Relationship between copper tonnage of Chilean base-metal

porphyry deposits and Os isotope ratios

Ryan Mathur Department of Geosciences, University of Arizona, Tucson, Arizona 85721, USA
Joaquin Ruiz
Francisco Munizaga Departamento de Geologia, Universidad de Chile, Casilla 13518-C.21, Santiago, Chile

ABSTRACT three distinct intervals of magmatic activity: 60,

Re-Os isotopes in pyrite and chalcopyrite from early high-temperature hypogene altera- 40–30, and 10– 4 Ma (Sillitoe, 1988). These
tion assemblages in Chilean porphyry copper deposits identify the source of Os and, by infer- deposits have been reviewed in detail (Zentilli,
ence, Cu in these ore systems. Typical concentrations for Os in both pyrite and chalcopyrite are 1974; Gustafson and Hunt, 1975; Alpers and
between 7 and 30 ppt (10 –12 g/g), and for Re between 0.200 and 10 ppb (10 –9 g/g). Re-Os Brimhall, 1988; Sillitoe, 1988, 1990; Skewes and
isochrons yield ages and initial 187Os/ 188Os ratios for the sulfides. The isotopic data reflect the Stern, 1994; Williams, 1994).
relative contributions of copper from the mantle and crust in Chilean porphyry copper The plutonic rocks associated with copper
deposits. Seven ore deposits that reside in different tectonic terranes and represent distinct mineralization range in composition from dio-
epochs of mineralization in Chile were studied. The initial osmium ratios of the first stage of rite to quartz monzonite. The deposits are char-
mineralization at each of the deposits range from 0.15 to 5. These values are more radiogenic acterized by a core of hypogene alteration
than the present chondritic mantle (~0.13), and indicate significant crustal contributions of Os (which includes the assemblages of biotite-
to the magmatic and/or hydrothermal systems. There is a strong correspondence between the orthoclase and quartz-sericite) and mineraliza-
total copper content and initial Os isotopic ratios in base-metal porphyry deposits. The larger tion with chalcopyrite, bornite, and sometimes
deposits have lower initial Os ratios than the smaller, less significant deposits. This relationship molybdenite, as the minerals that host the eco-
implies that larger deposits acquire a greater proportion of Os from the mantle. The initial Os nomic elements of interest. It is generally
ratio of samples in the central segment of porphyry copper deposits of northern Chile also thought that the early sulfides in these deposits
decreases with decreasing age of the deposit. A plausible interpretation of the Re-Os data is that are precipitated from dominantly magmatic
the later and larger deposits use regional tectonic and structural features that allow sampling fluids accompanying the intrusion. Later sulfides
of deeper more primitive magmatic sources. precipitate from meteoric waters that invade the
system as it cools (e.g., Beane and Titley, 1981;
Keywords: porphyry copper, Re-Os, Chile. Guilbert and Park, 1986; Beane and Bodnar,
1995). Low-temperature supergene processes
INTRODUCTION levels of the crust (e.g., Sawkins, 1990). Subse- that are not related to the primary magmatic-
Because of its chalcophile and siderophile quent development of large hydrothermal sys- hydrothermal system redistribute the economi-
nature, the Re-Os isotope system has long been tems (Titley and Hendrick, 1978; Lang and Titley, cally important minerals, making these deposits
recognized as a possible geochemical tool for 1998) around the intrusions deposited Cu-bearing economically viable, but presumably do not add
tracing the origin of metals and directly dating minerals in stockwork fractures. There are many more ore-forming elements to the system.
mineralization (Luck and Allegre, 1980; theories on the reason why these deposits form The ultimate source of the base metals (i.e.,
McCandless and Ruiz, 1993; Frei et al., 1998; (Clark, 1993; Williams, 1994). The possible con- copper and molybdenum) could either be mag-
McInnes et al., 1999). This is in contrast to most trols for mineralization include regional struc- matic with crustal or subcrustal components, or
other radiogenic isotopic systems, which typi- tures, a specific chemical and/or physical evolu- hydrothermal through leaching of the surrounding
cally consist of lithophile elements that concen- tion of the magmatic and hydrothermal system, host rocks (Sillitoe, 1988). Previous studies in this
trate in associated alteration silicates rather than and the sources and availability of sulfur and/or part of the Andean Cordillera yielded conflicting
sulfides. Furthermore, because Re is highly in- the ore-forming metals. Some studies have sug- results on this issue. Initial 87Sr/ 86Sr isotope ratios
compatible in the mantle compared to Os, the gested that the size of base-metal porphyry and rare earth element and trace element data
Re/Os ratio of the crust is highly enriched com- deposits may be related to mantle processes, and from the intrusive rocks indicate that the magmas
pared to that in the mantle (Drake and Jones, that the larger deposits should have a significant associated with the mineralization were derived
1983; Fleet and Stone, 1991; Fleet et al., 1999). mantle signature (Clark, 1993; McInnes et al., from the mantle, and interacted very little with
The time-integrated consequence of this ele- 1999). This study tests this hypothesis and cen- crustal rocks (Gustafson and Hunt, 1975; Shibata
mental behavior is that the crust evolves toward ters on using Os isotopes to decipher the source et al., 1984; Munizaga et al., 1985; Skewes and
elevated 187Os/ 188Os ratios in comparison to the of ore-forming metals. Stern, 1996). Pb isotopes, however, indicate a
chondritic mantle, which is ~0.13 today (Morgan more complex source for Pb that also includes
et al., 1995; Shen et al., 1996). GEOLOGIC BACKGROUND the crust (Puig, 1988; MacFarlane et al., 1990;
In this study, Os isotope ratios are used to Chile is the leading copper producer in the Tosdal, 1995).
trace the relative importance of the crust and world and has some of the largest base-metal por-
mantle as sources of Os in the porphyry copper phyry deposits known. The ore deposits formed ANALYTICAL TECHNIQUES
deposits of Chile. Due to the chalcophile nature in large hydrothermal systems during Cenozoic The sulfide samples analyzed for this study are
of this element, the sources of Os may be used as continental arc magmatism (Ramos, 1989). The associated with the early high-temperature altera-
proxy for the sources of Cu. ore-bearing porphyries intrude a collage of ig- tion silicates of either orthoclase-biotite or quartz-
Porphyry copper deposits, which are major neous, sedimentary, and metamorphic rocks that sericite assemblages (Table 1). The samples were
geochemical anomalies of Earth’s crust, are char- were accreted to the margin of South America wrapped in paper and crushed with a hammer in
acterized by large concentrations of chalcophile from the Late Proterozoic through the Cretaceous an effort to minimize contamination with plati-
elements. Their occurrence is linked to the intru- (Ramos, 1989; Richards, 1995). The suite of por- num group elements. After selection of pure sul-
sion of subduction-related magmas at shallow phyry copper deposits chosen for this study spans fide grains from the crushed rubble, the samples

Geology; June 2000; v. 28; no. 6; p. 555–558; 3 figures; 1 table. 555

 the sulfides from base-metal porphyry deposits
(Table 1) often yield isochrons from single hand
samples. Presumably, small platinum group
element–rich inclusions control the distribution
of Re-Os in sulfides, and record the Os isotopic
characteristics at the hand-sample scale.
Seemingly, Os is derived from the associated
intrusion in each of the deposits because early
fluids are thought to be principally magmatic.
The source of the Os may change through time as
the hydrothermal system evolves from one domi-
nated by magmatic to meteoric waters. For ex-
ample, Ruiz et al. (1997) and Freydier et al.
(1997) showed that the expected changes in the
source of the Os for El Teniente could be docu-
mented with well-constrained samples.
The initial Os ratios for the deposits studied
here were obtained from isochrons when pos-
sible. El Salvador (Fig. 1A) yields an isochron
with the expected age (Gustafson and Hunt,
1975) and initial Os ratio. Samples associated
with orthoclase-biotite assemblages and samples
associated with quartz-sericite alteration assem-
blages are on a similar isochron. The Re/Os ratios
of the samples associated with quartz-sericite are
more radiogenic than those of the samples asso-
were sieved and handpicked until enough clean cause the greatest source of error in these low- ciated with orthoclase-biotite, suggesting Re mo-
sulfides were obtained. Typically, 0.8–1.5 g of concentration analyses is the variable contribu- bility early in the history of the deposit. The Re
sample was dissolved in a Carius tube in 4:1 tion of the Os blank, errors for both the mobility, however, must have occurred within the
HNO3-HCl acid mixture. H2O2 was added to the 187Re/ 188Os and 187Os/ 188Os are calculated by error of the age of the isochron. Chuquicamata
acid mixture to ensure complete oxidation of the varying the concentration of the Os blank yields an isochron (Fig. 1B) from paragenetically
samples and spike equilibration (Shirey and (187Os/ 188Os for the blank = 0.18) between 1.5 late sulfides that correlates with the youngest Ar
Walker, 1995). Os was separated and purified in and 2.6 pg for Os. The calculated errors are ages reported (ca. 31 Ma, Reynolds et al., 1998).
a two-stage distillation process similar to that listed in Table 1 as percent deviation on the Two samples from a paragenetically earlier min-
described by Frei et al. (1998). Rhenium was mean. Age and initial 187Os/ 188Os ratios were eralizing event (35 Ma) yield an initial ratio that
extracted and purified through column chem- calculated with the program MACDAT-2, writ- is less radiogenic than the late mineralization.
istry. The isotopic measurements were made by ten by Clark Isachsen and Drew Coleman using Cerro Colorado samples form an error chron with
negative thermal ionization mass spectrometry the algorithms of York (1969). a reasonable age, but high error on the age and
(Creaser et al., 1991). initial isotopic ratio. The initial 187Os/ 188Os ratios
Internal reproducibility for the measure- Re-Os RESULTS that are reported in Table 1 include a conservative
ments is always better than 0.2% (2 σ); how- Typically the Os concentrations for pyrite and view of their error.
ever, this value underestimates the error of the chalcopyrite range between 7 and 30 ppt, except
analysis because of variable contribution of Os for the samples from Quebrada Blanca, which DISCUSSION
from the procedural blanks. Whole procedure contain 230 ppt. The Re concentrations for all the A central question in metallogenesis is what
chemical blanks during this study varied from samples analyzed range from 150 ppt to 20 ppb controls the size of ore deposits. This question
1.5 to 2.6 pg for Os and 25 to 35 pg for Re. Be- (Table 1). Large variations in the Re/Os ratios of has been difficult to address because the variables

20 15

Figure 1. Isochron plots of

Top three crosses
16
(A) El Salvador pyrite and pyrite from quartz-
Os/188Os

chalcopyrite isochron with sericite alteration zone 10

age of 39 ± 2 Ma, MSWD
12
0.797, and (B) Chuquica-
mata pyrite from quartz- 31 Ma quartz-sericit e
sericite alteration isochron Initial 1.01 ± 0.014
187

with age of 31 ± 2 Ma, 8

.

MSWD 1.01; two samples Bottom three ellipses 5

of chalcopyrite from ortho- are pyrite-chalcopyrite
clase-biotite alteration sug- 4
gest age of about 33 Ma. from orthoclase-biotite 35 Ma ort hoclase-biotite
MSWD is mean square
weighted deviate.
Initial 0.78 ± 0.29. alteration zone A Initial 0.15 ± 0.009 B
5,000 10,000 15,000 20,000 5,000 10,000 15,000
187
187
Re/ 188Os Re/ 188Os

556 GEOLOGY, June 2000

 8.E+ 07
Chuquicamata
7.E+ 07 Figure 2. Data from El
Teniente (Ruiz et al., 1997);
Estimated Tons of Cu

6.E+ 07 Y-axis scale is variable

5.E+ 07 depending on economy
(Long, 1995) and techno-
4.E+ 07 El Teniente logical advances; never-
theless, relative size com-
3.E+ 07
Collahuasi parisons between deposits
2.E+ 07 El Salvador remain constant. Agua
Agua Rica
Rica from Argentina is also
1.E+ 07 Quebrada Blanca
included in plot.
Cerro Colorado
0.E+ 00
0 1 2 3 4 5 6
Initial 187Os/ 188Os Ratio

that may control total tonnage of the ore metals deposits in northern Chile—i.e., Cerro Colo- There is a relationship between age and initial
are many and intertwined. Here we use Os iso- rado—are in the Arequipa terrane, which has a ratio of the earliest mineralization for the mid-
topes to constrain one of the possible controls— Proterozoic basement, whereas El Teniente is on Tertiary base-metal porphyry deposits. The old-
the source of the ore-forming metals. Possible the Chilenia terrane, which is floored by Ordo- est Re-Os age for this group of deposits is from
sources for the Os, and by inference Cu, are the vician metamorphic rocks. Tosdal (1995) sug- Quebrada Blanca, which yields an age of 45 ± 3
mantle wedge, the subducting slab and accompa- gested that differences in metal budgets for ore Ma. This age is slightly older than the 38 Ma
nying sediments, and the diverse reservoirs in the deposits between lat 26° and 28°S are linked to K-Ar age reported by Hunt et al. (1983). The ini-
upper and lower continental crust (e.g., Esser and the influence of crustal assimilation. Magmas tial Os ratio of this deposit is 1.28 ± 0.06, which
Turekian, 1993; Brandon et al., 1996; Chesley that contain large amounts of crustal components is the most radiogenic value for the mid-Tertiary
and Ruiz, 1998; Saal et al. 1998). may be diluted of base metals as the magmas base-metal porphyry deposits of central-northern
Figure 2 shows a correlation between the ini- slowly ascend to the surface and extensively Chile. The youngest deposit in this area of Chile
tial 187Os/ 188Os ratio of base-metal porphyry interact with the crust. This could be a possible is Chuquicamata, which is 35 Ma (Reynolds
deposits of Chile and their copper content. The geochemical explanation for the observed re- et al., 1998). Chuquicamata has an Os initial
copper tonnage for the deposits was obtained gional trend of copper versus Os initial ratios. ratio of 0.15. Figure 3 illustrates the trend and
from Long (1995). The analytical errors of the Initial Os ratios within ore deposits also illus- shows a strong correlation between age of miner-
Os isotopic values have already been addressed trate the isotopically complex nature of the alization and Os isotopic ratio. The trend toward
and we believe that they are conservative. The fluids that are involved in mineralization. The more primitive initial Os ratios with time sug-
error on the tonnage is difficult to evaluate. isochrons in Figure 1 from Chuquicamata and El gests that regional structural processes may have
Although the calculation for absolute copper Salvador show how the sources of metals asso- allowed later mantle-derived magmas to rise
content is based on economic assumptions and ciated with different hypogene alteration events without as much interaction with the crust in
may change, it is unlikely that the values would within an ore deposit can remain relatively simi- comparison to the earlier magmas.
change the order of the deposits. The correla- lar or change drastically. For example, in Chuqui-
tion shows that the largest deposits—Chuquica- camata, pyrite associated with quartz-sericite SUMMARY
mata and El Teniente—have the lowest initial from the later (31 Ma) alteration events has ini- Re-Os isotope measurements of pyrite and
187Os/ 188Os ratios. These low Os ratios (~0.15) tial 187Os/ 188Os of ~1, whereas chalcopyrite chalcopyrite samples from porphyry deposits of
are nevertheless more radiogenic than those ex- from an earlier system (35 Ma) associated with
pected for the mantle beneath continental arcs biotite-orthoclase alteration has an Os initial of
(0.13–0.15; Brandon et al., 1996), and indicate 0.15. Thus, the second mineralization event has 1.5
Os/188Os

that the crust—probably the lower crust—(e.g., a more significant crustal signature than the
Quebrada Blanca
Chesley and Ruiz, 1998) contributes Os to the earlier event. El Salvador presents a different
1
magmatic processes that produce base-metal por- case, in which the orthoclase-biotite and quartz-
187

phyry deposits. The smaller deposits have sericite alteration was produced by one event. El Salvador
Initial

187Os/ 188Os ratios that are very radiogenic (~5). Pyrite and chalcopyrite associated with both 0.5 Collahuasi
These highly radiogenic values require a signifi- orthoclase-biotite and quartz-sericite alteration
Chuquicamata
cant contribution of crustal Os to the magmatic are on the same isochron. In this case, the sources 0
and hydrothermal system. of metals for both alteration and mineralization 35 40 45
The Os isotope data plotted in Figure 2 do not events are approximately similar. These two Age of first mineralization event (Ma)
correlate with the age of the basement that under- deposits demonstrate the need to thoroughly
Figure 3.Timing of mineralization in porphyries
lies the deposits. This indicates that the size of the understand the age constraints of the mineraliza- of central Chile against initial Os. Note that
deposits and the source of the Os are independent tion and events within an ore deposit in order to initial Os isotope ratio of only first mineraliza-
of the age of the basement. For example, the interpret the isotope results. tion is plotted.

GEOLOGY, June 2000 557

Chile show a correlation between total copper Esser, B.K., and Turekian, K.K., 1993, The osmium Reynolds, P., Ravenhurst, C., Zentilli, M., and Lindsay,
tonnage and initial 187Os/ 188Os ratios. All the Os isotopic composition of the continental crust: D., 1998, High precision 40Ar/ 39Ar dating of two
Geochimica et Cosmochimica Acta, v. 57, consecutive hydrothermal events in the Chuquica-
isotope ratios are more radiogenic than those p. 3093–3104. mata porphyry copper system, Chile: Chemical
expected for purely mantle-derived magmas. Fleet, M.E., and Stone, W.E., 1991, Partitioning of Geology, v. 148, p. 45–60.
However, the larger deposits have the least radio- platinum-group elements in the Fe-Ni-S system Richards, D.R., 1995, Terranes and tectonic evolution of
genic Os initial ratios, indicating the least amount and their fractionation in nature: Geochimica et the Andes: A regional synthesis [Ph.D. thesis]:
Cosmochimica Acta, v. 55, p. 245–253. Tucson, University of Arizona, 451 p.
of crustal Os.
Fleet, M.E., Menghua, L., and Crocket, J.H., 1999, Par- Ruiz, J., Freydier, C., McCandless, T., Chesley, J., and
The good correspondence between size of the titioning of trace amounts of highly siderophile Munizaga, F., 1997, Re-Os-isotope systematics
deposit and Os isotope systematics has impor- elements in the Fe-Ni-S system and their fraction- from base metal porphyry and manto-type miner-
tant academic and economic consequences. The ation in nature: Geochimica et Cosmochimica alization in Chile: International Geology Review,
data show that the most significant source of the Acta, v. 63, p. 2611–2623. v. 39, p. 317–324.
Frei, R., Nägler, T., Schönberg, R., and Kramers, J., Saal, A.E., Rudnick, R.L., Ravizza, G.E., and Hart,
metals is in the mantle or lower crust, and that 1998, Re-Os, Sm-Nd, U-Pb and stepwise lead S.R., 1998, Re-Os isotope evidence for the com-
dilution by filtering of magma through higher leaching isotope systematics in shear-zone position, formation, and age of the lower conti-
crustal rocks makes the deposit smaller. The hosted gold mineralization: Genetic tracing and nental crust: Nature, v. 393, p. 58–61.
correlation developed here may be a useful ex- age constraints of crustal hydrothermal activity: Sawkins, F.J., 1990, Metal deposits in relation to plate
Geochimica et Cosmochimica Acta, v. 62, tectonics (second edition): New York, Springer-
ploration tool in that Os isotope measurements
p. 1925–1936. Verlag, 461 p.
may give an indication of the potential size of Freydier, C., Ruiz, J., Chesley, J., McCandless, T., and Shen, J.J., Papanastassiou, D., and Wasserberg, G.J.,
the deposit. Munizaga, F., 1997, Re-Os isotope systematics 1996, Precise Re-Os determinations and system-
of sulfides from felsic igneous rocks: Applica- atics of iron meteorites: Geochimica et Cosmo-
ACKNOWLEDGMENTS tion to base-metal porphyry mineralization in chimica Acta, v. 60, p. 2887–2900.
We thank Mark Baker for all his help in the labora- Chile: Geology, v. 25, p. 775–778. Shibata, K., Ishihara, S., and Ulriksen, C.E., 1984,
tory, and Miguel Tapia, Eduardo Fernandez, and Guilbert, J.M., and Park, C.F., 1986, The geology of ore Rb-Sr ages and initial 87Sr/ 86Sr ratios of late
Esteban Acuña from the Rio Algom Mining Com- deposits: New York, W.H. Freeman, 985 p. Paleozoic granitic rocks from northern Chile:
pany, Roberto Blondel and Walter Orquera, of El Gustafson, L.B., and Hunt, J.P., 1975, The porphyry Geological Society of Japan Bulletin, v. 35,
Salvador CODELCO, and Roberto Freraut and Jose copper deposit at El Salvador, Chile: Economic p. 537–545.
Rojas from Chuquicamata CODELCO, who assisted Geology, v. 70, p. 857–912. Shirey, S., and Walker, R., 1995, Carius tube digestion
in collecting the samples and provided relevant infor- Hunt, J.P., Munchmeyer, C., Marquardt, L., and Soto, for low-blank rhenium-osmium analysis: Ana-
mation about the ore deposits. BHP and Phelps- P., 1983, Applied geology at Quebrada Blanca lytical Chemistry, v. 67, p. 2136–2141.
Dodge kindly allowed us access to their mines and and Collahuasi, and the future of U.S. metal min- Sillitoe, R.H., 1988, Epochs of intrusion-related copper
logistic support while sampling. The work was also ing: Economic Geology, v. 80, p. 794–800. mineralization in the Andes: Journal of South
supported by the National Science Foundation Lang, J.R., and Titley, S.R., 1998, Isotopic and geo- American Earth Sciences, v. 1, p. 89–108.
through grants EAR-9708361 and EAR-9628150 and chemical characteristics of Laramide magmatic Sillitoe, R.H., 1990, Copper deposits and Andean evo-
FONDECYT through grant 1970403. We thank systems in Arizona and implications for the lution: Circum-Pacific Council for Energy and
Spencer Titley, De Verle Harris, John Chesley, genesis of porphyry copper deposits: Economic Mineral Resources, Earth Science Series, v. 11,
Fernando Barra, Hugh Rollinson, and an anonymous Geology, v. 93, p. 138–170. p. 285–311.
reviewer for their constructive comments on earlier Long, K., 1995, Production and reserves of Cordilleran Skewes, M.A., and Stern, C.R., 1994, Tectonic trigger
drafts of this paper. (Alaska to Chile) porphyry copper deposits, in for the formation of late Miocene Cu-rich breccia
Pierce, F.W., and Bolm, J., eds., Porphyry copper pipes in the Andes of central Chile: Geology,
REFERENCES CITED deposits of the American Cordillera: Arizona v. 22, p. 551–554.
Alpers, C., and Brimhall, G., 1988, Middle Miocene Geological Society Digest, v. 20, p. 35–68. Skewes, M.A., and Stern, C.R., 1996, Late Miocene
climatic change in the Atacama Desert, northern Luck, J., and Allegre, C., 1980, Osmium isotopes as mineralized breccias in the Andes of central
Chile: Evidence from supergene mineralization petrogenetic and geologic tracers: Earth and Chile: Sr and Nd isotopic evidences for multiple
at La Escondida: Geological Society of America Planetary Science Letters, v. 48, p. 148–154. magmatic sources, in Camus, F., et al., eds.,
Bulletin, v. 100, p. 1640–1656. Mac Farlane, A.W., Marcet, P., LeHuray, A.P., and Andean copper deposits: New discoveries, min-
Beane, R.E., and Bodnar, R.J., 1995, Hydrothermal Peterson, U., 1990, Lead isotope provinces of the eralization, styles and metallogeny: Society of
fluids and hydrothermal alteration in porphyry Central Andes inferred from ores and crustal Economic Geologists Special Publication 5,
copper deposits, in Pierce, F.W., and Bolm, J.G., rocks: Economic Geology, v. 85, p. 1857–1880. p. 33–42.
eds., Porphyry copper deposits of the American McCandless, T.E., and Ruiz, J., 1993, Rhenium-osmium Titley, S.R., and Hendrick, T.L., 1978, Intrusion and
Cordillera: Arizona Geological Society Digest, evidence for regionally-timed mineralization fracture styles of some mineralized porphyry sys-
v. 20, p. 83–93. events in southwestern North America: Science, tems of the southwestern Pacific and their rela-
Beane, R.E., and Titley, S.R., 1981, Porphyry copper v. 261, p. 1282–1286. tionship to plate interactions: Economic Geology,
deposits: Part II, Hydrothermal alteration and McInnes, B.I., McBride, J.S., Evan, N.T., Lambert, v. 73, p. 891–903.
mineralization: Economic Geology, 75th Anni- D.D., and Andrew, A.S., 1999, Osmium isotope Tosdal, R., 1995, Metal source differences in Cenozoic
versary Volume, p. 214–235. and noble metal recycling in subduction zones: porphyry Cu-Mo-Au deposits in the central
Brandon, A.D., Creaser, R.A., Shirey, S.B., and Carlson, Implication for metallogenesis of porphyry- Chilean Andes between 26° and 28°S: An influ-
R.W., 1996, Osmium recycling in subduction epithermal Cu-Au deposits: Science, v. 286, ence on the size of porphyry deposits?, in Whit-
zones: Science, v. 272, p. 861–864. p. 512–516. ing, B.H., et al., eds., Giant ore deposits: Society
Chesley, J.T., and Ruiz, J., 1998, Implications from the Morgan, J.W., Horan, M.F., Walker, R.J., and Gross- of Economic Geologists Special Publication 2,
Re-Os isotope systematics of the Columbia River man, J.N., 1995, Rhenium-osmium concentra- p. 125–138.
flood basalts: Earth and Planetary Science Letters, tion and isotope systematics in group IIAB iron Williams, W.C., 1994, Geochemical and structural
v. 154, p. 1–11. meteorites: Geochimica et Cosmochimica Acta, characteristics of ore deposits in northern Chile:
Clark, A.H., 1993, Are outsize porphyry copper deposits v. 59, p. 2331–2334. Comunicaciones, no. 45, p. 87–92.
either anatomically or environmentally distinc- Munizaga, F., Huete, C., and Herve, F., 1985, York, D., 1969, Least square fittings of a straight line
tive?, in Whiting, B.H., et al., eds., Giant ore Geochronología K-Ar y razones iniciales with correlated errors: Earth and Planetary Sci-
deposits: Society of Economic Geologists Special 87Sr/ 86Sr de la “Faja Pacifica” de “desarrollos ence Letters, v. 5, p. 320–324.
Publication 2, p. 213–283. hidrotermales”: 4th Congreso Geológico, Antofa- Zentilli, M., 1974, Geological evolution and metallo-
Creaser, R.A., Papanastassiou, D.A., and Wasserburg, gasta, Chile, Actas, v. 4, p. 4.357–4.379. genic relationships in the Andes of northern Chile
G.J., 1991, Negative thermal ion mass spectrom- Puig, A., 1988, Geologic and metallogenic significance between 26 and 29 South [Ph.D. thesis]:
etry of osmium, rhenium, and iridium: Geo- of the isotopic composition of lead in galenas Kingston, Ontario, Queen’s University, 446 p.
chimica et Cosmochimica Acta, v. 55, p. 397–401. of the Chilean Andes: Economic Geology, v. 83,
Drake, M.J., and Jones, J.H., 1983, Partitioning of p. 843–858. Manuscript received November 15, 1999
Re between metal and silicate phases [abs]: Ramos, V., 1989, The birth of southern South America: Revised manuscript received March 7, 2000
Meteoritics, v. 18, p. 291. American Scientist, v. 77, p. 444–450. Manuscript accepted March 29, 2000

558 Printed in USA GEOLOGY, June 2000