THE AMERICAN MINERALOGIST, VOL. 50, OCTOBER, 1965

DIFFERENCES IN THE MINOR ELEMENT COMPOSITIONOF BERYL IN VARIOUS ENVIRONMENTS1

Monrruon H. Sr.llrz, Warrncn R. Gnrrrrrrs eNp paur R.BAnNorr, U. S. Geolog,ical Survey, Denver, Colorado.

AesrRA.cr

The range in abundance of certain elements is difierent in beryls from different environ-ments, and these difierences are so large that the environment of most beryls can be deter-mined from their composition. 33 minor elements were found in beryl by spectrographicanalysis, but only 13 made up as much as 0.1 per cent of any beryl. Beryl from pegmatitecontains relatively small amounts of Fe, Mg, Sc, Ti and v, and may contain relativelylarge amounts of cs, Rb and Li. Beryl from wallrock adjacent to pegmatite generally hasmore Ti, Sc, V, Cr, Mg, Fe and Ni than beryl from pegmatite, and less Rb. Beryl comesfrom several types of veins and hence the amounts of the various elements may vary widely,Generally, however, beryl from veins contains only small amounts of cs, Rb and Li. someof the greater amounts of Mg, V, K, Fe, Cr and Ti occur in beryl from veins. Beryl fromgranite generally contains only minor amounts of Cs, Li, Rb, Cr, Ni, y and Zn, and largeamounts of Fe. Beryl from rhyolite contains only small amounts of Li, Na, Mg and Sc,but has the largest amount of all the beryls analyzed of Rb, Fe, Ti,Zn,Zr, Mn, As, B,Ga, Nb, Sn, Y and Yb.

INrnonucrroN

The physical properties of beryl, especially its indices of refraction andspecific gravity, have been correlated with its chemical composition(Winchell, 1947; Dorfman, 1952; Norton et al. l9S8; Schaller et al.,1962).These data have been valuable in the study of individual pegmatites,where the indices of refraction have been used to determine the Beo con-tent of the beryl (Staatz and Trites, 1955; Sheridan, 1955; Thurston,1955;staatz, et al.1963). The physical properties and the arkali metaland Beo content of beryl commonly vary from zone to zone in the samepegmatite (Page et al., 1953;Staatz, et at.,1963).

Beryl has been found in many pegmatites throughout the UnitedStates, and most of that described comes from this environment. Beryr,however, also occurs in the wallrock adjacent to pegmatite, in granite, inveins, and in rhyolite. Prior to the end of World War II, beryl hadrarely been described in these other environments, although in the unitedStates it had been reported from rhyolite in the Thomas Range, Utah(Hil lebrand, 1905), and in veins on Mt. Antero, Colorado (Landes, 1935),at Boriana, Arizona (Hobbs, 1944), and at Irish Creek, Virginia (Kosch-mann et a1,.,1942).

The generally accepted chemical formula for beryl is 3BeO.AhOa.6SiOz. It is well known, however, that beryl varies in chemical composi-

1 Publication authorized by the Director, U. S. Geological Survey.

1783

1784 M. H. STAATZ, W. R. GRIFFITTS AND P. R. BARNETT

tion and that the alkali metal oxides Cs:O, LirO, Na2O, K2O, and RbzO

may total as much as 8 per cent of the mineral (Folinsbee, 1941; Winchell'

1947). As Be2+, Si4+ "nd

Alt* have ionic radii of 0.34,0.39 and 0.57 A, re-

spectively, and Na+, K+, Rb+ and Cs+ have ionic radii of 0.98, 1.33, 1.49,

and 1.65 A (Mason, 1952), respectively, it is doubtful that any of these

alkali metal ions can substitute for berf ilium, aluminum or silicon'

Lithium, which has an atomic radius of 0.78 A, may substitute for

aluminum. A study of the atomic structure of beryl suggests the probable

position of the other elements. This structure consists of a series of hexag-

onal rings formed by six silicon-ox-vgen tetrahedra (Bragg and West,

1926). These rings are connected by tetrahedrally surrounded beryllium

and octahedrally surrounded aluminum and have open channels down

the center of the rings. It is probable that most of the alkalies in beryl are

located in these channels (Folinsbee, 1941). As Cs, which has the very

large ionic radius of 1.65 A, is evidentlv found in these channels, there is

the possibility that a large number of other elements with atomic radii

intermediate between Cs and Al may be similarly situated. The presence

of other elements in the beryl structure is also dependent, however, on the

charge requirements and the availability of the element'

For this study beryl was analyzed spectrographically and 68 elements

were Iooked for. Thirty-three elements besides beryllium, aluminum, and

silicon were found above the limit of detection. Of these 33, however, only

13 made up as much as 0.1 per cent of any beryl.

Beryls in granitic pegmatites occur in an environment rich in alkali

metals and poor in iron, magnesium, titanium, nickel, chromium, and

vanadium. Beryls from veins, granites, wallrock adjacent to pegmatites,

and rhyolite occur in environments that are generally poorer in the

alkali metals and richer in other elements. Beryl was studied from all five

environments to see if the minor element content varied in the different

environments.

TvpB ol'MltBnrer-

A total of 47 beryl specimens was used in this study-19 came from

pegmatites, 3 from country rock adjacent to pegmatites, 15 from veins, 9

from granite and 1 from a rhyolite. In two pegmatites one specimen of

beryl came from the wall zone and one from the core and show the type of

variation one might expect within a single pegmatite. Bervls collected

from the granite not only include segregations of beryl within the granit'e

but also beryl in aplite dikes within the granite and in veinlets made up

almost entirely of beryl within the granite. In Table 1, beryls used in this

study are grouped according to the environment, and the locality of each

specimen is shown. Double numbers occur where two spectrographic

,I,IINOR I':,LEUN,NTS IN B]I,RYL

Tanlo 1 LocerroNs ol Bonvls Usen tN Tnrs Sruoy

1785

SampleEnvironmcnt number or

numbersLocal i ty Remarks

Pegmatite 9 Helen Beryl pegmatite, Custer County, S. Dak From wall zonc.Pegmatite 5, 843 Helen Beryl pegmatite, Custer County, S Dal< From CorePegmatite 8, 846 Bull Moose pegmatite, Custer County, S Dak, From wall zone.Pegmatite 7 Ross pegmatite, Custer County, S Dak From intermediate zone.Pegmatite B34 Scott rose quartz pegmatite, Custer County, S From core

Dak.Pegmatite 854 Dan Patch pegmatite, Pennington County, S

Dak.Pegmatite 85 Pegmatite from Avon district, Idaho From rvall zone.Pegmatite 87 Pegmatite on Errington-Theil property, Ruby From unzoned pegmatite.

Range, Nev.Pegmatite 84 Pegmatite in Crystal Mountain district, Colo. From small podPegmatite B.55 Apache pegmatite, Petaca district, N MexPegmatite 840 Harding pegmatite, near Dixon, N. Mex, From wall zone,Pegmatite B41 Ilarding pegmatite, near Dixon, N Mex From core?egmatite 821, 848 Millard-Chandlerpegmatite, Rockingham Coun- Green beryl from core.

ty, N. H.Pegmatite B2O, 847 Millarcl-Chandler pegmatite, Rocl<ingham Coun- Pink beryl from same crys -

ty, N FI tals as 848.Pegmatite B1 Edwards pegmatite, Spruce Pine, N C. From intermediate zone.Pegmatite R2 Edwards pegmatite, Spruce Pine, N. C. I.rom replacement unitPcgmatite 83 Ral' pegmatite, Spruce Pine, B C From unrcned pegmatite.Pegmatite R52 ?atterson pegmatite, Shelby district, N. CPegmatite B53 Pegmatite in Lincolnton quadrangle, N CCountry rock adja- 86, 850 Avon district, Idaho Near pegmatite yielding

cent to pegmatite Il5Country rock adja- 822,849 Near Harding pegmatite, near Dixon, N Mex Near pegmatite yielding

cent to pegmatite 840.Country rock adja- 6,1ts42 Sunnyside mine, Petaca district, N. Mex.

cent to pegmatrteVein 814 Near Bagdad, Arix.Vein 839 Boomer mine, Lake George, Colo.Vein 813 Elfrida, .A.riz. Vug in vein.Vein B 19 Elfrida, Ariz. Beryl-muscovite mass in

same vern as sample B l9Vein BJs California mine Mt Antero, Colo.Vein 837 Mobawk tin mine, near Hil l City, S. Dak.Vcin 838 Emerald mine, Chivor district. Colombia.Vein 813 Black Pearl rvolframite mine, Bagdad, ArizVein Bq Windy claims, Ariz.Vein 810 Scheelite mine. Oreana. NevVein 817 Lakeview Scheelite mine, Nev.Vein 836 Lakevierv Scheelite mine, Nev Margin of vein,Vein 832 Laheview Scheelite mine, Nev Center o{ vein.Vein 88, B45 Wolframite vein, Victorio Mts , N. X{ex.\rein B33 Tungsten vein, El Karit, MoroccoGranite 816, B44 Near Bagdad, ArizGranite 4 Agua Verde, Pima County, Ariz. Vein in granite,Granite 2 Lone Pine, Calif Vein in graniteGranite 3 Sheeprock Mts, Tooele County, UtabGranite Bl2 Sheeprock Mts, Tooele County, UtahGranite 811 Sheeprock Mts, Tooele County, Utah Atlite dike in graniteGranite 1 Mt Antero, Colo.Granite B30 Mt Antero, Colo Aplite dike in granite.Granite B.3l Mt Antero, Colo.Rhyolite MHs-8-t-55Thomas Range, Juab County, Utah.

1786 M. H. STAATZ, W, R. GRIFFITTS AND P. R. BARNETT

analyses were made on different samples of the same specimen to checkthe precision of the method.

Coarse-grained beryi was separated by handpicking. Finer grainedberyl was crushed and the heavl'fraction from a bromoform separationwas further purified by passing it through a Frantz Isodynamic Sepa-rator. The beryl was then handpicked under a binocular microscope fromthe beryl-rich nonmagnetic fraction. These samples contained no readilyvisible impurit ies. However, two beryls from veins in carbonate rocks(Lakeview, Nevada, and Victorio Mountains, New Mexico) had anoma-lously high values for Ca and one of these had anomalously high X4g.These samples were reseparated and, with all the samples analvzed in thislot (830 to 855), were boiled in dilute HCI to remove all carbonateminerals.

SpBcurocnePHrc ANALYSES

Both semiquantitative and quantitative spectrographic methods wereused (Table 2). Results obtained by quantitative methods are italicizedin Table 2. Barnett made all the spectrographic analyses except forsamples MHS-84-55 and 1 9 which were done b1'J. C. Hamilton.

To check the analytical results both semiquantitative and quantitativemethods were used on some specimens, and for nine specimens (5, 8, B20,B2l, B16, 88, 6, B6 and 822) a second sample from the same materialwas g iven a new number (B43, B+6,847,848, 844, 845,842, B50 andB49) and resubmitted. In most specimens the difference in the analysesof the two samples is within the analytical l imit of error, indicating thatthe method of sampling was quite precise. Semiquantitative results aregrouped between the l imi t ing values 1.0,0.464,0.0215,0.01 per cent ,etc . , and are repor ted as 0.7,0.3,0.15 per cent , e tc . , which are the ap-proximate mid-points of these groups. Except for Ca and Cr, the agree-ment between semiquantitative and quantitative results are good. Ex-cluding these two elements, 62 per cent of the values obtained by the twodifferent methods f all within the same group (Meyers et al., 1961). For Caand Cr there is a definite bias, with the quantitative results being higherby about a factor of two.

Elements that were not looked for are represented bv a dash in Table 2.An element that was looked for but notfound is l isted as " ( " followed by'a value representing the approximate l imit of detection. This minimr.rmlimit is not only different for various elements, but can varv for the sameelement. For example, the presence of moderate amounts of Na wii l inpart mask the presence ol Zn and makes its l imit higher than if there arelesser amounts of Na. Likewise. the detection l imits for the alkalies are

1787MINOR DLII,MDNTS IN BERYL

i 9 iE q q

I O V V V o

h 6 6 h b n 6 6 &

5 6 r E r 5 5 r B r E r S - ' - - - -. ,

" 9 9 ? " . : V

€ € F * E \ F € E € E € C E € E F € € € BE 1 E E 6 3 t E S E i 3 S 8 - 3 3 8 B B B S g Sf v V d V

' V t V t V V v V v ; v V v

* : , q r q { i i qqq r ! $E$ I t r gg iEEs

: r l- l 1 i 1 : : r l i l r l i " r r . r r H 9 : , : : r r r :

s 6 N s m ] m + m s o s * -:

i - - r i , $ , Q E l r i { { , i , r t lv v r v r v ' ' v

sl l c . l : l : l 1 I 1 l 1 l r l r l i : l r l : l :- \ \ S \ h

= x t r : r e a e ! e . \ : \ : - 6 a

.E ,r - E E - - - e Ec=E - El

r € I E$€ I €q€ IEEE€gEEEBEsEsF E € € E F q € € E € E E E E € ! * E €q n n i r - i q q i E - e 3 3 8 s 3 s 3 sv v v v v v v v v v v V V v $ v V1 \ /

= = E s = = a _ 9 . = _ * _= 9 , = a ? . 9 ! 9 g a a g o o o o a o s o s5 5 , E s = = S 3 3 S 3 . = 3 g g = 3 9 S 3 =o r i < i ' ; ' ; . ; , ; , ;v " ' ? V V V V V V

t r \ N - * 6 5 5 P - - P €= . 9 e P . 9 , P a e . c . o o i d o o- , - . - E t = , s , = 3 8 8 8 3 3 ; 3l- v g v v

e I qqE r €q€ rFqEq:es€SssESI

l *l ' :

q r €$e q ie r €qqqEEE€q€r€qV V V V O V V V

i i t S r r t i r * t S : : r r r ] - r r r : r : TC V V V V o V V V

€ 9 3 i 3 * g y y P 3 3 g s . s o s s s 3 g 3 : g , e€ € ! € p r i J r + i t E E E a ' 5 a a a = i E 3 E! F ! ! ! g E ! ! ! 9 E E q q 9 E q d q q 4 E d q

y !! p Ff F !o "ro !0 "bo F "b! 9i 9! 9! 9! qo u M M s S u ba u F!

9 Nc j

i : 4 I o - h a @ 6 * ( F O F C N N O € e c q c q c 4 O O O O ; O C O O

N

F

z

tsz

4 7-: r:

. q

v - |2 , 6

. jZ H

4 J

j

v

O

oO

oea,

a

N

j

F

O

i dI V

El

- t s , \ , , : 3 r , \ , > i- E € :g E E 'F € C E:EErE E:#

r a c 6

: : ? E P , S e e 9 e , o d b o o o4 i q q q - q 3 3 , 8 3 3 3 ' 3 8

6 6 : : E E b n e ! \ e ! Eerea ea : r I t rEE € P P E e e g == == === 3 3 3 1 = = = = = E A | = = 6 6 , - - - - - E

O V V V V C

M. H. STAATZ, W. R. GRIFFITTS ANI) P, R. BARNETT

9 a . ^ : , : ,6 o N o N t s m o Q E 9 r :i o o o L c o o o t oo o o o o o o c o o E S S -O

N

- , - 5 ' * - ' E Ea 5 8 , 3 = 8 3

n 6 6 b

! t : x l = x t = =: x = L x x r = x l

6 6 N 6m f ; : - o r

l q q l " e l " ? l

6 S 6 6

e a e e e a a 6 ' a a e Q q ?5 a 5 E t E = - ? , 8 5 6 6 E 5

c

6 6 6 b S 6 N 6 6m - - -

: . - . Y l - i : : - i l : : : - - . : i _ :V V o V v

o

1788

\

,

F

6 S 6 6 h & 6 S 6 h 6 h

@ : : : : : : : : : : : : :

I V V V V V V 9 V V V V VI V

E = E r E: : : r -

6 b

5€SJseEEES: o a r E g

t 6

- -

a a

t . - . ^. I s s , - s s ^ n - . i o - . . i o

s o o s o o e m F i o s o o< I ? 9 : ? 9 ? ? ? ? ? : ? :

6 S \ 4 6 h 6 h 6 6 S h b

V V V V V V V S I V V V V V

Q E E 5 r r * E F Fd E 5 d 5 5 . 6 E .

V V V

- o P * - - = * 5 *= S 5 8 S d E S g 3 3O

- P q 6 6 e

cc :qqq6 6 6

E R S = S : 8 , 3a - . : : : - . : i .

a = q a F a q : : ? 9 =: > = = : = x x x x xO V V V V V V

t s r *Q Q : 9 ! t a !x x x x = -

= ,: i B ; : \ : # : * b r a

V V V V V V 9 V

e : e P * e 5 : , 9 ? et = = - = = r . = =

V] Vi N 6 : : b

M O e S O O G N

> | ? ? : ? ? : ? ? ? ? : ? : -I V V V O

4 6 A 6 6 h t s t s 6 6 D \ 66 6 6 o - € o o o o \ o : = e ! 9 ! 9 ! i ! 9 l =o o 6 0 0 Q c o c o a o s o

v v ' v t . j i v i v v o

cO N t s t s h

- - 5 € - - * - - *\ * o c i i c o o s

6 6 O N t s \\ 6 - h 6 4 h f ; O E

b b 6 r h c hN r o \ o o o o o a

V V V V V

3 8 3 3 S 3 3 c 5 S. V

3 . - e 5 - S - - *S V V V

v l l r * r : r l \ : r E r : q \ r \ \ r I S :l v v o v v v v

N - €

z = E

Q r a a

r o t s r E m - \ o i a- lF t s N s t s F c m m m s b m s m 6 s n n s e ]

l r E o i F s o

E 2

b N * * + s ba r N o l a s o c t s s

' \ / o' V

- l i J l l ? r : 1 l r l 1 : T r r 1 E I a b l \ s $ F I t sa v

* & & o R R d d m m A A d 5 5 & 3 . + o o - - * o - *

SAdAEEEASEEEfrE E E S d i l . . A d Ef i8EEfi f iEfr

s \ 3 3 3 3 * 3 S - ! * 6 65 8 8 3 3 8 3 3 8 8 8 3 8 8 3 3 8 E 5 3 5 5 : 3 3 3" i d v vv ' v d

6 I 6 S N d S + b b b S 6 N 6 t s $ t s $6 - , C G O S O S o F E - - - O - - O S O O

6 h Q 5 h 4 h 6 6d b c 6 s s \ o s i - \ t s t s s - $ m m 6 m \

: : : : : : : _ . - i - : - 1 - . : : : : : ' i ' i * i r l - 1 ' 1 : : IO V V V V V O

= m O - 6 6o h m c c o N - n t s m N N - t s t s $

8 ' 3 8 8 8 | 3 i t : 8 r 5 3 3 3 3 3 8 5 8 ' 3 I Ni d ' v o

t -g I e & 3 * * - I 5 ; 5 e B s 5 5 : N 6 S B E E E E t & 31 -

EsEEs* , EE:sEeeE$s : *SeeEsEs sV g V

' V . j ' V ' V '

* 3 S * * * - * * * 3 * * * * * * S \ * q * - o - EV " c ;

. ' V V ' V " V ' V " o V V

MINOR ELEMENTS IN BIiRYL

V V V V V V V V V V

6 6 : S 6 6- t s r r E - \ - \ - - r - - o - - _ N m r -

* - * e 9 t r e , = F : i P = = = : = = ? ? 9 9 P i P eb i 6 - - - r S d E d E d S S d E E E : 5 6 6 - 5 d I' ' o . . . .

V O V V V V V V V V V V

6l l : : 1 P l T T \ l \ l l T l T l T s : : f l : l \ ?V V S

! T

g E

I

e l

. = I . o o A

P E € : e e Q E , F E , I I I r E E F € F , F- - 5 d - t 5 5 E t E - r 5 - - E 5 E E : 5 , d' c ;

\ / o

5 o P * * s + * q E q - q - - r * s - E * - - * - *

s \ $ * q \ $ $ $ b q \ h b , b sT S O O S t S t S t S S S t S E A A S \ I O l O S' v i " j i '

' , , ;

' ,

l lI I 9 9 6 6L A @ @ m N - o O * F ! E. . : m m - O - m 6 m m m q r ': - .

* F F - - r

. D l A

d

R

N

!

F

I79O M. II. STAATZ, W. II. GRIFFITT,' AND P. R. BARNT,TT

lor,ver b1' the quantitative than by the semiquantitative method becausemore sensitive analytical l ines were used for the quantitative.

V,q.nr,ttrox on CouposrtroN rN Drmonollr EwvrnoNuoNrs

Variation in chemical composition of beryl from different environmentscan be seen bv arranging the chemicai analvses of some elements ingroups according to their occurrence (Table 2). These variations are large

enough so that beryls f rom pegmatites can generallv be distinguished from

those from granite and from wallrock adjacent to pegmatites. Beryisfrom some veins can be distinguished from those from other environments

but berl. is from other veins cannot. Actuall1', vein beryl represents sev-

elal different t1'pes of veins and, as might be expected, the range in com-

position of these ber,vls is large. As onlv one berl ' l was obtained from

rh-volite-from the only United States localit l- known to us-litt le czrn be

said of the range of its minor elements. However, if ber.vl from otherrh1'olites is even somewh:rt similar in composition, it should be readily

distinguishable from beryl formed in other environments. The beryl from

the rhyolite wil l not be compared to the compositions of the other beryls,

but wil l be discussed separatel.v at the end of this section.In order to give a better visual picture of the differences in composition

found in different environments, the ranges in composition of the variousbervls are shown graphicallf in Fig. 1 for the alkali metals and for certain

other diagnostic metals. For a single beryl analvzed by both quantitative

and semiquantitative rnethods, only the quantitative data were used in

Fig. 1;where a quantitative analvsis was not available, a semiquantita-tive analr-sis was used. The compositionai range of beryl from a wallrockadjacent to pegmatite is based on onl1' six specimens, and when more

specimens become available the range may need to be lengthened for someelements. The compositional ranges for beryls from other types of en-vironments mav also have to be expanded. This expansion wil l includethe probable increasing of the ranges for l i thium, cesium, and rubidiumcontent of beryls from pegmatites. We believe, however, that althoughdetails of these graphs may need modification, the general reiation amongthe compositions of beryl from different environments wil l remain the

same.The alkali metals are generall-v- the most common minor elements

found in berl-l. The ranges of the metal content of two of them, sodium

and potassium, in beryl from diflerent environments are not distinctive,as they overlap one another (Fig. 1). The range in l ithium content inpegmatite berr' ' l is similar to that in beryl from wallrock adjacent topegmatite but is much greater than that from veins or granite. The rangesin content of the other two alkali metals (cesium and rubidium) in

MINOR LLEMENTS IN B]'RYL

P e g m o l i 1 e

p e q m o r r l e

1791

I A p p r o r i . o t e l i m i t o f d e l e c t i o n

x W o l l r o c k o d j o c e n l t o

lr$+l v"inoI o o -

o-ti. €F € bron'Ie

I R h y o l r l e

0 5 1 0Sodiom, werqht percen l

I A p p r o * , . o t e l i m i t o f d e t e c t r o n

{fl.{J,"n'",',"/ W o l l r o c k o d t o c e n l l o p e g m o l i t e,_

L i l h i u m , w e a q h t p e r c e n l

I O o o r o * , . o , " l i m i t o f d e t e c t r o n

o 0 5 r o 1 5

l ron , wergh l percen l

J 6ppro i lmote l rm i l o f de tec t ion

G r o n r l e

r R h y o l i l e

o

I

Polossrum,wergh l percen l

A p p r o r r m o l e l r m i l o f d e l e c l r o n

o 8Cesrum, we igh t percen l

J A p p r o x i . o t e l i m i t o f d e l e c l i o n

Mogn€s ium, re igh l percen l

I A p o r o x i . o t e l i m i l o f d e l e c l i o n

Ip-____fJ-"",,,",,,"! . - W o t t . o " f o d j o c e n i 1 o p e q m o t r t e

l e

ftHvern

^ ^A -Q c'on,reRhyot, te t

o r o 2 0 3Rubidium, werghl percenl

o o o 2 0 0 4 0 0 6 0 0 4S c o n d r u m , w e r q h l p e r c e n l

Frc. 1. Range in abnndance of various minor elements in beryl from various environments.

J a o o r o x l m o t e l i m i l o f d e l e c i i o n

o 0 2 a 4 0 6

. i - - i wol l rock odiocent to pegmotrte

o o z 0 4 0 6

1792 M. H, STAATZ. W. R, GRIFFITTS AND P, R. BARNETT

pegmatite bervl are much greater than that found in beryl in other en-vironments. Characteristicall l ' , bervi with high Li, Rb or Cs contentcomes from pegmatites rich in l i thium minerals and generally from theinner zones, where the l ithium minerals are ordinarilv concentrated. In aferv pegmatites, such as the Tin Mountain pegmatite in the Rlack Hil ls ofSouth Dakota, cesium is concentrated enough to form pollucite (Cs+Al+-SieO26.H2O). In most pegmat i tes, however, both cesium and rubid iumoccur as minor eiements in such minerals as lepidolite (Stevens, 1938) andbern'I. Hence, the higher content of Li, Rb and Cs in ber,vl from the innerzones of a pegmatite reflects the general increase of these elements in themagma that formed this part of the pegmatite.

Two other elements, iron and magnesium, mav make up more than 0.5per cent of some beryl (Table 2 and Fig. 1). I 'he range of the iron andmagnesium content of bervl from pegmatites is smalier than that of berylfrom other environments (Fig. 1). The range in iron content of beryl f rompegmatites overlaps in part the range of beryl from veins and rvallrockadjacent to pegmatite, but is less than that of any beryl from granite. Therange of magnesium content of bervl from pegmatite also overlaps theIower part of the range of ber1.l from other environments; most pegmatiteberl ' l contains l itt le magnesium.

Scandium, although occurring in only minor amounts, has a progres-sive increase in its range from pegmatite to granite (Fig. 1). This elementhas a verv small range in beryl from pegmatite, a l itt le larger range inberyl in wallrock adjacent to pegmatite, and a much larger range in bery'lfrom veins and granite. Almost ali of the beryl from granite containsmore Sc than that from pegmatite.

\ranadium and titanium occur in minor amounts, if at all, in beryl frompegmatite. The vanadium content of only a few beryls exceeds 0.005 percent from other environments; in two veins (Chivor district, Colombia,and Oreana, Nevada), however, the ber14 contained 0.05 and 0.07 percent vanadium. The titanium content of beryi rarell ' exceeds 0.01 percent, but is greater in a few beryls from veins and granite.

Minor amounts of. zinc are found in manv berl ' ls including almost all ofthose from pegmatite. The range in zinc content in beryl from pegmatiteis larger than that of bervl from any other environment. The overlap inthe ranges oI zinc content in beryl from different environments, however,is quite large, and, hence, the zinc content is not useful in separatingberl. ls from difierent environments.

Some of the minor eiements that are found in beryl in wallrock adjacentto pegmatite evidently came from the pegmatite, others were obtainedfrom the wallrock. Our data were derived f rom two iocalit ies: the Hardingpegmatite is represented b1' ber,vl sampies from the core (B41), wall zone

T,IINOR EL}).MIJNTS IN BIiRYL r793

(B40), and the adjacent wal l rock (Ba9); the Avon pegmat i te is repre-sented b)' bervl f rom the wall zone (B5) and the adjacent wallrock (B50) .From this small group of specimens the folJowing trends were observed:(1) The amounts of Li, Cs and Zn in berl ' ls in both the wallrock andadjacent pegmatite are similar, suggesting that these elements migratedfrom the pegmatite; (2) the amount of Rb in bervls in the wallrock ismuch less than that in the pegmatite, suggesting that only a l itt le of thise lement migrated f rom the pegmat i te ;and (3) the amounts of T i , Sc, V,Cr, NIg, Fe, Na, Ni and Co in beryls in the wallrock are greater than thatin the pegmatite, suggesting that these elements are largell. obtainedfrom wallrock. Nickel is of special interest, although it occurs in onl_vminor amounts, because the onlv bervls containing any of it came fromwallrock adjacent to pegntatite. Two of the three specimens containingCo also came f rom rval l rock l the other f rom a vein (B14).

Vein bervl is found associated with a varietrr of minerals, most of whichare generallv considered to be of high-temperature origin. It is commonlvassociated with tungsten minerals (scheelite or wolframite), and half ofthe beryl localit ies from which specimens were collected for this reportcontained tungsten. It is also commonlv associated with cassiterite,molybdenite or just qtartz. The emerald veins at Chivor (B38) are a raret-vpe and consist of pyrite, aibite, quartz and emerald (Mentzel, 1931).The ranges of some of the minor element contents in beryl from thesevarious types of veins are, therefore, large. We thought that a significantdifierence might occur between the tungsten-bearing veins and the non-tungsten-bearing veins but examination of Table 2 shows that the rangeof most minor elements is similar in these two groups of veins. The biggestand perhaps the only significant difference was noted in the titaniumcontent; the beryl in tungsten-bearing veins contained from 0.00015 to0.007 per cent, that in non-tungsten-bearing veins contained from 0 to0.07 per cent. As manv veins are poor in most alkaii metals, the berl-Ifrom veins have small ranges and low contents of Li, Cs, Na and Rb. Insome places rock t1.pes similar to both vein and pegmatite are found, andin these gradational types the minor element content of beryl is similar toboth.

Vein beryl is diff icult to distinguish from that in wallrocks adjacent topegmatite, as both types have a high and simiiar Fe and Mg content. Thelarger Li range in beryl from the walirock adjacent to pegmatite mav aidin distinguishing some of the bervl in these two groups.

Ber14 from granite contains relativell. minor amounts of Na, Li, Cs,Rb, Cr, Ni and Zn and the range of these elements is small. Probabll- thetwo most diagnostic minor elements are Fe and Sc. The Fe contents ofbervls from granite are all relativel.v large as compared to those from

1794 M. IT. STAATZ, W. R GRIFFITTS AND P, R. BARN]].TT

pegmatite (Fig. 1). Scandium has a very large range of values, with a

mean of about 0.03 per cent Sc. Although this range is large it wil i un-

doubtedll ' have to be increased, as the mineral bazzite, which occurs iu

drusl- cavities in granite near Baveno in the Alps (Spencer, 1916), is,

according to Fleisher (1955), a scandian beryl. Beryl from granite, as

noted above, can be easily distinguished from that in pegmatite but it is

rnore dil icult to distinguish from that occurring in veins and in walirock

adjacent to pegmatite. The presence of Ni or Cr and a high Li contentma-v aid in distinguishing berl ' l adjacent to a pegmatite from that in agranite. A low Fe content in ber1.' l from a vein might aid in distinguishingit from beryl in a granite.

The red beryl from the rh1'olite is probably the most distinctive of ali

the bervls studied and its content of several minor elements is unique.'Ihis bery'l has a relativeiy low Li, Na, Mg and Sc content; a relativelyhigh Ba and Cs content, and the highest content, of anv beryl analyzed,

of Rb, Fe, T i , Zn,Mn, As, B, Ga, Nb, Sn, Y, Yb and Zr ( ' fab le 2) . Of the

elements in this last group, As and Nb were detected only in this bervl.Because rhy6]i1" and granite commonly are of similar composition, berylsfrom these two environments might be expected to be similar. Surpris-inglr', this is not so, and the minor element content of the berl ' l fromrhy'olite falis within the range of that found for granites only for Na, K,

Li, Ca, Ba and Cu; for Mg and Sc it is less than this range;for all other

minor elements for rvhich we have data, it is greater. The beryl from the

rhyolite has several other anomalous characteristics. In most bervls, if

one of the alkali metals (Na, Li, Rb and Cs) except K is high, the others

are also. In the beryl from the rhyolite, the Cs and Rb contents are high,

Na is moderate and Li is low. Commonlv beryls with high Fe content

have a high Mg content, but this beryl has one of the lowest Mg contents

and highest Fe contents of any beryl found.

AcxNowrolGMENTS

Many of the samples were donated to us and we wish to express our

thanks to Gardner Gantz Ior the emerald from Chivor (838); to L' R.

Page for the bery ls f rom the Bul l Moose (8) ;Ross (7) , and Scot t (B34)

pegmatites, and from the Mohawk tin vein (Be7); to J. W. Adams for the

ber'1'I from Mt. Antero, Coiorado (1), Lone Pine, California (2), Agua

Verde, Arizona (4), Sunn.vside mine (6), Harding pegmatite, NewMexico (B40) (841) (822), California mine (B35), and Millard-Chandlerpegmatite, New Hampshire (B20) (B21); to D. M. Lemmon for the bervlfrom the Lakeview vein (B32) (836); and to Jules Agard for the berylfrom Ei Karit, Morocco (B33).

MINOR I'LEMENTS IN BERYL 1795

RelnnnNces

Bnncc, W. L aNl J. Wnsr (1926) Ttre structure of bervl, BerAlrSioOrs. Royal Soc. LondonProe. , ser . A, l l l , 691-714.

Donru,rw, N. D. (1952) on the question of determination ot the origin of beryl. Aka.d...a/azrfr "ISSR l)okl 82.623-624.

Fr.nrscnrn, Mrcnanr. (1955) Bazzite (a scandium beryl?). Am. Mineral 40,370.For.rNsnm, R. E. (1941) Optic properties of cordierite in relation to alkalies in the cordi-

erite-beryl structure. Am. M ineral. 26. 485-500.HrrternaNo, W. l'. (1905) Red beryl from Utah. Am. Jour Sci.,4th ser., 19, 330-331.Hones, S. W. (1944) Tungsten deposits in the Boriana district and the Aquarius Range,

Mohave County, Arizona. U. S. GeoI. Suraey BuII,.94O-I,247-264.Koscnuaxr, A. H., J. J. Gr.ess, aNo J. S. Vnav (1942) Tin deposits of Irish Creek, Vir-

ginia. [/. S. Geol,. Surtell Bull.936-K, 271-296LaNnns, K. K. (1935) Colorado pegmatites. Am. Mineral.20,319 333Mesot, BnreN (1952) Princi.ples oJ Geochemistry. John Wiley and Sons, Inc., Ner, york.Mrwrztr,, Cnanms (1931) Geology of Chivor no. 1mine. Am. Inst. Mining Metall. Eng.

Trans 213-216.Mvans, A. T., R. G. Hevons, ,LNo P. J DuNrow (1961) A spectrochemical method for

the semiquantitative analysis of rocks, minerals, and veins. LI. S. GeoI. Surte.y BuIl.ro84-I,207-229.

Nonrow, J. J., W. R. Gnmnrrrs, enn V. R. Wrr.uentn, (1958) Geology ancl resources ofberyllium in the United States.Znd Inter. Con[ PeaceJul Uses oJ Atomic linergy,Geneta, 1958, Proc.2,21 34.

Pacn, L. R..qNo ornnns (1953) Pegmatite investigations, 1942 1945, Black Hills, SouthDakota. U. S Geol. Suney ProJ. Poper 247.

ScHnr.r.tn, W T., R. E. Srnvrws, ano R H. J,luNs (1962) An unusual beryl from Arizona.Am. Ll,ineral. 47, 672-699.

SneRrnlw, D. M. (1955) Geoiogy of the High Climb pegmatite, Custer County, SouthDakota. U. S. Geol,. Suney BulI l0l5-C, 59-98.

L. J. SrnNcnn (1916) A (seventh) list of new mineral names. Mineral. Mog. 17,344,362.Sraerz, M. H. ,rno A. F. Tnrms Jr. (1955) Geology of the euartz Creek pegrhatite dis-

trict, Gunnison County, Colorado. LI. S. Geol. Surwy proJ. paper 265,-- L. R. Pecn, J. J. NonroN enn V. R. Wrr.uanrn (1963) Exploration for beryllium

at the Helen Beryl, Elkhorn, and Tin Mountain pegmatites, Custer County, SouthDakota. U. S. Geol,. Surwy ProJ. Paper 297-C, 129-197.

Srevnxs, R. E. (1938) New analyses of lepidolites and their interpretation. Am. Mineral.23.607-628.

TnunsroN, W R. (1955) Pegmatites of the Crystal Xlountain district, Larimer Counr1,,Colorado. U. S. Geol,. Surtey Bull. l0ll.

WrNcnrr.r., A. N. (1947) Elements oJ Optical, Minerology. John Wiley and Sons, Ner,v york.