<<

American Mineralogist, Volume 64, pages 742-747, 1979

Witherite composition,physical properties,and genesis

ARTHUR BaTDRSARI' AND J. ALnxaNonn SpBen Orogenic Studies Laboratory, Department of Geological Sciences Virginia Polytechnic Institute and State University B lacksburg, Virginia 2406 I

Abstract

Microprobe analysesof 17 natural specimensfrom various localities show sub- stitution of (up to I I percent),lesser amounts of ( < I mole percent) and no detectiblelead. Most witheritescontain lessthan 4 mole percentSr and 0.5 mole per- cent Ca. Lattice parametersand density vary regularly with compositionin the entire range. The equation,equivalent mole percentSrCO, (+l.l): -1102.58(drro,A) + 2515.74,can be used to determinethe approximatecomposition of witherite- solid solutions for X(SrCO3)< I L A plot of mean ionic radius ys.cell volume for natural witheriteslies above the plane connectingthe pure end-membersBaCOr-SrCOr-CaCOr, suggesting a small posi- tive excessvolume of mixing. Calcium substitution 1slimited to minor amounts becauseof the miscibility gap between witherite and orthorhombic CaBa(COr)r,. The limited Sr substitutionand negligible Pb substitution,however, are believed to depend upon the composition of the pre-existing sulfate (barite) from which witherite forms and the disequilibrium behavior of low-temper- ature solutions(<200'C) that crystallizeorthorhombic carbonates.

Introduction mated ARL SEMQmicroprobe at 15 kV and l0 This work is part of a systematic study of the nanoamps,employing Bence-Albeemethods of data chemistry and physical properties of the ortho- reduction. Standardsincluded synthetic BaCOr(Ba) rhombic .The probable conditions and and SrCOr(Sr)and natural (Ca) and mechanismsof witherite genesisare examinedin or- (Pb). The content was obtained by calcu- der to explain the limited chemical composition of lating the number of carbon accordingto the : natural witherites. relation C Sr + Ca + Ba + Pb. No elementsother than Ba, Sr, and Ca could be detectedwith a Kevex Experimental procedures X-ray energy-dispersiveunit. With standardsand un- The 17 witherite samplesexamined in this study knowns of similar compositions,Ba analysesshould are listed in Table l.'? In addition to the literature be accurateto about 2 percentof the amount present. summarizedby Palacheet al. (1951) and Deer et al. The relative errors for Sr and Ca arc higher because (1962), other data used in the following discussions of counting statisticsand the problem of estirnating of witherite compositions are from Frank-Kame- backgroundcounts for minor elements. netskii(1948), Gvakhariya (1953), De Villiers (1971), Unit-cell parameterswere determinedfrom smear and Sidorenko(1947). mounts of powderedwitherite using a Picker powder The witherite sampleswere analyzed on an auto- di-ffractometer.Scans at the rate of 0.5olmin were made with monochromatized CuKa radiation using BaF, (a : 6.19784) as an internal standard. All rPresent address:Department of Geology, University of Geor- peakswere measuredat one-half peak height and in- gia, Athens, Georgia 30602. dexed accordingto the data ofSwanson et al. (1954). 2To obtain a copy Table of l, order Document AM-79-100 from program the BusinessOffice, Mineralogical Society of America, 2000 Flor- The least-squares of Appleman and Evans ida Ave., NW, Washington,DC 20009.Please remit $1.00in ad- (1973)was used to refine the cell parameters. vance for the microfiche. Densitiesof severalsamples were determinedwith 0m.3-.wx/ 79 /0708-0742$02.00 742 BALDASARI AND SPEER: WITHERITE 743 a 2 nI pycnometer,using toluene (correctedfor tem- Equ ivo I ent Mol o/o Sr perature) as the displaoementmedium. Several mea- to 5 surementswere made for eachsample and were aver- aged. Sample (2), Table l, was synthesizedfrom J. T. Bakerreagent grade BaCO,Lot418067 at 550oCand I kbar. This sample served as a standard for micro- a probe analysis. E anr o Results Resultsof the study are given in Table l. Because o ""' of the compositional similarity of the majority of samples,Table I has been abridgedin Table 2 to tn- clude representativecompositions. The microprobe analysesshow substitutionsof up to I I mole percent SrCOr, lessthan I mole percentCaCOr, and no mea- t45 t46 t47 surable substitution of . Sample 19 is the most Meon lonic Rodius (A) strontium-rich of the witheritesstudied or reportedin Fig. l. Plot of cell volume and mean for witherite the literature. Higher calcium contentshave been re- samplesin Table l. Becausethe calcium contentsof the witherites ported (Sidorenko, 1947)but were attributed to the are small, the meanionic radius is alsogiven in termsof the equiv- presenceof calcite inclusions.The analysesof with- alent mole percentSrCO3 substitution in witherite. The ionic radii fornine-coordinatedBa2+ (1.4?A), SP* (l.3lA), and Ca2*(l.l8A) are from Shannon(1976). The unit-cell volumesused to draw the ideal mixing lines are BaCO3, 3M.24A3 (Swansonand Fuyat, Table 2. and chemical data for witherites 1953),CaCO3, 226.85A3(Swanson et al., 1954),and SrCO3, 25935 A3 (Speerand Hensley-Dunn,I 976). t8

Pb0 0.00 0,00 0.00 0.00 erites in this study show a mean of 3.3 mole percent ca0 0.08 0.11 o,24 0.02 sr0 0.68 ]-72 3.26 6.06 Sr and 0.3 mole percentCa, comparableto the natu- Ba0 17.r5 76.58 75.00 72.LO ral witherite compositions in the literature. co2* 22.50 22.78 23.L0 23.28 Total 100.41 101.19 r01.60 101.46 The unit-cell parameters exhibit a small, system-

Nunber of lons on the basis of 1 COe atic variation betweenthe extremewitherite compo- sitions (Table 2). Unit-cell volume varies nearly lin- Pb 0.0000 0.0000 0.0000 0.0000 Ca 0.0028 0.0037 0.0081 0.0008 early with mean ionic radius over the compositional Sr o.or29 0.0320 0.0599 0.1104 Ba 0.9844 0.9644 o.9320 0.8888 range (Fig. l). If there were ideal mixing in the Ba- Sr and Ba-Ca solid solutions,the samplesshould lie l,l.r.R, *** r.4672 t.4639 1.4580 L.452L on the lines joining end-member orthorhombic car- bonatesin plots of cell volume vs.ionic radius. How- v, 63 303.93(6) 302.87(9) 301.32(10)300.38(6) ever, the witherite sampleslie abovethe ideal mixing arA s.3L2(.7) 5.305(1) 5.297(2) s.293(1) b,L 8.999(1) 8.890(2) 8.877(3) 8.861(2) lines (Fig. l), suggestinga positive excessvolume of cr L 6.428(r) 6.42t(2) 5.408(2) 6404(1) mixing in this compositional range. The excessvol- ume may result from the large atoms pre- d22o calc, 2.2807 2.2779 2.2745 2.2722 d22o ^"^" . 2.2807 2.2777 2.2750 2.2?2L venting the structure from contracting around the smaller Sr and Ca when they are present in ocalc. 4.29 4.28 4.27 4.24 omeas. 4.31 4.28 4.26 4.22 small amounts. The approximateSr contentof witherite can be de- termined indirectly, u5ing Figure l, becauseCa is a *CaLculated to gioe C = Ba + ST + Ca + Pb minor constituent. A simpler method is to use the **The estimted standttd errors a?e giuen in puentheses and. xefer to the Laet decinaL place(s). single 4zo spacingof witherite. The equation derived ***Man i.onic radius 4 Anglesarke, Inrcashi"e, Eng1'and I|MNH #103262 by linear regressionis 1 3 s e tlting s tone MLne, Nor tlmb etLand, 18 Arkhys, Kaukoeus, USSR equiv.mole percent Sr CO3(+l.l) Colmbin NMC#34999 19 Cassiar DtstrLet" British : -l102.58 (d,'o,A)+ 2515.74 BALDASARI AND SPEER: WITHERITE

The 220 reflection, rather than the 132 peak used for Equivolenl Mol 7o Sr other orthorhombic carbonates,was chosenbecause ro5'o of its intensity, sharpnessand the lack of interfering peaks. Table 2 contains measured and calculated den- sities, which compare favorably. Density was found to vary linearly with the mean atomic mass of the cationsin the sample(Fig. 2a), fitting a linear regres- sion equation

meancation atomic mass: 86.48(p) - 235.01 427 which can be recast as Equiv Mol o/. SrCOJ i | = : -173.9 - r73 9(P) + 748 9 equiv.mole percentSrCO3(il) (p) + 748.9 E o 425 This equation differs from the regressionline for the CD calculateddensity of syntheticorthorhombic carbon- o 4?4 atesof the group IIa cations: a Ca, Sr, Ba, and Ra (Fig. a_ 2b). Although over the range of the end-member group IIa carbonatesthese cations show a linear rela- tionship, in detail they exhibit a larger deviation than expected for the errors in cell volume and atomic mass determination. Other metal cations that form orthorhombic carbonates,Sm, Eu, and Pb, lie well aboveboth lines. Thesepossess suitable sizes and va- lencesto form orthorhombic carbonates,but they are heavieratoms than the group IIa cations.

Discussion 60 too t40 t80 220

Natural witherite examined in this study and re- Cofion Atomic Moss (omu) ported in the literature showsonly a limited composi- Fig. 2. (a) Plot of calculateddensities ys. mean atomic massfor tional range; only Sr substitutesfor Ba in significant cations of witherite samples in Table l. Becausethe calcium con- amounts. tents of the witherites are small, the mcan atomic mass is also Barium has been rarely reportedin analysesof ce- given as equivalent mole percent SrCO, substitution in witherite. russite and above trace amounts (Palache (b) Plot of the calculated densities and atomic mass for cations of et al., 1951,p. 194-196;Deeret al., 1962,p.319-322). the orthorhombic carbonates. Unit-cell volumes used to calculate the densitiesin addition to those used in Figure I are alstonite Speer and Hensley-Dunn(1976) and Speer(1976) (als),267.3A3(Sartori, 1975);EuCQ, 259.10A3(Mayer et al., found that Ba is present in only minor amounts 1964);SmCO:, 260.72A3(Asprey et al., 1964\;RaCO3, 335.3A3 (<3600 ppm) in strontianite.These results combined (Wiegel, 1977,p. 370-371),and PbCOr, 269.61A1(Swanson and with the witherite analysesfrom Table I and from Fuyat, 1953). The calculated densities of witherite samples in the literature show that the orthorhombic carbonates Table I fall along the line from Fig. 2a. exhibit strong bimodal distributions on the CaCOr- BaCOr, SrCOr-BaCOr,and PbCOr-BaCO,joins. witherite and alstonite, CaBa(COr)r,fot awide range The rarity of intermediatecompositions in the ortho- of temperatures,pressures, and fluid corrpositions. rhombic carbonatesmust be either a result of a crys- This miscibility gap accounts for the limited sub- tal-chemicalfactor, such as a miscibility gap, or a re- stitution of Ca for Ba in witherite. Complete or flection of the fluid composition,or the mechanisms nearly complete solid solutions have been found in by which witherite crystallizs5. the SrCO.-BaCO, system by Cork and Gerhard Experimental work on the CaCO.-BaCO, join by (1931),Terada (1953),Chang (1965, l97l) and Terada (1953),Chang (1965, l97l), Bostrom e/ a/. Chang and Brice (1972) and in the systemPbCO3- (1969), and Chang and Brice (1972) has demon- BaCO, by Bostrdmet al. (1969)and Chang and Brice strated the presenceof a miscibitty gap between (1972) over a similar range of conditions. The ionic '145 BALDASARI AND SPEER: TTITHERITE

fate compositions.Hanor (1968) termed this "inert 35 behavior." More recently Thorstensonand Plummer 30 (1977) have explained this o behavior of two-com- o 25 ponent solid phases as "stoichiometric saturation" o (L and have interpreted the causeas "low-temperature ! kinetic problems." o N Orthorhombic carbonatesare known to form at E problems o low temperatureswhere thesekinetic may z occur. In lead--barite- deposits where witherite has formed from barite, the filling temper- t5 aturesof fluid inclusionsin barite have been reported as 150"-70oC(Brown, 1967),130"-50oC (Sawkins, b ro to ! 1966),and l50o-ll5'C (Hayaseet al., 1977).Pre- ;5 q z sumably the witherite formed at temperaturesat the of the range,after formation of barite. o ro 20 40 60 80 90 too low end Bo Sr The mechanism of witherite formation may also be Corbonote an important control of its chemistry. Witherite is Fig. 3. Frequencydistribution ofobservedbarite- com- commonly thought to occur when barite is altered by positionsand witherite-strontianitecompositions. The sulfatedata waters (Weller 1952; are from Hanor ( 1968)who normalized, the 2293 barite and 77 ce- the action of carbonated et al., lestineanalyses so that eachset ofdata represents50 percentof the Holland, 1967;Helz and Holland, 1965).Based on total. The 20 strontianite analyses are from Speer and Hensley- the associationand texture of the many witherite oc- Dunn (1976)and Speer(1976). The 24 witherite analysesare from currencesin Great Britain, Hancox (1934)postulated this study, as well as Frank-Kamenetskii (1948), Gvakhariya that in all instances the witherite was produced in (1953), Deer et al. (1962), De Villiers (1971), and Sidorenko (1947\. this way. Strontianite is similarly believed to form from a pre-existingsulfate, celestine,while cerussite radii of Pb'z*(1.354) and Sr'* (l.3lA) for nine- has been noted to form from .Figure 3 is a coordinated ions (Shannon, 1976) are nearly the frequency distribution of naturally-occurring Ba-Sr same, and these ions could be expectedto behave sulfates and carbonates.The and sulfate similarly with regard to solid solutions with Ba2* compositionshave a similar distribution, suggesting (1.47A). In contrast,Ca'* (l.l8A) is much smaller that the geochemicalseparation of Ba and Sr that oc- than Pb'z*or Sf*, which may account for the mis- curred in the sulfate systemat low temperatureswas cibility gap along the Ba-Ca join. The differencein preservedwhen the carbonatesformed from the sul- ionic radii between Ba and Pb or Sr is nearly the fates. This is possible since the sulfate-to-carbonate same as between Ca and Sr as well as between Ca reaction for each cation takes place under differenl and Pb, where miscibitty gaps have been found conditionsof a(CO')/a(SO.) in the mineralizing fluid (Holland et al., 1963;Chang, 1965,l97l; Bostrdm et (Barton, 1957; Gundlach, 1959). The low reaction al., 1969;Chang and Brice, 1972).Nevertheless, the temperaturesof carbonateformation would probably experimental work indicates the possibitty of com- prevent any carbonate-sulfateor carbonate-carbon- plete solid solutions on the BaCOr-SrCO, and ate solid equilibration. So the composition of with- BaCOr-PbCO, joins and suggeststhat the causesof erite may primarily reflect the geochemical separa- the limited witherite compositionslie in its genesis. tion ofcations in the sulfatesystem preserved by lack The bimodal compositionof the orthorhombic car- of reequilibration in the carbonatesas a result of "in- bonateshas its analoguein the orthorhombic sulfate ert behavior." It could be concluded that witherite ,especially the (Ba,Sr)SOosystem, which has formed from pre-existing barite should have a Sr receivedthe most attention and is reviewedby Hanor content similar to the barite. The histogramin Figure (1968).Disequilibrium betweenthe aqueoussolution 3 suggeststhat this is generally true. In two speci-fic and the (Ba,Sr)SOocrystals, which behaveas an un- instances,Frank-Kamenetskii (1948) and Sidorenko reactive precipitate, permits fractional precipitation (1947) found this to hold for deposits which they betweenthe lesssoluble BaSO. and the more soluble studied. SrSOo.This results in a geochemicalseparation of Thermodynamic calculations for equilibrium con- the two elements and a paucity of intermediate sul- ditions would predict that during the alteration of 746 BALDASARI AND SPEER: WITHERITE barite to witherite, these two coexisting phases should have differing Sr contents.The equilibrium constantfor the exchangereaction o BaSOo* SrCO,? SrSOo+ BaCO3 (l) o o E indicatesa strong partitioning of Sr into the carbon- O ate phaseduring alteration. Figure 4 summarizesthe o calculations showing the compositionsof coexisting sulfatesand carbonatesfor three temperatures.These o calculations were made for P : I atm, assuming ideal solid solution in the carbonate and sulfate F l9 Cossior phasesand using the free energiesof formation from Robie and Waldbaum (1968). The results suggest o 02 04 06 08 rO that as witherite forms from barite, witherite prefer- Bo Sr Sulfote Composilion entially takesup Sr, and the two mineralsbecome in- creasingly Ba-rich until the barite is consumedand Fig. 4. Calculateddistributions of Sr and Ba betweenwitherite- strontianite and barite-celestine solid solutions at diferent the carbonatephase reaches the initial bulk composi- tem- peratures.The plotted point representsthe compositionsof co- tion of the sulfate. Under disequilibrium conditions, pxisting witherite and barite from the Cassiar District. British the distribution of Ba and Sr betweenthe coexisting Columbia (samplel9). pairs during alteration of barite to witherite could have several possibilities.In order to predict composition.Rather the barite must have been selec- the compositions of the coexisting mineral phases, tively depleted in Sr. This barite-to-witherite altera- disequilibrium would necessitate considering sepa- tion pair would then follow the expectedequilibrium rately the controls and deviation from equilibrium of behavior. This barite-witherite pair is plotted in Fig- the dissolutionmechanism of the sulfate and precipi- ure 4. tation mechanism of the carbonate.This would be insteadof combining them as in the equilibrium case Summary by exchangereaction l. Figure 4 showsthat the equilibrium partitioning of The limited compositionof witherite resultsfrom a Ba and Sr between carbonate and sulfate phases miscibility gap along the Ca join with alstonite and makescoexisting barite-witherite pairs potential geo- low-temperature kinetic problems preventing the thermometers.But since natural witherite forms at reequilibration of witherite, aqueousfluid and ear- low temperatures where disequilibrium is a good lier-precipitatedPb and Sr carbonates,cerussite, and possibility, equilibrium betweenthe two solidswould strontianite. The orthorhombic carbonate composi- need to be demonstrated.In addition, since natural tions may be largely inherited from precursor sulfate witherite and barite have nearly end-membercom- minerals and reflect sinilar disequilibrium processes positions,the pairs will plot on the narrow lirnb of a in the sulfate system.These disequilibrium geochem- diagram such as Figure 4. Compositionaldeterrrina- ical separationsare governedby the diflering solubi- tions and calibration of the geothermometerwould lities of Ba, Sr, Ca, and Pb carbonatesand sulfates have to be exceptionallyaccurate in order to reduce from which the carbonates form. As sulfates react the error in temperatureestimates to acceptablelev- with carbonated waters, each nearly end-member els. sulfate is converted to a carbonate under different A barite-witherite pair from the CassiarDistrict, conditions. The equilibrium constant of the barite- B.C., from which sample (19) was taken, was exam- witherite exchangereaction predicts that under equi- ined in this study. The locality and geologyof the oc- librium conditions the Sr preferentially partitions currence is given by Woodcock and Smitheringale into the carbonate phase during the conversion of (1957). The barite is almost entirely replaced by sulfate to carbonate. This is confirmed in one in- witherite and containsless than 0.02mole percent Sr, stance and suggeststhat the most Sr-rich witherite whereas the coexisting witherite contains I I mole should be found as the early alteration products of percent Sr. Ifthis representsa systemclosed with re- Sr-bearingbarites. However, because witherite forms spectto Ba and Sr, the barite cannot be at its original at such low temperatures,disequilibrium is likely, BALDASARI AND SPEER: WITHERITE 747 and other barite-to-witheritealteration pairs may ex- Hayase,K., G. R. Mas and A. L. Bengochea(1977) Synthesis of hibit different compositionalbehavior. barium- solid solution and somc consid- erations on the genesisof barite-celestite deposits in Neuquen Acknowledgments Province,Argentina. f. Japan.Assoc. Min. Pet. Econ Geol.,72, 93-t02. We appreciate the loan of specimensby John S. White, Smith- Helz, G. R. and H. D. Holland (1965)The and geologic sonian Institution, and Louis Moyd, CanadianNational Museum, occurenc€ of strontianite. Geochim.Cosmochim. Acta,29, 1303- as well as the use of the facilities of the OrogenicStudies Labora- l315. tory, Virginia PolytechnicInstitute and StateUniversity. In addi- Holland, H. D. (1967)Gangue minerals in hydrothermaldeposits. tion, we would like to thank S. W. Becker, P. H. Ribbe, D. A. In H. L. Barnes, Ed., Geochemistryof Hydrothermal Depos- Hewitt, and J. S. Hanor for critically readingth€ manuscript. its, p. 382432. Holt, Rinehart, and Winston, New York. References M. Brocsik, J. Munoz and U. M. Oxburgh (1963) The coprecipitation of Sf+ with aragonite and Ca2+ with - ite between 90" and l00"C. Geochim. Cosmochim. Acta, 27, Appleman, D. E. and H. T. Evans,Jr. (1973) Job 9214:Indexing 957-977. and least-squaresrefinement of powder diffraction data. Natl. Mayer, I., E. Levy and A. Glasner (1964)Thc crystal structureof Tech. Inf. Sem., U. S. Dept. Commerce, Springfield, Virginia, EuSOaand EuCO3.Acta Crystallogr.,17, l07l-1072. DocumentPB-216 188. Palache,C., H. Berman and C. Frondel (1951)Dana's Systemof Asprey, L. B., F. H. Ellinger and E. Staritzky (196a)Compounds Mineralogy,Vol. 2, 7th edition. Wiley, New York. of divalent lanthanides-preparation, optical properties, and Robie, R. A. and D. R. Waldbaum (1968)Thermodynamic prop- .In K. S. Vorres, Ed., Rare Earth ResearchII, erties of minerals and related substancesat 298.15'K (25oC) p. 10'-20.Macmillan, New York. and one atmosphere(1.013 bars) pressute and high temper- Barton, P. B. (1957)Some limitations on the possiblecomposition atures.U. S. Geol. Sum.Bull. 1259. of the ore-forming fl uid. Econ. Geol., 52, 333-3 53. Sartori, F. (1975)New data on alstonite.Lithos, 8, 199-207. Bostrdm, K., J. Hanor, J. Blankenburg and R. Glaccum (1969) Sawkins,F. J. (1966)Ore genesisin the North Pennineorefield, in Somesubsolidus phase relations in the systemBaCO3-CaCO3- the light of fluid inclusion studies.Econ. Geol., 6 I, 385-40l. PbCO3at 500"C.Ark. Mineral. Geol.,5,47-53. Shannon,R. D. (1976)Revised effective ioiric radii and systematic Brown, J. S. (Ed.) (1967)Genesis of stratiform lead-zinc-barite- studips of interatomic distances in halides and chalcog€nides. fluorite deposits.Econ. Geol.Monogr. 3. A cta Crystallogr., A 32, 75 l-7 67. Chang, L. L. Y. (1965) Subsolidusphase relations in the systems Sidorenko,A. V. (1947) The genesisof witherites of the western BaCO3-SrCO3,SrCO3-CaCO3, and BaCOr-CaCO3.J. Geol., Kopet-Dag. C. R. (Dokl.) Acad. Sci. I/RSS, 5J, 149-151. 73,348-368. Speer,J. A. (1976) Pennsylvaniarninerals: II. Strontianite. Mtn- - (1971) Subsolidusphase relations in the aragonite-type eral. Record, 7, 69-7 l. carbonates:I.The systemCaCO3-SrCO3-BaCO3. Am. Mineral., - and M. L. Hensley-Dunn(1976) Strontianite composition 56, t66{F-1673. and physicalproperties. Am. Mineral.,61, 1001-1004. - and W. R. Brice (1972) Subsolidusphase relations in Swanson,H. E. and R. K. Fuyat (1953)Standard X-ray ditrraction aragonite-typecarbonates:Il. The systemsCaCO3-SrCO3- powderpattems.Natl. Bur. Stand Circ. 539,VoL II. PbCO3and CaCO3-BaCO3-PbCO3.Am. Mineral.,S7, 155-168. Cork, J. M. and S. L. Gerhard (1931)Crystal structureof the se- tion powder patterns. .iv4t . Bur. Stand. Circ. 539, Vol. III. ries of barium and strontium carbonates.Am. Mineral., 16,71- Terada,J. (1953)Crystal structurcof the Ba, Sr and Ca triple car- 77. bonate.J. Phys.Soc. Japan,8,158-164. Deer, W. A., R. A. Howie and J. Zussman(1962) Rock-Forming Thorstenson,D. C. and L. N. Plunmer (1977) Equilibrium cri- Minerals, VoL 5, Non-silicates.Wiley, New York. teria for two-component solids reacting witl fixed composition De Villiers, J. P. R. (1971) Crystal structure of aragonite,stron- in an aqueous phase----example:the magnesium calcrtes.Am. J. tianite, and witheite. Am. MineraL,56,758-767. Sci.,277, 1203-1223. Frank-Kamenetskii,V. A. (1948) Witherite from barite veins in Weller, J. M., R. M. Grogan and F. R. Tippie (1952)Geology of the northern Caucasus(Arkhyz and Dzhalankol). Chem Zentr., the fluorite depositsof Illinois. Illinois StateGeol. Sum.Bull.76. I, 144.(extracted from Chem.Abstracts, 44,7 l9l, 1950). Wiegel, F. (1977) Radium. In K. Swars,Ed., GmelinsHandbuch Gundlach, H. (1959) UntersuchuDgenzur Geochemiedes Stron- der AnorganischenChemie, Erganzungshand 2. Springer-Verlag, tians auf hydrothermalen Lagerstatten.Geol. Jahrb., 76, 637- New York. 712. Woodcock, J. R. and W. V. Smitheringale(1957) Fluorite-with- Gvakhariya, G. V. (1953)A barite-witherite association.Soobsh- erite deposit at Low€r Liard Crossing, British Columbia, The cheniyaAkad. Nauk Gruzinskor,SSS& 14, no. 5,267-272. (ex- Geology of Canadian Industrial Mineral Deposits. Common- tracted from Chem.Abslracts, 49, 2956, 1955). wealth Mining and Metallurgical Congress,6th, Yancouver,24- Hancox, E. G. (1934)Witherite and .Mining Mag.,5l,'76- 247. 79. Hanor, J. S. (1968)Frequency distribution ofcompositions in the Manuscript received, September 27, 1978. barite-celestiteseries.'lnr. "L Sci..J3. 1215-1222. acceptedfor publication, October 27, 1978.