An ISO/16O and D/H Study of Tertiary Hydrothermal Systems in the Southern Half of the Idaho Batholith

An ISO/16O and D/H study of Tertiary hydrothermal systems in the southern half of the Idaho batholith

R E CRISS* H P TAYLOR J Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, California 91125 l, JR. }

ABSTRACT acteristic of deep-level fluid circulation in geothermal systems such as Yellowstone National Park, Wyoming. In such regions, the During Eocene time, 37 to 49 m.y. ago, a series of large major zones of hydrothermal activity seem to be principally asso- hydrothermal systems was developed around a group of epizonal ciated with either (1) the caldera ring zones or (2) the central plu- granite plutons in the Idaho batholith. These systems involved deep tons (resurgent domes). and extensive circulation of fluids derived from low-Sl80 (~-16) and low-SD (— 120) meteoric waters. Water-rock interactions INTRODUCTION occurred at temperatures of 150 to 400 °C, lowering the ,sO/ leO and D/H ratios in the surrounding Mesozoic rocks (tonalite, gra- A variety of geological evidence (Craig, 1963; Taylor, 1968, nodiorite, and granite), such that the feldspar S,80 and biotite SD 1971, 1977; Taylor and Forester, 1971, 1979) and theoretical models values became as low as -8.2 and -176, respectively. These values (Norton and Knight, 1977; Norton and Taylor, 1979) proves that contrast markedly with the primary isotopic compositions of+9.3 ± large-scale circulation of ground waters commonly occurs around 1.5 and -70 ± 5, respectively. Widespread propylitization of the shallow intrusive bodies in the Earth's crust. Stable isotopic studies Mesozoic plutonic rocks accompanied these isotopic exchange have shown that the circulating fluid is dominantly meteoric water effects. Systematic mapping shows that anomalous SD and 6lsO in subaerial regions (Craig and others, 1956) and ocean water in values occur over more than 15,000 km2, indicating the extensive submarine environments (Craig, 1966; Wenner and Taylor, 1973; lateral dimensions of the ancient circulating systems. The former Gregory and Taylor, 1981 ), although fluids of other derivations are zones of intense hydrothermal activity are marked by low-'80 important in some environments (Clayton and others, 1966; White zones, which were mapped in the vicinity of the margins of several and others, 1973). Similar hot fluids are known to be responsible Eocene plutons (for example, at Rocky Bar) and also within a giant for the formation of many ore deposits (O'Neil and Silberman, (5- to 20-km wide, 60- to 40-km diam) ring zone that surrounds the 1974; Taylor, 1973, 1974a; Sheppard and Taylor, 1974; Ohmoto Sawtooth Mountains. The latter anomaly is coincident with the and Rye, 1970, 1974; White, 1974; Bethke and others, 1976). high-permeability ring fracture zone of an Eocene caldera system. Magaritz and Taylor (1976a, 1976b) discovered that, far from Most of the ore deposits in the southern half of the Idaho batholith being limited only to the shallow plutonic environments mentioned are epithermal and mesothermal Au-Ag veins that are located near above, widespread l80 depletions produced by meteoric-hydro- l8 18 the periphery of the low- 0 zones (that is, near the outermost 6 0 thermal activity were common in a number of deeper-seated plu- = 8 isopleth). This association links these deposits with the Tertiary tonic environments within several of the great Mesozoic granitic hydrothermal activity and has great potential as an exploration tool batholiths of the North American Cordillera. Taylor and Magaritz in the heavily forested region. Evidence is presented that the Eocene (1976, 1978) extended these <5,80 and <5D studies to the Idaho ground-water circulation pattern was affected over large lateral dis- batholith and discovered wide zones of strong D and l80 depletion tances (25 to 50 km) and great depths (5 to 7 km). These conclu- produced by hydrothermal circulation systems associated with a sions, together with the indications that large amounts of water group of crosscutting Eocene plutons. Their principal conclusions, 3 (>7,000 km ) were involved in some systems and that the circula- other than demonstrating the existence and large scale 104 km2) tion patterns probably are related to caldera ring structures, may be of the ancient geothermal systems, were that the aqueous solutions of particular importance in geothermal exploration and exploita- were derived from ordinary meteoric waters and that the Eocene tion of analogous modern systems. For example, the "fossil" magmatic activity provided the requisite heat. They also noted that hydrothermal activity mapped in the Idaho batholith may be char- the areas of ancient hydrothermal activity coincided with large

Contribution No. 3575, Publications of the Division of Geological and Planetary Sciences, California Institute of Technology. The Appendix of this paper, which includes all of the isotopic data together with sample descriptions and localities, is available from the authors on request. It may also be obtained by ordering GSA supplementary material 83-4 from Documents Secretary, Geological Society of America, 3300 Penrose Place, P.O. Box 9140, Boulder, Colorado 80301.

•Present address: U.S. Geological Survey, Menlo Park, California 94025.

Geological Society of America Bulletin, v. 94, p. 640-663, 26 figs., 2 tables, May 1983.

640

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 641

regions of rocks with anomalously young K-Ar ages, as mapped in reconnaissance by R. L. Armstrong (1974). Following up the reconnaissance studies of Taylor and Maga- ritz (1976, 1978), the goal of the present investigation was to care- fully map the l80/l60 and D/H distributions in the southern one-half (Atlanta Lobe) of the Idaho batholith. A companion study of the K-Ar age relationships (Criss and others, 1980, 1982) was also undertaken to determine whether any detailed correlations exist among these different isotopic variables. For a number of

Figure 1. Generalized geologic map of Idaho, modified after King and Beikman (1974), Bond (1978), and Rember and Bennett (1979).

EXPLANATION

| Cu | upper Cenozoic undifferentiated

Tertiary intrusive rocks

1111111 lower Tertiary volcanic and sedimentary rocks

Mesozoic intrusive rocks

Mesozoic and Paleozoic undifferentiated

Precambrian undifferentiated

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 642 CR1SS AND TAYLOR

geological and logistic reasons, we believed that the Idaho batholith and involve reaction of the silicate with excess fluorine gas in a was one of the best areas in which to make such a detailed study of vacuum line, purification of the released oxygen, combustion of the these types of processes. oxygen to carbon dioxide, and analysis on a double-collecting mass This paper discusses the geographic distribution of the stable spectrometer. Hydrogen extractions from hydrous silicates were isotopic variations, the implications of these data with respect to the performed both at Caltech and in the stable isotope laboratory of geology of the ancient geothermal systems in Idaho, and their bear- the U.S. Geological Survey at Menlo Park, in a manner similar to ing on the genesis and localization of associated economic minerali- that outlined by Friedman (1953) and Godfrey (1962); this tech- zation. Criss and Taylor's (1978) suggestion that a major Eocene nique involves thermal decomposition of the mineral in a vacuum caldera developed in the Sawtooth Mountains region will be exam- line, reduction of the released water to H2 with hot uranium metal, ined in detail, and it will be shown that the present-day Yellowstone and mass spectrometer analysis. All analyses are reported as per mil geothermal area is a close analogue to the Sawtooth hydrothermal deviations from the SMOW standard (Craig, 1961a). More than system. The stable isotopic data from the eroded terrane of the 400 5lsO and 75 <5D measurements are discussed in this text, includ- Idaho batholith provide information on the otherwise inaccessible ing the earlier data of Taylor and Magaritz (1976, 1978), which are deepest levels of modern geothermal systems associated with large incorporated in all figures. Complete data tables, including detailed silicic volcanic centers. Such information currently cannot be descriptions and exact localities for 238 samples, are given in the obtained from modern systems because of the difficulties associated Appendix, which is available on request from the authors and with deep drilling in such hot terranes. which is also stored in the data bank of the Geological Society of America (see footnote on p. 640). GEOLOGIC SETTING ISOTOPIC RELATIONSHIPS IN The Idaho batholith is a large (~40,000-km2) composite mass, THE IDAHO BATHOLITH made up of numerous granitic plutons, located in the northern Rocky Mountains of central Idaho and adjacent portions of Mon- Sl80 and 5D measurements allow "normal" (primary mag- tana (Fig. 1). Most of the batholith consists of rather uniform gran- matic) rocks to be distinguished from their hydrothermally altered ite and granodiorite (Larsen and Schmidt, 1958; Ross, 1963), counterparts; this forms the basis for mapping the ancient geother- although rocks of the western margin are predominantly tonalite mal systems. Moderate-temperature interactions of crustal rocks and quartz-diorite (Moore, 1959; Schmidt, 1964). The batholith has with low-l80 fluids such as meteoric waters commonly produce an extended intrusive history ranging from the Jurassic to the dramatic l80 depletions in the rocks and a corresponding l80 Eocene, although the majority of plutons are probably Cretaceous enrichment in the fluid (Craig, 1963). Even small amounts of hot, in age (Anderson, 1952; McDowell and Kulp, 1969; Armstrong and low-SD fluids may produce significant SD lowering in rocks, others, 1977). although the attendant 5D enrichment in the fluid is generally small The east-central portion of the batholith contains most of the because of the low hydrogen content of ordinary rocks (Craig, 1963; major Tertiary intrusives, many of which are themselves of batholi- Taylor, 1977). Observed isotopic changes in rocks are thus indica- thic dimensions (Ross, 1934; Reid, 1963; Bennett, 1980). The most tive of the fluids and processes that produced them, and it will be common and widely recognized lithology is pink granite, although shown below that isotopic data from coexisting minerals in altered Tertiary quartz-diorite and granodiorite are also present, as well as rocks form nonequilibrium arrays that are clearly diagnostic of a variety of porphyritic dikes (Ross, 1934). Several features distin- hydrothermal interactions. guish the shallow-level granite plutons from the older deep-seated

Mesozoic granitoids: the former display crosscutting contacts, mia- i80/i«o Variations rolitic and granophyric textures, high heat-generation values, and Eocene isotopic ages (Ross, 1934; Reid, 1963; Swanberg and Two histograms of <5I80 values, one for quartz and one for Blackwell, 1973; R. L. Armstrong, 1974; Bennett, 1980; Criss, feldspar (Fig. 2), summarize the data on plutonic igneous rocks 1981). These plutons are of paramount importance because their from the 25,000-km2 region of the Atlanta Lobe, although the meas- heat pulse generated a series of giant hydrothermal systems that had urements are not weighted for areal significance. Most of the a profound influence on most of the rocks in the region and also on quartz <5I80 values (mean = 10.2, standard deviation = 1.3) are the distribution of economic mineralization (Taylor and Magaritz, similar to, or slightly higher than, those of average granitic rocks 1976, 1978; Criss and Taylor, 1978, 1979; Bennett, 1980). elsewhere in the world (Taylor, 1968); therefore, most of them are The stratigraphic age of the Tertiary plutons is well established believed to be close to the original primary igneous values. The by the Casto pluton, which cuts Eocene rocks of the Challis Volcan- feldspar 5I80 values vary between -8.2 and +10.7, have a mean of ics (Cater and others, 1973; Axelrod, 1966; R. L. Armstrong, 1974). about +6.0, a standard deviation of 4.0, and are strongly skewed These andesite to rhyolite flows and tuffs are probably comagmatic toward low values, such that the mean is significantly lower than with the Eocene plutons (Hamilton and Myers, 1967) and formerly the most commonly observed value of +8.9. The 5lsO distribution must have covered most of south-central Idaho (Ross, 1962; R. L. for feldspar is thus dramatically different than that for quartz, even Armstrong, 1974). The Challis volcanic rocks and the Eocene dikes though analyses of the two minerals are in most cases from exactly and plutons define a common calc-alkaline trend that is distin- the same rock samples. guishable from the calcic trend of the Mesozoic granitoids (Criss, It is known from a variety of previous studies (see Taylor, l8 l8 1981). 1968) that the "normal" fractionation (A value = S Oqtz - S Ofeid) between quartz and alkali feldspar in plutonic granitic rocks is EXPERIMENTAL METHODS about 0.8 to 1.5 and between quartz and plagioclase is 1.0 to 2.5. These are probably close to the equilibrium values at temperatures The techniques used for <5I80 determination of silicate miner- somewhat below the solidus of the granitic magmas. The A values als are essentially the same as outlined by Taylor and Epstein (1962) of most rocks in the Idaho batholith (Fig. 2) are similar to those of

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 643

—I 1 1 1 1 1 r

QUARTZ

Figure 2. Histograms illustrating the distribution of S,80 measurements of quartz and feldspar from granitic FELDSPAR rocks of the Atlanta Lobe of the Idaho batholith. The strongly skewed feldspar §,80 distribution is due to meteoric- hydrothermal alteration effects.

Rl H ra Rra Rj fa iwsi UL _H3_

-8 -6 -2 10 12

6'e0

ordinary granitoids; that is, they display small (< 2 per mil) quartz- feldspar A values that reflect their high-temperature origin. How- ever, the rocks with low <5I80 feldspar (< 8 per mil) all have abnormally large Aq_f values, in some cases exceeding 14 per mil. If these large fractionations are interpreted as equilibrium features, they require formation of the rock, or subsequent recrystallization, at extremely low temperatures (as low as 20 °C or lower). However, the <5lsO of quartz in these "abnormal" rocks is usually similar to, or only slightly lower than, that of the normal rocks. Note that simple closed-system re-equilibration of a granitic rock under subsolidus conditions would produce a significant lsO enrichment in the quartz; in fact, the quartz would show a greater per mil lsO shift than the corresponding lsO depletion in coexisting feldspar because of the lower modal abundance of quartz (20% to 30%) compared to feldspar (60% to 65%). The above considerations indicate that: (1) most of the Idaho batholith rocks originally had relatively similar initial Sl80 values; (2) after formation, the 8180 values of many feldspars (hence, many of the whole-rock values as well) were markedly lowered, but the quartz Sl80 values were either preserved or only slightly lowered, because of the much greater resistance of quartz to hydrothermal ,80 exchange; and (3) none of the quartz Sl80 values appear to have been increased during subsolidus cooling, as would be required if closed-system ,sO exchange were an important process. These conclusions imply that the rocks with large A values all must have interacted with an external low-'80 reservoir of oxygen; the only plausible reservoir of sufficient size is the meteoric ground waters that fill the fractures and pore spaces of rocks in continental environments. The differential quartz-feldspar l80 exchange systematics dis- 8 0 quartz covered in the Idaho batholith (Fig. 3) provide the most complete and regular patterns yet found for this mineral pair anywhere in the Figure 3. Sl80 relations of coexisting quartz and feldspar in world. The tight cluster of points near the 45° line (termed the granitic rocks of the Atlanta Lobe. Rocks that show little or no "primary cluster") represents rocks with essentially "normal" evidence of hydrothermal alteration plot in the "primary cluster" Ai8OQ_F values of about 2 per mil (O'Neil and Taylor, 1967; Taylor, and have quartz-feldspar fractionations of about 2 per mil (the

1968; Clayton and others, 1972; Blattner and Bird, 1974). The total AQ_f = 2 line). Hydrothermally altered rocks show large non- S180 variation of quartz and feldspar along the 45° line is about 3 equilibrium quartz-feldspar fractionations and plot below the per mil, and this probably represents the primary Sl80 variation primary cluster.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 644 CR1SS AND TAYLOR

within the batholith. Inasmuch as quartz and feldspar together con- stitute more than 90% of most of these rocks, the primary whole- rock 8180 values of the samples can be fairly accurately calculated. Thus, the whole-rock S180 values are constrained to range from about +8.0 to +11.0, with the highest values occurring in the west-central portion of the Atlanta Lobe. These relatively high primary 6,sO values are consistent with those particular granitic magmas having been derived from, or having exchanged with, moderately high l80 metasedimentary or metavolcanic rocks at depth. The points lying below the "primary cluster" in Figure 3 all have large nonequilibrium quartz-feldspar fractionations clearly indicative of hydrothermal alteration. The slope of the envelope that encompasses the altered samples suggests that the feldspar exchange rate is about four times faster than that of quartz; more detailed discussions of these data are presented by Gregory and others (1981). Despite the relative inertness of quartz, it is clear that quartz typically undergoes some lsO depletion during hydrothermal alteration. Note that all of the rocks with low-5l80 quartz values (< +9) also have extremely large ( >6) quartz-feldspar fractionations, indicating that subsolidus hydrothermal alteration is probably responsible for all of the low S,80 quartz values in the Idaho batholith.

Figure 5. Graph of ô,80 values of coexisting quartz and biotite 6 (± chlorite) for granitic rocks from the Atlanta Lobe. Sl80 feldspar Analogous relationships in the Eocene plutons may be used to estimate the primary S180 values of those granites (Fig. 4). Extrap- olation to the AQ_F = 1.4 line indicates that the Rocky Bar plutonic 2 complex has a well-defined primary S^Oquartz"® 10-5, whereas more limited data indicate a value of about 9.5 for the Crags and Saw- tooth batholiths. 0 The quartz-biotite relationships (Fig. 5) have the same general form and explanation as the quartz-feldspar relations just discussed. The most important difference is that the "normal" points -2 constituting the "primary cluster" are concentrated about a 45° line represented by a quartz-biotite fractionation of 6 per mil. The

-4 length of this line segment is again about 3 per mil, and points that fall significantly below the line have been l80-shifted by the hydrothermal fluids. Note that although these biotite concentrates -6 were made as pure as possible, many of these "biotites" contain finely interlayered chlorite (see below).

-8 D/H Variations 6I80 quartz D/ H exchange between hydroxyl minerals and the hydrother- Figure 4. <5I80 values of coexisting quartz mal fluids occurred concurrently with the l80/160 exchange effects and feldspar from several Eocene granite plu- described above. However, because of the small amounts of bound tons of the Idaho batholith region. The hydroxyl water in rocks, the 6D values provide a much more sensi- primary <5180 of these hydrothermally altered tive indicator of small water/rock ratios and/or low intensities of plutons can be estimated by extrapolating the hydrothermal alteration. nearly linear arrays back to the primary The measured SD values in the biotite concentrates range from magmatic line (A = 1.4). SB (Sawtooth batho- -66 to -176 per mil and are more or less continuously distributed lith), CRAGS (Crags pluton), CASTO (Casto between these limits. However, a SD value of about -70 ± 5 charac- pluton), RB ("Rocky Bar"; that is, the Twin terizes all of the petrographically fresh, unaltered rocks from the Springs-Dismal Swamp intrusive complex). western and northwestern parts of the Atlanta Lobe. The 5D values

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 645

Figure 6. Graph of 3D values of biotite (± chlorite) versus the §,80 value of coexisting feldspar in Atlanta Lobe granitic rocks. Most samples show signs of isotopic exchange, from an original composition of about 6D = 18 -65 to -75, 5 Ofeid = + 8 to + 10.8, to lower values. The "inverted L-shaped" distribution of the data is characteristic of hydrothermally altered igneous rocks (Taylor, 1977). This plot includes the earlier data of Taylor and Magaritz (1976, 1978).

-8 -6 -4 -2 0 2 4 6 8 10 12 5 0 feldspar

of about -150 all come from highly altered rocks, many of which are Figure 6 compares the <5180 values of feldspar with the 8D also strongly depleted in 180. The intermediate values characterize values of "biotite" from the same rock. Rocks with normal feldspar rocks that have also been hydrothermally exchanged, but with rela- 6180 values (~ +9) are generally characterized by heavy 5D values tively small amounts of water; these typically show only minor lsO (~-70). This appears to be a typical primary ¿>D value for the Idaho depletions. Note that in most instances the analyzed "biotite" batholith and is consistent with the 5D range of about -50 to -80 separates include intergrown chlorite, especially in the low-SD previously established for most igneous and metamorphic rocks examples. Inasmuch as it was not possible to separate this chlorite that are known to be of deep-seated origin (Sheppard and Epstein, from the analyzed biotite concentrates, the chlorite contributes sig- 1970; Taylor, 1974b) Note that rocks with abnormally low Sl80 nificantly to the D/H ratios of some of the mineral separates, espe- values (< +8) are without exception strongly depleted in deuterium cially considering that chlorite contains about four times more (SD< -120). This result was expected because the hydrogen/oxygen stoichiometric hydrogen than biotite. The analytical data described ratio of rocks is small, and because low -lsO meteoric waters typi- below are not corrected for this effect and hence all refer to biotite- cally have low SD values (Craig, 1961b). Similar relations are exhi- chlorite mixtures (here termed "biotite"). bited in a plot of the 5D of "biotite" versus the Sl80 of "biotite" (see Fig. 11 below). The 5D values of "biotite" systematically decrease as the quartz-feldspar fractionations increase (Fig. 7), indicating progressive alteration and increasing water/rock ratios. Note that the SD value is much more sensitive to low degrees of alteration than either the feldspar Sl80 value or the quartz-feldspar lsO fractionation, but that the opposite is true for strong alteration; in the latter case, the SD values remain almost constant at -150 to -175. The relationships shown in Figures 6 and 7 are similar to those first described by Taylor (1977) and Taylor and Magaritz (1978), and they are effec- tively a result of the low hydrogen contents of rocks as compared to their total oxygen contents. Muscovite-biotite D/H fractionations in the Atlanta Lobe range from +10 to +100 per mil (Fig. 8). The shaded band in Figure 8 represents the "normal" 10 to 20 per mil fractionation between muscovite and biotite found in most igneous rocks; such values appear to represent a close approach to equilibrium and are relatively insensitive to temperature (Taylor and Epstein, 1966; Suzuoki 18 A quartz-feldspar and Epstein, 1976). Two of the samples (S 0 = 3.1, 5.0) have exchanged with large amounts of low-l80 hydrothermal fluids; Figure 7. Graph of the 6D value of biotite (± chlorite) versus note that these strongly l80-shifted samples and the least l80- the A,80 value of coexisting quartz and feldspar for Atlanta Lobe shifted samples all lie near the "equilibrium" band. Samples show- granitic rocks. The isotopic compositions of most samples have ing intermediate alteration or lsO exchange plot away from that been changed from their primary isotopic values of SD =»-70 ± 5 band along a crude (curved?) reaction path, implying that the D/ H

and Aq_f »1.0 to 2.5; some samples exhibit extremely large A exchange rate of "biotite" with H2O (which may predominantly values indicative of gross nonequilibrium produced by exchange represent alteration to chlorite) is much faster than that of coexist- with low-l80 hydrothermal fluids. This plot includes the earlier ing muscovite with H2O, a conclusion also reached by Magaritz and data of Taylor and Magaritz (1976, 1978). Taylor (1976a).

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 646 CR1SS AND TAYLOR

General Pétrographie Features of Altered Rocks

Several minéralogie and pétrographie changes attend the hydrothermal alteration of the granitic rocks. Most of the alteration assemblages are of the propylitic type, which includes, in addition to several textural changes, the development of abundant chlorite, sericite, and smaller amounts of calcite, epidote, sphene, opaques, and/or other secondary minerals. Similar assemblages are currently being formed at 250 to 325 °C in feldspathic sandstones of the Salton Sea geothermal system, California (McDowell and Elders, 1980). Another important and commonly observed feature is the clouding of feldspar. This process commonly selectively affects plagioclase relative to potassium feldspar, especially the calcic centers of normally zoned grains. Much of this turbidity reflects the growth of fine sericite crystals within the host, but calcite, saussurite, chlorite, fluid inclusions, and perhaps clay minerals are also formed. In some extremely l80-depleted rocks, the alteration is so intense that it can be seen in hand specimen, and the feldspars have a dull, almost chalky appearance. In such cases, the K-feldspar is also -180 strongly clouded, and albite twinning in plagioclase may be absent. -14« -120 -190 -80 In at least some altered rocks, large single crystals of sericite have 8D biotite grown within the plagioclase grains. Figure 8. <5D values of coexisting biotite (± chlorite) and muscovite (or sericite) for Atlanta Lobe granitic rocks. The numbers Chloritization of Biotite beside the data points represent the S180 values of coexisting feldspar. The dashed line indicates a plausible trajectory for this The most conspicuous and quantifiable pétrographie change mineral pair during progressive hydrothermal alteration, assuming observed in the altered rocks is chloritization of mafic minerals, that D/H exchange between muscovite and H2O is much slower principally the biotite. This change is easily seen with the pétro- than for coexisting "biotite." In most samples, the muscovite is graphie microscope, which additionally shows that the conversion primary and relatively abundant, with the notable exception of one to chlorite first proceeds along cleavage planes and grain boundar- sample (Sl80 feldspar = 8.7) that contains coarse sericite (that is, ies of the biotites. Definitive evidence for the chlorite mineralogy is hydrothermal white mica that grew in the presence of the aqueous provided by both chemical and X-ray diffraction methods. Rela- fluids). tions between the K2O and H2O contents of the "biotite" separates used in the K-Ar study of Criss and others (1982) show that all samples conform closely to a simple mixture of an annite- phlogopite solid solution (biotite) with an Fe-Mg chlorite (Fig. 9).

S.0

•4.0- Figure 9. Relationship between K2O

(wt %) and H20 (yumoles HiO/mg sample) of the "biotite" separates analyzed in this study. The linear relationship is e 3.8- thought to represent alteration of part of the biotite, which originally had K2O E a. = 9.0 and H2O =1.6, to chlorite group minerals with a much higher H 0 con- 2.0 2 tent (~5 ¿tmoles/mg) and no potassium. All water contents are slightly lower than predicted for stoichiometric (pure OH) biotite-chlorite mixtures. 1 .0

0.0 5 6

K20(Wt %)

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 647

lie within a few hundred metres of mapped contacts with large Eocene plutons and thus must have been heated to fairly high temperatures. The relatively restricted range of 5I80 values of most of the end-member chlorites (-2 to -8) is somewhat surprising in the light of the very low calculated S180 values of unshifted Eocene meteoric waters in this area (about - 16; see below), and the relatively small (- 1 ± 2) chlorite-water l80/l60 fractionations estimated for moderate temperatures (250 ± 100 °C) by Wenner and Taylor (1971). This suggests that most of the hydrothermal fluids probably were l80-shifted upward to Sl80 values of about -9 to -3 in essentially all of the analyzed localities in the batholith. The consistency of this l80 shift is remarkable, and it implies that over-all water/rock ratios could not have been large. Thus, pristine meteoric water with a <5I80 of-16 was not an important hydrothermal fluid in the Idaho batholith. Figure 11 indicates that whereas the unaltered, high-l80 bio- 4 S 6 tites have normal SD values of -70 ± 5 and 5l8Oa=+4.5 to +6 (Fig. K20 (Wt %) 5), the low-'80 chlorite end member has a <5D value of approxi- ls Figure 10. Relationship between <5I80 and K2O for the anal- mately -150 ± 10 and <5 O=-3 to -6 (Fig. 10). Theoretical mixing yzed "biotite" separates. Most samples lie within a narrow band lines are not straight in Figure 11, owing to the high H2O content of chlorite. Most of the data points fall outside the band, indicating representing mixtures between a high (~ 9 wt %) K20, "normal" (4.5 to 6 per mil) 5I80 biotite and a potassium-free, low-180 (-3.5 to-6 that this is not a simple mixing process of two end-members (that is, per mil) chlorite that formed in the presence of meteoric- the biotite itself undergoes some D/ H exchange). The four low-6D hydrothermal fluids. Four samples show anomalous relationships; samples in Figure 11 that fall well outside the curved mixing band l8 these may have been altered mainly at high temperatures above the are the same samples that depart from the S 0/K20 mixing band stability field of chlorite, as they roughly straddle the dashed reference line that extrapolates to 5lsO =-25 at K2O = 0. Inasmuch as this is an impossibly low <5,80 value for chlorite, these four samples cannot represent mixing between a normal-180 biotite and any type of hydrothermal chlorite.

-90-

The consistent shift of the points below the theoretical biotite- -100 chlorite mixing band probably results from the substitution of F" and CI" for OH" and also from partial oxidation, both of which -110- would lower the concentration of structural water in the "biotite." 1 -120 The figure also suggests that no clear-cut change in the Fe/ Mg ratio o occurs during chloritization, but rather that the chlorite inherits the CO Fe-rich composition of the biotite (Criss, 1981). -130 The relation between chloritization and hydrothermal alteration is dramatically shown in Figure 10. Whereas primary, unal- -140 tered biotite with about 9 wt % K2O has a normal Sl80 value of 4.5 -150 to 6.6, the low-potassium chlorites formed in the presence of the l8 I8 low- 0 hydrothermal fluids have extrapolated <5 0 values ranging -160 from about -3.5 to -6.0, with a maximum possible range of -2 to -8. SAMPLES WITH All intermediate points on the diagram represent biotite-chlorite -170' KpO 5.1 10 9.1 WT. % mixtures and lie on a linear mixing line (actually a band) between these extreme compositions. Thus, the biotite itself apparently does -160- 5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8ieO biotite not exchange l80/l60 readily with the fluid. The l80 changes in most of the "biotites" seem to be produced by transformation of the Figure 11. Graph showing 5D versus 5I80 values for "biotite" primary biotite to an alteration product (chlorite) that crystallizes (that is, biotite + chlorite) separates from Atlanta Lobe granitic in approximate isotopic equilibrium with the fluid. rocks. Assuming the <5D of chlorite is -150 ± 10, a curved mixing The four low-lsO samples with relatively high K2O contents "band" is also shown. Contours of K2O content in biotite (data (RH 14a, RH 85d, RC 39e, 1 7) represent clear-cut exceptions to the from Criss and others, 1982) are shown for the biotite-chlorite mix- above conclusion. These relatively pure biotite samples may repre- tures. Note that, as would be expected, the six anomalous points sent rocks altered by low-deuterium meteoric fluids at higher with ¿>D < -160 from Figure 10 lie well outside the mixing band; temperatures, for the most part within the stability field of biotite; these samples also have K2O contents that do not conform to the this is plausible, because two of these rocks (RH 14a, RC 39e) contours.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 648 CR1SS AND TAYLOR

in Figure 10. This suggests that <5D of the biotite end member in remaining biotite but rather aggregates of chlorite, sericite, these samples was lowered by the hydrothermal activity. Note that opaques, and sphene. Rocks assigned to the "moderately" and these biotites have lower SD values than the chlorite end member; "strongly" propylitized categories are the most common rocks of this is the expected equilibrium relationship (Taylor and Epstein, the batholith, and they exhibit less extreme development of the 1966; Suzuoki and Epstein, 1976). Note that O'Neil and Kharaka same petrographic features described above. (1976) demonstrated in hydrothermal experiments that D/H Overall, the histograms in Figure 12 show that l80 depletions exchange in clay minerals progresses at a much faster rate than does correlate very well with mineralogical evidence for progressive l8 l6 0/ 0 exchange, similar to the inferred behavior in micas and hydration and alteration of the rocks. The mean 5I80 value of chlorites from Idaho. rocks within each category becomes progressively lower as the "alteration intensity" increases. The range of <5I80 variation within Correlation between Sl80 and Degree of Alteration the "moderate," "strong," and "extreme" categories is very broad, however, and this proves that the oxygen isotopic exchange and Figure 12 is a series of histograms showing the feldspar <5I80 mineralogic hydration effects do not proceed at constant relative distributions in rocks exhibiting different degrees of petrographic rates in all rocks. In particular, it is possible for rocks to exchange ls alteration (see Appendix). The "intensity of alteration" was visually O with fluids at high temperatures (3=400 °C), beyond the stability 18 estimated from the appearance of the biotite and feldspar in thin range of many hydrous phases; many of the low- 0 rocks assigned section; the boundaries between the categories are arbitrary, but to the "moderate" class may have been altered primarily under these care was taken to maximize consistency. In terms of this classifica- higher temperature conditions. For example, two of the four tion, fresh or "weakly" propylitized rocks are uncommon in the deuterium-depleted rocks in Figures 10 and 11 (RH 14a, I 7) have ,8 region, occurring only along the western and northwestern margins relatively low feldspar 5 0 values (+ 1.2 and + 5.0) but show only of the Atlanta Lobe, where there is insignificant development of "moderate" propylitization. Note that although this type of feature chlorite, sericite, and turbid feldspar. The most intensely altered is relatively uncommon in granitic rocks, it is very common in rocks (namely, those assigned to the "extremely propylitized" cate- layered gabbros, where fresh, "unaltered" olivine-pyroxene-plagio- l8 gory) are fairly common in the east-central portion of the batholith; clase assemblages have been depleted in 0 during very high- these all have highly turbid feldspar, considerable sericite, and other temperature hydrothermal exchange (Taylor and Forester, 1979; hydrous secondary minerals, and these rocks generally have no

WEAK PROPYLITIZATION

-5

MODERATE

JZU W- JZsMr-

STRON6

£0 -0- JZ§- m -5

EXTREME

ppi pi P | , pi pi7pi | E| 2 4 -5 HOST ROCK 8 s' o feldspar Figure 13. Graph comparing the <5180 values of Tertiary dike Figure 12. Histograms of feldspar ôl80 analyses (including a rocks in the Atlanta Lobe (mostly whole-rock data) with feldspar few whole-rock measurements) for rocks assigned to different alter- from nearby granitic host rocks. The oldest dikes (quartz- ation categories based solely on petrographic inspection (see text). monzonite porphyries, QMP) are typically much lower in S,80 The <5,80 values tend to decrease with increase in the development than the adjacent host rocks, but younger dikes tend to have <5lsO of hydrous phases and feldspar turbidity, but many reversals and values similar to those of the host granitic rocks and thus plot near irregularities occur. the "1:1" line of equal values.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 649

Gregory and Taylor, 1981). This implies that hydrothermal altera- 4.6) has undergone minor l80 depletion from its probable initial tion in gabbros commonly occurs at higher temperatures than it value of +6. does in granitic rocks. Quartz Veins Tertiary Dike Rocks The 6I80 values of vein quartz in the batholith vary by more than <5I80 analyses of several lamprophyre, rhyolite, and porphyry 20 per mil (Table 1). Most veins are located just outside the geo- dikes provide important information about the timing of the graphic regions where low-180 host rocks occur. The vein quartz is hydrothermal activity in the Atlanta Lobe. At several localities, the generally high in 180, especially near the Au-Ag mines; this would S,sO values of these dikes are either similar to, or less than, the imply deposition at relatively low temperatures (50 to 150 ° C) if the Sl80 values of the host granitic rocks from the same outcrops (Fig. 5I80 value of the water was -16 (that is, pristine meteoric water). 13), as was observed by Magaritz and Taylor (1976b) for Tertiary Somewhat higher temperatures (to >400 °C) are more likely, dike rocks crosscutting the Coast Range batholith in British because appreciable lsO enrichment of the water (to Sl80 values of Columbia. This result indicates that none of the analyzed dikes -9 to -3) commonly occurred through interaction with the host completely postdates the hydrothermal activity. The fractures rocks. The relatively high Sl80 values of the granitic host rocks occupied by the dikes were probably the zones of influx of hydro- imply that the W/R ratios in most cases were low, so appreciable thermal fluids and in general the influx apparently continued even l80 shift of the water probably occurred. If all the quartz veins in after dike emplacement. Table 1 were deposited at about 250 °C, either the waters must have ls In cases where the Sl80 values of the dikes are significantly undergone a wide range of O shifts (H2O attaining values from lower than the host rock, it is probable that: (1) the fine grain size of -15 to +5.6) or else the fluids were formed by mixing of waters of the dikes renders them more susceptible to exchange, and (2) there different origin. are probably small-scale 180 gradients centered on the dikes and A vein from the low-l80 zone inside the Sawtooth batholith fractures. Similar features have been observed by Magaritz and (RK. 41b) is comparatively much lower in Sl80 (-6.2), which implies Taylor (1976b) and Forester and Taylor (1977). H owever, there are deposition at a minimum temperature of 240 °C, even assuming no also significant differences in the ages and petrographic characteris- l80 shift whatsoever in the water. The extremely low 6I80 value of tics of the Atlanta Lobe dikes, and these parameters appear to the host rock (-3.0) suggests that the W/R ratios at this locality correlate with the 6,80 data. With one significant exception (RH were very high, so the zero lsO-shift condition may have been 123b), most of the dike samples that fall well below the line defining approximated. Note that O'Neil and others (1973), O'Neil and Sil- the locus of equal 5lsO values in both host rock and dike are berman (1974), and Taylor (1973) found a similar situation in the "quartz-monzonite porphyry" dikes, a type that Reid (1963) consid- Bodie District in California and in other western Nevada Au-Ag ers to predate the emplacement of the Sawtooth batholith (Fig. 13). districts; the quartz veins in those deposits were commonly formed In contrast, almost all dikes that lie near the 1:1 line of Figure in equilibrium with unshifted meteoric water of the region. 13 are later-stage rhyolite and rhyolite porphyry, lamprophyre, On the other hand, quartz gangue in the Wood River District andesite, or diabase (latest), all of which are known to postdate immediately east of the Idaho batholith has a Sl80 value of +16.4 emplacement of the Sawtooth batholith (Reid, 1963). This observa- (Hall and others, 1978). This suggests either a nonmeteoric water, tion represents good evidence that the hydrothermal activity was which is very unlikely considering the fluid-inclusion <5D values of initiated shortly after intrusion of the large Eocene batholiths, such -110 to -120, or else that the 6,80 value of the water at this locality that the older dikes were altered for longer times and initially at had completely shifted to be in equilibrium with the fine-grained, higher temperatures than the younger dikes, and, secondly, that the high-'80 argillite country rock. Fluid-inclusion measurements indi- duration of the hydrothermal event must have been protracted. cate that the deposition temperature of this deposit was high (about Note that even the youngest analyzed dike (diabase RH 43c; 5lsO = 270 °C; Hall and others, 1978).

TABLE I. CALCULATED TEMPERATURES OF QUARTZ VEINS FOR DIFFERENT FLUID Sl80 VALUES

Sample ô'sO <5I80 Temperature (°C) for indicated fluid 6 l80 values*' no. quartz vein host rock -16 -12 -8 -4 0 T-R-S+ (unshifted)

RK 12b 0.7 -0.4Wr 134° 183° 254° 372° 782° 8-8-27

RK 20e 4.1 104° 141° 192° 268° 399° 8-8-20

RK 41b -6.2 -3.0 229° 331° 560° 10-12-15

RK 43d 14.6 8.4 41° 61° 86° 118° 160° 6-13-24 T-R-S^ = township, range, section **Quartz-water fractionation factors from Matsuhisa and others (1979)

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 650 CR1SS AND TAYLOR

TABLE 2. CALCULATED öl80 VALUES OF METEOR1C-HYDROTHERMAL FLUIDS

Mineral Sample ôValue Isotopic composition of fluid in equilibrium with mineral at indicated temperatures* no. 150 °C 200 °C 250 °C 300 °C 350 °C

Feldspar RH 18 <5I80 = -8.2 -20.9 -17.7 -15.3 -13.6 -12.2 Vein quartz RK 41b (5I80 = -6.2 -21.3 -17.7 -15.0 -13.0 -11.4

Biotite YAG-1033 SD = -176 -69 -91 -107 -118 -127 Muscovite I 7 ÔD = -156 -63 -86 -102 -114 -123

•Based on fractionation factors from O'Neil and Taylor (1967), Matsuhisa and others (1979), and Suzuoki and Epstein (1976).

Temperature, Isotopic Composition, and Origin of the ore fluids at Butte, Montana (Sheppard and Taylor, 1974). This Hydrothermal Fluids in the Atlanta Lobe suggests that meteoric ground waters probably were isotopically similar over a broad region in Idaho and western Montana Theoretically, both the temperature and isotopic composition throughout the early Tertiary. of a hydrothermal fluid can be calculated from isotopic measure- The above calculated 5D value is also in substantial agreement ments of two coexisting minerals in equilibrium with that fluid. with fluid-inclusion measurements of -110 to -120 per mil in vein Unfortunately, the situation in the Idaho batholith is complicated minerals from the nearby Wood River District, which, incidentally, by the different rates of exchange between the fluid and the pre- had homogenization temperatures of 244 to 307 °C (Hall and oth- existing plutonic minerals, and also by the variable proportions of ers, 1978). This SD value is also similar to, but slightly heavier than, water and rock involved in the alteration. Thus, in most cases, the modern meteoric water in Idaho (SD = -130 to -155; Hall and oth- minerals are probably not in equilibrium with the fluid, and in ers, 1978), consistent with both (1) a slightly warmer climate in the many of the cases where equilibrium was attained (such as with feldspar, or in the veins) the Sl80 value of the fluid probably was strongly shifted from its original isotopic composition. The above difficulties make it impossible to carry out conven- tional oxygen isotope geothermometry of coexisting mineral pairs. The approach used here, which is decidedly inferior to the ideal situation, is to first compute possible 8 i80 and 5D values of coexisting fluids at different temperatures, on the basis of the <5I80 and 5D values of the most isotopicallv extreme l80- and D-depleted samples, and then to look for some systematics or consistency in the results. Table 2 illustrates such calculations for the lowest-l80 feldspar (-8.2) and vein quartz (-6.2) and for the lowest-SD biotite (-176) and muscovite (-156). The lowest -l80 plutonic quartz and biotite values were not used for this purpose because of extremely low exchange rates and interlayering of biotite with chlorite. The Sl80 values of fluid calculated from the low-'80 feldspars (Table 2) are in good agreement with those calculated from the vein quartz, and the SD values of the fluid calculated from biotite are similar to those calculated from muscovite, regardless of temperature. The agreement is encouraging, but all of these values still represent only possible fluid compositions, because the temperature is not known. However, the band of calculated waters from Table 2 intersects the meteoric water line at a Sl80 value of about -15 and a 5D value of about -110 (Fig. 14). This is the only possible unshifted meteoric water that could have been involved in exchange with these particular minerals. The estimated temperature of about 260 °C at the "crossover" point is a minimum value; if these waters had Sl80(%o) previously undergone a small lsO shift, the temperatures would Figure 14. S180 and <5D values of waters in equilibrium with have to be higher (300 to 350 °C) and <5D would then be as low as the most ,80- and deuterium-depleted samples so far obtained from -115 to -125 (Fig. 14). However, the 260 °C value is reasonable and the Idaho batholith, calculated at various plausible temperatures of is compatible with the temperature information obtained from the alteration (ISO to 350 °C). The calculated band for these highly K-Ar data (see Fig. 23 below). Also, the 6D value of the calculated altered samples intersects the meteoric water line at approximately water is reasonable and is isotopically identical to the unshifted S,80 = -15, SD = -110 per mil, our best present estimates for pris- meteoric water that was the source of the early Tertiary main-stage tine, unexchanged Eocene meteoric waters in southern Idaho.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 651

Figure 15. Contour map of Sl80 values of feldspar (dots; including a few whole-rock analyses) from granitic rocks of the Atlanta Lobe. The Eocene plutons are shown in pattern: CP (Crags pluton), CB (Casto batholith), SB (Sawtooth batholith), RB (Rocky Bar complex). The low- zones are centered on these intrusions and are designated as: CRZ (Casto Ring Zone), SRZ (Sawtooth Ring Zone), RBRZ (Rocky Bar Ring Zone), SM (Soldier Mountains anomaly), AR (Anderson Ranch anomaly), LP (Lucky Peak anomaly), B (Banner anomaly), BB (Boise Basin anomaly), PHB (Pearl-Horseshoe Bend anomaly). Note that low-l80 rocks are not present in the west- central portion of the Atlanta Lobe. This figure includes the earlier data of Taylor and Magaritz (1976, 1978).

region about 40 m.y. ago, in agreement with the evidence for highest 5I80 feldspars (> + 10) in this "normal" zone are also asso- warmer world-wide temperatures at intermediate and high latitudes ciated with the highest Sl80 values in plutonic quartz + 12), during the Eocene (Savin, 1977) and (2) the evidence for a some- which proves that the +10 5I80 contour reflects a primary 5I80 what lower elevation of the batholith region at this time (Axelrod, characteristic of the batholith. In the east-central part of the batho- 1966). lith, immediately east of this extensive "normal" ,80 region, are several conspicuous areas of anomalously low 5lsO values. It is GEOGRAPHIC DISTRIBUTION OF THE clear that the largest anomalies (CRZ, SRZ, RBRZ) are spatially HYDROTHERMAL SYSTEMS associated with the Eocene batholiths, notably the Casto batholith (CB), the Sawtooth batholith and its outliers, and the Twin Springs- Regional patterns of Sl80 and <5D values in the Idaho batho- Dismal Swamp-Steel Mountain-Trinity Mountain (Rocky Bar) lith provide important information on the location, nature, and intrusive complex. Smaller anomalies are associated with the Boise origin of the "fossil" hydrothermal systems. A <5I80 contour map Basin (BB) and Pearl-Horseshoe Bend (PHB) diorite stocks. (Fig. 15) shows that most of the "normal," high-l80 rocks (Sl80 Several other low-'80 zones were found where no Eocene >8) are confined to the west-central portion of the Atlanta Lobe, intrusions had been previously mapped, but where such rocks were and to smaller areas far removed from the Eocene plutons. The conclusively identified during the present study. These are the And-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 652 CR1SS AND TAYLOR

erson Ranch (AR), Soldier Mountain (SM), Lucky Peak (LP), and Banner (B) anomalies. In addition to the association with the Eocene plutons, the ól80 anomalies mentioned above display several other significant geometrical features. All of the low-'80 zones have extremely steep isotopic gradients at their peripheries; locally, these gradients exceed 10 per mil per km of distance. Second, the size of the anomaly is crudely proportional to the size of the associated interior pluton. Thus, the small diorite stocks are associated with small l80 anomalies, whereas the three large Eocene batholiths are each sur- rounded by enormous (500-2,400 km2) low-l80 zones. However, although the outcrop areas of these latter three intrusions are approximately comparable to one another (about 300-500 km2), the Rocky Bar anomaly is much smaller than the other two. Another surprising feature is that the best-studied low-l80 zones appear to occur as annular rings more or less centered on the Eocene plutons. The sharp outer edges of these low-l80 ring zones can be likened in some respects to an isotopic "wall" or "moat" where the 5I80 values change abruptly by several per mil. The best example is the Sawtooth Ring Zone (SRZ) located near the center of Figure 15, but the Rocky Bar Ring Zone (RBRZ) also shows this feature, and with more extensive sampling the Casto pluton might also show such a clear-cut low-l80 ring zone. Thus, a core zone of higher ól80 values exists interior to the low-l80 ring zones in each of these areas. This core zone of higher <5,80 values (lower W/R ratios?) includes the wall rocks and the margins of the Eocene Saw- tooth plutons, but not the interior portions of these plutons. The geometry is probably similar in the Casto region, although the data are scanty. In marked contrast, the interior "high"-l80 zone in the Rocky Bar area is almost totally restricted to the centers of the Eocene intrusions. Incidentally, Figures 19 and 20 below should be consulted for more details about the high-l80 zones interior to the Figure 16. Aeromagnetic contours of Zietz and others (1978) SRZ and RBRZ; some of this information in Figure 15 was general- shown together with the +8 (5lsO contour from Figure 15. Dotted ized to improve clarity. and dot-dashed aeromagnetic contours divide the batholith into A close correspondence exists between positive aeromagnetic regions of low (L), intermediate (blank pattern), and high positive l8 anomalies (Zietz and others, 1978) and the low- 0 zones (Fig. 16). magnetization (H; crosshatched pattern). The strongly altered The highest magnetic anomalies are associated with Eocene gran- zones interior to the + 8 contour, especially the large Eocene plutons ites, which implies that these granites have a larger magnetic suscep- (stippled pattern), correlate very well with the positive aeromag- tibility than do the Mesozoic rocks of the region (Cater and others, netic anomalies. 1973; Tschanz and others, 1974). Measurements by Criss and Champion (1981) generally confirm this inference. Note that the areas a few rocks with 5D values higher than -150 exist interior to positive magnetic anomaly in the Sawtooth region is much larger the most deuterium-depleted rocks of the ring zones; this is analo- than the mapped Eocene plutons and is comparable in size to the gous to the <5I80 relationships described above for these core zones. low-l80 SRZ area mapped in Figure 15. This indicates that there is In contrast to the Sl80 pattern, the 6D values show relatively considerably more Eocene granitic material in the area than has gentle horizontal gradients. Furthermore, discernible D/ H altera- been mapped, or that such rocks are present in the shallow subsur- tion effects are found at very large distances from the Eocene plu- face, as discussed below. Positive magnetic anomalies also exist tons; in fact, most of the Atlanta Lobe is characterized by near other low-'sO zones where no Eocene rocks have been "abnormally" low <5D rocks, as originally pointed out by Taylor and mapped, as in the southeastern portion of the batholith. This inde- Magaritz (1976). This feature results because significant D/H pendently suggests that Eocene rocks exist in these areas, confirm- changes can occur even when W/R ratios are too low to permit any ing the inferences made from the <5I80 data and petrographic perceptible oxygen isotope changes, as discussed above. These peri- studies. pheral effects probably reflect exchange between the rocks and The 5D patterns (Fig. 17) provide several complements as well ground waters that were migrating radially inward to the zones of as contrasts to the ó,80 anomalies. First, there is a good over-all higher temperatures and high W/ R ratios. The large-scale convec- spatial correspondence between the 5I80 and SD patterns, which, tive system(s) clearly caused considerable lateral migration of the of course, could be expected from the systematics displayed in Fig- ground waters, with the effects extending outward at least 25 to 50 ure 6. Thus, the rocks in the west-central portion of the batholith km. These inwardly migrating waters probably would have had have essentially "normal" SD values of -66 to -90, but the values temperatures in excess of 150 °C, simply as a result of the geother- decrease markedly (to less than -150 per mil) in the vicinity of the mal gradient that existed in the region prior to the Eocene plu- Eocene plutons. Note that in both the Rocky Bar and Sawtooth tonism (Criss and others, 1982).

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 653

Figure 17. Contours of ÔD values in biotite (± chlorite) in the Atlanta Lobe. Contours are shown at 20 per mil intervals. The -160 contour is hachured, and although the data are limited, this low-deuterium zone corresponds very well with the SRZ and RBRZ areas from Figure 15. Normal ÔD values of about -70 per mil are found only in the west-central zone. Pronounced D/H alteration effects (<5D < -140) occur only near the Eocene plutons (diagonal lined pattern). For purposes of contouring, in areas where limited ôD data exist (that is, the extreme south), it is assumed that all areas with ôl80 < + 6 (Fig. 15) would also have ôD<-140 (see text).

A- TAYLOR S MAGARITZ, 1976 • - THIS WORK

Figure 18 shows how the K2O contents of the high-quality THE SAWTOOTH WILDERNESS ANOMALY— biotite separates used in the K-Ar study of Criss and others (1982) A POSSIBLE GIANT CAULDRON vary across the Atlanta Lobe. These data reflect the chloritization of biotite, with pure stoichiometric biotite having K20=9.3 wt% The most striking feature of the regional Sl80 map (Fig. 15) is and pure chlorite having K2O = 0 wt%. Most of the "biotites" in the the large low-l80 anomaly, termed the Sawtooth Ring Zone (SRZ), Idaho batholith have been partially altered to chlorite, with rela- associated with the Sawtooth batholith and its outliers. As dis- tively pure biotite being found only on the western and eastern cussed below, the detailed geometry of this anomaly is related to margins of the batholith and in the high-temperature hydrothermal the topography and to the aeromagnetic pattern in the region. contact aureoles immediately surrounding the Eocene plutons These considerations strongly suggest that the low-l80 anomaly is (where temperatures were high enough that biotite was stable even coincident with zones of ring fracturing related to emplacement of at high water/rock ratios). The basic geographic pattern shown by the Eocene plutons. It is suggested that these systematic depletions the K2O contours in Figure 18 is very similar to the 5I80 and <5D in lsO are related to structures produced by cauldron subsidence, maps of Figures 15 and 17. The most highly chloritized samples lie which then were modified by resurgent doming. well outside the Eocene plutons, typically within the highly frac- The Sawtooth Ring Zone is a large elliptical annulus of low- tured and altered ring zones described above. lsO rocks (Figs. 19 and 20), having an outer diameter of about 60

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 654 CR1SS AND TAYLOR

coincident with the wide Sawtooth Valley, and on the northwest it adjoins a northeast-trending ridge that is at least partly fault- bounded. These topographic features, especially Sawtooth Valley, suggest that the SRZ is more easily eroded than adjacent rocks either within or outside the ring. The relation between topography and 5I80 is less pronounced on the western and southern sides, and here it may be partly obscured by more recent faulting. It is important to emphasize that the lowest l80 rocks that make up the ring zone usually occur far outside the contacts of any of the Eocene plutons. This relation is best displayed within the wide (10 km) western portion of the SRZ. The analyses clearly show that the rocks in the western part of the SRZ in this area, those farthest from the Sawtooth plutons, are on the average lower in <5I80 (~0 per mil) than Mesozoic rocks between the SRZ and the Sawtooth batholith (avg Sl80=+4). In previously studied areas of meteoric-hydrothermal alteration (see Taylor, 1971, 1974b), the zones of most intense l80 depletion occur either (1) at the margins of the pluton or "heat engine," or (2) in highly fractured caldera ring zones. The most likely explanation for the SRZ isotopic anomaly is the latter, because the low-l80 rocks occur at an appreciable distance from the large central plutons, and they also appear to coincide with a region of intrusive-related faults. Such faults are likely to have formed in the rigid country rocks as a result of subsidence during eruption of the Challis ash-flow tuffs associated with the Eocene plutons; the faults would have provided high- permeability zones and major conduits facilitating the flow of hydrothermal fluids. There is no other plausible explanation for the increase in Sl80 inward from the SRZ. For example, a model in which the water simply undergoes a shift to higher <5180 values as it migrates radially inward cannot be reconciled either with the abrupt outer boundary of the SRZ or with the low-l80 zones in the cores of some of the Eocene plutons. Figure 18. Map of the southern portion of Idaho batholith, The aeromagnetic data of Zietz and others (1978) provide showing contours of K2O contents (wt %) of high-quality biotite much additional information on the geological character of the concentrates used for K-Ar dating of the granitic rocks (from Criss Sawtooth l80 anomaly (Fig. 16). Note that the large diamond- and others, 1982). Pure, unchloritized biotite (KjO 3=9.0) occurs shaped magnetic anomaly and the associated magnetic low over only (1) in the western part of the batholith and (2) within the zone Sawtooth Valley are comparable in size and position to the anom- of strong heating by the Eocene plutons. The lowest K2O samples alous ól80 region encompassed by the SRZ. Although these mag- (< 6.0) lie within the hachured contour, where the actual K2O con- netic features crudely mirror the topography, the extremely low tents are indicated. susceptibilities of most Mesozoic rocks in the region imply that these anomalies reflect the large susceptibility contrast between km (north-south) by 40 km (east-west). The western portion of the Eocene rock and adjacent Mesozoic rock, rather than merely ring zone forms a broad belt about 10 km wide, approximately topography. The interpretation favored here is that the diamond- twice the width of the eastern and southern portions of the ring, and shaped Sawtooth block is composed largely of magnetic rock, much perhaps three times the width of the northern portion. The SRZ is of which may be Eocene granite in the shallow subsurface, with centered on the Sawtooth batholith and smaller associated plutons large outcrops of this granite being marked by even higher positive immediately to the north and encompasses the entire Sawtooth anomalies. Wilderness Region. A schematic geologic cross section of the Sawtooth Mountains The cores of several plutons interior to the ring zone exhibit (Fig. 21) shows a large Eocene batholith, inferred from the aero- local low-l80 regions, with their margins and adjacent wall rocks magnetic anomaly in the region, the conical top of which has been being higher in l80. This type of l80 geometry was observed within modified by formation of a large (50-km-diam) fault-bounded caul- the Sawtooth batholith and also across the smaller Eocene plutons dron and by subsequent resurgent doming. A thick sequence of to the north (near Stanley Lake and near Grandjean). The contours associated volcanic rocks and caldera fill is inferred to have overlain in Figure 20 have, therefore, been drawn assuming that a similar, the epizonal Eocene batholith; this rock sequence has been removed radially symmetric l80 geometry exists along all the contacts of from the immediate area, but thick exposures remain immediately these plutons, for example, along the unsampled western side of the to the east and it is possible that drilling would reveal such rocks in Sawtooth batholith. Sawtooth Valley. The graph above the cross section somewhat The S,80 pattern and the topography provide abundant evi- schematically shows open-system W/R ratios inferred from <5I80 dence that the SRZ is structurally controlled. The anomalous and SD data along the present level of exposure. Extremely high region has a crude oval or diamond shape. On the east, the zone is W/ R ratios (S 1) are required by several of the low ól80 values (as

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 655

44° 30'

Figure 19. Map of a portion of the area shown in Figure 15 (eastern part of the Atlanta Lobe) covering 1° of latitude and 1° of longitude, showing measured <5I80 values of feldspars from granitic rocks (a few samples represent whole-rock analyses; see appendix). Stippled pattern indicates 5.3 5.6«»*0-6 8.2 • -1.0 Eocene plutons; Tcv = Challis vol- 9.2 80 1.9 « 8.7 canic field. »8.6 • 8.8 1-1 1.3 8 3 _ 8'1 ••3-3 0.1 • 6.1 7.4 # • #-0.4 „„ 8.7 »7.6 8.9 « 9.4 • »8.5 9.2 3? # »-0.4 9.3 8.5 ®7.2 8.0* 97

8A •l8.9 74-3.2 5-6» 5*9 M 7.0* Vli 8.2 • ^ 88 Jt-2.• 6 8 2

43°30' 115°30 10 15 20 115e ' I I KM

low as -7 per mil) near the outer margins of the SRZ. Peripheral to lying the diamond-shaped magnetic anomaly would explain the size the zone of ring fracturing, the hydrothermal fluids must have and shape of the SRZ and would be compatible with intrusion- migrated radially inward, as indicated by the fluid path lines in the related faulting and perhaps with cauldron subsidence, as proposed cross section. It may appear from the W/R graph that the amount by Criss and Taylor (1978). A cauldron model implicitly accounts of radially migrating fluid is not sufficient to account for the high for the annular shape of the low-'80 ring zone along the cauldron W/R ratios along and interior to the ring zone, but it must be ring fracture, as shown by Taylor (1974b) for the Silverton cauldron remembered that the cylindrical geometry requires the flux of these in the San Juan Mountains, Colorado. The model is consistent with fluids to fall off approximately as 1/r, as shown. Note that this 1/r the epizonal character of the Sawtooth batholith and with the thick dependence fully explains the enormous scale of the D/ H alteration pyroclastic units within the Challis volcanic sequence immediately effects in the Atlanta Lobe. About 50% of these inwardly migrating to the east. It is, in fact, very likely that large amounts of rhyolitic fluids probably ascended through the ring-fracture system. Some of volcanic rocks and ash-flow tuff once made up the roof rocks of the the intercaldera hydrothermal activity could reflect deep circulating SRZ area and have since been eroded away. In this model, the fluids that penetrated into the Eocene batholith after its solidifica- Sawtooth batholith would in part represent a resurgent dome tion, but the most intense activity may be due to subsequent inter- emplaced after caldera collapse and ash-flow tuff eruption, analo- caldera circulation systems developed about the resurgent domes. gous to the model of Smith and Bailey (1968) for the Valles caldera. In summary, a large intrusive mass coincident with and under- However, note that more detailed field work is required to estab-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 656 CR1SS AND TAYLOR

44° 30'

Figure 20. Detailed 1° x 1° map of the area around the Sawtooth Ring Zone and the eastern part of the Rocky Bar Ring Zone, showing Eo- cene plutons (dense stippled pattern), sample localities (dots), and S180 contours of the data-points shown in Figure 19. Contours are shown at 1 per mil intervals between +4 and +10; values below +4 are not contoured because the variations are too extreme (they go down to values as low as -7 to -8, and they change dramatically over short distances; see Fig. 19). The SRZ is the large, sharply bounded annulus of low- 180 rocks (§,80 < +4) which extends around the Sawtooth batholith and its outlying plutons. The lighter stippled pattern within the hachured +4 contour outlines the positions of samples with S,80 < +2. The solid black area in the upper right indicates a portion of the Challis volcanic field.

43°30' 115° 30' 0 5 10 15 20 1 I I I I KM

lish the existence of a continuous ring structure completely around menon; note also that the interior zones of low-l80 rocks that are the Sawtooth Mountains. characteristic of the Sawtooth plutons (hydrothermal activity associated with resurgent doming?) are missing from the Rocky Bar THE ROCKY BAR RING ZONE plutonic complex. Although most of the Rocky Bar complex is composed of pink, Another well-studied low-l80 area in the Idaho batholith is miarolitic granite and fine-grained granophyric rock, the southerly that developed around the Twin Springs-Dismal Swamp-Steel Trinity Mountain intrusive body consists of gray biotite-horn- Mountain-Trinity Mountain intrusive complex, here called the blende granodiorite and lacks miarolitic cavities. Furthermore, Rocky Bar Ring Zone (RBRZ). The RBRZ encompasses an area of Figure 16 shows that whereas the other granite intrusions are asso- 20 x 30 km and is composed of a 3-km-wide band of low-'80 rock ciated with a positive magnetic anomaly, the Trinity Mountain that conforms to the irregular boundary of the Rocky Bar intrusive pluton is not. The lsO analyses indicate that the RBRZ extends complex. The RBRZ lies about 1 to 4 km outside the plutonic around the Trinity Mountain intrusive and that the interior of this contact (Figs. 15, 20, and 22). Interior to the low-l80 ring zone, pluton has not been altered by the adjacent Dismal Swamp pluton. both the wall rocks and the pluton margins show only minor l80 These two features represent the only compelling evidence for an depletions, and the core zones of the Eocene plutons have lsO Eocene age for the Trinity Mountain pluton, although the presence contents characteristic of "normal"-180 igneous rocks. In contrast of sphene and hornblende distinguishes it petrographically from to the SRZ, the RBRZ is a smaller-scale, contact-zone pheno- typical Mesozoic igneous rocks in this portion of the Idaho batho-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 657

W/R

DISTANCE FROM CONTACT (KILOMETERS)

DISTANCE Figure 22. Vertical southeast-northwest "cross section" through the Dismal Swamp portion of the Rocky Bar intrusive complex, somewhat schematically showing how the <5I80 values vary with elevation near the intrusive contact. Note that the lowest Sl80 values occur well outside the intrusive contact, and that the formerly productive mines (crosses) lie immediately outboard of the low-,80 zone, as discussed in the text. Triangles represent two points projected from the north to the southern side of the Dismal Swamp pluton, assuming radial symmetry.

Figure 22 illustrates several properties of the RBRZ. As already mentioned, the lowest-180 rocks occur a few kilometres away from the Eocene pluton contact, with the distance perhaps increasing slightly with increasing depth. Note that the vertical l8 Figure 21. The lower figure is an interpretative cross section height of the low- 0 zone is at least one-half of its width, suggest- across the Sawtooth Ring Zone, showing a large (~50-km diam) ing that the original height far exceeded the width and has since cauldron block foundered into a large Eocene batholith. Most of been removed by erosion. Most importantly, the present level of this hypothetical batholith lies below the present level of exposure exposure appears to lie somewhere near the middle of the zone of (wavy line), except for the resurgent domes(?) in the Sawtooth intense hydrothermal activity. At any level below an elevation of 18 Mountains. Arrows schematically illustrate fluid-migration paths. 6,000 ft, which is the elevation of the lowest- 0 samples (-6.9; -8.2), l8 A thick volcanic cover (Challis Volcanics) was formerly present. the S 0 values increase with depth. Thus, the W/ R ratios probably The upper graph shows somewhat generalized open-system water/ decreased with depth below this level, because of a drop in permea- rock ratios across the structure. Relatively impermeable ash-flow bility as the fractures were closed by increased lithostatic pressure. units within the rhyolite section may have provided a partial cap for This relationship is substantiated by the low-elevation data points the deep hydrothermal system, allowing an approach to closed- projected from the north side of the pluton (triangles), and also by system conditions; in such a case, even larger water/rock ratios relationships in surrounding rocks. For example, the low-elevation would be required. samples immediately northwest of the Twin Springs pluton are generally higher in lsO than are typical rocks of the low-'80 zone. However, samples within a V-shaped (in plan) fault block imme- lith. Mutual crosscutting relations between the Trinity Mountain diately northeast of the Twin Springs intrusive are generally low in ls pluton and the other Eocene rocks have not yet been established O despite their low elevation. This relationship indicates that this (E. H. Bennett, personal commun.). fault block has been downdropped, as is shown on the map by Figure 22 shows a "cross section" through the Dismal Swamp Rember and Bennett (1979). intrusion. The great relief in this area allows a significant vertical sampling of these rocks. The data points used in the plot probably IMPLICATIONS REGARDING THE all represent samples from one major fault block. Figure 22 is not a CHARACTERISTICS OF DEEP CIRCULATION true cross section because the horizontal axis actually assumes IN MODERN GEOTHERMAI. SYSTEMS radial symmetry and depicts the distance inward to the irregular pluton contact. Furthermore, although the intrusive contact has The i80/l60 and D/H data obtained in this work provide been assumed to be everywhere vertical, the true attitude is not valuable insight into the nature of circulation in modern geothermal known. systems at deep levels that are presently inaccessible to view. The

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 658 CR1SS AND TAYLOR

most important information provided by the ancient Sawtooth Mountains geothermal system is the large scale of the hydrothermal circulation. The regional D/H map (Fig. 17) proves that the circulation system developed about the Sawtooth Mountains involved an area of more than 8,000 km2. The K-Ar data of Criss and others (1980, 1982) indicate that this meteoric-hydrothermal activity took place 37 to 46 m.y. ago (Fig. 23) and probably penetrated to considerable depths, on the order of 7 km below the Eocene surface. Thus, more than 50,000 km3 of rock in the Sawtooth Mountains were apparently affected by the circulating hydrothermal fluids. Rocks within the stippled area of Figure 23 were probably hydrothermally altered at temperatures above the Ar "blocking temperature" that is presumed to be about 250 ± 50 °C (Dodson, 1973, 1979). Owing to the high temperatures, much of the biotite within this zone has escaped replacement by chlorite, even though it all was thoroughly exchanged at high water/rock ratios (see Fig. 18). This result implies that all of the meteoric-hydrothermal alteration outside the stippled areas in Figure 23 took place at lower temperatures, probably mainly in the range 150 to 250 °C. This zone includes most of the extreme l80 depletions, such as the Sawtooth and Rocky Bar Ring Zones. The amount of fluid that circulated through the Sawtooth hydrothermal system may be estimated from the <5I80 data. If all analyses are equally representative of the 2,500-km2 anomalous ,80 region, and if the former vertical height of this anomalous zone was only 7 km (compared to the ~50-km diam), then this gigantic volume (17,500 km3) of altered rock had an average <5i80 value of +3.2, fully 5.3 per mil lower than its primary value of +8.5. The minimum amount of fluid (with Sl80 initial = -16) required to l8 balance this 0 depletion in the rock can be calculated if the max- 10 o 10KM minimi _J imum possible l80 shift (11 per mil, to SlsO = -5) is allowed in the fluid. This gives (Taylor, 1977): Figure 23. Map of the southern part of the Idaho batholith showing K-Ar ages (m.y.) of granitic rocks (underlined numbers = Wx 11 = 5.3 x R Eocene plutons; others = Mesozoic rocks). Contours are drawn at 5-m.y. intervals. Triangles indicate data from earlier studies (R. L. that is, Armstrong, 1974, 1975; Armstrong and others, 1977; McDowell and Kulp, 1969; Percious and others, 1967) and circles represent the

(W/ R)closed system = 0.48; (W/ R)open system = 0.39. new data of Criss and others (1982). The stippled area indicates the approximate limits of complete Ar loss from the Mesozoic rocks during the Eocene thermal event; this terrane presumably was This calculation requires a minimum of 7,000 km3 of hydro- l8 heated above the Ar-blocking temperature for biotite (» 250 °C) thermal fluid to cause the 0 depletions observed in the Sawtooth during the meteoric-hydrothermal metamorphism described in the area, and it seems certain that the actual amount far exceeded this text. quantity. This volume of fluid could be supplied in 20,000 yr by ordinary rainfall in the 8,000-km2 recharge region even if only 5% of the rain sank deeply into the ground. The 7,000-km3 figure can 50% of the required heat could have been provided by the ordinary, be compared with the 0.1 km3/yr discharge rate estimated by pre-Eocene geothermal gradient. This gradient was probably com- Fournier and others (1976) for the Yellowstone caldera. All other parable to that of typical, modern continental crust (about 30 things being equal, the latter figure requires that the Sawtooth °C/km), so that the rocks and fluid at 7-km depth would have had a Mountains hydrothermal activity persisted for 70,000 yr; this is a temperature of 200 °C before the Eocene magmatic event. This reasonable first-order value but could be too short by more than a aspect of the deep circulation is important, because it shows that factor of 10. not all of the thermal energy of geothermal systems need be pro- It is clear that the relatively small (~400-km2) exposures of vided solely by crystallizing magma, as is commonly assumed. Eocene granitic rock in the Sawtooth Mountains did not have Thus, in the Sawtooth Mountains model, the Tertiary magma enough thermal energy to heat these gigantic volumes of fluid and chamber is considered to have provided the lateral thermal gra- highly altered rock to temperatures of ~ 300 °C. Much of the miss- dients and the driving energy that triggered the geothermal circula- ing heat was provided by the hypothetical, large (>5,000-km3), tion, but only a fraction (=S65%) of the total thermal energy of the Eocene magma chamber at greater depths, as inferred from the fluid. The net result of the geothermal circulation system is for the aeromagnetic data. Furthermore, if the size and depth of the fluid to abstract heat from large bodies of ordinary rock, as well as hydrothermal system are as great as here proposed, then as much as from the sides of the magma chamber, and to redistribute and

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 659

release this energy in the central and upper portions of the system, these may exhibit little or no significant surficial expressions of as in the models discussed by Norton and Knight (1977) and Norton hydrothermal activity if impermeable cap rocks are present. and Taylor (1979). Most of the inward-migrating fluids (>50%) probably as- EVIDENCE FOR A METEORIC-HYDROTHERMAL cended through the ring-fracture system, and much of the intercal- ORIGIN FOR THE ORE DEPOSITS dera hydrothermal activity was probably associated with later resurgent intrusion and doming of the foundered cauldron block. Several important and widely separated ore bodies occur Note that many modern, caldera-related geothermal systems (for within the Idaho batholith. The discussion below will show that example, Yellowstone, Long Valley, Valles; see Fig. 26 below) dis- most of these ore deposits have a direct relation to the Tertiary play similar postresurgence activity and that in fact, the majority of meteoric-hydrothermal activity and that the <5I80 data obtained in hot springs are located either along the ring fractures or along the this study have implications with respect to the locations of mineral- margins of the resurgent domes, forming annular zones of discharge ized areas in the batholith. Criss and Taylor (1978, 1979) demon- analogous to those in the SRZ (see map in Smith and Christiansen, strated an empirical correlation between the 5lsO distribution and 1980). the location of ore deposits. They also proposed that large areas of The evidence presented here for deep, intense circulation in the the batholith have low mineral potential, but that several narrow ring-fracture zones may imply that geothermal reserves exist in irregular zones in the batholith are favorable sites for future young rhyolite plateau regions at depth along caldera ring fractures; exploration. Figure 24 shows the +8 S,80 contour, together with the locations of all mineralized lodes in the southern part of the Idaho batholith that actually produced ore, based in large part on unpub- lished data provided by James A. Noble. The great majority of the localities within the batholith are Au-Ag deposits, commonly with significant but subordinate Pb, Zn, Cu, W, and/or Sb values (Ber- gendahl, 1964). Almost all of the Au-Ag deposits within the batholith are simple fissure veins that occupy shear zones within host granitic rocks. The ore deposits are commonly associated with pre-ore hypabyssal porphyry dikes and syn-ore and post-ore lamprophyre dikes that also cut the host rocks. This proves that the ore deposits are Ter- tiary in age and not directly related to emplacement of the Mesozoic batholith (Anderson, 1951). Most of the veins contain iron and base-metal sulfides, often complex metal-S-As-Sb compounds, and precious metals that may be free or associated with the sulfides, usually contained within a gangue of quartz or altered (usually sericitized) wall rock (Anderson, 1939, 1943, 1947; Ross, 1941). There is abundant evidence in all districts for relatively shallow ore deposition in that the veins and lodes were deposited in open spaces in the rocks; for example, in many districts the gold is associated with late-stage quartz in drusy cavities (Anderson, 1939, 1943). This is considered to be strong evidence for an epithermal origin for many of the veins. Most of the mines are located in the southern and eastern portions of the Atlanta Lobe (Fig. 24), near the zones of Tertiary hydrothermal activity. Note specifically that in the large northwestern sector of the Atlanta Lobe, where no low-'80 rocks were found and where only weak or non-existent deuterium depletions were noted, there is a complete tack of productive mineralized areas. This close spatial correlation between the 5I80 contours and the ore bodies is less conspicuous near Idaho City and in the western Boise Basin near Quartzburg. This might be attributable to the paucity of Figure 24. Map of the southern portion of the Idaho batholith, isotopic data points in these areas. It is worth noting that the 6D showing Eocene plutons (pattern), the +8 per mil S,80 contour that values in the latter mineralized areas are generally lower than in outlines the periphery of the low-,80 zones, and all productive lode surrounding rocks, and it is possible that a more detailed SD and mines (filled circles; large triangles represent large producers). 5lsO study would show good correlations on a smaller scale. There is a distinct correlation between the +8 contour and mineral- Figure 25 shows the relationship between the mine locations ized zones within the batholith. Major districts are labeled. Note and the Si80 contours in the best-studied portion of the batholith, that the large, west-central portion of the batholith has negligible near the Sawtooth Ring Zone and the Rocky Bar anomaly. More productivity; only weak hydrothermal effects operated here, as than 75% of the mines, and probably 95% of the production, occurs indicated by both the deuterium and 180 data (compare with Figs. within a narrow (~5-km-wide) strip, constituting only 30% of the 15 and 17). total area, that straddles the +8 per mil Sl80 contour defining the

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 660 CR1SS AND TAYLOR

44°30 Because the geochemistry of these ore deposits has not been studied in detail, no models of ore deposition will be discussed here. However, the meteoric waters undoubtedly migrated radially inward toward the low-l80 zones, gaining heat and perhaps dissol- ving disseminated metals from large volumes of rock during transit. The reasons for deposition of quartz and metals just outside the major ring structures are not known, but there is no evidence to support the involvement of magmatic water. In this view, the +8 per mil <5I80 contour simply represents the location of those water/- rock ratios, structures, temperatures, and pressures most favorable for ore deposition at the present erosion level. In all likelihood, many ore deposits were also formed at even shallower depths in the overlying cover of ash-flow tuffs, but these have since been eroded away.

CONCLUSIONS

Widespread hydrothermal activity in the southern half of the Idaho batholith is related to a period of intense magmatism and tectonism termed the "Eocene event." This period was characterized by intrusion of epizonal granite batholiths, formation of the ooge- netic Challis volcanic field, block faulting, ring faulting, and ore deposition. The effects of the hydrothermal activity are easily monitored against the relatively uniform primary character of the Mesozoic granitoids, which had feldspar <5lsO values of 9.3 ± 1.5 and biotite <5D values of -70 ± 5. The Eocene meteoric waters in this region had <5180 and ¿>D values of about -16 and -120, respectively, and interaction and exchange between the rocks and deep-circulating fluids derived from these waters produced striking depletions of ,80 and 43°30 deuterium in the rocks, such that the feldspar S180 and "biotite" <5D values became as low as -8.2 and -176, respectively. Propylitization of the rocks accompanied these exchange effects. Figure 25. The prime prospecting area within the region shown ls in Figures 19 and 20 (eastern part of the Atlanta Lobe) is an ~ 5-km- Systematic mapping of the O and D depletions of the rocks wide band that straddles the +8 <5I80 contour. More than 75% of provides new information on the ancient hydrothermal systems in Idaho. The largest well-studied low-l80 anomaly, termed the Saw- the mines (solid dots) and all of the important producers (crosses) 8 lie within the narrow zone. tooth Ring Zone (SRZ), is a 5- to 20-km-wide annulus of low-' 0 rock (avg <5,sO ~2 per mil) that has an outer diameter of 40 to 60 outer perimeter of the former zones of intense hydrothermal circu- km. D/ H effects in the rocks are often discernible more than 25 km lation. Although this should be considered to be only an empirical outside the periphery of the lsO anomaly; the larger size of the relationship at the present time, the outer +8 <5I80 contour coincides deuterium anomaly is attributable to the fact that <5D values of very well with the most productive ore zone in the batholith. This rocks are extremely sensitive to hydrothermal alteration, even at relationship has been accurately established for the Atlanta, Rocky low water/rock ratios (Taylor, 1977). These peripheral zones are Bar, and Banner Districts, which contain the most important depos- clearly part of the recharge area of these gigantic hydrothermal its in this 1° x 1° region, and this is strong evidence for a direct systems. genetic relationship between ore deposition and the hydrothermal The low-l80 ring zone of altered rocks is centered on and ls activity that produced the distribution of 5 O contours in the surrounds the Eocene Sawtooth batholith and its outlying plutons; batholith. The winding strip thus constitutes an obvious "target various aspects of the appearance of the SRZ (size, shape, degree of zone" for future mineral exploration in the batholith. Furthermore, fracturing, intensity of propylytic alteration, extreme 180 and ls detailed <5 O studies probably could refine this target zone much deuterium depletions, and so forth) are similar to features observed more precisely. in younger, less deeply eroded ring-fracture zones of large calderas Note that, in addition to the barren areas that lie far outside of the western United States (Fig. 26). In fact, the scale of the the (strong) hydrothermal activity, large areas within and interior to Sawtooth Ring Zone and the size of the underlying Eocene granitic the zones of extreme l80 depletion also have negligible recorded batholith, inferred from the large aeromagnetic anomaly, is re- production and probably low mineral potential. This statement, markably similar to relationships observed in the Yellowstone however, neglects the current exploration interest in the Mo, U, and rhyolitic volcanic field in Wyoming, formed during the last 2 m.y. Be associated directly with the Eocene granite plutons in these areas (Fig. 26). The Sawtooth Mountains area has the following features and also the potential of black-sand placers (Kiilsgaard and others, in common with those known or inferred for the Yellowstone 1970; F. C. Armstrong, 1957; Bennett, 1980). In any case, the empir- region (see Eaton and others, 1975; Smith and Christiansen, 1980): ical association of Au-Ag mineralization with the +8 <5I80 contour (1) the existence of a major 30- to 50-km-wide, shallow-level silicic apparently applies only to the outside contour, and not to the +8 magma chamber; (2) the formation of a large coeval volcanic field contours interior to the ring zones. that includes several major ash-flow tuff units, each of which pre-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 661

0 10 20 MILES YELLOWSTONE J I— IDAHO NATIONAL BATHOLITH 0 10 20 30 40 KM PARK L I I —I I

Figure 26. Comparison of the low-180 areas of the Sawtooth Ring Zone and the eastern part of the Rocky Bar Ring Zone with the geologic and hydrothermal features associated with caldera development in the Yellowstone rhyolite volcanic field, Wyoming (all at the same scale). In the Yellowstone area, features developed during the past 2 m.y. as a result of multiple ash-flow tuff eruption are the caldera rim, areas of resurgent doming, and the ring-fracture pattern (Smith and Christiansen, 1980; Eaton and others, 1975). The main areas of hydrothermal activity at the present time are also shown (solid black). In Idaho, the positions of the Eocene plutons (resurgent domes?) and l8 I8 18 the annular low- 0 zones (<5 0 feid< +4) are repeated from Figure 20; the solid black areas within the low- 0 ring zones represent areas with <5180 < +2 (these are the areas of most intense Eocene hydrothermal alteration). Note the general similarities between the features on the two maps, particularly the association of hydrothermal alteration with either (1) the ring-fracture zones, or (2) the resurgent domes.

sumably is associated with a caldera-forming event; and (3) intense explain the high thermal and fluid discharge rates observed in some meteoric-hydrothermal activity, with the discharge vents of most modern geothermal systems. hot springs being distributed within a giant 50-km diam) annulus. In addition to the Sawtooth Ring Zone, several other low-l80 This close analogy provides good evidence that the SRZ is coinci- zones have also been mapped in the Atlanta Lobe, and their clear- dent with a high-permeability ring-fracture zone associated with a cut spatial association with the Eocene intrusions provides an very large Eocene caldera and conversely implies that the geometri- important mapping and interpretative tool in the region. Further- cal and thermal properties of the ancient Sawtooth Mountains more, most of the productive mines in the region are located near hydrothermal system are in many ways similar to the deeper parts the periphery of these low-l80 zones (along the "outer" +8 per mil of modern cauldron-related geothermal systems, such as the one contour); this association links these ore deposits with the Tertiary that underlies the Yellowstone volcanic field. hydrothermal activity and has great potential as an exploration Most of the hydrothermal activity in the Idaho batholith tool. apparently occurred at temperatures of approximately 150 to 400 °C and persisted to depths of approximately 7 km below the earth's ACKNOWLEDGMENTS surface. A significant proportion (>1/3) of the heat provided to the circulating fluids was supplied by the ordinary geothermal gradient This paper represents a portion of the doctoral research project in the older rocks, although the driving force for the circulation was of R. E. Criss at the California Institute of Technology. We thank clearly furnished by the thermal energy of the Eocene plutons. Mike Carr, John Coulson, Doug White, and Joop Goris for techni- These concepts and the evidence that the fluid-flow regime may be cal assistance, and R. L. Armstrong, M. Magaritz, J. R. O'Neil, dramatically affected over lateral distances of 25 to 50 km help to L. T. Silver, E. H. Bennett, R. T. Gregory, and D. L. Nelson for

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 662 CR1SS AND TAYLOR

helpful discussions. J. H. Jones, J. Given, and K. M. Criss aided in Institute of Technology, 401 p. the field sampling. Financial support was supplied by National Criss, R. E., and Champion, D. E., 1981, Magnetic properties of plutonic rocks from the Idaho batholith and their implications to the interpreta- Science Foundation Grants EAR-76-21310 and EAR-78-16874, by tion of regional aeromagnetic data [abs.]: EOS (American Geophysical Department of Energy Grant EX-76-G-03-1305, and by the U.S. Union Transactions), v. 62, p. 1056. Geological Survey. Criss, R. E„ and Taylor, H. P., Jr., 1978, Regional l80/l60 and D/H variations in granitic rocks of the southern half of the Idaho batholith and the dimensions of the giant hydrothermal systems associated with REFERENCES CITED emplacement of the Eocene Sawtooth and Rocky Bar plutons: Geolog- ical Society of America Abstracts with Programs, v. 10, p. 384. Anderson, A. L., 1939, Geology and ore deposits of the Atlanta District, 1979, Isotope evidence for the relationship of large-scale Eocene Elmore County, Idaho: Idaho Bureau of Mines and Geology Pamphlet meteoric-hydrothermal systems to mineral deposits and cauldron sub- 49, 71 p. sidence in the Idaho batholith: American Institute of Mining Metallur- 1943, Geology of the gold-bearing lodes of the Rocky Bar District: gical Engineers, Fall Meeting, p. 24. Idaho Bureau of Mines and Geology Pamphlet 65, 39 p. Criss, R. E„ Lanphere, M. A., and Taylor, H. P., Jr., 1980, Effects of 1947, Geology and ore deposits of Boise Basin, Idaho: U.S. Geological regional uplift, doming and meteoric-hydrothermal metamorphism on Survey Bulletin 944C, p. 119-319. K-Ar ages of biotites in granitic rocks of the southern half (Atlanta 1951, Metallogenic epochs in Idaho: Economic Geology, v. 46, lobe) of the Idaho batholith: Geological Society of America Abstracts p. 592-607. with Programs, v. 12, p. 408. 1952, Multiple emplacement of the Idaho batholith: Journal of Geol- 1982, Effects of regional uplift, deformation, and meteoric-hydrother- ogy, v. 60, p. 255-265. mal metamorphism on K-Ar ages of biotites in the southern half of the Armstrong, F. C., 1957, Dismal Swamp placer deposit, Elmore County, Idaho batholith: Journal of Geophysical Research, v. 87, p. 7029-7046. Idaho: U.S. Geological Survey Bulletin 1042-K, p. 383-392. Dodson, M. H., 1973, Closure temperature in cooling geochronological Armstrong, R. L., 1974, Geochronometry of the Eocene volcanic-plutonic and petrological systems: Contributions to Mineralogy and Petrology, episode in Idaho: Northwest Geology, v. 3, p. 1-15. v. 40, p. 259-274. . 1975, The geochronometry of Idaho: Isochron/West, no. 14, 50 p. 1979, Theory of cooling ages, in Jager, E., and Hunziker, J. C., eds., Armstrong, R. L„ Taubeneck, W. H., and Hales, P. O., 1977, Rb-Sr and Lectures in isotope geology: New York, Springer-Verlag, p. 194-202. K-Ar geochronometry of Mesozoic granitic rocks and their Sr isotopic Eaton, G. P., Christiansen, R. L., Iyer, H. M., Pitt, A. M„ Mabey, D. R„ composition, Oregon, Washington, and Idaho: Geological Society of Blank, H. R., Jr., Zeitz, I., and Gettings, M. E., 1975, Magma beneath America Bulletin, v. 88, p. 397-411. Yellowstone National Park: Science, v. 188, p. 787-796. Axelrod, D. I., 1966, Potassium-argon ages of some western Tertiary floras: Forester, R. W„ and Taylor, H. P., 1977, l80/l60, D/H and l3C/l2C American Journal of Science, v. 264, p. 497-506. studies of the Tertiary igneous complex of Skye, Scotland: American Bennett, E. H., 1980, Granitic rocks of Tertiary age in the Idaho batholith Journal of Science, v. 277, p. 136-177. and their relation to mineralization: Economic Geology, v. 75, p. Foumier, R. O., White. D. E., and Truesdell, A. H., 1976, Convective heat 278-288. flow in Yellowstone National Park, in United Nations Symposium on Bergendahl, M. H., 1964, Gold, in Mineral and water resources of Idaho: the Development and Use of Geothermal Resources, 2nd, San Fran- U.S. Geological Survey Special Report No. 1, p. 93-101. cisco, May 1975, Volume 2: Washington, D.C., U.S. Government Bethke, P. M„ Barton, P. B., Jr., Lanphere, M. A., and Steven, T. A., 1976, Printing Office (Lawrence Berkeley Laboratory, University of Califor- Environment of ore deposition in the Creede mining district, San Juan nia), p. 731-739. Mountains, Colorado: Part II. Age of Mineralization: Economic Geol- Friedman, I., 1953, Deuterium content of natural waters and other substan- ogy, v. 71, p. 1006-1011. ces: Geochimica et Cosmochimica Acta, v. 4, p. 89-103. Blattner, P., and Bird, G. W., 1974, Oxygen isotope fractionation between Godfrey, J. D., 1962, The deuterium content of hydrous minerals from the quartz and K-feldspar at 600 °C: Earth and Planetary Science Letters, east-central Sierra Nevada and Yosemite National Park: Geochimica et v. 23, p. 21-27. Cosmochimica Acta, v. 26, p. 1215-1245. Bond, J. G., 1978, Geologic map of Idaho: Idaho Bureau of Mines and Gregory, R. T., and Taylor, H. P., Jr., 1981, An oxygen isotope profile in a Geology. section of Cretaceous oceanic crust, Samail ophiolite, Oman: Evidence Cater, F. W., Pinckney, D. M., Hamilton, W. B„ Parker, R. L., Weldin, R. for <5l80-buffering of the oceans by deep (>5 km) seawater-hydro- D„ Close, T. J., Zilka, N. T„ Leonard, B. F., and Davis, W. E„ 1973, thermal circulation at mid-ocean ridges: Journal of Geophysical Mineral resources of the Idaho primitive area and vicinity, Idaho: U.S. Research, v. 86, p. 2737-2755. Geological Survey Bulletin 1304, 431 p. Gregory, R. T., Criss, R. E., and Taylor, H. P., Jr., 1981, Analytical models Clayton, R. N„ Friedman, I., Graf, D. L„ Mayeda, T. K„ Meents, W. F., of l80 systematics of coexisting minerals in hydrothermally altered and Shimp, N. F., 1966, The origin of saline formation waters: Journal plutonic rocks [abs.]: EOS (American Geophysical Union Transac- of Geophysical Research, v. 71, p. 3869-3882. tions), v. 62, p. 1064. Clayton, R. N„ O'Neil, J. R„ and Mayeda, T. K., 1972, Oxygen isotope Hall, W. E., Rye, R. O., and Doe, B. R„ 1978, Wood River Mining exchange between quartz and water: Journal of Geophysical Research, District, Idaho—Intrusion-related lead-silver deposits derived from v. 77, p. 3057-3067. country rock source: Journal of Research, U.S. Geological Survey, v. 6, Craig, H., 1961a, Standard for reporting concentrations of deuterium and p. 579-592. oxygen-18 in natural waters: Science, v. 133, p. 1833-1834. Hamilton, W„ and Myers, W. B„ 1967, The nature of batholiths: U.S. 1961b, Isotopic variations in meteoric waters: Science, v. 133, Geological Survey Professional Paper 554C, p. 1-30. p. 1702-1703. Kiilsgaard, T. H., Freeman, V. L., and Coffman, J. S., 1970, Mineral 1963, The isotope geochemistry of water and carbon in geothermal Resources of the Sawtooth Primitive Area, Idaho: U.S. Geological areas, in Tongiorgi, E., ed.. Nuclear geology on geothermal areas, Survey Bulletin 1319-D, 174 p. Spoleto: CNR, Lab. Geol. Nucl. Pisa., p. 17-53. King, P. B., and Beikman, H. M., 1974, Geologic map of the United States, 1966, Isotopic composition and origin of the Red Sea and Salton Sea exclusive of Alaska and Hawaii: U.S. Geological Survey. geothermal brines: Science, v. 154, p. 1544-1548. Larsen, E. S., Jr., and Schmidt, R. G., 1958, A reconnaissance of the Idaho Craig, H., Boato, G., and White, D. E., 1956, Isotopic geochemistry of batholith and comparison with the southern California batholith: U.S. thermal waters, in Conference on Nuclear Processes in Geological Set- Geological Survey Bulletin 1070-A, p. 1-33. tings, Proceedings: Second National Academy of Sciences National Magaritz, M., and Taylor, H. P., 1976a, Isotope evidence for meteoric- Research Council Publication 400:29. hydrothermal alteration of plutonic igneous rocks in the Yakutat Bay Criss, R. E„ 1981, An l80/l60, D/H and K-Ar study of the southern half of and Skagway areas, Alaska: Earth and Planetary Science Letters, v. 30, the Idaho batholith [Ph.D. thesis]: Pasadena, California, California p. 179-190.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021 TERTIARY HYDROTHERMAL SYSTEMS, IDAHO BATHOLITH 663

1976b, ,s0/l60 and D/H studies along a 500 km traverse across the OH-bearing minerals and water: Geochimica et Cosmochimica Acta, v. Coast Range batholith and its country rocks, central British Columbia: 40, p. 1229-1240. Canadian Journal of Earth Sciences, v. 13, p. 1514-1536. Swanberg, C. A., and Blackwell, D. D., 1973, Areal distribution and geo- Matsuhisa, Y., Goldsmith, J. R., and Clayton, R. N., 1979, Oxygen isotopic physical significance of heat generation in the Idaho batholith and fractionation in the system quartz-albite-anorthite-water: Geochimica adjacent intrusions in eastern Oregon and western Montana: Geologi- et Cosmochimica Acta, v. 43, p. 1131-1140. cal Society of America Bulletin, v. 84, p. 1261-1282. McDowell, S. D., and Elders, W. A., 1980, Authigenic layer silicate miner- Taylor, H. P., Jr., 1968, The oxygen isotope geochemistry of igneous rocks: als in borehole Elmore 1, Salton Sea geothermal field, California, Contributions to Mineralogy and Petrology, v. 19, p. 1-71. U.S.A.: Contributions to Mineralogy and Petrology, v. 74, p. 293-310. 1971, Oxygen isotope evidence for large-scale interaction between McDowell, S. D., and Kulp, J. L., 1969, Potassium-argon dating of meteoric ground waters and Tertiary granodiorite intrusions, western the Idaho batholith: Geological Society of America Bulletin, v. 80, Cascade Range, Oregon: Journal of Geophysical Research, v. 76, p. 2379-2382. p. 7855-7874. Moore, J. G., 1959, The quartz diorite boundary line in the western United 1973, lsO/160 evidence for meteoric-hydrothermal alteration and ore States: Journal of Geology, v. 67, p. 198-210. deposition in the Tonopah, Comstock Lode, and Goldfield mining dis- Norton, D., and Knight, J., 1977, Transport phenomena in hydrothermal tricts, Nevada: Economic Geology, v. 68, p. 747-764. systems: Cooling plutons: American Journal of Science, v. 277, 1974a, The application of oxygen and hydrogen isotope studies to prob- p. 937-981. lems of hydrothermal alteration and ore deposition: Economic Geol- Norton, D., and Taylor, H. P., Jr., 1979, Quantitative simulation of the ogy, v. 69, p. 843-883. hydrothermal systems of crystallizing magmas on the basis of transport 1974b, Oxygen and hydrogen isotopic evidence for large-scale circula- theory and oxygen isotope data: An analysis of the Skaergaard intru- tion and interactions between ground waters and igneous intrusions, sion: Journal of Petrology, v. 20, p. 421-486. with particular reference to the San Juan volcanic field, Colorado, in Ohmoto, H„ and Rye, R. O., 1970, The Bluebell Mine, British Columbia. I. Hofmann, A. W„ Giletti, B. J., Yoder, H. S., Jr., and Yund, Mineralogy, paragenesis, fluid inclusions, and the isotopes of hydro- R. A., eds., Geochemical transport and kinetics: Carnegie Institute of gen, oxygen and carbon: Economic Geology, v. 65, p. 417-437. Washington Publication 634, 1974, p. 299-324. 1974, Hydrogen and oxygen isotopic compositions of fluid inclusions in 1977, Water/rock interactions and the origin of H2O in granitic batho- the Kuroko deposits, Japan: Economic Geology, v. 69, p. 947-953. liths: Journal of the Geological Society of London, v. 133, p. 509-558. O'Neil, J. R., and Kharaka, Y. K., 1976, Hydrogen and oxygen isotope Taylor, H. P., Jr., and Epstein, S., 1962, Relationship between l80/l60 exchange reactions between clay minerals and water: Geochimica et ratios in coexisting minerals in igneous and metamorphic rocks. Part 1: Cosmochimica Acta, v. 40, p. 241-246. Principles and experimental results: Geological Society of America Bul- O'Neil, J. R., and Silberman, M. L., 1974, Stable isotope relations in epi- letin, v. 73, p. 461-480. thermal Au-Ag deposits: Economic Geology, v. 69, p. 902-909. 1966, Deuterium-hydrogen ratios in coexisting minerals of metamor- O'Neil, J. R., and Taylor, H. P., Jr., 1967, The oxygen isotope and phic and igneous rocks [abs.]: EOS (American Geophysical Union cation exchange chemistry of feldspars: American Mineralogist, v. 52, Transactions), v. 47, p. 213. p. 1414-1437. Taylor, H. P., Jr., and Forester, R. W., 1971, Low-l80 igneous rocks from O'Neil, J. R., Silberman, M. L„ Fabbi, B. P., and Chesterman, C. W., 1973, the intrusive complexes of Skye, Mull, and Ardnamurchan, western Stable isotope and chemical relations during mineralization in Scotland: Journal of Petrology, v. 12, p. 465-497. the Bodie mining district, Mono County, California: Economic 1979, An oxygen and hydrogen isotope study of the Skaergaard intru- Geology, v. 68, p. 765-784. sion and its country rocks: A description of a 55-m.y.-old fossil Percious, J. K., Damon, P. E., and Olson, H. J., 1967, Radiometric dating hydrothermal system: Journal of Petrology, v. 20, p. 355-419. of Idaho batholith porphyries: U.S. Atomic Energy Commission, Taylor, H. P., Jr., and Magaritz, M., 1976, An oxygen and hydrogen iso- Annual Progress Report No. COO-689-76, Appendix A-X. tope study of the Idaho batholith [abs.]: EOS (American Geophysical Reid, R. R., 1963, Reconnaissance geology of the Sawtooth Range: Idaho Union Transactions), v. 57, p. 350. Bureau of Mines and Geology Pamphlet 129, 37 p. 1978, Oxygen and hydrogen isotope studies of the Cordilleran batho- Rember, W. C., and Bennett, E. H„ 1979, Geologic map of the Hailey liths of western North America, in Robinson, B. W., ed.. Stable iso- quadrangle, 1:250,000: Idaho Bureau of Mines and Geology. topes in the earth sciences: New Zealand Department of Scientific and Ross, C. P., 1934, Geology and ore deposits of the Casto quadrangle, Idaho: Industrial Research Bulletin 220, p. 151-173. U.S. Geological Survey Bulletin 854, 135 p. Tschanz, C. M., Kiilsgaard, T. H., Seeland, D. A., VanNoy, R. M., Evans, 1941, The metal and coal mining districts of Idaho with notes on the R. D., Federspiel, F. E., Ridenour, J., Zilka, N. T., Tuckek, E. T., and nonmetallic mineral resources of the state: Idaho Bureau of Mines and McMahon, A. G., 1974, Mineral resources of the eastern part of the Geology Pamphlet 57, 263 p. Sawtooth National Recreation Area, Custer and Blaine Counties, 1962, Stratified rocks in south-central Idaho: Idaho Bureau of Mines Idaho: U.S. Geological Survey Open-File Report, 648 p. and Geology Pamphlet 125, 126 p. Wenner, D. B., and Taylor, H. P., Jr., 1971, Temperatures of serpentiniza- 1963, Modal composition of the Idaho batholith: U.S. Geological Sur- tion of ultramafic rocks based on l80/160 fractionation between coex- vey Professional Paper 475-C, p. C86-C90. isting serpentine and magnetite: Contributions to Mineralogy and Savin, S. M., 1977, The history of the earth's surface temperature during the Petrology, v. 32, p. 165-185. past 100 million years, in Annual Review of Earth and Planetary Sci- 1973, Oxygen and hydrogen isotope studies of the serpentinization of ences, v. 5, p. 319-355. ultramafic rocks in oceanic environments and continental ophiolite Schmidt, D. L., 1964, Reconnaissance petrographic cross section of the complexes: American Journal of Science, v. 273, p. 207-239. Idaho batholith in Adams and Valley Counties, Idaho: U.S. Geological White, D. E., 1974, Diverse origins of hydrothermal ore fluids: Economic Survey Bulletin 1181-G, 50 p. Geology, v. 69, p. 954-973. Sheppard, S.M.F., and Epstein, S„ 1970, D/ H and lsO/ ,60 ratios of min- White, D. E., Barnes, I., and O'Neil, J. R„ 1973, Thermal and mineral erals of possible mantle or lower crustal origin: Earth and Planetary waters of nonmeteoric origin, California Coast Ranges: Geological Science Letters, v. 9, p. 232-239. Society of America Bulletin, v. 84, p. 547-560. Sheppard, S.M.F., and Taylor, H. P., Jr., 1974, Hydrogen and oxygen Zietz, I., Gilbert, F. P., and Kirby, J. R., Jr., 1978, Aeromagnetic map of isotope evidence for the origins of water in the Boulder batholith and Idaho: Color-coded intensities: U.S. Geological Survey Geophysical the Butte ore deposits, Montana: Economic Geology, v. 69, p. 926-946. Investigations Map GP-920. Smith, R. L., and Bailey, R. A., 1968, Resurgent cauldrons: Geological Society of America Memoir 116, p. 613-662. Smith, R. B., and Christiansen, R. L., 1980, Yellowstone Park as a window MANUSCRIPT RECEIVED BY THE SOCIETY SEPTEMBER 17, 1981 on the earth's interior: Scientific American, v. 242, p. 104-117. REVISED MANUSCRIPT RECEIVED MAY 19, 1982 Suzuoki, T., and Epstein, S., 1976, Hydrogen isotope fractionation between MANUSCRIPT ACCEPTED JUNE 9, 1982

Primed in U.S.A.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/94/5/640/3444703/i0016-7606-94-5-640.pdf by guest on 30 September 2021