Mechanisms of Intrusion of the Coastal Batholith of Peru Into Its Own Volcanic Ejecta

Home , Batholith, Casma Group, Ring dike

Cauldron Subsidence and Fluidization: Mechanisms of Intrusion of the Coastal Batholith of Peru into Its Own Volcanic Ejecta

JOHN S. MYERS Geological Survey of Greenland, 0stervoldgade 10, Copenhagen, Denmark

ABSTRACT 10 to 60 km inland, and has subvertical walls and a flat roof. It was emplaced be- The Coastal Batholith of Peru shows how tween Late Albian and Eocene time (about diorite, tonalite, granodiorite, and granite 100 to 50 m.y. ago) into two groups of magmas rose through their last few volcanic rocks which were erupted from the kilometers by a process of magmatic stop- rising batholith. ing by their fluidized upper surface. The Structures and features relating to intru- magmas successively displaced both older, sive mechanisms are described together, more basic plutons and their own volcanic drawing on examples from different units debris downward and canje to rest within 3 of the batholith because the structure and km of the Earth's surface after loosing vol- intrusion mechanism of most plutons are atiles by eruptions through fissures and cal- broadly similar. A more complete account deras. There was a continuous association of the batholith, the volcanic rocks, and the between volcanic eruptions and plutonic in- regional geology is in press (Myers, in trusions from at least 100 to 30 m.y. ago in press). Plutonic rock names follow the a narrow belt parallel with the continental margin. The batholith is 50 km wide, more than 1,100 km long, and probably 15 km or less thick. It has steep walls and an extensively preserved flat roof exposed in mountainous desert with relief of 4,500 m. It consists of plutons and sheets which were intruded in five distinct episodes into the same tabular belt by repeated cauldron subsidence. Each subsidence was preceded by the formation Figure 1. Map showing the extent of the of small shear zones which were fractured Coastal Batholith of Peru (after Bellido and and fluidized, and accompanied by the rise others, 1972) and locations of the area described and of section A-B of Figure 12. of corrosive, turbulent gas-liquid-solid mix- tures. Individual plutons form relatively thin tabular bodies with flat roofs and relief of more than 4,500 m. This paper de- floors and steep walls which pass down- scribes the structure of the batholith and ward into ring dikes and upward into ring evidence of the mechanism of its intrusion. dikes and calderas. Key words: structural Earlier descriptions of other parts of the geology, igneous rocks, batholiths, vol- Coastal Batholith were made by Jenks canism. (1948), Cossio (1964), Cossio and Jaén (1967), and Stewart (in Garcia, 1968). Part INTRODUCTION of the batholith lying immediately south of this area was described by Cobbing and The Coastal Batholith of Peru is one of Pitcher (1972) and Cobbing (1973a). These the largest and best exposed of a chain of studies showed the batholith to consist of a Mesozoic and Tertiary batholiths which number of plutons of which tonalite, occur along the western margin of the granodiorite, and adamellite far exceed feftvj Coastal Batholith American continent. It forms an almost basic rocks and granite in total volume. The continuous outcrop 1,100 km long and 50 earliest intrusions are more basic than I" Calipuy Voleantes

km wide on the western flank of the Andes younger ones and local basic to acid dif- j v v | Casma Voleantes between latitudes 8° and 16° south (Fig. 1). ferentiation sequences occur within some The 80-km-long section of it which lies be- composite plutons. K-Ar minimum ages, • ' Crttacvous shtlf stdiments tween latitudes 10° and 10° 30' south was mostly of biotite and hornblende obtained Pr«-Crttac«ous rocks mapped on the scale of 1:50,000 during 12 from a wide area of the Coastal Batholith, Figure 2. Part of northern Peru showing the months spent in the field in 1968-1970. range from about 100 to 10 m.y. (Stewart close association of the Casma and Calipuy The shapes of the intrusions and the struc- and others, 1974). Groups of volcanic rocks with the Coastal ture of their country rocks can be clearly Between latitudes 10° and 10° 30' south, Batholith (after Cobbing, 1973b); the area de- observed here in mountainous desert with the batholith lies parallel with the coastline. scribed is outlined.

Geological Society of America Bulletin, v. 86, p. 1209-1220, 12 figs., September 1975, Doc. no. 50903.

1209 1210 J. S. MYERS

definitions recommended by the Interna- with amphibolite facies metamorphism. Dikes of microdiorite were intruded be- tional Union of Geological Sciences (Streck- Biotite forms a prominent schistosity, tween and during the emplacement of all eisen, 1973). The term "host rock" is used hornblende crystals are elongate parallel the complexes. In the Sayán area (Fig. 1), to include any rock into which a pluton, with fold axes, and cordierite contains the dikes intruded between the Santa Rosa dike, or sheet was intruded, whereas the spiral inclusion trails of opaque minerals. and Puscao — San Jeronimo Complexes give term "country rock" is restricted to non- The Tapacocha fold belt is an anticlinorium K-Ar whole-rock ages of 72 to 68 m.y. A plutonic rocks, mostly volcanic, into which of isoclinal folds associated with green- fifth group of plutons at Sayán, younger the batholith was emplaced. schist facies metamorphism. than the Puscao — San Jeronimo Com- plexes, give K-Ar mineral ages of ~30 m.y. VOLCANIC COUNTRY ROCKS Calipuy Group (P. A. Wilson, 1974, personal commun.).

The Coastal Batholith was intruded into The Calipuy Group is a sequence of ter- Major Structures the Casma Group of volcanic rocks which restrial andesite, dacite and rhyolite, lava, are chiefly marine and Middle to Upper and pyroclastic flows which extends for at The batholith has steep walls, subparallel Albian, and the Calipuy Group of terrestri- least 600 km along the western cordillera of with the coast and continental margin, and al volcanic rocks which unconformably northern Peru (Fig. 2). The rocks were first a flat roof. The roof is extensively preserved overlie them and are Late Cretaceous to described in detail by Cossio (1964) and in the northeastern part of the area (Fig. 3). early Tertiary in age (Fig. 2). Cossio and Jaén (1967) and are extensively Individual plutons have similar flat roofs intruded by the Coastal Batholith. In the and steep outward-dipping or vertical walls Casma Group northeast part of the area described, they (Figs. 4, 5, and 6). The plutons generally overlie the Tapacocha fold belt with form rectangular bodies with long sides The Casma Group is a sequence of ande- marked unconformity (Fig. 3) and their av- parallel to the batholith and short sides site lava and pyroclastic flows, andesite pil- erage thickness is in the order of 2 km. The normal to it [for example, the Huampi low lava, tuff, dikes, and sills, with smaller rocks are probably Late Cretaceous to early Piruroc granodiorite (Fig. 4); the Con- amounts of dacitic pyroclastic flows, chert, Tertiary in age, and they formed the land taderas, Llagumpe, and Maria Cristina plu- and volcaniclastic sediments. The sequence surface during intrusion of most of the tons of the Puscao unit, and northernmost extends for at least 500 km along the batholith. San Jeronimo pluton (Fig. 5)]. The transi- coastal region of northern Peru and inland tion from steep walls to flat roof generally to just beyond the eastern margin of the COASTAL BATHOLITH occurs within a vertical distance of 50 m Coastal Batholith (Fig. 2). In the area de- (Fig. 4, localities B and D). This relation- scribed here, it comprises six formations The batholith is a heterogeneous mass of ship is most clearly seen on a large scale in with a total thickness of about 6.5 km, al- coarse-grained igneous rocks which was in- the dome-shaped outcrops of the Puca Jirca though the actual thickness in any one place truded as numerous plutons and sheets. and Chasquitambo plutons of the Puscao is probably in the order of 3 to 4 km. The Each pluton is a distinct mappable intrusive unit. The Puca Jirca pluton forms a subcir- stratigraphic sequence is summarized in the body with steep walls and a flat roof. Many cular dome-shaped outcrop, 5 km in diame- legend of Figure 3 and described in detail in plutons are composed of identical rock ter, through which a valley cuts a section Myers (1974, and in press). Fossils indicate types, and each major mappable rock type 700 m deep (Fig. 5). At the highest levels that most of the lower part of the volcanic is called a magma unit. This unit is the fun- exposed, the contacts dip outward at 5°; sequence is late Middle Albian in age. In the damental division of the batholith and is whereas in the valley floor, they dip at 30° northeast corner of the area, the Casma analogous to the lithostratigraphic term to 40° outward. The Chasquitambo pluton Group conformably overlies a thick se- "formation." In the same way, a "com- crops out over an area of 300 sq km be- quence of shale and siltstone of probable plex" (or super-unit of Cobbing and tween altitudes of 400 and 2,400 m (Fig. 5). Hauterivian to Middle Albian age (Huayl- Pitcher, 1972) is a group of related units, At the highest topographic levels, its con- lapainpa Group). and a "subunit" is a division of a unit, tact dips at less than 10° outward; at the analogous to the lithostratigraphic terms lowest levels exposed, its contact dips Regional Deformation and Metamorphism "group" and "member." steeply Outward. of the Casma Group The main components of the batholith Some plutons form ring dikes, the largest are listed in Table 1 in order of age, based of which are the Anta and Corcovado The Casma Group was folded into large, on crosscutting relations, and are described dikes, each 1 km thick and in rings up to 20 upright, open folds with axes slightly ob- in detail elsewhere (Myers, in press). km in diameter (Fig. 5). The walls of the lique to the walls of the batholith (Fig. 3). It TABLE 1. COMPONENTS OF THE BATHOLITH was metamorphosed into greenschist facies . Granite porphyry dikes during this ductile deformation before up- San Jeronimo unit < ApUte sheets ^ Syenogranite plutons lift, erosion, and eruption of the Calipuy Puscao-San Jeromino complex -70-50 m.y. (I) Group of volcanic rocks. The Casma Group . Aplite sheets was most strongly deformed in the Canoas Puscao unit J Monzogranite (adamellite) plutons t Baranda granodiorite sheets syncline and the Tapacocha fold belt (Fig. Huampi Piruroc granodiorite unit Santa Rosa complex Corralillo tonalité unit 3), which almost coincide in location with -95-70 m.y. (I) Huaricanga tonalité unit the future western and eastern walls of the Puca Punta mylonite sheets Paccho complex diorite and tonalité units batholith. The northwestern part of the -95-90 m.y. (F) Canoas syncline is an open fold with little Patap complex gabbro and diorite units associated metamorphism. The fold is -100-95 m.y. (F) tighter toward the southeast and passes Note: The isotopic ages are from recent unpublished work of P. A. Wilson (I) and are K-Ar minimum ages of structurally downward into a complex separated biotite, feldspar, and hornblende from samples collected over a wide area of the batholith, including synclinorium of isoclinal folds associated the region described here, or guesses (F) based on these ages and fossil evidence (Myers, 1974). A. k: « sv A 'Ao - o « oV,v\-> 10°S

VWItf V"; V! 0/ oo//. A i, •I wV KSl-vi. A

SSA ^»TP o^M ^ \. - A v' "O > o

"Ó o o

^W^^^Ki\ iA//N */ml^ il'/s¿mi/

V. .'-il'jb 's. /sv /, / . Vvii vf "KMh / ^AILT^m^'-* I / /.I- ' I. / _ \ L I 1 M

V / I ^ / CANOAS 7 (/SYNCLINE/

^/Vl'** I / 1 tO km V^v'

v••«¿viL o ^ifiruSw ^ tf1//

78°W, lO^CS 77*30'W

COASTAL BATHOLITH '/W Post-tectonic tonalite, granodiorite and granite c. 100-50my Syn-tectonic gabbro and diorite (Patap complex)

CALIPUY VOLCANICS Andesite and dacite; lava and pyrodastic flows c.2000m c. 95-50 my

? Pararin Fm: andesite lava and pyroclastic flows 600m

°o"o Cerro Lupin Fm: andesite pillow lava 1,800m

Cerro Breas Fm: chert 800m CASMA VOLCANICS agglomerate c.105-95my La Zorra Fm: f 1,800m tuff j « o J

; °n<| Punta Cramadal Fm: andesite pillow lava 600m

Seftal Cochapunta Fm: chert 1,000m HUAYLLAPAMPA GROUP Shale, siltstone, limestone and conglomerate 2,&00m c.120-105my

s ncline ^^ Anticline y Figure 3. Map of the stratigraphy and structure of the volcanic country rocks and syntectonic and posttectonic intrusions of the batholith, with locations of sections 1 -4 of Figure 6. Approximate ages ^ooo/ Contours in meters after fossil evidence and unpublished age determinations by P. A. Wilson. 10°S

10°30'S 3/ 4/ 77°30'W Figure 4. Map of the Santa Rosa complex and older rocks, with location of sections 1-4 of Figure 6.

1214 J. S. MYERS

Figure 6. Sections across the batholith ornamented as on Figures 4 and 5, where their locations are also marked. The structure of the country rocks is indicated by lines representing bedding. Black lines within the batholith indicate Baranda sheets; Puca Punta mylonites are not shown. Horizontal and vertical scales are equal.

ring dikes are generally vertical, but east of Pluton Contacts identical appearance with diorite composi- the village of Anta (Fig. 5), the inner con- tion and appinitic texture. tact of the Anta dike curves gently over an Most pluton contacts are sharp, and in apparently foundered block of Corralillo many places contact effects by which to in- Internal Structures tonalite which is cut off by the younger terpret the emplacement order of contigu- Corcovado ring dike of the San Jeronimo ous plutons are absent or ambiguous. Am- Plutons are generally homogeneous and unit (Fig. 6, sect. 4). Within the Corcovado biguous contact features include thin peg- contain rounded dioritic xenoliths less than ring dike, rocks of the Casma Group lie on matite veins along the contact and a slight 15 cm long, which are uniform in size and top of the flat roof of the Patorumi pluton reduction of the grain size of both pluton distribution. In some places the xenoliths (Fig. 5). The stratigraphy of these volcanics and host rock. can be seen to be derived by brittle fractur- (Pararin Formation, Fig. 3) indicates that Stoping and spalling occurred at some ing and mechanical corrosion from volcanic relative to neighboring volcanic rocks, they contacts. Figure 7 shows a typical example country rocks and Patap diorite (Fig. 7). have subsided a vertical distance of more from the roof contact of the Corralillo tona- In other places, the xenoliths formed than 1,500 m within the Corcovado ring lite (Fig. 4, locality C). At the contact, angu- from microdiorite sheets which broke up dike. The Patorumi pluton and Anta ring lar blocks of country rock were broken off after intrusion into unconsolidated plutons. dike are almost identical Puscao rock types, by stoping and spalling (Fig. 7a). Within the The latter situation is well displayed in the and their close spatial association on either next 200 m downward within the pluton, roof zone of the Puca Jirca granite pluton side of the Corcovado ring dike suggests the angular fragments are progressively (Fig. 5), where various stages and kinds of that they were originally part of one intru- smaller, more rounded, and dispersed, as breakup of the sheets are preserved (Fig. sive body. These outcrops both enable the their distance from the contact increases 8). Flat-lying trains of pillowlike bodies thickness of a major pluton to be measured (Figs. 7b and 7c). In the same distance, and angular fragments of porphyritic mi- (here in the order of 2 km) and indicate that fragments of both andesite volcanic rocks crodiorite are abundant as far as 200 m a tabular pluton passes downward into a and Patap diorite were recrystallized to from the roof, in both the pluton and in the ring dike. form medium-grained, uniform xenoliths of Huaricanga tonalite host. The microdiorite CAULDRON SUBSIDENCE AND FLUIDIZATION, COASTAL BATHOLITH OF PERU 1215

locally heat-softened envelope soon after the emplacement of the pluton. There is no evidence of the origin of most xenoliths, but because of their general un- iformity and the evidence described above, they are considered to have been derived for the most part from either host rocks or dikes intruded from depth. There is no field evidence that any xenoliths were trans- ported by the magma from great depth below the batholith. Many plutons possess a foliation weakly defined by disclike dioritic xenoliths and the long axes of hornblende and plagioclase. The foliation formed during magmat- tDes* igt ic crystallization under stress, and it is generally subvertical and nearly parallel with the length of the batholith. It is locally more strongly marked at pluton contacts, parallel with the contacts. Some plutons are composite with older (darker) and younger (lighter) subunits. The large-scale relationship between such subunits is best displayed in the Chas- quitambo pluton where a younger subunit occurs as a dome within a shell of an older subunit (Fig. 5; Fig. 6, sects. 2 and 3). In some places, the contact between them is sharp and the younger subunit encloses angular fragments of the older subunit; but in other places, the contact is gradational. Dioritic xenoliths are more abundant in the older subunit and can be seen in various Figure 8. Disrupted, flat-lying microdiorite stages of digestion which locally contami- sheets (stippled) in the roof zone of the Puca Jirca nated the monzogranite and converted it to pluton (white) (Fig. 5). a. Disruption after con- granodiorite. solidation of a sheet, b. disruption both before and after consolidation of a sheet, c. pillows Baranda Sheets formed by disruption before consolidation of a sheet.

Figure 7. Stoping, spalling, and progressive Sheets of gray granodiorite to quartz-rich rounding and breakdown in the size of andesite granodiorite were precursors of the em- Chasquitambo, a thick flat-lying sheet oc- volcanics (stippled) in the Corralillo tonalite placement of Puscao monzogranite plutons. curs between the two subunits of the Chas- (white). Locality C, Figure 4; a. at the contact, b. Their fabric and grain size range from quitambo pluton (Fig. 5; Fig. 6, sect. 4). 100 m below the contact, c. 200 m below the medium-grained, isotropic to fine-grained, The sheet cuts the older, darker subunit contact. planar to protoclastic mylonite. The sheets above and is itself cut by the younger, are named after Cerro Baranda where they lighter subunit below. It appears to have pillows partly resemble basaltic pillow lava are well exposed (Fig. 5); most are dikes, occupied the initial fracture which became erupted into water, and dolerite pillows in but they are collectively referred to as the roof fracture of the younger subunit. granophyre described by Blake and others "sheets" because some occur with flat-lying South of the village of Anta (Fig. 5), the (1965). The pillows have fine-grained mar- attitudes. They are generally 5 to 10 m wedge-shaped end of the Anta ring dike gins and globular boundaries (Figs. 8b and thick, but some are as much as 500 m thick. consists of Baranda granodiorite; it grades 8c) and in different places are veined by In profile section, many form irregular net- westward, as the dike thickens, into normal both the Huaricanga tonalite and the Puca works (Fig. 9), and a few form irregular Puscao monzogranite which makes up the Jirca pluton. They enclose angular frag- pipes a few meters in diameter. They are main part of the dike. ments of the Huaricanga tonalite but none characterized by the diversity of their size, of the Puca Jirca pluton. Some of the pil- internal structure, and xenoliths on both Contact Metamorphism lows were deformed before consolidation regional and local scales; they are of great (Fig. 8c) and some crystallized with patchy importance in understanding the process of The contact metamorphism produced by metamorphic textures. Immediately beyond cauldron subsidence discussed below. the Huaricanga tonalite is typical of most this contact zone of trains of microdiorite Baranda sheets are generally abundant as units of the batholith. Intense recrystalliza- pillows and fragments, thin and irregular far as 2 km outside the walls of a Puscao tion of the country rock is limited to within but continuous sheets of microdiorite occur pluton, parallel with the walls and along 50 m from the contact, but major recrystal- in the Huaricanga tonalite. These features the contacts, such as on the western side of lization can be seen for as much as 300 m suggest that the trains of pillows and frag- the Maria Cristina pluton (Fig. 5; Fig. 6, from its walls and 1,000 m above its roof. ments were derived from microdiorite sect. 1). They also occur with flat-lying at- Beyond these distances, contact meta- sheets which were intruded into the roof titudes up to 2 km above the flat roofs of morphism is not easy to distinguish in the zone of the Puca Jirca pluton and into its Puscao plutons. Southeast of the town of field from prebatholith low-grade regional 1216 J. S. MYERS

reole of 1 kb. Mineral assemblages indicate host rocks. Therefore, the plutons were metamorphism in the hornblende-hornfels neither emplaced by dilation nor by lifting facies with temperatures of 550° to 650°C their roof — they replaced the rocks which (Turner, 1968). previously occupied their sites by exchange downward. A chemical, metasomatic re- DISCUSSION AND CONCLUSIONS placement is unlikely because of the homogeneity of enormous plutons with Major Structures sharp contacts, their high level in the crust, and the narrowness of their metamorphic Both the whole batholith and individual aureoles. The associations between ring plutons have flat roofs and steep walls. The dikes, tabular plutons, and calderas and the shapes and spatial association of the Anta lack of distortion of the country rocks sug- ring dike and Patorumi pluton suggest that gest that the plutons were emplaced mainly plutons are tabular and pass downward by a mechanical replacement of the type into ring dikes (Fig. 6, sect. 4). Irregular first described by Clough and others (1909) veins of Puscao monzogranite are abundant as cauldron subsidence. This is illustrated to the east of the small Murpa partial ring by Figure 10, which is a simplified exten- dike of Baranda granodiorite (Fig. 5) and sion of section 2 of Figure 6 to a depth of 10 are similar to offshoots above the roofs of km and to the early Tertiary land surface major Puscao plutons elsewhere. These re- contemporary with the last stages of the Figure 9. Profile section of subvertical lations suggest that the Murpa dike lies batholith's emplacement. When the mag- Baranda sheets in the Contaderas pluton (Fig. 5). above the walls of a major Puscao pluton. mas came close to the surface, explosive Together, these outcrops suggest that ring ejection of fragmented host rocks with magmatic gas and liquid through calderas metamorphism. Wollastonite, brucite, dikes are both the upward and downward and fissures provided additional space for monticellite, diopside, and garnet are re- extension of the walls of Puscao plutons, the rising plutons. stricted to the inner part of the aureole; and that the upward extensions are com- tremolite, epidote, and hornblende are most posed of Baranda granodiorite, whereas the The lack of distortion of the envelope, commonly seen in the outer part. Close to downward extensions are of monzogranite. particularly the lack of vertical compression the contact, patches with scapolite, calcite, The Corcovado ring dike, which is similar of the roof rocks and the absence of rim- and wollastonite crystals up to 30 cm long in internal structure to a Puscao Baranda synclines, indicates that the plutons were occur and indicate high fluid mobility in sheet, probably reached the surface and not emplaced as diapirs at this level. Ram- this part of the aureole. The size and min- formed a caldera which contributed ignim- berg (1970) found that there is a great dif- eral sequence of the aureole is similar to brite to the Calipuy Group. ference in the structures produced in centri- that of plutons in the Sierra Nevada It can be seen from the maps (Figs. 3, 4, fuge model experiments according to the Batholith described by Kerrick (1970). and 5) and sections (Fig. 6) that both the viscosity contrast between the components Stratigraphic evidence which indicates that whole batholith and its individual plutons simulating magma and host rocks. If there the roof of the Coastal Batholith was 2 to 3 were emplaced without any significant lat- is a low viscosity contrast between magma km thick suggests a total pressure in the au- eral or upward vertical distortion of their and host rocks, then the magma rises

I SJ I Son Jeronimo granite Huampi Piruroc granodiorite [---;.— ] Calipuy volcanics

Puscao granite 1-4 Corralillo tonalité I v I Casma volcanics

Movement of block Huaricanga tonalité [ ' . .| Older country rocks During intrusion of Puscao 2 and Movement of magma San Jeronimo granites [P" j Patap gabbro-diorite Figure 10. Simplified section showing the major structure of the batholith, which is made up of numerous plutons emplaced by repeated cauldron subsidence into their own volcanic ejecta. An extension of section 2 of Figure 6, showing the probable situation that existed during the last major episode of intrusion, ~50 m.y. ago. The present topography is shown by the thick irregular line. The thickness of Calipuy volcanics is a minimum thickness; it was probably 2 to 4 km. The batholith probably extends for about 5 km or less below the base of the section. Only two phases of Puscao granite were seen, and Puscao intrusions 3 and 4 are conjectured to emphasize the nested appearance of repeated cauldron subsidence at one place. Horizontal and vertical scales are equal. CAULDRON SUBSIDENCE AND FLUIDIZATION, COASTAL BATHOLITH OF PERU 1217 diapirically, independent of the structure of the batholith. The shapes of the Puscao plu- states. They are the product of heterogene- the host rocks. If there is a high viscosity tons are also the best exposed by the pres- ous simple shear and represent the localiza- contrast with viscosity of the magma much ent erosion level, and this enables the relation of high strain states where deformation less than that of the host, then the magma tionship between the small emplacement took place by ductile flow. Their distribu- is too weak to push the host rock aside, but structures and the overall structure of plu- tion indicates that high strain concentra- its high fluidity enables it to move rapidly tons to be clearly observed. tions were localized in narrow zones with along zones of weakness. The latter case is The structures which preceded the em- either subvertical or horizontal attitudes, associated with collapse of the more rigid placement of a major Puscao pluton by were parallel with and less than 2 km from overburden and with stoping. The follow- cauldron subsidence occur in the same the future margins of a pluton, and they ing discussion shows that the Puscao plu- order everywhere, although the complete defined a framework of either cylinders or tons and those of Santa Rosa and San sequence is not everywhere preserved. The rectangles: Jeronimo are clearly examples of this sec- complete sequence is summarized below in The fracturing of the shear zones indi- ond case in which the upper surface of the the order in which structures developed. cates a change of deformation state from magmas possessed high fluidity. 1. Small shear zones less than 1 cm wide ductile flow to brittle failure within the and 10 cm long developed in irregular net- same planar zones, perhaps caused by an Intrusion Mechanisms works within distinct planar belts up to 2 increase in strain rate, as the mineral as- km from the future margin of a pluton. The semblages do not indicate any marked Minor structures which provide informa- belts are subparallel with the future margin change in temperature or confining pres- tion about the intrusive mechanisms of plu- of a pluton; they are flat-lying above its sure. A similar association of shear zones tons are generally very small. They are most roof and subvertical outside its walls. The and microbreccias was described from clearly seen around the Puscao granite plu- shear zones were fractured. They now con- along the main boundary fault of the Glen- tons which are emplaced mainly into older tain microbreccias with angular fragments coe cauldron subsidence in Scotland by leucocratic, homogeneous plutonic rocks of of both schistose centers and less-deformed Clough and others (1909, p. 629). They ob- margins of the shear zones and undeformed served of the microbreccias ("flinty crush host rock in a matrix of banded ultramylo- rock") their "invariable connexion in the nite. Additional fractures and shear zones field with obvious signs of shearing. . . . developed in irregular networks and joined Under the microscope .. . banded structure many of the earlier shear zone belts to- is often strongly pronounced, and the rocks gether. themselves are found to be richly charged 2. Baranda granodiorite sheets de- with fragments derived from the adjacent veloped within 2 km of the future margins shear-zone." of a main pluton, both within and adjacent The fragments of microbreccias within to the belts of shear zones and microbrec- the fractured shear zones are both mixed up cias, parallel with them. Some wider mi- and disorientated, indicating that locally crobreccia zones pass gradationally into the the fragments were free to move indepen- Baranda sheets, whereas many other similar dently of each other. This suggests that zones are cut by younger Baranda sheets. streams of gas were introduced into the Angular fragments of host rocks were fractures from below and isolated the finest stoped from the walls of the sheets and grained rock fragments. Progressive stages were mechanically rounded. In some places, can be observed in the enlargement of the the Baranda sheets developed a flow band- microbreccia zones by stoping. The larger ing which was deformed before the sheets breccia zones more than a centimeter thick consolidated; in other places, the adjacent contain both rounded as well as angular softened host rocks were deformed and re- fragments, and rounded fragments are crystallized and became banded gneisses. progressively more abundant in larger brec- 3. A major pluton was emplaced by cia zones, and away from the zone margins. mass exchange with a rectangular segment These observations suggest that the of its host rocks which subsided within a rounded fragments are the result of framework of Baranda sheets. The pluton mechanical abrasion of angular stoped wall cut and stoped fragments from its im- fragments rather than being of primary der- mediately contiguous Baranda sheets. ivation from initially curved fractures. 4. In some plutons, a first major intru- Wide breccia zones are rarely seen, but sion was followed by the emplacement of a transitions occur between breccia zones second, more homogeneous and more felsic consisting of both round and angular frag- subunit within a carapace of the first sub- ments in a matrix of ultramylonite, and unit, preceded by a repetition of events 1 to Baranda sheets containing between 20 and 3. In some cases, sheets of microdiorite 75 percent by volume of both rounded and were intruded into both a pluton and its angular xenoliths of host rocks (Fig. 11). corroded phenocrysts of quartz (ql) and plagio- immediate envelope and broke up into True tuffisites as first described by Cloos clase (pi) with rounded secondary overgrowths trains of pillows. (1941) are of only minor occurrence in the of quartz (q2) and plagioclase (p2); b. mechani- 5. Flat-lying sheets of granophyre were batholith, and Baranda sheets greatly out- cally rounded xenoliths of older microdiorite number breccia zones in volume. (dense stipple) xenoliths in Huaricanga tonalite emplaced in the uppermost parts of some (irregular dashes) in a Baranda sheet (open stip- plutons during further subsidence before The Baranda sheets represent the third ple). Some xenoliths are rimmed by quartz and the pluton was stabilized. stage in the development of the surfaces of plagioclase (white) and are internally partly re- The initial shear zones are similar to weakness initiated by the formation of the crystallized with segregation of felsic and mafic structures commonly seen in homogeneous shear zones and then developed by stoping shells with radial fabric. rocks subjected to heterogeneous strain and widening of the microbreccia-filled 1218 J. S. MYERS

fractures. Sheets in which angular xenoliths whose importance as a geological process VOLCANIC AND PLUTONIC occur at the margin and are followed in- was first recognized by Reynolds (1954). HISTORY ward by progressively more rounded The state of fluidization in which solid and xenoliths suggest that the rounding of the (or) liquid particles are isolated and sus- The batholith and its associated volcanic fragments is related to the time they were pended in through-flowing gas is known rocks were intruded through sialic crust at detached from their original position. A from industrial experience to stimulate least as old as late Precambrian (Stewart similar situation was described by Cloos chemical reaction between its components and others, 1974). The Precambrian (1941) in the Swabian tuffisite vents and as well as to thoroughly mix them and gneisses are exposed to the west of the was interpreted by Reynolds (1969) as the bring about mechanical attrition of the batholith along the coast of southern Peru result of rapid flow of gas-tuff streams in solid particles (Reynolds, 1954, 1969). At and are overlain by lower and upper the center of the vents, bringing in material first, gas diffuses through a body or bed of Paleozoic shale and sandstone (Bellido, which had traveled farther and had been fine-grained rock and (or) liquid particles 1969). Abundant volcanic rocks first occur suspended within the vents for longer than but "at a particular rate of gas flow, the in the Permian Mitu Group of clastic sedi- that in the marginal parts of the bodies bed expands and the individual particles ments, but much larger volumes of volcanic where stoping was just beginning. In the become free to move. With increase in the rocks were erupted in the Late Triassic, case of the Baranda sheets, this conclusion rate of gas flow, a bubble phase forms and concurrently with initiation of the two is supported by the rare occurrence of ex- travels upwards through the expanded bed great Andean troughs of Peru, and also dur- otic xenoliths which are invariably more in which the particles are violently agitated; ing Jurassic time. rounded than xenoliths of the immediate the bed is now said to be fluidized" During Cretaceous time, the older rocks host rock. (Reynolds, 1954, p. 577). subsided in two linear belts, parallel with Some of the xenoliths are composite and Although in some cases the adjacent host the continental margin, which are called the consist of either Santa Rosa tonalite or Pus- rock of a Baranda sheet shows signs of ear- West and East Peruvian Troughs (Wilson, cao monzogranite which enclose older lier shearing or deformation during em- 1963) (Fig. 12, sect. a). The West Peruvian xenoliths of diorite or tonalite (Fig. lib). placement of a sheet, in many cases the host Trough was itself divided into western and Such composite xenoliths which are spheri- rock shows no signs of deformation. In the eastern belts by the Tapacocha axis (Myers, cal and in which the truncated surfaces of latter cases where Baranda sheets show 1974), a tectonic line which allowed the both rock types have been equally strongly contorted banding, this banding is western block to sink faster than the eastern smoothed indicate that the rounding was probably a flow structure. Planar banding block. Submarine eruptions of andesite lava caused by a mechanical process. The pres- may be the result of laminar flow, whereas and pyroclastics and the intrusion of ence of concentric felsic and mafic shells contorted banding may reflect turbulent numerous sills (Casma Group) took place around some spherical xenoliths indicates flow. The thorough mixing of fragments of through and onto the fractured western vigorous chemical reaction between the different size, source, and degree of round- block, while thinner shelf deposits accumu- fragments and their matrix, but the fracturing, together with the occurrence of locally lated on the eastern block. The age of am- ing of some concentric shells indicates the derived well-rounded fragments with as- monites interbedded with the volcanic continued mechanical attrition of the sociated turbulent structures, indicates that rocks (Myers, 1974) indicates that subsid- xenoliths. The occurrence together, in a the fluid agent in these cases was gas and ence was most rapid in the Middle and single part of one sheet, of diorite xenoliths not liquid (Reynolds, 1954; Lewis and Upper Albian (—100 to 95 m.y. ago). with angular shape and no sign of reaction Bowerman, 1952). Fragments of diorite and granodiorite in with their matrix, spherical xenoliths with The fluidized state of the uppermost part pyroclastic rocks of the Casma Group indi- concentric reaction shells, and similar of the magmas at the time of their intrusion cate that some plutons had already crystal- spherical xenoliths with concentric shells also explains the narrowness of their lized by this time, probably at shallow which are partly broken and unconform- metamorphic aureoles. Experimental ob- depth. The close spatial and temporal as- ably overgrown, indicates that vigorous servations (Lewis, 1969, p. 107-108) show sociation of the Coastal Batholith and processes of both mechanical attrition and that in a fluidized bed, "the particle move- Casma Group (Figs. 2, 3, and 10) suggests chemical reaction were operative at the ment at walls, or any solid surface im- that these volcanic rocks were erupted dur- same time. The irregular mixup of both dif- mersed in the bed, is so rapid as to produce ing early phases of intrusion which culmi- ferent kinds and sizes of fragments indicates a steep temperature gradient if the surface is nated with emplacement of the Coastal that these processes were combined with heated or cooled. Consequently heat trans- Batholith at its present level (Fig. 12). the thorough mixing of the particles con- fer to or from a surface is very rapid." The During middle to Late Cretaceous time tained within the sheet. Similar phenomena expected consequence is realized by a nar- (~95 m.y. ago), most of the Casma Group relating to the origin of orbicular xenoliths row aureole with a mineralogy which indi- was folded into Andes-parallel folds and were recently described from the Sierra cates rapid heating and only a short time at was strongly deformed in two narrow belts. Nevada Batholith of California by Moore maximum temperature (Atherton and The most intense deformation formed the and Lockwood (1973). Brenchley, 1972). Tapacocha fold belt, marking continued ac- The phenocrysts in the Baranda sheets The bounding fractures of the rising plu- tivity of an older tectonic line (Tapacocha also indicate a history of both mechanical tons appear to have been controlled by the axis). This fold belt is similar to the most corrosion and chemical reaction, because pre-existing framework of Baranda sheets. strongly deformed part of the Canoas syn- they are both rounded and have embay- The occurrence of Baranda sheets as partial cline which passes gradationally upward ments infilled by secondary growths of the ring dikes, together with the occurrence of into a more open structure of less deformed same mineral, but with rounded outer mar- larger ring dikes of monzogranite, suggests and less metamorphosed rocks, and there- gins conforming with the general curvature that the widening of the Baranda sheets by fore it is probable that these structures pass of the rounded outer surface of the pheno- stoping associated with fluidization pro- downward into zones of increasingly in- crysts (Fig. 11a). Some phenocrysts have vided the conduits up which the magmas tense deformation and metamorphism. secondary spherical overgrowths around were emplaced. The central block of older They may be the upward expression of spherical inner crystals. rocks subsided into the fluidized rim of the major shear zones in the basement which These phenomena suggest that the matrix magma as if in quicksand and was corroded were intermittently active throughout Cre- of the Baranda sheets was fluidized, a state by flow of the fluidized system around it. taceous time. Similar shear zones may CAULDRON SUBSIDENCE AND FLUIDIZATION, COASTAL BATHOLITH OF PERU 1219

Present coastline 80°W 73°W WEST PERUVIAN TROUGH - -EAST PERUVIAN TROUOH- 1 TA * EU. I MIO. Mid-Cretaceous sea level

-<-- PARACAS GEANTICLINE MARANON GEANTICLINE p BRAZILIAN SHIELD i r J r 1 i i

'It

50- ; >— MÍd^Cretaceous Moho km UPPER MANTLE

i r > 100 km

Present coastline 80°W COASTAL CORDILLERA 73°W BATHOLITH BLANCA BATHOLITH Present sea level

Figure 12. Sections across Peru along the line A-B on Figure 1, showing the location of the Coastal Batholith in relation to major earth structures. Section a shows the sum of Valanginian to Senonian block movements and eruption of Casma volcanics above the rising batholith. Divisions of the West Peruvian Trough are: Eu = eugeosynclinal and Mio = miogeosynclinal fades, with Tapacocha axis = TA. In section b, divisions of the 100- to SO-m.y.-old Coastal Batholith are: SR = Santa Rosa complex, P = Puscao unit, and SJ = San Jeronimo unit. The 12- to 3-m.y.-old Cordillera Blanca Batholith is also shown. Thickness of present continental and oceanic crust after James (1971), surface geology of section a after Wilson (1963) and Myers (1974), approximate ages of batholiths after P. A. Wilson (unpub. data) and Stewart and others (1974). bound the East and West Peruvian Troughs successive emplacement of the Santa Rosa ago) (Fig. 2), although individual volcanic (Fig. 12). and younger plutons by cauldron subsid- eruptions and cauldron subsidences were Large bodies of rising magma first ence. The oldest tonalite unit of the Santa probably intermittent and rapid events. reached the level of crust now exposed dur- Rosa complex has a minimum age of 96 This suggests that the Casma and Calipuy ing the ductile deformation which formed m.y. (P. A. Wilson, 1974, personal com- Groups of volcanic rocks and the Coastal the Tapacocha and Canoas structures. They mun.) which indicates that intrusion of the Batholith were all part of the same mag- crystallized both during and between pulses Paccho complex and part of the Santa Rosa matic sequence, in which the average com- of deformation as hornblende diorite and complex by cauldron subsidence occurred position of both volcanic rocks and plutons gabbro of the Patap complex. Many of less than 4 m.y. after the eruption of their became more silicic with time. Perhaps these rocks possess rhythmic layering, Albian host rocks, the intrusion of the zone-refining of magmas, initially derived which may have formed in association with Patap complex, and the ductile deformation from subducted oceanic crust and mantle, volcanoes within the Casma volcanic suc- of these rocks. The younger granite intru- became more effective as the magmas rose cession; they may have supplied heat for the sions of the Puscao - San Jeronimo com- to the surface through an increasingly greenschist and amphibolite facies meta- plex (—70 to 50 m.y. ago) rose into the thicker, hotter, and more silicic crust. In- morphism within the narrow deformation Calipuy Group to within 2 or 3 km of the termittent fracturing and intrusion of mi- belts. surface and added to these volcanic rocks crodiorite dikes indicate the continual pres- After major erosion of the folded Casma by eruptions from fissures and calderas, lo- ence or intermittent generation of inter- Group, perhaps associated with regional cated mostly above the walls of rising plu- mediate magma at the depth from which uplift caused by the rise of magmas which tons (Fig. 10). the fractures were propagated, during the formed the Santa Rosa complex, renewed This sector of the western Andes provides whole intrusive life span of the batholith. eruptions of andesite and dacite formed the a fine example of a batholith composed of The Coastal Batholith of Peru supports lower part of the terrestrial Calipuy Group numerous plutons of a small number of the view of Hamilton and Myers (1967) of volcanic rocks (—95 to 70 m.y. ago). magma types, each emplaced by cauldron that many western American batholiths Eruptions were associated with the em- subsidence into the same 50-km-wide belt are thin and crystallized beneath covers of placement of tonalite and granodiorite over a span of 50 m.y. or more. It shows the their own volcanic ejecta. Not all western within the same 50-km-wide belt already flat top of a batholith about 15 km thick, American batholiths, however, were em- outlined by the Patap complex (Fig. 10). emplaced into its own volcanic debris to placed in the same passive way as most of The hiatus between the eruption of the within 2 or 3 km of the surface. There was a the Coastal Batholith of Peru, even at the Casma and Calipuy Groups probably cor- close association between volcanic erup- same high crustal level. Many plutons in the responds with the pause between syntec- tions and plutonic intrusions from at least southern part of the Sierra Nevada tonic intrusion of the Patap complex and Albian to Eocene time (~ 100 to 50 m.y. Batholith are considered by Bateman and 1220 J. S. MYERS

others (1963) to have been forcefully em- rocks of the western Andes, near the Que- evolution of the central Andes: Geol. Soc. placed as they pushed wall and roof rocks brada Venado Muerto, Peru: Jour. Geolo- America Bull., v. 82, p. 3325-3346. aside and upward. The Boulder Batholith of gy, v. 8, p. 161-178. Jenks, W. F., 1948, Geology of the Arequipa Montana is considered by Hamilton and Bateman, P. C., Clark, L. D., Huber, N. K., quadrangle: Bol. Inst. Geol. Peru, no. 9, p. Myers (1974) to have been emplaced dur- Moore, J. G., and Rinehart, C. D., 1963, 105-204. The Sierra Nevada Batholith, a synthesis of Kerrick, D. M., 1970, Contact metamorphism in ing part of a period of thrust faulting and to recent work across the central part: U.S. some areas of the Sierra Nevada, Califor- have spread laterally, deforming its wall Geol. Survey Prof. Paper 414-D, 46 p. nia: Geol. Soc. America Bull., v. 81, p. rocks and carrying its volcanic roof with it. Bellido, E., 1969, Sinopsis de la geología del 2913-2937. Different batholiths therefore appear to Peru: Peru Serv. Geología y Minería Bol. Lewis, J. B., 1969, The fluidization of solid parti- have been emplaced into a variety of near- 22, 54 p. cles: Geol. Soc. London Proc., no. 1655, p. surface tectonic environments. The em- Bellido, E., Girard, G., and Paredes, J., 1972, 106-108. placement history of many batholiths is in- Mapa metalogénico del Peru, scale Lewis, J. B., and Bowerman, E. W., 1952, Fluidi- deed so long that a local tectonic environ- 1:2,500,000, in Bellido, E., and Montreuil, zation of solid particles in liquids: Chem. ment may change during intrusion of a L., 1972, Aspectos generales de la and Eng. Progress, v. 48, p. 603-610. single batholith, such as in the Sierra metalogénia del Peru: Peru Serv. Geología y Moore, J. G., and Lockwood, J. P., 1973, Origin Minería, Geología Econ. Bol. 1, 149 p. Nevada of California and between the of comb layering and orbicular structure, Blake, D. H., Elwell, R.W.D., Gibson, I. L., Sierra Nevada Batholith, California: Geol. Patap and younger complexes of the Skelhorn, R. R., and Walker, G.P.L., 1965, Soc. America Bull., v. 84, p. 1-20. Coastal Batholith of Peru. In spite of these Some relationships resulting from the inti- Myers, J. S., 1974, Cretaceous stratigraphy and local changes of environment, the close as- mate association of acid and basic magmas: structure, western Andes of Peru between sociation in space and time of major vol- Geol. Soc. London Jour., v. 121, p. 31-49. latitudes 10° - 10°30': Am. Assoc. Pe- canism, plutonic intrusion, and deforma- Cloos, H., 1941, Bau und Tätigkeit von troleum Geologists Bull., v. 58, p. tion over an extensive but narrow area of Tuffschloten. Untersuchungen an dem 474-487. western Peru reflects the constant location Schwäbischen vulkan: Geol. Rundschau, v. in press, Geología de los cuadrángulos de of but a single cycle of magma generation 32, p. 709-800. Huarmey y Huayllapampa: Peru Serv. and differentiation of great magnitude and Clough, C. T., Maufe, H. B., and Bailey, E. B., Geología y Minería Bol. long duration, contemporary with the 1909, The cauldron-subsidence of Glen Ramberg, H., 1970, Model studies in relation to Coe, and the associated igneous intrusion of plutonic bodies, in Newall, G., widely postulated subduction of oceanic phenomena: Geol. Soc. London Jour., v. and Rast, N., eds., Mechanism of igneous lithosphere below. 65, p. 611-678. intrusion: Liverpool, England, Gallery Cobbing, E. J., 1973a, Geología de los Press, p. 261-286. ACKNOWLEDGMENTS cuadrángulos de Barranca, Ambar, Oyon, Reynolds, D. L., 1954, Fluidization as a geologi- Huacho, Huaral y Canta: Peru Serv. cal process and its bearing on the problem Geología y Minería Bol. 26, 172 p. of intrusive granites: Am. Jour. Sci., v. 252, Field work was done during tenure of a 1973b, compiler, Geological map of the p. 577-613. postdoctoral research fellowship of the Western Cordillera of Northern Peru, scale 1969, Fluidization as a volcanological Natural Environment Research Council 1:500,000: London, British Governments agent: Geol. Soc. London Proc., no. 1655, (United Kingdom), held at the University of Overseas Development Administration, Di- p. 110-113. Liverpool, England. Thanks are expressed rectorate of Overseas Surveys, 2 sheets. Stewart, J. W., Evernden, J. F., and Snelling, N. to J. J. Wilson, then working at the Servicio Cobbing, E. J., and Pitcher, W. S., 1972, The J., 1974, Age determinations from Peru and de Geologia y Mineria, Lima, for his gener- Coastal Batholith of central Peru: Geol. their significance: A reconnaissance survey: ous assistance in Peru, and to W. S. Pitcher Soc. London Jour., v. 128, p. 421-460. Geol. Soc. America Bull., v. 85, p. for his enthusiastic introduction to the Cossio, A., 1964, Geología de los cuadrángulos 1107-1116. batholith. M. A. Myers provided assistance de Santiago de Chuco y Santa Rosa: Peru Streckeisen, A. L., 1973, Plutonic rocks, Serv. Geología y Minería Bol. 8, 60 p. classification and nomenclature recom- in the field and in manuscript preparation. Cossío, A., and Jaén, H., 1967, Geología de los mended by the IUGS subcommission on the P. A. Wilson kindly provided unpublished cuadrángulos de Puemape, Chocope, systematics of igneous rocks: Geotimes, Oc- age determinations. Thanks are also ex- Otuzco, Trujillo, Salaverry y Santa: Peru tober, p. 26-30. pressed to E. J. Cobbing, J. C. Escher, W. S. Serv. Geología y Minería Bol. 17, 141 p. Turner, F. J., 1968, Metamorphic petrology: Pitcher, T.C.R. Pulvertaft, and W. P. Taylor Garcia, W., 1968, Geología de los cuadrángulos New York, McGraw-Hill, 403 p. for criticism of the manuscript. de Moliendo y La Joya: Peru Serv. Geología Wilson, J. J., 1963, Cretaceous stratigraphy of y Minería Bol. 19, 93 p. central Andes of Peru: Am. Assoc. Pe- Hamilton, W., and Myers, W. B., 1967, The na- troleum Geologists Bull., v. 47, p. 1-34. ture of batholiths: U. S. Geol. Survey Prof. REFERENCES CITED Paper 554-C, 30 p. MANUSCRIPT RECEIVED BY THE SOCIETY 1974, Nature of the Boulder batholith of FEBRUARY 22, 1974 Atherton, M. P., and Brenchley, P. J., 1972, A Montana: Geol. Soc. America Bull., v. 85, REVISED MANUSCRIPT RECEIVED JANUARY 23, preliminary study of the structure, stratig- p. 365-378. 1975 raphy and metamorphism of some contact James, D. E., 1971, Plate tectonic model for the MANUSCRIPT ACCEPTED FEBRUARY 3, 1975

Printed in U.S.A.