American Mineralogist, Volume 72, pages 319-328, 1987

Petrologic and crystal-chemical implications of cationorder-disorder in kutnahorite ICaMn(COr)rl*

D. R. Pucon, E. J. EssBNnDepartment of Geological Sciences, The University of Michigan, Ann Arbor, Michigan 48109, U.S.A.

A. M. G.qlNnsDivision of Earth Sciences, National Science Foundation, Washington, D.C.20550, U.S.A.

Ansrucr

The crystal structures of kutnahorite from Bald Knob (BK), North Carolina, and SterlingHill (SH), New Jersey, have been refined. The BK kutnahorite (CaorMn'oMg'Feo) is dis-ordered in Ca-Mn distribution, but the SH kutnahorite is substantially ordered:(Ca".roMno 1u)(Cao rrMno rrXCOrL. Long-range cation order in kutnahorite is not detectableusing conventional powder X-ray diffraction techniques, but it may be measured by single-crystal techniques. The large ionic radius of Mn relative to Mg in dolomite leads to coupleddistortion of the Ca and Mn octahedra that may result in a low ordering potential. Thus,the BK kutnahorite lacks significant cation order despite slow cooling from amphibolite-facies regional metamorphic conditions. Long-range cation order in SH kutnahorite iscompatible with a low-temperature solvus between calcite and kutnahorite as well asbetween kutnahorite and rhodochrosite. Two-phase intergrowths of manganoan calcite(CarrMn,rMg) and calcian kutnahorite (Ca.rMnrrMgr) from SH are interpreted as due toprimary coprecipitation of calcite and ordered kutnahorite from solution in the two-phaseregion at temperatures below the solvus crest. Data on metamorphic Ca-Mn carbonatesindicate complete solid solution between calcite and rhodochrosite at 600qC with solvibetween kutnahorite-calcite and kutnahorite-rhodochrosite forming at lower temperatures.

INrnooucrrou

Kutnahorite was originally described by Frondel andBauer (1955) as having ideal composition CaMn(COr),and ordering of Ca and Mn, resulting in a dolomitelikestructure. The ordering was determined by the apparentpresence of hll, l: 2N + I reflections (based on hexag-onal axes) in the powder-diffraction pattern that are com-patible with space group (R3) of the ordered compoundbut incompatible with space group (R3c) of the disor-dered calcite-type structure. Schindler and Ghose (1970),Wildeman (1970), Lumsden and Lloyd (1984), and Lloydet al. (1985) showed that Mn2+ substitutes in both the Aand B sites of dolomite, but with a strong preference forthe B(Mg) site. This implies strong Ca-Mn order in man-ganoan dolomite, but the results are for very low Mnconcentrations and may not apply to CaMn(COr)r. Bol-zenius et al. (1985) have briefly described a structure re-finement of a magnesian kutnahorite of an unreportedcomposition but did not report on the inferred cationordering.

The binary system CaCOr-MnCO. was experimentallyinvestigated by Goldsmith and Graf (1957) and de Cap-itani and Peters (198 l), who found a solvus between rho-dochrosite and kutnahorite but none between kutnahorite

* Contribution no. 422 from the Mineralogical taboratory,Department of Geological Sciences, the University of Michigan.

0003404x/87/030443 r 9$02.00

and calcite. The ternary system CaCOr-MnCOr-MgCO.was evaluated at 500 to 800"C by Goldsmith and Graf(1960), who found wide solvi separating manganoan cal-cite, manganoan dolomite, and manganoan magnesite.They were unable to detect ordering reflections in syn-thetic kutnahorite and concluded that the similar scatter-ing powers of Ca and Mn result in very weak and perhapsundetectable ordering reflections even if Ca and Mn arehighly ordered. Goldsmith (1983) further pointed out thatshort-range order may occur in carbonates but be unde-tectable by X-ray diffraction. Capobianco and Navrotsky(1987) combined experimental and thermodynamic dataon synthetic CaCOr-MnCO. solid solutions to calculatea solvus for disordered carbonates cresting at ca. 330'Cand X.^.o, of 0.44. The calculated solvus is metastablerelative to a solvus or solvi involving kutnahorite withshort- or long-range order (Capobianco and Nawotsky,1987). Although the coherent and spinodal solvi were notgiven, their data suggest that disordered carbonates nearthe kutnahorite composition are unstable at Earth surfacetemperatures.

Despite the absence ofobserved ordering reflections inthe synthetic materials, the experimental solvus inferredbetween kutnahorite and rhodochrosite (Goldsmith andGraf,1957 de Capitani and Peters, l98l) suggests cationordering in kutnahorite by analogy with the systemCaCOr-MgCOr. This also would suggest a solvus between

3 1 9

320

TABLE 1. Microprobe analyses of manganoan carbonates

Oxide weight percent

8K.14 H-1 09405 sH-525K SH-525C

PEACOR ET AL.: CATION ORDER.DISORDER IN KUTNAHORITE

FeOMnOMgoCaOZnO

Sum

0.232.60.4

zJ-o

n.d.40.699.4

0.0030.4960.0110.490n.d.

Nofe: CO, calculated from stoichiometry; n.d. : not determined; sam-ple 525 from Sterling Hill for kutnahorite (K) and calcite (C).

kutnahorite and calcite, but this solvus has not yet beendemonstrated experimentally. The compositions of bothsynthetic and natural metamorphic phases that fall closeto the CaCO.-CaMn(COr), binary join are consistent withcomplete solid solution although some two-phase mate-rials and compositional gaps suggest low-temperatwe sol-vi (Frondel and Bauer, 1955; Goldsmirh, 1959, 1983;Winter et al., I 98 I ; de Capitani and Peters, I 98 I ; Essene,1983; Fig. l). The continuum of compositions could bedue to (1) crystallization of metastable, cation-disorderedcarbonates within a two-phase field as in the calcite-do-lomite system (Goldsmith, 1983); (2) location of a solvusat a temperature below that of the formation of the min-erals studied; or (3) the lack ofresolution offine-grained,two-phase intergrowths. Extrapolation of the experimentsin the temary system CaCO,-MgCOr-MnCO, (Gold-smith and Gral 1960) indicates that the solvus will in-tersect the CaCO3-MnCO. join at T < 400'C. This sys-tem demonstrates the importance of even minor Mg ingenerating a solvus assemblage that may be driven byordering of cations in the dolomite-type structure.

Lastly, the "dolomite problem" has long been an enig-ma to crystal chemists. It is not at all clear why Ca andMg are ordered in dolomite, resulting in a phase thatdetermines the form of the CaCOr-MgCO, system andhaving far-reaching consequences in natural carbonatesystems. Reeder (1983) has reviewed the factors that af-fect the relative stabilities of possible ordered compoundswith the dolomite structure. Interpretation is hindered bya lack ofdetailed knowledge ofsteric factors in the struc-tures ofseveral carbonates. As kutnahorite is one ofonlyfour known dolomite-structure minerals, it is essential todefine its detailed structure so that meaningful compari-sons of steric parameters can be made and to provideinsight into factors that lead to ordered dolomite-typestructures.

As part of a study of Mn minerals from Bald Knob(BK), North Carolina, Winter et al. (1981) characterizeda phase having a composition very near to that of stoi-chiometric kutnahorite, for which no ordering reflections

could be detected in single-crystal X-ray photographs. Thisspecimen contains only small amounts of other cationssuch as Mg. Because such additional components wouldhave significant effects on order-disorder relations andbecause relatively pure kutnahorites are unusual, we choseto refine its structure in order to directly determine thestate of order-disorder. In addition, Pete J. Dunn kindlymade available a specimen of Sterling Hill (SH), NewJersey, kutnahorite that has a composition nearly iden-tical to that from BK. Because some SH Ca-Mn carbon-ates are two-phase intergrowths (Frondel and Bauer, I 9 5 5;Goldsmith, 1959, 1983), the presence of a solvus is im-plied, and this in tum suggests ordering in these kutna-horites; we therefore chose to refine the structure of aspecimen from that locality.

Saupr-r cHARACTERIZATToN

The Bald Knob, North Carolina, sample (BK-14) isfrom a fine-grained metamorphic rock that equilibratedunder middle amphibolite-facies conditions at tempera-tures of 5 50-600"C (Winter et a1., I 98 l). This sample andothers at Bald Knob display an obvious gneissic layeringthat reflects differences in proportions and grain sizes ofMn carbonates and silicates, such that the carbonates mayvary significantly in composition between layers. How-ever, analyses of six different grains from sample BK-14gave results that are identical within analytical error. Al-though the analytical totals ofthe carbonate analyses inWinter et al. (1981) are erratic, reanalysis of kutnahoriteBK-14 with the University of Michigan automated ce-vrrca Camebax microprobe yields a formula close to thatoriginally reported but with better totals (Table l). Acleavage fragment was separated from the original spec-imen, and its composition was assumed to correspond tothe average (Ca. onMno ,oMgo o, Feo ooCOr) for the specimen(Table l).

The Sterling Hill sample (SH) was obtained from Har-vard University sample H-109405. The kutnahorite oc-curs as a coarsely crystalline vein cutting zinc ore (Fron-del and Bauer, 1955). This sample was chosen because asystematic study of compositions of Sterling Hill andFranklin, New Jersey, kutnahorites (P. J. Dunn, pers.comm.) had shown that it is nearly stoichiometric withonly minor amounts of Mg and thus is very close to theBald Knob sample in composition. Refinement of the SHcomposition during X-ray structure analysis is consistentwith the composition (CaorrMnoor)COr. The single crys-tal that was studied was ground away before microprobeanalysis could be obtained; analyses of the crushed grainsfrom which the single-crystal mount was selected yielded(Cao orMno orZno o, Mg o,Feo o, )COr. However, it is evidentthat the SH sample is somewhat variable around the idealkutnahorite composition. In the discussion that follows,it is assumed that the composition derived from theX-ray study is correct.

Powder-diffractometer patterns were obtained fromboth BK and SH kutnahorites. We attempted to indexthese patterns initially based on the indexing of Frondel

0.83 1 . 8

u.c25.31 . 4

40.3100.1

Atom fraction0.012o.4770.0130.4800.018

0.2v .b0.5

56.50 .1

32.599.4

0.8z I - o

1 . 137.80.8

3 1 . 199.2

FeMnMgCaZn

0.0080.3510.0250.6070.009

0.0020.1170.0100.8880.003

PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE 321

Trale 2. Powder diffraction data for kutnahorite(Sample BK-14, Bald Knob, North Carolina)

TneLe 3. Crystal data for Bald Knob and Sterling Hill kutnaho-rite

Space groupLattice parameters

a (A)

v(A')DitfractometerRadiation

DetectorAbsorption correction

p (cm ')

crystal shapeaverage vertex to vertex

distanc€ (mm)sin d/X limitReflections measuredNumber of reflectionsNumber of unobserved

reflectionsFinal residuals

F (all data)R (observed data)

Largest A/sigma

4.8732(81 4.894(1)16.349(6) 16.50(2)

336.2(1) 342.3(3)Weissenberg geometry o-scanMoKa, monochromated with graphite

crystalScintillation counter

50 50rhombohedron rhombohedron

0.26 0.20

0.65 0.65h > 0 ; h + k > 0 h > 0 ; h + k > 0

322 31978 54

3.9%3 .1%0 .12

disordered distribution of Ca and Mn and that the SHkutnahorite is at least partially ordered. In order to testthe BK sample for the possibility of short-range orderthat might occur in anti-phase domains of a size so smallas to be undetectable using X-ray difraction, a samplewas studied using transmission-electron microscopy. Anion-thinned sample was observed using the University ofMichigan JEoL JEM roocx electron microscope. The rela-tively featureless images so obtained are consistent witha randomly disordered structure.

Refinement was carried out with the program RFTNE2(Finger and Prince, 1975) using neutral scattering factorsof Doyle and Turner (1968) with equal Ca and Mn oc-cupancies initially assumed for the A and B sites of bothkutnahorites. The occupancies were not constrained bythe analytical data, and the only restriction was that thetotal site occupancies are 1.0. The weighting scheme ofCruickshank (1965) was used. All refinements were car-ried out assuming space group R3 for the ordered struc-ture. No correction for extinction was utilized becauseexamination of the structure factors implied that therewas no significant extinction factor. Refinement con-verged after refinement of anisotropic temperature fac-tors and occupancy factors for both cation sites. Includingunobserved reflections, the final R factors are 3.90/o forBK and 4.0o/o for SH kutnahorite. Final structure factorsare listed in Table 4,r structure parameters in Table 5,data from thermal ellipsoids in Table 6, and selected dis-tances and angles in Table 7. Distances and angles werecalculated using the program onrre (Busing et al., 1964),

' To obtain a copy of Table 4, order Document AM-87-332from the Business Ofrce, Mineralogical Society of America, 1625I Street, N.W., Suite 414, Washington, D.C. 20006, U.S.A. Pleaseremit $5.00 in advance for the microfiche.

SHilh

R3R3U100

340

1 02I0

1 11o312121

? 7 4

2.9392.4382.225

2.O421.8751.839

1 . 8161.5871.565't.486

1.4681.4561.4351.4071 .2911.256

J . / 5

2.9362.4372.2242.O442.0431.8751.8391.8171 .8161.5881.5661.4861.4681.4561.4341.4071.2921.257

0121041 1 01 1 3107202o240 1 80091 1 62111222140281 1 0

125300

0 2 , 1 0128

Notei The ordering reflections (107) and (009) are in-ferred to have an intensity oJ 0 based on single-crystalobservations

and Bauer (1955) in order to obtain data for least-squaresrefinement of lattice parameters. Ambiguities became ap-parent in the indexing of the ordering reflections, how-ever, and we therefore reindexed the pattern (Table 2).This showed that all observed reflections for both the BKand SH samples, as well as those of Frondel and Bauer(1955), are explained by nonordering reflections alone.

Cnvsr,l,r,-srRucruRE REFINEMENTS

Rhombohedral cleavage fragments of kutnahorite fromboth localities were mounted along c on a Supper-Pacediffractometer system. MoK" radiation, monochromatedby a flat crystal of graphite, was used. Intensities of re-flections with sin d/tr < 0.65 were measured and correctedfor Lorentz, polarization, and absorption effects, resultingin 322 and 319 reflections for the BK and SH kutnaho-rites, respectively. Absorption corrections were carriedout using a modified version of a program written byBurnham (1962a). Lattice parameters were determinedby least-squares refinement of powder-diffractometer data,obtained using CuI! with Si as an internal standard andutilizing the program rcrsq (Burnham, 1962b). The re-fined values are a: 4.8732(8) and c : 16.349(6) A forBK and a: 4.894(l) and c: 16.50(2) A for SH (TableJ t .

Before proceeding with measurements of intensity val-ues, single-crystal X-ray scans were carefully made at co-ordinates corresponding to ordering reflections. Scans forthe Bald Knob specimen showed no observable intensi-ties for any ordering reflections except a barely observ-able peak for (107). This confirmed the single-crystal filmobservations for which no ordering reflections were ob-served. By contrast, several ordering reflections were ob-served to have weak-to-medium intensities for the SHcrystal. These data imply that the BK sample has a highly

4.O%3.1"h0.09

322 PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

TneLr 5. Structure parameters for Bald Knob (upper lines) and Sterling Hill (lower lines) kutnahorite

00

A

B

o

0000000.2634(3)0.2s68(3)

0000000.9995(4)0.98s0(4)

Y2Y2

0.2495(2)0.2461(2)0.2496(2)o.2472(1)

RP23P33

a

A

B

U

0.01 15(3)0.01 20(2)0.001 2(3)0.01 08(2)0.01 22(9)0.01 24(1 0)0.0131(s)0.01 22(6)

0 .01 150.01200.00120.01080.01220.01240.0272(61o.0244(71

0.001 0(1 )0.001 1(1)0.001 0(1 )0.001 1(1)0.001 1(2)0.0013(1)0.001 9(1)0.001 7(1 )

0.00570.00600.00560.00540.00610.00620.0125(4)0.01 1 1(6)

000000

- 0.0001 (1 )- 0.0004(1 )

00

000

- 0.0003(1 )- 0.0007(1 )

standard errors in lattice parameters, and the full vari-ance-covariance matrix of the structure refinement.

DrscussroN

Site occupancies

The results for the BK and SH kutnahorites are verydifferent, as anticipated from the absence and presence ofordering reflections in single-crystal photographs of therespective samples. The BK kutnahorite is almost com-pletely disordered in Ca-Mn distribution; the SH kutna-horite is largely ordered.

The disorder of BK kutnahorite is clearly indicated bythe near-equivalence of structure parameters for both theA and B cation sites (Tables 5-7). The average M-O dis-tances, for example, are 2.277(2) and 2.267(2) A, respec-tively. Final refined occupancies correspond to site com-

TABLE 6. Thermal€llipsoid data for Bald Knob (upper lines) andSterling Hill (lower lines) kutnahorite

positions Ca" on,o.,Mno ,, and Cao s3(4)Mno o, for the A and Bsites, respectively. The high occupancy value ofCa rela-tive to Mn in the B site may in part be attributed toordering of the small amount of Mg in the site in that theform factor of Mg will be accommodated by that of Ca.

Refined occupancies for the SH kutnahorite areCao ro,o,Mno ,u and Cao rr,r,Mrto, for the A and B sites, re-spectively, which correspond to the compositionCa" rrMnoorCOr, neglecting the small contributions of Mgand Fe. By comparison, the composition implied by thelattice parameters is CaorrMnoor, assuming an ideal com-position in the CaCO.-MnCO, binary and Vegard's law.This also is in fair agreement with the composition asdetermined by electron-microprobe analysis.

The Ca-O and Mn-O distances determined by Eflen-berger et al. ( 1 9 8 I ) for calcite and rhodochrosite are 2.362and2.l90 A, respectively. Assuming the site occupanciesgiven above, linear interpolations between these dis-

Tner-e 7. Selected interatomic distances (A) and an-gles for Bald Knob and Sterling Hill kutnahorite

Bald Knob Sterling Hill

Amplitude(A)

Angle toa ' f )

Angle toAngle to b. (') c- f)

1

3

1

3

1

3

1

2

3

0.1 02(1 )0.1 04(1 )0.1 18(2)0.'t22(2)0.1 00(1 )0.09s(1 )0.1 18(2)0.124(210.1 05(4)0.106(3)0.1 24(8)0.1 37(9)0.094(2)0.093(3)0.1 s4(2)0.1 43(2)0.162(2)0.1 60(2)

onon

1 76(1 )174(2)93(2)e5(2)s3(2)e2(2)

9090e4(2196(2)36(1 1)36(7)54(1 1)54(7)

90 9090 90

90 9090 90

909000

909000

909000

el(2)8s(2)

126(121126(7)36(12)36(7)

A polyhedronCa-Oo-o'o-o'O-Ca-O'O-Ca-O"QE

B polyhedronMn-Oo-o'o-o'O-Mn-O'O-Mn-O"QE

Carbon polyhedronc-oo-oo-c-o

2.277(2)3.1s2(3)3.288(3)

87.59.(9)92.41.(9)1 .001 7(5)

2.267(2)3.146(3)3.264(3)

87.90.(e)e2.10.(e)1 .001 3(1 2)

1 .285(1 )2.225(21

120.0e(1)

2.351(2)3.244(313.404(3)

87.76'(8)92.74"(8)1.0022(3)

2.217(2)3.074(3)3.1 96(3)

87.76(7)92.24(8)1 .001 5(1 0)

1 .2s6(1)2.244(2)

1 19.9812)

tances result in predictions of M-O distances of 2.334and2.236 A, which are different than the observed valuesof 2.351 and 2.217 A. If such a linear interpolation isvalid, this implies either that the degree of order is higherthan indicated by the site occupancies, or perhaps thatsome stress factor prevents their adjustment to ideal val-ues.

Stereochemistry vs. ordering

Insofar as the SH and BK kutnahorites are the sub-stantially ordered and disordered equivalents of nearlystoichiometric kutnahorite, they provide a unique op-portunity to compare structure parameters in dolomite-type structures. The most striking comparison of parain-eters is in the thermal ellipsoid data, which are nearlyidentical for both structures (Table 6). This implies thatthe values for BK kutnahorite do not include a contri-bution due to positional disorder. Even though differentindividual oxygen sites must be associated with differentspecies in A and B sites, the positions of all oxygen atomsmust be virtually identical. This implies a lack of short-range order below the level ofdetectability by X-ray dif-fraction for BK kutnahorite, as also concluded from reuobservations. Reeder (1983) tabulated rms amplitudes ofthermal ellipsoids for available refinements of dolomite(Althofl 1977; Reeder and Wenk, 1983; Reeder, 1983)and ankerite (Beran and Zemann, 1977). These valuesare all smaller than those for BK and SH kutnahorites,including those for vibration parallel to c. Similarly,Reeder listed values for several calcite-type structures forrefinements carried out by Althoff (1977) and Effenbergeret al. (1981). They are also smaller than those for kut-nahorite, although those for calcite are almost as large.We have no explanation for the disparity. We re-empha-size, however, that the same factors must have equivalenteffects in both the ordered and disordered structures.

The octahedra in calcite- and dolomite-type structuresare trigonally distorted. Several authors (Althofl 1977;Beran and Znmann, 1977; Rosenberg and Foit, 1979; Ef-fenberger et al., l98l; Reeder, 1983) have discussed thesignificance ofthe distortion using the quadratic elonga-tion (QE) parameter of Robinson et al. (1971). It is afunction of rotation of the CO, groups about c, resultingin rotation in an opposite sense ofthe vertices ofboth Aand B octahedra. The rotation results in an increase inthe A-O distance and a decrease in the B-O distance.

Reeder (1983) has plotted QE versus mean A-O andB-O distances for both calcite- and dolomite-type struc-tures. In general there is an increase in QE with distance(and therefore with increase in ionic radius) for calcite-type structures, where no CO, rotation is possible and alloctahedra are equivalent. However, rotation in dolomitecauses a nonequivalence in both QE and mean bond dis-tance for the A and B sites. QE for the B site (1.0008-1.0009) is only slightly less than for magnesite (1.0009-1.0010), but that for the A site (1.0016) is significantlyless than that for calcite (1.0020). Rotation of the car-bonate groups thus results in smaller octahedral distor-

323

tion while accommodating the large difference in iomcradius of Ca and Mg. Reeder (1983) has emphasized thiscoupling of the steric effects of the A and B sites; boththe absolute and relative ionic radii are significant in sta-bilization of a dolomite-type structure.

Attempts to explain the lack of a dolomite-typestructure for CaFe(COr), compared to CaMg(COr)r,CaMn(COr)r, CaZn(CO')r, and CdMg(COr), and to ex-plain their relative stabilities on the basis offactors suchas ionic radii or octahedral site-preference energies havenot been successful. The coupling of dimensions of the Aand B sites through a common oxygen atom may be thecritical factor (combined with site-preference energies forcertain transition-element ions), but we can frnd no sim-ple function for correlating such parameters.

Lacking a more appropriate function, the parameter

QE appears to be the best available one for evaluatingthe relative stability of a dolomite-type structure. For BKkutnahorite, the QE values (Table 7) for the A and B sitesare only slightly different than the average value (1.0014)for rhodochrosite and calcite, as expected for a disorderedstructure (Reeder, 1983). The QE value for the A site ofthe ordered SH kutnahorite is much greater than the av-erages of the calcite-type structures (Table 7). Thus, theCO, group rotation results in interatomic distances ap-proximately consistent with ionic radii, but at the priceof significant A-site octahedral distortion. Although thedistortions in kutnahorite are not significant enough tocause instability of the structure, they may be importantin constraining the order-disorder transition to low tem-peratures. The B-O distances in SH kutnahorite areslightly larger and A-O distances smaller than those pre-dicted from radii, and site occupancies may possibly bedue to the coupling ofdistortions in the A and B sites.

Reeder (1983) and Reeder and Wenk (1983) havepointed out that the C atom is displaced from the planeof three coordinating oxygen atoms in dolomite-typestructures. The displacement is in response to differencesbetween the ordered cations in layers above and belowthe carbonate groups. In BK kutnahorite, this displace-ment is nearly zero (0.002 A), but in SH kutnahorite, itis 0.016 A. It therefore provides a significant measure ofthe degree of cation order as in the case of dolomite.

The original powder pattern of kutnahorite (Frondeland Bauer, 1955) was incorrectly indexed as to orderingreflections. The scattering factors of Ca and Mn are sosimilar that such reflections are much weaker than in do-lomite where they are a function of the large diferencein scattering power of Ca and Mg (Goldsmith, 1983). Ifthese reflections could be identified in powder patterns,the degree of ordering would be much more easily deter-mined than through single-crystal patterns. We thereforeobtained a diffractometer scan using a portion of the Ster-ling Hill sample, maximizing conditions for observationofweak reflections, but no ordering reflections could bedetected. This implies that powder-diffraction data shouldnot be used to infer ordering in nearly pure Ca-Mn car-bonates unless special care is used, as is the case with

PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

324 PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

/ de Copitoni I Peters (1981)

50Mo | "/" lMnCO3

Fig. 1. Manganoan carbonateswith less than 10 molToMgCO,and FeCO3 projected onto the CaCOr-MnCO, binary join modi-fied from de Capitani and Peters (1981). SN : Serro do Navio,MP : Muretto Pass, B : Buritirama, A : Alagna, S : Scerscen.Additional data are BK (Bald Knob, North Carolina; Winter etal., 1981), AI (Andros Island, Greece; Reinecke, 1983), and SH(Sterling Hill, N.J.; this work). Solid lines are the inferred solviof Goldsmith and Graf (1957) and of de Capitani and Peters(1981), and dashed lines are possible solvi consistent with ex-perimental and natural data.

Rietveld-type refinements. Single-crystal patterns, how-ever, clearly show ordering reflections.

Nomenclature

In describing kutnahorite we have faced a seeminglytrivial, yet complicating aspect of nomenclature. The namekutnahorite was clearly meant to be applied to the or-dered phase with ideal composition CaMn(CO.), (Fron-del and Bauer, 1955). The probability that the type ma-terial was ordered is high, even though ordering reflectionswere erroneously identified. The name kutnahorite shouldnot be applicable to disordered phases, sensu stricto.However, we recommend its use for all phases nearCaMn(CO.), in composition, with adjectives such as"disordered" or "calcian" specifying aspects of structureand chemistry as they become determined. No confusionseems to result from such usage.

Irnplications of ordered vs. disordered kutnahorite

The occurrence ofdisordered kutnahorite at Bald Knoband ordered kutnahorite at Sterling Hill presumably re-stricts the order-disorder transition to a temperature be-tween those of the formation ofthese two carbonates. Theformation of ordered kutnahorite in late, low-tempera-ture veins at Sterling Hill is inferred to indicate growthof an ordered phase within its stability field. Unfortu-nately the equilibration temperature of the SH carbonateis uncertain. Our prejudice, based on estimates of late-stage hydrothermal parageneses at Sterling Hill wouldplace the formation of the SH carbonate veins and theorder-disorder transition somewhere in the range of 200to 400'C with the former at lower temperatures than thelatter. The survival of disordered BK kutnahorite dem-

onstrates that the kinetics ofthe ordering process are slug-gish even for the geologic times attendant with slow cool-ing of a regional metamorphic terrane (Winter et al., l98l).

Compositions of natural Mn carbonates in thesystem CaCO.-MnCO,

Observation of solid solutions and possible soh'us pairsof natural Mn carbonates provides insight into their phaseequilibria. Data on natural carbonates close to the Ca-CO.-MnCO, join give some constraints on solvi alongthis join. Using estimated equilibration temperatures, deCapitani and Peters (1981) projected the compositions ofnatural carbonates onto a CaCOr-MnCO. binary in com-parison with experiments on the kutnahorite-rhodochro-site (KR) solvus. Their figure has been expanded to in-clude data from Winter et al. (1981), Reinecke (1983),and this paper (Fig. l). The compositions of the naturalcarbonates are generally consistent with the experimentsof Goldsmith and Graf (1957) and de Capitani and Peters(1981) on the KR solvus (Fig. 1). However, its locationis constrained only as a maximum because of the unre-versed nature of their experiments. Successful experi-ments involving homogeneous solid solutions with equi-librium distributions of cations and initially located withinthe I(R solvus are needed to reverse its location (Essene,1983). Attempts have yielded no detectable reaction toform two-phase materials (Goldsmith and Graf, 1957; deCapitani and Peters, 1981). Experiments would have toinvolve kutnahorites with the equilibrium ordering. Fora complete understanding of this system by analogy withCaCOr-MgCO, (Burton and Kikuchi, 1984), the temper-ature of order-disorder transition of kutnahorite in rela-tion to the equilibrium solvi and tricritical points needsto be determined for different CalMn ratios.

Few two-phase pairs of carbonates near the join Ca-CO,-MnCO, have been reported. Some analytical dataare available on coexisting Mn carbonates that are closeto the CaCOr-MnCO, join: kutnahorite-rhodochrosite(KR) from blueschists at Andros Island, Greece (Rei-necke, 1983), and calcite-kutnahorite (CK) from SterlingHill. From the appearance of the sample 525 (Fig. 2l\),we tentatively infer that it formed by epitaxial growthrather than by exsolution. Sample 526 shows a morecomplex intergrowth of earlier euhedral calcite and laterreplacive (?) calcite in a calcian rhodochrosite host (Fig.2B). The textures of the two-phase materials do not re-semble those commonly observed for high-temperaturecalcites containing exsolved dolomite. However, an exso-lution or replacement origin for the Sterling Hill inter-growths cannot at present be ruled out. In any case, thisassemblage provides direct evidence (Table l) for the ex-istence of a low-temperature solvus between calcite andkutnahorite, but its shape and location remains ill-con-strained.

Compositions of natural Mn carbonates in the systemCaCOr-MnCOr-MgCO,

Before accepting that the compositions of natural car-bonates in Figure I properly constrain the binary solvi,

G o l d s m i t h I G . o f ( 1 9 5 7 )

l-.rr.rl i

'-"i

IA I ,

Co CO3

PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

CoCO.

325

Fig. 2. Backscattered-electron images of manganoan calcite(dark) intergrown in calcian kutnahorite (light) from Sterling Hill,N.J. (A) (SH-525) shows irregular "arrowheads" of manganoancalcite in kutnahorite (length ofphoto 335 pm). Fine dark linesare cracks and cleavages. (B) (SH-526) shows hexagonal inter-gowths and late fracture fillings (?) of manganoan calcite in coarsekutnahorite (length of photo 720 pm). Spots with dark circlesare from surface dust.

it is important to consider the possible effects of addi-tional components on the solvi. Compositions of mostnatural manganoan carbonates are more completely rep-resented by the ternary system CaCOr-MnCOr-MgCO.because they commonly have 5 to l0 molo/o substitutionof MgCO. (Fig. 3). The solid solution of FeCO. is gen-erally much less important; analyses with > 5 molo/o FeCO,have been excluded from the plot. Despite the wide ter-nary solvi of Goldsmith and Graf (1960), few equilibriumtwo- or three-phase manganoan carbonate assemblageshave been reported. The Andros Island kutnahorite co-existing with rhodochrosite has significant MgCOr,whereas the Sterling Hill kutnahorite coexisting with cal-cite, though low in MgCOr, varies erratically in CaCOr.

Nilg CO3 Mn C03

Fig. 3. Manganoan carbonates with less than 5 molo/o FeCO,projected onto the CaCOr-MnCOr-MgCO, ternary in compari-son with the experimental solvi of Goldsmith and Graf (1960)at 600'C. References: Doelter (1912,1914,1917), Krieger (1930),Wayland (1942), Ham and Oakes (1944), Frondel and Bauer(1955), Goldsmith and Graf (1955), Lee (1955), Stankevich(1955), Graf and Lamar (1956), Muta (1957), Karado (1959),Van Tassel and Scheere (1960), Deer et al. (1962), Pavlishin andSlivko (1962), Sundius (1963), Saliya (1964), Trdlicka (196D,Gabrielson and Sundius (1965), Pisa (1966), Sundius et al. (1966),Trdlicka and Sevcu (1968), Calvert and Price (1970), Wenk andMaurizio (1970), Della-Salla et al. (1973), Koritningetal. (1974),Shibuya and Harada (1976), Kashima and Motomura (1977),Peters et al. (1977,1978, 1980), Sanford (1978), Cortesogno etal. (1979), Suess (1979), Dabitzias (1980), Melone and Vurro(1980), Sivaprakash (1980), Bel lo et al. (1981), Fukuoka{1981),Green et al. (1981), Tanida and Kitamura (1981, 1982), Winteret al. (1981), Zak and Povondra (1981), Pedersen and Price(1982), Iwafuchi et al. (1983), Nahon et al. (1983), Reinecke(1983). Point lying near center of three-phase triangle is fromNahon et al. (1983) and deserves further studv.

Except for carbonates along the magnesite-calcite binaryjoin, most one-phase manganoan carbonates lie outsidethe 600'C isotherm consistent with lbrmation at or belowthat temperature (Fig. 3). However, if disordered kutna-horites grow metastably at low temperatures, it is alsopossible that phases with metastable intermediate com-positions may form as authigenic or diagenetic mineralsin sediments (Suess, I 979; Pedersen and Price, I 98 2; Ber-an et al., 1983; Boyle, 1983). No low-Fe carbonates areknown to occur as solid solutions between magnesite andrhodochrosite (Essene, 1983, Fig. 3). A solvus could begenerated between magnesite and rhodochrosite by ex-pansion of the experimental two-phase region betweencalcian magnesite and calcian rhodochrosite at 600'C (Fig.3) down to the MgCO.-MnCO. binary at low tempera-tures (Fig. 4A). Alternatively, it is possible that appro-priate protoliths have not yet been found that would yieldthese solid solutions (Essene, 1983). Permissive evidencefor a low-temperature magnesite-rhodochrosite solvus isprovided by the experiments of Erenburg (1961), who

Co CO.

{iii?' ;' I i \ \ \,,"'.:j;)

326 PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

Mn CO.

Co CO3

Mg CO3 Mn COE

Fig. 4. Ternary diagrams showing the compositions of two-phase manganoan carbonates from Andros Island, Greece(Reinecke, 1983), and from Sterling Hill, N.J. (this work). Pos-sible solvi at 300 + l00qC are consistent with these data and theexperimental solvi of Goldsmith and Graf (1960) at higher Z.Diagram A shows no miscibility gap between rhodochrosite andmagnesite; diagram B extends the ternary gap down to intersectthis binary join.

precipitated two carbonate phases when starting with ahigh Mn/Mg ratio in aqueous experiments on this binary.The rare magnesian rhodochrosites (Fig. 3) need to beexamined carefully for coexistence with or exsolution ofmanganoan magnesite to test these two alternatives. Wehave speculated in Figures 4,{ and 48 as to possible re-lations of the 300'C isotherms in the ternary system usingthe two-phase pairs from Andros Island and Sterling Hillin combination with single-phase carbonates and the ex-periments at higher temperatures. If equilibrium prevails,a third magnesian kutnahorite-rhodochrosite pair (Tsu-sue, 1967) suggests either strong compositional effects onthe Ko for Mg-Mn distribution between these two car-

bonates (Fig. aA) or formation of a three-phase field (Fig.4B). Shibuya and Harada (1976) and Zak and Povondra(1981) have described patchy-to-oscillatory variationsbetween manganoan ankerite and magnesian kutnahoritethat they ascribe to disequilibrium processes. These car-bonates may bear upon possible solvi in the quaternarysystem CaCOr-MnCOr-MgCOr-FeCOr, but no reversedexperiments are available for comparison in this system.

AcxNowr-nocMENTs

We again thank W. B. Simmons, Jr., for introducing us to the BaldKnob manganese deposit. C. Capobianco and A. Navrotsky kindly pro-

vided us with a preprint of their manuscript in advance of publication.

C. B. Henderson, L. A. Marcotty, and W. C. Bigelow of the Universityof Michigan Microbeam Laboratory are acknowledged for technical as-sistance. We are gateful to R. J. Reeder, W. B. Simmons, Jr., and ananonymous reviewer for their insightful comments on our manuscript.We also acknowledge NSF Grants EAR-8212764, 85R-83 and EAR-8408168 in partial support ofthis research.

RnpnnBNcBs

Althoff, P.L. (197?) Structural refinements ofdolomite and a magnesiancalcite and implications for dolomite formation in the marine environ-ment. American Mineralogist, 62, 7 7 2-7 83.

Bello, R.M.S., Coutinho, J.M.V., Valarelli, J.W., Schultz, R.A., and Ya-

mamoto, J.K. (1981) Dados odicionais sobre a mineralogia dos pro-

tominerios de manganes de Serra do Navio-Amapa. Annaes Academiade Ciencias, Rio de Janeiro, 53,123-134.

Beran, A., and Zemann, J. (1977) Refinement and comparison of thecrystal structures of a dolomite and of an Fe-rich ankerite. TschermaksMineralogische und Petrographische Mitteilungen, 24, 27 9-286.

Beran, A., Faupl, P., and Hamilton, W. (1983) Die Manganschiefer derStrubbergschichten (N6rdliche Kalkalpen, Osterreich)-eine diagene-tisch gepragte Mangankarbonatyererzung. Tschermaks Mineralogischeund Petrographische Mitteilungen, 31, 17 5-192.

Bolzenius, 8.H., Farkas, L., and Will, G. (1985) Strukturuntersuchung annatiirlichem Kutnahorit (abs.). Fortschritte der Mineralogie, 63, 29.

Boyle, E.A. (1983) Manganese carbonate overgroMhs on foraminiferaltests. Geochimica et Cosmochimica Acta, 47, 1815-18 19.

Burnham, C.W. (1962a) rclse, crystallographic lattice-constant least-squares refinement program. Carnegie Institution of Washington YearBook 61, 132-135.

-(1962b) Absorption corrections for prismatic crystals and evalu-ation of end effect. American Crystallographic Association 1962 An'nual Meeting Program and Abstracts, 19.

Burton, B., and Kikuchi, R. (1984) Thermodynamic analysis ofthe sys-tem CaCO,-MgCO, in the tetrahedron approximation of the clustervariation method. American Mineralogist, 69, 165-175.

Busing, W.R., Martin, K.D, and Lerry, H.A. (1964) A Fortran crystallo-graphic function and error program. U.S. National Technical Infor-mation Service ORNL-TM-306.

Calvert, S.8., and Price, N.B. (1970) Composition of manganese nodulesand manganese carbonates from Loch Fyne, Scotland. Contributionsto Mineralogy and Petrology, 29,215-233.

Capobianco, C., and Navrotsky, A. (1987) Solid solution thermody-namics in CaCOr-MnCOr. American Mineralogist, 7 2, 312-3 18.

Cortesogno, L., Lucchetti, G, and Penco, A M. (1979) Le mineralizza-zioni a manganese nei Diaspri delle ofiolite Liguri: Mineralogia e ge-

nesi. Rendiconti Societa Italiani di Mineralogia e Petrologia, 35, 151-197

Cruickshank, D.W.J (1965) Errors in least-square methods. In J. S. Rol-lett, Ed. Computing methods in crystallography, p. 114. Pergamon Press,Oxford

Dabitzias, S.G. (1980) Petrology and genesis of the Vavdos cryptocrys-talline magnesite deposits, Chalhdiki Peninsula, Northem Greece.Economic Geology, 75, 1 138-l l5l.

de Capitani, C., and Peters, Tj. (1981) The solvus in the system MnCOrCaCO3. Contributions to Mineralogy and Petrology, 76,394400.

Deer, W.A., Howie, R.A., and Zussman, J. (1962) Rock-forming min-erals. vol. 5, Non-silicates. Longmans, Green and Co. Ltd., London.

Della-Sala, J., Fuentes, F C., Herrero, M., Iaserre, N., and Viand, J. (1973)Contribution to the knowledge ofthe Argentine rhodochrosite. Gemsand Gemology, 14, l4l-143

Doelter, C. (1912,1914,1917) Handbuch der Mineralchemie. SteinkoplDresden.

Doyle, P.A., and Tumer, P.S. (1968) Relativistic Hartree-Fock X-ray andelectron scattering factors. Acta Crystallographica, 424, 39U397.

Effenberger, H, Mereiter, K., and Zemann, J. (1981) Crystal structurerefinements of magnesite, calcite, rhodochrosite, siderite, smithsoniteand dolomite, with discussion of some aspects of the stereochemistryof calcite-type carbonates. Zeitschrift fiir Kristallographie, 156, 233-243.

Erenburg, B G. (1961) Some artificial mixtures ofcarbonates with calcitestructure. Zhurnal Strucktumoi Khimii. 2. 212-2 13.

Essene, E.J. (1983) Solid solutions and solvi among metamorphic car-bonates with applications to geologic thermobarometry. MineralogicalSociety of America Reviews in Mineralogy, 11,77-96.

Finger, L.W., and Prince, E. (1975) A system of Fortran IV computerprograms for crystal structure calculations U S National Bureau ofStandards Technical Note 854.

Frondel, C., and Bauer, L.H. (1955) Kutnahorite: A manganese dolomite:CaMn(CO.)r. American Mineralogist, 40, 7 48-7 60.

Fukuoka, M. (198 I ) Mineralogical and genetical study on alabandite fromthe manganese deposits of Japan. Memoirs of the Faculty of Sciences,Kyushu University, Series D. Geology, 24,207-251.

Gabrielson, O, and Sundius, N. (1965) Ca-rich kutnahorite from Lang-ban, Sweden. Arkiv liir Mineralogi och Geologi, 4,287-289.

Goldsmith, J.R. (1959) Some aspects of the geochemistry of carbonates.In P.H. Abelson, Ed. Researches in geochemistry, vol. l. Wiley, NewYork.

-(1983) Phase relations of rhombohedral carbonates. MineralogicalSociety of America Reviews in Mineralogy, ll,49-76.

Goldsmith, J.R., and Graf, D.L. (1955) The occurrence of magnesiancalcites in nature Geochimica et Cosmochimica Acta. 7.212-230.

-(1957) The system CaO-MnO-COr: Solid solution and decompo-sition relations. Geochimica et Cosmochimica Acta, I l, 310-334.

- (1960) Subsolidus relations in the system CaCOr-MgCO,-MnCOr.Joumal of Geology, 68, 324-335.

Graf, D.L, and Lamar, J.E. (1956) Properties of calcium and magnesiumcarbonates and their bearing on some uses ofcarbonate rocks. Econom-ic Geology, 50th Anniversary Volurne, 639-713.

Green, G.R., Solomon, M., and Walshe, J.L. (1981) The formation of thevolcanic-hosted massive sulfide ore deposit at Roseberg, Tasmania.Economic Geology, 7 6, 304-338.

Ham, W.E., and Oakes, M.C. (1944) Manganese deposits of the Bromidedistrict, Oklahoma. Economic Geology, 39, 4lZ-443.

Iwafuchi, K., Watanabe, C., and Otsuka, R. (1983) Thermal decomposi-tion of magnesian kutnahorite. Thermochimica Acta, 60, 361-38 1.

Kara'do,Z. (1959) Chemical analyses ofJapanese minerals. Journal oftheFaculty of Sciences, Hokkaido University, Series IV. Geology and Min-eralogy,4, l-93.

Kashima, N., and Motomun, Y. (1977) Mn-carbonate nodule from theSusaki Formation of the Shimanto Belt, Shikoku. Klushu UniversityScience Reports ofthe Department ofGeology, 12,273-278.

Koritning, S., Mielki, P., and Meixner, H. (1974) Kutnahorite from Forst-haus Zinecker, I-oelling, Carinthia. Karinthin, 7 1, 12U124.

Krieger, P. (l 930) Notes on an X-ray diffraction study ofthe series calcite-rhodochrosite. American Mineralogist, | 5. 23-29.

Lee, D. (1955) Mineralogy of some Japanese manganese ores. StanfordUniversity Publications, Geological Sciences, 5, l-64.

Lloyd, R.V., Lumsden, D.N, and Gregg, J.M. (1985) Relationship be-tween paleotemperatures of metamorphic dolomites and Esn deter-mined Mn (II) ratios. Geochimica et Cosmochimica Acta, 49,2565-2568.

Lumsden, D.N., and Lloyd, R.V. (1984) Mn (II) partitioning betweencalcium and magnesium sites in studies of dolomite origin. Geochimicaet Cosmochimica Acta. 48. l86l-1865.

Melone, N., and Vuno, F. (1980) Osservazioni cristallochimiche su uncarbonato di Ca, Mn e Fe di Campiglia Marittima. Mineralogica etPetrographica Acta, 24, I 5 l-l 59.

327

Muta, K (1957) On the carbonate minerals associated with ore depositsin Kyushu (I,II). Japanese Association of Mineralogists, Petrologists,and Economic Geologists, 41, 1G23,80-.87.

Nahon, D, Beauvais, A., Boeglin, J.-L., Ducloux, J., and Nziengui-Ma-pangou, P. ( I 98 3) Manganite formation in the first stage of the lateriticmanganese ores in Africa. Chemical Geology, 40,25-32.

Pavlishin, V.I., and Slivko, M.M. (1962) Isomorphic miscibility in theCaCO,-MnCO, Series. Mineralogicheskii Sbornik, 19 62, 445-449.

Pedersen, T.F., and Price, N.B. (1982) The geochemistry of manganesecarbonates in Panama basin sediments. Geochimica et CosmochimicaActa. 46. 59-68.

Peters, Tj., Valarelli, J.V., Coutinho, J.M.V., Sommerauer, J., and vonRaumer, J. (1 977) The manganese deposits ofBuritirama (Para, Brazil).Schweizerische Mineralogische und Pefographische Mitteilungen, 57,- t t 5 - 5 L t ,

Peters, Tj., Trommsdorff, V, and Sommerauer, J. (1978) Manganese py-

roxenoids and carbonates: Critical phase relations in metamorphic as-semblages from the Alps. Contributions to Mineralogy and Petrology,66, 383-388.

-(1980) Progressive metamorphism of manganese carbonates andcherts in the Alps. Akademia Kiado Budapest, 18,271:283.

Pisa, M. (1966) Mineralogenesis of the Pb-Zn deposit at Bohutin nearPribram. Sbomik Geologickych Ved, 7, 5-164.

Reeder, R.J. (1983) Crystal chemistry of the rhombohedral carbonates.Mineralogical Society of America Reviews in Mineralogy, ll, l-47.

Reeder, R.J., and Wenk, H.R. (1983) Structure refinements of some ther-mally disordered dolomites. American Mineralogist, 68, 7 69-77 6.

Reinecke, T. (1983) Mineralogie und Petrologie der Mangan- und Eisen-reichen Meta-sedimente von Andros, Kykladen, Griechenland. Carolo-Wilhelmina zu Braunschweig Dissertation, NaturwissenschaftlichenFakultiit der Technischen Universifit.

Robinson, K, Gibbs, G.V., and Ribbe, P.H. (1971) Quadratic elongation:A quantitative measure of distortion in coordination polyhedra. Sci-ence. 172.567-570.

Rosenberg, P.E., and Foit, F F. (1979) The stability of transition metaldolomites in carbonate systems: A discussion. Geochimica et Cosmo-chimica Acta. 43. 951-955.

Saliya, D.G (1 964) Vein carbonates from a gold deposit in the ArmenianSSSR Trudy Kavkazskoho, Instituta Mineralnogo Syr'ya Mineralogiii Okhrany Nedr, SSSR, 1964,271-279 (not seen, extracted from Chem-ical Abstracts, 64,7874e, 1966).

Sanford, R.G (1978) Regional metamorphism ofultramafic bodies andtheir contact zones, western New England. Ph.D. thesis, Harvard Uni-versity, Cambridge, Massachusetts.

Schindler, P., and Ghose, S. (1970) Electron paramagnetic resonance ofMn'?* in dolomite and magnesite, and Mn?* distribution in dolomite.American Mineralogist, 55, 1889-1896.

Shibuya, G., and Harada, S. (1976) Mineralogical studies on kutnahoriteand other carbonate minerals from the Hoei Mine, Oita Prefecture.Journal ofthe Mineralogical Society ofJapan, 13, 1-25.

Sivaprakash, C. (1980) Mineralogy ofmanganese deposits ofKoduru andGarbham, Andhra Pradesh, India. Economic Geology, 75, 1083-1 104.

Stankevich, L.L. (1955) Calcian rhodochrosite from the Kerch deposits.Akademii Nauk SSSR Doklady, 105, 1328-1331.

Suess, E. (l979) Mineral phases formed in anoxic sediments by microbialdecomposition of organic matter. Geochimica et Cosmochimica Acta,43,339-352.

Sundius, N. (1963) Rhodochrosite from LEingban. Arkiv ftr Mineralogioch Geologi, 3,293-295

Sundius, N., Parwel, A., and Rajandi, B. (1966) Carbonates from Vaster-Silvberg. Arkiv Iiir Mineralogi och Geologi, 4,413-415.

Tanida, K., and Kitamura, T. (1981) Studies in the system lime-man-ganese-oxygen. I. Phase equilibria in the system CaO-MnO at 1 100"and 1400'C in air, with the mineralogy and thermal transformation ofkutnahorite from Fujikara mine, Iwate Prefecture. Tohoku DaigakuSenko Seiren Kenkyusho lho, 37, 33-42.

-(1982) Mineralogy and thermal transformation of kutnahorite fromFujikura Mine, Iwate Prefecture. Journal of the Japanese Associationof Mineralogists, Petrologists and Economic Geologists, 7 7, 227 -234.

Trdlicka, Z. (1964) Beitrag zur Differential-Althermoanalyse der Gang-mangancalcite und Rhodochrosite von Tscheschoslowakei. Acta Univ-ersitatis Carolinae Geologica, 2, 159-167.

PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

328

Trdlicka,Z., and Sevcu, J. (1968) Chemisches und RoentgenographischesStudium des Kutnohorits von Kutna Hora. Acta Universitatis CarolinaeGeologica, 3, 175-189.

Tsusue, A. (1967) Magnesian kutnahorite from Ryuima mine, Japan.American Mineralogist, 52, 17 5l-l'1 61.

Van Tassel, R., and Scheere, J. (1960) Contribution d le petrographie desroches carbonat6es du Westphalien Belge. Soci6t6 Belge G6ologie Bul-let in.69. 234-276.

Wayland, R.G. (1942) Composition, specific gravity and refractive indicesof rhodochrosite from Butte, Montana. American Mineralogst,2T, 614-628.

Wenk, H.R., and Maurizio, R. (1970) Geological observations in the Ber-gell area (S W. Alps). II. Contact minerals from Mt. Sissone-Cima de

PEACOR ET AL.: CATION ORDER-DISORDER IN KUTNAHORITE

Y azzeda. Schweizerische Mineralogische und Petrographische Mittei-lungen, 50, 349-354.

Wildernan, T.R. (1970) The distribution of Mn'* in some carbonates byelectron paramagnetic resonance. Chemical Geology, 5, 167-177.

Winter, G.A., Essene, E.J., and Peacor, D.R. (1981) Carbonates and py-roxenoids from the manganese deposit near Bald Knob, North Caro-lina. American Mineralogist, 66, 27 8-298.

Zak,L.,and Povondra, P. (1981) Kutnahorite from the Chvaletice Pyriteand Manganese Deposit, East Bohemia. Tschermaks Mineralogischeund Petrographische Mitteilungen, 28, 55-63.

Mam;scnrsr R.EcErvED Mancs 23, 1986MeNuscnrpr ACCEFTED NovervrBen 7, 1986