Bethke and Rye, 1979

8/11/2019 Bethke and Rye, 1979

1/20

Economic GeologyVol. 74, 1979, pp. 1832-1851

Environment f Ore Deposition n the Creede Mining District,San Juan Mountains, Colorado' Part IV. Source of Fluids

from Oxygen, Hydrogen, and Carbon sotope StudiesPHILIP 5[. BETHKE AND ROBERT O. RYE

Abstract

The hydrogen sotopic composition f fluids responsible or formation of the near-surface silver-base metal vein deposits at Creede was measured by direct analysis ofinclusion luids n sphalerite, uartz, and rhodochrosite nd was estimated rom analysesof illite and chlorite. The oxygen sotopic composition as determined directly on in-clusion luids n sphalerite nd was estimated rom analyses f quartz, llite, rhodochro-site, siderite, and adularia. The carbon isotopic composition was estimated fromanalyses of rhodochrosite nd siderite. The ranges in isotopic composition or waterand CO2 in the fluids associated ith the formation of each of the minerals is givenbelow (number of determinations iven in parentheses):

Mineral a Dn2o%o a sO 2o o a a('0o2%oSphalerite -81 to -54 (4) -10.1 to -4.5 (4)Quartz -97 to -86 (4) -5.9 to 1.8 (18)Illire -62 to -50 (8) -1.6 to 1.2 (7)Chlorite -64 to -55 (10) -2.2 to 0.8 (10)Adularia 4.2 (1)Rhodochrosite --82 to --78 (2) 4.2 to 9.4 (9) --5.7 to --4.2 (9)Siderite 4.9 to 9.9 (6) -6.9 to -2.7 (6)

The 8Di-i,,o nd 8sOI_i,.oalues f fluids ssociated ith the formation f sphalerite,quartz, llite/chlorite, nd carbonate minerals iffer substantially rom one another, ndthese differences ppear o have been maintained hroughout he depositional istory,regardless f the positions f the minerals n the paragenetic equence.

The data suggest hat waters from three coexisting eservoirs ed the vein system

alternately nd episodically uring vein formation, nd apparently here was ittle mix-ing of the fluids from the different reservoirs. The hydrogen, xygen, and carbonisotope ata suggest hat the carbonate aters were deep seated, robably ominantlymagmatic, n origin. The sphalerite nd illite/chlorite waters must have been domi-nantly meteoric n origin and substantially xygen hifted y exchange ith the volcaniccountry ocks. The quartz waters were also oxygen shifted meteoric waters but weresome 40 per rail lower in deuterium ontent han the sphalerite nd illite/chloritewaters.

We propose hat the quartz luids entered he vein system rom reservoirs eneaththe mountainous reas o the north in the vicinity of the present Continental Divide, butthat the sphalerite nd llite/chlorite luids entered he vein system rom a topographi-cally ow area o the south long he structural moat of the Creede aldera. The differ-ence n 8D between he two meteoric waters may reflect differences n altitude of therecharge reas or the wo reservoirs r may be due o isotopic volution f the closed-basin lake and interstitial waters in the moat surrounding he Creede caldera.

Introduction

THS paper is the fourth in a series n which thegeologic, ydrologic, nd chemical nvironments fdeposition f the middle Tertiary silver-base metalores of the Creede mining district are examined. Wepresent here the results of the initial phase of ourinvestigations f the carbon, oxygen, and hydrogenisotopic omposition f the ore fluids. This study was

begun n an effort o establish he sources f the orefluids and to gain an tlnderstandlng f their thermalhistory. Although hese bjectives ave been met na broad sense, t is clear that the Creede district hashad a much more complex hydrologic history thanoriginally anticipated. Specifically, he data suggestthat the ore fluids were dominated by meteoric wa-ters from at least two distinct sources hroughout

1832


2/20

CREEDE MINING DISTRICT: PART IV 1833

most of the vein-filling istory. Each had a differentisotopic ignature, ntered he hydrothermal ystemfrom a different echarge rea, and was associatedwith the deposition f a specific mineral assemblagerecurrently hroughout he period of mineral deposi-tion. A maglnatic omponent as not been unequi-vocally detected, ut neither can it be excluded,either as a contributor of the ore-forming fluidsthemselves r of some critical components f thosefluids.

Geologic nd mineraloyic ackground

The ore deposits f the Creede district are near-surface silver-base metal veins that fill faults andfractures n welded uff of quartz latitic to rhyoliticcomposition Figs. 1 and 2). Nearly all the pro-duction of the district has come from veins filling

the Amethyst, OH, P-vein, and the Bulldog Moun-tain faults or fault systems. n the first paper n thisseries, Steven and Eaton (1975) summarized hegeologic istory f the district, ased rimarily n he

OO'

37 o4'

107 00'

t

\\

\\

':.:-).:?.:

LA GARITA

CREEDE ,

8 MILES

1

KILOMETRES

Postcaldera lavas

(Fisher QuartzLotite) relatedto the Creedecaldera

EXPLANATION

Intrusive rock

Postcaldera sedimentary'rocks (Creede Formation)marginal to the Creedecaldera

Postcaldera lavas relatedto the Son Luis caldera

Geologic contact

Fault

8or and Oo// o downfarown side

Caldera margin

Horizontal Inclined

Compaction foliotion in caldera core

Fro. 1. Relationship f the Creede ein system o calderas n the Central San Juan CalderaComplex. Area of Figure outlined y box. See Steven nd Eaton 1975) or descriptionof relationship f veins o history f caldera evelopment; rom igure of Steven nd Eaton(1975).


3/20

1834 P. M. BETHKE .4ND t. O. RYE

\\

\\

\\

EXPLANATION

1Older undifferentiated ash flow tuffs

Fisher Quartz Latite

Tf, Quartz latite flow

Tfi, Volcanic neck

.'.:..'.fCreede Formation

Geologic contact

?

Fault active just before andduring mineralization

Dashed where uncertain or minor.Bar and ball on downthrown side

Older fault

Dotted where buried. Hachureson downthrown side

0 1 2 KILOMETERSI I ]

Fro. 2. Location map for various veins referred to in text. Area of map shown in Figure 1.For description of geologic history, see Steven and Eaton (1975); modified from figure 3 of

Steven and Eaton (1975).

work of Steven and his coworkers (Steven andRatt, 1965; Rattd and Steven, 1967; Steven et al.,1967; Lipman et al., 1970) and on the early work ofEmmons and Larsen (1913, 1923). The recentpublication on the calderas of the San Juan Moun-tains by Steven and Lipman (1976) is an excellentdescription f the history of the volcanic activity andof the nature of the volcanic country rocks which en-close and underlie the mineralized veins of the

Creede district. Bethke et al. (1976) have shownthat mineralization followed the last dated volcanism

in the Creede district by 1 to 2 million years.The mineralogy and paragenesis f the Creede

ores have been described n several recent reports(Bethke and Barton, 1971; Barton et al., 1977;Wetlaufer, 1977; Hull, 1970; Steven and Rattd,1965). In spite of much effort, the study of thedetailed ime-space elationships s not yet complete,


4/20

CREEDE MINING DISTRICT: PART IF

TAsL 1. Summary of the Paragenesis of Creede Ores

1835

Stage Characteristics Distribution

E stage (youngest)

D stage

C stage

B stage

A stage (oldest)

Fibrous pyrite with some marcasite and stibnite;generally high in orebodies and commonly on cross-cutting fractures

Relatively coarse-grained sphalerite, galena, chalco-pyrite, and quartz, some hematite; silver mineralsnotably absent; probably stage of illire alteration;subdivided into three substages on basis of colorbanding in sphalerite: inner light, middle dark, andouter light

Volumetrically minor, sits on deep etch of earlier Bstage; fluorite overgrowing siderite-manganosideriteand quartz; most fluorite deeply etched--commonlycompletely removed

On OH, P, and northern Amethyst veins, relativelyfine grained sphalerite, galena, chalcopyrite, chlorite,hematite, pyrite, and with some tetrahedrite-tennantite; on southern Amethyst vein and on Bull-dog Mtn., the vein system onsists f banded arite-sulfide with quartz; sulfides, mainly sphalerite andgalena with much tetrahedrite-tennantite and othersulfo-salts; native silver common

Primarily quartz with minor chlorite and sulfide onOH and P veins and on northern ? f Amethyst vein;on southern part of Amethyst vein and on the BulldogMtn. vein system, A stage consists mainly of bothquartz and rhodochrosite

District-wide

Northern parts of OH, P, and Amethyst veins;

poorly developed on Bulldog Mtn. veinsystem as developed thus far

Recognized only in northern parts of OH, Pand Amethyst veins thus far

District-wide

District-wide

although these relationships are probably betterknown for Creede han for any other district. Forthe purposes of the present study, the parageneticsequence s described n terms of five stages ofmineral deposition as outlined in Table 1. Therelationships etween stable sotopic omposition ndmineralogy and paragenesis s an important part ofthis study and some aspects of Table 1 deserveemphasis. Quartz is a major component f stages Athrough D, whereas sphalerite s present in sig-nificant quantities only in stages B and D, the twoore stages. Carbonate minerals are present only inthe A (rhodochrosite) and C (siderite-mangano-siderite) stages, nd the C stage occurrence s volu-metrically very minor. Chlorite appears to bepresent n stages A, B, and D, but its main period ofdeposition was in the B stage. Illite occurs as aproduct of wall-rock alteration n the upper portionsof the vein systems nd, although he alteration zone

may have developed ver a longer period, the ma-terial we have analyzed appears o represent he Dstage.

Procedures

The hydrogen, oxygen, and carbon isotopic com-positions of the hydrothermal luids responsible orthe deposition of the Creede ores were determineddirectly from analyses of the fluids in fluid inclusionsor were estimated rom calculations ased on analy-ses of the isotopic compositions f minerals presumed

to have been in equilibrium with the hydrothermalfluids. The amount of CO2 obtained when fluid in-clusions were opened for analysis by crushing invacuo was too little either for analysis or to affectthe oxygen isotopic composition of the water.Further, in the moderately oxidizing fluids whichexisted at Creede (Barton et al., 1977), isotopeeffects due to the presence f reduced species CO,CH4, H2) can be ignored (Ohmoto, 1972; Ohmotoand Shettel, 1974).

Materials analyzed

We have measured he aD values of the hydro-thermal fluids directly from analyses of water infiuid inclusions n carbonate minerals and quartz,and both the {180 and aD values directly fromanalyses of water in fluid inclusions n sphalerite.\Ve have estimated he hydrogen and/or oxygen so-topic composition of the ore fluids from analyticaldata on illites formed by wall-rock alteration andon vein-filling chlorite, quartz, adularia, and car-bonate minerals. The samples were chosen o pro-vide a geographic distribution and to represent heentire history of vein filling as indicated n Table 1.

In previous reports (Bethke et al., 1976; Barton et al.,1977), the intensely altered rocks found along the upperparts of all vein systems n the Creede district were describedas "sericitic." X-ray diffraction studies indicate that thematerial is actually an illite with from 10 to 20 percentinterlayered smectite (P. H. Wetlaufer and P.M. Bethke,unpub. data, 1978).


5/20

1836 P. 3[. BETHKE .4ND R. O. RYE

North South

/....reede FormationPresenturface x\ . .. .. ...- .Albion . "--J' Commodore L

,, ,L ..... , , ,i I I i , , , ,, - Commodore L


6/20

CREEDE 3IINING DISTRICT: P.4RT IF 1837

temperature of deposition f the Creede ores rangedfrom 190 to 270C and had a median value of about

250C according o fluid-inclusion homogenizationmeasurements eported by Roedder (1965). It isknown hat many of the analyzed amples rew oversubstantial emperature anges, but all fluid calcula-tions have been made assuming sotopic quilibriumat 250C. Also, the isotopic omposition f the orefluids probably changed with time. Therefore, cal-culated sotopic compositions eported here representaverage values integrated over a finite interval ofmineral deposition. An additional uncertainty, com-pounding those due to temperature and parageneticeffects, arises rom the inexactness f our knowledgeof the mineral-water fractionation factors (See dis-cussion by Taylor, 1974a, and Bottinga and Javoy,1973.). There are uncertainties n the experimentaland theoretical curves and their extrapolation tolower temperatures, and uncertainties caused by

variations in the compositions of some minerals.For this study we have used the quartz-water curvebased on the partial equilibrium data of Clayton et al.(1972) as suggested by Taylor (1974a) and themuscovite-water, chlorite-water, and alkali feldspar-water curves summarized y Taylor (1974a, Figs. 2and 4). We have assumed hat the isotopic composi-tions of fluids n equilibrium with the Mn-Fe carbon-ate minerals at Creede can be approximated by thecalcite-water and calcite-CO2 curves of O'Neil et al.(1969) and Bottinga (1968). Finally, the $8On2ovalues estimated rom the mineral analyses have notbeen corrected or activity effects due to the salinityof the Creede ore fluids (Truesdell, 1974). For thesalinity range of 4 to 12 weight percent NaC1 re-ported by Roedder (1965), the reported values or$80 of the fluids mav need to be corrected by asmuch as -2.0 per mil. As these various uncertain-ties are gradually eliminated, our calculated valueswill obviously need revision. The revisions, how-ever, should not be large and should not alter ourprincipal conclusions.

Isotopic Compositions of the Fluids

The results of the isotopic analyses of minerals arelisted n Table 2. The isotopic ompositions f fluids,either calculated from the mineral data or measured

directly on inclusion luids, are also listed. Thetable is arranged o group the samples y vein sys-tem. \Vithin each vein system, the samples arelisted in order of increasing epth. Coexisting min-erals known to have been in contact with one another

are indicated by brackets, but in no sample doescompelling extural evidence suggest hat the pairedminerals equilibrated with the same luid.

In Figure 4, the hydrogen and oxygen isotopic

compositions f the fluids are shown as functions ofboth mineralogy and the paragenetic sequence. Adistinct grouping of isotopic values by mineral isevident n Figure 4 and Table 2, and we shall dis-cuss he data in terms of the isotopic ompositions ffluids from which each of the minerals s presumedto have grown.

Sphalerite fluids

Samples f inclusion luids in four sphalerite speci-mens were analyzed for $DH2 and 8xsOn2. Thefluids analyzed were from samples which representall three major substages of the typically coarse-grained D stage (see Table 1) and which were col-lected rom both he OH and Bulldog Mountain veinsystems. The 8DH20 nd 18OH20 alues n all but thelatest substage sample MB-S-188) fall in a fairlynarrow range ($Dn2o = --54 to -70%0 and: -5.8 to -4.554). These initial data suggest hatthe 8DH2o and $8On2o alues of the ore fluids duringall but the time of deposition f the very latest coarsesphalerite were fairly uniform n both time and spacefor the OH and Bulldog Mountain veins; thus, thedata are consistent with the conclusions of Steven and

Eaton (1975) and Barton et al. (1977) that themajor veins in the Creede district were filled pene-contemporaneously uring he same hydrologic vent.

The $Dno and $sOn2o values of fluids in thevery late sphalerite (MB-S-188) are about -80 and- 10 per mil, respectively, nd are distinctly ighterthan those of the earlier D stage sphalerite luids.

Fluid-inclusion data reported by Roedder (1965)show that filling temperatures f this very late sub-stage average about 210 20C, significantly owerthan the 250 30C reported or the earlier D stagesphalerite. Both the isotope nd filling temperaturedata strongly suggest hat a fundamental hange ookplace in the hydrothermal system during the finalstages of sphalerite deposition, but much more de-tailed fluid inclusion heating and freezing studiescoupled with isotopic determinations are necessaryto document nd evaluate he nature and significanceof such a change.

Quartz fluidsThe $sO values were measured on 17 samples f

quartz from 14 localities n the OH, Amethyst, andBulldog Mountain vein systt.ns and on one samplefrom a small hanging-wall structure about 300 mwest of the Amethyst vein. The samples werechosen o span he deposi'ional equence, ut we can-not clearly document ny quartz belonging o stageB. At this time it is very difficult to correlategrowth zones between quartz and sphalerite sampleswithin stages. It is important o point out that most,


7/20

1838 P. M. BETHKE AND R. O. RYE

if not all, of the quartz samples studied probablygrew over significant ime and temperature ntervals.

Four quartz samples, ll from the OH vein, werechosen for measurement of the aD of the waters

trapped as fluid inclusions. Two of these samplesare clearly ntergrown with D stage sphalerite. Onesample was an unusually coarse (1 x 2.5 cm) andclear rug-filling, A stage, quartz crystal. The fourthsample was a 1-cm milky quartz zone which grew onC stage luorite and was in turn overgrown by earlyD stage sphalerite.

The values of 88Om0o alculated rom the mineral

data (assuming a temperature of equilibration of250C) range from --6.9 to 0.8 per mil (Table 2).This range is much larger than that determineddirectly on the sphalerite nclusion luids (excludingsample MB-S-188). If the range in temperaturefrom_ fluid inclusion studies is taken into account in

the calculation of fluid composition, he range iseven larger. On the other hand, it is also possiblethat the observed variation in a80 of quartz isacutally due to differences n temperature of dep-osition; the entire range in a80 of quartz can, infact, be accounted or by precipitation rom fluids of

TABLE 2. Relative Deuterium (D), 80, and :C Contents of Minerals and AssociatedFluids from the Creede Mining Diqtrict

Minerals Flu idsL

Location 0 Stage (SeeSample (See Fig. 3) Table 1) Mineral aD t*O taC D{o t802o artCoo2

OH vein

PMB CE-251-65 130 ft sublevel at 6-R D IllirePMB X-94-59 30 ft sublevel above Am 5 near D Quartz

19R

PMB DL-277-67 Am 5 level near K2R D IllitePMB DM-279-67 Am 5 level between K1 and K2R D IllirePMB AA-109-59 Am 5 level near 17R D QuartzPMB AA-109-59 Am 5 level near 17R B ChloriteMB-S-182-59 Am 5 level between 15 and 16R C/D uartzMB-S-188-59 Am 5 level between 15 and 16R D SphaleritePMB BA-200-65 7 level near K5R D IllitePMB BC-206-65 7 level between K3 and K4R D IlliteER-127-65 7 level between 16 and 17R D SphaleritePBB-33-106-59 7 level between 15 and 16R B ChloritePBB-40-119-59 7 level between 8 and 9R C SideriteER-119-65 30 m above 9 level between 18 C Siderite

and 19R

PMB-BY-244-65 9 level between K4 and K5R D QuartzPMB-BO-227-65 9 level between K2 and K3R C Siderite

PBB-28-90-59 9 level near 17R C SideritePBB-27-86-59 9 level between 15 and 16R A QuartzMB-K-93-59 9 level between 12 and 13R D Sphalerite

PBB 67-16-67 Above 10 level between K4 and 5R D QuartzPMB-CA-246-65(1) 10 level near K5R A QuartzPMB-CA-246-65(2) 10 level near K5R A QuartzPMP-CA-246-65(3) 10 level near K5R A QuartzPBB-66-15-67 10 level between 19 and K1R A QuartzPMB-50-59 11 level between 10 and 11R C Siderite

PMB-47A-59 11 evelear0R A QuartzMB J-47A-59 1 level near 10R A AdulariaPMB N-58-59 1 level between 8 and 9R B ChloritePBB 13-46-59 12 level near 13R B ChloriteNJP IX-59 12 level near 10R B ChloritePBB 25-83-59 12 level near Volunteer R B ChloritePBB 112-36-59 15 m above C5 level, Volunteer R C SideriteNJP IV-59 C5 sublevel between 16 and 17R B ChloritePBB 8-30-59 C5 sublevel near 14R B ChloriteNJP VI-59 C5 sublevel near 14R B Chlorite

--91 4.8 --53 0.2

8.0 --92 -2.3

-88 3.0 --50 --1.6

--92 4.0 -54 -0.66.0 --4.3

-108 1.8 --61 0.87.5 --97 --2.8

--78, --81 -10.1-89 3.2 --51 --1.4--88 3.9 --50 --0.7

--54, --55 --5.8-111 -0.5 --64 -1.5

16.7 -8.2 9.916.4 --7.5 9.6

7.9 -86 -2.414.9 --6.1 8.114.6 --6.3 7.8

7.4 --2.9

-55, -65, --4.5--67

9.3 --1.08.9 --1.48.1 --2.2

8.7 --1.6

8.3 --96, --95 --2.011.7 --4.0 4.9

6.0 --4.3

3.0 --4.2--108 --0.2 --61 --1.2--102 0.5 --55 --0.5--102 --0.4 --55 --1.4--110 0.0 --63 --1.0

16.5 --7.0 9.7--102 1.5 --55 0.5

--103 --0.2 --56 --1.2--102 --1.2 --55 --2.2

--6.9

--6.2

--4.8

--5.0

--2.7

--5.7

Amethyst vein system

PBB-449-131-59 (1) Am 5 level 30 m S of Albion x-cut D QuartzPBB-449-131-59 (2) An 5 level 30 m S of Albion x-cut D QuartzPBB-449-131-59 (3) An 5 level 30 m S of Albion x-cut D Quartz

5.2 --5.16.7 --3.64.4 --5.9

PMB FW-331-68

PMB FX-332-68

PMB JJ-518-71

PMB KP-558-71

Hanging-wall structures west of Amethyst vein systemC4 level, 300 m W of Man R A Rhodochrosite 11.0C4 level, 300 m W of Man R D Quartz 7.9C4 level, 230 m W of Amethyst B Chlorite --102 --0.3

vein N end of workingsC4 level, 400 m W of Man R D Illite --100 5.8

--6.9

--55

--62

4.2

--2.4

--1.3

1.2

--5.6


8/20

CREEDE MINING DISTRICT: PART IV 1839

TXBLI':2.- (Continued)

SampleLocation a

(See Fig. 3)

PMBKO-557-71 9550 level

PMB KA-542-71 9550 level

PMB KA-542-71 9550 levelPBB 132-8-74B 9360 level

PMB KE-547-71 9360 level,

PBB 145-37-74 9360 level

PBB 145-37-74 9360 levelPMB MU-642A-74 9360 levelPBB 147-50-74 9360 level

PBB 147-50-74 9360 level

PM B GC-338-68 9360 level

PMB LP-596-74 9200 level

PMB NB-658-74 9200 level

123 stopeA180 stopeA180 stopeA243 x-cut

A215 x-cut

145 x-cut

A145 x-cutA145 x-cutA59 x-cut

A59 x-cutA72 x-cut

F91 x-cut

F91 x-cut

PBB 108-32D-68 An 5 level at P vein x-cuO

Stage (SeeTable 1) Mineral

Bulldog Mountain vein systen

D Illitc

A Rhodochrosite

A QuartzA Rhodochrosite

D SphaleriteA Rhodochrosite

A QuartzA Rhodochrosite.\ Rhodochrosite

A QuartzA RhodochrositeA RhodochrositeA Rhodochrosite

P vein

D Illitc

Minerals Fluids t, 2

D tO laC DIt2O tsOl12O (laCco2

--96 --58

13.6 --6.6 6.8 --5.3

10.1 --0.2

14.2 --6.2 --78 7.4 --4.9

--62, --70 --5.415.2 --6.5 8.4 --5.2

12.1 1.8

14.6 --6.2 7.8 --4.9

14.6 --6.3 7.8 --5.0

8.4 1.9

16.2 --5.5 9.4 --4.2

14.8 --6.6 8.0 --5.3

11.2 --7.0 --82 4.4 --5.7

--93 3.7 --55 --0.9

t Values determined directly frmn analysis of fluid inclusions in italics.0-Calculated values assume equilibration temperatures of 250C; see text for fractionation factors used.a Abbreviations used to describe location: An = Anethyst, C = Commodore, Man = Manhatten, R = Raise, x-cut = crosscut; other abbreviations

such as K as in K4R or A as in A180 stope follow cmnpany usage.4 Locality PBB 108-32D-68 not shown in Figure 3; locality is approxbnately 235 n NE (behind plane of section in Fig. 3) of K1R.

constant 8sO}i,o over the temperature ange of 190 to 270C suggested y the fiuid inclusion studies ofRoedder (1965). These uncertainties can be re-moved only by careful selection and analysis of in-dividual growth zones known, from fluid inclusionstudies, o have been deposited over a narrow tem-perature interval.

The deuterium content of the quartz fluids isparticularly interesting. Regardless of whether thequartz appears to have grown either before, con-temporaneously ith, or after the D stage sphalerite,the 8D_oo values of fluids from inclusions in thequartz range from -86 to -96 per mil; substantiallymore negative han those or the sphalerite nclusionfiuids which range from --54 to -70 per mil (ne-glecting the very late sphalerite of MB-S-188).

Carbonate luids

Fifteen carbonate samples were analyzed forO}i2 and 8aCco2and the inclusion luids of twosamples were analyzed for 8D}i2. The samples

represent both the early rhodochrosite f the A stageand the later siderite-manganosiderite f the C stageand were taken from the Bulldog Mountain, OH,and Amethyst structures. One sample was collectedfrom a small hanging-wall vein 300 m west of theAmethyst vein. Wetlaufer (1977) has describedthe textural, compositional, nd paragenetic elationsof both carbonate generations.

The O values of all carbonate samples angefrom 11.0 to 16.7 per mil, and all but three samplesare within the range 13.6 to 16.7 per mil. Thecorresponding *C values range from -4.0 to -8.2

per mil (Table 2). No correlation appears o existbetween the *O or aC values and the chemical

composition r position of the carbonate samples ntime or space n the ore deposit. The fairly narrowrange in O and 8aC of the minerals indicates hatthe temperatures and aCco and sOi,o values ofthe hydrothermal fluids were fairly uniform in thedistrict from one period of carbonate deposition oanother, even though the periods were separated byone of the two main stages of carbonate-free ulfidedeposition. The calculated 8x*O}i= values of thefluids in equilibrium with the carbonates at 250Crange from 4.2 to 9.9 per mil. For the same reasonsdiscussed or the quartz fluids, the actual rangein 88Ou2 may have been substantially arger orsmaller. Although considerable uncertainty existsregarding these calculated luid compositions, t isclear that the 8Oo values of the carbonate luidswere much higher than those of the fluids in equi-librium with any other mineral investigated.

Assuming pH less han 7.1 (Barton et al., 1977,

indicated a pH of 5.4 for B and D stage sphaleritedeposition), CO2would be the dominant carbonatespecies n the fluid at 250C. The calculated 8xaCvalues of CO2in the fluids are very close o those ofthe minerals and have an approximate range of-2.7 to -6.9 per mil.

The waters in inclusion luids n two A stage rho-dochrosites rom the Bulldog Mountain mine wereanalyzed or D}i_oo. The waters have values of -82and -78 per mil and are distinctly different from thequartz fluids and from all but the latest sphalerite


9/20


STAGE

cSTAGE

STAGE

-lOO 90 -8o -70 6o -5o

(a) D%o

Sphalente RhodochrosteIlhteQuartz Sdente Chlonte. Adularla

STAGE

STAGE

10 -8 6 -4 2 o 2 4 6 8 10

(b) 18 O/o

FIC. 4. Relationship of aDHzo and a8OH=o alues of fluidsto mineralogy and position in depositional sequence. Degreeof uncertainty in paragenetic position indicated by verticalbar. Mineral symbols offset vertically to avoid overlap.

fluids. Although he rhodochrosite amples nalyzed

were selected o have a high probability of containingan overwhelming roportion of primary fluid inclu-sions, we cannot be certain that this was the case inthe absence f systematic eating and freezing studies.We are less confident, herefore, of the 8D2o valuesof the carbonate fluids than of those of either the

quartz or sphalerite luids.

Illite fluids

Seven samples of illite were analyzed or 8D and880, and one for aD alone. Of these, five werecollected long the OH vein at the base of the zone

of intense wall-rock alteration (see Barton et al.,1977) and one from the same zone on the P veinsome 700 m east of the OH vein. One sample(PMB KP) was taken from a hydrothermally lteredboulder in a coarse clastic tongue of the CreedeFormation, which fills the paleo-stream channel in-cised into the wall of the Creede caldera and is trun-

cated by the Amethyst fault, and one sample wastaken from the 9550 level of the Bulldog Mountainmine.

The structui-e and composition f the illites havenot yet been ully characterized, ut from X-ray dif-

fraction profiles all appear to have 1M structures,with as much as 20 percent smectite nterlayering.Most samples are very fine grained; the materialsanalyzed were taken from a


10/20

CREEDE MINING DISTRICT: PART II7 1841

positional nformation mmaterial or the purposesof this study. As judged rom he settling ates dur-ing mineral separation, he materials sed or analy-sis were less han St* n diameter.

Calculated values of asOn2o and aDn2o for thechlorite fluids show about the same range as thoseof the illite fluids (0.8 to -2.2o and -55 to-65%). In view of the uncertainty n our knowl-edge of the D and sO fractionation actors betweenchlorite and water (particularly high iron chlorite),the actual ranges n DH20 and sOu._,o should notbe considered well established.

idularia fluids

The single sample of adularia analyzed for sowas a split of the material used for K-Ar agedetermination Bethke et al., 1976). Its parageneticposition s fixed as early in the depositional istory(A stage). The vein quartz with which it is inter-

grown was also analyzed. The textural relations aretoo complex o establish he codeposition f the pair,but the calculated 8OH2o f the fluid in equilibriumwith the adularia at 250C is -4.2 per mil (Table2), essentially he same as the value of -4.3 per railcalculated or the quartz, suggesting hat isotopicequilibrium was achieved between he mineral pair.

Summary of isotopic compositions [ fluidsThe oxygen and hydrogen sotope data for the

various fluids are summarized on the now-standard

aD vs. a8OH= plot in Figure 5. Most analyzedmodern meteoric waters (except those that have

undergone xcessive vaporation) ie within a fewper mil of the meteoric water line, and presumablyall paleometeoric aters, at least those as recent asthe middle Tertiary, would also fall along this line(Craig, 1961; Taylor, 1974a). The isotopic com-positions of almost all primary magmatic waters areinferred to lie within the box labeled "Deep-SeatedVVater" Rye, 1966; Taylor, 1974a, and 1977), andpresumably waters derived from the crystallizationof shallow magmas at Creede would also have iso-topic compositions ithin this range. The isotopiccompositions f Creede luids, as reported n Table 2,are plotted on the diagram. The solid symbols epre-sent samples wherein both DH20 and asOuo_o avebeen determined. The open symbols epresent quartzand carbonate samples whose oxygen isotopic com-position alone was measured; hese are plotted onthe diagram at the average 8DH2 value of quartz orcarbonate fluid inclusions. Three modern surface

waters (two from springs and one from mine drain-age) are plotted on the meteoric water line at theirmeasured hydrogen sotopic compositions.

Three important observations can be made re-garding he data plotted n Figure 5:

1. The fluids associated with the deposition ofeach of the different mineral groups (sphalerite,quartz, carbonates, nd hydrous minerals) had sub-stantially different isotopic compositions.

2. Only the carbonate luids plot close o the fieldfor deep-seated aters; the other fluids plot at posi-tions considerably displaced oward the meteoric

- 4O

-60,

o -80

- lOO

-120

-16

} Sphalerlte 0 Quartz [] Rhodochroslte

&. Siderite

llhte* Chlonte

i i i [ i i i i L I I I f

-12 -8 -4 0 4 8 12

180%o

FZG. 5. The (SDn.:o nd (5On_oo alues of Creede ore fluids showing fields occupied by fluidsresponsible or the formation of different minerals. Solid symbols epresent samples whereinboth (Dn,o and (On_oo have been determined. Open symbols represent quartz and carbonatesamples whose oxygen isotopic composition lone was measured; these are plotted at theaverage H=O value of quartz or carbonate luid inclusion (see text).


11/20

1842 P. M. BETHKE hVD R. O. RYE

water line, suggesting considerable nvolvement ofmeteoric waters.

3. These other waters (with the exception of thefluid inclusions n the very late sphalerite) all intotwo distinct 3Dn2o groupings, ne ncluding phaler-ite and phyllosilicate luids at -58---10 per mil,the other representing quartz fluids at --91 + 6per mil.

The significances nd implications of these obser-vations are discussed n the following sections.

Relationship between sotopic Compositionsof Fluids and Mineralogy

The isotopic composition f the fluids responsiblefor the deposition f each of the various minerals oc-cupies a relatively small field on the 3D vs.plot (Fig. 5). Furthermore, only the illite andchlorite and the siderite and rhodocrosite fields

overlap; the fields for other minerals are widelyseparated. This association of different mineralswith hydrothermal luids of different isotopic com-positions has been observed n at least two otherdeposits, but the situation at Creede is moredefinitive. At the Pasto Bueno wolframite-sulfide

deposit n northern Peru, two periods of wolframitedeposition were related to relatively low aD watersthat were presumably meteoric in origin (Landisand Rye, 1974). At the Panasqueira, ortugal, tin-tungsten deposit, quartz and wolframite were associ-ated with fluids having an average aD2o ofper mil, whereas other minerals such as cassiterite,

apatite, and arsenopyrite were deposited rom fluidsthat had distinctly more negative aDn2o values(Kelly and Rye, 1979). At both of these deposits,but especially t Panasqueira, he origin of the fluidswas difficult to interpret because he 3sOno valuesof the fluids were very high and rather uniform, in-dicating that regardless of their origin, the fluidshad undergone a high degree of oxygen isotope ex-change with the country rocks. At Creede, however,the sO values for the hydrothermal waters shownearly a complete range between those typical ofvirtually unexchafiged meteoric waters and thosetypical of magmatic water.

In addition to the above-mentioned considerations,it is particularly important to reiterate that, for thoseminerals which we have sampled t various positionsin the depositional equence, his specificity of iso-topic composition f the fluid associated with a par-ticular mineral is maintained regardless of the posi-tion of that mineral in the depositional equence. hecarbonate and quartz data in particular illustrate thispoint. As shown n Figure 4, the oxygen sotopiccomposition of the A stage rhodochrosite waterscovers essentially he same ange as does hat of the

C stage siderite-manganosiderite luids. Both aremuch more enriched n sO than he fluids epresent-ing any other mineral. Quartz is an important min-eral or the whole depositional equence, nd hrough-out that sequence he isotopic compositions f thequartz-depositing luids appear to have been sub-stantially different from those associated with thedeposition of any other mineral.

There are several mportant mplications f theabove discusison. First, the oxygen and hydrogenisotope data require that the various minerals weredeposited roln fluids having different sotopic om-positions and, therefore, different origins and/orhistories of rock-water interaction.

Second, he paragenetic nd isotope data suggestthat these different fluids occupied he vein system(at least in the vicinity of the present orebodies)episodically nd repetitively hroughout he historyof vein filling, and there is little evidence f largescale mixing of the different fluids.

Third, although many of the nfinerals are inter-grown (even as coarse crystals ining vugs, a typicalrelation for D stage sphalerite and quartz), theymust have grown alternately from the differentfluids. The most striking examples re the intimatelyintergrown quartz and rhodochrosite of samplesPMB-KA, PBB-145, and PBB-147 (Table 2). Inthese samples, he quartz is finely intergrown as aminor phase in massive rhodochrosite. Thesesamples were specifically elected or sampling asmineral pairs. In PBB-147, textural evidence ug-

gests hat at least some of the quartz actually veinsthe rhodochrosite, ut in the other two samples, notextural evidence suggests anything but mutualgrowth. The sO values of these quartz fluids arethe highest we have found for quartz, but the calcu-lated oxygen sotopic compositions f the quartz andrhodochrosite luids still differ by 4 to 6 per mil andlnUSt therefore represent wo distinct fluids. Thisis true whether the two fluids 1nixed and coprecipi-tated the quartz and rhodochrosite, he difference nsO being due to kinetic factors, or whether thedifferences n 1sO represent equilibrium fractiona-tion during alternate deposition of the two minerals.

Fourth, the evidence that different minerals, al-though intergrown in the salne hand specimen, weredeposited from fluids of different origins and/orhistories of rock-water nteraction has very import-ant implications egarding he assumption f chemi-cal equilibrium between minerals. It may be as-sumed, until proven otherwise, that fluids fromdifferent sources or with different histories of rock-

water interactions will differ in chemical composi-tion, and therefore minerals deposited rom differentfluids will not be in total chemical equilibrium. This


12/20

CREEDE MINING DISTRICT: PART IF 1843

is particularly mportant or minerals whose hemicalcomposition s used in equilibrium arguments butmay be somewhat ess critical where the presence orabsence f one or more minerals s the equilibriumrequirement see discussion y Barton et al., 1963).The chemical arguments f Barton et al. (1977) forthe OH vein were based on the observed presenceof quartz, sericite (illire), potassium eldspar, chlo-rite, pyrite, hematite, and chalcopyrite; he lack ofobserved bornite, covelike, magnetite, and pyrrho-tire; and the FeS content of sphalerite. These evi-dences still seem valid, but the multiple fluid modeldeveloped here may explain the recurrent hydro-thermal leaching of sphalerite and certain other post-depositional hanges n sphalerite and other sulfideminerals described by those authors.

The complexity of the hydrologic egime mpliedby the above considerations s one of the most sur-prising and intriguing results of this study. Thestrong contrasts among the isotopic compositions fthe fluids responsible or the deposition f the vari-ous minerals together with the regularities n thecovariation of temperature and salinity determina-tions of fluid inclusions with paragenetic positionfound by Roedder 1977) provide a powerful ool inour continuing attempts o unravel the complexities.

Origins of the Hydrothermal Fluids

The 8DH2o and 88OH2o f hydrothermal luids are,in a general sense, he products of the original iso-topic compositions f the source luids, modified by

the degree of isotopic exchange between he fluidsand the rocks which they traversed. The isotopicsignature of the water may be further modified bymixing of fluids, by boiling, or by introducing orproducing significant amounts of reduced speciessuch as CO, H20, or CH4. Boiling has beendemonstrated or the Creede fluids (Barton et al.,1977), but at 250 the fractionation of both sO andD between iquid and vapor is too small for boilingto affect the isotopic composition f the fluid sig-nificantly unless the salinities of the fluids were in-creased strongly due to boiling, as in a vapor-dominated ystem. None of the fluid inclusion ob-servations have shown the presence of such highsalinities at Creede, nor have they suggested argedifferences in salinities between fluid inclusions in

quartz and sphalerite Roedder, 1965; J. T. Nash,written commun., 968). (See discussion y Trues-dell et al., 1977, on the effects of boiling on the iso-topic compositions f fluids n the Yellowstone geo-thermal system.) Further, the chemical environ-merit, as outlined by Barton et al. (1977), precludesthe existence f significant mounts f reduced peciesin the Creede fluids. Therefore, we must look to

original differences o explain the variations n iso-topic composition f the fluids.

The geologic history of the Creede district andthe close ime relationship of ore deposition o vol-canism suggest hat the only sources hat need beconsidered as contributors to the ore fluids are

lneteoric and magmatic waters. We may now con-sider what contribution each of these sources mayhave made to the constitution of the various iso-

topically distinct waters.

Carbonate fluids

The range of calculated sO values (4.2 to 9.9/,)is essentially he same or both the rhodochrosite ndsiderite luids. These O contents re much higherthan those known for any other mineral in theCreede veins and fall within or very near the "nor-mal" isotopic ange for magmatic water. The directdeterminations of 8D on fluid inclusions from two

rhodochrosite amples -78 and -82o) fall nearthe lighter end of the normal $D range or magmaticvaters. These oxygen and hydrogen sotope datasuggest hat either the carbonate fluids contained avery substantial magmatic component r that theywere meteoric waters that underwent extensive

oxygen isotope exchange with deep-seated ountryrocks.

The carbon isotope data are consistent with theabove conclusion. The range of 8aC values of thevein carbonates -8.2 to -4.0) lies within thatof deep-seated arbon as indicated by analyses of

carbonatites nd carbonates n igneous ocks Ohmotoand Rye, 1979). These values are also typical ofnumerous hydrothermal carbonates associated withfelsic intrusives in various parts of the worldthroughout geologic ime. The 8aC values of -8 to-5 per mil have been observed n hydrothermal andcarbonate minerals that occur in ore deposits n alltypes of country rocks and irrespective of whether ornot carbonate rocks are known, or inferred, to be inthe area. The convergence f the carbon, hydrogen,and oxygen isotopic data on values typical of mag-matic environments strongly suggests a magmaticsource for the carbonate fluids.

However, under certain extreme conditions theobserved sotopic compositions f the vein carbon-ates conceivably ould be produced by a deeply cir-culating meteoric water interacting, at moderate em-perature, with a buried marine carbonate unit. InFigure 6 we have ploted he 880 and $1aC data forthe carbonate minerals. Also plotted, for comparison,are the data of Steven and Friedman (1968) onCreede Formation travertines. The Creede Forma-

tion is a clastic unit composed f volcanic ash anddebris deposited n a shalloxv ake which occupied


13/20

1844 P. M. BETHKE ,,tND R. O. RYE

0p

2

10 110 * i 12 14 16 18 20 22 24 26b180%o

Fro. 6. Relationship between *mC and *80 for carbonateminerals (mineral data, not fluid data, plotted). Squaresrepresent rhodochrosite samples (A stage); triangles repre-sent siderites (C stage). Data for travertines deposited nCreede Formation as reported by Steven and Friedman(1968) shown as circles for comparison.

the moat of the resurgent Creede caldera. Thetravertines were deposited while the volcaniclasticsediments of the Creede Formation accumulated, andthey predate the vein carbonates by as much as 2million years and are volumetrically much more im-portant (Steven, 1969: Bethke et al., 1976). Stevenand Friedman (1968) concluded hat the travertineswere deposited by meteoric water at low tempera-tures and that they derived their carbon from dis-solution or decarbonation f a thin wedge of Meso-zoic sedimentary ocks containing marine carbonatebeds which they presumed o underlie the volcanicrocks of the central San Juan Mountains.

If the vein carbonates ormed from circulatingmeteoric waters which interacted with such deepmarine carbonates, pH of 9 or greater must haveobtained (assuming a temperature of interaction ofapproximately 300C), and total carbon, but onlypartial oxygen, sotope exchange between he fluidsand the carbonate would be required. In our opin-ion such required conditions re excessive. We sug-gest that the convergence f the $3C, 8180, and aDvalues of the carbonate luids on values typical ofdeep-seated or lnagmatic fluids favor a magmaticorigin for both the CO2 and H.O components ndrequire, at the least, a deep, high temperature historyfor the carbonate fitlids.

Sphalerite fluids

The D values of the fluid inclusions n sphalerite(Fig. 5) are well within the range of values fornormal magmatic waters and are much heavier thanthe values of present-day surface water. The $aOcontent of the fluid inclusions, however, is muchlower than that of normal magmatic waters and liesclose to the meteoric water line. The sphaleritefluids must have been either: (1) meteoric waters

whose deuterium content was significantly higherthan present surface waters and whose 1O contentwas shifted to higher values by reaction with thevolcanic wall-rocks; 2) magmatic waters whose 1Ocontent shifted o much ower values by low-tempera-ture exchange with the wall rocks; or (3) mixturesof the two.

For years t has been recognized hat meteoric wa-ters can reach large 8sO values, even as high asthose ypical of magmatic luids, as a result of ex-change with high-temperature ountry rocks (Craig,1963). The 8'C) values of magmatic waters, nor-really near 6 to 8 per mil (Rye, 1966; Taylor,1974a, 1977), can decrease s a result of low-tem-perature exchange with country rocks which host theore deposit. The actual decrease n 8sO values oflnaglnatic water depends n the degree of exchange,the water/rock ratio, the temperature of equilibra-tion, and the original alSO of the rocks. The lowest

asO values of exchanged magmatic fluids will beproduced when the water/rock ratio, temperature,and 8O of the country rocks are all as low aspossible.

The rocks immediately enclosing he productiveparts of the veins at Creede (the Bachelor MountainMember of the Carpenter Ridge Tuff) underwentextensive potassium metasomatism uring an earlierhydrothernml vent several million years prior to oredeposition Ratt and Steven, 1967). This introduc-tion of potassium ffected a large volume of rock inthe Creede district and produced rocks havinggreater han 8 weight percent K20 and having K20/Xa20 weight ratios as high as 18. \Ve have mea-sured the oxygen sotopic and chemical compositionof a large number of these potassium-metasomatizedwall rocks (R. O. Rye and P.M. Bethke, unpub.data, 1977). The potassium metasomatism as beenfound to be accompanied y an $1sO shift in therocks o lower values, and a good correlation existsbetween 80 and the K20/Na20 ratio. The lowestas() value observed hus far in the country rocks is4.3 per rail. If the ore fitlids exchanged with rocksthat had such ow $sO values at temperatures as lowas 200C under conditions of very low water/rockratios. then the O of the fluid could have reachedvalues as low as -4 per rail. For several reasons,however, this is an unreasonable lower limit.

First, to achieve such a low value would requireboth complete water-rock oxygen isotope exchangeand a small water/rock ratio, an exceedingly nlikelycombination n an open issure system such as existedduring mineralization. Second, ignificant xchangeis unlikely to have occurred at a temperature as lowas 200C. In most open hydrothermal systems,the fitlids are not in equilibrium with the host


14/20

CREEDE MINING DISTRICT: PART II/ 1845

rock at the site of ore deposition but, rather, withhigher temperature rocks much deeper n the system(Rye, 1966; Rye and Sawkins, 1974). Third, hostrocks having $80 values as low as 4.3 per mil con-stitute ' a small volume of the total host rock in theBachelor Mountain unit. Such low values have not

been observed n the overlying and underlying unitsnear the vein systems. Unless the volcanic and Pre-cambrian ocks deep within the hydrothermal lumb-ing system were somehow depleted n 80 prior toore deposition, most of the host rock with which thefluids could have exchanged probably had a80values considerably arger than 4.3 per rail. A morerealistic lower limit for a80 values of exchangedmagmatic waters in the Creede system would prob-ably be 2 to 3 per rail.

Thus, on the basis of a80 data alone, the sphaler-ite fluids must have had a major meteoric compo-nent. Although in no way do the data rule out thepossible contribution of a magmatic component, hesphalerite luids must have been dominated y watersof meteoric origin.

Illire and chlorite fluids

If the isotopic ompositions f the fluids calculatedfrom those determined on the minerals are correct,the aDn2o ange of the illite and chlorite fluids wasessentially identical to that of all but the latestsphalerite luids, but the asOn2o values of the fluidswere several per rail larger. In spite of the previ-ously discussed uncertainties in the values of the

mineral-water fractionation factors used to calculatethe fluid composition, we feel that this difference n88Ox2o s probably real and that the 8Dn.o cor-respondence s reasonable. The largervalues may indicate a larger magmatic contributionto the fluid, a higher degree of isotopic exchangewith the country rocks, or only partial oxygen iso-tope exchange during the formation of the illite andchlorite. The available evidence provides no basisfor choosing among these possibilities. The samearguments presented or the source of the sphaleritefluids apply to the illite and chlorite fluids and, evenif a larger magmatic contribution was the canse ofthe a80_oo difference, the fluids must still havebeen dominated by meteoric waters.

Quartz fluids

The oxygen isotopic compositions f the quartzfluids calculated rom the mineral compositions overa broad range and overlap both the sphalerite andillite-chlorite fluid compositions. Most of thevalues were obtained early in the study; the hydro-gen determinations were made ater. In the absenceof data for the hydrogen sotopic composition f the

quartz fluids, their $D was initially assumedto be also about -60 per rail, and the quartz fluidsseemed o bridge the gap between he sphalerite ndillite-chlorite luids (Bethke et al., 1973). Whenthe hydrogen sotopic ompositions f inclusion luidsin four samples of quartz were determined, t be-

came apparent hat this assumption was wrong andthat the quartz was deposited rom a fluid having anisotopic composition strongly contrasting with thatassociated with the deposition of any other mineral.The low values of aDn,.o, as well as all the argumentsconcerning oxygen isotopic shift through exchangewith country rock, virtually require that the quartzfluids also originated primarily as meteoric watersbut having a lower deuterium content than themeteoric waters responsible or sphalerite, llite, andchlorite formation.

The above-mentioned onsiderations n the originof the fluids appear to require that at least threedistinct waters occupied he Creede vein system e-currently during the depositional history. One ofthese fluids, the one responsible or depositing bothcarbonate stages, must have been of deep-seated,probably magmatic origin. The other two waterswere dominantly meteoric in origin but differedgreatly in their hydrogen sotopic omposition. reedeis not uuique n having waters from different sourcesinvolved during vein formation. We have pointedout previously hat the wolframite-sulfide eposits tPasto Bueno in northern Peru and the tin-tungstendeposit at Panasqueira, Portugal, appear to have

been formed by fluids from more than one source.In addition, the studies of Kamilli and Ohmoto(1977) at Colqui and of Sawkins and Rye (1977)at Caudalosa show that these two Peruvian vein

deposits lso t)resent evidence of waters from diversesources. ,Vhat does appear uniqne about Creede(and perhaps Panasqueira) is the paragenetic evi-dence which requires hat the three waters coexistedin separate eservoirs which fed the Creede vein sys-tem el)isodically hroughout he period of vein filling.

Speculations on Recharge Areas and 8DVariations of Meteoric Waters

The discovery hat two meteoric waters whose aDvalues differed by as much as 40 per rail were in-volved in mineral deposition n the Creede vein sys-tem, apparently alternately throughout most of thelife of the system, is one of the most significant, ifperplexing, results of this study. Such a situationcould arise in two general ways. Either (1) the twometeoric waters were originally different n hydrogenisotopic composition r (2) the hydrogen sotopiccomposition of the surface waters over the entirearea was uniform, but waters in different reservoirs


15/20


underwent substantially different isotopic evolutionprior to entering the vein system.

The first alternative has at least two possiblevariations. Either (a) the two waters were gen-erated at different times and entered he system roma single echarge rea or (b) waters of substantiallydifferent sotopic omposition ntered he system dur-ing the same interval but at two or more separaterecharge areas feeding different reservoirs. Sparsedata from both the eastern (Taylor, 1974a, b) andwestern Forester and Taylor, 1972; Taylor, 1974a;Casaderail nd Ohmoto, 1977) San Juan Mountainsshow that the aD values of surface waters which have

existed n the San Juan Mountains during the past30 million years have varied by at least 70 per mil.This variation must be the result of several factors

related to the geologic history of the San JuanMountains and of the western United States. These

factors nclude high relief, regional uplift, and changesin the relative influence (perhaps in both time andspace) of Pacific and Gulf of Mexico air masses.Although it appears o be generally true that thedistribution of aD values of meteoric water in the

western United States during much of Tertiarytime was nearly the same as that of the present day,it is clear that one must be very careful in applyingsuch an assumption o specific deposits n the SanJuan Mountains, or in many other areas of presentlyhigh relief.

The aD values of meteoric water at a specific e-charge area for a hydrothermal system can change

considerably ith daily, seasonal, nd yearly climaticvariations but probably not by as much as 40 per mileven in areas of high relief and where mixing ofdifferent air masses occurs. In any event, becausethe residence ime of meteoric waters n hydrother-mal systems s much longer than annual weathercycles, short term variations n the 8D of meteoricwater at a single recharge area would probably behomogenized within the hydrothermal reservoir. Weconsider t unlikely that the observed differences n8D were generated by such short term fluctuations.

The aD values of meteoric waters at a specificlocality may also change over long periods of time.For example, aD values of meteoric waters can be-come more negative during general uplift of an areaor as a result of a decrease in the annual mean tem-

perature of precipitation. The aD values of theore fluids at Climax, Colorado, became progressivelydepleted n deuterium with time, and one interpreta-tion was that ore deposition was going on duringuplift (Hall et al., 1974). At Creede, however, hequartz fluids are much more depleted n deuteriumrelative to the sphalerite (and illite and hlorite)fluids, even hough he quartz was deposited hrough-

out the paragenetic sequence. This requires thattwo fluids must have coexisted during the period ofvein filling. It is conceivable hat, as a result ofgeneral uplift, a single, sotopically oned reservoirhaving an early, heavy water at depth and a light,younger water on top might have been produced. nthe absence of convective overturn, such a reservoirmight have persisted or a significant ength of time.However, the range of homogenization emperaturesof fluid inclusions n quartz is about the same as thatfor those in sphalerite (Roedder, 1965), and it isdifficult o imagine how such a vertically zoned res-ervoir could have escaped onvective mixing if boththe lower and upper parts were heated o approxi-mately the same emperature by any source of heatat depth. It seems much more likely that the twowaters, whatever the origin of their differences nD, coexisted in two separate reservoirs.

The geology and paleotopography of the Creede

area as described by Steven and Eaton (1975),Steven and Lipman (1976), Steven (1968), andSteven and Ratt (1965) are compatible with thepresence of two distinct recharge areas, differingsignificantly in altitude. The ore occurs alongfractures hat make up the north-northwest-trendingCreede graben structure Fig. 1). To the north, thefractures can be traced into the ring fracturesystem of the San Luis Peak caldera near thepresent Continental Divide. To the south, along thepresent position of the headwaters of the Rio Grande,the fractures transect the ring fracture zone of theCreede caldera and cut the volcaniclastic sediments

of the Creede Formation, which partially fill themoat of the caldera. It is probable that the veinsystem was fed by waters from both of these areasand that they supplied waters differing in aD by asmuch as 40 to 50 per mil, with both being heavierthan the present-day surface water whose D is-112 per rail. Accepting the proposition hat thequartz waters entered the vein system rom one re-charge area, and the sphalerite and illite/chloritefluids from another, xve may inquire as to the causesof the differences n the hydrogen sotopic omposi-tion of the meteoric waters in the two areas.

The aD value of meteoric water is largely a func-tion of the annual mean temperature of precipitationwhich is related to both altitude and latitude. The

presence of an unevolved meteoric water having a8D of -55 per mil at the time of ore deposition nthe Creede district would require that 25 millionyears ago the Creede district was much lower thanat present and that few significant mountain rangesexisted between Creede and the Pacific Ocean and/or the Gulf of Mexico. These conditions are rea-

sonable n the light of available geologic evidence.


16/20

CREEDE MI,1NG DISTRICT: PART II7 1847

Reconstruction of the western United States dur-

ing the Eocene indicates that prior to basin andrange faulting the distance rom Creede o tile PacificOcean may have been 300 km less than today(Hamilton and Myers, 1966). Also, altitudes intile western Uuited States were considerably less.Most significantly, however, the Creede area wasabout a kilometer lower than its present altitude(Steven and Eaton, 1975) and was part of a large,poorly drained volcanic plateau surlnomlted by a fewisolated hills (Steven, 1968, and oral conlmun.,1976). The Sangre de Cristo Range did not exist(Taylor, 1975). In the absence of the Sangre deCristos, no major mountain barrier existed betweenthe Gulf of Mexico and the Creede district; thus,the $D values of rain falling on the caldera wall andsupplying recharge for the ring structure of thecaldera could have been about -55 per rail.

On the other hand, a $D value of about -95 permil or less would not have been unreasonable forprecipitation falling near the present ContinentalDivide north of Creede, because he high plateauand surmounting hills surrounding the Creede cal-dera were as nluch as 1,200 m higher than the lakelevel in the caldera moat (Steven and Lipman,1976) and because he divide region could havereceived substantial precipitation from Pacific airmasses. In areas of abrupt altitude changes, suchas the east side of the present Andes, the isotopiccomposition of meteoric water can change quiteabruptly over short distances.

A second possible cause of the differences n aDbetween he two recharge areas is particularly ap-pealing; such a difference might be due to a differ-ence in isotopic evolution within the reservoirs. Theclastic Creede Formation is a potential reservoirwhere meteoric waters could evolve isotopically oheavier values of fid and imBO. According o Stevenand Ratt (1965), the Creede Formation accumu-lated in an arcuate closed asin ormed by the sub-sidence and later resurgent doming of the Creedecaldera. The bulk of tile material illing the moatlikebasin consists f thin-bedded ilty shales and sand-stones and some tuff beds, all of which accunmlated

in a shallow ake or playa euvironment. Along themargins, these beds intertongue with stream sedi-ments, primarily sandstones nd conglomerates, ndwith fanglomerates and talus-regolith deposits.Large masses of travertine, deposited rom mineralsprings concurrently with sedimentation, are wide-spread throughout tile Creede Formation. Much ofthe clastic material which makes up the CreedeFormation was deposited as air-fall and ash-flowruffs from the Fisher Quartz Eatire volcanoes hicherupted along he ring fracture of the Creede caldera

at the tilne the Creede Formation was accumulating.Much of this material, along with material derivedfrom the weathering of the surrounding ash-flow uffsheets was reworked by stream and wave action. Inshort, the lnoat of the Creede caldera was the siteof a shallow closed-basin ake, or playa, in whichsilicic volcanic ash, the bulk of it glassy when de-posited, and the weathering products of silicic vol-canic rocks, accumulated to a thickness of about 1km or lnore (Steven and Lipman, 1976).

The climate during Creede ime was similar to thatat present (Steven and Eaton, 1975). It would beexpected hat under such conditions a shallow closed-basin lake would evolve chemically oward an alka-line saline lake through the combined processes ofevaporation and diagenesis. Many similar closed-basin lakes exist in the western United States at

present (Jones, 1966; Eugster and Hardie, 1979).Neither the degree nor trend of the chemical evolu-

tion of Lake Creede can yet be evaluated becauselittle is known of the mineralogy of the lacustrinefacies of the Creede Forlnation. No evaporite acieshave been reported, but Steven and Van Loenen(1971) have shown hat some of the tuff beds havebeen converted o clinoptilolite or clinoptilolite-smec-tite mixtures and, according o Steven and Ratt(1965), calcite was precipitated n thin limestonebeds, as cement between clasts. and as travertinearound spring orifices. \Vhat little is known of themineralogy of the Creede Formation supports theevolution of a shallow. moderately alkaline, saline

lake and ground-water reservoir during the deposi-tion of the Creede Formation.The processes f evaporation and diagenesis par-

ticularly the hydration reactions of glass to clay)both lead to isotopic, as well as chemical, evolutionof the lake waters. The fractionation of hydrogenand oxygen isotopes during evaporation rom stand-ing bodies of water was first investigated experi-mentally by Craig et al. (1963) and most recently bySofer and Gat (1975). As evaporation akes place,the residual water is enriched in both deuterium and

80. The relationship between the enrichment trendsis such that lid = 5 x li*O, consistent with the

trends observed n manv natural evaporite systems.In contrast to the effects of evaporation on the

isotopic compositions of the lake and interstitialwaters, the effects of diagenetic reactions, such ashydration of glass to smectite or zeolite. or the pre-cipitation of calcite, cannot be estimated without amuch better knowledge of the quantitative mineral-ogy of the Creede Formation. However. by analogywith the isotopic evolution of waters in closed basins(cf. Taylor. 1974a), we would expect that the resultof such evolution. ike that caused by evaporation,


17/20

1848 P. M. BETHKE IND R. O. RYE

would be to increase both the deuterium and 80contents of the waters.

Assuming n average hickness f the lake beds nthe Creede Formation of 750 m, and outer diameterof 20 kin, an inner diameter of 8 km for the moat,and an arc of deposition f 270 (Figs. 1 and 2), thetotal original volume of Creede Lake sedimentswould have been nearly 100 km 3. If these sedimentshad an average effective porosity as low as 0.1, thethe Creede Formation would have been an enormous

reservoir ontaining 0 km 3 of meteoric water, whoseisotopic composition probably was substantiallyevolved oward arger values of aD and $80. Thus,whether or not there was a significant difference nisotopic omposition f meteoric waters falling ondifferent parts of the central San Juan Mountains25 million years ago, subsequent volution of thewaters in different reservoirs could account for the

contrasting sotopic compositions f the quartz wa-

ters and of those responsible or sphalerite (andillite/chlorite) formation. Waters entering the sys-tem from the mountain areas to the north would

have exchanged oxygen with the volcanic rocks asthey became heated but would have remained almostconstant in deuterium content. These waters would

have been responsible or the bulk of the quartzdeposition. \Vater entering the vein system romthe low moat area to the south would have evolved

isotopically hrough evaporation nd diagenesis olarger values of aD (and asO) and would have beenresponsible or sphalerite deposition nd for illiteand chlorite formation.

Summary and Conclusions

In the preceding aragraphs, e have discussedvarious possible lternatives whereby meteoric wa-ters of substantially different isotopic compositionsmight have been generated n the Creede area andentered the vein system at various times. Theparagenetic vidence equires hat two isotopicallydifferent reservoirs must have coexisted and fed the

vein system lternately. For that reason, we stronglyfavor either: (1) the alternative whereby he 8Dof waters in the two reservoirs differed because the

two widely separated echarge areas were receivingprecipitation f different sotopic omposition ue odifferences in altitude and in the relative influencesof Pacific and Gulf of Mexico air masses, r (2) thealternative whereby the differences n aD betweenthe two waters resulted rom the isotopic evolutionof water in and above the lacustrine Creede Forma-tion sediments hrough vaporation odified y dia-genetic hanges. he difference n these wo alter-natives s illustrated chematically n Figure 7.

Present information is not sufficient o decide be-

tween these alternatives or even to assure that either

is valid. Possibly, aspects of both alternatives wereimportant. What does seem certain, however, isthat the bulk of the quartz was deposited rom fluidsentering the vein system rom the mountainous reato the north in the vicinity of the San Luis Peakcaldera, whereas the sphalerite, llite, and chloritewere formed from fluids which entered he vein sys-tem from the moat area of the Creede caldera. The

carbonate luids must have had a deeper source andlikely represent a substantial magmatic contributionto the hydrothermal system. In our continuingstudies, we will integrate he stable sotope data withtemperature, salinity, and chemical data on fluid in-clusions on a very detailed paragenetic basis. Wehope his will enable us to discuss he history of thehydrothermal fluids in more detail and, perhaps, toconstruct a more precise chemical, sotopic, and hy-drologic model of ore deposition.

In summary, he major conclusions f our studymay be listed as follows:1. The isotope data indicate that different fluids

having different sources and/or wall-rock exchangehistories were involved n the deposition f differentgenerations f minerals. Thus, detailed luid-inclu-sion, stable sotope, and chemical studies hold tre-mendous potential for developing chemical-hydro-logic 1nodels or the Creede hydrothermal system.

Since fluids fronl different sources are likely tohave different chemical parameters, serious restric-tions apply to the use of minerals deposited romdifferent fluids in arguments or chemical equilib-rium, particularly if these arguments require morethan the presence r absence f one or more phases.The equilibrium assemblages ssumed or the OHvein by Barton et al. (1977), however, appear o re-main valid.

2. Direct and indirect determinations of the

xsOH20 f the hvdrothermal luids require that thefluids contained a substantial meteoric componentduring deposition f sphalerite nd quartz and prob-ably also during ormation f chlorite nd llite. Asubstantial magmatic component may have beenpresent during he deposition f carbonates. hedata do not rule out some magmatic contribution othe fluids esponsible or the deposition f all min-eral phases.

3. The cart)on sotope data as well as those oroxygen nd hydrogen uggest hat the carbon n thevarious carbonate fluids was derived from a deep-seated. probably magmatic source.

4. Direct measurements f the aD of the hydro-thermal luids equire hat two meteoric waters wereinvolved n the ore fluids, one having a aD of --55


18/20


per mil or larger, another having a 3D of -95 permil or smaller.

5. The presence of two meteoric waters of dif-ferent 3D values requires hat the Creede hydrother-mal plumbing system had two recharge areas. Thisis supported by topographic and structural recon-

structions f the area during ore deposition Stevenand Eaton, 1975).

6. The differences in 8D between the two meteoric

waters may reflect either climatic differences n therecharge areas for the two reservoirs due to differ-ences n altitude and related orographic factors or

- 4O

-6O

-lOO

-120

QuartzPresenturfaceater

DEEP-SEATED

WATER

.:::::::::::: .....................

Sdente / Rhodochroste

-12 -8 -4 0

t80% o

4 8

- 4o

-6o

o -80o

-lOO

-120

'/ '":..':::-Sphalerite

QuartzPresenturfaceater

DEEP-SEATED

WATER

Siderite / R hodochroste

-16 -12 -8 -4 0 4 8

(b) 80%

Fro. 7. Graphical illustration of alternative ways of producing two isotopically differentmeteoric waters. (a) Original difference n D due to precipitation at differing altitudes and/orfrom different air masses. (b) Originally homogeneous meteoric water undergoing differentisotopic evolution in different reservoirs; water in Lake Creede becomes heavier due to evapora-tion and diagenetic reactions.


19/20


isotopic evolution of the waters in the moat of theCreede caldera due to evaporite and diagenetic pro-cesses, r possibly ome combination f the two.

7. The large range of 8D values of meteoric watersin the Creede district as well as the San Juan Moun-tains as a whole indicates that considerable caution

must be exercised n assuming wide uniformity of8D distribution of meteoric water in the westernUnited States at any time,, uring the Tertiary.

8. The large range of aD values of meteoric wa-ters also indicates that caution must be used in as-

suming hat a large range of aD values n hydrother-mal fluids can be considered evidence of mixing ofmagmatic nd meteoric luids without definite nowl-edge of the paleogeography, limatology, nd historyof rock-water interactions.

Acknowledgments

We wish to express our appreciation o Messrs.B. T. Poxson and T. B. Poxson of the EmperiusMining Company nd to the staff of that companyfor the cooperation nd nterest hey have shown nour studies at Creede, which so far have been con-ducted primarily on their properties. We are alsoindebted o the staff of the Homestake Mining Com-pany, operators f the Bulldog Mountain mine, whohave been most cooperative n helping us extend ourstudies o that vein system; in particular M. M.Roebet has given reely of his time and shared hisknowledge f the geology f the Creede istrict with

us. All our work at Creede has benefited enormouslyfrom our many consultations ith T. A. Steven. Hisadvice on the geologic history and paleogeographyof the Creede district were central to our argumentsconcerning he origin of the differences n the 8D ofthe meteoric waters. Joseph F. W'helan made someof the 380 measurements. Pamela H. Wetlaufer

selected, repared, nd documented ll the carbon-ate samples nd has contributed o both he develop-ment and presentation f the ideas herein presented.Paul B. Barton, Jr. collected many of the samplesanalyzed, elected materials or fluid-inclusion naly-sis, and has been a continuing ource of stimulationand advice. Although we absolve ur colleagues fall blame, we warmly acknowledge heir contribu-tions o this study.

P.M. B.U.S. GEOLOGICAL SURVEY

RESTON, VIRGINIA 22092R. O. R.

U.S. GEOLOGICAL SURVEY

DENVER,COLORADO0225July 7, 1978; Juy 17, 1979

REFERENCES

Barton, P. B., Jr., Bethke, P.M., and Toulmin, P., III,1963, Equilibrium in ore deposits: Symposium on themineralogy of sulfides, InternaC Mineralog. Assoc., 3rdGen. Mtg., Washington, D.C. 1962, Mineralog. Soc.America Spec. Paper 1, p. 171-185.

Barton, P. B., Jr., Bethke, P.M., and Roedder, E., 1977,Environment of ore deposition in the Creede mining dis-

trict, San Juan Mountains, Colorado: Part III. Pro-gress toward interpretation of the chemistry of the ore-forming fluid for the OH vein: ECON. G.oL., v. 72, p.1-24.

Bethke, P.M., and Barton, P. B., Jr., 1971, History offilling of the OH vein, Creede, Colorado [abs.]: ECON.GEOL., V. 66, p. 1265.

Bethke, P.M., Barton, P. B., Jr., Lanphere, M. A., andSteven, T. A., 1976, Environment of ore deposition n theCreede mining district, San Juan Mountains, Colorado: II.Age of mineralization: ECON. G-OL., v. 71, p. 1006-1011.

Bethke, P.M., Barton, P. B., Jr., and Rye, R. O., 1973,Hydrogen, oxygen and sulfur isotopic compositions ofore fluids in the Creede district, Mineral County, Colo-rado [abs.]: ECON. G-o-., v. 68, p. 1205.

Bottinga, Y., 1968, Calculation of fractionation factors forcarbon and oxygen isotopic exchange n the system calcite-

carbon dioxide-water: Jour. Phys. Chemistry, v. 72, p.800-808.Bottinga, Y., and Javoy, M., 1973, Comments on oxygen

isotope geothermometry: Earth Planet. Sci. Letters, v.20, p. 250--265.

Casadevall, T., and Ohmoto, H., 1977, Sunnyside mine,Eureka mining district, San Juan County, Colorado:Geochemistry of gold and base metal ore deposition in avolcanic environment: ECON. GEOL., v. 72, p. 1285-1320.

Clayton, R. N., O'Neil, J. R., and Mayeda, T., 1972, Oxy-gen isotope exchange between quartz and water: Jour.Geophys. Research, v. 77, p. 3057-3067.

Craig, H., 1961, Isotopic variations in meteoric waters:Science, v. 133, p. 1702-1703.

-- 1963, The isotopic geochemistry of water and carbon ingeothermal areas, in Tongiorgi, E., ed., Nuclear geologyon geothermal areas, Spoleto: Pisa, Consiglio Nazionale

della Richerche, Lab. Geologia Nucleare, p. 17-53.Craig, H., Gordon, L., and Horibe, Y., 1963, Isotopic ex-change effects in the evaporation of water: Jour. Geophys.Research, v. 68, p. 5079-5087.

Emmons, W. H., and Larsen, E. S., Jr., 1913, A preliminaryreport on the geology and ore deposits of Creede, Colo-rado: U.S. Geol. Survey Bull. 530, p. 42-65.

1923, Geology and ore deposits of the Creede district,Colorado: U.S. Geol. Survey Bull. 718, 198 p.

Eugster, H., and Hardie, L. A., 1979, Saline lakes, inLerman, A., ed., Physics and chemistry of lakes: NewYork, Springer-Verlag, p. 237-293.

Forester, R. W., and Taylor, H. P., 1972, Oxygen andhydrogen isotope data on the interaction of meteoricground waters with a gabbro-diorite stock, San JuanMountains, Colorado: Internat. Geol. Cong., 24th, Mon-treal 1972, Sec. 10, Geochemistry, p. 254-263.

Hall, W. E., Friedman, I., and Nash, J. T., 1974, Fluid in-clusion and light stable isotope study of the Climax molyb-denum deposits, Colorado: ECON. GEOL., v. 69, p. 884-901.

Hamilton, W. B., and Myers, W. B., 1966, Cenozoic ectonicsof the western United States: Rev. Geophysics, v. 4, p.509-549.

Hull, D. A., 1970, Geology of the Puzzle vein, Creede min-ing district, Colorado: Unpub. Ph.D. thesis, Univ.Nevada, Reno, 151 p.

Jones, B. F., 1966, Geochemical evolution of closed basinwater in the western Great Basin, in Rau, J. L., ed.,Second symposium on salt: Cleveland, Ohio, Geol. Soc.Northern Ohio, v. 1, p. 181-200.

Kamilli, R. J., and Ohmoto, H., 1977, Paragenesis, zoning,fluid inclusion, and isotopic studies of the Finlandia vein,


20/20


Colqui district, central Peru: Ecoa. GEOL., v. 72, p. 950-982.

Kelly, W. C., and Rye, R. O., 1979, Geologic, fluid inclusionand stable isotope studies of the Panasqueira tin-tungstendeposits, Beira Baixa, Portugal: Econ. Geol., v. 74, p.1721-1822.

Landis, G. P., and Rye, R. O., 1974, Geologic, fluid inclu-sion, and stable isotope studies of the Pasto Bueno tung-sten-base metal ore deposit, northern Peru: Ecoa. GEOL.,

v. 69, p. 1925-1059.Lipman, P. W., Steven, T. A., and Mehnert, H. H., 1970,Volcanic history of the San Juan Mountains, Colorado,as indicated by potassium-argon dating: Geol. Soc.America Bull., v. 81, p. 2329-2352.

Ohmoto, H., 1972, Systematics of sulfur and carbon isotopesin hydrothermal ore deposits: Ecoa. GEo., v. 67, p. 551-578.

Ohmoto, H., and Rye, R. O., 1979, Isotopes of sulfur andcarbon, in Barnes, H. L., Geochemistry of hydrothermalore deposits, 2nd Ed.: New York, John Wiley.

Ohmoto, H., and Shettel, D. L., 1974, Effect of [o.o on thehydrogen and oxygen isotopic compositions of minerals athigh temperatures and pressures [abs.]: Geol. Soc. Amer-ica Abstracts with Programs, v. 6, no. 7, p. 898.

O'Neil, J. R., Clayton, R. N., and Mayeda, T., 1969, Oxygenisotope fractionation in divalent metal carbonates: Jour.Chem. Physics, v. 51, p. 5547-5558.

Ratt, J. C., and Steven, T. A., 1967, Ash flows and relatedvolcanic rocks associated with the Creede caldera, SanJuan Mountains, Colorado: U. S. Geol. Survey Prof.Paper 524-H, 58 p.

Roedder, E., 1965, Evidence from fluid inclusions as to thenature of the ore-forming fluids, in Symposium on prob-lems of postmagmatic re deposition, Prague 1963: Prague,Czechoslovakia Geol. Survey, v. 2, p. 375-384.

1977, Changes in ore fluid with time, from fluid in-clusion studies at Creede, Colorado, in Proc. Internat.Assoc. Genesis of Ore Deposits, 4th Symposium, Varna,Bulgaria, 1974: Sofia, Bulgaria, IAGOD, v. 2, p. 179-185.

Rye, R. O., 1966, The carbon, hydrogen and oxygen isotopiccomposition of the hydrothermal fluids responsible for thelead-zinc deposits at Providencia, Zacatecas, Mexico:Ecoa. GEOL., v. 61, p. 1399-1427.

Rye, R. O., and Sawkins, F. J., 1974, Fluid inclusion andstable isotope studies on the Casapalca Ag-Pb-Zn-Cu de-posit, central Andes, Peru: Ecor. GEOL., . 69, p. 181-205.

Sawkins, F. J., and Rye, R. O., 1977, Fluid inclusion andstable isotope studies indicating mixing of magmatic andmeteoric water: Caudalosa silver deposit, central Andes,Peru, in Proc. Internat. Assoc. Genesis of Ore Deposits,4th Symposium, Varna, Bulgaria, 1974: Sofia, Bulgaria,IAGOD, v. 2, p. 110-116.

Sorer, Z., and Gat, J. R., 1975, The isotope composition ofbrines: Effect of the isotopic activity ra tio in salinesolutions: Earth Planet. Sci. Letters, v. 26, p. 179-186.

Steven, T. A., 1968, Critical review of the San Juan pene-plain, southwestern Colorado: U.S. Geol. Survey Prof.Paper 524-1, 19 p.

1969, Possible relation of mineralization to thermalsprings in the Creede district, San Juan Mountains, Colo-rado: A discussion: Ecor. GEo., v. 64, p. 696-698.

Steven, T. A., and Eaton, G. P., 1975, Environment of oredeposition n the Creede mining district, San Juan Moun-tains, Colorado: I. Geologic, hydrologic, and geophysical

setting: Eor. GEo., v. 70, p. 1023-1037.Steven, T. A., and Friedman, I., 1968, The source of traver-tine in the Creede Formation, San Juan Mountains, Colo-rado: U.S. Geol. Survey Prof. Paper 600-B, p. B29-B36.

Steven, T. A., and Lipman, P. W., 1976, Calderas of theSan Juan volcanic field, southwestern Colorado: U.S.Geol. Survey Prof. Paper 958, 35 p.

Steven, T. A., and Ratt, J. C., 1965, Geology and structuralcontrol of ore deposition n the Creede district, San JuanMountains, Colorado: U.S. Geol. Survey Prof. Paper487, 90 p.

Steven, T. A., and Van Loenen, R. E., 1971, Clinoptilolite-bearing tuff beds in the Creede Formation, San JuanMountains, Colorado: U.S. Geol. Survey Prof. Paper750-C, p. C98-C103.

Steven, T. A., Mehnert, H. H., and Obradovich, J. D.,1967, Age of volcanic activity in the San Juan Mountains,Colorado: U.S. Geol. Survey Prof. Paper 575-D, p.D47-D55.

Taylor, H. P., Jr., 1974a, The application of oxygen andhydrogen isotope studies to problems of hydrothermalalteration and ore deposition: Eor. GEo., v. 69, p. 843-883.

-- 1974b, Oxygen and hydrogen sotope evidence or large-scale circulation and interaction between ground watersand igneous intrusions, with particular reference to theSan Juan volcanic field, Colorado, in Holmann, A. W.,Giletti, B. J., Yoder, H. S., Jr., and Yund, R. A., eds.,Geochemical ransport and kinetics: Carnegie Inst. Wash-ington, pub. 634, p. 299-324.

-- 1977, Water/rock interactions and the origin of HO ingranitic batholiths: Geol. Soc. London Jour., v. 133, p.509-558.

Taylor, R. B., 1975, Neogene tectonism in south-centralColorado: Geol. Soc. America Mem. 144, p. 211-226.

Truesdell, A. H., 1974, Oxygen isotope activities and con-centrations n aqueous salt solutions at elevated tempera-tures: Consequences or isotope geochemistry: EarthPlanet. Sci. Letters, v. 23, p. 387-396.

Truesdell, A. H., Nathenson, M., and Rye, R. O., 1977,The effects of subsurface boiling and dilution on the iso-topic compositions of Yellowstone thermal waters: Jour.Geophys. Research, v. 82, p. 3694-3704.

Wetlaufer, P. H., 1977, Geochemistry and mineralogy of thecarbonates of the Creede mining district, Colorado: U. $.Geol. Survey Open-File Rept. 77-706, 134 p.

Bethke and Rye, 1979

Documents