Dennis G. Lewellen,Deparlrrenlol Geoogy
0059
Un versty ol KentlrckyLexnoton Kerrtlcky'10506
Petrographic Differences Between Two Sedimentary Rock Bodies of
the Eocene-AgeManastashFormation,Kittitas County, Washington
Abstract
Ioiliesof1lr. Ilara<r$
l f o r r n e r i , r !nh o $ l I d r a i g r i l l
. i ? . i n . r e r s e " .r l r r . t z . o n l e f l
Lnrrr:*,
n l ,r. grr rF z" ,i 'r'Lr''.
t x r r r * r , l r , , c k s ' , 1r i i l l i t i n s r o n t , , , s i t i , , n
lntroduction
rocks are fairly rridespread
Although sedirnentar,r'
along thc [1anksof thr rorth and cenlral Cascadc
\lountains of Washington, the latcral relationships bctrveengeographicall,l separatedeposits
are not lrell understood.Correiationsare difficult
to establish, bccause of a lirck of either prorni'
nent nrarker bcds or obvious guide fossilsuithin
these lithologicall.vsimilar units. Thc sedirnentarr rocks are predominantl! composed of
arko-qic sandstones whic'h are l8-'18 percent
fcldspar (mostl-"-plagioclase)and 40-7B percent
q u a r l z( F r i / r P l ll a i q t . P l a l t l ' " i l r , r r r a i r ri n. i n tercalatcd shales suggest thal most of thc
sedirnentarrrocL bodies arc Eoceneto Palcocene
(?) in age (Nervnranl98l). Radiometric dates of
*ith thc sedimentarl
volcanic rocks interbeclclecl
rocks arc generalll Eocenein age, but the resulls
var,v rvidel.''due to alteration of the volcanic
m a t c r i a l a f t e r d e p o s i t i o n( l r i z z e l l 1 9 7 9 ,T a b o r
et al. l9B4).
Recent studics hare shc,n'nthat the nineral
compo-.itionof sand changcsrith grain size and
that this rclationship rrral |e usclul in solving
problerns in the Manaslash !'ormalirrn crrnctrlt_
ing lhc soLlrcerocks, clirnate, dcpositional cn'
vironmrrtt, or agc rclation-rhipsbetweensedimen'
tar,r rock bodies(Flores I96?, Davis and Ehrlich
19i,1, Basu l9ifi). lhis studv utilizcs similar
in ordcr to resolvea problem of origin
melhc-rds
arrd age relationshil)sof hro sela.alc, but })roximal, boriiesof Tertiar! sedincntarl rocks in cen_
tral s.ashingtort.
T h e s t u d l a r e a i s l o c a t e ( la b o u t 8 k r n s o u t h
o f C l c I l l u r n ,\ V a s h i r g t o n .v h e r c 1 . 6 t o 3 k m o l
160
N o r t h r c s tS c i e n c tV
, ol. 60,No 3, 1986
vounger volcanic flows, tuffs, and volr:aniclastic
rocks separate t1fir -.edinentarv roc'k bodies of
the Eocene'ageN{anastashFcrrmation(stc areas
A and B on Figure l). SedirnentatJrocks ix arcil
A have a thickness of approxirnatelv 1,000 rn,
rhilt the rock- irr area B lrirre a nr:,rirnurrr
thickness of i50 m. Both areas unconformabi)
overlie cr;-stalline rocks composcd of -.chist.
phvllite, and quartz-diorile gneiss (Stout 1962),
and both are unconformabh orerlain [rv Eocene_
age volcanicflows, luffs, and lolcaniclastitrrocks
Both rock bodies are corrlposedmostlYof arkosic
sandstonewilh rninor amou ts of shale,siltstone,
conglomeratc, and coal. IIowever, arca B rocks
are substantiall"v coar-qer grai ed and more
quartzosethan area A rocks, $hich are compitrastutliesirtdicatethat
tirel,r leldspar-rich.PrevioLrs
these rocks were deposilcd b1 fLuvial processe-.
(Lewellcn l9B3)n'ithin a *arm, tellperate to -subtropical clirnate (Stour 19.5?),alrd that area B
rocks mal be olcler thar area A rocks (Nervman
l98l).
Differences in the lrcnds of grarrt slle ver_
sus mineral compositionof the sedimerltar!rocks
of area A and B werc alalvzed in an effort to
resolr,ethr problern-.of dcpositionaland age rela'I
tionships. hc presently disjulrc'tbodie-.mirl be
rornnants of a olce larger sedinlentarl urlit, if
cliffcrcrces in mineral contenls are a rellcclion
o f d i l f e r i n g g r a i n s i z e st : a u s c db r c l e p o s i t i o n a l
processes.Conr.ersclr,nlinelalogical cliifcrcnces
g r e a t e r l h a n t h o s c a s s o c i a t e dn i t h g r a r n _ s l z e
repre_
r ariationsrrar- indir:arcthat the -.andsl()nes
scnt differcnt ages! source areas, or other
u n r e l a t e dd c p o s i t i o n a fl a c l o r s .
Methods
Fourteen sandstone samples fronr area A and
thirteen fronr area B ucrc collectcd at sites that
r " f l " , 1 l h ' . : r r c r lr n , l . t r r l i g r r p h i , ' r r r i u t i o n i. n
the two rock bodies(Figure l). Thin sectionswere
prepared arrd stainedfor potassiumfeirisparand
plagioclaseaccording to the tcchniqucs of Lainz
e r a l . ( l q o 4 )a n , l\ u r m . r n ( 1 q 7 4 )T. h . e p p r u r i m a r e
grain sizc of each sample $as cxprcsse{iaccorct-
F i g r r . L L o . a r i . n d a l o l t h . s t u d ) a r . n a n d t h c s a . d s t o n cs a d p l . s .
( c l . E l u m ,w o 3 h , s L m n o r r h )
r r.,rss Somple locotion
Petrographic Differences Betveen'fwo Sedimentarr Rock Bodies
t6t
TABI.E
l
Crain size and relarire perccnl ,luantities of
quartz and fcldspar n,.asu..d in each sanple.
Feldspar
Gram Size
5l
m-2
39
iJ{)
m.l3
n'16
m 1?
m-23
nt2 t'
n-29
m.33
n.31
3;
n, 36
m3?
or-38
r39
m-12
m.13
nr-4;
22
3l
.19
.13
3B
39
35
't3
3;
32
91
88
6i
a2
?8
8?
!13
69
6i
line
;8
69
5l
5i
5'l
62
65
57
63
4a
9
l2
33
l8
22
t3
'i
u0
3l
33
20
59
8ll
6g
2t)
:12
Resultsand Discussion
fine
ing lo the Went orth scale (ver1-finc, fine,
medium, or coarse-grainedsandstone)bascd on
average measurement of tlrc lortg axis of l0 to
15 feldspar or quartz grains randomly sclectcd
through the use of a randorn digit table. Detrital
c o m p o n p n t sr r e r r .d r t e r r n i r r e dh r p o i n t e u u n t i n g
an average of 800 points per sample (Chales
l 9 l a ) . C a t e g o r i . su f r " ' u g n i z c d d e t r i l u ' r . . r e
monocrlstalline quartz, polycrystalline quarlz,
plagioclase,potassium feldspar, volcanic lithics,
sedimentary lithics, metamorphic lithics,
microgranular lithics, micas, clav-sizedmaterial,
and a nriscellaneouscategorv for those grains
l h a l d i d n o l f i l i n l , ' a n 1 r , I r h c . , t h e rc a t e g o r i e s .
ln order to dctcrnine the grain size and
mineral composition relationship of thc sirnds t o n e . r .i t h i n ' . r , h r r , r . l i n r u r - r c g r p ' s i o nt s q u o tions i{ere derivcd for grain size fersus feldspar
and quartz contents.
T-tests {ere used to compare thr'slopes and
intercepts of sinilar regressionequations of thc
dilTerent rock bodies so that mineral composition {ariations not attribulirblc to grain size
changcs could be ascertained.Nonsignificalt t-
162
Levellen
test results, determined at a 95 perccnt confidence lerel, for the comparisonof the slopeand
interceptsof the linear rcgressionequationsfrom
th. r"cl L"tlie' irJicale rhal difl, r"n,,. in
m i n e r r r lc o n t e n t a r e a l l r i h u l o h l , ' t u g r . r i n - s i z e
trends. Significant t-testresultsfor thc slopc, in'
tercept, or bolh indicate differencesgreater than
those assoc;atedwith grain size varialions and
require other explanalions.
The orineral composition and the grain'size
mcasurernertsfor each samplearc givcn in Table
l. In general, sandstone samplcs are vert-fine
grained in area A and coarsc grained in area B.
The average conrpositionof the samplesin area
A is 40 pcrccnt quartz,58 percent feldspar, and
3 percenl rock fragments, whcrcas area B
samplcs alerage 67 percent quartz,20 percent
feldspar, and 13 percent rock fragments. fhese
sets of ave.agesarc very similar to those of the
sandstoncunits derived fron upliftcd bl,rcks of
Holocene to Itrecambrian age granitic rocks
studied by Dickinson and Suczek (1979).
Figurc 2 gives the linear-regressiondata of
grain sizeon quartz and ieidsparcontent for each
r r c a . T h " . e d a t a i n d i rJ l e 1 1 1 13 c g r s i n . i z e
feldspar content increases,and quartz
decreases,
T-testsindicate that the slopes
content decreases.
of likc regressionequations from cach area are
not significantll differcnt. However,t-testsof the
i n l c r l ' c p l s o f s i m i l a r r e s r e s s i u nl . q u a l i o s a r e
significantly different. Although some of the
variations in mineral composition are attributable to grain size,additional conrpositional
differencesnot related to grain size exist between
the rocks of areas A and B (Table 2).
'l AflLE 2- Results
(h)andtheinter.epts
of t teslsof theslorrc
(.r)ofrheres.e$ionequarions
lromarcasA andB.
Hlpothesis
'I
test"
Ho:trq - tr3
H 1 : b 1+ b g
I = 0 . 8 7 ? b( l l = 2 3
H u : a , 1= a g
H 1 : a , q+ a g
t = 3.021'df
a.r:.0;.t=2.069
b itifferencca
s re nonsignifi.ant
c d i f f e r e r c e s! . e s i g n i l i c a n t
23
loo
90
o
80
o
70
o o
o
o
o o
60
q)
c
o 50
Syx= 8.87
(J
|{
40
Syx:6.96
30
o
20
80
Yx=6.96
o
c)
60
CL
Y x = 8 . 87
50
o
c 40
o
\J
E
tl-
o o
30
o
o
o
o
o
t
2
G r o i n S i z e ( i n p h i u n i t s)
d e c r e o s e s- i
l-igure 2. S.alr.r plot oi grai. si1,i i.rsui
p!en
qurrlz and felispar content ol27 samptesfrori area A (solid.in'l.sl
arll !r!.d B
ctnet
Petrographic Differences Betreen Tuo Sedimcntalv llock Bodies
163
The data indicate that the rocks of areas A
and B probably represent two separate deposi'
tional events, probably of differing ages. Such
a conclusion is consistent with the fossil plant
evidence of Newman (l98l).
Disregarding grain-size trends, the higher
quartz contents of area B sandstonemay reflect
climatically induced differencesin the degree of
rleathering and soil formation processesin the
source rocks. For example, sand samples collected in modern streams draining granitic rocks,
not more than 19 km from the source rock, in
the warm moist climate of the southern Appalachians, contain less feldspar and more quartz
for a given grain size than those from the cool
dry northern Rocky Mountains (Basu 1976).
Linear'regressionequationsfor the Appalachian
data are similar to those of area B, whereas equations for the modern Rocky Mountain data are
nearly identical to those of area A (Table 3).
ion equations
1.om
TABLE 3. Conparisonof linear-regress
stud).
thisstudyandfron Basu's(19?6)
QuartzConienr
Regressedon
Crain Size
Source
F e l d s p a rC o n t e n t
Rcgressedor
Grain Sizr
Y = 77.83 6.33! Y = 2 2 . 1 ? + 6 . 3 3 *
B a s u( 1 9 7 6 )
Y = 86.62 9.75x Y = 1 3 - 3 8 + 9 . 7 5 x
Area B
source region appears to be small.
A more likely explanation is that compositional differencesin quartz and feldspar content
reflect different source rocks. Suitable source ter'
ranes include a wide variety of crystalline rocks
extending from central Washington to central
l d a h o .A l t h o u g hm o " t o f t h e " er o , " k sa r e g r a n i t i c
or gneissic,which would yield sedirnentssimilar
to the composition of the rocks in area A, others
are lov' to medium-grade metamorphic rocks cut
by numerous quartz veins. These quartz"rich
rocks could presumably provide substantial
q u a r t zd e t r i t u s a. n d t h e a c c o m p a n l i n gm i c a c e o u s
minerals of the metamorphic rocks could be
broken down during transport and carried
beyond the area of deposition as fine-silt and clay
particles.A mixed sourceterrane of granitic and
low- to medium-grade metamorphic rocks could
provide the detritus represented in the
Manastash Formation by the rocks of area B,
*hich are nearly 70 percent quartz.
Different source rocks most likely account for
the mineralogical differencesbet*een the rocks
of areasA and B. As such the detritus was either
derived frorn two geographically separate source
terranes or from erosion of the same site through
time. In the latter case, the older rocks of area
B were derived from low- to medium-grade,
quarlz'rich metamorphic rocks, whereasarea A
r e c e i v e dd c t r i t u s f r o m a n u n d e r l l i n g i n l r u s i v e
g r a n i l i ' "p l u t o n o r g n e i s se r p o s e da f t " r e r o " i o n
of the overlying rocks.
Roclies
i :18.17
3 . 6 ? ri = 5 1 . 8 3 +
i:52.56
4 . 5 9 r Y = 4 ? . . 1+4, t . 5 9 x
3.67x
BasL,(1976)
Paleobotanical data from the Manastash rocks
indicate a warm, &et climate at the site of deposition, thus the observed compositional differences
betweenareasA and B could reflect climatic differences in the source terranes, particularly if an
age difference exists between the two areas. Considering the short distances between the secllment
source and the site of deposition, however, the
chances of radically different climates seem
remote. Basu's sediments were collected within
l9 km of their source and the likelihood ofrapid
climatic changeswithin l9 km in the Manastash
164
Lewellen
Conclusions
Analysisof mineralcompositionand grain size
adjacentbodiesof early
of two geographically
rocksseparated
by
Tertiaryarkosicsedimentary
youngervolcanicrocksin centralWashingtonindicatethat the two rock bodiesare mineralogically distinctafter grain sizeeffectsare takeninto
account.The rock bodiesare thereforeinferred
to be of differentages;a conclusionthat is in
accordwith previouslyreportedpaleobotanical
findings.Differencesin mineralcompositionof
the two bodiescan be attribuledto differences
of the
in climateor petrographiccharacteristics
sourceterranefrom which the sedimentswere
is
derived.However,the mostlikely explanation
in the sourcerock
that petrographicdifferences
are relatedto an unroofingsequence
causedby
uplift and erosion lhat exposed rocks of differing conposlton.
Acknowledgments
I wouldlike lo thankDr. JohnC. Fermfor his
adr ice, guidance, and critical evaluation of this
papt'r, and Dr. Charles W. Walkcr for his
guidance and support in doing the research.
Thanks are extendcdto lhe State of Washington,
Division of Ceologyand Earth Resourcesfor pro'
viding financial and malerial support and to
Sigma Xi, the Scicntific Research Societ,v,lor
providing a Crant'in'Aid.
Literature Cited
B a s L rA, . l 9 ? 6 . P e r r o l o g ro f H o l o c c n cl l u v i a ls a n dd e r j r e df r c m p l u t o n i .s o u r c er o c k s :i r n p l i c { l i o n st . , p a l c o . l i n ! t i . i n t e r p r e r a
t i o n . J . o l S e d .P e t r o . , 1 . 1 : 6 9 . 1 - 7 0 9 .
C h a r e s ,F . 1 9 1 9 .A s i m p l ep o i n t c o u n r e rf o r t } i n s e c t i o na n a l y s i sA. m . M ; n t r a l o g i s t3 4 : l ' l l
l l n i t e d S t a t e sC
. eol.
o n d d r n a m i f si n l h c s o u t h e a s t e m
D a r i s , I f . \ ! . , a n d R . t h r l i . h . l 9 ? 4 . L a t eP a l e o z o i c . r u s l n l c o ' n p o s i t ia
S o c .o f A m . S p e c i a lI ' a t . r l 4 U : l 7 l 1 8 5 .
s ull.
D i c k i n s o nW
, . R . , a n d C . , { . S u ( z c k .l 9 ? 9 . P l a r et e c t o n i c sa n d s a n d s t o . e . o n p o s i t i o n s, ^. r n . . 4 s s o co.l t l ' t r o . C e o l o g i s t B
62121642142.
F l o r e s ,R . I L l 9 6 7 . V a r i a t i o ni n r n i n e r a lc o n t o ! i t i o D d u . i n g r r a n s p o r tJ. o t S c d . I ' e l r o . 3 2 : 2 3 52 3 9
F r i z z e l l , V - . { .l 9 ? 9 . } ' c l t u , l o ga
} n d s t r a t i g r a p h yo f p a l c o g e n en o n m a r i n es a n d s l o n e sC,a $ a d c R a n g e ,W a s h i n g t o nL. S C e o l .
S o r v . O p e nF i l e R e p . 7 9 - l 1 4 9 .
L a i n z ,R . V . , R . l l . S t e r e n sa
. n d M . B . N o r m a n .1 9 6 ' t .S r a i n i r g o f p l a g i o c l a slec l d s p a ra n d o t h e r n n r e r a l sr i t h F . D a n d C .
R e d n o . 2 . U . S . C c o l . S r n . P r o f . P a p e t5 0 1 - 1 3 : t s 1 5 2 ' B l 5 : 1 .
, i t t i r a sC o u n l y .W a r h i n g r o n .
. h e s t r u c r u r ea n d d e p o s i r i o n aeln v i r o n m e nor f t h e N l a n a s t a slho r m a r i o nK
L e q e l l e nD
, . G. 1983T
E J - r - r '\.d . 1 i , g n I n i r 'r . i t . .T h . i Newnan, K. R. l9Bl. I'tl!noiogi. biosrratigraphlof somr:early Tertiarl nonnurine formalionsi. ..nbal ard westernWashington.
C e o l . S o c .o f A n . S p c c i a lP a p e r 1 8 1 : 4 96 5 .
N o n n a n ,I f . B . t 9 7 , 1 .I m p r o r e d t f c h n i q u e sf o r s e l e c t i r es t a i n i n go l f e l d s p a .a n d o t h e r m n r e r a l su s i r g a n a r a n t h .t J - S -C e o l .
S u r v .J . o J R e s .2 : ; 3 7 9 .
porrioDof thc !lt. Sluart quadrangle,Sashlnglon. Uniw:rsitr of trashington.Thesis.
Stour.Nt. L. i957. Ceologrot thc sourhwesrern
. e o l .S o c .o l A m . I l r r l l .7 5 : j 1 137i j , l .
l a s c r d eN l o u n t a i n sW, a s h i n g r o nG
. 1 9 6 , 1C. e o i o g yo l a ! d r t o t t h e s o u t hc e n r r a C
T a t r o r ,R . W . . v . A . F r i z r e l l ,J r . , J . , { . \ ' a n c c . .a n d C . t r . N a e s e r .1 9 8 1 .A g c s a n d s t r a t i g r a p h ,o"f- l o w e ra n d m i d d l e l t r L i a r v
s e d i m e n r . r rd r J v o l c a n i cr o c k so f I h e c e n t i a lC a s . a d e sV. a s h i n g l o n rr p p l i ( a l i o n t o l h . t e . t o n i . h i s t o r r o f l h e S l r x i g h l
C r e e kF a u l t . G s r l . S o c .o f A n . B u l 1 .9 5 : 2 6 - 1 4 .
Receitetl 4 f ebruary 1985
Acceptedfor publication1.3August 1985
Petrographic Differences Betwcen Two Sedimentart Rock Bodies
r65