American Mineralogist, Volume 71, pages 428439, 1986

Mineralogy and paragenesis of "pocket" clays and associated minerals incomplex granitic pegmatiteso San Diego County, California

EucrNn E. Foono. H.lRRv C. Srlnxnv.rJosnpn E. Tlcclnr" Jn.U.S. Geological Survey, Denver, Colorado 80225

Ansrru,cr

The gem- and specimen-bearing, complex granitic pegrnatite and aplite dikes and bodiesof Cretaceous age emplaced into the Southern California batholith locally contain fracturesand cavities ("pockets") that are usually clay filled. Some pockets, however, contain onlya thin coating of cookeite or sericite, deposited as "snow on the roof." Other pocketscontain pseudomorphs of lepidolite after elbaite, bavenite after beryl, and clays and micasafter spodumene. Pseudomorphism took place under nearly closed system conditions attemperatures similar to those at the time of pocket formation. Most pockets contain frag-mented primary minerals enclosed in a matrix of Ca-Na zeolites; white, pink, or redbeidellite; Li-tosudite; Ca-Mg montmorillonite; and rare calcite. Paragenetic relationshipsindicate a decrease in Na relative to Ca during the crystallization of zeolites and a decreasein Li and Al-concomitant with an enrichment of Ca, Mg, and Si-for the layer silicates.The most prevalent minerals found are stilbite, laumontite, cookeite, Li-tosudite, beidellite,Mg-Ca montmorillonite, palygorskite, and calcite. Sparse amounts of nontronite, heuland-ite, and todorokito have also been identified. Deposition ofzeolites, clays, and carbonatesin open pockets took place subsequent to crystallization of the primary pocket minerals,under hydrothermal conditions with temperatures ranging from approximately 400 to150'C. The major amounts of Ca and Mg, represented by as much as 8.4 wto/o MgO inpalygorskite and more than l0 wto/o CaO in minerals such as laumontite and bavenite,were provided by hydrothermal alteration ofgabbro-norite and/ortonalite host rocks. Red-brown (iron-stained) kaolinite is locally present as a final phase of deposition in somepockets and is probably indicative of a transition from an alkaline to an acidic environment.

INrnooucrroN AND STATEMENT oF pRoBLEM

The gem-bearing complex granitic pegmatites of SanDiego and Riverside Counties, California, have been ob-jects ofstudy since their discovery in the 1870s. Exploi-tation, initially for gems and later for specimen-qualityminerals, began in 1898 with the discovery of the Himala-ya dike system in the Mesa Grande district (Rynerson,1967; Foord, 1976). Scientific studies of the pegmatitesbegan about 1903 and include those of Kunz (1905),Schaller (1925), Jahns and Wright (1951), Hanley (1951),Jahns (1954), Simpson (1965), and Foord (1976). Jahns(1979) presented a summary of the Stewart and Himalayamines. From the time that mining first began, it was notedthat the best-quality minerals occurred in specific portionsof pegmatite dikes and bodies and were frequently con-centrated in crystal-lined, clay-filled, or open voids andcavities referred to as "pockets" by the miners. Thesecavities range from several centimeters across to exten-sive, multiply-projecting, and branching cavities as muchas 3 m or more in maximum dimension. Many pocketscontain fragments of minerals that were once attached to

t Present address: I 636 South Yarrow Court, lakewood, Col-orado 80226.0003{04v86/03044428$02.00 428

the pocket walls but are now "floating" in a matrix ofclays, zeolites, and other low-temperature minerals. A fewpockets, referred to as "dry pockets," contain no clays,zeolites, or carbonates, but do contain small amounts ofcookeite as a coating on the minerals present. Coatings ofcookeite, with or without associated minerals, that arefound only on the top surfaces of various primary pocketminerals such as quartz, feldspars, micas, and tourmaline,and not underside or overhanging surfaces are known as"snow-on-the-roof' coatings. Pockets in the Himalayadike system are strongly controlled vertically by dike ge-ometry and zonation, but they may extend laterally, inthe plane of the dike, for 3 to as much as 30 m or more.Additional details on the Himalaya dike system and thepockets themselves are given in Foord (1976,1977) and,Jahns (1979). Some pocket clays appear to be partial orcomplete replacements (pseudomorphs) of pre-existingminerals, most frequently beryl, elbaite, and Nb-Ta oxides(Himalaya dike system) and spodumene and phosphates(Pala district). Apart from that of Herbert (1982), little orno specific attention has been devoted to the detailed claymineralogy of the pocket clays. Most of the earlier studieswere done prior to use of X-ray powder diftaction foridentification purposes. Of the early work, only that ofSchaller (1905, 1925) and Laudermilk and Woodford

FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES 429

(1934) included chemical analyses ofpocket clays frompegmatites in the Southern California batholith. As earlyas 1880, howevor, Brush and Dana (1880) had examinedclays and other minerals pseudomorphous after spodu-mene in pegmatites at Branchville, Connecticut. The pinkclay analyzed by Schaller (1905) was termed "halloysite"by Ross and Shannon (1926) after optical study. Kaolinitepseudomorphs after spodumene were described bySchwartz (1937) from the Etta mine, Black Hills. BothSchallor (1925) and l,audermilk and Woodford (1934)attributed the origin of the pocket clays to weatheringrather than to hydrothermal action.

Schaller (1925) identified two types of pocket clay inthe Pala district (l) a white, gray, or pink clay, derivedfrom decomposition of the pegmatite minerals, and (2) adeep red-brown, sticky clay, derived in part from the sur-rounding gabbro-norite. Jahns and Wright (1951) attrib-uted their formation to both hypogene and supergene pro-cesses. The hypogene clays contained members of boththe smectite and kaolinite groups; the supergene clayscontained only members of the kaolinite group. Foord(197 6, 1977) found white to pink Mg-Ca montmorillonitein pockets within the Himalaya dike system, Mesa Grandedistrict. Red-brown, disordered kaolinite was also de-scribed, and was clearly deposited later than the lighter-colored montmorillonite.

The following questions are addressed in this paper: (l)What were the sources of the elements incorporated intothe pocket-filling clays and pseudomorphous minerals? (2)How, when, and under what conditions were the mineralsformed? (3) What is the paragenesis? Samples of late-stagesecondary minerals and pocket fillings were collected fromfour of the major pegmatite districts in San Diego Countyfor mineralogic and petrologic study. In addition, samplesof the host norite were collected from the Himalava dikesystem for chemical and mineralogical study.

S,lprpr-rs AND ANALyTTcAL METHoDS

Thirty-nine of fifty-three samples examined (Table l), are fromthe Himalaya pegmatite-aplite dike system. Two samples of claypseudomorphous after spodumene, variety kunzite, were ex-amined from the Katrina mine, on Hiriart Mountain; five sam-ples ofpseudomorphous and pocket clays are from the OceanView (Elizabeth R) mine on Chief Mountain; one sample of deeppurple-red clay is from the White Queen mine, Hiriart Mountain;and one sample of red-brown pocket-filling clay is from the Tour-maline Queen mine, Queen Mountain. All are from the Paladistrict. Four samples of pocket clay from the main dike and onefrom the "garnet dike" on the Little Three mine, Ramona district,and one sample of material coating pocket minerals from theMaple Lode mine, Aguanga Mountain, were examined as well.Locations of the four districts, as well as the individual mines,are given in Jahns and Wright (1951), Jahns (1979), Simpson(1965), and Foord (1976).

All samples were prepared for X-ray diffraction analysis ac-

'? To receive a copy of Table I, order Document AM-86-296from the Business Office, Mineralogical Society of America, 1625I Street, N.W., Suite 414, Washington, D.C. 20006. Please remit$5.00 in advance for the microfiche.

cording to accepted clay mineral procedures (Starkey et al., I 984).Bulk splits of samples were first X-rayed as unoriented packs.

Then the clay-sized (<2 pm) fraction was removed by settling inwater or centrifuging and run as air-dried, randomly orientedpatterns. Oriented patterns were made after air-drying al 25'C,after glycolation at 60.C, and heating at 400 and 550"C for 0.5h each. Other size fractions for approxirnately 30 samples, suchas >2, l-2,0.5-1, 0.24.5, and <0.2 pm, were also examinedby X-ray difraction. Some samples were examined at other tem-peratures and conditions when necessary (e.g., Greene-Kelly testor treatment with glycerol). Many of the samples were examinedwith a petrographic microscope to confirm rnineral identifica-tions and estimate relative amounts of individual minerals inmineral mixtures.

Twenty mineral separates and rock samples were analyzed formajor elements by X-ray fluorescence spectroscopy using a Philips3PWl600 simultaneous wavelength-dispersive spectrometer ac-cording to the method of Taggart et al. (1981). Samples wereincorporated in lithium tetraborate fusion discs prepared by themethods of Taggart and Wahlberg (1980a, 1980b). Because ofthe difrculty of purifying the clay fractions, most of the mineralseparates were analyzed as a half-weight sample (0.4 g); rocksamples and some of the more abundant clay separates wereanalyzed as normal full-weight samples (0.8 g). Six-step semi-quantitative emission spectrographic analyses were made of fif-teen samples of clay and other minerals. Rare alkalis (Li, Rb, Cs)were spectrographically determined for three samples. LirO wasdetermined by induction-coupled plasma spectroscopy for fiveselected samples known to contain Li.

BlcxcnouND TNFoRMATIoN AND oBSERvATToNS

In the Pala and Mesa Grande districts, pegmatite dike

emplacement was along nearly flat lying planes of weak-

ness in the oldest unit ofthe Southern California batholith,

the San Marcos Gabbro (Krummenacher et al., 1975;

Foord, 1976). In the Ramona district, the slightly younger

Green Valley Tonalite is the host to pegmatite-aplite dikes.

The older, higher-temperature units of the batholith be-

haved as brittle media with development of sheet-struc-

ture in response to a combination of intrusive pressures

from below and tectonic unloading.A "snow-on-the-roof ' coating of cookeite, Li-

Alo(Si3Al)O,0(OH)r, andlor fine-grained K-feldspar with

or without albite is found usually only covering minerals

on the bottom of pocket cavities. Pocket roofs are gen-

erally completely free of these coatings. Deposition from

stagnant and rest, subcritical liquids may be inferred from

this widespread material because of its physical distri-

bution as described above, the very fine grain size, and

in some cases, an unusual chemical composition, as at theLittle Three mine where in one large pocket a boron-richK-feldspar possibly representing a pressure-quench prod-

uct was deposited. Furthermore, rotation to the west of

the batholith as a whole is indicated by the distribution

of the "snow-on-the-roof'on the pocket minerals, which

was deposited under a vertical gravity gradient, and by

the present average 30" dips ofthe pegmatite-aplite dikes

in that direction Goord. 1976).

3 Use ofbrand names is for descriptive purposes only and doesnot constitute endorsement by the U.S. Geological Survey.

430 FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES

Crystallization temperatures of primary minerals in theHimalaya and Little Three pegmatite-aplite dike systemsare thought to have ranged from approximately 700"C(basal aplite) down to 520-540'C at the central pocketzone and at pressures of 2 kbar (Taylor et al., I 979; Sternet al., 1986). Temperature and pressure estimates of 45G-500"C and 2-3 kbar for pocket formation have been madeby London (1986).

So-called "dry pockets" form less than l0lo ofall pocketsand are best developed in the Himalaya dike system. Thesepockets contain their primary minerals in varying statesof preservation, coated with cookeite or muscovite. Crys-tals of tourmaline, qtartz, feldspar, mica, beryl, apatite,and Nb-Ta oxides that have been broken offduring pocketrupture are all loosely or compactly cemented together inthe bottoms of such pockets. The cement may consist onlyof cookeite and rarely of muscovite, but more frequentlyof both primary minerals and the micas. Pockets of thistype, and particularly "clay-filled" pockets, may containa lower-temperature and lower-pressure assemblage ofzeolites, clay minerals, micas, borates, and carbonates.

The individual Himalaya and Little Three dikes, whichmay be from several centimeters to 2 m thick, are thinnerby an order of magnitude or more than many of the dikesatPala such as the Stewart (20 m thick), the Katrina (16m thick), and the Ocean View (10-13 m thick). Depthand temperature of emplacement of all of these dikes wasabout the same, 7 km and 730-700"C (Taylor et al., 1979)and under structural conditions as presented above. Themore competent and thicker (>2 m) dikes were not able(or incompletely able) to release their elevated internalvapor pressures in areas of final crystallization, and as aresult the final crystallizing products "stewed in their ownjuices" (Jahns and Burnham, 1969; Jahns, 1982, pers.comm.). Thinner dikes (<2 m), such as the Himalaya andLittle Three dikes, generally were able to release theirinternally derived vapor pressures and pocket fluids (Jahns,1981, pers. comm.). Thus, pocket cavities are well de-veloped in the Himalaya and Little Three dikes, and aminimum of mineral replacement has occurred, whereasin the Stewart, Katrina, and Ocean View dikes, pocketsare not as well developed, and mineral replacement hastaken place in both the pockets and the rest ofthe dikesthemselves.

Mineral replacement producing pseudomorphs of pock-et minerals, such as clay minerals after spodumene (kunz-ite) in some of the pegmatite bodies in the Pala district,appears to have occurred either prior to or in the absenceof pocket rupture. Large, thick, competent dikes-such asthe Stewart, the Ocean View, and the Katrina in the Paladistrict-contain spodumene, some of it as gem-qualitykunzite. In all ofthese dikes, the spodumene andlor kunz-ite is commonly etched and chemically corroded and isin a matrix of white, pink, or red clays and micas. Thematrix often perfectly preserves the cleavage ofthe pre-existing kunzite, and a spongeJike network is frequentlydeveloped with splinters and fragments of kunzite sur-rounded by pure clays and micas. Both Ca-Mg mont-

morillonite and cookeite or kaolinite have been observedto form as pseudomorphs after spodumene (Herbert, 1982).

Most replacement appears to be postrupture and thusat temperatures of less than 500"C, as in the case of theHimalaya dike system, where pseudomorphs of lepidoliteafter elbaite, bavenite after beryl, rynersonite and fersmiteafter stibiotantalite, as well as others have been found.Pollucite is found sparingly in the pocket zone of theHimalaya dike system (Foord, 1976) and is often asso-ciated with montmorillonite and other minerals, but mostpollucite does not appear to be altered to illite, kaolinite,qvartz, and montmorillonite (see, e.g., Cernf, 1978).K-feldspars are frequently corroded and solution etchedand have partial coatings of minerals such as cleaveland-ite, lepidolite, and cookeite. Some K-feldspar crystals havea sharp and distinctive overgrowth ofglassy ordered or-thoclase (Prince et al., 1973; Horsky and Martin, 1977).Elbaite crystals in some pockets arb found panially etched(never a mixture of unetched and etched together) andcorroded. Corrosion has occurred on the prism faces butnot the pyramidal or pedion faces. Some doubly termi-nated pink crystals of elbaite tipped by g.reen tourmalinehave the pink portion nearly completely removed, but thegreen ends are essentially untouched, lustrous, and smooth.Figure I shows a crystal of pink-green elbaite coated bymuscovite. The green cap is unattacked, and the termi-nation beneath the muscovite coating is sharp and lus-trous. Other tourmaline crystals show some corrosion onall parts, but the ends are much less affected. This is be-lieved to be a function of the chemical and structuralcharacter of tourmaline. Most crystals are partially orcompletely coated with cookeite or later-formed zeolitesand clays. The corrosion of the feldspar and tourmaline(and sometimes beryl) is believed to take place in pocketsthat have released their internal vapor pressure by rup-turing or slower release and then resealed themselves bysubsequent crystallization. All of these phenomena haveoccurred prior to deposition of cookeite from the finalstagnant rest liquid. Some pockets show evidence ofhav-ing adjusted to altered chemistry three or more times(Foord,1976).

Ruptured pockets may contain only zeolites and calcitethat cement previously crystallized minerals, or they maycontain some clay minerals as well. Usually, laumontiteand beidellite-montmorillonite are the dominant matrixminerals; however, some ruptured pockets appear to con-tain only cookeite and clay minerals, with little or nozeolites. In a few pockets, the presence of fibrous paly-gorskite, (Mg,Al)rSioO,o(OH)'4HrO, indicates a domi-nance of Mg as the divalent cation. An analysis of pure,white, elastic palygorskite from the Himalaya mine is giv-en in Table 2 (analysis l0). Pure masses of palygorskiteseveral centimeters thick or more and several meters inextent were found during mining in 1984.

Rare todorokite, (Mn2*,Ca,Mg,Ba,Sr)Mn{-Or'HrO,occurs either as a coating on nearly all other minerals oras concretionary pellets (Foord, 1976). Only red-brownkaolinite appears to be younger on the basis of textural

FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES 431

relationships. The todorokite is the only mineral foundin the Himalaya dike system to contain >0.1 wto/o eachof Sr, Ba, and Cu. Analysis 13 (Table 3) shows that 500ppm Ni, 150 ppm V, and 150 ppm Co are also presen!these latter elements were probably derived from hydro-thermal alteration of the host sulfide-bearing norite-gab-bro as they are present in the host and not the dike system.

Pseudomalachite, Cu'(PO4),(OH)4'HrO, and crypto-melane, K(Mna*,Mn2*)rO,., two late-stage minerals fromthe Ocean View mine, were identified as new to the SanDiego County pegmatites. The pseudomalachite occurs assparse, thin, discontinuous green fracture coatings in feld-spar and quartz and as cleavage fillings in muscovite.Cryptomelane occurs in the pocket zone as compact, dense,blue-black masses several centimeters or more across.Sparse malayaite, CaSnSiOr, was identified from the Hi-malaya mine. The mineral occurs as tiny white hemi-spheres and rosettes, <0.2 mm across, on top of elbaite,qvartz, cleavelandite, and K-feldspar.

Deposition of zeolites commenced with stilbite, Na-CarAlrSi,rO16. l4H2O, and/or heulandite, (Na,Ca)'-r-Al3(Al,Si)2Si,3O36. l2HrO (sparse), followed by laumon-tite, CaAlrSioorr.4HrO. Both stilbite and laumontite arecommon in the dike system, but do not always occurtogether. These same three zeolites were first reported asoccurring in pegmatite pockets in San Diego County byRogers (1909) from the Victor pegmatite at Rincon andwere later described as widespread but minor constituentsof fracture fillings and pockets, associated with clay min-erals in the Pala district by Jahns and Wright (1951).Stilbite is also present, coating pocket minerals at theMaple Lode mine, Aguanga Mountain (Table I, ML-l).In some pockets in the new Himalaya workings (1980-1984), early clear rhombohedral calcite was followed byplaty, opaque, nearly pure calcite contemporaneous withthe deposition of laumontite. Table 3 is a compilation ofsemiquantitative emission spectrographic analyses of theselate-stage pocket-filling minerals including stilbite, lau-montite, calcite, cookeite, montmorillonite, Li-tosudite,beidellite, palygorskite, todorokite, pseudomorphousbavenite and rynersonite (Foord and Mrose, 1978), fine-grained, locally porcelaneous aggregates of B-containingK-feldspar, and finally the orthoclase-sanidine(?) and al-bite that coat primary pocket minerals. The three feldsparcoatings (analyses l-3, Table 3) represent very late pre-cipitates on all pocket minerals on the floors of the pock-ets. The B-rich K-feldspar from the Little Three mine isunusual and not completely characterized. The two sam-ples of stilbite and laumontite (analyses 4 and 5) are typ-ical for the species and contain trace amounts of othermetals. There is little ditrerence in the chemistry of thetwo generations of calcite (analyses 6 and 7) from theHimalaya mine; traces of Mn, Sr, and Y are present inboth. The two beidellites and two montmorillonites (anal-yses 8-11) contain varying amounts of different elementsthat reflect details oftheir origin and some contamination.An analysis of thoroughly water-washed palygorskite(analysis 12) confirms that major (about 2.4 wto/o Na'O)

Fig. l. Etched and corroded tourmaline crystal showing pref-erential dissolution of the prism zone and coating of muscovite.

sodium was removed. The todorokite (analysis l3), thefirst occurrence to be reported from granitic pegmatites(Foord, I 97 6), contains I . 50/o Ba, I . 50/o Mg, 0. 30lo Cu, 0.0 5 0/o

Ni, 0.150/o Sr, and 0.03o/oZn. The pseudomorphous bav-enite after beryl (analysis 14) contains a trace amount ofstibiotantalite. The rynersonite (analysis l5), pseudo-morphous after stibiotantalite, contains negligible REE.

Ca and Mg, which form major components in the clays,zeolites, and calcite encountered in pockets that have re-leased their internal vapor phase, are not present in anyof the other pocket minerals, having been depleted byfractional crystallization. Therefore, they must have comefrom some other source than the pockets. The source isnot the pegmatites but principally their host rocks. Abun-dant Ca and Mg are present in the host norite-gabbro ortonalite. The composition of the montmorillonite devel-oped in the altered envelopes ofrock adjacent to the dikesis extremely similar to that of the montmorillonite foundin the pockets. For many pockets, the source of the Si,Al, and Fe in the contained clays and pocket fillings isalso the host rocks. A progressive and continuous miner-alogic and geochemical sequence has been observed in thepegmatite pocket contents (Figs. 2 and 3). The first mineralto be deposited as "snow-on-the-roof is cookeite, indic-ative of availability of Li and Al. If a given pocket con-tinued to interact with circulating hydrothermal solutionsand did not become closed, depletion of Li but continuedavailabilty of Al resulted in crystallization of Li-tosuditeand subsequently beidellite (Table 2). Continued fluid cir-culation and alteration ofthe norite adjacent to the apliteand pegmatite dikes resulted in an increase of Ca and Mgand formation of Ca-Mg montmorillonite, palygorskite,zeolites, and calcite. Still later, sparse amounts of todo-rokite and nontronite were deposited, and final depositionofFe-stained kaolinite occurred after a significant hiatusin time and at low temperatures and pressures. The tex-tural observations are consistent with the observed chem-ical trends.

! ;

I E + :

x + " >

3 . i f "

FOORD ET AL.:"POCKET" CLAYS IN GRANITIC PEGMATITES

I. . . . t 1 t .

l o- l o

i ' t o '

o . . . t . co o - a l @ o- l ' "

o a oo o l

N O O O N 6 0 0 d Oo d N o o N o o o o H N d ! E

4 < b E A 2 ' ! f o E J J J F Z

o c o N o E

P S : o o o o i o E

@ i N O O O S

s 6 o +

s i i o

o 6 s o o o o N

6 i - i i N 6 i

. 9 3 - o o o o 9 9 . : *

6 t 6 t€ r . , . . ! . !

Q i 6 o

$ t s o o i o o o

€ i O O o o N

o 6 0 0

o

s 5 . - i o o o 9 9 :

J a e =

: . i € 5

432

e

ax

>|

6. F

o =t r d i

t r ( t( ! v

k o. d H

u 6o €€ *

_a i!:>o

JI

o

C)

O

c^lo

€F

€o

o

o

C)Jo

ooo

o

(,

uv

oo

o

F

FOORD ET AL.: *POCKET" CLAYS IN GRANITIC PEGMATITES

o z z 6 6 Q o a 2d a o Z Z

z o z o o a o

433

. i . i = J o o o o o No 6 N 9 NO N

o o = v o z o o N oo o o

o i o o E S E S : "

o N 6 O

i o o o N u o o o o o o o i

O F

o n i w o z o n o zo

o i o o o z o N h z

* n4 i

z o o z z z z o J z o o E z o o = o N . i t i z o o g z o 3 oE 3 R 3 3 I 3 E : = = 2

d z / z

> @ E

o q a g

n P E I

o o E e 6

9 . 3 ' 8

l ^ . " E

P E d ^

2 q 6 2 2 0 2 2 0 c

o z o o o o o 2 z z6 0 0 N o or a a z

z 6 0 z z F F z z a o 2 z z o z z o a 6

z F o z 2 o 6 o z z ^ o z z o 6 o o E EO 6 2 N d O t s < O o

5z 6 o z z a t s z z z z a z z o E z o x 6 d z z o z z a o z

Ez 6 o z o z o q z z z a z z o z 2 6 Z F o 2 = ? 1 2 ? 2 2 2

o 6 2 0 0 a o 6 2N Q O Z Z

d a ' = = = o o =

z z z z z a a

i z z o z z a a zo z z

6 z z z z z a a z 2 Z

o w 2 z z 6 o o 6 oN o i r

z

+

o z z 6 z z a 2 2

o z z o z o a a zo o z z

o i o o o z o o 6 z z N 4 z o a q o z z o o z z o z 2 o z FO O O H6 0

O O T Z O Z Z Z 6 O z Z a Z Z Z Z Z Z Z 2 O Z Z 2 o Z Z O QO i o O

o o o o

o o E z o z z z 6 z z z z z z o z ? z z q q z z 2 ^ 2 z o oo 6o a

o o N z t s z o z o z z z o z z z z z z z z o z z z z z z > o6 0

o o 6 v o z o h o o z z z z z o z z z z z a z z 2 z z a E 6

i O a O 6

o o o o o i o o i o z o o 2 z z 6 0 z 2 0 6 2 2 0 2 2 0 E 46 0 N O o O

N N Z O 6 0 0 0 z6 - O o O

o o o

o F z o o o o o u

N O O o

o E z a o o a o zO N O z Z

< ; d d c i o z $ - - o u ! - a = z z 6 z z z r z o N z z q z z

o o o o o z * o t s o 2 0 F 2 2 0 F o o N o o z o o z z o z zo N N o o 6 6 0

E = = = =N N X N A * *

o E 6 i C S ( O i O t s ! 6 O C ! 6 i C l i i 6 O 6 s ! ! 9 !E ! O H = < O 4 4 O O O O J = Z Z d h A 6 6 ? P > > N N O < Z Y 9 9 9 r E a >

434

M l n e r a l s

FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES

L a t e - s t a g e P o c k e t P o s t - r u p t u r epocket mineral rupture growth +/-growth pseudonorphlng

C r y s t a l l i z a t i o n o fh y d r o l h e r @ L c 1 a y 6 , z e o l l t e sand carbonates

e l b a i c e

q u a r I z

f e l d s p a r s

nLcas

p o l l u c i c e

beryl

c a s s l t e r i t e

Nb-Ta oxides

---f

f ine-grainedalbl te-Kf sc o a E l n g

l e p i d o l i t epseudomorph6

after tourual lne

ryners o ni tea n d f e r s m l t e

bavenlte

E t o k e s i t e

Ealayalte

h a o b e r g l t e

herder I t e

cookelte

L1-tosudite

b e l d e 1 1 1 t e

Ca-Mgnontnori l lonl tepalygorskl te

heulandite

e t i l b l t e

lauooni te

c a l c i t e

nontroni le

t o d o r o k l t e

kaoL ini te

rhombohedral Dlaty

E a r l v + T i m g € L a t e

Fig.2. Paragenetic diagram for late-stage pocket minerals in the Himalaya dike system.

Figure 2 is a paragenetic diagram of minerals formedprior to complete crystallization of pegmatite pockets ex-tending to those minerals deposited under hydrothermalhypogene conditions and supergene weathering condi-tions. A schematic summary of the major assemblages ofminerals identified from pegmatite pockets studied, andtheir postulated evolution and depositional sequence isgiven in Figure 3. The paragenesis was determined bydepositional sequences, textures, and structures observed,as well as by chemical studies.

CHnprrsrny oF THE Hosr cABBRo-NoRrrEThe source for many or all of the elements constituting

the pocket clays and the other hydrothermal and super-gene minerals found in the pockets is the gabbro-noritehost. The aplite andlor pegmatite dikes of the Himalayadike system are contained in a gabbro-norite, which is

frequently altered from an extemely hard, blue-gray, hy-persthene-augite-olivine-bearing, porphyritic (plagro-clase) or equigranular norite-gabbro, to a soft, grayish-white to brownish-gray friable material preserving theoriginal igneous texture. Biotite, iron-rich hornblende, andvermiculite are developed from the mafic minerals in thefresh rock, along with abundant, white, Ca-Mg mont-morillonite replacing the plagioclase feldspar. A discon-tinuous, but distinctive hematite-red-colored band 2 cmthick is present within the altered norite above and belowthe dikes. This band or line is referred to as the "red heatline" by the miners. Chemical analyses ofthe norite aboveand below the line were essentially the same (analyses 4and 5, Table 4). The origin of this band is not yet clearbut may represent oxidation or alteration fronts createdduring alteration of the norite. The zones of altered rockare up to 20 cm or more wide. Where several branching

FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES

Ennry Time ---) Lnre

Cookeite coat ing on al lpr imary minerals as tsnow-

o n - t h e - r o o f '

c o o k e i t e + s t l l b i t e + l a u n o n t i r e + / -

calcl te + buff montmori l loni te (none or very

s p a r s e )

cookeite -> Li- tosudite - ;>

(pearly) (waxy-pink or buff)

beidel l i te - ;> Ca-Mg montnori l loni te J( p i n k o r w h i E e , ( p i n k o r w h i t e ,waxy) earthy)

-> palygorskl te( w h i t e , e l a s t i c )

cookelte -> Li- tosudite --> beidel l i te -->

Ca-Mg montmori l loni te +/- heulandite +/-

s t i l b i t e + / - l a u m o n t L t e

I , 2 , 3 , o r 4 , p l u s n o n t r o n i t e + / -( d e e p b r o m )

todoroki te + kaot ini te( b 1 a c k , e a r t h y ) ( r e d - b r o m , s t t c k y )

assmblages, not al l n inerals l isted may be developed, I lerderl te and/or hanbergiEe nayE h m a n d p r e - d a t e d e p o s i t i o D o f c o o k e L t e .

Mineralogical assemblages and evolution of pocket clays in the Himalaya dike system.

435

l 'DRy PocrETs'

Primary pocketn i n e r a l s ( e . 9 .

q u a r t z , t o u m a l i n e ,f e l d s p a r s , n i c a s ,and associatedmlnerals

' CLny - r t l l Ep

Poc rers '

Note- In any of the f lven r a a a n r i h . l 1 ^ f

Fig. 3.

dikes are present, the norite is frequently completely al-tered between the dikes.

Locally, the presence of tourmaline (schorl-dravite se-ries) in the host norite suggests the introduction ofB (Foord,1976). Small dikelets of fine-grained (<0.1 mm) "pres-sure-quench" aplite emanate from the pocket zone andextend either upward or downward toward the marginsof the pegmatite-aplite dikes. These stringers have in-truded the norite and resulted in its tourmalinization, withconcomitant development of hornblende, biotite, ver-miculite, and montmorillonite. An analysis (Table 4, no.3) shows decreased amounts of CaO and MgO and in-creased amounts ofNarO, B2O3, Nl2O, and HrO. The areasof B-metasomatized host rock are all spatially and tem-porally related to formation of the pocket zone of theHimalaya dike system. This metasomatism occurred ear-lier and in an environment of higher temperature andpressure than the hydrothermal alteration involving theformation and deposition of the pocket clays.

The bulk chemistry oftwo samples (Table 4, nos. I and2) ofequigranular and porphyritic varieties ofnearly freshnorite was determined. Six analyses of the highly alterednorite, in the form of three pairs of samples from threedifferent locations in the San Diego mine, were also per-formed (Table 4, nos. 4-9), along with one sample (Table4, no. 3) of B-metasomatized norite. At two of the threelocations, samples were taken above and below the dike.The analyses have been recalculated on a dry-weight basisto account for water in hydrous minerals formed during

alteration (LO\ 5.73-9.15 wo/o). The content of FerO,remains essentially unchanged when comparing fresh andaltered norite. Significantly, however, the content of MgOdecreases from l0 to 7 wto/o. and that of CaO from I 1 to6 wto/o. The contents of SiOr, AlrOr, NarO, I(rO, TiO2,PrO, and MnO increase slightly. All of these observationsprovide strong evidence that the pocket clays were derivedfrom the host norite.

DrscussroNStructural formulas for the ten analyzed clays, which

include Mg-Ca montmorillonite, beidellite, Li-tosudite,and palygorskite, are shown in Table 5.

The analyses of the beidellites and particularly the Mg-Ca montmorillonites and the Mg-poor montmorilloniteare similar to the results given for two pink montmoril-lonites from pegrnatite pockets from the Pala district(Wells, 1937; Ross and Hendricks, 1945) and to mont-morillonites from Greenwood, Maine (Wells, 1937),Branchville, Connecticut (Brush and Dana, 1880), Clare-mont, California (Laudermilk and Woodford, 1934), andEmbudo, New Mexico (Ross and Shannon, 1926). Schal-ler (1905) described pink halloysite from the Stewart mineat Pala. The purity ofthis material and its X-ray propertiesare unknown. The Li-tosudite (Table 2, analyses 3 and 4)is a combination of di-trioctahedral cookeite and diocta-hedral beidellite. Li-tosudite was first described by Ru-sinova et al. (1976) and discused by Eberl (1978b). Syn-thetic Li-tosudite, along with kaolinite and quartz, was

Pocket ruptureand addit ionalg r o v t h o f t h eprimary ninerals+/- f ine-grainedc o a t i n g o f K - f s+ / - a l b i t e , a n dc o r r o s i o n , + / -pseudomorphing

of pr lnary

minerals

I

B

+

I

z

s c d o 6 6 6 oc o { i i 6 i 6

. . t . . . :

6 @ € 6 O O ' Oo

i N i €{ t i o o o ' o

o

N 6 0 6 6 do 6 N o o N ! 6n o o o i o o o o o o

v 9

o o o 6 6 o .N N o o o - i i

- N O O N O O O O O ' O

. o

t o @ + 6 s € ,o d o s N - i {

o N o 6 N i o i o o oo

o € r a N i i :

6 6 i h F i o i o o ' oo

o o o d d o i F , +N J i O

6 N N O O O ' €

o s 9 $ 6 N€ o i N o i ; 6

o o 6 i o i o o o o ' oo

6 N 6 C N i 4 j O i i N

6 6 @ 6 d i O O O O ' <o

aoo

N O O O N 6o d F o o N o o o o H !d i o & s 6 d i N c o o6 < & E O Z J ? e = J F

FOORD ET AL.: "POCKET" CLAYS IN GRAMTIC PEGMATITES

N i F Or r D r 6 @ o o d N <

i r - 4 o o 6

6 O N -

6 r o : o o o

o N 6 N 6 + 6 0 S6 N H i O r 6

o @ s a N o o o o o o

v

6 6 6 0

\ o o o o o 6r ! l

o o N -6 O o H r N 6 6 d r o

o i o r o o N

o o 6 N

o 4 o i o o 6

6 € + s { o : F o l +- o 6 0 N - 6 1 -

6 N N O O O N I €

j i 6 N 6$ o € 6 0 0 H N o < 6

o 6 6 i o r o o o o ov

, . " 8

@ d 6 6 ! r r + O i €

o 6 6 O N i O O O O O - v

O

o

N O - O N 6o N t s o o N c o o o Hr i o b d 6 N . i N r o6 1 b U O Z V t s t u = J

b

! 9 q

;

436

a. F

IO

6

ro

v

E t rw O

Fh't 1)d J aE E

9 X

E ( )9 ? s

t E a* .i5

6 iO X

E Hc d E

oI

oo

tr

t!

X

$

(lF.

FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES

Table 5. Calculated structural formulas for analyzed clays and palygorskite

1 . ( A 1 1 . 6 2 M c 0 . 2 8 F e 9 . 9 7 ) l r . 9 T ( S 1 3 . 9 6 4 r o . o q ) r q . o o 0 r o ( 0 H ) 2 [ N a o . o 9 M n o . o 8 C " o . o 6 K o . o 5 ] r o . 2 O

, 2 ,8H20

2 ' ( A l 1 . 7 3 F e o . o 5 M n o . 0 r M 8 0 . 2 r ) s 2 . 6 6 ( s i 3 . 6 9 A r o . 3 r ) r ! . o o 0 r o ( 0 H ) 2 [ M s g . r o c a o . o 7 7 t o . z T

. 3. 8H^0

3 . ( A 1 1 . g 3 M 8 6 . t 6 F e o . o 1 ) t 2 . 6 9 ( S i 3 . 6 7 A 1 0 . 3 3 ) r , r . o o 0 t o ( 0 H ) 2 [ K o . r 6 C a o , r t M 8 o . o r ] r o . 3 r

. 1 . 5H20

r r . ( a r . ' . g t l l g a . r 3 F e O . o 2 ) r z . o o ( S i 3 . 5 Z A 1 o . q 3 ) r + . O O 0 r o ( 0 H ) z I C " o . r 5 K o . r O M g O . o Z M . O . O r ] r 0 . 3 3

. 2 . \ H 2 0

5 . ( A l z . o r F u o . o z ) r r . o o ( S i 3 . o o A 1 1 . o o ) r q . o 0 0 1 o ( 0 H ) 2 [ M s 6 . 3 6 c " o . . r 3 N r o . o z K o . o 4 ] r o . q 9

,2 .8H2O

6 . ( A 1 1 . g 7 M B 6 . 1 2 F e o . o . t ) r 2 . O O ( S i 3 . 5 4 A 1 0 . 1 6 ) r q . o o 0 r o ( 0 H ) 2 [ M s o . 1 6 C " o . r r N " o . o q K o . o 2 ] r 0 , 3 3

, 3 . o H 2 o

7 . ( A 1 1 . z 6 M c o . 2 q ) r r . o o ( s i 3 . 6 9 A 1 0 . 3 r ) r 4 . o o 0 r o ( 0 H ) 2 . 0 [ M g o . r I c a o . r o K o . o r ] r 0 . 2 5 ' 3 . 3 H 2 0

8 . ( A 1 1 . 5 7 M e 6 . 4 3 ) r 2 . 6 9 ( S i 3 . 9 1 A l o . o 9 ) r q . o o 0 t o ( 0 H ) 2 . 6 [ M s 6 . 1 6 c u o . r ! K o . 9 3 J 5 q . 2 7 ' 3 . 3 u a o

9 . ( A 1 1 . 5 6 M 8 9 . 4 q ) r 2 . o O ( s i 3 . 9 2 A 1 o . o g ) r l . o o 0 r o ( 0 H ) 2 . 0 [ M s O . r r C u O . r 5 K O . o r ] r 0 . 2 6 . 3 . 6 H 2 0

' 1 0 . ( A 1 1 . 1 1 M s 6 . 6 6 F e o . o r r K o . o r l C a o . o 3 ) r z . o 3 ( s i 3 . 8 9 A r o . r r ) r l . o o 0 r o ( 0 H ) ' 2 ' 0 5 H 2 0

Notes - Numben ing is as g iven in Tab le 2 . Smect i te fo rnu las ca lcu la ted on the bas is o f 22( 0 , 0 H ) , p a : - y g o n s k i t e o n 2 1 ( 0 , 0 H ) , L i - t o s u d i t e b y s u b t r a c t i n g 5 0 w t . S o f c o o k e i t e ( i d e a l )

a n d c a l c u l a t i n g t h e f o r m u l a f o r L h e s m e c t i t e c o m p o n e n t o n L h e b a s i s o f 2 2 ( 0 , 0 H ) . T h es t r u c t u r a l f o r m u l a f o r t h e n o n t n o r i l l o n i t e ( n o . 1 ) 1 s a f E e r c o r r e c t i o n f o r 2 0 v L . f iadmixed cooke i te . The s t ruc tu ra l fonmula fo r t ,he be ide l l i te (no . 5 ) i s a f le r cor rec t ionf o r 3 0 w t . $ a d m i x e d l e p i d o l i t e , 1 0 w t . t c o o k e i t e a n d ' 1 0 w t . t r e l b a i t e . T h e s t r u c t u r a lf o r m u l a f o n t h e p a l y g o r s k i t e ( n o . ' 1 0 ) i s a f L e n s u b t n a c t i o n o f N a 2 0 a s w a t e r - s o f u b l e N a C l .

generated from Wyoming bentonite saturated with LiCl wto/o confirmed that much of the Na was removed by theand reacted at 400'C for 30 d, but was not produced after single washing.30 d at 300"C (Eberl, 1978a). The Li-tosudite in the Hi- A chemical analysis of a sample of cookeite, containingmalaya dike system may not necessarily have formed in some microcline, albite, and stibiotantalite, confirmed thethis temperature range because the experiments were not optical and X-ray diffraction identification of cookeite.reversed. However, these temperatures are reasonable on All of the samples of beidellite and Ca-Mg montmoril-thebasisofoxygenisotopicandfluid-inclusiondata(Tay- lonite are dioctahedral and have douo spacings of 1.49-lor et al., 1979) obtained on higher temperature and pres- 1.5 I A. Beidellite cannot be distinguished visually fromsure primary pocket minerals and lower-temperature fluid the Ca-Mg montmorillonite in the pockets and must beinclusions in later secondarv calcite. identified by complete chemical analysis or by the Greene-

X-raydiffractiondataindicatethatthepalygorskite(Ta- Kellytest(Starkeyetal., 1984).Themineralmaybemuchble I , Him- l0) is pure, but the analysis of several samples more widespread in this and other pegmatites than real-indicates 2.42 wto/o NarO. By omitting Na from the anal- ized at present. The one sample of nontronite (Table I ,ysis, a more reasonable structural formula is obtained SDTM 4C) examined was trioctahedral with a do"o of 1.53(Table 5, analysis l0). To determine whether the Na was A. Nontronite and todorokite are both uncommon min-present as a residual salt from an intragranular brine so- erals in the Himalaya dike system and represent deposi-lution in the palygorskite, a 4.07-g sample was disaggre- tionfromlocallyspecializedonvironmentswiththeavail-gated in triply distilled ultrapure water and centrifuged. ability of the necessary Fe and./or Mn. Both occur late inThe supernatant liquid was analyzed semiquantitatively the crystallization history, and only the red-brown ka-for Na and qualitatively for Cl. Cl was determined by olinite postdates them.adding AgNO, solution, which forms a dense precipitate AsnotedinTable l, threesamples(SDTM-2EV; SDTM-of AgCl if Cl ions are present; Na was determined by 2EJ, pink and white) of montmorillonite collapse on heat-atomic absorption spectrophotometry. These determi- ing at 400'C to a doo, spacing of 9.2 A rather than thenations indicate that most of the Na is present in a water- normal 9.8 A spacing. The reason for this smaller spacingsoluble form. The residue was dried and analyzed by XRF has not been determined, but may be due to substitutionspectrometry, and the decrease in NarO ftom2.42to 1.26 of F for OH in the 2il layer resulting in a thinner octa-

43',1

438 FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES

hectahedral sheet. An XRF analysis of a split of SDTM-2EV (Table l) gave (in weight percent) SiO, 58.7, AlrO319.1, FerO. <0.04, MgO 3.76, CaO 1.63, NarO 0.19, KrO4.27,TiO2 <0.02, MnO <0.02, PrO, <0.05, LOI (920"C)l2.7,Total 100.35. After taking into account the admixed20-30 wto/o K-feldspar, the bulk chemistry of this mont-morillonite is similar to that of the two montmorillonitesgiven in Table 2 (analyses 8 and 9).

Two of the pocket clays examined (Table 2, analysis l;Table 3, analysis I l) were a striking purplish red, whereasall others were either white, pink, brown, or red brown.The purplish-red color correlates with a significantcontentof Mn, > l0lo MnO, in both samples. Much of the Mn inpink pegmatite clays from New England is present as Mn3*,based on optical spectroscopy studies (D. Sherman andN. Vergo, 1985, pers. comm.).

Calcite has been found locally in pockets in the Hi-malaya dike system associated with laumontite, as a pock-et cement. Only small amounts of later, buff-coloredmontmorillonite were associated with the calcite. Calcitehas been reported from orogenic and anorogenic graniticpegmatites (Gillson, 1927; Hurlbut, 1958; Foord andMartin, 1979). A specimen (no. 15614) in the DenverMuseum of Natural History, collected by H. Truebe, wasidentified as calcite by R. R. Cobban. The calcite occursin beryl and smoky quartz pocket cavities in the anoro-genic Mount Antero granite of Oligocene age (Switzer,1939). Quartz, feldspar, beryl, and fluorite are associatedwith the calcite. The calcite itself closely resembles thecalcite from the Himalaya dike system, having a platyhabit, clear yellow cores, and opaque tan-buff rims thatare solution-etched. The transition from early, clear, yel-low, rhombohedral calcite to later, opaque, cream-col-ored, platy calcite parallels the observed development ofhabits of calcite from hydrothermal veins and MississippiValley-type deposits (A. V. Heyl, 1984, pers. comm.).Five measurements of fluid-inclusion-filling and/or ho-mogenization temperatures for the platy calcite yieldedpressure-uncorrected temperatures of 205-2 I 5"C. Freez-ing-point-depression temperatures between - l5 and- l8"C were also determined. Salinities, expressed asequivalent weight percent NaCl, range from 8 to 12.

CoNcr,usroNs

We confirm the observations by Schaller (1925) andJahns and Wright (1951) that there are two distinct typesofpocket clays-hypogene and supergene. The source formany of the elements incorporated in the zeolite-clay-carbonate pocket-filling assemblage is the immediatelyadjacent norite. The pocket minerals were deposited un-der moderate- to low-temperature hydrothermal condi-tions at some time after the crystallization of the primarypocket minerals. A sharp discontinuity between the white-pink hydrothermal smectite clays and the red-brown su-pergene kaolinite is apparent in all the pegmatites. Ka-olinite is currently being deposited in some of the fracturesand pockets that have access to meteoric waters. The hy-drothermal smectites were probably deposited in an al-

kaline environment, whereas the supergene kaolinite was

deposited in an acid or less alkaline environment, on the

basis of general knowledge of depositional conditions(Krauskopf, 1967). Many of the pegmatites show a con-

sistent and widespread trend ofchanging depositional en-

vironment, reflected in both chemistry and mineralogy.

Li-tosudite and palygorskite are first reported here from

a pegmatite. This study and others of the complex peg-

matites of San Diego County, California, indicate the ex-

istence of a continuum between magmatic and hydro-

thermal conditions.

Acxxowr.nocMENTS

We wish to acknowledge the enthusiastic support and the fivesamples provided by Elizabeth Herbert from the Ocean Viewmine. George Ashley provided two samples from the Katrinamine, Louis B. Spaulding provided four samples from the LittleThree mine, Roland Reed provided material from the OceanView (Eliz. R) and Maple Lode mines. One sample from theWhite Queen mine was provided by Norman Dawson, and sam-ples from the Tourmaline Queen and Stewart pegmatites wereprovided by Pala Properties International. This study would nothave been possible without the support and assistance ofEugeneB. Rynerson of the San Diego Tourmaline Mining Co. and thatof Linley and Winnifred Hall of Himalaya Gem Mines, Inc.Additional crucial samples from the Himalaya mine were pro-vided by William F. Larson of Pala International. Mr. GarthBricker (Pala International) provided the samples of malayaite.All of the owners and operators of the gem- and specimen-pro-ducing pegmatite mines are gratefully acknowledged for theircontinued support and interest. The inspiration for this and otherstudies of the San Diego County pegmatites has come from Rich-ard H. Jahns and Waldemar T. Schaller. Both had a lifelonginterest in the pegmatites of San Diego and Riverside Counties,and theirwork has established a firm base on which future studiescan be built.

Six-step semiquantitative emission spectrographic analyses andspectrographic rare alkali determinations were made by N. M.Conklin of the U.S. Geological Survey. LirO determinations byICP were made by Julian Gray and George Riddle of the U.S.Geological Survey. XRF major-element analyses were performed

by E. Robb, R. Stern, K. Stewart, and A. Bartel, also of the U.S.Geological Survey.

The manuscript has benefitted from the perceptive reviews ofB. F. Bohor, Dennis Eberl, J. L. Krumhansl, R. C. Ewing, andB. C. Chakoumakos.

RnrnnrNcns

Brush, G.J., and Dana, E.S. (1880) On the mineral locality atBranchville, Connecticut: Fourth Paper. Spodumene and theresults of its alteration. American Journal of Science, Article24, Third Series, 20 (l l8), 257 -284.

Grn!, Petr. (1978) Alteration of pollucite in some pegmatites ofsoutheastern Manitoba. Canadian Mineralogist, I 6' 89-95'

Eberl, Dennis. (1978a) The reaction of montmorillonite to mixed-layer clay: The effect of interlayer alkali and alkaline earthcations. Geochimica et Cosmochimica Acta, 42, l-7.

- (1978b) Reaction series for dioctahedral smectites. Claysand Clay Minerals, 26,327-340.

Foord, E.E. (1976) Mineralogy and petrogenesis oflayered peg-matite-aplite dikes in the Mesa Grande district, San DiegoCounty, California. Ph.D. dissertation, Stanford University,Stanford, California.

FOORD ET AL.: "POCKET" CLAYS IN GRANITIC PEGMATITES 439

- (1977) Famous mineral localities: The Himalaya dikesystem, Mesa Grande district, San Diego County, California.Mineralogical Record, 8, 461-47 4.

Foord, E.E., and Martin, R.F. (1979) Amazonite from the PikesPeak batholith. Mineralogical Record, 10, 37 3-384.

Foord, E.E., and Mrose, M.E. (1978) Rynersonite, Ca(Ta,Nb),O6,a new mineral from San Diego County, California. AmericanMineralogist, 63, 7 09-7 14.

Gillson, J.L. (1927) The granite of Conway, New Hampshire,and its druse minerals. American Mineralogist, 12,307-319.

Hanley, J.B. (1951) Economic geology of the Rincon pegmatites,San Diego County, California. California Division of MinesSpecial Report 7-B.

Herbert, E.L. (1982) Clay mineralogy of granitic pegmatites inthe Pala district, San Diego County, California. M.S. thesis,San Diego State University, San Diego, California.

Horsky, S.J., and Martin, R.F. (1977) The anomalous ion-ex-change behavior of "ordered" orthoclase. American Mineral-ogtst,62, I l9l- l 199.

Hurlbut, C.S. ( I 9 5 8) Additional data on bikitaite. American Min-eralogist, 43, 7 68-7 7 0.

Jahns, R.H. (1954) Pegmatites of southem California. In R.H.Jahns, Ed. Geology of southern California. California Divisionof Mines Bulletin 170. 37-50.

- (1979) Gem-bearing pegmatites in San Diego County,Califomia: The Stewart mine, Pala district, and the Himalayamine, Mesa Grande district. In P.L. Abbott and V.R. Todd,Eds. Mesozoic crystalline rocks: Peninsular Ranges batholithand pegmatites, Point Sal ophiolite, 1-38. Department of Geo-logical Sciences, San Diego State University, San Diego, Cal-ifornia.

Jahns, R.H., and Burnham, C.W. (1969) Experimental studiesof pegmatite genesis: I. A model for the derivation and crys-tallization of granitic pegmatites. Economic Geology, 64, 843-864.

Jahns, R.H., and Wright, L.A. (1951) Gem- and lithium-bearingpegmatites of the Pala district, San Diego County, California.California Division of Mines Special Report 7-A.

Krauskopf, K.B. (1967) Introduction to geochemistry. McGraw-Hill. New York.

Krummenacher, Daniel, Gastil, R.G., Bushee, J., and Dupont,J. (1975) K-Ar apparent ages, Peninsular Ranges batholith,southern California and Baja California. Geological Society ofAmerica Bulletin, 86, 760-768.

Kunz, G.F. (1905) Gems, jewelers' materials, and ornamentalstones of California. California Mining Bureau Bulletin 37.

Laudermilk, J.D., and Woodford, A.O. (1934) Secondary mont-morillonite in a California pegmatite. American Mineralogist,19.26V267.

London, David. (1986) Formation of tourmaline-rich gem pock-ets in miarolitic pegmatites. American Mineralogist, 7 |, 396-405.

Prince, Edward, Donnay, Gabrielle, and Martin, R.F. (1973)Neutron diffraction refinement ofan ordered orthoclase struc-ture. American Mineralogist, 58, 500-507.

Rogers, A.F. ( 1909) Minerats from the pegmatite veins ofRincon,San Diego Co., California. The Quarterly (Stanford Univer-sity),208-218.

Ross. C.S., and Hendricks, S.B. (1945) Minerals of the mont-morillonite group, their origin and relation to soils and clays.U.S. Geological Survey Professional Paper 205-B.

Ross, C.S., and Shannon, E.V . (1926\ Minerals of bentonite andrelated clays and their physical properties. American CeramicSociety Joumal, 9, 77 -96.

Rusinova, O.V., Drits, V.A., and Gorshkov, A.I. (1976) Struc-tural-mineralogical characteristics of mixed-layered cookeite-montmorillonite. Izvestia Akademia Nauk SSSR, Seriya Geo-logicheskaya, 10, 95-104 (in Russian).

Rynerson, F.J. (1967) Exploring and mining for gems and goldin the west. Naturegraph Publishing, Healdsburg, California'

Schaller, W.T. (1905) Mineralogical notes. l. Halloysite. U.S.Geological Survey Bulletin 262, l2L.

- (1925) The genesis of lithium pegmatites. American Jour-nal of Science, 5th series, 10,269J79.

Schwartz, G.M. (1937) Alteration of spodumene to kaolinite inthe Etta mine. American Journal of Science, 33, 303-307.

Simpson, D.R. (1965) Geology of the central part of the Ramonapegmatite district, San Diego County, California. CaliforniaDivision of Mines Special Report 86.

Starkey, H.C., Blackmon, P.D., and Haufl P.L. (1984) The rou-tine rnineralogical analyses of clay-bearing samples. U'S. Geo-logical Survey Bulletin 1563.

Stern, L.A., Brown, G.E., Jr., Bird, D.K., Jahns, R.H., Foord,E.E., Shigley, J.E., and Spaulding, L.B', Jr. (1986) Mineralogyand geochemical evolution ofthe Little Three pegnatite-aplitelayered intrusive; Ramona, California. American Mineralogist,7 t. 406-427 .

Switzer, George. (1939) Granite pegmatites of the Mt. Anteroregion, Colorado. American Mineralogist, 24, 7 9 l-809.

-laggart, J.8., and Wahlberg, J.S. (1980a) New mold design forcasting fused samples. Advances in X-ray Analysis, 23,257-2 6 t .

- (1980b) A new in-mufle automatic fluxer design for cast-ing glass discs for X-ray fluorescence analysis. Federation ofAnalytical Chemists and Spectroscopy Societies meeting,Philadelphia, Pennsylvania, September, I 980.

'taggart, J.8., Lichte, F.E., and Wahlberg, J.S' (1981) Methods

ofanalysis ofsamples using X-ray fluorescence and induction-coupled plasma spectroscopy. In U.S. Geological SurveyProfessional Paper 1250, 683-687.

Taylor, B.E., Foord, E.E., and Friedrichsen, Hans. (1979) Stableisotope and fluid-inclusion studies ofgem-bearing granitic peg-matite-aplite dikes, San Diego County, California. Contribu-tions to Mineralogy and Petrology, 68, 187-205.

Wells, R.C. (1937) Analyses of rocks and minerals. U.S. Geo-logical Survey Bulletin 878, 108.

MlNuscnrpr RECEIVED Mencn 5, 1985MnNuscnrpr AccEprED Srprevngn 21, 1985