which this author is familiar. On Plate 3 of USGS Professional
Paper 748. Drewes shows an Earp outcrop area about 2.5 Kilometers
west-northwest of the Apache Ridge area. Unfortunately, this author
has not -observed this Earp outcrop or others northwest of that
site. Drewes' general description for the Earp Formation is
"Reddish and yellowish-brown sandstone, siltstone, and marlstone",
which does not match the observation on Apache Ridge.

PAGE

48