Bookstrom 1977
Bookstrom 1977
Bookstrom 1977
Abstract
.?
Fro. 1. The Main pit at E1 RomeraI as photographedfrom its easternrim in 1970. Cord6n
La Liga lies in the background,to the west of the pit. Maj or geological boundariesand features
are indicatedby the black lines and labels. The pit benchesare at 10-m vertical intervals. and
the pit floor is at the 310-m elevation.
geologyof E1 Romeral was summarizedby Ruiz the determinativecurves of Heinrich (1965). In-
et al. (1965, p. 230-232). dices of refraction of other minerals were estimate•
Methods
(to onedecimalplace),measuredin index oils usin•
white light (to two places),and in oilswith a sodium
Open-pit exposureswere mapped at a scale of light source(to three places). Pyroxeneand amphi-
1: 1,000,and outcropsin a 13-km• area surrounding bole crystalswere properly orientedwith a spindl•
the mine were mappedat 1:5,000. Reconnaissance stage for determinationof their optical properties.
traversesbeyondthat area were plottedat 1: 50,000. Opaquemineralswere identifiedin polishedsectio•
About 1,700 sampleswere collected,studied, and by reflected-lightmicroscopic methods,accordingtc
classified,and derailedmineralogicand petrographic determinativecharacteristicstabulatedby Uytenbo-
analyseswere doneon nearly300 samples. gaardt and Burke (1971).
Determinative dam for the minerals of the Romeral Other physical and chemicalmethods used i•
area have beengiven previously(Bookstrom,1975, mineralogicaland geochemicaldeterminationsar•
AppendixC, p. 374-396). Approximately600 plagio- briefly describedin the text or tableswherethey are
clasedeterminationswere made,using the universal first mentioned. These include X-ray diffraction,
stage. The Michelle-Levy and Ritmann methods X-ray fluorescence, emissionspectrographic analysis,
were routinely used to determinecompositionsof microprobeanalysis,titration, colorimetrictests,and
polysynetheticallytwinnedplagioclase crystals.The determinationsof ionic concentrations by specificion
Federov method, in conjunctionwith determinative electrode.
curvesby Hutton (1971), was used to determine In estimatingmodal compositionsof rocks 60•
both the compositionand structuralstate of at least pointswere initially countedper sample,and then a
one plagioclasecrystal from each rock type. Un- 1,500-pointcount was made exclusivelyfor opaque
twinned plagioclasegrains were identifiedby mea- mineralspresentin smallquantifies. Averagegrain
suringthe X' index of refractionon (001) and using sizes were determined,and the Holmes correction
3•.4GNETITE DEPOSITS OF EL ROMER.4L,CHILE 1103
72 ø 7 Iø 70 ø 69 ø
-- 26 ø
--27 ø
L/mirof Ch/leon•
.i Iron t•rov/nce/--"
,CERROIM•,N
Copiap6
ß
-- 28 ø
? /'
Area of closely spaced
magnet/re - apat/te deposits
ntofagasta ,
:
-- 29 ø
EL
ALGARROBO
,EL
TOFO .
ELROMEHAL
_ 30 ø Lo
Serena /
.•,••'•...'
ß N
f 0
I •
I00km.
I
,'
--31 ø
.:
I dfi i •ß1
Fro. 2. Index map,showingthe limits of the Chilean•ron province,and the locationsof the
iron minesof E1 Romeral,E1 Algarrobo,and Cerro Iman (after Ruiz et al., 1965,part II,
fig. 4).
Fro. 3. Generalizedeast-westcrosssection,showingthe geologyof Chile at the latitude of
El Romeral (29ø 43' S). This sectionwas constructedœromthe GeologicMap oœChile (Ruiz
and Corvalgn,1968), œroma contourmap oœthe Peru-Chile trench (Scholl et al., 1968), and
from a cross section of the inclined seismic zone (.lames, 1971). Ps= Paleozoic schists,
phyllites, and quartzites; Pg = Paleozoic granite; Trv'- Triassic volcanic rhyolite; Jv =
Jurassic volcaniclasticandesires (La Negra Fro); Jm=Jurassic marine limestones; Jgd=
Jurassic granodiorite; JKqm = Jurassic-Cretaceousquartz monzonite; Kap = Cretaceous La
Liga andesiteporphyry; Ka -- Cretaceousandesiresills and marine limestones(Arqueros Fro);
Kd = CretaceousRomeral diorite; Kin = CretaceousRomeral magnetite deposits; Kgd -- Cre-
taceousgranodioritic batholiths; Kv, = Cretaceousandesitesand continental sedimentaryrocks
(Quebrada Marquesa Fro), Kv•=Cretaceous andesiresand continental sedimentary rocks
(Vifiita Fro); KTg = Cretaceous-Tertiarygranitic rocks; Tv = Tertiary volcanicrocks.
factor was applied for very small opaquegrains The Paleozoic basement rocks of E1 Romeral are
(Chayes,1956). Suchgrainscommonlyconsistof unconformably overlainby erostonalrenmantsof
intergrowths of minerals that are identifiable in volcaniclasticisland-arcandesites(Fig. 3), possibly
polishedsectionbut not in thin section. Therefore, correlativewith the La Negra Formation (Garcia,
visual estimateswere made of the volumetricper- 1964)of Jurassic age. Northandeastof E1Romeral
centagesof the variousmineralsin suchintergrowths the Jurassicvolcaniclastic rocks are overlain by
in eachof 10 grainsin the polishedbillet from which Early Cretaceous(Neocomian) marine andesite
the correspondingthin sectionhad been cut. The flowsand sills of the ArquerosFormation (Aguirre
averagepercentagesper grain were then multiplied and Egert, 1965; Dedi6s, 1967). These andesitic
by the total opaque-mineralcontent of the rock to strata are warpedinto broad,north-trendingfolds,
give an estimateof the percentageof each opaque metamorphosed to low-grademeta-andesites,andin-
mineral in the rock. trudedby a successionof north-trending
batholiths
of calc-alkaline
composition
(Thomas,1967;Aguirre
Regional geologicsetting and Egert, 1970).
The strike-slipAtacamafault systemparallelsthe
Along the coastof Chile, metasedimentaryrocks, coastof northern Chile for about 1,000 km. Discon-
intruded by granitic plutons,are exposedin eroded tinuous breaks delineate its trend from Coptap6,
anticlines and upthrown fault blocks. The meta- southto E1 Romeral(St. AreandandAllen, 1965).
sedimentaryrocks are phyllites,schists,and quartz- The Romeralore depositslie within an anastomosing
ites, in decreasingorder of abundance. According setof strike-slip
faultsthatmayrepresent
the south-
to radiometric dating by Munizaga (1967) and ern end of the Atacama fault system.
petrologicwork by Gonz•lez-Boronino(1970), the
metasedimentaryrocks of central Chile were de- Local geologicsetting
posited during Carboniferous time, strongly de-
formed and regionally metamorphosed during The magnetitedepositsof E1 Romeralare in a
Permiantime,intrudedby grantticplutonsandtherm- composite sliverof metasedimentaryrocksof Paleo-
ally metamorphosedduring Pertoo-Triassic time, zoic age and andesiticand dioriticrocksof Early
and weakly redeformedand remetamorphosed dur- Cretaceousage. This sliver is about 4 km wide
ing Early Triassic time. The metasedimentary rocks (E-W) andat least8 kmlong(N-S). It isbounded
of E1 Romeral are structurallyand mineralogically on the west by a preore batholithof quartz mon-
similar to thoseof central Chile and appearto have zonite and on the east by a postorebatholithof
a similar history. granodiorite(Fig. 4).
MAGNETITE DEPOSITS OF EL ROMER•IL, CHILE 1105
/+
•./ t + + + +
8 + + +
• + + + +
//I + + + + +l"
•+ + + +
+ + + +,', . .
+ + +
+ ß + +
+ + + + + +
+ + + + + + + •...3• •.•
Romero/d•or/te+ +
. +
+ (weslmoss) + +
+ + + + + .•" feldspath/c
+ + + + [ ....... rOCN ......
+ + + +
+ + +•
+ + + + •,:..',',..dtor•te/.::}•
''•'
•'•:..::::::::'/• '''1
--/2.•_t•.• •.• .... :•:•: ....... •'
+ + ,-
+ + +
+ + + +.. ........ ....+ ¾..
+ + + t:7½::::::•:{-•>,"l•
+ + + /'11 ....................
+ + + :::::::::::::::::::::::::::
+ + +• '•' .•+
+ + + .' .... t . . zI ' ' ß
Plutonic rocks
Kap Kd Postdiorite,preoredikes
hiean 4- S.D. Mean 4- S.D. Kp•nd Kaqd
Plagioclasephenocrysts
Core 3.0 4- 1.1 (An54 4- 5) 35.0 4- 2.9 (An47 4- 4) 14.2 (An64) 25.4 (An28)
Inner rim 9.4 4- 3.5 (An39 4- 7) 8.0 4- 0.7 (An38 4- 4) -- --
Rim 5.0 4- 1.9 (An30 4- 4) 17.8 4- 1.5 (An24 q- 2) 5.8 (An48) 10.3 (An24)
Groundmass 43.7 4- 7.4 (An38 4- 4) -- 37.0 (An64) --
Unzoned -- 4.0 4- 2.3 (An36 4- 6) -- 29.0 (An27)
Secondary -- 1.2 (An27) -- --
--
Range Range
Actinolite 20.4 0-53 25.5 19-30 -- 18.6
Hornblende 6.7 0-24 -- -- 27.3 --
Clinozoisite 3.3 0-6 2.3 0.3-8 -- 4.2
Augite 3.2 0-17 2.0 0-7 8.4 --
Sphene 1.6 0-5 2.0 0.5-3 tr 3.2
Penninite 1.5 0-10 0.1 0-0.5 -- --
Quartz 1.1 0-6 1.9 0-3 -- 9.6
I lmenite 0.3 tr-1 0.1 0-0.4 0.6 --
Magnetite 0.3 0-1 0.1 tr-0.3 1.2 --
Hematite 0.1 0-0.1 tr 0-0.1 0.1 --
Chlorapatite 0.1 0-0.5 0.2 0-0.7 0.5 --
Ruffle tr 0-0.3 tr 0-tr -- --
Biotite -- -- 0.1 0-0.7 5.2 --
Tourmaline -- -- tr O-tr -- --
Pyrite -- -- tr 0-tr -- --
Chalcopyrite -- -- tr 0-tr -- --
Orthoclase .... tr --
"gash" fractures, probably related to left-lateral the nearestnumber in the series 1, 1.5, 2, 3, 5, 7,
movementon the Romeral fault (Figs. 6 and 17). for eachorder of magnitude): Ca 1,000-10,000,Mg
The intramineral dikes are dioritic in composition 2,000-5,000, Ti 1,500-2,000,V 500-1,000, Mn 500-
(Table 2) andapliticin texture (Fig. 16E-l). Their 700, As < 500, Zn < 200, Sb < 100, B 30-50, Ga
primary constituentsare plagioclaselaths (mostly 15-30, Cr < 10-300, Co < 10-50, Pb < 10, Ni 5-20,
0.03-0.3 mm and Ana0_•), disseminatedsubocta- Mo 3-5, Cu 2-5. (The pulverizedand washedmag-
hedral grains of impure titanomagnetite,and nilnor netic separatesfrom unmineralizedsamplesof intra-
relict hornblende. Secondaryconstituentsinclude mineraldiorite aplite appearedto be free of silicate
abundant actinolite and various amounts of clino- material at 60x magnification. The analysts,Sky-
zoisite, chlorite, scapolite, magnetite, pyrite, and line Labs,Inc., considerthe "accepted"valuefor each
chalcopyrite(Fig. 16 E-2). elementto be within ---1 stepof the rangereported
Many samplesof intramineral diorite aplite con- at the 68 percent confidencelevel and within ñ2
tain several types of Fe-Ti-oxide mineral grains stepsat the 95 percentconfidence level). The unit
(Fig. 7). Disseminated suboctahedral grainsof im- cell size of impure titanomagnetitefrom an unmin-
pure titanomagnetite(0.01-0.1 ram) are gray with eralized intramineral dike is 8.408-----0.008A, as
a slightbrownishlavendertint in reflectedlight, and comparedto 8.395 _• for pure magnetite(Lindsley,
they are slightly anisotropic. These grains com- 1965). Data for this determinationwere obtained
monly containtrace, almostsubmicroscopic, hematite from three X-ray diffractometer runs, using Ni-
lamellaealongoctahedralplanes(Fig. 7a), and they filtered CuK• radiation and a corundum internal
more rarely containintergrowthsof ilmenite, pseu- standard.
dobrookite,and/or maghemite(Table 2). Semi- The disseminatedsuboctahedral,impure titano-
quantitativeemissionspectrographic analysesof two magnetitegrains are interpretedto be of magmatic
samplesindicatethe following trace elementcontents origin. By contrast, magnetitethat is very pale
for the impure titanomagnetitegrains (in ppm, to pinkish to brownishgray and completelyisotropic
1108 ,,4RTHUR ,,4. BOOK'STROM'
Plagioclase
Phenocrysts 27.2 (An29) -- -- -- 7.0 (An2 7)
Groundmass 20.5 (An28) 9.9 (An30) 30.2 (An29) $3.9 (An29 + 2) 48.0 (An27)
Secondary .... 3.0 oligoclase
Actinolite 37.9 40.2 49.2 27.2 --
Penninite -- 6.3 0.2 15.9 27.3
Calcite .... 0.2
Epidote 12.5 -- 0.2 -- --
Green biotite -- 6.8 0.2 0.3 7.1
Sphene 0.8 tr -- tr --
Chlorapatite 0.4 -- -- -- tr
Scapolite -- 34.0 -- -- --
Nontronite .... 3.0
Montmorillonite -- tr -- -- --
Ilmenite -- 0.1 tr tr tr
Magnetite 1 0.7 1.7 -- 2.3 3.9
Magnetite 2 .... tr
Magnetite 3 -- 1.1 18.9 0.2 tr
Titanomaghemite-Pseudobrookite -- -- -- 0.1 •
Hematite tr tr -- tr --
Pyrite -- -- 1.4 • --
Chalcopyrite tr -- 0.1 tr tr
Bornite tr ....
the orebody. Intramineral (IM-2) diorite aplite but if remobilization was the dominant process,
dikesare postoreon the westernside of the orebody, then the replacementbodiespresumablywould have
but in its southeastern part they have beenpartially been more evenly distributed, and would have been
replacedby magnetite,actinolite,clinozoisite,scapo~ just as abundant in the late intramineral dikes as in
lite, chlorite, and pyrite (Fig. 16E-2); and in its the earlier ones. However, the replacementbodies
centralpart they are absent,presumablybecausethey are not evenly distributed,and they are much more
have been completelyreplacedthere. Intramineral developedin the earlier dikes than in the later ones.
diorite aplite (IM-3) dikes are predominantlypost- Thus it would appear that postdike hydrothermal
ore at the outer edgesof the orebody,but they have replacementwas the dominantprocessin the forma-
been partially replacedby magnetitein its central tion of veinlets and replacementbodiesin the intra-
part. The late intramineral (IM-4) porphyritic mineral dikes. Assuming this to be true, it would
diorite aplite dikes cut IM-1, IM-2, and IM-3 dikes. appear that replacementfronts shiftedfrom west to
They are predominantly postore in age but are cut east as the Main orebody grew, and then retreated
by sparsemagnetite-actinoliteveinlets,which in turn toward the central part of the orebodyas ore deposi-
are cut by sparser veinlets containing oligoclase, tion waned.
chlorite,calcite,quartz, magnetite,chalcopyrite,chal-The dioritic modal compositionsof the intramin-
cocite, and covellite. eral dikessuggestthat they are geneticallyrelatedto
Two explanationscan be offered for the veinlets the Romeral diorite. Their postdiorite ages and
and replacementdepositsin the intralnineral dikes. their more sodicplagioclasessuggestthat they may
One is that dike eraplacementcausedremobilization representresidual diorite magmas. Chemicalcom-
of ore minerals, which were then redepositedin the positionsfor relativelyunalteredsamplesof La Liga
dike rocks. The other is that the dikes •vere em- andesite porphyry, Romeral diorite, and intramin-
placed during ore depositionand were altered and eral diorite aplite (IM-3), which are given in Table
replacedduring postdikehydrothermalalterationand 3, indicate that the intramineral diorite aplite is en-
ore deposition. Both processes probablyoccurred, richedin iron with respectto both La Liga andesite
M.dGNETITE DEPOSITS OF EL ROMER.dL, CHILE 1111
Fro. 8. Main and North Extension orebodies. This ore-zone map is based on exposuresin
the Main pit in the year 1970, and on assaysof samplesfrom 1,450 10-m percussiondrill holes
on the 320 to 300 rn levels (elevations).
1114 .4RTHUR .4. BOOKSTROM
-
I
,•-- SECTION 11,800 N. N76OE -
o lOOm
I I
ELEV.
Pi! prol'//e (Meter•)
IM dikes
Pzp /'
Ka(a)
1a)• // Pzp
200-
//' I /
/ /
/ /
/I /
/
/
i/I /
/
Kd(ap) I/
v
\
/
Kgd
\
\
Ko(o) /
Kd
\
\
',
/
\ ,,.•
'\
/
--200
\
/ --200-
EXPLANATION
(ap)Aplitized
rock • Rameral
diorite
2-7 (analysts,SkylineLabs, Inc.). Unlike the im- The Main orebodyis penetratedby small veinlets
pure primary titanomagnetite of the intramin- of gangueminerals. The veinletsare discontinuous
eral dikes, the relatively pure ore magnetite is and are offset by other veinlets and fractures, indi-
optically isotropic and is very pale pinkish to cating recurrentfracturing accompanied by gangue-
brownishgray in reflectedlight. Its unit cell size, mineraldeposition.In generalthe followinggangue-
as determinedby X-ray diffraction,is 8.398-----0.007 mineralsequence canbe observed:
A, which is very close to that of pure magnetite 1. actinoliticveinlets(with magnetite,clinozoisite,
(8.395 A, according
to Lindsley,1965). apatite, and pyrite),
M/IGNETITE DEPOSITSOF EL ROJlEt•,4L,CHILE 1115
d.
FIG. 10. Textures and microtextures typical of the Romeral magnetite ores in reflected light:
a. Intergrowth o{ actinolite and poikiloblastic,skeletal magnetite crystals in actinolitized rock,
high-grade zone in western Main pit.
b. Magnetite ore cut by actinolite veinlets, which are in turn cut by a calcite veinlet and a
barren {racture, east Main pit.
c. Magnetite veinlets cutting actinolitized andesite porphyry, eastern margin of Main orebody.
d. Capillary magnetiteveinletsin actinolitizedrock, high-grade zone in western Main pit.
1116 ARTHURA. BOOKSTROM
o loom
I •
13•000N
'111111
I /
t
ßf../'-
/• q,....... I.
ß .. :J:
.! o½•(it?o/sYe,,,.f. -•..1•..
ß •''1'' f' ' •' '
•//• c• • •..... ,..•..
i•k .......
::•:::::•7::::::•:::::::
I..... el I ..... • .......
• ........ i ..... • M) ..... j % .......
:::::::::::::::::::::::::::::::::
........... •.... i•.•...,,. •fj .....
.............. f•to•
ß)• ...... •..•..)..t•, 'tl ...........
. i/r• ....... i ....... I..•c•I .....
•f•-•.,•-•::::::
•-•u•r•- • ......
:y ;•:•'•;•:: :::
rl' ' '•'•zl ......
ß . • .......... I...• ......
•M/c•echne.,. ........ • ........... .
ß• ..... • .....
ß • .... • • .... i ........ i. ............ ß
•/1' ' • ' ' '•1)• ..... I ....... • ............. V
• •....... • • ...... i ..... I...• ..........
• ....... • I• ...... • ..... r" •jL ......... :
:::::::::::::::::::::::::::::::::::::::
/--• ...... I•1 • ....... •' •' •' ''l/t ...........
•1 • ....... • • .' ...... '. 'l•l ' • ' ' '• ............
• • ..... • / •1 ....... 4•'i'l' ß '• '• ............ .
ß .......
:ty'•/•/•l:::::: ' '
.....
' •tl/ •...• /,k/• I1......
..4
ß•' I
.......
...• -• -•/ / •.
i •/ /I
Fro. 16. Textures and mineral assemblagesin sometypically altered rocks from the Romeral
area. Photos A through E, G, and H, are photomicrographstaken with transmitted light
with nicols at 80 ø. Photo F was taken in the field.
A. Dioritized quartzite. A relict mosaic of quartz grains (q) comprises17 volume percent
of the rock. This mosaic is disruptedby 70 percent of plagioclaselaths (p). Eight volume
percent of hornblende(h) is interstitial, and is associatedwith 2 volume percent of very fine
grained magnetite. Many of the plagioclaselaths have polysyntheticallytwinned cores (An30)
and irregular, untwinned rims (An20). The cores of some plagioclasegrains are altered to
very fine grained epidoteand montmorillonite(e-m).
B. Felty actinolite after phyllite. This is the sample of actinolite that was analyzed by
microprobe. The opaque material is magnetite and marrite. Minor chlorite is present along
the veinlet.
C. Actinolite-scapoliterock from 300 m southeastof the Main orebody. Actinolite (act) is
in randomly oriented prisms. Scapolite (s) is in fine-grained, light gray patches. Magnetite
veinlets and pods are black. Tourmaline (t) is intergrown with actinolite and magnetite in
the lower left part of the sample. An epidote (e) vein (about 0.6 mm thick) cuts the magnetite
(mt) veinlets. It contains minor magnetite, but the large black area is an open space.
D. Spheneporphyroblasts(sp) in a martix of randomly oriented actinolite prisms (act) and
magnetite (rot). From the actinolite sphene zone along the southeast margin of the Main
orebody.
E. Textures and mineral assemblagesof relatively unaltered versus strongly mineralized
intramineral dikes. Both samples are from the southeasternpart of the Main orebody at the
310-m elevation. Transmitted light, nicols at 80ø. 1. Texture of relatively unaltered intra-
mineral (IM-3) diorite aplite. Light gray lath-shaped grains are plagioclase (p) (An•_+•).
Darker gray prisms and interstitial material is actinolite (a). Black grains are magnetite (m),
most of which probably is magmatic. 2. Intramineral (IM-2) diorite aplite, largely replaced
by magnetite (m) and clinozoisite(cz).
F. White, oligoclase-rich rock with remnants of dark green, actinolitized phyllite from north
of the North orebody. The oligoclase rock is cut by a small actinolite-magnetite-apatite vein.
The age relationshipsshown here are consistentwith those seen west of the Main orebogy.
There aplitizedrockscontainremnantsof actinolitizeddiorite,and the aplitizedrocksare.,.cut
by large, widely scattered actinolite-magnetite-apatiteveins.
G. Gradation from Romeral diorite (left), containingplagioclase(pl), actinolite (act), and
nmgnetite (mt), to aplitic quartzofeldspathicrock (right), with secondary plagioclase (p2)
and quartz (q). This sample is from a 3 cm aplitized selvage along a thin veinlet containing
minor magnetite, biotite, tourmaline, quartz, and clinozoisite.
H. Completelyaplitized rock (after Romeral diorite). The rock consistsof plagioclase(An8)
(p), quartz (q), and minor tourmaline (t). Myrmekitic texture (m) is present in the lower
central and right-hand portions of the photo.
1124 ARTHUR ,4. BOOKSTROM
o ,, ß ß : o ø
o
m /, '.. ". •
--t4,000 N ( m. ,.
/! '":. •,
J
K• I
• N
./
: '-.
"'::...•
i:• •' ".... 0I , t I , 500
I m.
-- I•,000 N. 5•.
,' •: •• •
/, X
7_:• ) •kt Ut ;D
i ' / : •'•
/ ,') • • .............
/ /I ,'/ • [
'" I I I i I :'
/ /
/ / .
MAGNETITE DEPOSITS OF EL ROMERAL, CHILE 1127
• • • .• .. * -. ,
ing slickensides
on the fault in the North pit, how- Gently plunging slickensideson the fault indicate
ever, indicate that the latest movement on the north- strike-slipmovement,and severaloffset featuresin-
ern part of the fault was strike-slip. Part of the dicateabout200-----50m of right-lateralmovement.
right-lateral movement of the late North-northeast The North orebodyapparentlyhas beendisplaced
fault evidentlywas shuntedonto the northern part about200 m with respectto the North Extensionore-
of the East fault. body. Similarly, the early Romeral fault has been
Late North-northeastlault cut and displacedby about200 m in the right-
lateral sense.
The youngestmajor fault in the Romeralsystem As shownin Figure 17, severalintramineral dikes
is the postoreNorth-northeastfault, which lies west are offset10 to 50 m in a fight-lateralsensealong
of the Main orebody(Fig. 17). This fault cutsthe subsidiaryfaultsthatprobablyare relatedto themain
early Romeral fault, cuts the northern end of the right-lateral North-northeastfault.
North Extensionorebody,and lies southeastof the
North orebody. Its zone of brecciaand gougeis Regionallault pattern
10 to 50 m thick and containsfragmentsof magnetite The Romeralfault systemis oneof a discontinuous
and white argillizedrock in a dark gray matrix of series of breaks delineatingthe southern300 km
pulverized magnetiteand hematite. of the Atacamafault system,which parallelsthe
17. The Romeralfault system,showingthe late North-northeast
fault, the late faultsin
the Main orebody,the East fault, andthe early Romeralfault.
1 128 ,4RTHUR ,4. BOOKSTRO•],•
Chilean coast for about 1,000 kin. Several {dher no explosion breccias to indicate that it exceeded
major magnetite depositsof early Cretaceousage, confining pressure and rock strength. High fluid
such as E1 Tofo, E1 Algarrobo, and Cerro Imam, pressuresmay have triggered movementon the P,.o-
are located near north-trending shear zones and meral fault, but it is unlikely that they were the
faults of the Atacama system. At Cerro Imim, 300 ultimate cause of that movement, inasmuch as the
km north of E1 P,.olneral,the north-trending mag- fault is a regionalfeature. Fluid pressuremay have
netite depositis cut by en echelon,northwest-trend- beenlessthan lithostaticin the late magnetite-actino-
•ng dikes. If thesedikes,like the intramineraldikes lite-apatiteveinsthat apparentlyfilled openfractures,
of E1 P,.omeral,filled tensional fractures related to but in the tight veinletsand replacementdepositsof
strike-slipfaulting, their orientations•vould suggest the orebodiesit seemslikely that total pressureswere
early Cretaceous left-lateral movement. Postore about lithostatic,or approximately2 kb.
movementson the Atacama fault system,however,
apparentlyhave been right-lateral. At E1 P,.omeral Temperatures
there has been 200-+ 50 m of postore right-lateral
movement, and in northern Chile the Atacama Magnetite ore depositionat E1 P,.omeralwas pre-
fault is active, with right-lateral strealnoffsetsof as cededby formation of an amphibolitefaciesassem-
lnuchas 1.5 kin (St. Areandand Allen, 1965). blageconsistingof hornblende,secondaryplagioclase
(Anao-ao),and diopside. It was accompaniedby
Boundary Conditions for Ore Deposition formation of a transitional amphibolite-greenschist
and Alteration facies assemblageconsistingof actinolite, secondary
plagioclase(An27_+.•),clinozoisite,scapolite,sphene,
Depth and pressure
and apatite. It was followed by formation of a
The Romeral magnetite deposits probably are greenschistfaciesassemblage consistingof chlorite,
middle Xeocomianin age. They are youngerthan secondaryplagioclase(Ans-2s), epidote,quartz, and
La Liga andesiteporphyry, which is thought to sphene. Thus the ore-associatedmineral assem-
representa magmabody that fed andesitesills and blagesof E1 Romeral record a generally retrograde
flowsof the early NeocomianArquerosFormation, transition from amphibolite to amphibolite-green-
but they are olderthan the Punta de Piedra batho- schist,to greenschistfacies conditions.
lith, which is late Neocomianin age. Experimental studies of equilibrium phase rela-
The tops of the P,.omeralmagnetitedepositslay tions between amphiboliteand greenschistfacies
at about the level of the Paleozoic-Jurassicuncon- mineralsby Liou et al. (1974) indicatethat in basal-
formity. They probablyformedunder a coverof tic rocksthe assemblage actinolite4- oligoclase-----
epi-
Jurassicvolcaniclasticandesitesand early Neo- dote is stable at temperaturesbetween 550ø and
cornJan
andesitesillsand flows. No Jurassicsections 475øC,at 2 kb, and at oxygenfugacitiesin the mag-
have been measurednear E1 Romeral, where only netite stabilityfield and approximatingthosedefined
minor erosional renmants of the Jurassic section by the NXO buffer assemblage.\¾ithin this 550ø
remain, but an east-westcrosssectionat 26ø S by to 475ø temperaturerange the anorthite contentof
Ruiz et al. (1965) showsthe full westerneugeo- plagioclaseand the A1 content of actinolite vary
synclinal,island-arcfaciesto be 9,000 m thick and directly with temperature. Above 550ø actinolite
the easternmiogeosynclinal limestonefacies to be convertsto hornblende,oligoclaseconvertsto ande-
2,000 m thick. An intermediatethicknessof 5,500 sine, and the resulting assemblage,which may in-
m is suggested asa reasonable estimate
for thethick- clude clinopyroxene,is typical of the amphibolite
nessof the Jurassic sectionat E1 Romeralwhenthe facies. Below 475ø oligoclaseconvertsto albite, and
orebodiesformed. Neocomian Arqueros andesites the resulting greenschistfacies typically includes
andeastof E1 P,.omeraI,chlorite.
are presentto the southeast
wherethey rangein thickness from 850 to 1,230m Although the Romeral ore depositsformed by re-
(AguirreandEgert,1965). The Arquerossection placementof andesite,phyllite,and schistin an open
graduallythickens andthereforeit is esti- system, rather than by isochemicalalteration of
westward,
mated that the Arqueros andesiteswere at least basalt in a closedsystem,the mineral assemblages
1,250 m thick at E1 Romeral. Thus the Romeral of E1 P,.omeralare remarkably similar to thosepro-
magnetitedepositsprobablyformedat a depthof ducedexperimentally by Liou et al. (1974). Further-
approximately
6,750m, or about7 km. more, it is inferred that the Romeral deposits,like
Assumingthat pressurewasnearlylithostatic,this the experimentalassemblages, formed at about 2 kb
would suggestthat ore depositionoccurredat ap- pressure. Thus, even though the relatively alumi-
proximately
2 kb pressure.Fluidpressure mayhave nous compositionsof the Romeral host rocks may
beensomewhat greaterthanlithostatic,but thereare have tendedto lower the temperatureof the horn-
M.4GNETITE DEPOSITS OF EL ROMER./iL,CHILE 1129
blende-actinolite conversion somewhat it seems km, at about 2 kb pressure, and at temperatures
reasonableto suggestthat the Romeral magnetite mainly in the range 550ø to 475øC.
depositsand their accompanyingtransitionalamphib- 10. Left-lateral movement on the Romeral fault
olite-greenschistfacies minerals formed at tem- resulted in the formation of schistose,breccialike,
peraturesmostly in the range 550ø to 475øC. The and dikelike magnetite-actinolite-apatitecataclasites.
preoreamphibolitefaciesassemblage probablyformed 11. Ore depositionwas followed by emplacement
at temperaturessomewhatabove550ø, and the post- of a few minor albitite and quartz-albitepegmatite,
ore greenschistfacies assemblageprobably formed aplite,and granitedikes;by aplitizationof previously
at temperaturessomewhatbelow475ø. actinolitizedRomeral diorite and phyllite; by em-
placementof many biotitic diorite aplite dikes; by
Summary biotitization; by formation of actinolite-magnetite-
apatite veins; by chloritization; by argillization,
1. La Liga andesiteporphyry intruded Paleozoic martitization, and sericitization; by right-lateral
metasedimentaryrocks and Jurassic volcaniclastic movementon the late Xorth-northeastfault; and by
andesites.
emplacementof the postore granodiorite batholith
2. Romeral diorite intruded La Liga andesitepor- east of E1 Romeral.
phyry. It containsthe samesuiteof mineralsas the
andesiteporphyrybut is phaneriticand is somewhat Acknowledgments
depletedin iron and enrichedin silica with respect
to the andesiteporphyry. I am grateful to Drs. S. R. Wallace and G. L.
3. The Romeral magnetitedepositsare adjacent Hole, who arranged support for this project with
and subparallelto the eastern contactof the west Bethlehem Steel Corp., and I would also like to
massof the Romeral diorite pluton. thank F. Franquesa,F. Mufioz, and J. Johnsonfor
4. Ore depositionwas precededby crystallization logisticalsupportin Chile.
of the Romeral diorite; by emplacementof post- For their guidance,encouragement, criticismand
diorite, preore dikes, someof which are more mafic comment,I am indebtedto Drs. R. H. Jahns, C. F.
than Romeral diorite, and some of which are more Park, Jr., F. W. Dickson,C. O. Hutton, B. M. Page,
silicic; by formation of amphibolitefacies alteration \V. C. Luth, and J. G. Liou, all of the schoolof
minerals near diorite contacts and in the reentrant Earth Sciences,Stanford University.
or roof pendant between the eastern and western R. Laniz took the photomicrographs, R. Dahl did
massesof the diorite pluton; and by mobilizationof muchof the final drafting,and R. Foster typedthe
iron in hematite-bandedquartzites near the diorite final draft of the paper.
contactand redepositionof that iron as hematite80 Thanks also are due to Drs. \Vm. Kelly, E. Es-
to 145 m from the contact. sene, and B. J. Skinner, for critically reviewing
5. Ore depositionwas accompanied by left-lateral various versions of this manuscript. Any errors,
strike-slip movementon the Romeral fault, and by omissions,or misinterpretationsthat remain are the
eraplacementof four generations of intramineral responsibilityof the author.
diorite aplite dikesinto gashfractures,which opened CLIMAX MOLYBDENUM COMPANY
in responseto the left-lateralfaulting. DIVISION OFAMAX, INc.
6. The repeatederaplacement of intramineraldikes WESTERN DIVISION OFFICE
suggeststhat a bodyof dioritic magmaexistedbelow 13949 WEST COLF^X AVENUE
the Romeral magnetitedepositsas they formed. GOLr)EN,COZOR^r)O 80401
7. The primary compositions of intramineraldikes May 6, Septembcr24, 1976
suggestthat they representan iron-enriched,water-
enrichedresidualdiorite magma. REFERENCES
8. The pattern of mineralizationof the intramin-
eral dikes and the distributionsof apatite and pyrite Aguirre L., Luis, and Egert R., Earnesto, 1965, Cuadr/tngulo
Quebrada Marquesa: Chile, Inst. Inv. Geol. Carta 15.
suggestthat ore depositionbegan on the western [ Santiago]
side of the Main orebody,near the Romeral fault, - 1970, Cuadr•tnguloLambert (La Serena), Provincia de
Coquimbo: Chile, Inst. Inv. Geol. Carta 23. [Santiago]
shifted eastwardand upward as the orebodygrew, Bookstrom, A. A., 1975, Magnetite depositsof E1 Romeral,
and then retreated toward the central and lower Chile: physical geology, sequenceof events, and prozesses
parts of the orebodyas ore depositionwaned. of formation: Standard Univ. Ph.D. thesis [Ann Arbor,
Mich., Xerox Univ. Microfilms, Order No. 75-21, 848],
9. Magnetite ore depositionwas accompaniedby 431 p.
formation of actinolite, secondary plagioclase Burnham, C. W., 1967, Hydrothermal fluids at the magmatic
stage, in Barnes, H. L., ed., Geochemistryof hydrothermal
(An27__5),clinozoisite,scapolite,sphene,and apatite. ore deposits: New York, Holt, Rinehart and Winston,
It probablyoccurredat a depthof approximately7 Inc., p. 34-76.
1130 ARTHUR .4. BOOKSTROM
Casey, Raymond, 1964, The Cretaceousperiod, in Harland, studies of the phase relations between greenschist and
W. B., Smith, A. G., and Wilcock, B., eds., The Phanero- amphibolitein a basalticsystem: Am. Jour. Sci., v. 274,
zoic time-scale: Geol. Soc. London (•uart. Jour., v. 120 p. 613-632.
(supp.), p. 193-202. Melo, O., 1965, Romeral geologic survey: Bethlehem Chile
Chayes,F., 1956, Petrographicmodal analysis: New York, Iron Mines Co., unpublishedmap and report, 27 p.
John Wiley and Sons, Inc., 113 p. Munizaga, F., 1967, Geochronologiade los esquistosdel Sur
Dedi6s A., Pedro, 1967, Cuadr/tnguloVicufia: Chile, Inst. del Chile: Chile, ResumdnJour. Trab. Inst. Inv. Geol.,
Inv. Geol. Carta 16. [Santiago] X Anniv. [Santiago]
Deer, W. A., Howie, R. A., and Zussman,J., 1967, An Park, C. F., 1972, The iron ore depositsof the Pacific Basin:
introduction to the rock-forming minerals: New York, Ecoa. GEor•.,v. 67, p. 339-349.
John Wiley and Sons, Inc., 528 p. Ruiz F., Carlos, and Corvalfi.n, J., 1968, Mapa geo16gicode
Ford, W. E., 1966, Dana's textbook of mineralogy, 4th ed.: Chile (escala 1: 1,000,000): Chile, Inst. Inv. Geol.
New York, John Wiley and Sons, Inc., 851 p. [Santiago]
Garcia, F., 1964, Geologia del Norte Grande: Chile Em- Ruiz F., Carlos, Aguirre, L., Corval/tn,J., Klohn, C., Klohn,
pressaNacional del Petroleo, informe inedito. [Santiago] E., and Levi, B., 1965, Geologla y yacimientos metallferos
Gonz•des-Boronino, F., 1970, Series metamorphicasdel basa- de Chile: Chile, Inst. Inv. Geol. [Santiago]
mento cristalino de la Cordillera de la Costa, Chile central:Ruiz F., Carlos, Ortiz, F., Moraga, A., and Aguilar, A.,
Univ. Chile Pub. 37. [Santiago] 1968, Genesisof the Chilean iron ore depositsof Mesozoic
Heinrich, E. W., 1965, Microscopic identification of minerals: age: Internat. Geol. Cong., 23rd, Czechoslovakia 1968,
New York, McGraxv-Hill Book Co., 414 p. Rept., sec. 7, Proc., p. 328-338.
Holland, H. D., 1972, Granites, solutions,and base metal St. Amand, Pierre, and Allen, C. R., 1965, Strike-slip fault-
deposits: Ecox. GEor..,v. 67, p. 281-302. ing in northern Chile [abs.]: Geol. Soc. America Bull., v.
Hutton, C. O., 1971, Curves for determinationof composi- 71, p. 1965.
tions and structural states of plagioclasefeldspars by the Scholl, D. W., yon Huene, R., and RidIon, J. B., 1968,
Federov method: Stanford Univ., unpublishedcourse ma- Spreading of the ocean floor: undeformed sedimentsin the
terial. Peru-Chile trench: Science, v. 159, p. 869-871.
James,David E., 1971, Plate tectonicmodel for the evolu- Thomas B., Herbert, 1967, Geologla de la Hoja Ovalle,
tion of the central Andes: Geol. Soc. America Bull. v. 82, Provincia de Coquimbo: Chile, Inst. Inv. Geol. Bol. 23.
p. 3325-3346. [ Santiago]
Lindsley,D. H., 1965, Iron-titanium oxides: Carnegie Inst. Uytenbogaardt, W., and Burke, E. A. J., 1971, Tables for
Washington Year Book 64, p. 144-148. microscopic identification of ()re minerals: Amsterdam,
Liou, J. B., Kunioshi, S., and Ito, K., 1974, Experimental Elsevier Publishing Co., 430 p.