THE AMERICAN MINERALOGIST, VOL 5I, SEPTEMBER OCTOBER. 1966

VIRGINIA NIETAI,IICT }IINERALS:PERRIERITE AND CHEVKINITE

Rrcnano S. MrrcuBl4 (Jniaersity oJ Virginia, Charlottesville, Vtrginia.

Ansrn,lcrPerrierite occurs several piaces in Virginia, especially in Amherst, Bedford, and Nelson

counties. chevkinite is found only at a previously unknown iocality in Nelson county. Allchevkinite reported earlier from Amherst, Bedford, and Nelson counties is actually per-rierite. rndexed r-ray powder data are given, which serve to distinguish between these verysimilar minerals. Approximate uniL ceil constants for perrierite are a:13.61, b:5.68,c:11.73 A, p:113.5 ' ; for chevkin i te, o:13.60, b:5.82, c:11.22 A, p: t00.75". Semi-quantitative spectrographic data show no appreciable chemical difierences between per-rierite and chevkinite. Both minerals contain thorium. Most of the Virginia specimens areonly slightly metamict. Heat treatment in air improves f-ray patterns although it producesa slight shrinkage in unit cell dimensions. Chevkinite begins to transform to perrierite above600o C., and CeOz forms from both minerals at higher temperatures. Weathering productsassociated with perrierite include anatase, churchite, cerianite, bastnaesite. as well asminerals very close to ancylite and florencite.

fNrnooucrroN

Recent s-ray studies of metamict minerals which occur in Virginiahave shown that some of the materials heretofore reported as allanite orchevkinite are actually perrierite. The purpose of this paper is to giver-ray powder data pertaining to the identif ication of perrierite and chev-kinite, and to report all the known Virginia occurrences of these min-erals. The numerous occurrences of allanite in Vireinia are to be treatedin a separate paper.

Perrierite was first reported from Nettuno, Roma, Italy by Bonatti andGottardi (1950), who later recognized its chemical and crvstallographicsimilarit ies to chevkinite (1953, 1954). Fleischer (1951) and Jaffe et al.(1956) expressed some doubt regarding the validity of perrierite being aseparate species. However, a further consideration of the problem byBonatti (1959), and the subsequent publication of the structure of themineral by Gottardi (1960), showed that perrierite is a separate mineral.Lima-de-Faria (1962) demonstrated that chevkinite, when heated in airat very high temperatures, transforms to the perrierite phase and sug-gested that perrierite may be an oxidized form of chevkinite.

Since its init ial discovery several occurrences have been recognized.Bonatti (1959) showed that a material from Kobe-mura, Kyoto prefec-ture, Japan, originally named chevkinite by Takubo and Nishimura(1953) is actually perrierite. Lima-de-Faria (1962) has identif ied perrier-ite from two localit ies in India, as well as from Mozambique and BedfordCounty, Virginia. According to Portnov (1964) a strontian perrierite

PERRIERITE AND CHT,VKINITE 1395

occurs with ordinary perrierite in the northern Baikal region of the SovietUnion.

X-ney D.cte lon Pn,nnmnrtB aNp CnpvrrNrrn

Although some similarities exist between x..'ay data" for perrierite andchevkinite, as well as alianite, the data allow one to distinguish betweenthe minerals without difficulty. Differences in the intensities of reflectionsare of special diagnostic value. That r-ray powder data can be used todistinguish between the minerals has also been pointed out bv Bonatti

Tlrr,n 1. X-ney Poworn Dere. lon UNrnuroo pnnnrrrrtEFRoM AMTTERST CouNrv, Vrncrrra

CuKa Radiation. Cameras oI 11.46 cm diameter

hkl d(calc.) d(meas ) I(obs ) hklA A

d(calc.) d(meas.) I (obs )

001201200oo22021 1 01 1 120r1 1 1rt22030033 1 1r t 2 , 2 0 2401,3123 1 01021 1 34004033 1 3

1 0 7 66 6 86 2 45 . 3 85 2 4

4 9 94 6 54 . 3 94 0 83 8 43 5 q3 . 5 43 4 53 3 83 . 3 63 3 43 . 2 03 1 23 0 32 9 7

10.92

5 . 4 1

5 1 95 0 1

4 4 04 1 03 8 53 . 5 6

3 . 4 7

3 3 13 2 23 1 43 0 52 9 9

3 1 12040200211 1 340100420340+221220r143 1 4o22, 3r2222

2 9 52 9 32 8 42 7 52 7 12 7 22 6 92 6 82 6 22 . 6 12 5 82 . 5 72 5 32 5 12 5 0

mw

m

2 . 9 6

2 . 8 4

2 7 4

2 6 9

2 6 3

m -

2 5 6 v w

Z J T W

2 . 4 7 v v w2 28 vvw2 . 2 4 v w2-18 mz t t v v w

2 O i v w1 . 9 5 m

mw

s

and Gottardi (1953, 1954), Neumann et at. (1957) and Lima-de_Faria(1962). rn the light of recent work the suggestion by Jafie et al. (19s6)that the differences between patterns are only due to a great range ofchemical substitutions within chevkinite alone, no longer seems tenable.

Even though much of the virginia perrierite is slightly metamict andmust be heated to give the best x-ray data., most samples give diagnosticx-ray patterns before heat treatment. The measured interplanar spacingsfor the mineral, given in Table 1, represent the average of values obtainedfrom four sharp r-ray powder fi ims (11.46 cm diameter camera, cuKoradiation) made from unheated materiai from the Burley farm, Amherstcounty. using unit cell data and partially indexed powder data reported

1396 RICHARD S. MITCIIDLL

by Bonatti (1959) the following approximate unit cell values were deter-

mined for the Vi rg in ia mater ia l : a :13.61,6:5.68, c : l l '73 A' , a :b:c: 2.39{l:2.06s, g: 113.5". Interplanar spacings calculated from these

data reasonably match the observed data, and permit the observed values

to be indexed. In Table 1, all calculated spacings-allowed by space group

C2/m (Gottardi, 1960) are l isted for dr,r.r)2.50 A. All spacings down to

1.95 A were calculated, but so many different values match each of the

measured reflections in this range that inclusion of these data seems

pointless. The Virginia data compare well with values published by

Bonatti (1959) and Lima-de-Faria (1962) and also match diffraction

graphical data presented by Neumann et al. (1957).

Films made of heated metamict samples, except for some differences in

sharpness of lines, are nearly identical, although careful measurements

show a slight shrinkage of the unit cell size with increa,qed temperature.

Rarely the CeOr l ines 3.12 ( l l l ) ,2 .7I (200) , and 1 '91 A (220) appear in

x,-ray patterns of perrierite heated in air. These were noted in two sam-

ples from the Burley farm, Amherst County, one heated half an hour at

800" C. and another heated one hour at 10000 C. Other samples from the

same locality treated in the same way showed no CeOr.

Careful studies of numerous powder films of Virginia perrierite showed

occasional weak allanite impurity l ine-s, especially at 9.30 (001)' 8'09

(100), 4.65 (Tn), 3.24 (201) and 2.62 L (311)' There apparentlv is no

relationship between these lines and the heat treatment of the sample

since they also appear on unheated samples. The fact that obvious allan-

ite does not occur at the Virginia perrierite localities makes the appear-

ance of these lines problematical.The measured interplanar spacings for unheated chevkinite, given in

Table 2, are the average of values obtained from four r-ray powder films

(11.46 cm diameter camera, cuK" radiation) of material from near

Nlartin's Store, Nelson County. Using *-ray data reportedby Jaffe et al.

(1956) for a comparison, it was determined that the Virginia chevkinite

has app rox ima te va lues o f o :13 .60 , b :5 .82 , c : t L .22 A , a "b : c :2 ' 3321

l:1.92", P:100.75'. Since the observed data reasonably match inter-

planar spacings calculated from these values, the data can be indexed.

AII calculated spacings allowed by the chevkinite space group C2f m are

Iisted in Tabte 2 f or dnrr ) 2.8 L. Values down to 1.99 A were calculated,

but these data are not included because so many different values match

the measured reflections in this range. The observed data reported here

compare well with data for untreated chevkinite reported from various

localities by Kauffman and Jaffe (1946), Jatre et al. (1956), and Lima-de-

Falia (1962). There are, however, minor differences in the indexing of

lines compared with results reported earlier by Jaffe et al. (1956).

P ERRI ERIT E AN D CII EV KI N IT E

Samples of chevkinite were heated for one hour each in air at 200o C.and 400o C., and at 1000 intervals from 6000 C. through 1000o C. X-raypatterns of samples heated through 6000 C. were essentially the same asthose for untrea,ted materials, except for a slight shrinkage of the unitcell, in agreement with results obtained by Jaffe et al. (1956) . The 700o C.and 8000 C. samples showed minor perrierite lines with the chevkinite.At 900o C. perrierite predominated over chevkinite. The 10000 C. samplecontained distinct lines for CeOz along with perrierite and a few chevkin-

T.ter-n 2. X-nlv Powtnn D,nra rox UNTREATED CnrvrrNrrnlnou ManrrN's Sronn, Nrr.soN CouNrv, VrnorNre,

CuKa Radiation. Cameras of 11.46 cm diameter

1397

I(obs )d(calc ) d(meas )I(obs.)L r . r d l c a l c . ) d ( m e a s . )

A A

00120020roo21 1 02011 1 12021t1l t 2202t t 20033 1 03 1 12034014003 1 13121 1 3402401

t1 026 6 86 2 5

5 . 2 9

+ . 7 04 . 6 73 . 9 83 . 9 13 . 7 03 . 6 7

3 5 23 5 13 . 3 83 3 43 2 33 2 03 1 43 1 33 . 0 4

4 9 3

4 6 7

2031 1 3o20021312313, 00440320422040222r022221rt4222a l r

5 1 0

2 . 9 92 . 9 32 . 9 12 8 r2 . 7 92 . 7 6

2 . 7 32 6 72 6 52 6 42 . 5 72 5 52 5 32 4 72 4 62 4 3

2 . 9 4 m2 9 O v w

2 . 7 9 w w

2 -74 ms

2 . 6 + w +2 5 8 w w

2.53 vvw

2 . 4 + w w2.35 vvw2 1 8 w2 . O 8 w w1 9 9 v w1 . 9 0 v w1 . 6 5 w

3 5 5

3 . 3 43 2 23 . 1 8

3 . 1 13 . 0 3

m

i te l ines. These results generaliy are in agreement with Lima-de-Faria(1962) who observed that chevkinite becomes perrierite when heated inair, and that CeOz also is often formed.

Occasionally allanite impurity lines were observed in chevkinite dif-fraction patterns, especially values near 9.30 (001) and 8.09 A (100).Because of some similarities between strong allanite reflections and cer-tain chevkinite reflections it is difficult to differentiate between some ofthem when both are present on one r-ray fiIm, d.g., chevkinite 4.67 (III),3 .55 (310) , 2.94 ( t l3) and 2.74 A 14m1, and a i lan i te 4.65 (111) , 3.S2(2ll),2.95 (SOt) and 2.74 h (021). The presence of the allanite is some-

1398 RICIIARD S MITCHELL

what problematic since allanite specimens were not found in the deposit

studied. Allanite l ines were found primaril-v in unheated samples.

CuBurcer- Dar,q ror< PBnnrpnrrp ,qNl CgsvrlwttB

Speciaily selected pure samples from most of the perrierite and chev-

kinite localit ies discussed in this paper were subjected to semiquantita-

tive spectrographic analyses. Considering the iimits of accuracy in the

determination of larger percentages with this method one rvill see a close

similarity between these data and quantitative data assembled by Jaffeet at. (1956) for these minerals from other localit ies. A study of Table 3

Teel-e 3. Snureue,Nrrrlrrvr Spncrnocnapnrc Dere ox Prnnrrnrtr

aNn CnnvrtNrrE rRoM VrtcrNre

Elements reported as oxides Analyst: F. W. Barley,

American Spectrographic Laboratories, San Francisco

Perrierite Chevkini te

MgAICaScTi

MnFe1\ l

CuY

Nb

SnB aLaCePrNdSm

Pb

0 2 52 .4 5

05t 0

0 11

8002

2 51 52

.008

.0045

1 0 .5

443

. 0 3

.007

0 34 5

08

00808

t0005

. 0 0 132

. 2 5

0 1 5.003

7 51 0

6 .. 5. 3. 304

. 0 5

0 53 53

031 2 5

005l . )

0 42 54 .

. o 71 2 5

.008! 2

1 0003

. 2 52

0 1005

5 .1 015 .

. 5

. 3

. 203

. 0 1

0 . 42 53

. 0 51 2 5

o2l . )

1 0 ..002.002. 3. l . t

_ l 5

. 0 2

.0015

1 0 .. 5

. 43

. 1 206025

0 325 .

. 0 5

0 1. 2

T 2

002. 504

r . 2 5. 0 3. o l007

t t <

f . . )

52

. 208

. 0 8

.o7

GdDyYb

v2997 d v3049d v2984d vJ039d

o 73 . 55

I1 0

0031 5

1 0

001. 5- l t

. 7

06001

1 01 0 .2 57 . 51 .

. 3

. 30605

v2996d v3022d v2994d

5 .8

. 54 ,

44

. 206

. 0 4

S, Cd, Th Not determinable because of interference.

Sifnondetectables: balance in all seven samples.

V2997,1, Amherst County, Burley farm.V3049d. Amherst County, Wares GaP.

V2984d. Bedford County, north of Chamblissburg

V3039d. Bedford CountY (?).

V2996d. Nelson County, east of RoselandV3O22d Nelson CountY, on Hat Creek.

V2994d Nelson County, southeast of Martin's Store.

PERRIERITE AND CIIDVKINITE 1399

shows no outstanding diflerences among the Virginia perrierites exceptfor the fact that the Bedford County specimen (V29S4d) tends to besomewhat richer in some of the rare earths than the other specimens.Differences between perrierite and chevkinite are also nearly negligible,except for slightly larger amounts of Ti, Fe and Nb in the chevkinite. Thechemical similarit ies between the two minerals have already been dis-cussed by Bonatti (1959). Although Th was not detected in the semi-quantitative analyses, because of interference, a detailed study of theradioactivity of the specimens indicated its presence (T. Will iamson,pers. comm., 1965). Eakins (1891) also reported thorium in his analyses ofspecimens (in this study shown to be perrierite) from an unspecifiedlocality in Bedford County.

Occunnnwcns or Vrncrnre PBnnrnnrrB

It is unfortunate that only one of the occurrences of perrierite pre-sented in this discussion, the Burley farm in Amherst County, is an insilzz deposit. The other materials described either were found as floatmasses, or are museum specimens whose exact localities are now vague orIost. Only one of the Virginia perrierites has been mentioned in the l itera-ture previously. Lima-de-Faria (1962) identified perrierite in a museumspecimen from an unspecified iocality in Bedford County. Until now threeof the localities included here have been considered chevkinite occur-rences in the l iterature. These are the Burley farm, Amherst County(Pegau, 1932), Bedford County (Eakins, 1891), and Hat Creek, NelsonCounty (Price, 1888).

Amherst County, Bwrley farm, northwest o.f Amherst. Numerous specimensof perrierite have been collected from deeply weathered pegmatites onthe Burley farm about 5 miles northwest of Amherst. The property isless than a mile south of U. S. Highway 60 on State Road 715. Severalshallow pits, which were dug in the weathered deposits years ago, aresti l l visible near the place where the meadow joins the woods on the highhil l across the road and southeast of the Burley home. Pegau (1932)briefly described the deposit. He believed the material to be allanite withchevkinite intergrowths. He reported large irregular masses, someweighing 500 pounds, in pockets in a hypersthene granodiorite countryrock.

The writer has found several primary minerals associated with per-rierite in the weathered deposit. Several reddish-gray subhedral zirconmasses, up to 2] inches across, were found. These exhibit a distinctorange-yellorv fluorescence with long-wave ultraviolet radiation. I lmenitemasses, an inch or more across, often showing good parting, are common.

14OO RICITARD S. MITCHDLL

Numerous large pieces of magnetite with excellent parting were found

intimately associated with a weathered material resembling mica schist.

Fractured gray quartz and pieces of kaolinized feldspar are common in

the pits. No specimens of rutile, chevkinite, or allanite, reported by

Pegau (1932) were found by the writer. A specimen in the Lewis Brooks

Museum collection (V2997d) labelled by Pegau as "allanite with bands

of tscheffkinite" was shown in this study to consist only of perrierite.

Slight traces of allanite impurity lines were found, however, in r-ray

data of some perrierite samples from this deposit.

The perrierite is massive, black, and has a dull to pitchy luster. Com-

monly narrow pitch-black bands are found in the dull black portions.

Most of the perrierite collected by the writer is externally altered to a

light brown earthy material. Numerous pseudomorphic masses, with

only traces of perrierite at their centers, have been observed. X-ray pow-

der patterns of this earthy material show that anatase is the chief con-

stituent. Preliminary studies of white and pink earthy to porcelaneous

materials filling fractures in these masses have yielded r,-ray data for

churchite, cerianite, and a member of the plumbogummite group which

is very close to florencite.Most r-ray powder patterns made of unheated perrierite were very

sharp and distinct. Occasionally, however, the effects of metamictization

were noted on some samples. Separate samples from one piece (V2987d)

were heated in air for one hour each at temperature intervals of 1000 C.

from 2000 C. to 1000o C. The noticeable changes include sharpening of

lines, slight decreases in interplanar spacings, and the formation of CeOz

in the 1000o C. sample. A study of the pitchy and dull bands in the Pegau

specimen mentioned above showed no appreciable differences in r-ray

data between the materials of different luster, except for the fact that

CeOz lines appeared when the dull material was heated at 800" C. for

half an hour.

Amherst Counly, at Wares Gap. Mt. B. Smith Campbell of Route 4,

Amherst, Virginia, reported (pers. comm., 1963) that he found over 320

pounds of an allanite-like mineral as float pieces at Wares (Weirs) Gap in

Amherst County. The locality is about 3.25 miles southwest of the Burley

farm occurrence. After a dil igent search of the area the writer was unable

to find traces of the mineral, but he is fortunate to have a large specimen

of the material through the kindness of Mr. Campbell.

The specimen (V30a9d) available for study measures 16 by 9 by 8

inches, and is over 80 per cent perrierite. Anhedral to subhedral milky

and smoky qvartz is embedded in the perrierite, especially on one side of

the mass. Small amounts of feldspar are associated with the quartz.

PNRRIERITE AND CHEVKINITD, UNI

Numerous grayish-red zircon crystals, ranging up to 2.5 inches long andhalf an inch square, and a fragment of light yellow apatite half an inchIong, are associated with the perrierite. The specimen obviously ispegmatitic. The perrierite is black and vitreous with some large dullareas. Although the mass is essentially fresh, in places the perrierite isaltered to a Iight moderate brown earthy substance. X-ray studies ofthis showed anatase with traces of quartz. Fractures in the fresh perrieriteare incrusted with dull pale yellowish orange to grayish orange pinkmaterial which is a mixture of bastnaesite and a mineral with r-ray datavery close to ancylite. Weak perrierite x-ray patterns were producedfrom unheated samples. Heating at 8000 C. for half an hour in air in-creased the quality of the patterns.

Bed,ford, County, north oJ Chamblissburg. Two specimens in a collection ofallanite in the Lewis Brooks Museum, assembled by Prof. T. L. Watsonin about 1916, were shown by x-ray study to be perrierite. One specimen(V2984d) was labelled "Powell farm, High Knob, Bedford County,Virginia, JluJy 21, t916." The other (V2981d) had "Stewartsville, Vir-ginia, Mrs. N. A. McMannaway, September, 1915." An allanite speci-men (V2985d) also had the Powell farm label. With the McMannawayspecimen were also pieces of black vitreous garnet and magnetite. Thewriter made a serious attempt to find the localities indicated by theselabels. He found that both the Powells and l\{cMannaways lived near thepresent W. E. Dooley farm which is off State Road 616 north of Chamb-lissburg. The writer did not find perrierite here, but did find two masses olfloat qrartz containing allanite near the edge of a meadow on a ridgelocated just west of the roadway leading to the Dooley house.

The two specimens of perrierite available for study are each less thanan inch square. Thev are dull to vitreous and black. Unheated samplesgave diffuse r-ray patterns of perrierite. Sharper patterns resulted whenthe materials were heated in air at 800" C. for half an hour.

A third specimen (V3039d) in Watson's allanite collection is labelled"Bedford County(?), Virginia." This specimen, which is a l itt le overaninch square, looks like the Chamblissburg material in every respect. Thepowder of the mineral is the same light brownish gray which is somewhatdifferent from the Amherst and Nelson County specimens. The r-raypowder patterns of both unheated and heated samples show perrieritewith a few very faint allanite lines.

Bed,Jord County. Eakins (1891) reported chevkinite from an unspecifiedJocality in Bedford County. Several more or less rounded masses, coatedwith a brownish-yellow alteration material, were found. The largest mass

I4O2 RICHARD S. MITCHDLL

weighed over 7.5 pounds. On fresh surfaces the mineral showed a bandedstructure of Iustrous black and dull black material, the bands varyingfrom mere lines to over one-eighth inch in thickness.

Five pieces of the Eakins (1891) material were obtained for this studyfrom the U. S. National Museum (USNM 83896, one piece; USNNI78512, four pieces). X-ray studies of unheated samples of each of theseshowed definite, but generally weak to very weak, perrierite patterns.The r-ray patterns were somewhat improved, but sti l l rather weak, foreach of the samples when heated half an hour at 6000 C.; heat treatmentat 800o C. for half an hour produced much improved perrierite patterns.Very weak allanite lines were observed in one of the unheated perrierites.An r-ray study of the alteration crust on two of the samples showed ana-tase to be the chief component.

Jafte et. aI. (1956) gave r-ray data for heat-treated material from Vir-ginia which they called chevkinite. Although the exact locality was notgiven, it is implied from the chemical analysis quoted by them, that thematerial was that of Eakins (1891), shown here to be perrierite. Theytreated their samples by heating them over a Bunsen burner for an hour.The fact that data for this material match data for definite chevkinitefrom other localities treated in a similar fashion is expected, now that weknow perrierite and chevkinite yield the same products when heated inair at high temperatures. The writer found no evidence to suggest thatboth perrierite and chevkinite occur among the Eakins (1891) materials,or in other samples from Bedford County.

Another specimen from Bedford County, in the Brit ish Museum (BNI86189), was shown by Lima-de-Faria (1962) to be perrierite. A small chipof the same specimen was studied by the writer. The mineral is somewhatmetamict, but unheated samples have very definite perrierite *-ray pat-terns. Samples heated in air at 600" C. for half an hour and at 800" C. foran hour gave excellent perrierite data. According to nlr. P. G. Embrey(pers. comm., 1965) of the Nluseum, this specimen was purchased in1903 from a James R. Gregory and Company. Although its earlier historyis not known, this may be from the Eakins (1891) locality.

The writer was also fortunate to obtain another Bedford County speci-men from the British n{useum (RNI 82694) The mineral was labelled"chevkinite" and was purchased in 1897 from a Mr. J. Graves, althoughit bears a G. L. English and Company label (P. G. Embrey, pers. comm.,1965). This specimen is very similar to the Eakins (1891) material andmav be from the same locality. The sample is one of the most metamictones observed in this study. An unheated piece showed only the mostintense r-ray l ine for perrierite. A sample heated in air aL 600o C. for

P I'RRI ERIT Ii A N D CI] EV KI N IT E

half an hour gave a faint, but pure perrierite pattern. When heated at800" C. for an hour the :r-ray pattern was typical for perrierite.

h-elson County, east of Roseland,. Through the kindness of the late N,Iiss

J. J. Glass, United States Geological Survey, the writer had available asmall specimen (V2996d) labelled "chevkinite" from 4 miles east ofRoseland. This isolated mass weighing 1| pounds when found was col-lected from near a stream by L. J. Kidd in 1940. It was believed to havecome from the nearby Blue Ridge.

The mineral is dull black with some black vitreous bands. Untreatedpowder yielded the perrierite *ray pattern, but the l ines were weak anddiffuse. Heat treatment in air at 8000 C. for half an hour gave sharp per-rierite patterns.

Nelson County, on Hat Creek. Price (1888) reported chevkinite frorn HatCreek, 4 miles east of Massie's Ntil ls. The mineral occurred as a loosemass weighing over 20 pounds. It appeared to have been broken lrom aveinlike deposit about 2 inches thick. The original piece was fresh exceptfor a brown crust attributed by Price (1888) to contamination of adja-cent rock.

A small specimen obtained from the U. S. National l{useum (USNNI47569: V3022d) is dull black with a few thin black seams of vitrer.rusIuster, l ike that described bv Price (1888). Unheated samples gave pat-terns for perrierite which were usually very sharp and distinct. Samplesheated at 8000 C. for half an hour gave similar results. X-ray fi lms ofheated and unheated material frecuentlv also showed the presence ofweak allanite l ines.

OccunnBNcn on VrncrNra CnovrrNrro

Although chevkinite has been considered to occur in Virginia for morethan 75 years, since its description by Price (1888), the present workshows these earlier materials are actually perrierite. These invalidatedchevkinite localit ies are the Burley farm, Amherst County (Pegau, 1932),Bedford County (Eakins, 1891), and Hat Creek, Nelson County (Price,1888). The only chevkinite encountered by the writer is from a previouslyunreported in situ deposit near X{artin's Store, Nelson County.

Nelson County, southeast of Martin's Store. A. chevkinite-bearing pegma-tite occurs in a road cut on the northeastern side of State Highway 6 at aplace about one mile southeast of Martin's Store (or four miles northwestof the intersection of U. S. Highway 29 and State Highway 6). The ex-

1403

r404 RTCHARD S. MTTCIIELL

posure is in a cut on a former curve of the road before the road wasstraightened out, and is across from where the Rockfish River now flowsnext to State Highway 6. The dike, which is in a biotite gneiss, is over twofeet wide and is exposed vertically in the cut for at least 15 feet. It con-sists mainly of medium- to coarse-grained blue-gray qtartz and whitefeldspar. The core of the dike consists of fine-grained biotite masses fillingspaces between subhedral feldspar crystals. Associated with the fi.ne-grained biotite are tiny pink garnet crystals and a few anhedral to sub-hedral chevkinite masses not more than three inches across. The depositshows no evidence of weathering, and it has not been mined.

The masses of chevkinite are intimately associated with biotite andqtrartz. The mineral (V2994d) is black, locally with a slight brownishtint, and is both dull and vitreous. A bright yellow amorphous substanceoccurs as irregular inclusions in some of the pieces. The chevkinite showsno evidence of weathering. X-ray studies show it to be scarcely meta-mict. although it does contain Th. A discussion of how the mineral fromthis locality behaves when heat-treated was given earlier in this paper.

AcrNowr-BoGMENTS

The writer wishes to express his thanks to individuals who suppliedspecimens for study, especially to Mr. P. G. Embrey of the Brit ishMuseum (Natural History), Mr. P. E. Desautels of the U. S. NationalMuseum, and the late Miss J. J. Glass, of the U. S. Geological Survey.Dr. T. G. Williamson, of the University of Virginia Reactor Facility,kindly studied the radioactivity of the specimens. Mr. R. J. Bland, Jr.,of Richmond, Virginia, assisted the writer in collecting materials in thefield. This research was supported in part by a grant from the SigmaXi-RESA Research Fund.

Rrlrruxcns

Boxrrtr, S. (1959) Chevkinite, perrierite and epidotes. Am. Mineral,.44, 115-137.- AND G. Gorre,ror (1950) Perrierite, nuovo minerale ritrovato nella sabbia di Net-

tuno (Roma). Rm.iL Acc. Naz. Lincei Ser. 8,9,361-368.- AND G. Gorrenor (1953) Nuovi dati sulla perrierite. Relazioni tra perrierite,

chevkinite ed epid,oti. Rend. Soc. Mineral. Ital, 9,242-243.G. Gorr,rpor (1954) Nuovi dati sulla perrierite. Relazioni tra perrierite, chev-

kinite ed epidoti. Renil. Soc. Mineral. Ital. lO,2OS-225.Errrrs, L. G (1891) New analyses of astrophyllite and tschefikinite. Am. Jour. Sci (3) 42,

34-38 ; same paper (1892) U .S. Geol, Sun. BuJl. 90, 4l-44.Fr.nrscnt, M. (1951) New mineral names-perrierite. Am. Mi,nera.l,.36r926

Gorranor, G (1960) The crystal structure of perrierite. Am. Minerdl. 45, l-14.

Jerrn, H. W , H. T. EvlNs, Jr. .qr.ro R. W. Cnapuen (1956) Occurrenceand ageof chev-kinite from the l)evil's Slide favalite-ouartz svenite near Stark, New Hampshire. ,42r.

Mineral. 41. 474-487 .

PERRI ERITE AN D CIIEV KI NITE 1405

Kaurru.nN, A. J., Jr. ,lru H. W. Jrrru (1946) Chevkinite (tschefikinite) from Arizona.

Am. MineraL. 31, 582-588.Lrue-oo-Fanre, I. 0962) Heat treatment of chevkinite and perrierite. Mineral. Mag- 33,

42-47.Nrulte.xN, H., T. Svnrurnur eNo P. C. Sannd (1957 )X-ray powder patterns for mineral

identification. III Silicates. Ailonitl'. Norshe Viilenskaps-Akod'. Oslo. I' Mat.-Notunt.

K t . 6 .Pncau,A. A.(1932) Pegmatitedepositsof Virginia. VirginioGeol.Surtt.Bttll '.33.Ponrrov, A. M. (1964) Strontian perrierite in the northern Baikal region (in Russian).

DohI. Aka.d..lfaaft S.S.S.R. 156, 579-581.Pnrcn, R. C. (1888) Analysis of tscheffkinite from Nelson County, Virginia. Am. Chem-

Jour. 10,38-39.Taruno, J. aro S. Nrsnruuna. (1953) on tscheffkinite from Kobe-mura, Kyoto Prefecture,

Japan. Me.m. Col,l. Sci., Unia. Kyoto Ser.8,2O,323-328.

Manu,scriptrecehted. January 25,1966; accepteil,Jor publi'cation, February 10, 1966.