Volcanic evolution of the Amealco caldera, central Mexico

Transcription

1 Geological Society of America Special Paper Volcanic evolution of the Amealco caldera, central Mexico Gerardo J. Aguirre-Díaz* Estación Regional del Centro, Instituto de Geología, Universidad Nacional Autónoma de México, Apartado Postal 376, Guanajuato, Guanajuato, México Fred W. McDowell Department of Geological Sciences, University of Texas at Austin, Austin, Texas 78713, United States ABSTRACT The Amealco caldera is a well-preserved Pliocene volcanic center, 11 km in diameter, located in the central part of the Mexican Volcanic Belt. It is one of seven calderas known in the belt. Compared to those of the other calderas, the Amealco products are less evolved, and include only a minor volume of rhyolite. According to the stratigraphic record and K-Ar data, the inferred volcanic history of the Amealco caldera is as follows. Caldera-related activity started ca. 4.7 Ma with eruptions of pumice fallout and pyroclastic flows apparently of Plinian type. These events were followed by eruption of far-reaching surges and pyroclastic flows that deposited three widespread ignimbrites named Amealco I, Amealco II, and Amealco III. By about 4.7 Ma at least 77 km 3 (Dense rock equivalent, DRE) of trachyandesitic-trachydacitic magma were evacuated from the magma chamber and caused caldera collapse. After this climatic stage, pyroclastic activity continued, probably as tephra fountains from ring-fracture vents, that erupted pumice flows and fallouts that were accompanied by mud flows forming deposits of local extent. Both tephra and mud-flow deposits make up a DRE volume of 2.35 km 3. This was followed by 4.3 Ma trachyandesitic lava domes that were emplaced through several ring-fracture vents, making up a DRE volume of 3.8 km 3 ; the domes form the caldera s present topographic rim. At about 4.0 Ma, a modest-sized volcano had formed on the western flank of the caldera that erupted several trachyandesitic lava flows and fallout tephra (both lava and tephra deposits = 0.8 km 3 ). Between 3.9 and 3.7 Ma, 10 intracaldera lava domes were emplaced, accompanied by tephra eruptions that produced relatively small deposits (volume not quantified) that were later reworked and redeposited as lake deposits within the caldera; five of these lava domes are trachyandesitic (4.3 km 3 ) and five are rhyolitic (2.4 km 3 ). The central lava domes are interpreted here as the viscous, gas-poor magma that usually erupts at the end of a caldera cycle, and thus may mark the end of the volcanic evolution of the Amealco caldera. Volcanic activity continued adjacent to the caldera for at least 1.6 m.y. after the emplacement of the central lava domes. These events include bimodal volcanism at 3.7 Ma of basaltic-andesite lava from a volcano just north of the caldera rim (Hormigas volcano) and emplacement of several rhyolitic lava domes to the southwest of the caldera (Coronita rhyolite). At 2.9 Ma a rhyolite (obsidian) lava dome complex was *Present address: Unidad de Investigacion en Ciencas de la Tierra, Campus Jusiquilla, Universidad Nacional Autónoma de México, Apartado Postal 1-742, Centro, Querétaro, Qro., 7600 México. Aguirre-Díaz, G., and McDowell, F. W., 1999, Volcanic evolution of the Amealco caldera, central Mexico, in Delgado-Granados, H., Aguirre-Díaz, G., and Stock, J. M., eds., Cenozoic Tectonics and Volcanism of Mexico: Boulder, Colorado, Geological Society of America Special Paper

2 2 G. J. Aguirre-Díaz and F. W. McDowell emplaced 15 km to the north of the caldera (El Rincón rhyolite). The last volcanic episodes are represented by a 2.5 Ma andesitic lava dome (Garabato andesite) and a 2.2 Ma andesitic scoria cone on the southern margin of the caldera (El Comal andesite). The southern portion of the caldera was displaced by the Epitacio Huerta fault, a major west-northwest south-southwest normal fault, which is part of the more regional Chapala-Cuitzeo-Acambay graben system. The Epitacio Huerta fault was active mostly prior to 2.5 ± 0.3 Ma as the 2.5 Ma Garabato andesite was only slightly displaced by this fault. Faulting apparently ended before 2.2 ± 0.1 Ma, because the El Comal scoria cone is not displaced in spite of its position on the trace of the Epitacio Huerta fault. INTRODUCTION Relatively few calderas have been recognized within the Mexican Volcanic Belt, a complex east-west trending continental volcanic province that crosses Mexico between lat 19 and 21 N (Fig. 1). There are several tens of volcanic system complexes within the belt, but only seven have been identified as calderas, including, from west to east, La Primavera, Los Azufres, Amealco, Mazahua, Huichapan, Aculco, Los Humeros, and Las Cumbres. From these, only the La Primavera and Los Humeros calderas, respectively at the extreme west and east ends of the Mexican Volcanic Belt, have been studied in detail (Mahood, 1980, 1981; Mahood and Drake, 1983; Mahood and Halliday, 1988; Ferriz and Mahood, 1984, 1987; Verma, 1984). Within the central sector, the Los Azufres complex has been the focus of several studies, but its origin as a caldera is still debated (Cathelineau et al., 1987; Dobson and Mahood, 1985; Ferrari et al., 1993; Pradal and Robin, 1994). This chapter focuses on a physical reconstruction of the major volcanic events that formed the Amealco caldera. It is based on the geologic mapping and detailed measurement of several stratigraphic sections that include the caldera products and those of peripheral volcanoes. The field information is supplemented with stratigraphically controlled K-Ar ages and major element chemistry. REGIONAL GEOLOGIC SETTING The Amealco caldera is a Pliocene volcanic center in the central part of the Mexican Volcanic Belt (Fig. 1). The belt is a continental margin volcanic arc that has been related to the subduction of the Cocos oceanic plate along the Middle America trench beneath southern Mexico (Urrutia-Fucugauchi and Del Castillo, 1977; Nixon, 1982; Suárez et al., 1990; Singh and Pardo, 1993). The Mexican Volcanic Belt is a complex arc that can be divided into sectors with their own characteristics, including volcanic style, structure, morphology, age, and chemistry. At least three sectors can be recognized, the western, central, and eastern. The western sector is that between the western coast and the Chapala lake (included); the central sector is between the Chapala lake and the Popocatepetl-Ixtaccihuatl north-south trending volcanic chain (both limits excluded); and the eastern sector is between this chain (included) and the eastern coast (Fig. 1). Two main structural features characterize the central Mexican Volcanic Belt: the east-west oriented graben system of Chapala-Cuitzeo-Acambay (Suter et al., 1991, 1995), also known as the Chapala-Tula fault zone (Johnson and Harrison, 1990), and the Taxco San Miguel de Allende line (Demant, 1978; Nixon et al., 1987), or the Querétaro fracture zone (Johnson and Harrison, 1990), which crosses the central part of the belt with a northnorthwest south-southeast orientation (Fig. 1), and which could be regarded as Basin and Range style normal faulting. These two regional fault systems intersect south of the city of Querétaro. The Amealco caldera is at this intersection (Fig. 1). LOCAL GEOLOGICAL SETTING The first formal reference to the Amealco caldera was made by Sánchez-Rubio (1978, 1984). Carrasco-Núñez (1988) also studied the Amealco caldera and provided a set of chemical data on the volcanic rocks that was published by Verma et al. (1991) together with numerous isotopic (Sr and Nd) analyses. The Amealco caldera is 11 km in diameter (Fig. 2). Its southern portion was displaced by the Epitacio Huerta normal fault system, which bounds the Acambay graben on the north. These faults are part of the east-west trending Chapala-Cuitzeo- Acambay graben system (Fig. 1). The north-northwest southsouthwest Taxco San Miguel de Allende fault system is older than the Amealco caldera, and the east-west Acambay graben is younger than the Amealco caldera, and has seismically active segments (Suter et al., 1995). Representative schematic measured sections of the Amealco caldera products are shown in Figures 3 and 4. Locations of these and other measured sections in the mapped area are shown in Figure 5. The major ignimbrites of the Amealco Tuff are the Amealco I, Amealco II, and Amealco III. Tephra 1 is the tephra beneath the Amealco I ignimbrite and includes Amealco zero, a minor but distinctive ignimbrite only observed in the western deposits (Fig. 3). Tephra 2 is between the Amealco I and Amealco II, and tephra 3 is between the Amealco II and Amealco III (Fig. 3). K-Ar data are provided in Table 1. Representative major-element chemical analyses of the Amealco caldera products and of peripheral volcanism are provided in Table 2, and the total alkalisilica chemical classification is shown in Figure 6. All distances mentioned are with reference to the center of the caldera. It is inferred that the Amealco caldera was formed by

3 Volcanic evolution of the Amealco caldera, central Mexico 3 22 N 20 N 18 N 16 N 105 W 103 W 101 W 99 W 97 W 95 W 22 N 100 km C Normal faults Lake LP Volcano Caldera Town G Chapala Colima graben Middle America Trench 105 W 103 W 101 W 99 W 97 W 95 W Miocene-Pliocene rocks TSM Cuitzeo Pliocene-Quaternary rocks Mo LA Ma Q AMEALCO CCA T N H Ta P M LH M E X I C O Mexican Volcanic Belt 500 km O V Amealco 20 N 18 N 16 N Figure 1. Index map of major volcanoes, calderas, and major fault systems in the Mexican Volcanic Belt. The Amealco caldera is in the central portion of the belt, at the intersection of the Taxco San Miguel de Allende (TSM) and Chapala- Cuitzeo-Acambay (CCA) fault systems. Towns: G: Guadalajara, C: Colima, Mo: Morelia, Q: Querétaro, T: Toluca, M: Mexico City, P: Pachuca, V: Veracruz, O: Oaxaca, Ta: Taxco. Calderas: LP: La Primavera, LA: Los Azufres, Ma: Mazahua, H: Huichapan, LH: Los Humeros. Modified after Nixon et al. (1987). Inset shows regional index map of the Mexican Volcanic Belt. collapse, following the model of Smith (1979) and Smith and Bailey (1968). However, it apparently did not undergo resurgence after collapse, although high-standing central lava domes were emplaced toward the end of the caldera s evolution. Figure 7 shows the volcanic evolution of the Amealco caldera. Eruptions are characterized as Plinian, vulcanian, or strombolian, based on the observed qualitative characteristics of the deposits, rather than quantitatively, using grain-size and distribution analysis. K-Ar GEOCHRONOLOGY We obtained 25 K-Ar ages (Table 1) for units that represent the entire range of the Amealco caldera activity, and include some units of both precaldera and postcaldera volcanism. Some units have been dated with two or more samples from separate localities. Except for minor rhyolite, volcanic units in the Amealco region lack potassium-rich phenocryst phases. For these samples of Pliocene age we relied primarily upon glass or crystalline groundmass to achieve reasonable precision. For some samples (samples Am-1, Am-179, Am-63, Am-81; Table 1) paired matrix and feldspar ages were obtained, with acceptable agreement in three of four cases. K-Ar analyses were performed in the Department of Geological Sciences of the University of Texas at Austin, in a laboratory that has been commonly used for mid-tertiary or older samples. The techniques include flame photometry for K analysis and isotope dilution for Ar analysis using a Nuclide 3 in gas-source mass spectrometer. Details of the techniques can be found in Aguirre-Díaz et al. (this volume). Analytical uncertainties were mostly <5% of the age, and only a few exceeded 8%. The worst case was 22% for Am-122, a sample with high atmospheric argon content. For map units with multiple age determinations, an assigned age is calculated from a mean value that is weighted inversely in proportion to the precision of the individual results. VOLCANIC EVOLUTION OF THE AMEALCO CALDERA Precaldera volcanic rocks (>4.7 Ma) Most of the precaldera rocks are lava and pyroclastic rocks of intermediate or felsic composition. A K-Ar age of 5.7 Ma was obtained from basaltic andesite (sample Am-67, Table 1) that underlies caldera products 28 km to the north-northeast of the caldera, and a precaldera felsic ignimbrite 33 km to the west yielded a K-Ar age of 4.7 Ma (sample Am-84, Table 1).

4 4 G. J. Aguirre-Díaz and F. W. McDowell N N to Coroneo N W W W to Galindo W W S. Mateo Sierra El Rinc n cc Epitacio Huerta EPITACIO HUERTA FAULT E Palomas volcano Las Hormigas volcano CG Garabato Molinos 120 SJ Tm AMEALCO Tz Tlc Tch Tg Tc El Comal volcano 120 Tsr Teb LA SMT Loma Linda Tenango km A il N to Aculco N N W to Tepuxtepec AGE (Ma) Quaternary Pliocene- Quaternary Tepuxtepec reservoir W POSTCALDERA UNITS Reworked pumice and alluvium. Andesites, basaltic-andesites and related cones. El Comal cone and related andesitic lava flow and pyroclastic rocks. Garabato Andesite: hornblende andesite dome. El Rinc Huichapan Tuff. n Rhyolite lava domes W Lerma river M W to Temascalcingo W PRECALDERA UNITS Silicic lava dome. 5.7 Intermediate (andesitic?) lava flows and domes. Caldera outline Hormigas Andesite: basaltic andesite lavas. Coronita Rhyolite lava domes. CALDERA UNITS Santa Rosa Andesite: trachyandesitic central domes;tz: Zancudo, Tg: El Gallo, Tc: Capando, Tsr: Santa Rosa. Tm: La Mesa. La Cruz Rhyolite: central domes; Tlc: La Cruz, Tch: Chiteje, Teb: El Barco. Palomas Dacite: trachydacitic lavas and pyroclastic rocks. Spatter-lava cone. Campana Dacite: trachydacitic rim dome. Amealco Andesite: trachyandesitic rim domes. Brick Pumice: pumice flow and fall, with epiclastic deposits. Amealco tuff: trachyandesite-trachydacitic ignimbrites and interlayered surge, fallout and mud-flow deposits. Normal fault Volcano Town or village Paved road Unpaved road Lake River Figure 2. Simplified geologic map of the Amealco caldera and peripheral volcanism. CC: Cerrito Colorado cone, E: El Espía dome, SJ: San Juan Hedó, LA: Los Arcos, SMT: San Miguel Tlascaltepec, M: Mexquititlán, CG: Cañada de García. Plinian eruptions and associated pyroclastic flows (4.7 Ma) The earliest activity related to the Amealco caldera occurred around 4.7 Ma, when Plinian eruptions produced widespread pumice fallout and pyroclastic flows (e.g., Amealco zero ignimbrite; Figs. 3 and 7). This stage produced the earliest pyroclastic deposits of the Amealco Tuff that mainly consist of layered, white, medium- to coarse-grained, angular, pumice lapilli. Thickness of the initial fallout deposits is 3 m at a locality 25 km to the east, less than 1 m 15 km to the west of the caldera, and practically zero 30 km to the west. They

5 EXPLANATION 49.8 m Amealco III covered 62 m Huichapan ignimbrite Huichapan undifferentiated tephra Amealco III 73.5 m 70 Amealco III Intermediate lava Vesiculated intermediate lava Platy-jointed lava Brecciated lava Glassy lava Undulated pumice fall deposit Lithics-rich surge deposit Rounded lithics Fluvial deposit Recent alluvium Amealco II 32.5 m Ameal co III Ameal co II m Ameal co II Ameal co I 24 Amealco III Amealco I Amealco II Recent talus Ignimbrite with black pumices Coarse pumice fallout deposit Fine pumice fallout deposit Pumice flow deposit Ash-flow deposit Finely layered surge deposits Lithic-rich surge deposits Black, uncollapsed pumice fragments Black fiamme covered covered Amealco I Ameal co I Amealco II White, uncollapsed pumice fragments Accidental, angular lithics Vitrophyre of ignimbrite Monolithologic matrix-free deposit Mud-flow deposit Mafic lava flow Felsic ignimbrite 6 0 Section 1 Amealco Zero 6 0 Section Section Amealco I Section Section 11 Weathered pumice deposit Metamorphic rocks Paleosoil or varved lake deps. Reworked deposit Pumice veins Figure 3. Representative measured sections of the Amealco Tuff. Locations are shown in Figure 5.

6 6 G. J. Aguirre-Díaz and F. W. McDowell Figure 4. Representative measured sections of the Brick Pumice and Amealco Andesite caldera units. Locations are shown in Figure 5. Section symbols are shown in Figure 3. are also absent 28 km to the north-northeast of the caldera, and apparently 30 km to the south as well. Thus, paleowind transport of the Plinian column s material may have been dominantly to the east. Apparently a column collapse caused the pyroclastic flow that formed Amealco zero ignimbrite, which is observed only to the west of the caldera interlayered with pumice fall deposits (section 1, Fig. 3). The Amealco zero ignimbrite is nonwelded, dark gray when fresh, or red when weathered. It is mainly composed of black pumice, with minor amounts of lithics in a pumiceous matrix. The physical aspect of this deposit is like a reddish scoria flow. The Amealco zero yielded a minimum dense rock equivalent (DRE) volume of 0.75 km 3. Whole-rock pumice analysis of the Amealco zero yielded the most mafic composition of all the Amealco caldera products (SiO 2 = 61.6 wt. 3 ; this and subsequent values of SiO 2 are normalized volatile free; Table 2, Fig. 6). Caldera formation and emplacement of the Amealco Tuff ( Ma) Between 4.7 and 4.5 Ma caldera collapse occurred in response to the eruption of voluminous and widespread pyroclastic flows and pumice fallouts that produced the major ignimbrites and interbedded tephra of the Amealco Tuff (Figs. 3 and 7). The first voluminous pyroclastic flow deposited the Amealco I ignimbrite. This ignimbrite was preceded by eruption of surges, minor pyroclastic flows that deposited unwelded ignimbrites, pumice fallouts, and mud flows. This pattern was repeated at least two more times to produce the tephra 2 and Amealco II ignimbrite, and the tephra 3 and Amealco III ignimbrite (Fig. 3). Each of these cycles (tephra and major ignimbrite) may have formed a transient caldera, which was destroyed by subsequent events, and/or was buried by younger pyroclastic deposits. However, only one caldera structure is now evident. Alternatively, it is possible that multiple collapse events occurred in this single caldera (e.g., Druitt et al., 1994). The similar volumes of each cycle (major ignimbrite plus surge and fall tephra) support the model of several calderas. The first hypothetical caldera is related to a DRE volume of 21 km 3 (Amealco zero, tephra 1, and Amealco I); the second caldera is related to a DRE volume of 32 km 3 (Amealco II and tephra 2); and the third caldera is related to a DRE volume of 24 km 3. According to an empirical relationship of ignimbrite volume versus caldera diameter (Cas and Wright, 1987, p. 233), each cycle should have produced a caldera of about 11 km diameter (for 21 to 32 km 3 ), which is the actual caldera diameter. Several unconformities are present within the Amealco Tuff,

7 Volcanic evolution of the Amealco caldera, central Mexico 7 Figure 5. Distribution of the Amealco Tuff based on the distribution of the major ignimbrites Amealco I, Amealco II, and Amealco III. Figure indicates location of the measured sections. H: Huimilpan; T: Tlalpujahua. as indicated by paleosoils, irregular surfaces due to erosion, or thin layers of varved lake deposits (Fig. 3). These are particularly evident after major pyroclastic flow eruptions that produced the ignimbrites Amealco I, II, and III. Apparently, pauses occurred between the volcanic episodes that formed the Amealco Tuff. On the basis of mapping, the Amealco Tuff has an estimated minimum distribution of 2880 km 2, and a total DRE volume estimate of 77.7 km 3. These are minimum volume estimates because they do not include the intracaldera tuff, the distal lapilli-ash fallouts, and the coignimbrite ash-cloud deposits, which were eroded away. A detailed description of the procedure to estimate the volume of the Amealco Tuff can be found in Aguirre-Díaz (1993, 1996). The age of the Amealco Tuff is derived from five K-Ar ages (three on glass and two on plagioclase separates) from four separate localities (Table 1). The range of ages is from 4.54 to 4.74 Ma, and the weighted mean of 4.68 ± 0.10 Ma is chosen as the best estimate. The mean is close to the ages for the three glasses, which are more precise than those for the plagioclases because of their higher potassium contents. The plagioclase ages are lower, but their errors overlap with both the weighted mean and the individual glass ages. The bulk of the Amealco Tuff is trachyandesitic-trachydacitic in composition; most pumice fragments analyzed have a SiO 2 range from 61.6 to 65.5 wt% (Table 2, Fig. 6). The mineralogy observed in the Amealco Tuff is consistent with this composition, because it contains plagioclase, hypersthene, augite, ilmenite, titanomagnetite, accessory apatite, and rare olivine. Mingling of glass is clearly evident in the Amealco Tuff. The major ignimbrites and the interlayered tephra deposits of the Amealco Tuff contain pumices with distinct compositions (documented by electron-probe microanalysis; Aguirre-Díaz, 1993), and different colors (black, yellow, and white). Field and laboratory observations (Aguirre-Díaz, 1991, 1993) have shown that the different glasses are juvenile and are well mingled, even at the microscopic scale (mingled shards). The different liquids were ejected simultaneously to form unzoned deposits that do not show sorting of the different glasses. The eruptive mechanism that caused these deposits could have been similar to that proposed by Sparks et al. (1977), in which the different magmas could have been simultaneously erupted after stirring, caused by mafic magma input at the base of a zoned subcaldera magma chamber. This mechanism must have occurred repeatedly to produce the multiglass pyroclastic deposits of the Amealco Tuff, i.e., the Amealco I, II and III. Additional features indicate a complex eruptive history for the Amealco Tuff. For example, fallout deposits consisting of angular, coarse fragments of a single rock type, black glassy lava, are found intercalated in the sequence (section 1, Fig. 3). These deposits could be interpreted as vulcanian-type eruptions that destroyed lava bodies, perhaps domes, that may have been obstructing the vents. Thick stacks of pyroclastic surge deposits, that traveled as far as 28 km from source, indicate repeated phreatomagmatic pulses of great explosivity (section 11, Fig. 3). Stacks of mud-flow deposits are also interbedded with ash-flow deposits as far as 25 km from the source, and were common during the activity that produced the Amealco Tuff. Some of these deposits are lithic rich (20 35 vol% lithic material), others are mostly composed of sand-size material. Brick Pumice, rim lava domes (Amealco Andesite), Campana dome ( Ma) Explosive activity continued in the Amealco caldera after eruption of the Amealco Tuff. However, the pyroclastic material issued during these events was less voluminous and more localized (Fig. 2). These pumice deposits were named the Brick Pumice by Sánchez-Rubio (1984). The Brick Pumice was emplaced in the interval between 4.6 ± 0.1 Ma (the age of Amealco Tuff) and the 4.3 Ma emplacement of lava domes that form the rim of the caldera, named the Amealco Andesite (Fig. 7). The Amealco Andesite was contemporaneous with the last events of deposition

8 8 G. J. Aguirre-Díaz and F. W. McDowell TABLE 1. K Ar AGES OF AMEAICO CALDERA REGION Unit Sample Location Mat.* K Weight Ar x Ar Age ±1σ Assigned Lat Long Age (N) (W) (%) (g) (scc/g) (%) (Ma) (Ma) (Ma ± 1σ) Pre-Amealco Am Wr ± Am Fds ± Amealco Tuff Am Gl ± Am-1 Fds Am Fds Am Gl Am Gl Amealco And. Am Gms ± Am Wr Am Wr Am Wr Campana Daci. Am Gl ± Palomas And. Am-41b Gms ± La Cruz Rhyo. Am Gl ± Am-63 Fds Sta. Rosa And. Am Gms ± Am Gms Coronita Rhy. Am Gl ± Am-179 Fds Hormigas And. Am Gms ± Huichapan Tuff Am Gl ± Am-81 Fds

9 Volcanic evolution of the Amealco caldera, central Mexico 9 TABLE 1. K Ar AGES OF AMEAICO CALDERA REGION (continued - page 2) Unit Sample Location Mat.* K Weight Ar x Ar Age ±1σ Assigned Lat Long Age (N) (W) (%) (g) (cm 3 /g) (%) (Ma) (Ma) (Ma ± 1σ) El Rincón Rhy. Am Gl ± Garabato And. Am Wr ± El Comal And. Am-19b Gms ± *Material used for K-Ar analyses: Gl = glass; Wr = whole rock; Gms = groundmass; Fds = feldspar. Scc/g = standard cubic cm/gram. 40 Ar = radiogenic argon content of sample, in percent of total 40 Ar. **Errof of age at one sigma, see text for details. Assigned Age = weighted mean of the different ages obtained for a particular unit. 40 K/K = x 10-4 moles/mole; λβ = x /yr; λε + ε = x /yr. TABLE 2. REPRESENTATIVE WHOLE-ROCK CHEMICAL ANALYSES Product Ameaico Ignimbrites Rim Lava Domes Central Domes Sample Am-39 Am-35 Am-22 Am-21 Am-255 Am-256 Am-133b Am-195 Am-53 Am-61 Unit Zero Ame-1 Ame-II Ame-II Ame-III Ame-III Ame. And. Ame. And Santa Rosa Santa Rosa Latitude Longitude Quadrangle F14C88 F14C86 F14C86 F14C86 F14C86 F14C76 F14C86 F14C86 F14C86 F14C86 SiO TiO Al 2 O Fe 2 O FeO MnO MgO CaO Na 2 O K 2 O P 2 O H 2 O H 2 O CO Total Mg# CIPW Norm (wt. % volatile free) Q C Or Ab An Wo Di Hy Mt Il Ap Total TT index

10 10 G. J. Aguirre-Díaz and F. W. McDowell Figure 6. Total alkali-silica volcanic-rock classification diagram (Le Bas et al., 1986) of representative Amealco caldera products and peripheral volcanism. Symbols in figure are: Amealco Tuff ignimbrites: closed square; Amealco Andesite: open diamond; central lava dome (both rhyolite and trachyandesite): open square; Palomas Dacite: plus sign; Campana Dacite: X; Hormigas Andesite: open circle; Garabato Andesite: triangle; Comal Andesite: closed circle; Coronita Rhyolite: closed diamond. Abbreviations: b, basalt; ba, basaltic andesite; a, andesite; d, dacite; r, rhyolite; bta, basaltic trachyandesite; ta, trachyandesite; td, trachydacite (see Table 2 for chemical data). of the Brick Pumice, because they occur interlayered in several sites (sections 19 and 20, Fig. 4). The Amealco Andesite (3.8 km 3 DRE) was emplaced as ring-fracture lava domes (Fig. 7), forming a continuous rim of trachyandesitic lava domes and associated lava flows (SiO 2 = wt%, Table 2, Fig. 6) that formed the topographic expression of the caldera (Fig. 2). Several pumice flows consisting of coarse-grained, unsorted pumice deposits with a pumiceous matrix, and less frequently, pumice fallouts consisting of coarse-grained, sorted pumice deposits without matrix, were erupted through ring fracture vents (Fig. 7). The pumice flow and fall eruptions may have been foam fountains of sub-plinian type, because they produced massive, coarse-grained pumice flows that extended as far as 17 km from the caldera. Fallouts probably were also part of these eruptions, because pumice-fall deposits occur interlayered with the pumice-flow deposits. Initially, pumice flow dominated the activity, but pumice fallouts predominated at the end. The pumice flows are thick next to the caldera (>15 m, no base exposed), but thickness decreases quickly with distance; they are on the order of 2 m thick 2 or 3 km away from the rim. Close to the caldera rim lavas are interlayered with these tephra deposits (sections 19 and 20, Fig. 4). Farther (>2 km) from the caldera rim the pyroclastic deposits are interlayered with epiclastic deposits (section 22, fig. 4). Both pyroclastic and epiclastic deposits are mapped as the Brick Pumice unit, which makes up a total DRE volume of 2.4 km 3. The Campana dome was also emplaced at about 4.3 Ma (0.8 km 3, DRE). Although it also forms part of the rim of the caldera, this dome is higher (more viscous?) and is chemically a little more evolved (SiO 2 = 66.8 wt%; Table 2, Fig. 6) than the other rim domes. The uncertainty in the K-Ar ages of both the rim and Campana dome allows the Campana dome to be significantly later than the rim domes (Table 1), as the field relationships indicate. Palomas volcano (4 Ma) The Palomas volcano may have been already formed by 4 Ma (Fig. 7). However, the large uncertainty (±0.4 m.y., Table 1) of this K-Ar age allows the formation of the Palomas volcano any time between 4.4 and 3.6 Ma. Palomas lavas overlie the Brick Pumice to the south of the Palomas cone (Fig. 2). Palomas activity included scoria flows and tephra and lavaflow eruptions with a trachydacitic composition (SiO 2 = 65 wt%; Table 2, Fig. 6). These emissions gradually built a small composite cone that stands only 200 m above the adjacent lava plateau. Most of the Palomas lavas flowed to the west of the vent (Fig. 2), and formed a blocky lava (aa type) plateau on top of the Amealco Tuff, making up a volume of 0.8 km 3 (DRE). The lava plateau extends to 8 km to the west of Palomas cone, where it is displaced by the Epitacio Huerta fault (Fig. 2). Central lava domes: La Cruz Rhyolite and Santa Rosa Andesite ( ) At 3.9 ± 0.1 Ma, central lava domes of rhyolitic composition (SiO 2 = wt%; Table 2, Fig. 6) were emplaced mostly in the northern part of the Amealco caldera (Fig. 7). These domes formed the La Cruz Rhyolite, with a DRE volume of 2.4 km 3. They are exogenous domes of Pelean type (Blake, 1990), having steep sides, crumble breccias, and block and deposits. At 3.7 ± 0.25 Ma, several lava domes of trachyandesitic composition (SiO 2 = wt%; Table 2, Fig. 6) were emplaced in the center of the caldera and formed the Santa Rosa Andesite (Fig. 7), with a total DRE volume of 4.3 km 3. The uncertainty of the K-Ar ages overlap, but the ages suggest that the rhyolitic domes were earlier than the intermediate lava domes, as the field relationships indicate. The lavas of the undated trachyandesitic Capando dome (Fig. 2) overlie the rhyolite of the Chiteje de la

11 ~ 3.9 Ma Plinian eruptions from an inferred central vent produced pumice-fall deposits, surges, and nonwelded ignimbrites (e.g., Ignimbrite Zero) Ma Emplacement of rhyolitic central lava domes (La Cruz Rhyolite) Ma Caldera collapse(s?) associated to major ignimbrite eruptions from ring-fracture vents, accompanied by pumice fallouts, pyroclastic surges, mud flows, and vulcanian eruptions (Amealco Tuff) Ma Emplacement of trachyandesitic central lava domes (Santa Rosa Andesite). 3.7 Ma Eruption from ring-fracture vents of coarse-grained pumice flows and sparse pumice fallouts (Brick Pumice). ~ 4.3 Ma Emplacement of Coronita Rhyolite lava domes west of the caldera, and of basaltic andesite of Hormigas Volcano next and to the north of the caldera rim. 3.5 Ma Emplacement of ring-fracture lava domes and related lava flows with sporadic tephra eruptions (Amealco Andesite). ~ 4.3 Ma Eruption of felsic pyroclastic flows from a source east of the Amealco caldera (Huichapan caldera?) that produced the Huichapan Tuff. 2.5 Ma Pumice-flow and fallout activity persisted after the emplacement of the rim lava domes (Brick Pumice). Emplacement of the Campana dome. ~ 4.0 Ma Displacement of Amealco caldera by normal faulting along the Epitacio fault system. Emplacement of the Garabato lava dome. The dome was little displaced by faulting. Reworked material filled the tectonic depression south of the caldera. 2.2 Ma Emplacement of the Palomas volcano adjacent to the caldera margin. Eruption of lavas and pyroclastic rocks. Emplacement of El Comal Cone on the fault trace that displaced the caldera. Reworked material continue to fill the tectonic depression south of the caldera. Figure 7. Volcanic evolution of the Amealco caldera and peripheral volcanoes.

12 12 G. J. Aguirre-Díaz and F. W. McDowell Cruz dome. The same arrangement is observed between the lavas of the Santa Rosa dome and El Barco domes (Fig. 2). The relatively long time between emplacement of the rim domes (4.3 ± 0.1 Ma) and that of the central domes (3.9 ± 0.1 Ma) was sufficient for the magma remaining in the subcaldera magma chamber to undergo differentiation. It is suggested that a rhyolitic cap was formed at the top of the chamber as a result of this differentiation. This rhyolitic magma layer was underlain by trachyandesitic magma, which likely made up the bulk of the magma chamber. Foundering of blocks above the already fractured and faulted caldera roof may have squeezed out the viscous rhyolitic magma and produced the rhyolitic lava domes of the La Cruz Rhyolite; as block foundering proceeded, the underlying trachyandesitic magma was then squeezed out to produce the trachyandesitic lava domes of the Santa Rosa Andesite. Except for the emplacement of the central lava domes, no evidence of resurgence was observed. The negative evidence includes: (1) absence of a caldera moat, (2) absence of exposures of the intracaldera facies of the Amealco Tuff, and (3) absence of outward tilting of the intracaldera layered pumice deposits of the Brick Pumice (section 26, Fig. 4). All these features would be expected if the caldera had been inflated by resurgence; thus, there apparently was no resurgence in the Amealco caldera volcanic evolution. The central lava domes are regarded as exogenous. These lavas were viscous and did not flow far from the vent, but produced high-standing and steep-flanked domes. Coronita Rhyolite and mafic Hormigas volcano (3.7 Ma) Two rhyolitic domes were emplaced near the Amealco caldera, just south of San Mateo (Fig. 2), at about 3.7 Ma (Fig. 7). This age is the weighted mean of a glass age of 3.5 Ma and a feldspar age of 3.9 Ma (Table 1). There is no overlap of errors, but there is no reason to discard either of the two; thus, the weighted mean of the two results was used as the most representative age for this unit. The Coronita Rhyolite is a much more evolved rock relative to the Amealco caldera products (SiO 2 = 75.9 wt%; Table 2, Fig. 6), and probably was derived from a magmatic system different from that of the Amealco caldera. Several similar rhyolitic domes were emplaced around the Sierra Puruagua, which is a large volcanic complex 15 km to the west-southwest of Amealco caldera (Fig. 5). Thus, the magma chamber from which these domes originated was probably beneath the Sierra Puruagua volcanic complex. Also at 3.7 Ma, another small volcano, Hormigas, developed next to the caldera rim (Figs. 2 and 7), erupting a total DRE volume of 0.8 km 3. The volcano apparently only consists of olivinebearing, andesitic lava flows (SiO 2 = 55.8 wt%; Table 2, Fig. 6). Pyroclastic deposits are not observed. The Hormigas lavas reached as far as 5 km to the north of the vent. Similar to the Coronita Rhyolite, the products of the Hormigas volcano may have been derived from a different magmatic source than that of the Amealco caldera products (Aguirre-Díaz, 1993). Huichapan Tuff (3.5 Ma) The Huichapan Tuff is a voluminous and widespread felsic ignimbrite with associated surge, ash-flow, and mud-flow deposits that blanketed the eastern portion of the area at about 3.5 Ma (Fig. 7). The total volume of this unit has not been estimated, but the ignimbrite can be followed for more than 50 km to the east. The tuff probably originated from the Huichapan caldera by several major explosive eruptions. The caldera is about 40 km to the east of the contact between the Amealco and Huichapan Tuffs. This contact is about 20 km to the east of the Amealco caldera. The ignimbrite is welded and has well-developed columnar jointing (the cover photo of Cas and Wright, 1987, is this ignimbrite). Because of its proximity to the Amealco caldera, this ignimbrite has been misinterpreted as an Amealco ignimbrite (Sánchez-Rubio, 1984). The ignimbrite of the Huichapan Tuff is practically 100% glass, with sparse quartz and K-feldspar (both 2 10 vol%). From both glass and feldspar, K-Ar ages of 3.6 ± 0.1 and 3.4 ± 0.2 Ma, respectively, were obtained (Table 1). The weighted mean of these ages, 3.5 ± 0.2 Ma, was used as the most representative age of this unit. The Huichapan Tuff may be correlated with the San Francisco Tuff of Herrera and Milán (1981) and Milán et al. (1993) because of their similar physical aspect and mineralogy and same stratigraphic position. In addition, the Huichapan Tuff can be followed to the east from the contact with the Amealco Tuff to a position very close to the Huichapan caldera, where it is buried by the lavas of the Nopala volcano, which postdates the San Francisco Tuff and was emplaced on the eastern rim of the Huichapan caldera. El Rincón Rhyolite (2.9 Ma) The Sierra El Rincón is a rhyolitic lava dome complex that was emplaced after the Amealco Tuff at 2.9 Ma (Table 1) and is about 15 km north-northwest of the Amealco caldera (Fig. 2). The name of this unit is proposed as El Rincón Rhyolite. The rock is a perlitized obsidian. It is practically aphyric. The only phases observed were altered, mafic, euhedral crystals that only make 4% of the volume, still preserving a euhedral shape of olivine. The obsidian is a peraluminous, high-silica rhyolite (SiO 2 = 75.3 wt%, sample Am-122, Table 2). Although this unit, as the Huichapan Tuff, was not erupted by the Amealco caldera, it is an important peripheral unit. Both the Huichapan Tuff and El Rincón Rhyolite are useful to constrain the timing of the Amealco Tuff. Caldera displacement by normal faulting, Garabato Dome, El Comal cone ( Ma) The southern portion of the Amealco caldera was displaced by normal faulting between 3.7 and 2.5 Ma, which are the ages of the central lava domes and the Garabato dome, respectively (see following). Faulting took place along the south-facing Epitacio Huerta fault system, which changes orientation from east-west to

13 Volcanic evolution of the Amealco caldera, central Mexico 13 southeast-northwest from east to west (Figs. 1 and 2). The fault system is the northern boundary of the Acambay graben (Fig. 5). The displacement along this fault system must have been larger than m, the common height of the rim lava domes, as the southern caldera margin is totally missing. To the east of the Amealco caldera there is another fault system with an orientation similar to that of the Epitacio Huerta fault system (Fig. 2), and with scarps of m, with opposite sense of movement (northfacing faults). We and other workers (Suter et al., 1991) propose that this contrast in style of faulting may be due to an older structural discontinuity with a northwest orientation, corresponding to the Taxco San Miguel de Allende fault system (or Querétaro fracture zone, Fig. 1), which is overlain by the Amealco caldera. At 2.5 Ma, Garabato dome was emplaced from a conduit apparently located along the Epitacio Huerta fault system. The Garabato dome is hornblende andesite with an SiO 2 content of 62.6 wt% (Table 2, Fig. 6), and a DRE volume of 0.1 km 3. The dome was displaced by a normal fault, presumably part of the same system. However, the fault displacement of the lavas of the Garabato dome is only 25 m (Sánchez-Rubio, 1984). Apparently most displacement of the caldera had already occurred when the Garabato dome was emplaced. Thus, faulting mostly occurred between 3.7 and 2.5 Ma, perhaps closer to 2.5 Ma. The last volcanic episode in the Amealco caldera area was the development of the monogenetic El Comal scoria cone, at 2.2 Ma, making up a DRE volume of 0.15 km 3. It was formed predominantly by strombolian eruptions. A lava flow breached the cone on its northeastern flank. The El Comal products became chemically more evolved with time, from basaltic andesite (56.3 wt% SiO 2 ) to andesite (62 wt% SiO 2 ) (Fig. 6). The cone is in an advanced degree of erosion, but steep flanks of scoria-fall deposits are preserved. The cone was emplaced on the trace of the Epitacio Huerta fault system (Figs. 2 and 7). As the cone is unfaulted, fault movement must be older than 2.2 Ma. Elsewhere, between Epitacio Huerta and Coroneo (Fig. 2), there are reports of historical seismicity along this fault system that damaged the church of Coroneo by the turn of this century (Urbina and Camacho, 1913). There is also field evidence for more recent faulting in the form of colluvium deposits in contact with the Amealco Tuff along the Epitacio Huerta fault system, near Epitacio Huerta (Suter et al., 1995). These facts suggest that the Epitacio Huerta fault trace may be a system of several segments that have independent histories of movement. CONCLUSIONS The Amealco caldera is a well-preserved Pliocene volcanic center, 11 km in diameter, located in the central part of the Mexican Volcanic Belt. The caldera was formed by several discrete volcanic events: (1) eruption of 77 k 3 of trachyandesitic-trachydacitic magma that deposited the Amealco Tuff and caused caldera collapse between 4.7 and 4.5 Ma; (2) eruption from ring fracture vents of 2.4 km 3 of magma as tephra that formed the Brick Pumice unit; (3) eruption of 3.8 km 3 of trachyandesitic magma through several ring-fracture vents as lava domes at 4.3 Ma (Amealco Andesite); (4) emplacement of 0.8 km 3 of trachydacitic magma at 4.3 Ma on the northwestern caldera rim as the Campana dome; (5) development of Palomas volcano at 4.0 Ma, which represents 0.8 km 3 of trachydacitic magma; and (6) emplacement of 10 lava domes in the center of the caldera between 3.9 and 3.7 Ma, of which 5 are trachyandesitic (Santa Rosa Andesite) with a magma volume of 4.3 km 3, and 5 are rhyolitic (La Cruz Rhyolite) with a magma volume of 2.4 km 3. After emplacement of the central lava domes volcanic activity continued within and next to the caldera during the next 1.6 m.y. These events are: (1) eruption of basaltic andesite magma from Hormigas volcano at 3.7 Ma; (2) emplacement of rhyolite lava domes, also at 3.7 Ma (Coronita Rhyolite); (3) emplacement of a 2.5 Ma hornblende andesite dome (Garabato Andesite); and (4) the formation of the andesitic El Comal scoria cone at 2.2 Ma. In addition to the peripheral volcanism next to the caldera, the Amealco Tuff is overlain 20 km to the east of the caldera by a 3.5 Ma widespread felsic ignimbrite and related tephra deposits (Huichapan Tuff), and by a 2.9 Ma rhyolitic lava dome complex 15 km north of the caldera (Sierra El Rincón). The southern portion of the Amealco caldera was displaced at least m by normal faulting along a segment of the Epitacio Huerta fault system. Faulting on this segment must have occurred between 3.7 and 2.5 Ma, and closer to 2.5 Ma, because the 2.5 Ma Garabato dome is displaced only 25 m by this faulting. At least in the vicinity of the caldera, faulting cannot be later than 2.2 Ma, the age of the El Comal cone, which is not displaced, although it is on the trace of the fault. ACKNOWLEDGMENTS This was greatly improved by comments on earlier versions by Daniel S. Barker and Bruce N. Turbeville. We thanks Scott Thieben, who provided assistance and supervision to gather the chemical analyses. This study was supported in part by endowments to Daniel S. Barker, the Geology Foundation of the Department of Geological Sciences of the University of Texas at Austin, a scholarship to Aguirre-Díaz from Dirección General de Asuntos del Personal Académico of Universidad Nacional Autónoma de México (DGAPA-UNAM), and by grant IN to Aguirre-Díaz from Programa de Apoyo a Proyectos de Investigación e Inovación Tecnológica of DGAPA-UNAM. The K-Ar laboratory at the University of Texas at Austin has been supported by National Science Foundation grants EAR and EAR to McDowell. REFERENCES CITED Aguirre-Díaz, G. J., 1991, Mixing of liquids in ignimbrites of Las Américas Formation, Central Mexican Volcanic Belt: Eos (Transactions, American Geophysical Union), v. 72, p Aguirre-Díaz, G. J., 1993, The Amealco caldera, Querétaro, México: Geology,

14 14 G. J. Aguirre-Díaz and F. W. McDowell geochronology, geochemistry, and comparison with other silicic centers of the Mexican Volcanic Belt [Ph.d. thesis]: Austin, University of Texas, 401 p. Aguirre-Díaz, G. J., 1996, Volcanic stratigraphy of the Amealco caldera and vicinity, central Mexican Volcanic Belt: Revista Mexicana Ciencias Geologícas, v. 13, p Blake, S., 1990, Viscoplastic models of lava domes, in Fink, J. H., ed., Lava flows and domes, IAVCEI Proceedings in Volcanology 2: New York, Springer- Verlag, p Carrasco-Núñez, G., 1988, Geología y petrología de los campos volcánicos de Los Azufres (Mich.), Amealco y El Zamorano (Qro.) [tesis maestría]: México, D. F., Facultad Ingeniería, Universidad Nacional Autónoma de México, 148 p. Cas, R. A. F., and Wright J. V., 1987, Volcanic successions. Modern and ancient: London, Allen & Unwin, 528 p. Cathelineau, M., Oliver, R., and Nieva, D., 1987, Geochemistry of volcanic series of the Los Azufres geothermal field (México): Geofísica Internacional, v. 26, p Demant, A., 1978, Características del Eje Neovolcánico Transmexicano y sus problemas de interpretación: Universidad Nacional Autónoma de México, Instituto de Geología, Revista, v. 2, p Dobson, P. F., and Mahood, G. A., 1985, Volcanic stratigraphy of the Los Azufres geothermal area, Mexico: Journal of Volcanology and Geothermal Research, v. 25, p Druitt, T. H., Brenchley, P. J., Gökten, Y. E., and Francaviglia, V., 1994, Evolution of the Acigöl caldera, Turkey: International Volcanolology Congress: International Association of Volcanology and Chemistry of Earth s Interior (AVCE), Abstracts volume: Ankara, Turkey, Middle East Technical University. Ferrari, L., Garduño, V. H., Pasquarè, G., and Tibaldi, A., 1993, The Los Azufres caldera, México: The result of multiple nested collapse: Reply: Journal of Volcanology and Geothermal Research, v. 56, p Ferriz, H., and Mahood, G. A., 1984, Eruption rates and compositional trends at Los Humeros volcanic center, Puebla: Journal of Geophysical Research, v. 89, p Ferriz, H., and Mahood, G. A., 1987, Strong compositional zonation in a silicic magmatic system: Los Humeros, Mexican Volcanic Belt: Journal of Petrology, v. 28, p Herrera, F., and Milán, M., 1981, Estudio geológico de las zonas geotérmicas de Yexthó, Pathé y Taxidó, estodos de Hidalgo y Querétaro: Informe técnico C.F.E (inédito). Johnson, C. A., and Harrison, C. G. A., 1990, Neotectonics in central Mexico: Physics of the Earth and Planetary Interiors, v. 64, p Johnson, W. M., and Maxwell, J. A., 1981, Rock and mineral analysis (second edition): Malabar, Florida, Robert E. Krieger Publishing Company, 489 p. Le Bas, M. J., Le Maitre, R. W., Streckheisen, A., and Zanettin, B., 1986, A chemical classification of volcanic rocks based on the total alkali-silica diagram: J. Petrol. v. 27, p Mahood, G. A., 1980, Geological evolution of a Pleistocene rhyolitic center Sierra La Primavera, Jalisco, Mexico: Journal of Volcanology and Geothermal Research, v. 8, p Mahood, G. A., 1981, Chemical evolution of a Pleistocene rhyolitic center Sierra La Primavera, Jalisco, México: Contributions to Mineralogy and Petrolology, v. 77, p Mahood, G. A., and Drake, R. A., 1983, K-Ar dating young volcanic rocks: A case study of the Sierra La Primavera, Jalisco, México: Geological Society of America Bulletin, v. 93, p Mahood, G. A., and Halliday, A. N., 1988, Generation of high-silica rhyolite: A Nd, Sr, and O isotopic study of Sierra La Primavera, Mexican Neovolcanic Belt: Contributions to Mineralogy and Petrology, v. 100, p Milán, M., Yáñez, C., Navarro, I., Verma, S. P., and Carrasco, G., 1993, Geología y geoquímica de elementos mayores de la caldera de Huichapan, Hidalgo, México: Geofísica Internacional, v. 32, p Nixon, G. T., 1982, The relationship between Quaternary volcanism in central Mexico and the seismicity and structure of subducted ocean lithosphere: Geological Society of America Bulletin, v. 93, p Nixon, G. T., Demant, A., Armstrong, R. L., and Harakal, J. E., 1987, K-Ar and geologic data bearing on the age and evolution of the Trans-Mexican Volcanic Belt: Geofísica Internacional, v. 26, p Pradal, E., and Robin, C., 1994, Long-lived volcanic phases at Los Azufres volcanic center, Mexico: Journal of Volcanology and Geothermal Research, v. 63, p Sánchez-Rubio, G., 1978, The Amealco caldera: Geolological Society of America Abstracts with Programs, v. 10, p Sánchez-Rubio, G., 1984, Cenozoic volcanism in the Toluca-Amealco region, central Mexico [Master thesis]: London, Imperial College of Science and Technology, University of London, 275 p. Singh, S. K., Pardo, M., 1993, Geometry of the Benioff zone and state of stress in the overriding plate in central Mexico: Geophysical Research Letters, v. 20, p Smith, R. L., 1979, Ash-flow magmatism, in Geological Society of America Special Paper 180, p Smith, R. L., and Bailey, R. A., 1968, Resurgent cauldrons, in Coats, R. R., Hay, R. L., and Anderson, C. A., eds., Studies in volcanology: Geological Society of America Memoir 116, p Sparks, R. S. J., Sigurdsson, H., and Wilson, L., 1977, Magma mixing: A mechanism for triggering acid explosive eruptions: Nature, v. 267, p Suárez, G., Monfret, T., Wittlinger, G., and David, C., 1990, Geometry of subduction and depth of the seismogenic zone in the Guerrero gap, Mexico: Nature, v. 345, p Suter, M., Aguirre, G. J., Siebe, C., Quintero, O., and Komorowski, J. C., 1991, Volcanism and active faulting in the central part of the Trans-Mexican volcanic belt, in Walawender, M. J., and Hanan, B. B., eds., Geological excursions in southern California and Mexico: Department of Geological Sciences, San Diego State University, p Suter, M., Quintero, O., López-Martínez, M., Aguirre-Díaz, G., and Farrar, E., 1995, The Acambay graben: Active intra-arc extension in the Trans- Mexican Volcanic Belt, Mexico: Tectonics, v. 14, p Thornton, C. P., and Tuttle, O. F., 1960, Chemistry of igneous rocks: I. Differentiation index: American Journal of Science, v. 258, p Urbina, F., and Camacho, H., 1913, La zona megasísmica Acambay-Tixmadejé, Estado de México, conmovida el 19 de noviembre de 1912: Boletín del Instituto Geología de México, v. 32, 125 p. Urrutia-Fucugauchi, J., and Del Castillo, L., 1977, Un modelo del Eje Volcánico Transmexicano: Boletín de la Sociedad Geológica Mexicana, v. 38, p Verma, S. P., 1984, Alkali and alkaline earth element geochemistry of Los Humeros caldera, Puebla, Mexico: Journal of Volcanology Geothermal Research, v. 20, p Verma, S. P., Carrasco-Núñez, G., and Milán, M., 1991, Geology and geochemistry of Amealco caldera, Qro., Mexico: Journal of Volcanology Geothermal Research, v. 47, p MANUSCRIPT ACCEPTED BY THE SOCIETY AUGUST 31,1998 Printed in U.S.A.