garnets
Geochemistry
ofzoned
lromtheSanPedro
mine,
SantaFeGounty,
NewMexico
by Charles
G.feeandW,W.Atkinson,
Jr., Department
of Geological
Sciences,
University
of Colorado,
Boulder,
C080309
Abstract
The San Pedro mine, in north-central New Metco, contains chalcopyrite-bearing garnetite skarn as
replacementsof 14 favorable carbonate
beds. Mineralization is concentrated
along or near marble line boundary
surfacesbetween garnet and marble in
each bed, such that orebodiesare arranged in a manto configuration,
cite-hematite-chlorite-chalcopyrite
alt e r a t i o n a s s e m b l a g ea n d s e c o n d generationgarnetveinletsand vug linrn8s.
Microprobe analvsisof some zoned
and unzoned San Pedro garnets provides an important insight into the nature of short- and long-term variation
in hydrothermal fluid chemistry.Early
main-stagefirst-generationgarnetshav'e
stacked
oneaboveanother.
andradite
coresandzonedrims.The
Early high-temperature contact
metamorphism resulted in the formation of iron-poor calc-silicates in mar-
rim zones alternate between purely andraditic and more grossularitic (up to
9 mole percent Al) compositions. This
ble and grossulariticgarnet (gr) in
interbedded orbicular hornfels. Subsequent infiltration of an iron, silica,
and metal-bearing magmatohydrothermal fluid causedthe metasomatic
replacementof marbleby massivegarnet-chalcopyriteskam. This main stage
of disseminated ore deposition and
skarn formation was in turn followed
by a low-temperature skarn-destructive phase involving the entry of meteoric water into the hydrothermal
system,which produced a quartz-cal-
patternof oscillatory
zoningis present
to a much greaterdegreein the crosscutting second-generationvein garnets (up to 457oAl). Unzoned garnet
presentin orbicularhornfelshas an extremelyhigh Al content,containingup
to 8L mole percentgrossular(8lVa Al).
Although variation of many factors
could be responsible,a likely causeof
a shift to more aluminous garnet compositionsis a decreasein CO, in hydrothermal fluid, caused by boiling.
Another possibility is a change from
o
c)
Songre de
Criso
f
Mountoins
ofintrusive
centers
thatmakeupthe25-milelong north-trending San Pedro-Ortiz porphyry belt. From north to south the centers
are known as La Cienega, Los Cerrillos, the
Ortiz Mountains,the SanPedroMountains,
and South Mountain. The Sandiauplift lies
to the southwest of the porphyry belt, and
the Sangre de Cristo Mountains lie to the
northeast.The belt lies about 12 mi from the
easternedge of the Rio Grande rift (Fig. 1).
Exceptfor the younger La Cienegaintrusion, the porphyry belt appears to be Oligocenein age based on K-Ar dates of 38 to
26 m.y. for the contemporaneousEspinaso
Vertebrate
faunaof U-BarCave.
HidalgoCounty
p.74
Photoessayon DogSprings
Memberof the Spears
Formation
p.78
Historical
reviewof uranium
production
fromthe
TodiltoLimestone
p. 80
Service/News
p. 85
Indexfor volume7
p. 87
Statfnotes
p. 88
Goming
soon
o.
o
!
o
a
General geology
The SanPedromine, which containsa small
copper-bearingskarn, is located in the San
PedroMountains of north-centralNew Mexico. The mountain range belongs to the set
Alsoin thisissue
z
J e r n e z u p l i ft
lithostatic to hydrostatic pressure accompanied by fracturing. Finally, an
influx of meteoricwater may also have
been involved.
aa Terliory monzonilic
FI9UR-Et-h*t
maq -oj north-central New Mexico showing location of the San Pedro Mountains, from Atkinson, 1976.
Petrography
and stratigraphyof
Sevill+Trident
exploration
wells
"lost"fusulinids
Needham's
LateHolocene
displacement
alongthe
OrganMountains
fault
Geotechnical
testingin NewMexico
I MILE
MONZONTTE
_
-
::l\[,'- i ) r,
:jjl) 1 r !,:
uoaeroLms.l
lENf
I-"-J-J-J-
i'i'i'l- 't I rl
;l;:;:{-) t./,;
'.'.',1r.-rr
?_
p€ BASEMENT
Gornelized Lms.
FIGURE2-Geologic cross section of the San Pedro Mountains, Santa Fe County, New Mexico, generalized from Atkinson, 1976'
volcanics, which erupted from vents in the
Cerrillos and Ortiz Mountains areas(Kautz
et al., 1981).The intrusive rocks of the belt
as a whole and of the San Pedro Mountains
in particular tend to be rather homogeneous
monzonitic porphyries. The source of the
mineralization at the San Pedro mine was
probably the San Lazarus stock, a composite
pluton made up of porphyritic monzonite
and quartz monzonite containing molybdenum-bearingquartz veinlets.
Sedimentaryrocksdip about 14oto the east
in the San Pedro mine area, where the mineralizationis hostedby the upper part of the
Pennsylvanian Madera Formation, a sequence of alternating shale and carbonate
beds. Massive and disseminatedmineralization occurs in replacementsof 14 chemically receptive Iimestone beds separatedby
barren shale. This favorable seriesis 270 ft
thick, and is sandwiched belween an underlying monzonite porphyry sill and an
overlying rhyolite sill.
Hydrothermal fluid expelled from the adiacentSan Lazarusstock was constrainedto
ilow between the pre-existing sills and
through the favorable series, which resulted
in the metasomatic replacement of limestone/marble beds by garnet-chalcopyrite
skatn (Fig. 2). Shale units were simultaneously converted to diopside-plagioclasequartz hornfels. Sharply defined marble line
boundary surfacesmark the position of the
metasomatic front and separate skarn from
marble within each limestone bed. Mineralization is concenhatedalong and near these
marble line boundaries, with ore grades decreasingrapidly from the skarn-marble contacts into the skarn, toward the source of
mineralization. Ore generally does not extend more than 200 ft inward from the marble lines and does not occur in marble. At
San Pedro, this style of mineralization has
produced a stack of skarn-hosted,marbleline orebodies, each of which has a manto
configuration.
San Pedro skarn
The character of a given skarn-type ore
depositis the result of the interplay of many
factors, such as the chemistry of the mineralizing hydrothermal fluid and its evolution through time, the establishment of
localized activity gradients, host rock composition and permeability, volatile compoNovember 1985 Neu MexicoGeology
sition, temperature,pressure,and the relative
importance of metimorphic and metasomatic processes(Einaudi et al., 1981).Most
skarnjappear to go through a similar set of
developmental stiges as a iesult of variation
through time in thi relative impact of these
factors. Metamorphic effects generally precede metasomatii ones, with late-stagealteration being overprinted onto them both,
which produces distinctive patterns of mineralogicalzonation and alteration(Roseand
Burt, 1979).
metamor. At San.Pedro,- early contact
phism of .impure limeston-eand/or dolomite
resulted in formation of light-colored calcsilicates such as tremolite, diopside, and
wollastonite.This was followed by two periodsof mineralization.Disseminatedorewas
deposited.first as.Fe, Sr, and metal-bearing
fluids infiltrated the carbonate beds, resulting in formation of massivegarnetite skarn
and earlychalcopyritedeposition.Late skarn
destruction and ore redistribution produced
vuggy ore. Here, garnet destructionand alteration associatedwith the entry of meteoric water into the hydrothermal fluid
produced a quartz-hematite-calcite-chloritechalcopyritemineral assemblage.The vugs
containing these minerals are lined almost
invariably with euhedral garnet. Crosscutting, second-generation garnet veinlets
also were produced during this stageof genesis. Finally, argillic alteration occurred,
forming talc, chlorite, montmorillonite, sericite, quartz, and limonite. Typically, temperafures for the formation of skarns have
been found to be 500-700'C for early contact
metasomatism, 400-600'C for garnet skarn,
and 200-400'C for the late hydrous alteration
phase (Rose and Burt, L979;Einaudi et al.,
1981).It is expectedthat temperaturesyet to
be determined on San Pedro rocks will fall
in these ranges. The resulting San Pedro deposit exhibits the following zonation from
the inside outward: gametite skam/vuggy ore
with argillic alteration; gametite skam/vuggy
ore; marble line with garnetite skarn and disseminatedore; marble i wollastonite-diopside-tremolite; and unaltered limestone/
dolomite.
Interaction between the infiltratin g hydrothermal fluid and the host limestone/marble
is responsible for skarn formation and minerclization. Therefore, it is clear that the thermochemical character of the hydrothermal
solution must have varied considerably over
time and distance to have produced such a
complex pattern. Some insight into the_nature of local and temporal variation in fluid
chemistry a!_S-an.Pedro-may be. obtained
through detailedmicroprobe analysisof some
zoned and unzoned garnets'
Garnet analyses
An the San Pedro garnet analyses were
performed at the Unive'rsity of Colorado with
I V4C-ZS elecfronmicropiobeusing a Kevex
energy dispersive systeh at an operating
voltije of iS tV a 2jmicron beam diameter]
u su.ttlpl"current of 200 microamps, and 100r".o'd count times. Standardi employed
were wollastonite for Ca and Si, an&rallite
garnet for Fe, and augite and garnet for Al.
torrections w"re made usinglhe Magic IV
Cr, M[, and
computer program (Colby, 196"8).
New AAexnc@
GEOLOGV
. Science
andService
Vofume
7, No. 4, November
1985
Edltolr Deborah A. Shaw
Published quarterly by
New Mexico Bureau of Mines and Mineral Resources
a division of New Mqico Institute of Mining & Tchnology
BOARD OF REGENTS
Ex Officio
Toney Anaya, Gownor of Nru Metico
Superintendeil
of public Insttuction
Morgan,
Alan
Appointed
SteveTorres, Pres.,7967-199L,Socorro
Judy Floyd, Sq.lTreas.,f977-1987,Ins Cruces
Gilbert L. Cano, 1985-191, Albuquerque
Lenton Malry, L985-7989,Albuquerque
Donald W. Mords, 1983-1989,LosAlamos
New Mexico Institute of Mining & Technology
LaurenceH.Lattman
......
Presidmt..
New Mexico Bureau of Mines & Mineral Resources
.......
FrankE.Kottlowski
D i / K t o r. .
......
GeorgeS.Austin
DeputyDircctor
Sub{riptions: Jssued quarterly, February, May, AsB\sL
price
November; subscription
$6.00/calendaryear.
Editorial mqttq : Contributions of material forconsideration
in future issuesof NMG are welcome.Articles submitted
for Dublicationshould be in the editor's hands a minimum of five (5) months before date of publication (February, May, August, or Novenber) and should be no
longer than 20 typewitten, double-spacedpages. Addressinquiries to Deborah A. Shaw, editor of Nru Merico Geology,New Mexico Bureau of Mines & Mineral
Resources,Socoqo, NM 87801
reproducible
@ithoutper'
Published
as publicdomain,therefore
mission.Sourcecreditfequested,
Cilculation:7,6N
Printer:Univetsily of New Mexico Printint Plant
Mn were not determined becauseof an interestin Fe-Al variation and the lack of measurableamountsof theseelements.Tracesof
Mn were found in San Pedro garnets.
Threesampleschosenfrom drill corewere
analyzed,at60points by microprobe.Of these,
sampleR-15 (673ft) containsskarn garnets.
Contact-metamorphic garnets present in
samples R-tS (575ft) and R-15 (773.5ft) occur in orbiculesin the interbeddedhornfels.
In all cases,San Pedro garnets are of the
andradite (ad)-grossular(gr) solid solution
series (CarFerSi.O,r-CarAlrsi.O,r).
Table 1
presentsrepresentativeanalysesof San Pedro
garnets.
In sampleR-15 (673ft) two generationsof
metasomaticgarnet are present. The firstgeneration garnet grains contain yellowbrown isotropiccoresbounded by zoned rims,
consistingof oscillatory zones of yellow, isotropic gamet, altemating with paler, weakly
anisotropiczones. Twenty-one points were
analyzed in a traverse across a first-generation gamet (Figs.3, 4A). The core (C) is nearly
pure andradite(Xad - 100mole percent;Xgr
: 0%), as are the yellow, isotropic zones in
the rim (R). The weakly anisotropic zones
contain up to 9 mole percent grossular.Although the precisenature of chemicalvariation within zones cannot be resolved with
the spacingof analysesused, the correlation
betweenanisotropicgarnetand the presence
of aluminum is clear.
The massivefirst-generationgarnet is cut
by grossular-richsecond-generationgarnet
veins (V; Fig. 4B). Locally, the second-generation garnet cements a breccia of the first.
The second-generation garnet is strongly anisotropic, showing sector twinning and weak
to strong oscillatory zoning. Birefringence
varies from 0.003 to 0.015, attaining first-order
red colors. In some veins, an isotropic band
that has sharp boundaries with the anisotropic garnet ii present. Other sharp changes
in birefringence are present locally, but in
general, changes are smooth but nonsystematic'
r!
F I R S T G E N E R A T I O N G A R N E T( R - ' 1 5 , 6 7 3 , )
C O M P O S I T I O N A LV A R I A T I O N ,R I M T O C O R E
RIM
CORE
rsorRoPtc
TABLE1-Representative gamet analysesin weight
percent. A, first-generation, core andradite, sample R-15 (673 ft); B, first-generation, rim garnet,
sampleR-15 (673 tt); C, second-generation,
vein
garnet, sample R-15 (673ft); D, garnet in hornfels
orbicule, sample R-15 (676 It); * the low total may
be accountedfor bv the presenceof H,O in hvdrogrossular,as indicated'by loss on igniiion; Xgr,
mole percent grossular
Si
AI
Fe
Ca
Xgt
9^
L#
o0)
o
9
(Lc)
>o
9a
"E
F
UJH
E5
oxide
sio,
Alro3
Fe2O3
CaO
Total
z
36.26
0.00
29 75
34 32
100.33
36.47
0.97
28.72
34.28
99.72
37.38
5 11
22.20
36.04
700 73
37.75
74.02
597
34.87
91.95*
No. of ions per 24 orygens
6.097
6.115
6.083
0.000
0.180
0.980
3.761
3.554
2.719
6.777
6.769
6.285
0.000
0.048
0.265
6.229
2.772
0.746
6.265
0-788
o
0.o
o.25
0 . 75
0.50
DISTANCE FROM RlM, in millimeters
FIGURE 3-Microprobe analysesof first-generation garnet, zoning profile, and mole percent Brossular (Xgr) plotted, versus distance from rim edge, in mm Boundaries between isotropic and
anisotropic optical zones indicated by vertical dashed lines.
FIGURE4-Photomicrographs of San Pedro garnets. Sample R-15 (673 ft); crossed polarizers; length of each white scalebar equals 1 mm. A, Firstgenerationgarnetite, showing isotropic andradite cores(C) and zoned rims (R); B, Second-generationgarnet vein (V) cutting first-generationgarnetite.
Note the highly birefringent nature of the vein material The location of probe traversesare shown by heavy solid lines.
New Mexico Geology November
1985
A traverseacrossone vein garnet (Fig. 5)
was analyzed at 18 points, revealing oscillatory compositionalzoning of a much greater
amplitude and higher frequency than that of
the first-generation garnet. Zones at the center of the vein are very andradite-rich (Xgr
: 2.5 mole percent),but these give way toward the vein wall to hishlv aluminous compositions as the Al co-ntdntof the garnet
quickly attains its maximum (Xgr : 45 mole
percent), then falls off erratically with the
amplitude of its variation being as much as
25 mole percent, eventually coming down to
near the vein
much lower values (21,-22Ea)
walls. The mean grossular content for vein
garnetsis 25.7Vo,as opposed to a value of
2.0% for the first-generation garnet (seeFig.
6).
Harris and Einaudi (1982) have documented a style of garnet compositionalzoning for calcic, Fe-rich skarn at Yerington,
Nevada, which is similar to that found at San
Pedro. These authors note the presenceof
grossularitic veinlets cutting early garnet
having isotropic andraditic cores and aluminous and compositionallyvariable rims.
This same pattem is shown clearlv bv the
analysisof iirst- and second-generaiiongarnets in sample R-15 (673 ft; seeFigs. 3 and
s).
Sweeney(1980)has studied skarn garnet
in the Bingham district, Utah, and has found
it to be very similar in nature to the firststage garnet at San Pedro. Bingham garnet
is mostly nearly pure andradite, with a few
thin aluminous zones present in each crvstal. The zoning was foind to exist on u ,r6ty
fine scale, with many thin zones only a micron or two wide. Anisotropic garnet also
correlated with Al content, although the
maximum grossularcomponent was no more
than 30Vo.
Alteration of massive vein wall garnet to
a mixture of chlorite, hematite, calcite, and
quaftz occurs in places along the vein itself
and in patchesscatteredthroughout the skam.
Elsewhere, fresh first-generation andradite
dominates. Remnant garnet in the altered
patcheshas an Al content that is only slightly
lower than the AI content of the vein garnet
itself (seeFig. 6; maximum : 25.3 mole percent,mean = 19.9mole percent),which suggests that the composition has been altered
by diffusional exchange.
The garnets found in orbicules in hornfels
were probed in two samples:R-15 (773.5ft)
and R-15 (675 tt). The results of the analysis
show that these unzoned garnets are distinctive by virtue of their highly aluminous
compositions. The mean grossular contents
for samplesR-15 (773.5ft; 3 analyses)and
676 ft; 7 analyses) are 53.7 and 81..3mole
percent, respectively.
Interpretation and conclusions
The compositions of the garnets found in
orbiculesin hornfels are the most aluminous
found at San Pedro. The unzoned character
of these garnets and their very high Al contents suggest that the hornfels environment
of formation was relativelv unaffectedbv the
November 1985 NerrlMexico Geology
V E I N ,A L T E R E DV E I N W A L L & U N A L T E R E DG A R N E T S
C O M P O S I T I O N AVLA R I A T I O N( R - 15 , 6 7 3 ' )
60
E
U
F
2
U
I
I
VEIN
.9
=
u
50
I o
o
z
I
I
A L T E R E DV E I N W A L L
9^
o
CD
!Eao
6
ao
og
Io)
>o.
O= eo
I
F
IIJ
2
E
o:E
Io
o
I UJ
IE
I u,
F
III L
TPzo
a-
a
I
F
J
I
z
l
10
o
1.5
1.O
0.5
D I S T A N C EF R O M V E I N C E N T E R ,i n m i l l i m e t e r s
FIGURE S-Microprobe analysesof second-generationvein, first-generation altered vein wall, and
first-generation unaltered gatnet zoning profiles. Mole percent grossular (Xgr) ig plotted versus
distance from vein center. Boundaries between zones are indicated by vertical dashed lines.
LARIATION
G A R N E TC O M P O S I T I O N A V
SKARN GARNETS
(ALL FROM B-15,673')
HORNFELS GARNETS
(ORBICULAR)
9
zF
6^to
o<
Fc
gg
*g'
8E
F.-
6
H8
o4
2
9
j32
;H :
z
F
zo
G
;2
;F
O
u
z
U
o
d;
9>
;
F
IIJ:
o
v
L
ci
U
G
ul
f
18
U
i:
i=
b
N
o
ct
E
926
5T
J
2
3
s'
7
I
6l
o
c
I
I
T
h
o
F
F
:
T?
?t
rl
E
3
tr
I
t_
o
G
E
FIGURE 6-Microprobe analyses, gamet comPositional range, mole percent grossular (Xgr), and
skarn and hornfels varieties. Bars indicate range and diamonds represent mean values for each
garnet type. Numerals placed above upper bar show the number of analysesperformed on each
variety.
infilhating hydrothermal fluid responsiblefor
the deposition of the skarn garnets. The orbicules must have had a Iocally derived composition and a metamorphic rather than
metasomaticorigin.
Tracy (1982) defined growth zoning as
zoning developed in a mineral becauseof a
continuous or discontinuous change in the
composition of material supplied to the
growing surfaceof the crystal. Negligible rates
of solid-statediffusion are also required to
preserve such zoning. Skarn formation results from interaction between an infiltrating
hydrothermal fluid and carbonatehost rocks
over a wide range of temperatures (200700"C).Garnet zoning shows that the composition of that solution varies over time and
with distance from its source. Details of this
variation are provided by analysis of the
growth-zoned garnets present in the San
Pedro skarn.
First-generation garnets have andradite
coresand zoned rims. The rim zones alternate abruptly between purely andraditic and
more grossularitic(up to 9 mole percentAl)
compositions.Therefore,compositionaldifferencesas expressedby the oscillatoryzoning observed in the first-generation garnet
rims reflect short-term variation in the composition of the hydrothermal fluid.
Without more information, it is not possible to suggest exactly what controls the
composition of the solution depositing the
gamet. It is likely that it was largely a hypersalinefluid exsolvedby an igneoussource,
probably buffered at its source by igneous
minerals in deeper rocks. After it left the
source, it may have mixed with unknown
amounts of meteoric or connate water and
traversedunderlying sedimentaryrocks and
Precambrianbasement rocks affected to unknown degreesby contactmetamorphismand
the early portions of the mineralizing fluid.
The chemishy of such solutionsis complex.
The stability of andradite vs. grossular is affected not only by such obvious factors as
the Fe/Al ratid, birt by activities of species
that would affectthe Fe'*/Fe't ratio, such as
for, fsr, pH, and others.
Sweeney (1980)has proposed a mechanism for producing zoned ad-gr garnets
based on T-)Go, relations. In Figure 7, the
lowermost T-X;, stability curv6 for grossular lies at high6r temperatures than"that
for andradite, requiring higher temperatures, and/or lower X.o. to stabilize grbssular. Calculatedcurves frir intermediai=e
solid
solutions lie between the two curves. An earIier study of fluid inclusions at Bingham by
Moore and Nash (1974)showed that hvdrothermal solutions underwent a change in
pressure from lithostatic to hydrostatic during the period of ore deposition. Sweeney
showed that such an event could change
conditions from those depositing pure andradite to those depositing grossular-rich
garngt. She also examined the lowering of
X.o, by boiling as a mechanismfor the shift
to rirore grossular-richcompositions.For high
valuesof X.o, boiling would have little effect.
However, Eihaudi et al. (1981)report that,
- Greo
Grgo--------=
An+
Grroot Qtz
{"d - -'
Adl
c)
o
F
/,
Tr+3Cc+ZQtz
P f t r i d= P s o l i d= I , O O O b o r s
Q,2
0,6
Q,4
0,8
r,O
Xcoz
FIGURE 7-T-Xco, diagram. Dashed lines represent the position of the garnet univariant boundaries
as a function of variation in aluminum content, where Gr,6 is pure Ca-Al garnet (grossular), Ad is
pure Ca-Fe garnet (andradite), and Grzo,Gro, Grr, Grro, Grao,and Grroate andradite-grossular solid
solutions. The arrow indicates the rapid shift toward grossularitic compositions expectedto result from
isothermal boiling of the hydrothermal fluid. Adapted from Sweeney, 1980.
in most skarns, X.o" values vary from initial
values of 0.2 to Iattir values of 0.05. At low
CO, concentrations,boitng would reduceCO,
quickly, and solutions would shift rapidly
into the field of grossularitic garnet (seeFig.
Z). Studiesof fluid inclusions in the Bingham
dishict (Moore and Nash, 1974;Starkins,1983)
indicateextensiveboiling at temperaturesless
than 500'C, under which boiling would easily reduceCO, content to stabilizegrossular.
It must always be borne in mind, of course,
that with many variablesinvolved, other factors may cause the changes.For example,
pulses of higher temperature fluids could be
responsible.Meteoric water might also enter
with a low CO, content.
At San Pedro, a period of fracturing and
local brecciationpreceded the change from
first- to second-stagegamet. A rapid survey
of fluid inclusions in thin sections indicated
that vapor-rich and coexisting liquid-rich inclusions are indeed present in garnet, which
suggeststhat solutions did boil, although they
are so sparsethat they cannot yet be related
to specificevents. Releaseof pressure and
consequentboiling accompanying fracturing
may therefore have been important.
In general,during the late destructivestage
of skarn evolution, temperatures are greatly
reduced, pH decreases, release of hydrothermal fluid from the magmatic source may
be restricted, and dilution of the fluid by the
entry of meteoric water is expected (Einaudi
et al., 1981).The result is garnet-destructive
alteration and sulfide redistribution. Such
events are apparent at San Pedro. We expect
that dilution of the hydrothermal fluid by
meteoric water causesan extreme decrease
in the activity of the chloride ion (ClJ. The
reduced availability of this important complexing agent could greatly lessenthe ability
of the hydrothermalfluid to carry ferric iron,
without affectingAl solubility (Barnes,1979).
Garnet nucleated from such a fluid would be
grossularitic rather than andraditic. The formation of second-generationgamet veins and
vuggy garnets (associatedwith vuggy ore)
may have occurred in this manner at San
vein garnet is
Pedro.The second-generation
very aluminous (up to 45 mole percent grossular),showsoscillatoryzoning, and cuts firstgeneration garnetite. Here again, the oscillatory nature of the zoning may have been
due to the episodic releaseof fluid from the
cooling magmatic source.
In places, metasomatic exchangebetween
this fluid and previously deposited vein-wall
garnet may have altered the original (firstgeneration) garnets to the extent that they
approach the compositions of the vein garnets themselves.Thesealtered vein-wall garnets show elevated Al contents and contain
patches of fine-grained chlorite-calcite-specular hematite, an assemblagetypical of the
garnet-destructive stage of skarn formation.
The lack of pervasive alteration of first-generation garnet by fluids of the second generationmay indicatethat meteoricwater influx
was restricted. Alternativelv. the alteration
process may have been aborted by loss of
fluid pressure due to fracturing as shown by
the cross-cuttingalterationalong gamet veins.
The structural setting of the San Pedro deposit between two thick sills may have enhancedearly Fe-rich skam formation but also
t3
New Mexico Geolo3y November 1985
may have limited the circulation of meteoric
water during the skarn-destructivestage.
ACKNowLEDGrvrnNrs-This
paper is derived from a doctoral thesis in preparation
at the University of Colorado (Boulder).Financial support for field work and analysis
was generouslyprovided by the New Mexico
Bureau of Mines and Mineral Resources.
Funding to cover the cost of thin section
manufactureand accessto the SanPedromine
property was kindly given by the Goldfield
Corporation.This paper has benefited from
critical reviews by J. Renault, f . Grambling,
and J. L. Mufloz. Specialthanks go to Linda
A. Kroff for her continuoussupport throughout this project. We thank the Cooperative
Institute for Researchin EnvironmentalSciencesfor the use of its word-processingservrces.
References
Atkinson, W. W., Jr., 1976, Zoning and paragenesisof
oresin the SanPedroMountains;ln Woodward,L. A.,
and Northrop, S. A. (eds.),Tectonicsand mineral resourcesof southwestern North America: New Mexico
GeologicalSociety,SpecialPublication 6, pp.187 797.
Barnes,H. L., 1979,Solubilitiesof ore minerals; in Barnes,
H. L. (ed.), Geochemistryof hydrothermalore deposits: John Wiley & Sons, New York, pp. 401,460.
Colby, J. W., '1968,Quantitative microprobe analysis of
thin insulating films; la Advances in x-ray analysis:
Plenum Press,New York, v. 11, pp. 287-305.
Einaudi,M. T., Meinert, L. D., and Newberry, R. J., 1981,
Skarn deposits:EconomicCeology, 75th Anniversary
Volume, pp.317-391,.
Hanis, N. B., and Einaudi, M. T., 1982,Skarn deposits
i n t h e Y e r i n g t o nd i s t r i c t ,N e v a d a - m e t a s o m a t isck a r n
evolution near Ludwig: EconomicGeology,v.77, pp.
817-899.
Kautz, P. F., Ingersoll,R. V, Baldridge,W S., Damon,
P E., and Shafiqullah,M., 1981,Geologyof the EspinasoFormation, Oligocene,north-central New Mexico:
GeologicalSocietyofAmerica, Bulleiin, v. 92, pp.23182400.
Moore, W J., and Nash, J. T., 1974,Alteration and fluid
inclusion studies of the porphyry copper ore body at
Bingham,Utah: EconomicGeology,v.69, pp. 631.-645.
Rose,A. W., and Burt, D. M., 1979,Hvdrothermalalteration; in Barnes,H. L. (ed.), Geochemistryof hydrothermal ore deposits: John Wiley & Sons, New York,
2nd edition, pp. 773-235.
Starkins, R. D., 1983, Fluid inclusion study of the Carr
Fork skarndeposit,Binghammining district,Utah: Unpl.tblishedM.S. thesis, University of Colorado, Boulder,
Colorado, 134pp.
Sweeney,M. 1., 1980,Geochemistry of garnets from the
North Ore shoot, Bingham district, Utah: Unpublished
M.S. thesis, University of Utah, Salt Lake City, Utah,
1s4pp.
Trary, R. J., 1982, Compositional zoning and inclusions
in metamorphic minerals: Mineralogical Society of
America Review, v. 10, pp. 355-394.
launa
report
0nthevertebrate
Preliminary
of U-BarCave,
NewMexico
HidalgoCounty,
(ElPaso),
TX79968
ElPaso,
of Texas
Paleobiology,
University
of Vertebrate
byArthurH. Harris,
Curator
Introduction
Current excavationsin southwesternNew
Mexico are producing a major late Pleistocene vertebrate assemblagethat is adding
greatly to our knowledge of the biota and
past ecology of the state. Since December
1983, evidence of more than 90 vertebrate
taxa has been produced in U-Bar Cave. The
complexity and extent of the cave deposits
and faunas will require an extensiveperiod
of study; however, the importanceof the early
findings to understandingthe Pleistocenebiology of New Mexico is such that these preIiminary data should be made availablenow.
U-Bar Cave (site LA 5689)is in the U-Bar
Limestone(Cretaceous)about 6 mi from the
Mexicanborder in the "boot heel" of southwestern New Mexico. The altitude is about
5150ft. The cave is surrounded by Chihuahuan desertscrubmixed with Upper Sonoran woodland elements.The main chamber
of the cave (Fig. 1) is about 315 ft long and
avefagesabout 50 ft wide.
The cavehas produced archaeologicalmaterial sinceabout 1935(Lambertand Ambler,
1961),and it was the site of formal excavations by Lambert and Ambler in 1960.They
consideredthe archaeologicalmaterialasmost
probably assignableto the Animas Phaseof
the CasasGrandesculture, with a likelv date
315
entronce
3r2
297
294
291
u
288
s t€st
prt
M i n e rs p i l
2A
?42
,
u:,'
*
276
271
LO SI
d e b r is
piles r
210
.I
264
I
D7
South
debns
pile
243
37
234
O
5
lOm
|.++f/------1J
Enfronce
225
trLULIUN
Pincushion
cactus
November 1985 New MexicoGeology
O
15 3ofl
JKLMNOPQRST
FIGURE 1-Sketch map of U-Bar Cave showing Srid system, features,and regions
ieferred to in the text.