AND ANDORITE(?) (2Pbs'agzs'3sbzss)FROM OREGON*

CRYSTALLOGRAPHY OF VALENTINITE (Sb,O3) AND ANDORITE(?) (2PbS'AgzS'3SbzSs)FROM OREGON*

Walonltan T. Scnnr,r,nn, U.S. GeologicalSuraey, Washington, D. C.

CoNrnwrs Introduction. 651 Valentinite. 652 Axial Ratio. 652 Forms and Angles. 653 Discussion of Prism Zone 657 Cornbinations. 66r Andorite(?). 662 INTRODUCTION A set of ore specimensfrom the Ochoco mining district, Oregon, sub- mitted for mineralogical study by G. F. Loughlin of the U. S. Geological Survey, showed a number of unusual minerals. Their occurrenceis men- tioned in a recent report by Gilluly, Reed, and Park (1933). The locality is Gold Hill (Howard district), Ochoco Creek area, northeastern part of Crook County, central Oregon. The matrix of the specimenssubmitted was largely carbonatel calcite, manganian dolomite with <,r index : 1.690-1.693,and calcianrhodochrosite with a:1.783. One specimencontained a few hair-like metallic crystals, which break when bent, probably due to a basal cleavage. They are provisionally assigned to jamesonite and the following measurements were obtained on one of the thicker hair-like crystals.

Ta.rr,a 1. MresunBunxrs or JlltnsoNrrn

Measured ]amesonite

(110):(T10):101"51' l0r"2l' (1T0):(IT0): qs :: I0l 22 (110):(120): 19 26 t9 r7 (110): (00t) cleavage:90"-92' 90 00

A few of the hair-like crystals, however, were very flexible and could easily be bent double, showing the presenceof a second hair-like mineral with no basal cleavage.The material seemsto correspond to the mixture of "feather ore" previously called warrenite, which, however, is a mixture of jamesonite and zinckenite, as previously describedby Schaller (1911).

x Published by permission of the Director, U. S. Geological Survey'

651 652 THE A MERICAN M I NERA LOGIST

A single prismatic crystal of another mineral, provisionally referred to andorite and describedin the following paper, was found on one specimen. Other minerals present are: aragonite, pyrite, tetrahedrite, and two other metallic minerals, one lead-gray and another prismatic dark reddish-black, which could not be identified on account of the naucitv of material. Apparently this locality might yield several of the rarer silver sulpho- salts and anyone having accessto the locality should make a careful searchfor well crystallized minerals of this character. A number of small non-metallic crystals in one of the cavities of a specimen proved to be valentinite, antimony oxide, Sb2O3,and their goniometric measurements showed a large number of crystal forms, many of them new. They are described on the following pages.

VALENTINITE The optical properties and chemical reactions of these crystals are typically those of valentinite. The optical axial planes for red and blue light are normal to each other and the negative axial angle is small, with strong dispersion. The doubly terminated crystals are small, about a millimeter in length, and have no attraction as specimens.They would probably pass unnoticed unless one was making a detailed examination of the mineralogy of the material.

Axral Rarro

The a-axis is calculated from the average measured { angles of the prisms of the seven measuredcrystals. The measurementsfor Bl340l and C{450} are omitted as they show much greater divergencefrom the calculated angles than do those of the other prisms. Likewise, those prisms whose s angle is very large (greater than 81o) are also omitted as they are too close to {100} to yield ratios of comparative value. The c-axis is calculated from the average measurementsof the p angle of the new sidedome Z{0.11.8}.The axial ratio so obtainedis a:b:c:0.393O: 1:0.4350. This ratio is close to those recently given by other authors, as follows:

a:b:c Spencer (1907): 0.3938:1:0.4344 Ungemachl (1912): 0.3928:7:0.4333 Cesdro(1925): 0.3952:l:0.4342 Schaller: 0. 3930:1 :0 .4350

1 ungemach gave the axial ratio as 0.3936: 1 : 0.4339. According to a personal communica- tion from Prof. Palache (June 9, 1936), cesdro, in a note, calls attention to an error in these values The correct ratio is: 0.3928:1:04333. JOURNAL MINERALOGICAL SOCIETY OF AMERICA 653

The average of these four values gives the axial ratio of valentinite as aib:c:0.3939: 1 :0.4339.

Fonus exo ANcrBs A total of 22 forms were found on the 7 crystals measured. These in- clude the two pinacoids o{100} and b{010}, nineteen prisms of which 15 are new for valentinite, and the new side dome f 10.11.81.

Frc. 1. Idealized drawing of valentinite from the ochoco district, crook county, Oregon.The crystal is about a millimeter high' Forms: 6[010], rz{110}, I{0'11'8}' The narrow striated face B adjoining b[010], shows the position of,4[350]' 8[340]' and C[450]. The broad striated area in front includes the prisms from D[10.9.0] to S[910].

The general habit and distribution of forms on these crystals ale shown in Figure 1. The dominant forms are m{ll}l and the new side dome ?{0.11.8} which is the only terminal form present. The prisms other than ml ll0l are all line f aces.The side pinacoid 6 { 010} is generally narrow but wider than a'line face whereaso{100} is a line face. The facesof m arevertically striated and thoseof ?{0.11.8} are horizontally striated parallel to their intersection. The reflections from the unit prisms weregood; all the otherswere poor. 654 TEE AMEMCAN MINERAINGIST

Terr-o 2. ValoNrnvrrr': Fonus ,tNn ANcr,ps New forms are starred

Average No. I No. Calculated measured Difierence crystals I faces @ angle { angle

6[010] 7 10 0009' 0'00' 0'09 o{100} z 2 89 42 90 l6 x,4 00 [350] 3 3 56 48 43 4Bls4Dl 56 05 2 62 44 62 18 *c{450} 26 J 64 07 63 47 20 m1110l 7 28 68 *D 39 68 30 09 [10.S.0] 5 8 /u .to +E1760l 7029 07 I 7 71 26 7t 2l 05 d{s40} J 7 72 33 *F[430] 723t 02 9 /J JJ /J JJ 02 "G[ 10.7.0] 4 6 7432 *H 7435 03 {14.9.01 6 8 7548 48 *,r 75 00 [ 530] + 7645 7642 *1

o The p angle.

Tasre 3. Verexrrurrn: Maesunnarnxrsor d ANGLEor mlll\l

Average of fair and Limits good reflections

68'08'-69"25l 68"23', 68 32 -68 41 68 36 68 43 -69 03 68 46 68 38 -69 30 68 55 68 26 -68 46 68 37 68 34 -68 46 68 40 68 01 -68 38 68 37

Average:68'39' JOT]RNAL MINER-ALOGICAL SOCIETY OF AMERICA 655

Tasln 4. VelnNrrnrrr: Mresunnunxrs or d ANGLEol OrHERPRTsMS

.4{3s0} B$4al

Calculated 56"43', 62"18', Average of measurements 5648 62 44 2- 56 13 5- 62 45 +- 573r 6- 62 37 5- 56 39 Individual measurements

ols40] F{430} G110.7.01El14.e.ol "/{s30}

Calculated 72031', I.) JJ 74" 35', 75"48', 7642', Average of measurements lz.tJ 7335 7432 7548 7645 (t- 7227 r- 7329 2- 7425 t- 7609 2- 762r t; 7230 2- 7350 2- 7445 2- 7528 2- 765r lL- t?- 7230 2- 7337 2- 7423 2- 7536 3- 7639 72 20 2- 735r 3- 7442 +- 755r J- tl rl Individual 3- 73 26 5- /5 5v measurements 7257 5- 7427 6- 7635 3- 73 26 6- 75 48 5- 73 23 72 18 6- 74 3r 6- 76 rO 5- 7326 72 52 6-- 735r 7- 75 20

pl2r0l Ll73ol n[310]

Calculated 77"40', 78"52', 8,0"25', 81'36', 82"31', Average of 7748 7854 t0 24 814I 82 42 measurements 2- 7836 3- 80 06 1- 81 36 r- 82 29 2- 7853 4- 80 42 2- 8l 49 4- 82 07 Individual s- 79 04 4- t0 42 4- 81 39 6- 82 55 measuremenf,s +- 79 10 5- 80 07 6- 83 17 5- 78 24 s:7s rB 656 TEE AMEMCAN MINERALOGIST

Calculated 85'30' 87'30', Average of measurements 8547 8735 2- 8559 +- 87 39 Individual 4- 85 58 5- 87 24 measurements 5- 85 24 7- 87 43

Teer,n 5. Var,rNrrltrre: Mresunrutxrs or ?10.11.81

Difierence from calculated Measured p angle

10.11.81

0"56', 31049/ 0'59', 32

JI o42 30 42 07 39 011 30 53 04 50 034 3l 12 23 109 034 3109 20 106 052 3101 72 58 023 3023 26 20 010 3051 n') 48 010 3052 03 49 039 30 15 34 t2 039 3036 t.) 33

Average (omitting crystal no. 2) : 30"53'. Calculated:30 49 .

The indices {0.11.8} seem to be the simplest which can be given to this new side dome. The calculated p angles for other indices, close to {0.1 1.8 } , are as f ollows:

p {06s}:27'30' l0a3|:3ss3 {0.11.8}:s0 4e {075}:3117 {032} :33 sa Average of measured angles:30"53' JOT.]RNALMINER, LOGICALSOCIETY OF AMEMCA 657

The closestcalculated angle for a simple form is that for {043} a form not listed for valentinite, but not a single one of the 13 measurements is as low as 30o03' though two measurements approach this value. As the last column in the preceding table shows, the differences between the measuredangles and 30"03'are considerable.The difierencesbetween the average measured angle for T and the nearest known forms, {065} and [ 032] , are also considerable. Differencefrom {065}:3'23' Difierencefuom {032]r:2 ll A discussionof the side dome seriesof forms from (001) to (010) does not help in deciding the correct indices for ?. Taking the zone segment from (011) to (021),in the middle of which ? lies,and reducingit to the general type 0 @, ? becomesf instead of the normal ]. The form {043} would yield } but as just shown the measurementsdo not agree with these indices.

7 o T X 011 098 065 11.8 o32 053 02r 96 ll 35 I 85 8 ZJ

V-It 11 3 _ 0 =, t2 Dz-1) t+ 7 -'l /1\- / -l1\ 3 l l | .ly'z: 0 l -t l, I +l \l/ \4,/

DrscussroNol Pnrsu ZouB The following discussion of the prism zone follows the well known method of Goldschmidt's comparison with normal series. It not only served to confirm the correctnessof the indices of many of the new prisms but it has been of greatest value in deciding the correct indices of three of the new prisms with more complex indices. The poor reflections obtained from most of these new prisms with a consequent considerablevariation in the measured angles do not suffice, alone, to correctly determine the indices. Including the two pinacoids D{010} and o{100} there ate now 26 forms in this zone, including the two given by Cesbro(1925)' This zone consistsnaturally oI two sectionsextending from 6{010} to mlll}l and Irom m,l110| to o[100]. In the first sectionfrom 6 to m,the $ anglesof the forms difier from adjacent ones by many degreesexcept for the two forms B{340} and C{450}, where the differenceis only lo29'. B:utfor the second section from m to o, the forms are closely crowded together and are about one degree apart. 658 THE AMEMCAN MINERAINGIST

The first sectionis:

b p nABC lf, 010 160 130 t20 350 340 450 110 h 1 I 1334 o I R 6 J 2s4s

I I h:0 ( ) 1--34 y'ra: . (+) I ,+3(4)€

The two forms p{160} and C{450} are extra. Splitting the zone at nlI20\: R r11 1334 6TV TT4s 1

t2 IJ ;;1 0-t- @ JJ o2-v 42

1) 1 1 ;2q o;23 I-l

,l 1 Nz: '72€ TT- oTt3 €

All the forms in this section can be transferred to parts of the normal series. The secondsection, from m{ 110} to o { 100} is a long one with 19 forms and runs continuously with no well defined break. The form with the simplest indices,except for the two end membersmlllTl and o{100} is pl210l which moreover lies in the middle of the zone. This zone is therefore split into two sections at pl2l}l, which are considered separately. The first section from m to p gives:

tnDE aFGHJ Rp 110 10.9.0 760 540 430 10.7.014.9.0 530 950 2rO

h 1 10754r07459 R 96437935

II113524 l- l: 96437935

711135 -:--24 1-r- E53244 JOURNAL MINEKALOGICAL SOCIETV OF AMEKICA

122 35 TT7 zn t22 35 y'fr: (8) 7 s "T T2 All the forms agree with the normal seriesntr4 except for (8), equivalent to 1( {950} , which is extra. If this zone section be f urther split at 1, equivalent to F { 430} , the two sections will agree with I[a.

122

4J.t T'

1) t2 IJ ts-l: 2 a', | -1) JJ

1) IJ 17- 3 (7) 33 T"T For the form D{10.9.0} the measurementsalone would hardly dif- ferentiatebetween {11.10.0}, {10.9.0}, and {980} as the followingcom- parison shows. 6 D, average measured 70'36', Calculated for { 11.10.0} 7018 {10.e.ol 7o2e {oso} 7042 Of the eight measurementsof D, two agreebetter with {11.10'0f, one better with {10.9.0}, four better with {980}, while one is half way be- is the tween {10.9.0}and {980}. But the discussionshows that {10.9'0} correct symbol. In the discussion {11.10.0} would yield $ and {980} would yield f in the normal seriesI[n instead of [, and {11.10.0} would yield f and {980} would yield f in the normal seriesof //g instead of *. For the complex indices of G, namely {10.7.0} the question might arise whether the simpler ones {750} might not be better. The comparison of measured and calculated angles shows: 6 G, averagemeasured 74"32' Calculatedfor {10.7.0} 74 35 Calculatedfor [750] 74 17

Of the six faces of G measured, the angles are closer to {750} for only two of the measurements. The discussion would yield f instead of $ the for /y'.r, and I instead of I for /y'a. The discussion therefore verifies 660 THE AMERICAN MINER,4LOGIST

measurementsthat the more complexindices {10.7.0}are to be preferred for the correct indices of G. The most complicated indices of all the new forms here described are thoseof 11{1a.9.0}.The discussionshows however that theseindices readily fall into the normal series. simpler indices which suggest them- selvesare {320}, [850], and {11.7.0}.The comparisonof measuredand calculated anglesis as follows. o H , average measured 75"48', Calculated for { 850 } 76r0 {11.7.0 } /o 5() {14.e.0} 7< L7 I32ol lJ tl

Of the eight measuredfaces of 11,one (7 5'2O,) agreesvery closely with the calculatedvalue for {320}ltwo agree very closely with the carcu- lated value for {850}, and two agreeclosely with the calculatedvarue for {11.7.0}.As all the reflectionsmeasured were poor, there is no justifica- tion for claiming that the measurements show four distinct forms instead of one. The only question is as to the correct indices of this form. And according to the measurements, the only question is between {14.9.0} and {11.7.0}. The measurementsalone are not sufficientto decide this question as the following comparison shows. The difierences (A) between the measurementsand the calculated anglesfor {14.9.0} and f 11.7.0|are as follows:

Crystal no. Measurements A { 14.9.0 J ^t11.7.01 | 76"09' +21' +13', 2 7528 _20 _28 2 7536 _r2 -20 4 755r +03 -05 s 75s9 +11 +03 6 7548 0 -08 6 7610 +22 +14 7 7520 _28 -36

There is practically no choice, the measurements agreeing about as well with one as with the other. The discussionhowever shows that {11.7.0} reduces to $ for /y'a. This value of $ does not belong in the normal seriesry'+ and moreover is more complicated than f,, derived from {14.9.0}, which does belong in /y'a.fn the normal seriesiy':, {11.7.0} becomesf which cloesnot be- IOURNAL MINERALOGICALSOCIETY OF AMERICA 661

Iong in .ly'rand is more complex than $ derived from { 14.9.0} , which does belong in /y'a. The discussion therefore shows that the correct indices f.ot H are {14.9.0}, a lact which cannot be determined from the measurements alone. Although the form 1({950} is extra in both Iy'r and ly'3,it causesno disturbance in the normal seriesand the measurementsof the five faces of K, three of which agree closely with the calculated value, fully sub- stantiate this form. The secondsection of the zone, from p{210} to o{ 100}, is as follows:

2r0 730 830 310 17.5.0 720 4r0 510 910 100

h 78 1a 7 3 h .)J 5 2

o 'u 743 tl 7 5 a 632 10 4 z 2

111 7 3 3 a-l: o.tz 10 + 2 z

l2 7 3 1 33 5 2

I2 3 17- 1 (7) 33 (+)2 This section of the zone gives nearly a perfect normal 1tr8series, with only one member (f) missingand two extras.The form {7201 of Cesiro, to which the letter R is assigned,fits in perfectly. The form {17'5'0} is undoubteclly vicinal to {720][, Cesbro giving a calculated difference of only 0o11'. The form S{910} is extra in the normal seriesbut is well substahtiated by the measurementsof three faces.

CounrwarroNs The combinations observed on the seven measured crystals of valentinite are shown in the followins table. 662 TEE A MEMCAN MINERALOGIST

Tesrr 6. Var,pxrrNrrp: CousrNerroNs Ossnrvno oN TEE SBvBNMrasr,-nro Cnisrar,s

Form

b{010 } b b b b b a{100} a & .4{3s0} A I B 13401 B B

c{4s0} L L rnllr0l n, m, ,n 1n NL D[10.e.0] D D D D D E1760l E E E E o Is40] 6 6 6 6 q Fl430l F D l7 G{10.7.0} G G G G Hl14.e.0l H H H "rls30l r J J 1(feso ] K K K pl2r0l u p p L1730l r Ml830l M r{310} T T 1rl410l 1r .lr P{s10} P P s{e10} s s T T T T "{0.11.8}

ANDORTTE(?) The single prismatic crystal referred to in the introduction to valen, tinite was about a millimeter long and of a dark gray to nearly black

Frc. 2. rdealized drawing of crystal of andorite(?). crystal about a millimeter high Forms: * [ 120],1. 12301, mlIt\l, 01 3201, 1 {03 1 1, 7 t02tl, F l012l. JOURNAL MINERALOGICAL SOCIETY OF AMEMCA 663 color. It was detached and measured on the goniometer with the results shown below. The crystal was Iost when it was attempted to apply micro- chemical tests. A striated prism zone, terminated essentially only by side domes, giving poor reflections, made it impossible to decide whether the sym- metry was orthorhombic or monoclinic. An extended comparison of the measured.angles with those listed in the literature supplemented by a comparison of the habit with those of possibleminerals, as given by Gold- schmidt's Atlas d,er Krystallformen, lead to the conclusion that the measured crystal was either andorite, 2PbS.AgzS.3SbsSr,or freieslebenite, 3Pbs.2Agrs.2Sbzss,with a considerablepreference for andorite. A total of 40 faces were measured, of which about half could be cor- related with 13 forms of andorite and with 15 forms of freieslebenite. The facesmeasured are listed below in the order of measurement,with the corresponding forms and angles for andorite and for freieslebenite. No attempt was made to interpret the measurements of these prism faces(nos. 7,g,17,23,26,and 34) for which no agreementcould be noted with the known forms of either andorite or freieslebenite. The axial ratios used.are, for andorite, eibic:0.6771:1:0.4458, and for freieslebenite,a:b:c:0.5371 :l:0.9277, 0:87o46' .

Tswn 7, Alroonnn eun Fnernsr-nBENTTE: GOUpARTSONOl MtaSUnBl Ar{cr-Bs

Note Measured Andorite Freieslebenite book Ietter Form Form

1 Al 44"06', 90'00'l | 230 M'34' 90"00' 350 45"39',90'00' 2 o 45 22 90 00 ) J b 66 34 90 00 320 6543 90 00 430 66 15 90 00 A 8522 90 00 810 8548 90 00 d 8724 90 00 100 90 00 9000 100 90 00 90 00 6 e 82 20 90 00 510 82 18 90 00 510 83 18 90 00

JI 68 12 90 00 8 66 04 90 00 320 65 43 90 00 430 66 15 90 00

9 J3 64 47 90 00 10 J1 63 07 90 00 430 6305 90 00 11 fs 61 39 90 00 110 59 36 90 00 5451 90 00 t2 J6 54t7 90 00 110 5554 90 00 560 13 45 12 90 00 230 4434 90 00 350 45 39 90 00 t4 36 54 90 00 r20 3627 90 00 90 00 15 Lr 35 29 90 00 250 34 17 16 36 36 90 00 120 3627 90 00 17 38 33 90 00 18 40 37 90 00 19 43 00 90 00 230 44 34 90 00 664 THE AMERICAN M I N ERA LOGIST

T.qsr,n 7. (Concluded.).

Note I Measured Andorite Freieslebenite No. book letter Form

20 1) 461l 90 00 350 45 39 90 00 2l u 5359 90 00 450 53 45 90 00 22 , 5603 90 00 110 560 .)+ .)r 90 00 23 s 5704 90 00 24 f .)d l/ 90 00 110 59 36 90 00 25 q 63 12 90 00 430 6305 90 00 26 ? 68 12 90 00 27 0 70 20 90 00 210 7t l8 90 00 28 n 66 33 90 00 320 65 43 90 00 29 'rn 64 03 90 00 430 6305 90 00 30 I 57 05 90 00 110 5554 90 00 JI h 54 22 90 00 54 51 90 00 32 j 4449 90 00 230 44 34 90 00 +5 39 90 00 33 't 36 32 90 00 120 3627 90 00 34 h 37 15 90 00 B 019 52 23 031 000 .)J IJ 032 136 54 18 36 D 019 52 27 031 000 032 136 54 18 J' L 019 47 07 052 000 48 06 u.)4 155 49 14 38 E 255 38 20 021 000 41 43 011 224 42 52 034 312 34 52 39 El 41 08 \ sz+" 745r 41 07 214b 7421 40 41 J 312 7717 45 2l 60 09 141" 20 16 62 15 I.)J" 19 54 62 22

a Form not known for andorite. b Form not known for freieslebenite.

Faces number 39 and 40 are extremely minutel they appear as mere bright specks on the goniometer but gave distinct though poor reflections. The measurementsagree fairly well with the calculated values for pyramids of fairly simple indices for both andorite and freieslebenite, though none of these pyramids are known forms for either mineral. Grouping the measurementsof faces of the known forms together, on the assumption that the crystal measured was andorjte, the results shown in the following table are obtained. The closest corresponding angle for freieslebeniteis added to the last column. JOURNAL MINERALOGICAL SOCIETY OF AMERICA 665

Taerr 8. Axnonrrr(?): Coupenrsox ol Mresunro 'q.rn CalcutetnD d ANGLES or KNoww Pnrsrus

Average of Calculated Calculated Measurements measutements andorite freieslebenite

'-37"15', lso"sz-38 ', kIr2ol 1?(?O\"" " 36 36"27' 12501:34"17 [36s4

(++oo -45 22 +o {3s0}:4s3e )u -46 14037 1l [4s12

(

nl21ol 70 20 70 20 7l 18

'/ [510] 82 20 82 20 82 18 {5t01:3313

A closer agreement of the measured prism angles with those of freieslebenite can be obtained by turning the measured crystal 90o around the c-axis, though then there is no agreement between the dome faces.

Tesln 9. ANoourn (?): Coltlenrsox ol Meesunro Pnrsu ANGlns, rl Cnvsr.lr- fs TunNro 90' Anouno lre c-Axrs

Form 90"-d Freieslebenite

kIr20l 53'03', [4501:53'45' u2301 45 28 {3s0}:4s3e mlrr0l l2s0l: s4 17 o1320l ,A )a '[140]:23 95 n1210l 19 40 {1s0}:184e l/[s1o ] 740

n Form not known for freieslebenite. 666 TEE AMERICAN M INERALOGIST

An idealized drawing of the crystal is shown in Figure 2. The faces of the unit prism ml 110] are drawn as distinct faces, whereas actually they are rounded and strongly striated vertically. The forms |-l20l , {230]1,{110}, {430}, and {320} are the common prisms, each occurring with at least three faces. The lower portions of both faces of y{031 } are striated. The crystal from the Ochocodistrict resemblesin generalhabit Figure 2 of andorite and Figure 17 of freieslebenite, as listed in Goldschmidt,s A tlas d,erKry stallformen. The absenceof any pinacoids (the measurement of face No. 5, with O:87"24' does not necessarily demonstrate the occurrence of {100}) favors the reference to freieslebenite, for andorite crystals commonly have pinacoids present, especially the side pinacoid ,{010}. Thus, in the 15 combinations listed by Koch (1928) D is present on every crystal of andorite as it likewise is in the five drawings listed in Goldschmidt's Atlas der Krystallformen. Faces of D[010] are frequently absent on crystals of f reieslebenite. The generally better agreement of the measured angles with those of andorite, however, would seem to favor the reference of the measured crystal to andorite. Additional and better material, though, needs to be examined before a definite identification can be made.

RnrunrNcBs Cosino,G. (1925):Acad.. roy. Belgique, Cl. Sci.,Mem. (fasc. 4),l-57. Kocn,S. (1928): Cbl. Mi.n., A,30. Gtr.luty,J. Reno,J. C.,and Pem, C. F., Jn. (1933):U. S. Geol.Szrra., Bull.84GA, 127-r30. ScnAr.mn,W. T. (1911):IJ. S. Geol,.Su.ra.,Bttll.4gDr27. SrnNcen,L. l. (1907):Min. Mag., 14,328-331. Uncruacr, H. (1912):BuJl. Soc.Jranc. M.i,n.,35, 539-552. IJrcnuecE's values have beenadopted by Gor,oscmuo,r(1923), Atlas d.er Krystallformen, 9, 45-Heidelberg.