ECONOMIC

Vo.. XXVI.

GEOLOGY
MAY, I93I No. 3

THE TIN

ORES OF UNCIA-LLALLAGUA,
FRIEDRICH AHLFELD.

BOLIVIAY

INTRODUCTION.

THE town of Uncia lies at an elevationof 3,815 meters,about

lOO km. southeastof Oruro, in the heart of the Bolivian Eastern

Andes. Its latitude is 18 27' 58" South and its longitude 60 37' 41" west of Greenwich. The ore deposits are in the very centerof the Bolivian metalliferous province. Toward the north and towardthe souththe intensity of mineralization gradually decreases. Uncia lies at the junction of the N.-S. axis of
the Andes of Southern Bolivia and the NW.-SE. axis of the

Andes of northern Bolivia, that is, Uncia lies at the bend in the

range. The intensive opening up of the sedimentary shellat this

bendis apparently the cause of the intrusion of unusually large quantities of igneous rock. It is the reasonfor the striking widening of the rangeand the numerous igneous intrusions in the EasternAndesbetween latitudes18 S. and 19 S. The bendin the axis of the rangeis alsothe cause of the abundance of ore deposits 'withinthis area--Oruro, Negro Pabellon, Morococala, Poopo,Avicaya,Antequera, Huanuni,Challapacheta, the Pujru-VilaApacheta group,Colquechaca, andmanyothers.
ROCK FORMATIONS.

Paleozoic,probablyDevonian,unfossiliferous sandstones,
quartzites and slateswith NW.-SE. strike form an anticline west
Translated from German manuscript by Joseph T. Singewald, Jr. 16 241

242

FRIEDRICH

alHLFELD,

of Uncia in the crest of which the little stock 2 occurs that contains

the world's largest tin ore deposit. Immediately east o.f Uncia
a small mass of Cretaceous Puca sandstone is folded into the older

sediments. In addition to the Uncia stock,volcanicrocks are represented by two small rhyo-andesite occurrences northeastof Uncia and extensivetuffaceousrhyolitic flows that extend from Negro Pabellon to a little westof Catavi. They are younger than the ore deposition. Northeastof Uncia and at Catavi are hot springs. Lindgren hasdescribed the latter. They havedeposited largesinterterraces consisting chieflyof aragonite,but includesiliceous portionsthat containopaland dark portionscarryingmanganese hydroxidegel with an appreciable WO content. These springsare evidence of the very young--probably Pliocene--ageof the ore deposition. The Uncia igneousintrusion has the form of an ellipsewith

axes 2,ooo m..longparallel to thestrikeof therange andx,7oo m.

long at right angles thereto. The mine workingsindicateit extendsverticalin depth. It is a singleintrusion. The rock is so deeplyaltereddown to the lowestmine workings (775 m. below the surface) that it cannot be classified with certainty. Feldspars are completely decomposed and are only recognizable throughtheir

outlines. Biotite'has alteredto aggregates of minute sericite

scales. Only the frequently crusheddihexahedralquartzesare unchanged. I am indebted to Dr. Brendlerof Hamburg for an analysis of a specimen from the Siglo XX level. It is compared
2 Throughout the manuscript the author uses the term "Lakkolith"
U'ncia intrusion and other similar bodies of igneous rock in the Bolivian

for the
Andes.

These intrusions do not conform to the generally accepted shape and geologic relations of a laccolith. On questioning the author on this point he suggested using Steinmann's term "stehende Lakkolith," and showed by a sketch an intrusive mass suddenly tapering at the bottom into a small feeder such as usually shown in
an ideal section of a laccolith. The main mass intrudes the strata and is not forced

between them, and the vertical extent is much greater than the horizontal. A translation of the term "stehende Lakkolith" would give so misleading a concept to the English reader that it appears to me better to use the word "stock," the reader keeping in mind that the stock is believed by the author to diminish in

cross-section quite abruptly in depth and not maintain or even increaseits diameter
with increasing depth. In support of this idea the author states that the mine workings at Potosi have clearly demonstrated such a shape for that intrusion.-J. T. S. Jr.

TIN

ORES OF UNCIA-LLALLAGUA,

BOLIVIA.

243

belowwith an analysis of a fresh dacitefrom Oruro given by

Kittl. 8

Llallagua (Brendler)
Molecular Per Cent

Oruro (KittO
Molecular Per Cent
73.422 9.397

Analysis
SiO ............... TiO .............. AIO3 .............. FeO .............
FeO ............... CaO ...............

Analysis
64-73
14.o4

67.75
0.24

75.367
o.2ol

18.45
0.95

12.1o8

2.04

0.84
1.39

0.395 1.911 1.oo6 2.311

0.63
2-33 4-03 1.52 2.43

0.2665 e.148
4.911

MgO ..............
KO ............... NaO .............. PO ...............
MnO ..............

5.67
2.46

4.038
.663

2.578 1.765
5.1o9 0.259 o.o75

4.63
0.54
0.08

-- HO ............

0.32

o.31

+ HO ............

3.53

For the Oruro rock the valuesof A, C, and F after Osann are'
A (alkali feldspar) ............................. C (lime feldspar) .............................. F (dark constituents) ........................... 6.874 mois. 4.9IX " 4.992 "

According to the terminology of Osannthe rock is a dacite. Because of its great alterationthe Llallagua rock can not be definitely determined. It differsfrom the Oruro rockespecially
in its lower lime content. On this. basis it would have to be

classed with the quartz porphyries. Alteration, however,commonly results in a decreasein lime and an increasein alumina. 4

It is very likely,therefore, that the freshLlallaguarockis richerin

lime and similar to the dacite of Oruro and other Bolivian localities.

The Llallagua rockhasbeen altered not onlythrough weathering but alsoby pneumatolysis. Pneumatolytic alteration--evidenced by the occurrence of tourmaline--is restrictedto certain

portions, especially to the vicinityof the veinsandcountry rock

a Erwin Kittl, "Analisis de Roeas eruptivas de Bolivia," Revista Minera de Bolivia, I927, Afio 2, p. 5I.

4Waldemar Lindgren, "Replacementin the Tin-bearing Veins of Caracoles,

Bolivia," EcoN. Gv. ot.., vol. 2i, p. 142, i926.

244

FRIEDRICH

AHLFELD.

inclusions. Such rock has a dark color caused by minute microscopictourmaline needles. The pneumatolyticalteration is so complete in places that almostnothingis left of the original rock. Thin sections show much secondary quartz and pseudomorphs
after mica and hornblende.

Although the rock often has a tuffaceousappearance in thin section, physical-chemical considerations suchas the high temperature of ore formation lead me to the conclusion that the intrusion

is not a neck but a stock which had a thin sedimentaryshell. That no tuffs or glasses have beenfound pointsto the sameconclusion.

The contactaction of the magma on the intruded rock varies greatlyat the differentexposures but is generallyslight. On the 600 m. level at Uncia, the countryrock of the easterncontactis changedto a hard hornfels, indicatingthat the intensity of the contactaction increases with depth. The condensation of the gases and solutions that carriedthe orestook placemainly within the stock. Only a few veins,the contactvein for example,extend for a shortdistance into the countryrock. This is dueto the great absorptive capacityof the porousquartz porphyry for the solutions. They condensed mainlynearthe apexof the intrusion.
THE VEINS.

There are two. types of veins. One type are fillings of true faults with occasionallyconsiderable displacement. They are persistent and regular in. strike and dip. To this group belongs the importantSan Fermin-San Josdvein (Fig. ) which lies near the westcontact and dipsto ihe east. It is the real "trunk vein" and is joined in depthby a number of smallerveins. Other are Forastera, Contacto,Reggis-Inka, and Salvadora. They are up to m. in width, and the San Fermin locally,even 8 m. They have well-definedwalls and commonlya drusy and brecciated filling. The strike ranges between N. 3o E. and N. 50 E. Between the veinsof the preceding type are countless smaller veins representing replacement of the wall rock along minute fractures. They are generallyparallel to. the large veins, but

TIN ORES OF UNCL4-LLALLAGUA,

BOLIVIA.

245

some strike at right angles to them (cruzeras). They are of limited extent along the strike and are characterizedby irregularity and non-persistence in their ore content. Their usualwidth is 5 to 30 cm. They become leanerin depth. The filling is compact and drusesare rare. A characteristic phenomenon is that someof the large veinsdo

Fro. x. Sketch of the Uncia Llallagua intrusion with the veins and faults. The broken line is the boundary of the stock on the surface.

not extend to the surface. An explanationof this will be suggestedfurther on. In Uncia-Llallagua9 main veins and about 3 smallveinshavebeenworked (Fig. ).

246

FRIEDRICH

AHLFELD.

Major tectonic movements occurred withinthe stock subsequent

to ore deposition. The northeasternportion with the veins Carnevalitoand Nueva sank along the Stanton and Diaz faults The faults are barren of ore and filled only with clayeycomminuted material.
PARAGENESIS OF MINERALS.

The mineralfillingall the veinsis quitesimilardespite many variations. The followingminerals were recognized which are described in paragenetic sequence (from near magmatic to more
remote magmaticaffiliations: Hypogene Minerals.

Tourmaline.--Abundant in the countryrock of the veinsand in certainbarrenshearzones. Occurs irregularlydistributed in quartzin smallmicroscopic needles and shows'weak pleochroism (Fig. 2). 5 Almostneverabsent whencassiterite occurs, but de-

creases in 'quantity upwards. The greisen accompanying the

SanFerminveinon the deepest levelof the mine (775 m. below the surface)consists onlyof tourmaline, quartz,anda little cassiterite. Mica and topaz are absent. In part the tourmalineresembles achroite and is withoutpleochroism. Quartz.--Associated with the tourmaline is muchfine-grained quartz as a pneumatolytic alteration of the country rock. Because of thisthesilica content of some of therockis veryhigh. The frequentlywell-developed trapezohedral crystalsaccompanying cassiterite in druses are high-temperature quartz formed under56o. It is contemporaneous with the cassiterite or older. .Theabundant chalcedony is probably supergene. Cassiterite.--Occurs almostexclusively in twins of the Zinn-

waldtype. They rarelyattainto a sizeof x cm.,and are generallymuchsmaller.Transparent crystals are rare; they are
commonly black. An analysisby G. Greene of a cassiterite crystalyielded 94.o6 per centSnO2and 4.3o per centFe. The
5 The microphotographs of this paper were made with the kind aid 'of Mr.
Ehrenberg.

0 G. Greene, "Solubility of Tin Minerals,"Eng. and Min. ]our., vol. x22, p. 47,
026.

TIN

ORES

OF UNCII-LL/-1LL.dGU.4,

BOLIVL4.

247

crystalsare made up of zonal layers of various degrees of color intensity. The mostabundantform of twin is after (o), and is usuallyin polysynthetic-cyclic aggregates. In brecciated veins, especially in the San Fermin and Salvadora,cassiterite crystals

, .,'5:''

.., :

. '..

. .

.-i

. '.:"

. .,'., ?

Fx. .

F.

3-

Fx. . Tourma]ine-greisen, country rock. San Fermin vein, Siglo XX level. Thin sealion, X 74Fx. 3- Cassiterite crystal (high relief) surroundedby stannite with unmixing partic]es of chalcopyrite. The cassiterite crystal is strongly corroded. San Fermin vein, 330 m. below surface. Polished section,
X47.

fillcavities between quartz crystals and coat country rk fragments.

n the largeore shoots, as thoseof the San Fermin vein, cassiterite is the principalconstituent of the ore and often curs as-

sociated with wolframite, bismuthinite, and pyrrhotite, lessfrequently with stannite andapatite, withoutgangue minera]s. In oneof my specimens from the 6o m. levelof the Salvadora shaft,in addition to the usualtwinned cassiterite, are spherical
aggregatesof faintly colored needles. inkelmann describes

248.

FRIEDRICH

AHLFELD.

cassiterite with gel structur.e in rhyolite from Uncia. It would appear that cassiteritepresumablyformed from sols is more
abundant near the surface. I have not observed such in the

deeper portionsof the deposit. .4rsenopyrite.--Occurs isolatedin depth,as in the deepest level of the San Fermin, in veinletsin greisen,apparentlyof pneumatolytic origin. Since it doesnot occur in association with other minerals,its genetic positioncan not be definitelydetermined. 14/olframite.--Wolframite of normal composition frequently accompanies cassiterite, especially on the Uncia side. Occasionally its crystalsare coveredby cassiterite crystals. Manganeserich varieties (hiibnerite) are formed under hydrothermalconditions.

Bismuthinite.--Commonly accompanies the cassiterite,especiallyin the Llallaguasection, whereit is foundin nearlyall veins. Its lancet-like crystals,attaining a length of to cm., never show free terminations but are alwaysembedded in cassiterite or pyrrhotite. Occasionally its crystals are coatedwith cassiterite. In someveins the bismuthinite contentvaries proportionally with
that of cassiterite.

Acicularbismuthinite, similarto jamesonite, was seenin druses in arsenopyrite.

Native Bismuth.--Occasionally abundant andin masses weighing up to several kilograms (San Fermin vein). It is neverin crystalform, but together with bismuthinite constitutes the filling between quartzcrystals. Sinceit occurs to a depthof 4oo m., I do not agreewith Kittl 7 in ascribing its formation from bismuthinite through a process of reduction in thezone cfoxidation. I believe thatit wasdeposited directly fromsolutions under pneumatolytic conditions and that the predominance of nativehismuch nearthe surface and of bismuthinite in depth,as frequently observed in Bolivia, is a phenomenon of primary difference in depth. Pyrrhotite is the most important mineral associated with cassiterite in most' of theveins. It forms massive compact masses
in which cassitrite,bismuthiniteand wolframite are embedded
7 Kittl, "Algunas observaciones sobre la formation de Bismuto nativo," Rev. Minera, Soc.Argentinade Min. y Geol.Afio 2, 93o, PP. -3-

TIN

ORES

OF UNCL4-LL.4LL.4GU.4,

BOLIVIA.

249

frequently with idiomorphic boundaries.The abundance of pyrrhotitehasescaped the attentionof mostobservers because it has been almost entirely convertedto pyrite and marcasite. Only occasionally, especially within the sediments, doesone encounter fresh remnantsor remnantsin process of alteration. .4patite is the only hypogenephosphate that occursin appreciablequantity, even though mainly microscopic. It is confinedto a few veins (San Fermin). Occasionally it is found in
drusesas rose-colored tabular crystals,but its usual mode of occurrenceis in blue-greenmassesintergrown with cassiteriteand pyrrhotite. It is of pneumato.lytic origin. Monazite.--This is also a pneumatolytic mineral which was found onceon the 60 m. levelof the Salvadoravein as small,light rose-colored crystalsintergrown with cassiterite and nacrite. Sta.nnite.--Stanniteis more commonthan generallyrecognized

butis confined to certain veins. As a mineral of thehigh-tem-"

peraturehydrothermalphase,it is more abundantnearer the surface. Accordingto the former superintendent of the Uncia Mine, Engineer Hans Block, someparallelstringersof the Salvadora vein consisted almostsolelyof stannite. The mineral was found in suchquantitynear the surfaceat Uncia that the erectionof a silverleaching plant was seriously considered in t9o5. The mineral .isespecially abundant microscopically in the oresof the Polo

vorin and San Fermin veins. In the latter vein it is in part at leastolder than bismuthinite but alwaysyoungerthan cassiterite, the strongly corroded crystals of whichit encloses (Fig. 3). It includes unmixing particles of chalcopyrite. The embedding of cassiterite crystalsin stannite,which is so frequentlyseenin Boli.vian ores,canbe explained in two ways. On the onehand, stannite may belong to a later phase of ore deposition and may havebeendeposited by alkalinesolutions that dissolved cassiterite. On the otherhand,it mayhavebeen deposited by the same solutions that precipitated cassiterite, bismuthinite, and pyrrhotite,

and its deposition accompanied by resorption of the previously

precipitated cassiterite.On thebasis of my observations on other deposits, I ammore inclined to holdto thelatterexplanation. The stannite usually occurs massive. However,on the 2o m.

25o

FRIEDRICH

AHLFELD.

level of the Salvadora vein, magnificent twinnedcrystals up to 4 cm. in sizewere encountered on crystals of cassiterite and covered with crystals of hfibnerite. In depth,below4oo m., stannite disappears completely. 3'phaleriteappearsto. be limited to the upper portion of the deposits as a hydrothermal mineral. No material was available to me for investigation. The sphaleriteis said to be rich in cadmiumand to have been found in the Contactovein together
with franckcite.

Fra.nckeite.--Thismineralwas not determined analytically, but in polished surfaces by comparison with franckeitefrom Huanuni. It occurs as a very thin coatingand in tabular crystalson pyrite derivedfrom pyrrhotite. It was not seenwith fresh pyrrhotite. At a depthof about300 m. on the Contacto vein, I observed a4 cm. stringerwith cassiterite on both walls and franckeite, pyrite derivedfrom pyrrhotite,and sideritein the middle. A polished section showedcassiterite filling the spaces betweenthe franckeite crystals. This may representcontemporaneous depositionof

franckeite and cassiterite. At greaterdepthsfranckcite was not

observed.

Hiibnerite.-It occurs near the surface in separate veinlets of moreremote magmatic affiliation. It possesses the typical brown color and radiatingprismatic structure. Excellentcrystals are
found in druses. In the Contacto vein it is associated with

cassiterite. In a specimen from the Salvadoravein cassiterite

crystals encrust hiibnerite crystals, that is, are younger than the
hfibnerite.

3'iderite. mOccurs sparingly with pyrrhotite. Pyrite.--Appears to playa subordinate roleasa hypogene mineral. Occasional it coats cassiterite. Most of the pyritehas been derived from pyrrhotite.

Rarehypogene minerals arechalcopyrite andgalena.

3'upergene Minerals.

Since theoxidation zone haslongsince been worked out,I was

unable to study theminerals in place butwasdependent onspecimens fromolder collections. At theoutcrop of most veins were

TIN

ORES OF UNCIA-LLALLAGUA,

BOLIVIA.

:251

lage quantitiesof cavernous limonite derived from pyrrhotite. Davy mentions native arsenicand realgar derivedfrom arsenopyrite. Greenockite as a yellow, pulverulent coatingon "pyrite is said to havebeenquite common, especially' in the San Fermin
vein.

Concerning supergene bismuth minerals, I 'observed bismuth

ochre in cassiteritepseudomorphic after bismuthinite. Among the specimens of bismuthinitefrom the upper levelswere some with a'weaker metalliclustre. On polished sections it was seen

thatthebroad crystals of bismuthinite were opening upintoleaves

alongcleavage planes and were bent.intoice crystaland feathery forms. They possessed a darker color than the fresh bismuthinitc and lower hardness. The mineral is an ox3/sulphide of bismuthand apparentlyidenticalwith ka.relinite,Bi4 SOs (Fig. 4). A separation of this mineral for analysis is impossible on account of its intimate intergrowthwith bismuthinite. The min-

eralmusthavebeen quiteabundant in theupper levels at Llallagua With a limitedsupply of oxygen, stannite decomposed to chalcopyrite,chalcocite, bornitc and corellite. In the presence of moreabundant oxygenthe latter mineralswere oxidized. Copper sulphate migrateddownward,reactedwith fresh sulphides and
gave rise to copper-richcementationzones with native copper.
The tin content of the stannite in the form of metastannic acid

appears not to have migrated,but to have beendeposited in situ as cassiterite. In the specimens of stannitefrom near the surface that were availableto me, I couldfind no supergene cassiterite. Suchmay havebeenpresent, but overlooked in mining. Koeber-

lin 8 states that largequantities of stannite were oxidizedin the

San Fermin' vein. He describes beneath the limonitic zone a

Iom.

thick zone o.f rich cassiterite,marcasite, and chalcanthite.

The latter, in the almostcomplete absence of other copper minerals,must haveoriginatedfrom the weatheringof stannite. The phosphorus contentof the primary ores and the ready solubility of the apatitegave rise to the formationof numerous and in places large quantities of supergene iron and aluminum
s Koeberlln, "Geologic Features o Bolivia's Tin-bearing Veins," Eng. and Min.

]our.'Press,vol. sex,pp. 636-642, x926.

252

FRIEDRICH

ZtHLFELD.

Fro. 4- Bismuthinite (white) altering to karelinite (gray).

black is bismuth ochre and limonite. Salvadora vein.

The

Polished section,

X 235.

Fro. 5. Pyrite pseudomorphic after pyrrhotite with basal cleavage (white) and volframite (gray). San Fernfin vein. Polished section,
< I2.

Fc. 6. Pyrrhotite altering to pyrite. San Fermin vein. Polished

section, ,, 2.

TIN

ORES OF UNCIA-LLALLAGUA,

BOLIVIA.

:253

phosphates. In the SanFermin vein,400 m. beneath the surface,

I observed cavernous masses of wavellite that in placesfilled the
entire width of the vein. Wavellite is the most common of the

phosphates and occursvery frequentlyas a coating on quartz crystals and cassiterite and as pseudomorphs after large tabular

marcasite crystals.In thewavellite druses are occasionally the

iron-aluminumphosphates described by Gordon, vauxite (light blue), metavauxite, and paravauxite(both colorless). Vivianite, which is especially characteristic of the Bolivian deposits, occurs in beautifultransparent crystals both in drusesin massive cassiterite and togetherwith phosphate-allophane and pyrite in separate veins. The abundantoccurrence of vivianite ceases at about 36o m. belowthe surface. Severalof the phosphates that I observed have not yet beennamed,as, for example,a hydrousiron phosphate that togetherwith vauxite forms white silky tufts. On heatingit turns brown. In openfracturesthat cut the Forasteravein, I observed at a depth of 45 m. besides wavellite an abundance of chalcedony, in the outer layersof whichminutequantities of sulphides are visible

in polished sections. They include pyrite, wurtzite;a white isotropicore of the smaltitefamily, a light greenmineralintergrown with the white that is perhaps tetrahedrite, and an unknownanis'tropic rose-coloredcopper mineral. These chalcedonydeposits may have been formed by either ascending or descending solutions.

Beforeleavingthe consideration of the minerals, the pyrite and marcasite derived from pyrrhotite should be mentioned. They are widespreadand abundant in nearly all the veins. Pseudomorphsof pyrite after pyrrhotite are found in all the Bolivian deposits. Polishedsections show that the transformation begins along the boundariesof pyrrhotite with other minerals and advancesinto the pyrrhotite individualsto form a network of pyrite lathsenclosing freshpyrrhotitegrains (Fig. 6). Even when the transformationis complete, the basalparting of the pyrrhotite is well preserved(Fig. 5)- A subtraction of substance appearsto accompany the change, because countless druses are left lined with minute pyrite crystals. To explain the chemistryof this trans-

254

FRIEDRICH

A HLFELD.

formation

is difficult.

Either

a subtraction of iron or an addi-

tion of sulphurmay have occurred. A subtraction of iron (as

sulphateor carbonate,is an oxidation process confinedto the zone of oxidation. In Llallagua almost all pyrrhotite'is transformed to pyrite,evenat depths to whichoxidationprocesses do not reach. Fresh pyrrhotiteoccurs only in small nests. Even at depthsof 4oo m. or as much as 600 m. there is little or no pyrrhotite. In other Bolivian deposits, however, as at Tasna, there is much pyrrhotite at depth; there the transformation does not seem to have extended deep. Similar observations are at hand for other deposits from otherpartsof the world. A subtraction of iron can

haveoccurred onlyas a double transformation (perhaps through the formation of acidiron carbonate). Newly formedcarbonates or hydroxides oughtthento occurin association with the pyrite. But such is notthecase. At Llallagua even thereadily oxidizable bismuthinite within the pyrite pseudomorphs is preserved unaltered. Another explanation must thereforebe sought. The change musthavebeenbrought aboutthrougha directaddition

of sulphur molecules. This could takeplace through the action

the reaction:

on thepyrrhotite of thermal waters carrying H-OS a/cco. rdingto

FeS q- H_oS---FeS-o q- H2.

It is reasonable to assume somesolution and transportation of

pyrrhotite at thesame time. The formation of pyritefrompyrrhotite would notthenbea process restricted to thezone of oxidation but would 'berestricted to the action of thermal waters con-

taining hydrogen sulphide.The process requires furtherelucidation.

Summaryof Hypo#eneMineralization.

The following general statements canbe madeconcerning the distribution and quantity of the hypogene minerals in the Llallagua-Llncia deposits.

. In largequantity: pyriteand marcasite derived from pyrrhotite.

2. In quantity: tourmaline, quartz,cassiterite. 3- In appreciable quantity: bismuthinite, wolframite, stannite.

TIN

ORES OF UNCI.4-LL.4LL.4GU.4,

BOLIVI.4.

:255

4. Sparingly: arsenopyrite,native bismuth, pyrite, franckeite, hfibnerite,sphalerite (?). 5. Sporadically or in minutequantity: apatite,monazite,siderite, nacrite,chalcopyrite.
,

ORIGIN

OF

THE

ORES.

The sequence of mineralsin the Uncia-Llallaguadepositcorresponds to the orderthat hasbeenworkedout for the magmatic
cycle from physical-chemical considerations. Three zones are

recognized in ascending order,eachonecharacterized by the predominance of differentminerals. In the deepest portions of the deposittourmalineis most abundant. It is not confinedto the immediate wall rockof the veins,but is widelyand irregularly distributed throughthe countryrock. Higher up. it is more restricted in its distribution and disappears entirelynear the surface.

The second zoneis characterized by an abundance of heavy metals. Cassiterite appears first. It is found in small quantity asa constituent of thegreisen in the deepest mineworkings (in theSanFerminveinat 775 m. below the surface). The lower
boundary of tin ore deposition in the Uncia section liesbetween 55o m. and 6oom. belowthe surface; in the Llallagua section,

whichis nearer the periphery of the intrusion, it lies deeper.

Arsenopyrite is the oldestof' the sulphides.The formationof tourmaline ceased whenthe deposition of cassiterite wasinitiated.

It occurs in thewallrockof thetin veins asan older deposition, butnotwithinthecompact cassiterite masses characteristic of the

second zone. As a consequence of rapidly falling temperature,

cassiterite wasdeposited verysuddenly.It occurs in theformof

shoots (bolsones) of unexampled richness but of verylimited vertical extent.These shoots arefound at greater depths near the periphery of thestock ontheSanFermin vein, asa consequence

of themore rapid cooling under theinfluence of theproximity of theintruded rock, thanin thecenter, asfor example in the Salvadora vein, where they extend close to thepresent surface. Together with cassiterite were deposited bismuthinite, wolframite, apatite, and pyrrhotite. Quartz continued tobedeposited in con-

256

FRIEDRICH AHLFELD.

siderable quantities. The nature of the mineralizationsuggests very concentrated acid solutions with strong gas phaseand temperature of 400 to 500. The wall rock of the ore shootscontains tourmalinebut has also beenintensivelysericitized. The third zone,farthestremoved from the magma,is character-

ized byless cassiterite andthepredominance of sulphides aridsulphosaltsof tin, lead, copper, antimony, and zinc. These are lacking in the lower parts of the mine at depthsbelow 45 m., and increasein abundance upwards. Stannite appearsfirst and
even occurs in the rich cassiterite shoots. Then follow in order

franckeite,hiibnerite,sphalerite, siderite,all mineralsof the hydrothermal sequence.Their investigationis made difficult because the uppermost part of the deposit hasbeenleached by oxidation processes.Furthermore,the upper parts of the mine are workedout and no longeraccessible.Koeberlin'sstatement that a numberof the mostimporfantveinsdo not extendto the surfaceis perhaps to.be explained by the fact that in their uppermost parts they did not carry cassiterite but easilydestroyed sulphides and sulphosalts and that only limoniteremains. The transformationof pyrrhotite into pyrite and marcasite doubtless occurred at the close of the periodof mineralization.CONCLUSIONS.

The formation of the Uncia-Llallaguadepositis considered to have been a single continuous process. Differencesin mineral contentare not regardedas due to deposition by different solutions ascending at different times,but by one and the samesolution which changed its composition through fractionaldistillation and therebygave rise to various primary differences in depth. This view can not be provenabsolutely, but is probablefrom the standpoint of our understanding of the physical-chemical principlesof ore deposition. Let us summarizeoncemore the resultsof the preceding considerations. High tensionsolutions rich in volatile constituents penetrated the top of a small stock,the tuff-like quartz porphyry of which was ideallysuited for the deposition of ores. Mineral-

TIN

ORES

OF UNCIA-LLALLAGU,4,

BOLIVIA.

257

ization began with the formation of tourmaline greisen. The solutions, penetrating the countless fractures of the quartz porphyry,decreased in pressure so rapidly that the bulk of their

heavymetal contentwas precipitated in a relativelysmall space. Precipitationbegan with little sulphide. With falling temperature the initially acid solutionsbecameneutral and sulphosalts (stannite) were precipitatednearer the surface. Most of the heavy metal content had been depositedbefore the solutions reached the surface. To what extent ore deposits of the hydrothermal phasewere depositedin the eroded crest of the stock and the former sedimentary coverwe can not know. Analogy. with other relateddeposits rendersit likely that hydrothermal chalcopyrite-pyrite veinsandargentiferous galena-sphalerite veins
existed.

The deposit is a transition between the intrusive-pneumatolytic tin veinsin northern Boliviaandthe extrusive-hydrothermal.ones in southern Bolivia. It is more closely relatedto the former, however, because of theprominence of tourmaline, cassiterite, and pyrrhotite, a characteristic assemblage of many of the northern deposits (i.e. Araca). On the other hand,in the northerndeposits, the well-defined primarydifferences in depth that signify an unusually quickcooling of the solutions are lacking. Likewise theydo'notshowthe simultaneous deposition of minerals which undernormalconditions are widelyseparated, asbismuthinite and
stannite or cassiterite and franckcite.

Finally,attention should becalled to the factthatthegreatenrichments that Koeberlinascribed to secondary processes are satisfactorily accounted for in another way by the preceding explanation.
M,v, G,.

17