You are on page 1of 35



faulting in the late Tertiary and continuing to

the present (Hunt and Mabey, 1966). Large struc-
tural relief developed during Artist Drive and
Hypotheses about the origin of sedimentary Furnace Creek ti me forming a hydrologically
borate deposits are based largely on concepts closed basin in which about 1,500 m of sediment
developed from geologic observations first made accumulated. The clasts in the Furnace Creek
early in this century. Many of these interpreta- Formation are mostly derived from the nearby
tions need revision to bring them into agreement Funeral and Black Mountains which lie to the
with the current geologic data base and with northeast and southwest respectively (Figs. 1 and
modern hypotheses concerning saline lacustrine 2).
An understanding of the age of the major bo-
Previous work that includes some of our own on rate deposits in the Furnace Creek Formation,
the origin and diagenesis of borate deposits has their origin and facies relationships, as well as
several problems which need to be resolved: 1) their diagenetic and thermal history, are inte-
age of deposits; 2) source of boron; 3) deposi- grated to provide a geological model of borate
tional systems and associated boron localization deposition in a heliothermal, saline, perennial
mechanisms; 4) mineralogy of primary borate pre- lake.
cipitates; and 5) origin of colemanite deposits.
We utilize data on Death Valley borate deposits
gathered during surface geologic mapping, dril-
ling, and mining of these deposits during the AGE OF DEPOSITS
last two decades. Our study is limited to these
data because few reports on the geology of Death The published ages for the Artist Drive and
Valley borate deposits are available from prior Furnace Creek Formations vary from ~ligocene(?)
exploration or mining operations. This informa- to late Miocene to early Pliocene. This is a
tion is integrated with depositional mcdels for semantic problem based on the arbitrary, past
perennial and ephemeral saline lake deposits assignment of the Clarendonian, Hemphillian, and
(Hardie and others, 1978; Collinson, 1978; Ken- Blancan Ages to the Pliocene. This usage was
dall, 1984; Dean and Fouch, 1983; Fouch and Dean, followed by McAllister (1976, 1973, 1970) in
1982) allowing a preliminary facies analysis of defining the Artist Drive and Furnace Creek For-
the Furnace Creek Formation which hosts borate mations by relating fossil diatoms to the North
deposits in the Death Valley region of Cali- American mammalian chronology. Based on this
fornia. chronology, the uppermost Artist Drive Formation
was Clarendenian (early~liocene). The Furnace
Large borate deposits in the Death Valley Creek Formation was Hemphillian (middle Plio-
region consist mainly of those within the lower cene).
Furnace Creek Formation. Borate deposits higher
in the formation, Recent ephemeral lake deposits The North American mammalian chronology has
and surface efflorescences are very minor in been redefined (Savage and Russell, 1983). The
comparison. The borate deposits are scattered Clarendonian is now middle to late Miocene (-12
within an 8-km-wide area extending about 40 km to -8.4 mybp) and the Hemphillian is now late
northwest from the Lila C Mine to the Harmony Miocene (-8.4-5.3 mybp) or earliest Pliocene.
Borax Works (Figs. 1 and 2). Large-scale borate Potassium-argon dating restricts the Artist Drive
deposition requires several geological factors to from about 8.0 to 6.3 mybp and the Furnace Creek
be in operation simultaneously. Initial concen- from about 6.3 to 5.4 mybp (Fleck, 1970). Com-
tration of boron is typically by hydrothermal parison of radiometry to mammalian fmsils yields
extraction of boron from the country rock with a good agreement between paleontology and radiome-
variable magmatic contribution. Borates have try as summarized in the table below.
relatively high aqueous solubilities. Rapid
evaporation in an arid, hot environment helps
produce boron supersaturation in these borate- early
rich fluids and inhibits borate dissolution. The Pliocene Blancan
process that formed the late Miocene borate de- ....................... . ..
5 3 m y ------------------
posits in the Death Valley region was an intense -5.4 m.y.
one that concentrated boron by a factor of about Hemphi llian Furnace Creek Fm.
10,000 compared to the crustal average of 10 ppm -6.3 m.y.
boron (Krauskopf, 1967). Artist Drive Fm.
late -8.0 m.y.
The Furnace Creek For mation presently under- Miocene --------------8.4 m.y
lies about 1600 sq km, although it was probably
more extensive in the past. Borates underlie
----------- 12.0 m.y .-----------------
less than 1% of this area so boron-concentrating
mechanisms were very localized. The Furnace
Creek Formation is composed mostly of interbedded Source: Fleck (1970); Savage and Russell (1983).
clastic and volcanic rocks. These rocks are
basinal deposits laid down in an elongate north-
west-trending sag that resulted from initiation Both the Artist Drive and Furnace Creek Formation
of the youngest phase of Basin and Range block are Hemphillian. The uncertainties in radiometry

and fossil ages are permissive of a Clarendo- such as boron, lithium, and arsenic (Ellis, 1979)
nian(?) age for the lower Artist Drive Forma- in residual fluids. These elements are also
tion. often strongly concentrated in saline lakes. The
transfer of the elements between these two envi-
The Death Valley area lies in a zone transi- ronments is not as direct as previously supped
tional between the northern and southern portions (starting with Gale, 1913).
of the Basin and Range province f at on, 1980;
Zoback and others, 1981) so affinity with nor- The contention that spring waters are domi-
thern- or southern-style tectonic processes is nantly magmatic is not supported by the hydrogen
unclear. The youngest phase of Basin and Range and oxygen isotope composition or the inert-gas
faulting did not develop before about 10 to 7 content of the water. Equilibration of magmatic
m.y. ago (Stewart, 1978) in the northern Basin fluids with host rocks is rapid so exotic con-
and Range and about 13 m.y. ago in the southern stituents, such as boron, lithium, arsenic, etc.,
Basin and Range (Zoback and others, 1981). The yield no indication of fluid origin (~llisand
age of the Artist Drive and Furnace Creek and the Mahon, 1967). Thermal-spring waters typically
timing of the latest phase of northern faulting have a meteoric deuterium content like that of
are in good agreement. Initial crustal extension local rainfall. The oxygen isotope composition
typically is a sag which is followed by block is heavier. Oxygen isotopes in the meteoric
faulting. The Artist Drive and Furnace Creek water are modified by high-temperature exchange
Formations were deposited in these sags with the with oxygen-rich rocks. These rocks contain
conglomerates in the Furnace Creek suggesting little hydrogen, so the deuterium content is
that block-faulting had begun. These formations relatively unaffected by this exchange process
and their depositional basins were subsequently (Drever, 1982). The dissolved inert gas concen-
broken up by block faulting which has continued tration in thermal-spring waters is consistent
to the Recent. Death Valley is a post-Furnace with atmospheric gases dissolved in the local
Creek example of this process because the bound- meteoric water (Ellis and Mahon, 1977). This
ing faults of this graben truncate both the Ar- quantitative evidence shows that water in hydrc-
tist Drive and the Furnace Creek Formations. thermal systems is mostly meteoric (Ellis, 1979)
rather than magmatic. There is considerable
opportunity for meteoric water to be incorporated
BORON SOURCE with magmatic water (Hem, 1972). The near-sur-
face magmatic-water proportion is typically re-
IGNEOUS SOURCE duced to 5% or less by mixing with meteoric water
(White, 1957). Once outside the magma body,
The extensional tectonic regime and associated magmatic fluids promptly lose their identity
bimodal volcanism in the Death Valley region and (Barnes, 1972).
the Basin and Range province correspond with
ridge-trench interactions at the plate boundary Fluid inclusion analysis indicates that boron-
to the west. Evolution from a subducting bounda- rich fluids from magmatic sources commonly prti-
ry into a transform one coincided with the begin- cipate in the formation of pegmatite veins, or in
ning of crustal extension and wrench faulting the metamorphosis of preexisting rocks. Borax,
between triple-points migrating north and south ulexite, and boron-rich fluids have been identi-
(Zoback and others, 1981; Atwater and Molnar, fied in fluid inclusions of skarn and greissen
1973). Subduction is also associated at some deposits (Yakubova, 1952, 1955; Malinko and
point with the history of other borate terranes others, 1976; Thomas and Bauman, 1980). Boron-
(Peru trench, Himalayan trench, etc.). SuMuct- bearing minerals formed deep within the Geysers-
ing trenches may be a strong primary boron-con- Clear Lake geothermal system are from boron-rich
centrating systems. Boron is concentrated by fluids associated with a nearby magmatic source
adsorption on marine illite (Walker, 1975). This (~ternfeldand others, 1983). Boron-bear ing
suggests one source for boron involving adsorp- minerals in skarn orebodies are formed by post-
tion and concentration of boron in marine illite magmatic processes (~arsukovand Egorov, 1960:
followed by complex interactions at depth after Lisitsyn and Malinko, 1971).
subduction. This interaction could lead to the
re-emergence of the boron as an odd-sized late- Fluid inclusion and deuterium-oxygen isotope
magmatic element in igneous systems (Watanabe, evidence suggest that the amount of aquews boron
1967, in Walker, 1975; Seyfried and others, 1984; that is derived from from magmatic sources may be
Ozol, F377). small. This is because magmatic aqueous boron
can react with the country rock, or form pegma-
Subduction is associated with magmatism and tites rather than routinely making its way undi-
volcanism. Energy is supplied, via this igneous luted to the surface or near surface. Boron,
activity, to shallower rocks and may initiate once combined in silicates such as tourmaline, is
boron remobilization from them rather than di- not readily remobilized at low temperature
rectly supplying boron. As subduction is shut- (Graham, 1957; Sims and Bingham, 1968). Moderate
off, extension, with associated bimodal basalt- to high temperature hydrothermal systems are
rhyolite volcanism and magmatism is initiated capable of extracting significant amounts of
(Stewart, 1978). Compressional and extensional boron from rock (~llisand Mahon, 1967). Some or
terranes have a strong magmatic-volcanic c o m p all magmatic boron may reach shallow depths but
nent which is frequently boron-rich. Late mag- another method of prcducing an additional boron
matic processes strongly concentrate elements source is probably required to prcduce the boron
seen in orebodies. This additional source ap- Fine-grained clastic rocks and marls contain up
pears to be hydrothermal activity. to 300 pprn boron. Tertiary volcaniclastic rocks
have 50 to 700 pprn boron in unstable glass from
The correlation of subduction and a boron-rich which boron is readily mobilized during diagene-
terrane is so strong that it cannot be ignored. sis. The boron in the source rocks of the Death
The subduction system may provide boron or energy Valley region are not available for input into
or both. If the boron concentration in magmatic saline lakes unless remobilized.
fluids is great enough, simple dilution by mete-
oric water axld explain the small magmatic water Remobilization of boron by hot-water leaching
cornpent in boron-rich spring waters. In this of volcanic rocks has been demonstrated in the
case, the boron is largely deep magmatic while laboratory by Ellis and Mahon (1967). They de-
the water carrying it is largely meteoric. The termined that 50 to 80% of the boron in volcanic
subduction system may simply supply energy, rocks is liberated by hot-water leaching before
rather than boron, which drives shallow hydro- the rocks show macroscopic hydrothermal altera-
thermal (meteoric)systems that leach boron from tion. Basalt and andesite liberate boron very
relatively near-surface source rocks. readily. Boron concentrations of 20-30 pprn in
natural hydrothermal fluids could be produced
solely by the reaction of hot water on volcanic
rock and without any magmatic contribution (Ellis
and Mahon, 1967).
Tertiary rocks hosting the borate deposits,
Paleozoic marine rocks, and Precambrian meta- Local heat sources for hydrothermal convection
morphic rocks of the region are volumetrically cwld be provided by the dike swarms, and granit-
large and are potentially rich sources of boron ic plutons common throughout the region. Heating
for hydrothermal leaching (Table 1). Precambrian by compaction could lead to sedimentary hydro-
metamorphic rocks contain about 70 pprn boron. thermal activity and may locally be a more impor-
Precambrian to Paleozoic carbonate and coarser- tant heat source than magmatic heating.
grained sedimentary rocks have relatively low
boron contents ranging from (10 to 20 pprn boron. Cenozoic granitic intrusions into Precambrian
and Paleozoic rocks exposed in the Panamint Range
Table 1. Spectrographic analysis of boron in contain only 10 to 30 pprn boron (~untand Mabey,
rocks from the Death Valley region. 1966). This low concentration may be the re-
sidual boron remaining after hydrothermal leach-
ing rather than the initial boron content. The
CnFCmm boron may have been lost in residual fluids. The
Rcck Unit Cenozoic intrusives show all the petrologic
characteristics of hydrothermal alteration by
meteoric water (Taylor, 1979) described in the
TEmIAKY Death Valley region by Hunt and Mabey (1966):
Artist Drive Formtion 50-100 turbid feldspar; alteration of primary hornblende
Titus Canyon Formation 100-700 to a pseudomorph composed of epidote, calcite and
Furnace Creek Formtion 70-100 magnetite; ubiquitous granophyric or micrographic
Funeral Formtion intergrowths of turbid alkali feldspar and
Spring deposited travertine (10 quartz; and epidote-quartz-sulfide veins crosscut
and calcite veins both the intrusive and host rocks. Other hydro-
thermal effects in the local plutons are oligo-
PAIM)U)IC clase phenocrysts with strongly argillized bor-
limestone/dolomite (10-50 ders in an argillized enclosing gro~mdmass,and
quartzite (10-20 euhedral feldspars altered to sericite and clay
shale 15-300 minerals (Hunt and Mabey, 1966).
altered rock at thrust fault (150
Boron concentrations can be exceptionally high
PFECMBRIAN in geothermal fluids derived from sedimentary
schist and gneiss host rocks (~llis,1979). The geothermal systems
at Ngawa (New zealand) and Larderello (1taly)
as above, subjected to discharge fluids containing 1020 and 780 pprn
Tertiary alteration in boron respectively (Ellis and Mahon, 1977). The
the Amargosa thrust Larderello system occurs in Triassic to Jurassic
conplex mudstone, anhydrite, limestone, and quartzite.
The Kizildere (New Zealand) geothermal system,
limestone/dolomite (10-15 within Miocene and Pliocene sandstone, limestone,
dolomitic shale 20-150 and anhydrite underlain by Paleozoic mica schist,
quartzite 10-200 also has an anomalously high boron content (Ellis
conglomerate 15-20 and Mahon, 1977). The Broadlands (New zealand)
shale or schist 30-300 geothermal system shows increasing boron content
shale 30-200 from west to east. This increase correlates with
diabase 50 an increasing association of thermal water with
hydrothermally altered graywacke and argillite
Smrce: Hunt and Mabey (1966). (Ellis, 1979). Geothermal systems with high

boron contents are typically associated with needed to yield large deposits. Hydrothermal
sedimentary host rock suggesting that much of the leaching alone is typically insufficient based on
boron is being extracted from them rather than comparison of borate deposition, location of
from metamorphic or igneous host rock. subduction-extension zones, and hydrothermal cen-
ters. Many hydrothermally active areas have m
Hydrothermal leaching of local rocks is demon- borate orebodies and often no borate deposits at
strated by the apparent loss of about 50 ppm all.
boron in altered Precambrian metamorphic rocks
below the Amargosa thrust sheet (Table 1). This
metamorphic and thrusting event is middle Miocene Table 2. Location of samples plotted on Figures
to late Piocene (Fleck, 1970; Hunt and Mabey, 3 and 4.
1966) which corresponds to the time of borate
deposition in the Furnace Creek Formation To be
effetive in producing boron for precipitation,
the hydrothermal system needs to vent at the
surface usually as thermal springs. 1) Aetna Spring
2) Arrowhead Hot Spring
There are other, less volumetrically important 3) Baker Soda Spring
sources of boron in emphemeral lake systems. 4) Big Bend Hot Spring
Rettig and others (1980) suggest the major source 5) Calistqa Hot Spring
of boron in Uyuni salar, Boliva is weathering of 6) Cooks Spring
Recent rhyolitic volcanic rocks within its basin 7) Crabtree Hot Spring
Risacher and Eugster (1979) conclude that the 8) Deadshot Spring
boron in spring waters is recycled from earlier 9) Fales Hot Spring
evaporite deposits in the playa complex. 10) Fouts Spring 1
11) Fouts Spring 2
12) Grovers Hot Spring
Thermalsprings in the Furnace Creek Formation 13) Hunt Hof Spring
14) Liskey Hot Spring
Direct evidence for the existence of ancient 15) L m g Valley
boron-bearing thermal springs associated with the 16) Mark West Spring
Furnace Creek borate deposits is generally lack- 17) Mercey Hot Spring
ing. Indirect evidence is strong and includes 18) Napa Rock Scda Spring
the ecology of fossil diatoms that prefer hot 19) One Shot Spring
water and are associated with opalized beds in 20) Orrs Hot Spring
the upper portion of the Furnace Creek Formation 21) Pilger Estates Hot Spring
(McAllister, 1970). These diatomaceous deposits 22) Point Arem Hot Spring
are stratigraphically higher than major borate 23) Salt Spring
deposition. Less conclusive evidence is the 24) Saratqa Spring
close association in time and space of volcanic 25) Seigler Spring
vents and borate-deposition centers suggesting a 26) Skaggs Spring
penecontemporaneous heat source for thermal 27) Scda Spring
springs in the Furnace Creek basin. 28) Sulfur Bank Mine
29) Tuscan Hot Spring
Presently inactive Holocene springs once de- 30) Vichy Spring
posited minor amounts of borate minerals on older 31) Warner Hot Spring
alluvial terraces in the Death Valley region. 32) Wister Mudpots
These springs are younger than the major borate
deposits and their boron content may be derived
in part from remobilization of older borate de-
posits (Mc~llister,1970). Travertine depasits 33) Rhcdes Marsh
and calcite veins in the Funeral Formation con- 34) Scda Lake
tain less than 10 ppm boron (Table 1) and it
appears that the main pulse of boriferous fluids
had ceased before deposition of these rocks.
Boron content in thermal springs of geothermal
systems (activehydrothermal systems) in Cali-
SOME IMPLICATIONS OF A HYD- BORON SWRQI fornia and Nevada apparently is controlled by
temperature (Fig. 3, Table 2). A strong negative
A strong association exists between Cenozoic correlation between boron and temperature results
subduction zones and borate deposits. This is a from thermal waters cooling as they take succes-
consequence of both active and ancient hydrother- sively longer paths through rock, or from a more
mal systems induced by deeper magmatic intrusions complete extraction of boron in an old, and hence
in these zones (Ellis and Mahon, 1977). Hydro- cooler, hydrothermal system. A longer fluid path
thermal systems are not restricted to tectonic allows boron to be extracted from rock for a
belts and any area where hydrothermal systems longer period of time. Active borate4epositing
have surface discharge in an arid climate has springs are low temperature and support the con-
potential for borate deposits but at least some tention that warm springs, rather than hot
subduction-derived (magmat ic) boron is probably springs, contribute much of the borate to these


Death Valley Salt Pan, CA (1) mean annual surface temperature 24

maximm air temperature 57

maximum ground surface temperature 88 soil measuremnt

general summertime ground temperature >50-58 soi1 measurement ;

locally >58%
Heliothermal lakes (2) brine temperature 30-70

Deep Springs Lake, CA (3) mean annual air temperature 12 -

discharge lo0
above local springs
fault-line springs flowing into lake 20
Saline Valley, CA (4) water temperature 38 in "salt lake"
Searles Lake, CA (5) mean annual surface temperature 19.1 at Trona, CA

geothermal gradient 91c/km

bottomhole temperature log 27.5 260mdepth

projected temperature at base of 115

sedimentary fill (1000 m)
Rbdes Marsh, NV (6) spring in center of borate 16 fluid contains
depiting playa 55 ppm boron

Walker Lake, NV (7) oxygen-isotope paleotemperature 27-34 lacustrine limestone

Cmstal sabka (8) m a x i m temperature 40-50 optimal conditions
Rio Alumbria, Arg. (9) water temperature 15 spring depositing
Big S d a Lake, NV (10) sediment/water interface in stratified 13 constant all year
saline lake
water surface temperature 24 summer
water surface temperature 4 winter
Searles Lake, CA (11) 32 summer trona
Searles Lake, CA (12) brine temperature 23-25 at <lo0 ft depth
Kramer (Boron) CA (13) borax to kernite transition 53-63
geothermal gradient 66%/km assumed
Laboratory study (14) ulexite to colernanite >70 at 1 atm(?)
Laboratory study (15) inyoite to colemanite 38 at 500 atm
inyoite to meyerhofferite -- stable @ >98O, 1 atm

Sources: (1) Hunt & others (1966); (2) Sonnenfeld & Hudec (1980); (3) Jones (1965); (4)Hardie (1968);
(5) smith (1979); (6) Jones (1966); (7) Link & others (1985); (8) Borchert (1969); (9) Muessig (1966);
(10)Oremland & Des Marais (1983); (11)Smith & others (1970); (12)Haines (1959); (13)Christ & Garrels
(1959); (14) Kemp (1956); and (15) Inan & others (1973).

Table 3. Temperatures of evaporite deposition and diagenesis.


sedimentary basins (Table 3). Muessig (1966) tion of borates in an alkaline perennial or
found a 15% spring d v i t i n g ulexite, and Jones ephemeral saline lake environment; and 2) preci-
(1966) reported a 16Oc spring in the center of pitation induced by cooling and degassing at the
Rhodes Marsh, a borate-bear ing ephemeral lake, surface of boron-bearing thermal spring dis-
that contains 55 ppm boron. A 3 7 O ~thermal charge. These mechanisms are transitional be-
spring (2.4 ppm B) is feeding a saline lake which cause the source of boron for both is attributed
is depositing ulexite from a cool (12O~) brine to springs (Muessig, 1966). After deposition,
containing 34.8 ppm boron (~isacherand Eugster, borates dissolve slowly upon freshwater dilution
1979). Sedimentary basins with hicjh-temperature of the ephemeral or perennial saline lake (Karaz-
thermal springs tend to contain little borate m w hanov, 1963). This characteristic promotes the
temperature springs are far more likely to con- preservation of the borate depits by increasing
tribute bron to adjacent basins. the likelihood of burial before dissolution
Borate deposition was active over a relatively Spring aprons containing borates may occur in
short time during depition of the Furnace Creek a lake or along its margin. A boron-rich spring
Formation. The short duration of the boron in- may produce a proximal spring apron and may also
flux into the Furnace Creek basin may be related feed a chemical delta in a downstream lake. A
to the short-lived character of geothermal and chemical delta could have great morphological
hydrothermal systems which persist up to 3 m.y. similarity to a spring apron. Spring apron bo-
in the western United States (Silberman and rates are deposited largely by sukaerial cooling
others, 1972: Ellis, 1979; Barker, 1983). In the or degassing of C02 (Gulensoy and Kocakerim,
Death Valley region major borate deposition is 1978). In lakes, borates are depited by mys-
confined to the lower 120 m of the Furnace Creek ical-chemical changes in the fluid during mixing
Formation. Fleck (1970) estimated a mini mum of waters and interaction with the local environ-
depositional rate of 127 cm/1000 yr for the Fur- ment. Many spring fluids are relatively boron-
nace Creek Formation at the north end of the poor. In a closed-hydrologic basin, evaporative
Black Mountains. Sediment 120 m thick would re- concentration of these low-boron waters and other
quire about 100,000 yr for deposition at this brine modifications can occur which are suffi-
rate. Ellis (1979) indicates that, because of cient to cause deposition of borate deposits in
the self-sealing process in geothermal systems, the lake.
major outflow is episodic and occurs for brief
periods (about 1,000 yr) alternating with long
periods (10,000-100,000 yr) of conductive heating Spring Aprons
and minor outflow. Unsealing of the geothermal
system by tectonic activity or hydrothermal ex- Borate deposits from spring water can form
plosions triggers periods of major flow. The immediately about a vent or far from it. Muessig
timing of this mechanism could, on average, pro- (1966), and Norman and Santini (Chapter 4, this
vide, one major boron pulse into the Furnace volume) describe Holocene and active borate-
Creek basin over the brief borate deposition precipitating springs in South America. These
period indicated for the lower Furnace Creek springs produce aprons which are small, with the
Formation. largest containing only a few thousand short tons
of borate. The spring aprons are thin ranging up
Geothermal waters dominated by sodium chloride to a few yards thick and are young. They are
are often charged with carbon dioxide and are composed primarily of ulexite and calcite with
commonly saturated (or supersaturated) with cal- some borax and howlite. Owing to the high carbo-
cium carbnate (Ellis, 1979). The sodium/calcium nate content, lack of chloride and sparse sul-
ratio versus boron content of the spring water phate, Muessig categorized the Death Valley bed-
shows a general positive correlation with geo- ded depits as spring depits. The playa de-
thermal systems in California and Nevada (~ig. posits in South America are spring deposits only
4). This relationship suggests that hydrothermal slightly removed from their immediate source
extraction of boron is accompanied by increasing (Muessig,1966). Elsewhere, the separation can
sodium content in the brine. Calcium-dominated reach many tens of kilometers between the point
waters are rare in geothermal systems (see also that boron-rich fluids reach the surface and the
Waring, 1965). The deposition of calcium borates place where borates are precipitated.
requires either extensive modification of the
brine to a favorable [~a]/[~a] concentration Ephemeral Saline Lakes or Playas
ratio, diagenetic removal of sodium rather than
calcium, or extreme physical wnditions, such as Ephemeral saline lakes (playas) have been
relatively hicjh temperature, low [H~O], etc. subdivided by Hardie and others (1978) into ones
having either a moist or a dry mudflat surface.
The dry mudflat (dry playa) subenvironment of
DEPOSITION OF SEDIMENTARY BORATES ephemeral saline lakes is usually dry both at the
surface and in the shallow subsurface and is
~~ BORATE DEPOSITICN characterized by very minor, secondary, borate
efflorescences. Borate efflorescences of which
The two main mechanisms proposed for borate the Harmony area of Death Valley (McAllister,
deposition in the Death Valley region are: 1) 1970) is an example, are economically insignifi-
intense surface evaporation of dilute boron- cant. In contrast, the moist mudflat subenviron-
bearing solutions which induces direct precipita- ment of ephemeral saline lakes is wet below a

surface crust or is occasionally wet at the sur- 1) alluvial fan, 2) sand flat, 3) dry mud-
face. Borate and other saline minerals are abun- flat, 4) ephemeral saline lake, 5) perennial
dant in the moist-mudlat type. Examples of bo- saline lake, 6) windblown dune field, 7)
rate dewits in moist-mudflat ephemeral lakes perennial stream floodplain, 8) ephemeral
range in size and borate content from small, low- stream fldplain, 9) springs and spring-fed
tonnage deposits in South America (Muessig, 1966) pnds, and 10) shoreline.
to the large, high-tonnage deposit at Searles
Lake, California (Smith, 1979; and Chapter 8, The perennial saline lake subenvironment is
this wlume). the most likely analog for the Furnace Creek
Formation where large borate deposits occur. This
Three criteria for recognizing borate minerals subenvironment has these typical characteristics:
formed in a moist-mudflat ephemeral lake are (1)
the borates are not pseudomorphs; (2) they occur 1) the surface body of brine persists for many
within a restricted sedimentary horizon; and, (3) years (tens to thousands) without drying up;
they are forming at or near the sediment surface. 2) the lake may either be shallow (a few meters)
The borates form by growing in the ephemeral-lake or deep (hundreds of meters);
muds and are strictly a surface or near-surface 3) if deeper than a few meters, the lake is
phenomena (Muessig, 1959). usually stratified (meromictic);
4) substantial perennial inflow is required for
The 35 South American moist-mudlat ephemeral recharge of water and addition of solutes and
lakes described by ~uessig(1966) range up to clastic sediment;
several hundred square km in area and up to 5) perennial springs provide solutes and storm-
several million short tons of borate. All these generated flooding introduces elastics;
ephemeral lakes are moist beneath the surface and 6) evaporation is a primary and continuous pro-
many have perennial lagoons and old shorelines up cess causing surface brine concentration and
to 10 m higher than the mcdern surface. The bo- saline mineral nucleation;
rate is not a desiccation feature from standing 7) the essential feature of stratified perennial
water. Ulexite dominates the borate assemblage, lakes where evaporation exceeds inflow is the
carbonate is sparse or absent, other saline min- continuous repetition of inflow --> evapo-
erals, such as halite and gypsum are common, and, ration --> saline mineral precipitation -->
borate deposits are related to nearby springs sinking of brine and settling of chemical
(Table 4a, b). The Turilari (Argentina) ephemer- sediment.
al lake is analogous to the Kramer borate deposit
in California, except for size, and both deposits The sedimentary features produced by a peren-
crystallized from standing water in a shallow nial saline lake are areally widespread and lat-
lake (Muessig, 1966). Recent South American erally continuous saline-mineral laminae and thin
borate deposits are not good analogs for the clastic mud beds with partings. The sediment may
sedimentary and mineralogical characteristics of be varved, but chemical precipitation is often
the late Miocene Furnace Creek borate depits. continuous, and varves are more likely a record
of storm-flood clastic influxes punctuating the
continual "rain" of chemical sediment than a
Perennial Saline Lake seasma1 or annual record. Diagenetic structures
related to reducing conditions and bacterial
A closed, perennial lake (Langbein, 1961) has metabolysis and fenestral fabric from gas bubbles
outflow that is absent, intermittent, or limited. may occur. The lake may be shallow so wave
Most of the solutes supplied to it are trapped. mixing of sediment yields a relatively structure-
A lack of outflow at the lake implies that annual less strata.
evapration or underflow exceeds groundwater and
runoff input to the lake. Otherwise the basin Heliothermal lakes are an important type of
would overflow and disperse the concentrated saline perennial lake. This type has distinct
brine. Closed lakes range from shallow, near- chemical, sedimentological, and thermal charac-
ephemeral saline lakes, which are common, to deep teristics that are imprtant in explaining many
perennial ones which are relatively uncommon. of the geologic features, including borate depo-
Very large borate deposits are rare and are not sition, observed in the Furnace Creek Formation.
now forming in ephemeral saline lakes. Borate
deposits are more likely to be related to peren-
nial, closed, saline lakes. . Because no two mcd- He1iotherma1 Lakes
ern saline-lake systems are identical, it is
difficult to develop a single model covering all Heliothermal lakes are meromictic (density
saline or other lakes ( ~ e a nand Fouch, 1983; stratified), perennial, saline lakes that develop
Fouch and Dean, 1982). Ancient saline perennial a hot water lens due to absorption of solar
lake deposits often appear to be significantly radiation within brine layers overlain by more
different from those in existing saline lakes dilute layers (Sonnenfeld and Hudec, 1980). Mer-
(Eugster, 1980). omictic lakes do not develop convection and be-
come stagnant with hydrogen- sulfide-rich an-
Sedimentology of saline lakes has been sum- aerobic conditions near the bottom. Features of
marized in detail by Hardie and others (1978). heliothermal lakes (Sonnenfeld and Hudec, 1980)
They divided the sedimentary system into ten imprtant to borate deposition are that they:

ulexite (u) A X R X - - C C R C - A
probertite ( P ) C A -- -- -- X X - - -- --
inyoite (I) - -- -- - - -- R -- X -- X --
meyerhofferite (M) -- -- -- -- -- - R R R -- -- -
colemite ( C ) C P P P P P X P C -- -- -
borax (B) -- -- -- -- -- - P - P C L --
probertite/ulexite core X P - - - -- - -- -
colemanite rind X X -- X - X -- -- -
C to U w. depth ? X X -- I -- -- -- -
C to P w. depth -- P - - -- - - ? ?
U to P w. depth X -- ? R -- -- - -- -
dolomite at borate zone R -- - C - X X X -
carbonate increases near brate zone P P P ? ? X ? L A
stratabound P P P P P P P P -
stratified lenticular
stratified subtabular
interlocked nodules
disseminated stringers/nodules
intraformtional breccia: in ore
above ore
rrrarl nucleus in U or C nodules
grcwth zones: solid inclusions/marl
clay drapes over ore crystals -- --
C veins crosscut ore zones ? C
U veins crosscut ore zones X --
C replacing U X X
I to C/M G-R G-R SILIcn!rE
potassium feldspar

mixed-layer clays

T R Z Y Z ~
realgar/orpiment -- x c -- -- ? --
gypum/anhydrite R -- R ? - - - - X -- A
L< -
celestite ? x -- -- -- - -- C -- -- -- --
halite -- -- -- -- -- -- -- -- -- A C X
native sulfur -- C -- -- X --
other saline minerals R R R R R R R R R A C C

Table 4a. Mineralogic and petrologic characteristics of selected borate orekcdies.


Increased marl/limestone at:

basin margin
basin center
tufa munds
chemical delta
Maximum thickness of ore (m)

close association - - - - - - - X X -- -- X X
remte to borates X
-- -
spring munds X P
I<NBOUS lcsucm<H
altered basalts/palagonite pillow basalts X X G X X -- -- -- -- - -- -
tuff interbeds X X X X X - - X X X ? L ?
bimodal basalt/rhyolite volcanism X X X X X X X X -- X X X
S F D m -
graded bedding
ripple mark
soft sediment slumping
rip-up clasts
scour mark
pebble mudstones
edge-wise mglomrate
pebble conglomerate
subaerial unconformities
sublacustrine unconformities
fossil soil zones
fossil desert pavement
solution textures in borates
karst topography

fining u p a d cycle w. ore depsition x x - - - - - -- X -- -- -- -
laminatedshale/carbonateenclosingore A A A A A A A A A A -- -
laminated shale interfingers w/-rse X X I X X X X ? ? ? X ?
clastics near basin margin
ore ccextensive w. lacustrine beds A M - - - - - -- -- X X -- -
algal strolrratolites G G G G G - - - -- -- - - -
diatoms X X X X X X - - ? ? L ?
invertebrate trace

R = rare; X = present; C = common; L = locally abundant; P = prevalent; A = abundant; I = inferred; ? =

suspected presence; G = occurs in geologically related unit; w. = with. Double dash indicates the
feature has not been reported.

Sources: Death Valley Deposits, McAllister (1970), Evans & others (1976), Barker & Wilson (1976); Billie,
Norman & Johnson (1980), Countryman (1977); Boraxo, Wilson (1976); Terry, Barker (1980); Maria, Barker &
~ilson,1976; unpubl. Tenneco Mining Inc. reports; Lila C, Gale (1912), and unpubl. Tenneco Mining Inc.
reports; Gerstley, Gale (1912) and unpubl. Tenneco Mining Inc. reports; Kramer, Barnard & Kistler (1966),
Bowser (19661, Gates (19591, Dickson & Raab (1965); Kirka, Brown & Jones (19711, Inan (1972), Inan &
others (1973), Ataman & Baysal (1978); Emet, Brown & Jones (1971); Helvaci & Firman (1976), Ataman &
Baysal (19781, Helvaci (1984); Searles Lake, Smith (1965, 1979); South American playa and hot spring de-
posits, Muessig (1958, 1959, 1966); Norman & Santini (Chapter 4, this volume); Hurlbut & others (1973).
Table 4b. Sedimentology of selected borate deposits.

1) occur worldwide in arid areas with large grades into an ephemeral saline lake. If the GWT
deficits in annual water balance; exceeds the sediment surface elevation, a peren-
2) often have a broad, shallow shelf surround- nial lake results. There are several ways to
ing a relatively small and deep, high- change the relative relationship of the GWT to
density brine pool; the sediment surface, such as changes in rates of
3) commonly reach bottom brine-layer tempera- sediment accretion, subsidence, and groundwater
tures of 35O~-60c and occasionally reach recharge. Many basins initially fill rapidly
about 70c; with sediment after formation of the depression.
4) can last for centuries after thermal stra- Basin infilling gives rise to a regressive shore-
tiication is established; line and a coarsening-upward facies sequence.
5) have thermal (density) stratification that Sediment deposition is controlled by the inter-
will not turn over even under strong steady play of these physical parameters during the
winds: evolution of the typical perennial lake.
6) require a density difference between layers
of 15 g/l which is only reached in chloride The facies patterns in a perennial lake system
or sulfate brines and not in carbonate and (Fig. 5) are initial alluvial fan deposits giving
bicarbonate ones; and, way to ephemeral lacustrine depits followed by
7) have frequently gone unnoticed, because few a perennial lacustrine system in which basin
field scientists carry remote sensing ther- formation overwhelms basin infilling. Continued
mometers, so existing distribution maps are inilling coupled with cessation of basin form-
misleading. tion eventually cause regression in the perennial
lake resulting in an ephemeral saline lake fol-
lowed by a dry mudflat with final burial of the
FACIES ANALYSIS OF THE FURNACE CREEK FORMATION older rocks by alluvial fan deposits. These
environmental changes can occur extremely rapidly
Physical Controls on the Facies Sequence in closed basins (Hardie and others, 1978). A
in Closed, Lacustrine Basins example is the evapration of Lake Manly to form
the Death Valley ephemeral lake in the Quaternary
(Hunt and Mabey, 1966). Very well-developed allu-
The position of the groundwater table relative vial fans subsequently formed over much of the
to the sediment surface controls the depositional lake basin.
environment in the central portion of a closed
basin. If the groundwater table (GWT) is lower Facies in the Furnace Creek Formation
than the sediment surface elevation, a dry mud-
flat deposit forms. As the GWT approaches the The rock record in the Furnace Creek Formation
sediment surface, the environment of deposition generally reflects carbonate-volcaniclastic-bo-







Modified from Kendall ( 1984).

Figure 5. Hypothetical sedimentary cycle in a perennial lake - mud flat complex.



Holocene Surficial - Younger and older alluvium with some talus rubble; local-
deposits ly includes evaporites, travertine, and landslide depos-

QUA- Olivine basalt mainly in flws with basaltic agglomerate

Pleistocene at vents; basal gravelly mudstone and sandstone with lo-
Funeral 450 m cal sedimentary breccia; main sedimentary unit is partly
?- ? - Formtion to to well lithified conglomerate, with a sand and m d ma-
600 m trix, and mddy conglomeratic sandstone; local travertine
Pliocene and well-lithified breccia.
Greenwater 360 m Vitrophyre, mstly in flows and locally devitrified,
? - ? - volcanics interbedded with tuff-breccia and bedded conglomerate.

Tfu (upper me-r): Mudstone with basal pumiceous tuff;
grades into sandstone and-conglomerate.
Tf (main member): Calcareous, tuffaceous mudstone and
shale with marlstone, sandstone, and minor limestone;
interlaminations of micrite and minor sparite; minor
pyrite when unweathered; contains gypsum, dispersed
saline minerals, and mjor borate deposits.
Tfcu (upper conglomerate member): Conglomerate inter-
tonguing through m c h of the formation but not with Tfc.
TERTIARY Miocene Furnace 1500 m Tfg (gypsiferous member): Gypsum and minor anhydrite with
Creek to variably abundant mudstone interbeds; fibrous gypsum
Formation 2100 m (selenite)veins and masses; minor limestone in upper
portion; minor borate deposits.
~fa/~b/~p (volcanic rocks): Altered basalt flows (Tfa)
with minor and scattered borate veins; basaltic flows and
intrusions (Tfb) with local borate veins; tuff breccia
and minor vitromyre (TQ) partly altered; Tfa and Tfb
occur as many interbeds throughout the formation.
Tfc (basal conglomerate member): Conglomerate with a
sandy or mddy matrix; locally composed of limestone
flat-pebble (edgewise or chipstone) clasts in thick
lenses; contains mjor borate deposits.
locally conformable or unconformable

Artist Dominantly lacustrine mudstone and sandstone with minor

Drive >I200 m volcanic conglomerate and tuff; interbedded massive,
Formation varigated tuff-breccia, partly altered and well
lithifi d .

Source: McAllister (1976, 1973, 1970), Barker and Wilson (1976), Wilson (1976), Wilson and Emmons (1977),
Countryman (1977), Fleck (1970).
Figure 6. Generalized stratigraphy of the Furnace Creek and Amargosa Valley borate areas of the Death
Valley region, California.

rate lacustrine deposition with a varying shore- 1976), tufa mounds (McAllister,1970), and reed
line position. The initial basin fill is repre- plant fossils (Axelrod, 1940) in growth position
sented in part by lag and alluvial fan-derived (in sec. 12 and 13, T26N, R ~ E )are associated
cobble conglomerates (Tfc; Fig. 6) that grade and with paleoshorelines of the lake. The fossil
interfinger into finer-grained laminated lacus- diatom assemblages are consistent with brackish
trine tuffamus clastics ( ~ f )in the basal Fur- water and thermal spring activity.
nace Creek Formation. The groundwater table
exceeded the sediment level owing to continued During the regression of the saline lake,
subsidence so the basal conglomerate locally was lacustrine sedimentation is superseded by a wet
formed during transgression of the expanding mudlat possibly followed by a dry mudflat Final
perennial saline lake. The "chipstone" facies regression of the Furnace Creek lake is indicated
(Barker and Wilson, 1976, p. 24), an informal by the Funeral Formation cobble conglomerate
unit of Tfc (McAllister, 1970; 1973) associated facies (Fig. 6) which completes basin infilling.
with the basal cobble conglomerate facies, re-
cords the disruption of carbonate flats in an Borate deposition in the basal Furnace Creek
ephemeral lake to form an intraformational edge- Formation occurs in the marl- or limestone-rich
wise conglomerate during perennial lake trans- portions of this unit. Thermal springs were dis-
gression (Fig. 7). Similar intraformational charging into a deep perennial lake as shown by
conglomerates formed during a transgression are the deposition of borates enclosed by the well-
observed in the Green River Formation (Eugster stratified, uniform laminated shale/marl facies.
and Hardie, 1975). Carbonate flats are common on The parallel and laterally continuous laminated
the periphery of chemically zoned ephemeral lakes shale/marl facies shows few sedimentary struc-
(Hunt and others, 1966; Collinson, 1978). Else- tures formed by current action suggesting that
where the Tfc represents sediment in alluvial fan the lake bottom was usually below wave base.
complex. Quiescent anoxic conditions allowed limestone
laminae and thicker mudstone laminae and beds to
Sedimentary rocks deposited in the deep pe- be laid down and preserved. Reducing conditions
rennial saline lake phase are the uniform cal- are suggested by pyrite in unweathered shale/marl
cite-laminated marl/shale (Tf) typical of most of facies (Wilson and Emmons, 1977).
the Furnace Creek Formation. Other evidence of a
lacustrine environment are basalt flows that Borates are in the basal cobble conglomerate
increase in abundance basinward and are phreati- facies (Tfc) and in the gypsum facies (Tfg) of
cally altered to palagonite, fragmented flows, the Furnace Creek Formation (Fig. 6). Borate
and pillow lavas fa; McAllister, 1970) by flow- deposition began with the precipitation of so-
ing into the lake. dium-calcium borates in the basal conglomerate
facies as observed in the Inyo orebody (~ilson,
The presence of algal stromatolites (in sec. 1976). The basal conglomerate here represents a
27, T26 N, R2E SBM and other areas; personal distal alluvial-fan/fluvial facies. Localization
communication, D. E. Emmons, Tenneco Mining, of the ore in a limited portion of the cobble



Figure 7. Schematic block diagram of the perennial lake - mud flat complex of the
Furnace Creek Formation. Modified from Eugster and Hardie (1975) and Kendall
( 1 884). Stratigraphic nomenclature from McAllister ( 1 973).

conglomerate facies could reflect a nearby borif- 1966). The beds at Searles Lake, California, are
erous, thermal-spring that discharged into these thicker than the South American depsits and have
gravels causing precipitation of borates. These a large clastic compcnent with many free-growing
borates are prcbably late synsedimentary phases borate crystals. The Death Valley, Kramer and
and nat a diagenetic precipitate because the cal- Kirka borate deposits are thickly bedded with a
cium-borate colemanite, not sodium-calcium bo- small clastic component and little evidence of
rates, appears to be the stable phase when bo- free-growing, matrix-supported equant crystals of
rates are in contact with ground water (Foshag, the South-American or Searles-Lake types.
1921; Inan and others, 1973). This relatiaxship
is supported by the apparent present-day deposi- The geometry of the Death Valley borate de-
tion of colemanite in water-filled vugs along a posits also indicates that they are lacustrine
fault in the Boraxo pit occurs at shallow depths (Table 4a). Spring-apron borate deposits form
(Wilson, 1976). prismatic fan-shaped deposits on sloping sur-
faces. Planoconvex deposits occur on level
ground. Both of these configurations unconforma-
Sedimentary Environment bly overlie pre-existing sediment or basement
rocks (Muessig, 1966). A perennial-lake evap-
Most of the Furnace Creek sediments were de- orite deposit is typically biconvex in shape and
posited in a perennial lake system. Evidence conformably interdigitates with the enclosing
able 2) in support of this includes: 1) the nonevaporite lacustrine sediments.
laterally continuous limestone laminations in
mudstone: 2)paucity of evidence for subaerial
exposure; 3) rare current structures; 4) fluvial Localization of Borate Deposition in the Basin
sediments only at the basin margin; 5) shoreline
features around the periphery of the present out-
crop; 6) subaqueously altered basalts; and 7) al- One characteristic of Death Valley borate
tered tuffs with a mineral assemblage indicating deposits that does not appear to fit a general
an alkaline, moderately saline environment (Bark- lacustrine model is that saline minerals deposits
er, 1980). Paleontologic evidence includes: 1) in a lake typically are distributed throughout
the presence of fossil diatom genera that today much of the basin and may underlie all of it.
are found in lake environments: and 2) presence Borate deposits in the Furnace Creek Formation
of Lyonothamnus mohavensis leaves and an uniden- are not extensive, underlying less than 1%, when
tified reed-like plant, locally in growth p i - compared to the areal extent of the lacustrine
tion, that was interpreted to indicate a shallow facies in the entire Furnace Creek basin. Some
saline lake in an arid climate (Curry, in Axel- strong localizing process must have been opera-
rod, 1940). These depsits show charactFristics tive. Local concentration and precipitation of
of neither a dry mudflat environment, where the soluble borates apparently occurs in a restricted
laminated sediments are mudcracked and disrupted, lake lagoon or a deep brine pool in a heliother-
nor a moist mudflat environment. The moist mud- ma1 lake. Alternative possibilities are the for-
flat environment is characterized by recurring & mation of chemical deltas at the lake inlets due
situ salt crystallization which completely de- to brine mixing or the formation of sublacustrine
stroys the internal sedimentary structures. Pro- spring aprons by the cooling of hot, boriferous
longed subaerial expsure induces extensive poly- fluids upon contact with the cooler lake brine.
gonal mudcracking, saline crusts, and thin dis-
continuous lamination, if any survives. Wind- Heliothermal lake bottom structure best ex-
reworked sediments, current structures, and flood plain the strong localization of borate depsits
depits are comma in moist mudflats (Hardie and in a basin. Heliothermal perennial lakes common-
others, 1978), but not in the Furnace Creek For- ly have a shallow shelf area surrounding an are-
ration. ally restricted deep brine pool. Quiet condi-
tions are induced by depth and thermal stratifi-
The origin of the Death Valley bedded borate cation so the more evaporated, dense brines ac-
deposits requires a system in which very thick cumulate in the deep pool. Borate deposition in
borate beds could form with a minor clastic com- this environment would produce thick, areally-
ponent and little evidence for subaerial exposure limited deposits in the deep prtion which would
(Table 4a, b). These deposits are 1-3 orders of interfinger with laminated sediments deposited
magnitude larger than modern spring or ephemeral within the quiet stratified waters. The shallow
lake depits (Muessig, 1966) and are similar in shelf area could have an associated carbonate
magnitude to the largest modern lacustrine ana- flat which is disrupted during transgression
logs (Eugster, 1980). The large size and lack of forming lag or edge-wise conglomerate. The high
draping (except at Monte Blanco) probably elim- temperature within the heliothermal lake may
inates a spring apron origin for most deposits of allow the primary borate precipitated here to be
borate in the Death Valley region. The borate a lower hydrate such as colemanite or probertite.
purity and thick beds within the deposits is These sedimentological characteristics of helio-
quite different from modern borate-depositing thermal lakes fit the observed facies relation-
systems. The ephemeral lake deposits in South ships and lithologic attributes of the Furnace
America have beds of borate up to a few meters Creek Formation better than a simple, unstrati-
thick interbedded with voluminous clastics and fied perennial-lake model.
often composed of many free-growing equant crys-
tals supported by a non-borate matrix (~uessig,

PRIMARY BORATE MIBERALS : if an appropriate [Na]/[ca] is reached at an

THE ORIGIN OF COLEMANITE appropriate [H20]. The two minerals have similar
infinite-polymer chain structures that may be
Colemanite ( c ~ ~ B ~ o ~ ~ is' ~thought
H ~ o ) by many kinetically difficult to form (Christ and others,
to form only diagenetically by alteration of 1958; Clark and Appleman, 1964). The occurrence
higher-hydrate borates. This would occur by sim- of probertite on the Death Valley ephemeral lake
ple dehydration of primary inyoite suggests that this is not an absolute barrier at
( C a 2 ~ 6 ~ 1 1 ' 1 3 ~ 2 0t)o m e y e r h o f f e r i t e the extreme temperature encountered on the ephem-
(Ca2B6~11'7~20) to colemanite. Another mechanism eral lake floor. The formation of primary cole-
of colemanite formation is the alteration of manite deposits must be the result of changing
primary ulexite (NaCaB509'8~20) to colemanite by geochemical conditions in the depositional envi-
dehydration and sodium removal (Rodgers, 1919; ronment. It is typically not the result of later
Foshag, 1921; Muessig, 1959). diagenetic alteration of synsedimentary borate
minerals of higher hydration and/or lower sodium
The highest hydrates of the calcium- and so- content. Support for this hypothesis comes from
dium-calcium-borate mineral series are most sta- models of brine evolution, the possible ranges of
ble under surface conditions in South America the [H20] in natural systems, and the thermal
(Muessig, 1959). This working hypothesis pre- history of the Furnace Creek basin.
sents a problem when used outside South America
because some Death Valley borate deposits are
mostly colemanite which is the lowest hydrate in BRINE EVOLUTION MODELS
the calcium-borate series. The geology of some
deposits indicates that colemanite is a primary
borate. For example, the Terry, Maria, and the Brine evolution in saline lakes can produce
Lila< deposits are colemanite. Almost no other calcium borates in a sodium-rich environment.
borate is present and no clear textural or miner- Inyoite, the high-hydrate calcium borate is also
alogical evidence for any borate precursor ex- observed as a primary mineral within the moist
ists. A similar condition occurs in the cole- mudlat sediment of a South American ephemeral
manite deposits at Emet, Turkey (Helvaci and lake adjacent to an area where ulexite precipi-
Firman, 1976). tates in a thermal-spring apron (~uessig,1966).
Similar chemical conditions can also produce
In zoned borate deposits, a secondary origin primary colemanite in a perennial saline lake
for colemanite is indicated by its crosscutting even though it is a low-hydrate in its series.
relationship in sodium-calcium borate deposits, The existence of colemanite-precipitating condi-
porphyroblastic blocky crystals and radial aggre- tions in modern saline lakes has not been ob-
gates replacing ulexite on the edges of the de- served. There appear to be no modem analogs for
posit a able 4a). Experimental data (Fig. 8) some ancient saline deposits (Eugster, 1980).
show that these processes occur as either the The lack of modern analogs does not preclude the
activity of water (i.e. [H20]) or the activity existence of appropriate conditions in saline
ratio [Na]/[ca] decreases or brine temperature lakes in the past. Central to the problem is
increases (Christ and others, 1967). that boron-bearing brines from hydrothermal
fluids are rich in sodium (Fig. 4; Ellis, 1979),
The high-hydrate-first hypothesis is invalid as are brines in saline lakes of western North
under modern surface conditions on the Death America (~ugsterand Hardie, 1978). Bristol Dry
Valley playa. Probertite ( N a C a ~ '~5H20),
0~ the Lake in California, where hydrated calcium chlo-
lower hydrate in the sodium-calcium borates, and ride salts now precipitate, is sodium-rich (Na/ca
meyerhofferite, the intermediate hydrate in the = 1.3; Eugster and Hardie, 1978). Basin proces-
calcium borates, and possibly colemanite itself, ses that concentrate calcium in residual brines
are forming in and on the surface crust of the need not produce calcium-rich fluids to pre-
modern playa (Hunt and Mabey, 1966; McAllister, cipitate calcium salts.
1970). The replacement of inyoite by meyerhof -
ferite or colemanite is rare, although it occurs The most feasible mechanism for enriching
at Monte Blanco (Rodgers, 1919), and inyoite calcium relative to sodium in a brine is mineral
itself is not widely distributed in modern or precipitation. Other potential mechanisms are
ancient deposits (Muessig, 1958; Table 4a). We ion exchange between sediment and brine (Bowser,
conclude that the inyoite dehydration mechanism 1966), reverse osmosis (Ellis and Mahon, 1977),
for the formation of colemanite is volumetrically and Soret-effect solute migration (Hanor, 1979).
insignificant . The latter two mechanisms are controversial in
their efficiency and in their importance at near
Geochemistry of hydrated sodium, calcium, and surface conditions in geologic systems.
magnesium borates suggests that colemanite pre-
cipitates under surface conditions of low [H20]
and a relatively low [~a]/[ca] (~hrist and
others, 1967). Extreme desiccation causes pro- Mineral Precipitation By Evaporation
bertite to form rather than ulexite on the Death
Valley playa and in some spring aprons (Hunt and
Mabey, 1966; McAllister, 1970). The range of The Hardie-Eugster (1970) model of mineral
[H20] at which probertite forms overlaps that of precipitation with continuous evaporation in
colemanite (Fig. 8). Primary colemanite can form saline lakes suggests that several chemical di-

C%&OI I* 5 4 0

Notes: Dashed lines enclose m e t a s t a b l e f i e l d s

[-I = A c t i v i t i e s
C l o s e d s y s t e m f o r boron; O p e n f o r c a t i o n s a n d w a t e r
X : Y a x i s r a t i o i s I : 10
Solids a r e i n equilibrium w i t h s a t u r a t e d solutions
Source: Christ, Truesdell, a n d Erd, 1967, f i g u r e 16.

Figure 8. Schematic phase relations i n t h e s y s t e m N a 2 0 2 & % 4 H 2 0 - 2 C a 0 3&0,5H20-H20.

N o t e t h e overlap o f logP207 o f colemanite w i t h t h e ulexite and probertite field, which indicates
t h a t colemanite c a n f o r m in a s y s t a m t h a t precipitates ulexite or probertite with t h e
appropriate N a / C a ratio.

vides must be crossed to produce a calcium-en- The Halite Problem

riched brine. The Furnace Creek evaporites were
apparently deficient in magnesium as shown by Halite is generally the most common mineral in
only rare magnesium-bearing minerals in playas continental and marine evaporite deposits (Har-
of the region (Droste, 1x1). We do not consider die, 1984). Halite is uncommon in brate depos-
the magnesium-bearing phases to be important in its except in ephemeral-lake types (Table 4a).
the brine emlution to calcium-borate precipita- The absence of halite in the Furnace Creek Forma-
tion. tion argues against an ephemeral lake origin for
these rocks. The absence of halite raises a
The Hardie-Eugster model predicts that even question concerning the lack of the sdium-chlo-
dilute saline fluids, upon sufficient evapora- ride-rich fluid that the brine evolution path de-
tion, will soon commence to precipitate calcite. scribed above would have generated in the Furnace
This is the first crucial chemical divide in Creek basin. Perhaps this Na-Cl brine was trans-
producing a calcium-enriched brine. If the cal- ported to Death Valley proper when the Furnace
cium concentration is greater than the total Creek basin was breached in the Neogene. Death
alkalinity (i.e., the sum of carbonate species) Valley formed mainly after Furnace Creek time
in the brine, calcium levels will increase in the making it a possible sink for regional ground-
brine. This path initially produces a sulfate- water flow. The Death Valley playa contains
and chloride-rich brine. In the Furnace Creek halite, up to 300 m thick (Hardie, 1984), preci-
borate deposits, this stage in brine evolution pitated during the evaporation of a pleistocene
produced abundant borates as shown by the typical lake (Hunt and others, 1966). This sdium chlo-
association of limestone with sodium-calcium and ride-rich lakewater may be derived, in part, from
calcium borates. the Furnace Creek brine.
The next chemical divide in a calcium-enriched Halite is readily soluble in groundwater and
brine is gypsum precipitation. If the calcium perhaps it was removed in solution during dia-
concentration exceeds that of sulfate, the re- genesis. Beds both within and above borate de-
maining brine will be a sodium-calcium-chloride- posits are often brecciated (Table 4a) which is a
enriched brine. Pyrite is common in the Furnace feature commonly associated with dissolution of
Creek Formation (Wilson and Emmons, 1977) sug- salt beds. This brecciation could be the result
gesting reducing conditions in the brine. calcium of complete solution of a much more soluble
exceeded sulfate in the Furnace Creek brine. phase, such as halite, in the subsurface. In
Gypsum precipitation with calcite in the Furnace borate deposits, brecciation is usually ascribed
Creek evaporites is later than the deposition of to volume decreases related to borate dehydration
the main colemanite and ulexite borate deposits rather than saline mineral solution. Beddedor
(McAllister, 1970). These younger gypsum (origi- disseminated halite is very rare in the Furnace
nally anhydrite?) deposits are small, are finely Creek Formation. It seems unlikely that all salt
intermixed with colemanite (McAllister, 1970) and and most evidence of its presence, such as crys-
have apparently co-precipitated with it. The tal casts, could be removed by leaching. The
residual sodium-calcium-chloride brine was not brine may not have generally reached halite pre-
precipitated within the Furnace Creek basin. cipitating concentrations. Borate precipitation
This brine evolution path (Hardie and Eugster, probably occurred under special conditions in a
1978) has produced the present4ay calcium-rich restricted portion of a heliothermal lake with
saline waters of Bristol Dry Lake ( ~ a / =~ a
1.3) other portions of the lake below halite satura-
and Soda Lake (~a/Ca= 5 ) , California. tion.

Boratdepositing brines with calcium contents

less than total alkalinity follow a path of cal- ACTIVITY OF WATER IN BRINES
cium depletion. The zonation of the evaporite
minerals at the Kirka borate deposit in Turkey is
attributed to an increasingly sodium-rich brine Sedimentary processes that could potentially
based on the precipitation-of calcium-bearing reduce the activity of water, i.e., [H20], to
phases (Inan and others, 1973). Calcite was levels at which low-hydrate brates could preci-
precipitated initially. This enriched the resid- pitate include evaporation and increasing temper-
ual fluid in sodium cations and borate anions so ature. Generally, as [H20] is decrease at con-
that, upcn further evaporation, ulexite precipi- stant temperature, borate higher hydrates lose
tated. Depletion of a significant portion of water to form secondary lower hydrates (Hanshaw,
calcium in the brine via this mechanism leads to 1963). If the [H20] in the brine reaches a level
the precipitation of sodium borate as borax. where a lower hydrate is stable it will form
rather than a higher hydrate. Estimates of [H~OI
Brine evolution by precipitation in a semi- in natural brines are few and direct measurements
closed system forms zoned deposits with a primary are generally unavailable. The sparse data from
mineralogy of calcite to ulexite to borax. The chemical and natural systems indicate that in
brine appears to be sufficiently sodium-rich both ephemeral lakes and perennial lakes it is
that, over time, nearly all calcium was removed possible to precipitate low-hydrate brates.
from the brine.
E&hemeral Lake Environment
Activity of H20 in a brine is numerically
equal to the relative humidity of the atmosphere
in equilibrium with the brine, provided the brine
and atmosphere are at the same temperature re-
ver, 1982). Hunt and others (1x6) report a mean
annual relative humidity of about 17.5% at a
station near the Death Valley ephemeral lake.
The relative humidity reaches a low of 11% in the
summer and a high of 27% in the winter. General-
ly, near surface water /sedi ment temperature is
approximately equal to the mean annual air tem-
perature (~retener,1981), and the mean annual
humidity is used as an estimate of the [H20] on
the Death Valley ephemeral lake. This extremely
low [H20] is required to crystallize the prober-
tite, meyerhofferite, and colemanite(?) observed
on the Death Valley ephemeral lake. An [B20] of Figure 9. Marl (dark zones) alternating kith
below 0.75 (Drever, 1982) on the Death Valley bands of radial acimlar colemanite (translucent)
ephemeral lake is also indicated by the dehydra- from Death Valley, California. Close associatian
tion of gypsum to form anhydrite rinds at mean of marl and colemanite suggests they are penecon-
annual surface temperature of 20c (Hunt and temporanecus. Death Valley borates commcrily have
others, 1966). The presence of tincalconite on cores and bands of white to light brow,; miail.
the Death Valley playa (Hunt and others, 1966) The calcite in these borates is characteristi-dl-
implies an [H20] of less than 0.58 at 20c ly orange under cathcdoluminescence (Appendix I)
(Hanshaw, 1963). Kernite (Hunt and others, 1966) and yellow under vertical ultraviolet light exci-
in the shallow sediments on the ephemeral lake tation. The borate minerals themselves did not
also suggests a low [H20] (Christ and Garrels, fluoresce. The fluorescence of borates (Gaft and
1959). others, 1977) may be caused by mixtures of the
fluorescent calcite and nonfluorescent borate.
Under isothermal conditions, the humidity of Field of view about 5mm wide.
the atmosphere controls the upper limit of brine
salinity. Soluble salts can only accumulate when
the atmospheric humidity is less than the [H20]
at which they are stable. If the atmospheric final brine constituents but the [H20] at early
humidity is greater than that in equilibrium with mineral precipitation divides alcng the evolution
a given brine, evaporation will not occur path of the brine is relevant in both cases. If
(Drever, 1982). Instead, water comes out of the total alkalinity is less than the calcium ion
atmosphere and into the brine thus diluting it. concentration, the paths taken by nonmarine and
The low ambient humidity at Death Valley makes marine brine will be the same, especially with
virtually any saline mineral stable, relative to respect to the initial minerals precipitated
[H20], on the ephemeral lake surface. (Hardie and Eugster, 1970; Hardie, 1984). Simple
evaporation of seawater (Kinsman, 1976) indicates
Climatic conditicns on coastal sabkas are not an [H20] during carbonate precipitation from 0.98
good analogs to those on ephemeral lakes (con- to 0.93; from 0.93 to 0.76, for gypsum-anhydrite;
tinental sabkas). Typical daytime humidity above from 0.76 to 0.67 for halite, and less than 0.67
the Persian Gulf coastal sabkas is 34% to 45% for potash salts.
insma man, 1976) which is much higher than on the
ephemeral lake at Death Valley. Under optimal Borates are closely associated with calcium
conditions the mean annual temperature in a carbonate (Table 4a) and are penecontemporaneous
coastal sabka may exceed that of an ephemeral with it in the paragenetic sequence (~ig.9).
lake by ~ O ~ C - ~ aOable
~ C 3) and we assume ancient This association appears to be a primary genetic
systems are similar. The expcted mineral assem- texture because there is some geochemical evi-
blages are siqnificantly different between these dence that calcite enhances the coprecipitation
environments. The major ions in the brine are of krate (~itanoand others, 1978; 1979). With-
similar (Hardie, 1984) and the proportions dif- in the range of [H~o]for calcium carbonate pre-
fer. cipitation (0.98 to 0.93) only the highest hy-
drates ulexite and inyoite would be stable at
temperature less than 20c (~igs.10 and 11).
Perennial Lake Environmt This stability relationship is in accordance with
Muessig's (1959) predictions. Muessig studied
The activity of water in perennial lakes is only systems with a high ~ a / ~
a as indicated
poorly known We use the evaporation of seawater by abundant halite and only rare calcium-bearing
to mcdel CH 0] in non-marine hydrothermal fluids. saline minerals on the South American playas.
Hydrothermal fluids contain major ions similar to Expriments indicate that a saturated CaC12 brine
those in seawater but in differing proportions would have an [H20] of 0.31 ranging up to a
(Hardie, 1984). Evaporation of hydrother ma1 CH 01 of 0.75 for a saturated NaCl brine (Lerman,
fluids leads to differences in the proportions of 1939). Although a natural brine is much more

S o u r c e : A f t e r H a n s h a w , ( 1963). l09[~,0]

F i g u r e 10. P r e l i m i n a r y c a l c i u m b o r a t e p h a s e r e l a t i o n s in t e r m s o f t e m p e r a t u r e a n d
m24. T h e reciprocal o f absolute t e m p e r a t u r e and log^,^ h a s been
inverted t o OC and p24, r e s p e c t i v e l y , t o aid interpretation.

0 -0. I -0.2
Source: A f t e r Hanshaw, ( 1 9631.
log [Hz 01

Figure I I. P r e l i m i n a r y sodium-calcium p h a s e r e l a t i o n s in terms o f t e m p e r a t u r e a n d

. A x e s changed a s described in Fig. 10.

complex, the laboratory data suggest that a cal- THERMAL HISTORY OF ?HE FURNACE CREEK 'I'ROUGH
cium-enriched chloride brine would tend to have
lower [H20] than a Na/cl enriched brine. The Fluid Inclusion Evidence
mineral precipitation path leading to a CaCl? -
rich brine could produce b t h the low Na/~aratio Two-phase fluid inclusions (Appendix I) in
and the low [H~O]necessary to produce primary outcrop samples from the Furnace Creek borates
colemanite precipitation with limestone. are locally abundant. These inclusions occur
along healed fractures and have variable vapor-
Brine temperatures of 35~-700~attained in to-liquid ratios diagnostic of secondary and/or
heliothermal lakes make the calcium borates necked-down fluid inclusions (Figs. 12 and 13).
inyoite and meyerhofferite unstable at any [H~o] Freezing point depressions of these inclusions
(Fig. 10). Probertite, rather than ulexite, are near OOC indicating they contain a brine of
would be stable at [H20] of 0.9 or less in helio- very low salinity. Breaching these inclusions on
thermal lakes (~ig.11). Calcium brate stabili- a crushing stage yields mostly water vapor plus a
ty is strongly temperature controlled. This non-condensible gas, possibly air. These fea-
relationship causes a primary assemblage of cole- tures indicate that the inclusions formed during
manite with calcite during even minor brine heat- late diagenesis and weathering of the borate
ing at the reduced [H~o]possible in calcium-rich deposits. The vadose zone in the Furnace Creek
brines. region is now at least 150 m thick (Wilson and
Emmons, 1977). Uplift of these rocks into the
The [H20] of gypsum-anhydrite precipitation vadose zone causes fractures and pores to fill
(0.934.76) encompasses the stability of coleman- with a mixture of air and meteoric water held by
ite (less than 0.79) in the brine. The associa- capillarity. Infrequent groundwater recharge
tion of gypsum-anhydrite beds with lenses and dissolves some borate in the relatively fresh
stringers of colemanite in the Furnace Creek water and later reprecipitates it during evapora-
Formation (McAllister, 1970) appears to be a tion. This process traps a variable mixture of
reascaable primary assemblage.

Figure 12. Transmitted light view of a polished Figure 13. Detailed view of Figure 12 section.
section of colemanite from Played Out Mine, Death Vapor-rich fluid inclusions appear dark due to
Valley, California. Sample contains common se- internal reflection of light at the chamber mar-
condary fluid inclusions with variable liquid to gin. Some liquid may be condensed in narrow
vapor ratio. The somewhat regular pattern of the portions of the inclusions (arrows). Liquid
inclusion planes suggests a cleavage control of inclusions show a near-spherical bubble and less-
the now-healed fractures. Field of view is about pervasive internal-reflection effects. Field of
view is 0.3 mm.

air and water in the inclusions along healed aqueous solutions, undercooling of some 40c is
fractures and within pores. Experiments indicate required to overame metastability effects and to
that borates are slow to dissolve in fresh water nucleate a vapor phase (Roedder, 1984). Rocks
(Karazhanov, 1%3), but that dissolution would be that form at less than about ~ O ~ C - ~will
O ~ not
enhanced by C02-rich meteoric water precolating contain two phase aqueous inclusions at ambient
down fractures (Gulensoy and Kocakerim, 1978). temperature (20c-30c). These data support the
Soil m2 is often an order of macpitude greater lower temperature range during borate diagenesis
than K O 2 in the atmosphere enhancing this ef- in the Furnace Creek Formation.
fect. The ubiquitous secondary fluid inclusions
make it difficult to distinguish primary inclu-
sions in these samples. Vitrinite Reflectance Evidence
Primary fluid inclusions in borates can be The apparent maximum temperature range of
destroyed by recrystallization of soluble borates between 60c to lloOc can be compared to the
at depth. The high solubility of borax indicates thermal maturity of disseminated organic matter
that recrystallization of the rock during burial (OM) in the Furnace Creek Formation. Vitrinite
would expulse excess brine and produce the ob- reflectance (R,,,) is the best optical indicator of
served mrsely crystalline interlocking texture thermal maturity of OM (Stach and others, 1982;
(Christ and Garrels, 1959). Much of the "excess Tissot and Welte, 1984). Burial increases R, in
brine" in evaporites is in fluid inclusions. respcPlse to the geothermal gradient by irrevers-
Burial of soluble brates could eliminate these ible devolitization reactions. Because there is
inclusions and cause brine expllsion and migra- no reverse reaction, OM thermal maturation does
tion. not undergo retrograde diagenesis and R records
the peak burial temperature. Peak buriay temper-
Many identifiable primary fluid inclusions ature is here estimated from Barker and Pawlewicz
(along zoned crystals, Fig. 14) in the Furnace (in press) who calibrated an empirical geother-
Creek borates are single phase. This is demon- mometer based on R,.
strated by the stretching of the inclusion during
overheating and re-nucleation of two-phases upcn Sufficient OM was concentrated from shale only
cooling. Single-phase, primary fluid inclusions from the Boraxo deposit. Ten other borate ore
imply a low-temperature origin for the rock. In and shale samples from the Death Valley area
contained insufficient OM for R analysis.
Sparse to abundant translucent O# with some
obvious contributions of opaque OM recycled from
pre-existing rocks, were also found in borate
samples from the Frazier district, Tick Canyon,
and Playedat claim (Death Valley), California.
The lack of OM could show that the contribution
of terrestrial plants to the sediment may be
small in this arih region, that the OM is trapped
at the lake margins, or that extensive oxidation
during weathering of the Furnace Creek Formation
has removed OM that was originally deposited in
the sediment.
Vitrinite reflectance measured on polished
mounts (Appendix I) of OM concentrates from the
Boraxo borate deposit indicates a thermal maturi-
ty of 0.4% Rm. This is equivalent to a peak
burial temperature of about 30c-35Oc. The
limited reaction duration at maximum temmrature
indicates that the vitrinite reflectance may be
somewhat low for the temperature reached. Sys-
tems with extremely short heating durations may
Figure 14. Doubly polished thin section of zoned have a vitrinite reflectance lower than expected
colemanite. Growth-zoned secondary colemanite from empirical geothermometers (Barker, 1983).
crystal filling a fracture crosscutting bedded The thermal immaturity of these samples is sup-
colemanite and marl, De Belly mine, near Cork- ported by translucent orange to brown OM that has
screw Canyon, California. Matrix is to the left. a woody to amorphous structure. The translucency
Zones in the colemanite are composed of single of OM decreases with increasing diagenesis tem-
phase primary fluid inclusions, and opaque to perature until it becomes opaque. The presence
translucent solid matter. EDS analysis of this of even slightly translucent woody OM implies a
and most borate samples shows a low silicon and low thermal maturity of less than 0.5% R, equiva-
aluminum background with sporadic readings of lent (Tissot and Welte, 1984).
alkali metals typical for trace amounts of clay
minerals in sedimentary rocks. These samples
were polished with diamond and are not contami- Clay Mineral Evidence
nated with alumina- or silicon-based abrasives.
Field of view is 0.7 mm. The basic relationship used to estimate burial

temperature from clay mineralogy is the conver- similarities throughout the Neogene, we use the
sion of smectite to illite as temperature and/or current temperature regime in the Death Valley
pressure increases (Weaver, 1967; Keller, 1963). region as an analog for surface thermal condi-
Most studies of this reaction are related to tions during Miocene borate deposition.
petroleum migration/generation in marine sedi-
ments (~owerand others, 1976; Aronson and Hower, Heliotherma1 effects on concentrated saline
1976; Yeh and Savin, 1977; Dypvik, 1983; and lake waters may elevate the maximum brine temper-
Bruce, 1984). These techniques should be appli- ature above the regional surface water tempera-
cable to saline nonmarine conditions. ture (Sonnenfeld and Hudec, 1980). We use 15Oc
to 50c as a typical range of potential brine
The threshold temperature required to initiate temperatures, including those of stratified lakes
smectite dehydration ranges from 160% (71c) in able 4). This temperature range is wider than
the sediments of the Mississippi River delta to the 2 5 O ~to 35% proposed for borax precipitation
more than 300~ ( 1 5 0 ~ ~in) the Niger Delta in the Miocene lake at Kramer, California (Christ
(Bruce, 1984). Smectite disappears in sediments and Garrels, 1959).
under the North Sea over a short temperature
range of 14g0 and 167% (65O-75T) equivalent to Borehole temperature measurements in basins
an Rm of 0.5% (Dypvik, 1983, after Foscolos and similar in geology to that of Death Valley (Deep
others, 1976). Tourtelot (1979) modified zones Spring Valley, CA; Silver Peak playa, NV; and
of diagenesis from Curtis (1977) in which smec- Fish Lake, NV) indicate geothermal gradients
tite conversion begins at about 70c while noting ranging from 2g0c/km to 4g0c/km (Sass and Munroe,
that this was a broad interface. 1974). Adecline ingradient following Miocene
volcanism and related geothermal systems is pro&
The transformation of smectite to illite de- able. This decline yields a reduced local geo-
pends on temperature but it also depends on fixa- thermal gradient, so similar reduced range (30c
tion of potassium ( ~ o w e rand others, 1976) and to 50T/km) of geothermal gradient is used to as-
liberation of silica during Al-for-Si substitu- sess burial temperature in borate deposits.
tion (~oscolosand Powell, 1979) and possibly
liberation of magnesium and iron (~ucheckiand The borate deposits in the basal Furnace Creek
others, 1977). Formation were deposited by 6.3 f 0.1 my ago
(Fleck, 1970). The entire 1500 m of the Furnace
If a marine to non-marine saline pore water Creek Formation was deposited over a maximum time
analogy is viable, smectite would be an increas- span of 1.2 my ending about 5.4 f 0.2 my (Fleck,
ingly minor clay in the Furnace Creek sediments 1970). Deposition of the Greenwater and Funeral
depending on how far they had exceeded a burial Formations followed the Furnace Creek Formation.
temperature of 65O to 75O~. In addition, dis- The local folding and erosion truncating the
seminated silica and chlorite, from the displaced Furnace Creek Formation (~c~llister, 1970) sug-
magnesium and iron, would form in the shale as a gest that it was already being unburied by the
result of smectite transformation. Neither auth- time these formations were deposited. We assume
igenic quartz nor chlorite is prominent in the that the maximum burial of the Furnace Creek
shale of the Furnace Creek Formation. The in- borates is equal to its stratigraphic thickness.
verse relationship expected between smectite and Burial to 1500 m, under a surface temperature of
illite, if burial diagenesis is significant, is 15O to 50c and a geothermal gradient of 30 to
lacking. Smectite is the dominant clay (~roste 50~~/km, indicates the maximum diagenetic temper-
in McAllister, 1970, p. 5; Countryman, 1977;
- ature reached in the borate deposits in the basal
Wilson, 1976; Norman and Johnson, 1980) with Furnace Creek was between about 60c and 1 2 0 ~ ~ .
chlorite very sparsely present in some deposits. Rapid burial after deposition is indicated by the
The clay mineralogy from X-ray diffraction limited amount of post-Artist Drive time (abut 6
studies of selected borates ores from the Death my) for 1500 m of burial and subsequent deep
Valley region a able 5) support the predominance erosion to take place. Fleck (1970) estimated
of smectite over illite and the paucity of chlo- the erosion rates could easily be in excess of 50
rite. Abundant smectite occurs (Table 3a) in cm/yr. This erosion rate is equivalent to an
many Death Valley borate deposits as well as uplift of at least 1 km in 20,000 yr. The expo-
worldwide (Ataman and Baysal, 1978). The R, and sure to maximum temperature in these rocks must
the volumetric relationship between the smectite, have been geologically brief. The extremely
illite, and chlorite supports burial temperatures rapid deposition rate for the Furnace Creek For-
below 65O to 75O~. mation may have temporarily reduced the geotherm-
al gradient by the quick build-up of a mass of
relatively cool sediments. If equilibrium was
Burial History Reconstruction not re-established before uplift, the lower maxi-
mum burial temperature would be favored.
The climate was arid in the Miocene during
development of the Furnace Creek trough (~c~llis- Burial history reconstruction, fluid inclu-
ter, 1970). The mean surface temperature was sions, vitrinite reflectance, and clay mineralogy
about 1 5 O ~and has remained stable to the present of the borate deposits suggest that the Furnace
(Savin, 1977). The mcdern regional surface water Creek borates reached a maximum temperature be-
temperature ranges from ~ O O Cto 1 5 O ~(Blakey, tween 70~-1200~. The exposure to the maximum
1966) and is similar to the mean surface tempera- temperature was brief and borate burial diagene-
ture. Because of the geological and climatic sis was minimized.

The brief exposure to maximum temperature has region.

produced an incomplete conversion of Furnace
Creek borates to dehydrated, more stable ghases. 2) The sedimentary environment in epkmral
Christ and Garrels (1959) concluded that the lakes, heliothermal saline lakes and some unstra-
sporadic occurrence of kernite in borax, along tified perennial saline lakes can produce the
with the slow growth =d difficult nucleation of [~a]/[~a] ratio, CH 01 and water temperature
this phase indicate a slow approach to equilib- conditions necessary Zor primary mlemanite dep-
rium with burial diagenetic conditions. This sition. The heliothermal lake model best ex-
observation is relevant to colemanite origin plains the strong localization of borates in the
because borax and kernite have structural and lacustrine basin within the isolated deep-por-
chemical differences similar to those between tions of this type of lake. The long-term sta-
inyoite and colemanite (Christ and Garrels, bility of the thermal stratification allows the
1959). The polymineralic zoned borates of Death quiet deposition of laminated sediment enclosing
Valley (~c~llister, 1970: Kistler and Smith, the borate deposits. The basin morphology and
1975) may represent an incomplete conversion chemica1 para meters developed in typical helio-
reaction during burial diagenesis. If nearly thermal lakes are very conducive to borate forma-
monomineralic colemanite deposits were also the tion.
result of burial diagenesis, they should show
evidence of incomplete reaction, which they do 3 ) The bulk of the boron-rich fluids entering
not. the Furnace Creek paleo-lake were derived from
hydrothermal leaching of boron-bearing country
rock. The contribution of solutes from magmatic
Origin of Colemnite Deposits: an Interpretation fractionation products is unknown but could be
Sedimentary textures in ancient borate depos-
its, borate geology of recent ememera1 lakes, 4) Coarse clastic debris derived from local
brine evolution models, the low [H20] possible in sources around the basin margin eventually en-
saline lakes, and the thermal history of the croached upon a closed, saline-lake basin. In-
Furnace Creek Formation, together suggest that traformational, edgewise conglomerates formed
the nearly monomineralic colemanite deposits are during a transgression of a perennial lake shore-
primary synsedimentary features rather than the line over marginal marl mudflat flooring the
residuum from diagenesis of non-colemanite borate basin. Borate deposition commenced, during or
deposits. soon after transgression, within the laminated
shales of the lacustrine facies. Later infilling
In zoned borate deposits, crosscutting rela- of the basin gives rise to a regressive mrsen-
tionships and replacement features in the cal- ing upward sediment sequence.
cium-borate enclosing a sodium-calcium borate
core indicate a secondary origin of colemanite 5) Only quiet, deep, or stratified water allow
from diagenetic reactions. The conversion of deposition of finely laminated mudstone with
sodium-calcium borates to the low hydrate cole- individual laminae traceable for a considerable
m i t e is famred by burial with increasing tem- distance.
perature (Figs. 10 and 11) and immersion in
groundwater. The residual brine after borate 6) The most abundant borate is often colemanite,
precipitation is sodium-rich compared to the the lowest hydrate in the calcium-borate series.
dilute groundwater derived from the adjacent This colemanite has textures and sedimentary
mountains and a diagenetic reaction to form so- associations that indicate that it is primary and
dium-poor phases such as colemanite may be fa- has not formed by diagenetic alteration of other
vored. The temperature of conversion of ulexite borates .
to colemanite is above 70c (Table 3). This is
just within the range of the maximum burial tem- 7) The thermal history of the Furnace Creek
perature reached in the basal Furnace Creek For- Formation suggests that a low burial temperature
mation for what must have been a very short time. of between 7O0c and 1 2 0 and
~ ~ short duration of
Under these conditions, partial reaction of so- heating produced zoned colemanite deposits from
dium-calcium borates to colemanite is not unrea- incomplete conversion of higherh-hydrate, scdium-
sonable and supports our interpretation. rich borates during burial diagenesis.

The observations in this paper support these
1) Borate deposits in the late Miocene (Hemphil-
lian) Furnace Creek Formation result from the We thank the following individuals who contri-
precipitation of high and low hydrate sodium- buted borate samples: S. J. Lefond, J. L. Wilson,
calcium borates and low hydrate calcium-bearing J. W. Minnette, K. G. Papke, C. E. Schroeter, C.
minerals in a heliothermal, perennial, saline D. Chandler, R. Currier, F. G. Johnson, J. C.
lake. Ephemeral-lake borate deposits are not Norman, and E. S. Montgomery. R W. Richardson
volumetrically important in the Death Valley provided scanning electron microscopy and pho-

tography. R M. Pollastro provided initial x-ray book: Las Vegas to Death Valley and return,
diffraction data and mineral identification, and Nevada Bureau of Mines and Geology Report 26,
M. Bowie did additional work. M. J. Pawlewicz p. 22-23.
and B. C. Crysdale prepared kerogen concentrates
from borate samples and measured vitrinite re- Barnard, R. M., and Kistler, R. B., 1966, Strati-
flectance. Technical support was supplied by M. graphic and structural evolution of the Kramer
Bowie, J. Hingtgen, M. Wooldridge, J. Brannan, I. sodium borate body, Boron, California: in J.
Rae, and L. McNeil at the New Mexico Bureau of L., Rau, ed., 2nd Symposium on Salt, v. 1,
Mines and Mineral Resources. Northern Ohio Geological Society, Cleveland,
p. 133-150.
The manuscript was read by W. E. Dean, M. H.
Link, J. L. Wilson, and J. C. Norman, whose Barsukov, V. L., and Egorov, A. P. 1960, Some
efforts and comments are greatly appreciated. We chemical characteristics of conditions of
remain solely responsible for all statements and formation of hypogene borate deposits: (Geo-
conclusions in this paper. kimia, 1957, English translation) Geochemis-
try, no. 8, p. 790-801.

REFERENCES Barnes, H. L., 1972, Hydrothermal solutions: &

R. W. Fairbridge, ed., Encyclopedia of Geo-
Aronson, J. L., and Hower, J., 1976, Mechanism of chemistry ?Znvironmental Sciences, VAn Nostrand
burial metamorphism of argillaceous sediment: Reinhold, New York, p. 571.
2. radiogenic argon evidence: Geological So-
ciety of America Bulletin, v. 87, no. 5, p. Baskin, D. K., 1979, A method of preparing phyto-
738-744. clasts for vitrinite reflectance analysis:
Journal of Sedimentary Petrology, v. 49, no.
Ataman, G., and Baysal, O., 1978, Clay mineralogy 2, p. 633-635.
of Turkish borate deposits: Chemical Geology,
V. 22, p. 233-247. Blakey, J. F., 1966, Temperature of surface
waters in the coterminous United States: U.S.
Atwater, T., and Molnar, P., 1973, Relative mo- Geological Survey Hydrologic Investigations
tion of the Pacific and North American plates Map HA-235, scale 1:5,000,000.
deduced from sea-floor spreading in the Atlan-
tic, Indian, and south Pacific Oceans: in R Borchert, H., 1969, Principles of oceanic salt
L. Kouach and A. Nur, eds., Tectonic problems deposition and metamorphism: ~eologicalSocie-
of the San Andreas fault system: Stanford ty of America Bulletin, v. 80, no. 5, p. 821-
University Publications in Geological Science, 864.
V. 13, p. 136-148.
Bowser, C. J. , 1966, Chemical zonation of the
Axelrod, D. I., 1940, A record of Lyonothamnus in borates of Kramer, California: & J. L. Rau,
Death Valley, California: Journal of Geology, ed., 2nd Symposium on Salt, v. 1, Northern
V. 48, p. 526-531. Ohio Geological Society, Cleveland, p. 122-
Barker, C. E., 1980, The Terry borate deposit,
Amargosa Valley, Inyo County, California: Brown, W. W., and Jones, D. D., 1971, Borate
California Geology, v. 33, no. 8, p. 181-187. deposits of Turkey: in A. S. Campbell, ed.,
Geology and History of Turkey, Petroleum Ex-
, 1982, A rapid method for concentrating ploration Society of Libya, Tripoli, p. 483-
sedimentary organic matter for vitrinite re- 492.
flectance analysis: Journal of Sedimentary
Petrology, v. 52, no. 2, p. 663-664. Bruce, C. H., 1984, Smectite dehydration--its
relation to structural development and hydro-
, 1983, Influence of time on metamorphism of carbon accumulation in northern G U M of Mexico
sedimentary organic matter in liquiddominated basin: American Association of Petroleum Geol-
geothermal systems, western North America: ogists Bulletin, v. 68, no. 6, p. 673-683.
Geology, v. 11, no. 7, p. 384-388.
Chorlton, L. B., 1973, The effect of boron on
Barker, C. E., and Reynolds, T. J., 1984, Prepar- phase relations in the granite-water system:
ing doubly-polished sections of temperature unpubl. M.S. thesis, McGill University, 95 p.
sensitive rocks: Journal of Sedimentary Pe-
trology, v. 54, no. 2, p. 635-636. Christ, C. L., Clark, J. R., and Evans, H. T.,
Jr., 1958, Studies of borate minerals (111):
Barker, C. E., and Pawlewicz, M. J., 1985, The The crystal structure of colemanite,
correlation of vitrinite reflectance with C ~ B ~ O ~ ( O'H20:
H) Acta Crystallographia, v.
maximum temperature in humic kerogen: in G. 11, p. 76i-3770.
Buntebarth and L. Stegena, eds., Paleogeother-
mics, Springer-Verlag, New York, in press. Christ, C. L., and Garrels, R. M., 1959, Rela-
tions among sodium borate hydrates at the Kra-
Barker, J. M., and Wilson, J. L., 1976, Borate mer deposit, Boron, California: American Jcur-
deposits in the Death Valley region: Guide- nal of Science, v. 257, p. 516-528.

Christ, C. L., Truesdell, A. H., and Erd, R., , 1977, Chemistry and Geothermal Systems:
1967, Borate mineral assemblages in the system Academic Press, New York, 392 p
NaZO-Ca0-Mg0-B203-H20: Geochimica et Qsmo-
chimica Acta, v. 31, p. 313-337. Eugster, H. P., 1980, Geochemistry of evaporitic
lacustrine deposits: in F. A. Donath, ed.,
Clark, J. R., and Appleman, D. E., 1964, Penta- Annual Review of Earth and Planetary Sciences,
borate polyanion in the crystal structure of V. 8, p. 35-63.
ulexite, N ~ C ~ B ~ O ~ ( O H ) ~ 'Science,
~ H ~ O : v. 145,
p. 1295-1296. Eugster, H. P., and Hardie, L. A., 1975, Sedimen-
tation in an ancient playa-lake complex: the
Collinson, J. D., 1978, Lakes: in H. G. Reading, Wilkins Peak member of the Green River Forma-
ed., Sedimentary ~nvironmexsand Facies, tion of Wyoming: Geological Society of America
Elsevier, New York, p. 61-79. Bulletin, v. 86, p. 319-334.
Countryman, R. L., 1977, The subsurface geology, Eugster, H. P., and Hardie, L. A., 1978, Saline
structure, and mineralogy of the Billie borate lakes: in A. Lerman, ed., Lakes--Chemistry,
deposit, Death Valley, California: unpubl. ~eology,Physics:Springer-Verlag, New York,
M.S. thesis, University of California, Los p. 237-293.
Angeles, 129 p.
Evans, J. R., Taylor, G. C., and Rapp, J. S.,
Curtis, C. D., 1977, Sedimentary geochemistry: 1976, Mines and mineral deposits in Death
environments and processes dominated by in- Valley National Monument, California: Cali-
volvement of an aqueous phase: Philos-ical fornia Division of Mines and Geology Special
Transactions of the Royal Society of London, Report 125, 61 p.
V. A286, p. 353-372.
Fleck, R J., 1970, Age and tectonic significance
Dean, W. E., and Fouch T. D., 1983, Lacustrine of volcanic rocks, Death Valley area, Califor-
environments: in P. A. Scholle, D. G. Bebout, nia: Geological Society of America Bulletin,
and C. H. ~ooreTeds.,Carbonate Depositional v. 81, p. 2807-2816.
Environments, American Association of Petro-
leum Geologists, Tulsa, p. 98-130. Foshag, W. F., 1921, The origin of the colemanite
deposits of California: Economic Geology, v.
Dickson, F. W., and Raab, W. J., 1965, Origin of 16, p. 1199-1214.
stibnite associated with borax at Kramer,
California (abs.): Geological Society of Amer- Foscolos, A. E., and Powell, T. G., 1979, Minera-
ica Annual Meeting Program, p 4546. logical and geochemical transformation of
clays during burial diagenesis (catagenesis):
Drever, J. I., 1982, The Geochemistry of Natural relation to oil generation: Developments in
Waters: PrenticefIall, Ehglewood Cliffs, New Sedimentology, v. 27, p. 261-270.
Jersey, 388 p.
Foscolos, A. E., Powell, T. G., and Gunther, P.
Droste, J. B., 1961, Clay minerals in sediments R., 1976, The use of clay minerals and inor-
of Owens, China, Searles, Panamint, Bristol, ganic and organic geochemical indicators for
Cadiz, and Danby Lake basins, California: evaluating the degree of diagenesis and oil
Geological Society of America Bulletin, v. 72, generating potential of shales: Geochimica et
no. 11, p. 1713-1722. Cosmochimica Acta, v. 40, p. 953-966.
Dypvik, H., 1983, Clay mineral transformations in Fouch, T. D., and Dean, W. E., 1982, Lacustrine
Tertiary and Mesozoic sediments from North and associated clastic depositional environ-
Sea: American Association of Petroleum Geolo- ments: in P. A. Scholle and D. Spearing, eds.,
gists Bulletin, v. 67, no. 1, p. 160-165. Sandstone Depositional Environments, American
Association of Petroleum Geologists, Tulsa, p.
Eaton, G. P., 1980, Geophysical and geological 87-114.
characteristics of the crust of the Basin and
Range province: in B. C. Burchfiel, J. E. Gaft, M. L., Gorobets, B. S., Malinko, S. V.,
Oliver, and L. T. Silver, chmn., Continental Nekrasova, R A., 1977, Fluorescence of boron
Tectonics, National Academy of Sciences, minerals: Acad. Sci. USSR, Dokl., Earth Sci.
Washington, D.C., p. 96-110. Section, v. 236, no. 1-6, p. 164-167.
Ellis, R J., 1979, Explored geothermal systems: Gale, H., S., 1912, The Lila C borax mine at
in H. L. Barnes, ed., Geochemistry of Hy-
- Ryan, California: Mineral Resources of the
drothermal Ore Deposits, 2nd ed., J. Wiley, United States, 1911, p. 861-865, 2 pl.
New York, p. 632-683.
, 1913, The origin of colemanite deposits:
Ellis, A. J., and Mahon, W. A. J., 1967, Natural U.S. Geological Survey Professional Paper 85,
hydrothermal systems and experimental hot p. 3-9.
water/rock interactians (part 11): Geochimica
Cosmcchimica Acta, v. 31, p 519-538. Gates, G. R., 1959, Clay mineral composition of
borate deposits and associated strata at Bo-

ron, California: Science, v. 130, p. 102. mineralogical and chemical evidence: ~eologi-
cal Society of America Bulletin, v. 87, no. 5,
Graham, E. R., 1957, The weathering of some bo- p. 725-737.
ron-bearing materials: Soil Science Society of
America Proceedings, v. 21, p. 505-507. Hunt, C. B., and Mabey, D. R., 1966, General
geology of Death Valley, California-stratig-
Gretener, P. E., 1981, Geothermics: Using temper- raphy and structure: u.S. Geological Survey
ature in hydrocarbon exploration: American Professional Paper 494-A, p. ~l-M65.
Association of Petroleum Geologists Course
Notes, v. 17, 156 p. Hunt, C. B., and others, 1966, General geology of
Death Valley, California--hydrologic basin:
Gulensoy, H., and Kocakerim, M. M., 1978, Solu- U.S. Geological Survey Professional Paper 494-
bility of colemanite mineral in C02-containing B, p. B1-B138.
water and geological formation of this miner-
al: Bulletin of the Mineral Research and Ex- Hurlbut, C. S., Aristarain, L. F., and Erd, R.
ploration Institute of Turkey, foreign edi- C., 1973, Kernite from Tincalayu, Salta, Ar-
tion, no. 90, p. 1-19. gentina: American Mineralogist, v. 58, p 308-
Haines, D. V., 1959, Core logs from Searles Lake,
San Bernardino Co., California: U.S. Geologi- Inan, K., 1972, New brate district, Eskishehir-
cal Survey Bulletin 1045-E, p. 139-317. Kirka Province Turkey: Transactions/~ectionB,
Institute of Mining and Metallurgy, v. 81, p.
Hanor, J. S., 1979, The sedimentary genesis of B163-B165.
hydrothermal fluids: in H. L. Barnes, ed.,
Geochemistry of Hydrothermal Ore Depsits, 2nd Inan, K., Dunhan, A. C., and Esson, J., 1973,
ed., J. Wiley, New York, p. 137-172. Turkey: mineralogy, chemistry and origin of
Kirka borate deposit, Eskishehir Province:
Hanshaw, B. B., 1963, Preliminary relations in Institute of Mining and Metallurgy, Transac-
the system Na2B407-Ca2 B6Ol1-H20: U.S. Geolog- tions, Section B (Aug.), p. B114-B123.
ical Survey Professional Paper 475-B, p. B24-
B27. Jones, B. F., 1965, The hydrology and mineralogy
of Deep Springs Lake, Inyo County, California:
Hardie, L. A., 1968, The origin of the Recent U.S. Geological Survey Professional Paper 302-
non-marine evaporite depsit of Saline Valley, A, 56 p.
Inyo County, California: Gecchimica et Cosmo-
chimica Acta, v. 32, p. 1279-1301. , 1966, Geochemical evolution of closed
basin water in the western Great Basin: & J.
, 1984, Evaporites: marine or non-marine: C. Rau, ed., Second Symposium on Salt, North-
American Journal of Science, v. 284, p. 193- ern Ohio Geological Society, Cleveland, p.
240. 181-200.
Hardie, L. A., and Eugster, H. P., 1970, The Karazhanov, N. A., 1963, The kinetics of the
evolution of closed-basin brines: Mineralogi- dissolution of borates and other natural
cal Society of America Special Publication, v. salts: Journal of ~ppliedChemistry, U.S.S.R.,
3, p. 273-290. v. 36, p. 2560-2567.
Hardie, L. A., Smoot, J. P., and Eugster, H. P., Keller, W. D., 1963, Diagenesis in clay miner-
1978, Saline lakes and their deposits: a sedi- als--a review: International Series, Mono-
mentological approach: International Associa- graphs in Earth Sciences, v. 13, p. 136-157.
tion of Sedimentologists Special Publication,
V. 2, p. 7-41. Kemp, P. H., 1956, The Chemistry of Borates: Part
I: Borax Consolidated Ltd., London, p. 1-90.
Helvaci, C., 1984, Occurrence of rare borate
minerals: veatchite-A, tunellite, teruggite Kendall, A. C., 1984, Continental and supratidal
and cahnite in the Emet brate deposits, Tur- (sabka) evaporites: R. G. Walker, ed.,
key: Mineralium Depsita, v. 19, p. 217-226. Facies Models, 2nd ed., Geoscience Canada
Reprint Series, v. 1, p. 259-296.
Helvaci, C., and Firman, R. J., 1976, Geological
setting and mineralogy of Emet brate deps- Kinsman, D. J. J., 1976, Evaporites: relative
its, Turkey: Trans. Inst. Mining Metall. Sec- humidity control of primary mineral facies:
tion B, Appl. Earth Sci. 85B, p. 142-152. Journal of Sedimentary Petrology, v. 46, p.
Hem, J. D., 1972, Magmatic water: in R. W. Fair-
bridge, ed., Encyclopedia of ~ a e m i s t r yand Kistler, R. B., and Smith, W. C., 1975, Boron and
Ehvironmental Sciences, Van Nostrand Reinhold, borates: in S. J. Lefond, ed., Industrial
New York, p. 666. Minerals a Rocks, 4th Ed., Society of Mining
Engineers of AIME, New York, p 473496.
Hower, J., and others, 1976, Mechanism of burial
metamorphism of argillaceous sediment: 1. Kitano, Y., Okumura, M., and Idogaki, M., 1978,

Coprecipitation of borate-boron with calcium Cleveland, p. 151-159.

carbonate: Geochemical Journal, v. 12, p. 183-
189. Norman, J. C., and Johnson, F. C., 1980, The
Billie borate orebody, Death Valley, Califor-
, 1979, Influence of borate-boron on crystal nia: in D. L. Fife and A. R. Brown, eds.,
form of calcium carbonate: Geochemical Journ- ~ e o l e a n dMineral Wealth of the California
al, v.13, p. 223-224. Desert, Smth Coast Geolqical Society, Santa
Ana, p. 268-275.
Krauskupf, K. B., 1967, Introduction of Geochem-
istry: Md;raw+Iill, New York, 721 p. Oremland, R. S., and Des Marais, D. J., 1983,
Distribution, alxmdance, and carbon isotopic
Langbein, W. B., 1961, Salinity and hydrology of composition of gaseous hydrocarbons in Big
closed lakes: U.S. Geological Survey Profes- Soda Lake, Nevada: An alkaline, meromictic
sional Paper 413, 20 p. lake: Geochimica et Cosmochimica Acta, v. 47,
p. 2107-2114.
Lerman, A., 1979, Geochemical Processes, Water
and Sediment Environments: J. Wiley, New York, Ozol, A. A., 1977, Plate tectonics and the pro-
481 p. cesses of volcanqenic-sedimentary format ion
of boron: An SSSR Izvestiya, ser. Geol., no.
Link, M. H., Roberts, M. T., and Newton, M. S., 8, p. 68-75 (Russian text; transl., 1978,
1985, Walker Lake basin: an example of late International Geology Review, v. 20, no. 6).
Tertiary(?) to Recent sedimentation in a basin
adjacent to an active, strike-slip fault: in Renner, J. L., and others, 1976, Selected geo-
K. T. Biddle and N. Christie-Blick, edsT thermal resources data: hydrothermal convec-
Strike-slip Deformation, Basin Formation and tion systems in the states of Alaska, Arizona,
Sedimentation, Society of Ewnomic Paleontolo- California; Colorado, Hawaiai, Idaho, Montana,
gists and Mineralqists, Special Publication, Nevada, New Mexico, Oregon, Utah, Washington,
in press. and Wyoming: U.S. Department of Commerce,
National Technical Information Service, pbli-
Lisitsyn, A., Ye., and Malinko, S. V., 1971, cation PB-250 377, 352 p.
Physimhemical conditions for develupment of
borosilicates and borates in limy skarn forma- Rettig, S. L., Jones, B. F., and Risacher, F.,
tion: International Geology Review, v. 13, no. 1980, Geochemical evolution of brines in the
12, p. 1773-1789. Salar of Uyuni, Boliva: Chemical Geology, v.
30, p. 57-79.
Malinko, S. V. and others, 1976, Inclusions of
boron-bearing hydrothermal solutions in quartz Risacher, F., and Eugster, H. P., 1979, Holocene
crystals based on (n, alpha) radiography: pisoliths and encrustations associated with
Acad. Sci. USSR, Dokl., Earth Sci. Section. v. spring-fed surface pools, Pastos Grandes,
228, no. 1-6, p. 117-120. Bolivia: Sedimentology, v. 26, p 253-270.
McAllister, J. F., 1970, Geology of the Furnace Rodgers, A. F., 1919, Colemanite pseudomorphous
Creek borate area, Death Valley, Inyo County, after inyoite from Death Valley, California:
California: California Division of Mines and The American Mineralogist, v. 4, p. 135-139.
Geology Map Sheet 14, 9 p., 1:24,000.
Roedder, E., 1984, Fluid inclusions: Mineralqi-
, 1973, Geologic map and sections of the cal Society of America Reviews in Mineralogy,
Amargosa Valley borate area-southeat contin- v. 12, 644 p.
uation of the Furnace Creek area-Inyo County,
California: U.S. Geological Survey Map 1-782, Sass, J. H., and Munroe, R. J., compilers, 1974,
1:24,000. Basic heat flow data from the United States:
U.S. Geological Survey Open-file Report 74-9,
, 1976, Columnar section of the main part of nonconsecutive pages.
the Furnace Creek Formation of Piocene (Clar-
endonian and Hemphillian) age across Twenty Savage, D. E., and Russell, D. E., 1983, Mamma-
Mule Team Canyon, Furnace Creek borate area, lian Paleofaunas of the World: Addison-Wesley
Death Valley, California: U.S. Geological Publishing Co., Reading, Mass., 432 p.
Survey Open-file Report 76-261, 1 p.
Savin, S. M., 1977, The history of the earth's
Muessig, S., 1958, First known occurrence of surface during the past 100 million years:
inyotte in a playa, at Laguna Salinas, Peru: Annual Reviews of Earth and Planetary Sci-
The American Mineralogist, v. 43, p. 1144-147. ences, v. 5, p. 319-355.
, 1959, Primary borates in playa deposits: Selley, R C., 1976, An Introduction to Sedimen-
Economic Geology, v. 54, p. 495-501. tology: Academic Press, New York, 408 p.
, 1966, Recent South American borate d e p - Seyfried, W. E., Janecky, D. R , and Motti, M.
its: in J. L. Rau, ed., Second Symposium on J., 1984, Alteration of the oceanic crust:
~alt,v.1, Northern Ohio Geological Society, implications for geochemical cycles of lithium

and boron: Geochimica et Cosmhimica Acta, v. Formation and Occurrence, 2nd ed.: Springer-
48, no. 3, p. 557-569. Verlag, Berlin, 699 p.
Silberman, M. L., and others , 1972, K-Ar ages of Tourtelot, H. A., 1979, Black shale--its deposi-
volcanic rocks and gold-bearing quartz-adular- tion and diagenesis: Clays and Clay Minerals,
ia veins in the Bodie Mining District, Mono v. 27, no. 5, p. 313-321.
County, California: Economic Geology, v. 67,
p. 597-604 Walker, C. T., ed., 1975, Geochemistry of Boron:
Benchmark Papers in ~ e o l o g ~ / 2 3Dowden,
Sims, J. R., and Bingham, F. T., 1968, Retention Hutchinson, and Ross, Stroudsburg, Penn., 414
of boron by layer silicates, sesquioxides, and P
soil materials, I1 sesquioxides: Soil Science
Society of America proceedings, v. 32, p. 364- Waring, G. A., 1965, Thermal springs of the
369. United States and other countries of the
world--a summary: U.S. Geological Survey P r e
Smith, G. I., 1966, Geology of Searles Lake-- fessional Paper 492, 383 p.
guide to prospecting for buried continental
salines: fi J. L. Rau, ed., 2nd Symposium on Weaver, C. E., 1967, The significance of clay
Salt, Northern Ohio Geological Society, Cleve- minerals in sediments: B. Nagy and V. Co-
land, p. 167-180. lumbo, eds., Fundamental Aspects of Petroleum
Geochemistry, Elsevier Publishing Co., New
, 1979, Subsurface stratigraphy and geochem- York, p. 37-76.
istry of late Quaternary evaporites, Searles
Lake, California: U.S. Geological Survey Pro- Werre, R. W., Jr., and others, 1979, A novel gas-
fessional Paper 1043, 130 p. flow fluid inclusionheating-freezing stage
(abs.): Geological Society of America Ab-
Smith and others, 1970, Salt crystallization stracts with Program 11, p. 539.
temperatures in Searles Lake, California:
Mineralogical Society of America Special Pa- White, D. E., 1957, Thermal waters of volcanic
per, v. 3, p. 257-259. origin: Geological Society of America Bulle-
tin, v. 68, p. 1637-1658.
Sonnenfeld, P., and Hudec, P. P., 1980, Helio-
thermal lakes: in A. Nissenbaum, ed., Hyper- Wilson, J. L., 1976, Geology and engineering as-
saline Brines and Evaporitic Environments, pects of Boraxo pit, Death Valley, California:
Elsevier, New York, p. 93-100. unpubl. M.S. thesis, University of Southern
California, 95 p
Stach, E., and others, 1982, Stach's Textbook of
Coal Petrology: Gebruder Borntraeger, Berlin, Wilson, J. L., and Emmons, D. L., 1977, Origin
535 p. and configuration of the oxidized zone in
Tertiary formations, Death Valley region,
Sternfeld, J., Keskinen, M., and Blethen, R., California: Geology, v. 5, no. 11, p. 696-698.
1983, Hydrothermal mineralization of a Clear
Lake geothermal well, Lake County, California: Yakubova, V. V., 1952, An experiment in the study
Geothermal Resources Council Transactions, v. of inclusions in minerals of pegmatites from
7, p. 193-197. Murzinka (~rals ) : Mineral. Muzeya Akad. Nauk
SSSR Trudy, v. 4, p. 102-121 (in Russian).
Stewart, J. H., 1978, Basin-range structure in
western North America: a review: Geological
Society of America Memoir 152, p. 1-31, plate
-, 1955, An attempt at study of inclusions in
minerals of Mursinka (~ral)pegmatites. Min-
1-1. eral Muzeya Akad. Nauk SSSR Trudy, v. 7, p.
132-150 (in Russian; translated in Inter-
Suchecki, R. K., Perry, E. A., and Hubert, J. F., national Geology Review, 1959, v. 1, no. 8, p.
1977, Clay petrology of Cambro-Ordovician 52-66).
continental margin, Cow Head klippe, western
Newfoundland: Clays and Clay Minerals, v. 25, Yeh, H., and Savin, S. M., 1977, Mechanisms of
p. 163-170. burial metamorphism of argillaceous sediments:
3. 0-isotope evidence: Geological Society of
Taylor, H. P., 1979, Oxygen and hydrogen isotope America Bulletin, v. 88, no. 9, p. 1321-1330.
relationships in hydrothermal mineral depos-
its: in H. C. Barnes, ed., Geochemistry of Zoback, M. L., Anderson, R. E., and Thompson, G.
~~drothermal Ore Deposits, 2nd ed., J. Wiley, A., 1981, Cainozoic evolution of the state of
New York, p. 236-277. stress and style of tectonism of the Basin and
Range province ofthe westernunited States:
Thomas, R., and Bauman, L., 1980, Ergebnisse von Philosophical Transactions of the Rayal Soci-
thermometrischen und kryometrischen Untersu- ety of London, v. 300, p. 407-434.
chungen an Kassiteriten des Erzgebirges: Z.
geol. Wiss., Berlin, v. 10, p. 1281-1299.
Tissot, B. P., and Welte, D. H., 1984, Petroleum

APPENDIX I ples of both fracture and polished borate sur-

faces. The samples were analyzed on a Jeol Model
35C SEM at 25 kv and 100 ma. Qualitative elemen-
tal analysis to confirm the mineralogy of the
sample was by a Kevex energy dispersive spectrom-
Fluid Inclusions eter interfaced with the Jeol SEM.
Fluid inclusion temperature determinations
were made in a U.S. Geological Survey gas flow
heating stage (Werre and others, 1979). The
heating stage was calibrated using artificial Polished thin sections (Barker and Reynolds,
fluid inclusions (available from Synflinc Inc., 1984) of selected borate ores were analyzed using
State College, PA) with k m w n homogenization tem- a Nuclide ELM-2a Luminoscope. Operating condi-
perature (Th) and composition allowing direct tions were a chamber pressure of 15-35 thermo-
measurement of stage gradients and the necessary wuple microns, acceleration potential of 16-17
temperature corrections. These standards were kv, and beam current of 0.4-0.6 ma.
used to calibrate the stage in the range of -
56.6O~to +374.1c. The thermocouple was cali-
brated to the digital thermometer (outside of the Vitrinite Reflectance
stage) using the freezing point of mercury,
water, zinc, tin, and aluminum over a range of - Sedimentary organic matter (OM) was prepared
39% to +660c. These measurements indicate the from selected borate samples by digestion in
accuracy of the temperature determination is f
1c in heating mode and 0.2O~ in freezing mode.
Selected borate samples were doubly polished into
dilute HC1 to remove carbonate and acid soluble
saline minerals. OM was isolated from the acid
insoluble residue (Barker, 1982) and mcunted and
rock sections suitable for fluid inclusion analy- polished askin in, 1979). Reflectance was mea-
sis using the method of Barker and Reynolds sured on a Zeiss Universal microscope fitted with
(1984). Roedder (1984) discusses the petrography a MPM-Ol microphotometric system restricted with
of fluid inclusions and the interpretation of a pinhole diaphram to read a 3 micron spot on the
phase transition temperatures. sample at 500x total magnification using a
40x/0.85 n.a. lens under oil immersion ( N ~ - -- ~ ~ ~
1.518). The system was calibrated using a Zeiss
Scanning Electron Micros- leuwsapphire standard that has a reflectance of
0.58 percent when illuminated by filtered 546 nm
Scanning electron microscope (SEM) observa- light.
tions utilized palladium-gold alloy coated sam-