PaleoBios, Volume 17, Numbers 2-4, Pages 13-27, September 13,1996 Fossil mammals as a tool in Andean stratigraphy: dwindling evidence of Late Cretaceous volcanism in the South Central Main Range Andre R. Wyss1, Reynaldo Charrier2, and John J. Flynn3 'Department of Geological Sciences, University of California, Santa Barbara, CA 93106, email: wyss@geology.ucsb.edu Departamento de Geologia, Universidad de Chile, Casilla 13518 Correo 21, Santiago, Chile, email: rcharrie@tamarugo.cec.uchile.cl Department of Geology, Field Museum of Natural History, Roosevelt Rd. at Lake Shore Drive, Chicago IL 60605, email: flynn@fmppr.fmnh.org ABSTRACT A major central Andean volcanic/volcaniclastic unit, the Abanico (=Coya-MachaIi) Formation, was long considered Late Cretaceous in age—at least in part, and as lacking chronologically useful fossils. Recent unanticipated discovery of Cenozoic mammalian remains in this unit in central Chile, in conjunction with new isotopic geochronometric evidence, overturns both of these views, necessitating considerable revision of the accepted geotectonic history of the region. Two fossil mammal faunas have been recovered in the Rio Tinguiririca drainage (34° 50' S) of the Andean Main Range. The first discovered and better sampled of the two faunas represents a new biostratigraphic interval, of transitional Eocene-Oligocene age (pre-Deseadan, post- Mustersan, South American Land Mammal Age, SALMA) (31.4 - 37.7 Ma); it is derived from a horizon directly overlying Mesozoic units. The second fauna is unquestionably significantly older (?Early Eocene, Casamayoran SALMA) than the first. This contradicts previous stratigraphic interpretations of the region which place horizons producing the older fauna some 2000 m higher in the local stratigraphic section than those producing the younger, indicating an unforeseen degree of tectonic complexity for this Andean segment. Two additional Cenozoic mammal faunas have been discovered within the Abanico (=Coya-MachaIi) Formation north of the Tinguiririca valley region. Aside from a preliminary description of a primate skull, neither of these faunas has yet been studied, but preliminary biostratigraphic and radioisotopic information indicates both to be younger than the younger fauna from the Tinguiririca region. This has important geological ramifications inasmuch as both faunas occur in what was previously considered the oldest part of the formation, its oft-cited "eastern strip" occurring north of about 34° 40' S. Consequently, deposition of much (if not all) of the Abanico (=Coya-Machali) Formation cannot be attributed to an episode of Late Cretaceous volcanism generally posited for the region. These finds of well preserved, temporally distinct, and geographically widespread fossils in the Abanico (=Coya-Machali) Formation are of demonstrated importance to the interpretation of central Andean tectonics, establishing fossil vertebrates as an important new source of geochronologic information for this region. INTRODUCTION Despite the considerable attention the central Andes have received as the world's "type" example of a "simple" subduction-related orogenic belt, cur- rent knowledge about the timing of certain events in the geologic history of the region, particularly during the late Mesozoic and early Cenozoic, re- mains surprisingly limited. Such information, how- ever, is fundamental to understanding the relation- ship between patterns of Andean tectonism, sub- duction, and subsidence in bordering regions, and is central to interpreting the geotectonic history of the central Andean Cordillera in general. The bulk of the Andean Main Range in central Chile consists of an approximately 10 km thickness of Jurassic to early Late Cretaceous marine and terrestrial detrital back-arc, and Late Cretaceous to Cenozoic continental (mainly volcanic and volcaniclastic) arc deposits (Charrier, 1981a). Ow- ing to excellent biostratigraphic control in marine units of the lower half of this sequence, their ages and those of bracketing terrestrial units are tightly constrained, making correlation to units in adjacent western Argentina straightforward. Such clarity fails to extend, however, to age determination and correlation for the post-Neocomian continental de- posits comprising the upper 5 km of this sequence. Problems stem from, (1) the previous near complete lack of paleontologic control for these units, and (2) pervasive low grade metamorphism which has lim- In The Uses of Vertebrate Fossils in Biostratigraphic Correlation (C. J. Bell and S. S. Sumida, eds.). PaleoBios v. 17(2-4) Page 14 Fossil Mammals in Andean Stratigraphy Wyss et al. ited the utility of "whole rock" isotopic age determi- nations. Our recent discovery of abundant fossil mammals from at least three temporally distinct and geographically widely spread stratigraphic lev- els within this 5 km sequence, as well as the advent of single crystal laser fusion ^Ar/^Ar dating, cir- cumvent these earlier limitations, making possible refined age estimates and facilitating correlations for a major stratigraphic sequence in this segment of the Andes. STRATIGRAPHIC CONTEXT As mentioned, the lower half of the cover rock sequence on the Chilean slope of the Andean Main Range contains multiple fossiliferous marine inter- calations, making correlation to units on the Argen- tine slope and associated basins unambiguous. The youngest of these unproblematic correlations is be- tween the marine Chilean Banos del Flaco Forma- tion (plus its equivalent Lo Valdes Formation) and the Mendoza Group of Argentina, which are widely accepted as Tithonian to Hauterivian in age (Klohn, 1960; Biro-Bagoczky, 1964, 1984; Davidson and Vicente, 1973; Thiele, 1980; Figure 1). Age control for the 5 km of continental sedi- mentary and volcanic deposits overlying the Chil- ean Banos del Flaco (= Lo Valdes) Formation is meager. The 3 km thick, dominantly sedimentary Colimapu Formation rests conformably on the Banos del Flaco Formation and has been assigned an Albian age on the basis of charophytes (Martinez and Osorio, 1963). Its base, however, is widely regarded as Aptian (see Malumian et al., 1983, table II therein, for example), and the unit as a whole is generally correlated to the Rayoso Group of Argen- tina (Figure 1). Until recently the Colimapu Forma- tion charophytes represented the stratigraphically highest biostratigraphic control in the Andean Main Range sequence of Chile. There are no published radioisotopic ages for the Colimapu Formation. The next youngest unit in the Chilean Main Range sequence, the dominantly volcanic and volcaniclastic Abanico (=Coya-Machali) Formation, poses the greatest uncertainties in terms of age and correlation. North of approximately 34° 40' S this unit forms two parallel, north-south oriented swaths (Figure 2). Its eastern swath, once assigned a poorly constrained Coniacian through Maastrichtian age, rests unconformably on the Colimapu Formation and has been roughly corre- lated to the Neuquen (Klohn, 1960; Yrigoyen 1979; Malumian et al, 1983) and Malargiie (Barrio, 1990, figure 5 therein) groups in Argentina. The Abanico (=Coya-Machali) Formation was realized to contain age diagnostic fossils only during the last few years (see below). 40K/40Ar analyses of this unit pub- lished during the late 1970's and early 1980's yielded highly variable (Paleocene-Miocene) age estimates (Vergara and Drake, 1978; Drake et al., 1982). Low grade metamorphism called the reli- ability of these determinations into question (ibid.), and prior to our recent discovery of Cenozoic mam- mal fossils a Late Cretaceous age was assumed for at least the eastern swath of this unit (see Charrier and Muhoz, 1994). The 2500 m thick, dominantly volcanic and volcaniclastic Farellones Formation unconformably overlies the Abanico (=Coya-Machali) Formation over much of the western flank of the central Andean Main Range. The Abanico (=Coya- Machali) and Farellones formations are closely similar lithologically, to the point that in the absence of evidence of their intervening unconformity, they are often impossible to distinguish in the field. The Farellones Formation is not known to contain biostratigraphically useful fossils. It was initially considered Late Cretaceous through early Cenozoic in age (Aguirre, 1960), but more recent ^K/^Ar studies suggest it to be Miocene to perhaps Pliocene in age (Charrier and Munizaga, 1979; Vergara et al., 1988). As it is generally less metamorphosed than the Abanico (=Coya-Machali) Formation, radioisotopic age determinations for the Farellones Formation are generally considered credible (ibid.). It merits note, however, that several ^K/^Ar age determinations for the Farellones Formation overlap rather broadly with those from the underlying Abanico (=Coya-Machali) Formation, Figure 1. Simplified stratigraphic columns of the western slope of the Andean Main Range (left), and eastern slope of the Main Range and Neuquen Basin (right) at approximately 35° S, showing correlations (after Davidson and Vicente 1973, Yrigoyen, 1979, Charrier 1981a, 1981b, and Charrier et al, in press). The main point of departure from most previously proposed schemes is the suggested correlation of the entire Abanico (=Coya-Machali) Formation to the Agua de la Piedra Formation-and equivalents-rather than to the Neuquen and Malargiie groups. The new mammal faunas and recently published radioisotopic dates discussed herein are all derived from the Abanico (=Coya-Machali) Formation. The precise stratigraphic position of the dated and fossiliferous levels within the formation is obscured by marked tectonic complexity. As discussed in the text, a Late Cretaceous age for some small portion of the Abanico (=Coya-Machali) Formation cannot presently be ruled out. Breaks in stratigraphic columns signify unconformities. Wyss et al. Fossil Mammals in Andean Stratigraphy Page 15 Andean Main Range Chilean Slope Andean Main Range Argentine Slope and Neuquen Basin Quaternary volcanics E c •m j o (> O JS F en n a m m ¦a en < Q. F hal ll o m o o 2 c m ro >% X} o < o ii c • Tristeza Fm. Palomares Fm. Santa Maria Conglomerates Cenozoic Coihueco Fm. Pircala-Carrizo fms. Roca-Jaguel fms. Loncoche-Allen fms. Maastrichtian Jurassic Cretaceous Rio Colorado Fm. Rio Neuquen Fm. Campanian Santonian Coniacian Diamante Fm. Huitrfn Fm. Cenomanian Albian Aptian Agrio/ Mulichinco/ Quintuco Vaca Muerta fms. Barrem./Neocom. Tithonian Tordillo Fm. Kimmeridgian Auquilco/ La Manga/ Lotena fms. Oxford./ Callov. - Upper Cuyano Gp. Bajocian 600 m Nacientes del . Teno Fm. I_____I I I Correlations of Chilean and Argentine sequences. Conformable contact Unconformable contact Page 16 Fossil Mammals in Andean Stratigraphy Wyss et al. possibly reflecting incorrect attribution of the dated sample to the two formations. RECENT FINDINGS Abundant Cenozoic mammals have been dis- covered in volcaniclastic sediments from at least three stratigraphic levels within the upper (post- Neocomian) part of the Chilean slope Main Range sequence during the past several years. Two of these faunas are from near the town of Termas del Flaco in the valley of the Rio Tinguiririca (34° 50' S) (Figure 2). The younger of these faunas, recently designated the Tinguiririca Fauna (Wyss et al., 1994a), has been recovered from several sites (local- ity #1, Figure 2) in deposits formerly mapped as pertaining to the Colimapu Formation (and hence considered Aptian/Albian in age); a transitional Eocene-Oligocene age, however, is clearly indicated by the fossils (Wyss et al., 1994a), ""K/^Ar determi- nations (Wyss et al., 1990), and ^Ar/^Ar analyses (Flynn et al., 1991; Wyss et al., 1993,1994a; Charrier et al., in press), the latter of which securely bracket the fauna between approximately 31.4 - 37.7 Ma. The formational assignment of this horizon is open to little question. Lithologic criteria and differences in strike and dip clearly differentiate these deposits from the Cretaceous unit which they unconformably overlie (Charrier et al., 1990; Charrier et al., in press); based on recent mapping and sedimentological analysis, this fossil bearing unit is referred to the Abanico (=Coya-Machali) Formation (Charrier et al., 1994, in press; Zapatta, 1995). It is important to note, in this connection, that the Tinguiririca Fauna is not assignable to the Divisaderan South American Land Mammal Age (SALMA) as suggested in two recent reviews (Marshall and Sempere, 1993; Sempere et al., 1994). Rather, the Tinguiririca Fauna represents a new biostratigraphic interval interposed between the Deseadan and Mustersan SALMAs (Wyss et al., 1993; 1994a). A taxonomic list for the Tinguiririca Fauna is given in Table 1. The second fossiliferous horizon in the Tinguiririca valley occurs as a series of localities approximately 14 km northwest of the sites produc- ing the Tinguiririca Fauna (Wyss et al., 1992a, b). This mammal assemblage, designated the Tapado Fauna (Wyss et al, 1994a; locality #2, Figure 2), is derived from deposits which have consistently been mapped as belonging to the Abanico (= Coya- Machali) Formation. Only a limited number of Tapado Fauna specimens have been prepared and identified to date, and no radioisotopic analyses have yet been carried out for this area. Nonethe- less, available biostratigraphic evidence (Table 2) is sufficient to provisionally assign this fauna to the ?Early Eocene Casamayoran SALMA (Wyss et al., 1994a). These two Tinguiririca region faunas sug- gest, therefore, that previous 40K/mAi estimates in- dicating a Tertiary (rather than Mesozoic) age for the Abanico (=Coya-Machali) Formation are per- haps more plausible than generally assumed, and that the greater part of the unit (at least in this area) is substantially younger than previously thought (Wyss et al., 1990; Charrier et al., 1990, in press; Charrier and Muhoz, 1994). As discussed, north of the Tinguiririca region, between approximately 34° 40' and 32° 50' S, expo- sures of the Abanico (=Coya-Machali) Formation are divided into parallel eastern and western swaths by the Farellones Formation (Figure 2); the struc- tural and stratigraphic relationships of the two Abanico (=Coya-Machali) Formation swaths in this area remain uncertain (Charrier, 1981b). Given the then accepted Late Cretaceous age of the eastern swath, radioisotopic evidence from the Tinguiririca region was used to suggest derivation of the two early Tertiary mammal faunas of this region (34° 50' S) from levels likely corresponding to the western swath of this formation further to the north (Charrier et al., in press). That the bulk of the Abanico (=Coya-Machali) Formation in the Tinguiririca region is considerably younger than previously thought, called for reconsideration of the age of the formation's eastern swath. Successful paleontological reconnaissance in the eastern swath of the Abanico (=Coya-Machali) Formation north of the Tinguiririca region during January, 1994 provided precisely such an evalua- tion (Charrier et al., 1994; Wyss et al., 1994b, 1994c; Flynn et al., 1995). The northern of the two newly discovered Cenozoic mammal faunas occurs in ho- rizons near the core of a tightly folded anticline in the drainage of the Rio Maipo approximately 110 km north of the Tinguiririca region (locality #4, Figure 2). The second of the new eastern swath faunas is from along the Rio Las Lenas (of the Rio Cachapoal drainage basin) approximately 60 km north of the Tinguiririca valley (locality #3, Figure 2), also from highly folded volcaniclastic sediments. Ongoing preparation, study, and radioisotopic analysis will elucidate the taxonomic composition and age of these northern faunas. Both faunas con- tain hypsodont notoungulates, marsupials, diverse rodents, and other groups of eutherians, often rep- resented by exceptionally preserved material in- cluding skulls. Notable is a nearly complete an- thropoid skull recovered from the Las Lenas local- Wyss et al. Fossil Mammals in Andean Stratigraphy Page 17 Figure 2. Simplified map of the Chilean Main Cordillera between 33° 30' and 35° 30' S showing exposure of the mammal-bearing Abanico (=Coya-Machali) Formation, and position of various localities discussed in the text (formation is identified by stipple fill pattern). 1 identifies the primary Tinguiririca Fauna (of transitional Eocene- Oligocene age) producing locality. Additional sites producing this fauna occur approximately 1 km west of the tip of the arrow and immediately north of the Rio Tinguiririca (see Wyss et al., 1994a). 2 identifies the primary Tapado Fauna (?early Eocene) producing locality. An additional site producing this fauna occurs immediately north of the Rio Tinguiririca. 3 identifies a site discovered in January, 1994 producing early Miocene mammals from the valley of the Rio Las Lenas in the upper Rio Cachapoal drainage basin (Flynn et al, 1995). Sites producing this fauna occur north of the river as well. 4 identifies a mammal locality of similar age as the Las Lenas Fauna in the upper drainage of the Rio Maipo, also discovered in January, 1994. Note two parallel swaths of Abanico (=Coya-Machali) Formation exposure north of approximately 34° 45' S; these are separated by a swath of the Farellones Formation (shown in diagonal fill). Page 18 Fossil Mammals in Andean Stratigraphy Wyss et al. ity (Flynn et al., 1995), representing the earliest well known New World monkey. Preliminary biostrati- graphic and radioisotopic evidence indicates both faunas to be significantly younger than the two faunas from the Tinguiririca region (i.e the Tinguiririca and Tapado faunas); ""Ar/^Ar analy- ses of matrix associated with the Las Lenas monkey have yielded an age of 20.09 ± 0.27 Ma (ibid.). Al- though they appear to be roughly contemporane- ous, the precise temporal relationship of the two new eastern swath faunas remains to be determined. Finally, these two faunas suggest that the eastern and western swaths of the Abanico (=Coya-Machali) Formation are roughly similar in age, contrary to earlier suggestions. IMPLICATIONS The newly demonstrated occurrence of Eocene and younger fossil mammal faunas at four geo- graphically distant locales in the Abanico (=Coya- Machali) Formation has broad implications for in- terpretation of central Andean tectonic evolution. Perhaps most obvious are, (1) the effects this has on correlations between the Chilean and Argentine sequences, which in turn influences current discus- sions concerning: (2) the timing and relationship between deformation, uplift, and volcanism in the Andes, and changes in subduction at the plate mar- gin; (3) the timing and linkage between Andean tectonism and patterns of subsidence in bordering basins (such as in the neighboring Neuquen re- gion); and (4) Andean structural style. These issues are considered in turn below. Correlation As previously noted, the youngest secure cor- relation between the stratigraphic sequence of the central Chilean Andean Main Range and associ- ated back-arc deposits is between the Tithonian- Hauterivian Banos del Flaco Formation in Chile and the Mendoza Group in Argentina (Klohn, 1960; Davidson and Vicente, 1973; Figure 1). Among the correlations suggested by new paleontologic and radioisotopic evidence, the one most at odds with previous proposals concerns the Abanico (=Coya- Machali) Formation. The eastern swath of this unit Table 1. Faunal list for the Tinguiririca Fauna of the Tinguiririca valley, central Chile. Identifications are based on direct comparison with relevant collections in North and South American museums more fully detailed elsewhere (Wyss et al., 1994a). Of greatest biostratigraphic importance are the polydolopid, notostylopid, and "notopithecine" (latest known occurrences for these groups), and the rodents, "advanced rhynchippines," and interatheriines (earliest known occurrences). The previously undocumented co-occurrence of these taxa, plus additional biostratigraphic information demonstrate a pre-Deseadan, post-Mustersan "age" for this fauna. (Modified from ibid.) Marsupialia Notoungulata (continued) Groeberiidae Interatheriidae gen. et sp. nov. "Notopithecinae" Polydolopidae gen. et sp. nov. Polydolops, sp. nov. Interatheriinae Didelphimorphia inc. sed. gen. et sp. nov. A gen. et sp. nov. Archaeohyracidae Eutheria Pseudhyrax cf. entracytheroides Edentata cf. "Bryanpattersonia sulcidens" Dasypodidae inc. sed. new taxon A Tardigrada new taxon B Pseudoglyptodon sp. nov. new taxon C Litopterna new taxon D Adianthidae Homalodotheriidae Indaleciinae Trigonolophodon cf. elegans gen. et sp. nov. Notohippidae Rodentia "Rhynchippinae" ?Dasyproctidae Eomorphippus sp. nov. gen. et sp. nov. "Eomorphippus" cf. pascuali (second taxon, unstudied) new taxon A Notoungulata new taxon B Notostylopidae Isotemnidae inc. sed. ?gen. et sp. nov. Leontiniidae inc. sed. Wyss et al. Fossil Mammals in Andean Stratigraphy Page 19 Table 2. Preliminary faunal list for the Tapado Fauna of the Tinguiririca valley central Chile. Identifications are based on comparison with relevant North and South American museum collections; in some cases material remains incompletely prepared and identifications are consequently provisional. Many other specimens from this locality are unprepared and are presently visible only in cross section—these are not included in this tabulation. Secure placement of this fauna within the SALMA sequence awaits further study; the exclusive occurrence of brachydont herbivorous taxa is perhaps the single most persuasive piece of evidence that the Tapado Fauna is older than the Tinguiririca Fauna. Our provisional assignment of the Tapado Fauna to the Casamayoran SALMA is based largely on comparisons with an undescribed fauna from northern Argentina. It is worthy to note, in this regard, that despite its relative geographic proximity to Patagonia, the Tapado Fauna shows several distinct resemblances to faunas from well north of that region (as does also the Tinguiririca Fauna; Wyss et al, 1994a). Eutheria "Notopterna" "Notonychopidae" gen. et sp. nov. "Didolodontidae" Emestokokenia sp. nov. Notoungulata Notostylopidae ?gen. et sp. nov. Archaeohyracidae cf. Eohyrax isotemnoides Notohippidae gen. et sp. inc. sed. Isotemnidae gen. et sp. inc. sed. Interatheriidae gen. et sp. inc. sed. has customarily been assigned a poorly constrained Coniacian through Maastrichtian age, usually be- ing loosely correlated to the Neuquen (Klohn, 1960; Yrigoyen 1979) or Malargiie (Barrio, 1990, figure 5) groups—or both (Malumian et al., 1983)—in Ar- gentina (Figure 1). On the contrary, the four dispar- ate fossil-producing and/or single crystal radioiso- topically dated sites we have identified to date are consistent in supporting a much younger age for this unit. Rather than correlating with sediments of the Riograndico Supersequence (=Neuquen + Malargiie groups) in Argentina, the Abanico (=Coya-Machali) Formation should more appropri- ately be considered a western temporal equivalent of the Agua de la Piedra Formation (and roughly correlative units such as the Palomares Formation and Santa Maria Conglomerates, Yrigoyen, 1979, 1993; Figure 1). This proposed correlation of the Chilean Abanico (=Coya-Machali) and the Argentine Agua de la Piedra formations implies drastic upward shifts in correlation of the Farellones Formation (the unit overlying the Abanico Formation) to the Ar- gentine section as well. The Farellones Formation has been regarded by some authors as representing the western volcanic equivalent of the Malargiie Group (Yrigoyen, 1979), or as interfingering with the Santa Maria Conglomerate (Ramos, 1988). If, as we suggest, the Abanico (=Coya-Machali) Forma- tion is the rough temporal equivalent of the Santa Maria Conglomerate, the Farellones Formation (which is separated from the Abanico (=Coya- Machali) Formation by an angular unconformity - Klohn, 1960; Charrier, 1981b) must correlate to still younger units in the Argentine sequence (e.g. the Collon-Cura Formation and equivalents; see Charrier, 1981a). Additionally, the suggested cor- relation of the Abanico (=Coya-Machali) Formation to the Agua de la Piedra Formation in Argentina implies that the unconformity separating it from underlying units (generally the Colimapu Forma- tion, of probable Aptian-Albian age) represents much more "missing" time than previously thought; the length and indeed the existence of this unconformity are matters of current debate (Charrier, 1981a, 1981b; Charrier et al., 1994, in press; Godoy, 1988,1991). Tectonics and Subduction History The central Andes have long been the focus of geophysical and modelling studies of orogenic pro- cesses, owing to the relatively simple geometries of the plate interactions involved. A primary goal of such studies has been to relate the history of sea- floor spreading and subduction of the Nazca - and before it the Farallon - plate with patterns of tec- tonism on the continental margin (Charrier, 1973; Frutos, 1981; Malumian and Ramos, 1984). An im- portant limitation, historically, has been the degree to which the stratigraphic and tectonic history of the Andes can be calibrated temporally. Ironically, the portion of the Andean stratigraphic sequence best controlled chronologically (Anisian- Neocomian, based on marine fossils), is the interval for which the subduction history of the oceanic plate is virtually unknown. Where subduction his- tory is best understood (Cenozoic), age control for the Andean stratigraphic sequence has been nota- bly deficient - particularly prior to the Miocene. New fossil mammal faunas of the Abanico (=Coya- Page 20 Fossil Mammals in Andean Stratigraphy Wyss et al. New Chronologic 40K/ 40Ar Data 10- 20- I 30- 40 — 50- *F. 3,4 * * Farallon Plate Breakup _ Eocene Plate Reorganization Figure 3. Plot of trench normal convergence rates for the Andean margin at 30° S 72° 30' W since anomaly 21 time (from Pilger, 1984). 4BK/40Ai column shows published whole rock dates for the Abanico (=Coya-Machali) Formation from compilation by Drake et al. (1982). These data were also presumably included in Pilger's (1984) compilation (i.e. ibid. fig. 9) which showed a correlation between volcanic activity in the overriding plate with variations in convergence rate. It should be pointed out, however, that the significance of these dates (particularly the younger ones) has been questioned by most central Andean stratigraphers due to the influence of low temperature metamorphism. Thus, until recently it was unclear whether the late Paleogene was actually magmatically quiescent (as held in prevailing tectonic opinion), or whether volcanic rocks of that age exist but simply have not been identified as such due to alteration. New chronologic information (right column) suggests the latter to be the case. This new information consists of single crystal laser fusion '"'Ar/^Ar ages bracketing the Tinguiririca Fauna - identified as F, - (Wyss et al, 1993), and one such analysis (Flynn et al., 1995) associated with the Las Lenas Fauna (F3) indicated by asterisks. Approximate biostratigraphically based age estimates for the Tapado (F2) and the Rio Maipo (F4) faunas are also given. Dozens of late Neogene ages for the Farellones Formation (not shown, see Drake et al., 1982), demonstrate high levels of volcanic activity during this interval of relatively rapid plate convergence. The new chronologic information, however, seems to suggest a more complex relationship between the history of the downgoing slab and continental magmatic activity prior to this time. Machali) Formation help to rectify this deficiency by permitting a more detailed assessment of the correlation between subduction processes and pat- terns of tectonism on the continental margin. The post-Neocomian marine withdrawal from the Lo Prado intra-arc basin in the Coastal and Main Range (see, Charrier, 1981b; Ramos, 1988; Charrier and Munoz, 1994) and back-arc (Mendoza- Neuquen Basin) settings in this Andean region is widely attributed to greatly increased oceanic spreading rates in the South Pacific. This in turn is linked to changes presumed to have occurred at the convergent boundary along western South America, including increased subduction rates and a shift from a "Marianas" to a "Chilean" type subductional regime (see Uliana and Biddle, 1988, and references therein). As a result, the Pacific margin of South America is predicted to have been characterized by active ("Chilean-type") arc magmatism during this interval. Indeed, at the latitude of the new fossil mammal-producing sites in the Chilean Main Range, the voluminous Abanico (=Coya-Machali) Formation has been viewed as representing the product of this volcanic episode (Ramos, 1988; Barrio, 1990—see particu- larly figure 5 therein). The new paleontologic and Wyss et al. Fossil Mammals in Andean Stratigraphy Page 21 radioisotopic data reviewed here suggest, instead, that deposition of most - if not all - of the Abanico (=Coya-Machali) Formation corresponds to a vol- canic interval much younger than the one to which it is usually ascribed. Indeed, this new information points to the need to reassess whether any Late Cretaceous volcanic deposits are preserved in this sector of the Main Range (Charrier et al., in press). The western margin of southern South America was influenced by early Cenozoic plate reorganizations. A probable, ca. 50 Ma old change (between anomaly 22 and 25) in plate motion re- sulted in the establishment of NE-SW directed con- vergence between the Farallon and South Ameri- can plates, a pattern which persisted until about 25 Ma (Pilger, 1984; Pardo-Casas and Molnar, 1987). A reorientation of this convergence vector to an essentially E-W direction occurred near anomaly 7 time (ca. 25 Ma), a change coinciding with fragmen- tation of the Farallon plate (ibid.). Trench-normal convergence rates at the latitude of central Chile (30° S) for the past 50 Ma show a short period of elevated rates near anomaly 18 time (ca. 40 Ma), followed by an interval of low convergence rates between anomalies 13 (36 Ma) and 7 (25 Ma), with an interval of renewed high rates (10-12 cm/yr) characterizing most of the past 25 Ma (ibid.; Figure 3). To highlight the implications of the new geo- chronologic data for the Abanico (=Coya- Machali) Formation, it is instructive to consider this subduction history in the context of various earlier interpretations of the timing of Andean tectonism in central Chile. The first Cenozoic plate motion change (part of global Eocene reorganization) and high rates of trench-normal convergence have been linked with compressional deformation within the Andes ("Compression fini-eocene" of Aubouin et al., 1973; "incaic orogenesis" of Coira et al., 1982), as well as with a period of relative volcanic quiescence (Coira et al., 1982). Deformation and decreasing magmatic activity of the Andean arc have been interpreted as the result of low angle subduction geometry produced by the subduction of increas- ingly younger oceanic crust (Coira et al., 1982; Pilger, 1984; Uliana and Biddle, 1988). Similarly, the 25 Ma modification of conver- gence direction between the Nazca and South American plates to an E-W orientation has been invoked to explain coeval changes in the Andean arc, including an ostensible early Neogene reactiva- tion of the magmatic belt. This "breakout episode," which is sometimes interpreted as marking the ini- tiation of Farellones Formation deposition (Munizaga and Vicente, 1982; Vergara et al., 1988), has been postulated (Ramos, 1988) to result from increased convergence rates along the margin at that time (Pilger, 1984; Pardo-Casas and Molnar, 1987). Aside from the incongruity of invoking high convergence rates to account for early Cenozoic cessation of magmatism on the one hand, and early Neogene "breakout episode" on the other, the new chronologic data reported above indicate that the volcanic activity associated with deposition of the Abanico (=Coya-Machali) Fm. cannot be attributed to Late Cretaceous subduction related magmatism. This unit, which unconformably overlies Mesozoic deposits in this region, corresponds instead to a Late Eocene-Oligocene interval of slow convergence following Eocene plate reorganization. The latter event may thus be linked to deformation (at least tilting), erosion (uplift), and magmatic quiescence, with subsequent magmatic activity associated with extensional basin development (see below) coincid- ing with slow spreading and subduction rates (and possibly higher subduction angle). Fossil mammal faunas and associated ^Ar/^Ar dates from this re- gion demonstrate, therefore, a substantial thickness of volcanic deposits corresponding to a temporal interval previously considered magmatically quies- cent. Adding the caveat necessary for all negative evidence, the lack of evidence of Late Cretaceous volcanism in the Main Range does not necessarily imply that it did not occur, but only that there is little or no positive depositional evidence of volcan- ism during this time. Thus, while numerous inconsistencies and questions remain, the improved age constraints for various volcanic units in this Andean region pro- vided by fossil mammals and single crystal radio- isotopic dating promise to shed light on the rela- tionship between the subduction history of the Farallon and Nazca plates and continental margin tectonics. Even at this early stage, however, it is evident that the relationship between these pro- cesses is more complex than once thought. Subsidence There are presently two opposing views on the timing and connection between Andean tec- tonism and patterns of regional subsidence in the bordering Neuquen region of Argentina. Sedimentologic distinctions between the Neuquen and Malargiie groups of the Neuquen Basin are attributed to changes in style of basin subsidence and provenance direction. Deposition of the younger Malargiie Group is considered re- Page 22 Fossil Mammals in Andean Stratigraphy Wyss et al. lated to the onset of asymmetric subsidence and a western source of siliciclastic sediments. One cur- rent model ascribes these changes to formation of a foreland basin in proximity to active arc volcanoes (Jordan 1981; Barrio 1990). Subsidence itself, ac- cording to this model, is an isostatic response to adjacent thickening of crust due to encroachment of a fold and thrust belt (ibid.). A second model, proposed by Legarreta et al. (1989), suggests that Malargiie Group deposition preceded the onset of orogeny, and represents in- stead the last stage of back-arc accumulation. These authors link latest Cretaceous basin development to changes in subduction regime, the results of which are argued to include activation of the mag- matic belt with concurrent elevation.of the arc re- gion, and subsidence in the back-arc domain caused by magmatic loading. Cited in support of this model are Late Cretaceous-mid Cenozoic eastward migra- tion of arc magmatism (Montecinos, 1979; Parada et al., 1988; Parada, 1990) and a Late Cretaceous-Pale- ocene increase in arc activity previously widely ac- cepted for this region (e.g. Aguirre et al., 1974). Sempere et al. (1994) implicitly accepted a "Legarreta-type" model (i.e. post-Neuquen Group existence of a retro-arc basin), but otherwise made no mention of the controversy. At issue, then, is nothing less than the timing of the onset of orogeny and retro-arc sedimentation in central Andean Ar- gentina. Data provided by the new Abanico (=Coya-Machalf) Formation fossil mammals and single crystal radioisotopic dating provide a ratio- nale for helping to select between these two models and shed light on the chronology of the process. Evidence for a Late Cretaceous-Paleocene pulse of volcanic activity in the central Andes has previously rested almost exclusively on the tremen- dous thickness (circa 3 km) of the primarily volca- nic Abanico (=Coya-Machali) Formation, the most widespread and voluminous unit of the Chilean central Main Range (Figure 2). Significant revision upwards in the age of much (if not all) of this unit indicated by the new paleontological and radioiso- topic data just reported argues definitively against its attribution to a Late Cretaceous pulse in volcanic activity (although the formation is sufficiently com- plex structurally that a Late Cretaceous age for some small portion of it cannot be ruled out). The pres- ence of Eocene and younger fossil assemblages in the Abanico (=Coya-Machali) Formation thus ap- pears discrepant with any model invoking the trig- gering of coeval subsidence and deposition of the Malargiie Group in a retro-arc domain by activity in a volcanic arc. The more Cenozoic fossils continue to appear at varied stratigraphic and geographic locales in the Abanico (=Coya-Machali) Formation, the more the existence of any direct evidence of a Late Creta- ceous pulse of volcanism in the central Andean Main Range (see also above) is called into question, and the more it suggests that if development of the Malargiie Basin is indeed associated with magmatic activity to the west, the volcanic arc was probably located on the eastern side of the present Coast Range or Central Depression where Late Cretaceous volcanic deposits and granitoids are known (Spichiger, 1993). It should be pointed out, how- ever, that evidence of a younger age for the Abanico (=Coya-Machali) Formation and adjustment of a major correlation that once supported a coeval vol- canic loading model, does not in itself lend cre- dence to the alternative hypothesis (which also re- lies on contemporaneous volcanic activity). All that can presently be stated is that the hypothesis invok- ing an incoming fold belt to induce loading to ac- count for deposition of the Malargiie Group is not contradicted by the new geochronologic evidence from the Abanico (=Coya-Machali) Formation. Re- garding deposition of the Abanico (=Coya-Machali) Formation itself, its structural features and the new geochronologic evidence discussed herein point to the existence of a late Paleogene extensional basin (unrelated to those of the Mesozoic) in which this unit was deposited (Charrier et al., 1994). The un- conformable contact between the Abanico (=Coya- Machali) Formation and underlying Mesozoic units in the Tinguiririca valley suggests an erosional epi- sode related to slight uplift and tilting, or mild deformation (Charrier et al., in press). The age of this erosional period is bracketed by the initiation of deposition in the Abanico Basin (at a minimum Tapado Fauna time) and the top of the Malargiie Group. Structural Style Early regional geologic studies of the Chilean Andes (carried out mainly by Europeans, e.g. Briiggen, 1950; Zeil, 1964) did not report nappes, low angle thrust faults, or other evidence of strong compressional tectonic events characteristic of the Alps. Influenced by the current morphostructural features of the range, these authors regarded defor- mation in the Chilean Andes to have been caused mainly by extensional processes. Comprehensive geologic mapping of Chile, commencing in the late 1950's (e.g. Thomas, 1958; Klohn, 1960), revealed the existence of a systemati- cally folded stratigraphic sequence punctuated by Wyss et al. Fossil Mammals in Andean Stratigraphy Page 23 several angular unconformities, pointing to a more complex structural history than posited earlier. Concurrent studies in Argentina (e.g. Groeber, 1951, 1952) identified correlative Mesozoic deposits, all of which were considered to have been deposited within a single basin and to be related by an intri- cate system of faults and folds. More detailed regional studies of the central Chilean Main Range began to be produced in the 1970's (Davidson, 1971; Vicente, 1972; Charrier, 1973, 1981b; Davidson and Vicente, 1973; Thiele, 1980). These studies, focussing on the tectonic, mag- matic and paleogeographic evolution, brought to light systematically developed low and high angle thrusts, and pervasive folding (sometimes tight and oppositely verging) in various units in the region. Still more recently, considerable extension has been documented in the Andean Main Range of the Copiapo region (27° S latitude) for the Cretaceous (Mpodozis and Allmendinger, 1991) and early Cenozoic (see various papers in the 1994 Chilean Geological Congress), as well as for the early Ceno- zoic of the central Chilean Main Range (Charrier et al., 1994). In these areas, movement along origi- nally extensional faults, later became inverted. Pa- leontological evidence reflecting this in the Tinguiririca valley consists of: (1) the occurrence of strata within the Abanico (=Coya-Machali) Forma- tion containing fossil mammals (the Tapado Fauna, locality 2, Figure 2) older than those (the Tinguiririca Fauna, locality 1, Figure 2) contained in the basal levels of the unit (unconformable on the Mesozoic series) near Termas del Flaco; and (2) the stratigraphic superposition of the former strata over the latter (Wyss et al., 1993,1994a; Charrier et al., in press). This, in conjunction with various sedimen- tologic and structural features (see Charrier et al. 1994, in press), indicates that the Abanico (=Coya- Machali) Formation was deposited in an extensional basin formed on a Mesozoic basement, with subse- quent inversion involving at least local layer paral- lel thrusting. CONCLUSIONS There has previously been little reliable radio- isotopic or paleontologic age control for the post- Neocomian, pre-Miocene, portion of the strati- graphic sequence forming the western slope of the central Andes. Such control, however, is funda- mental to interpretations of late Mesozoic through early Cenozoic tectonics of the central Andean Cor- dillera and its bordering basins. A primary aim of our ongoing work in the region is to add to the currently meager geochronological data base for the Abanico (=Coya-Machali) Formation, a major continental volcanic and volcaniclastic unit of the central Chilean Main Range. Toward that end, pa- leontological reconnaissance in the Main Range be- tween 33° 45' and 35° 05' S has thus far yielded four Cenozoic mammal faunas; in view of the previous lack of chronologically useful fossils for this vastly distributed unit and a Late Cretaceous age assumed by several authors, this series of discoveries marks a welcome addition to understanding of the tec- tonic history of the central Andean Main Range. Nowhere in the relatively small portion of the Abanico (=Coya-Machali) Formation thus far sur- veyed has biostratigraphic or single crystal radio- isotopic evidence revealed a Mesozoic age. Two mammal faunas of probable early Eocene, and tran- sitional Eocene-Oligocene age have been recovered from the Tinguiririca valley, the latter having been calibrated radioisotopically. Two additional fau- nas have been recovered from sites in the Abanico (=Coya-Machali) Formation approximately 60 and 110 km north of the Tinguiririca region respectively. The former of these presently has one associated (early Miocene) radioisotopic age determination; in conjunction with preliminary biostratigraphic in- formation, both of these more northern faunas are clearly younger than the youngest fauna from the Tinguiririca region. Assuming that future work confirms similar Paleogene-early Neogene ages for other stratigraphically and geographically disparate por- tions of the Abanico (=Coya-Machali) Formation, the following geologic inferences may be drawn. First, the Abanico (=Coya-Machali) Formation is considerably younger than once assumed and its eastern and western swaths are similar in age; rather than representing the near-arc volcanic equivalent of the Cretaceous Riograndico Supersequence of Argentina, it more likely corre- sponds temporally with the Agua de la Piedra For- mation (and its correlates) of that region. This in- fluences, secondly, decisions concerning the timing and linkage between Andean tectonism and causes of subsidence in adjacent basins, suggesting that deposition of the Malargiie Group (of the bordering Neuquen region in Argentina) cannot be ascribed to "volcanic loading"—at least not that caused by the Abanico (=Coya-Machali) Formation. Third, the presence of thick, Eocene and younger aged levels in the Abanico (=Coya-Machali) Formation clouds the simple relationship between rates of oce- anic crust subduction and patterns of volcanism on the over-riding continental margin once assumed for this segment of the Andes. Late Eocene and Page 24 Fossil Mammals in Andean Stratigraphy Wyss et al. Oligocene volcanism and basin formation in this region corresponds to an interval of slow rates of convergence previously considered associated with a period of magmatic quiescence. If indeed a Late Cretaceous volcanic episode characterized this seg- ment of the Andean Main Range, as has been sup- posed, there is remarkably little direct evidence of it. And finally, the tectonic superposition of older fossiliferous strata over younger in the Tinguiririca valley demonstrates more structural complexity for this region of the Main Range than previously ap- preciated (see also Charrier, 1981b). In addition to representing a major new source of evidence about Cenozoic mammal evolution in South America, the unexpected, abundant, and well-preserved fossils of the Abanico (=Coya-Machali) Formation prom- ise to offer continued significant advances in eluci- dating Andean tectonic history. ACKNOWLEDGEMENTS Don Luis F. Capurro, Director, and Daniel Frassinetti, Chairman of the Department of Geol- ogy, of the Museo Nacional de Historia Natural, Santiago, are thanked for their continued sponsor- ship of our work. We are grateful to Stuart Sumida for inviting our participation in this symposium. Christopher Bell, Judd Case, and Anne Weil pro- vided helpful comments which significantly im- proved the manuscript. We are indebted to Michael Woodburne for a particularly thorough and insight- ful critique. Fieldwork was made possible by sup- port from NSF grants DEB-9020213 and 9318126 to ARW, E-3305-9212 from the Departamento Tecnico de Investigacion, Universidad de Chile to RC, and DEB-9317943 from the NSF to JJF. LITERATURE CITED Aguirre, L. 1960. Geologia de los Andes de Chile Central, Provincia de Aconcagua. Instituto de Investigaciones Geologicas (Chile), Boletin 9. 70 pp. Aguirre, L., R. Charrier, J. Davidson, C. A. Mpodozis, S. Rivano, R. Thiele, E. Tidy, M. Vergara, and J.-C. Vicente. 1974. Andean magmatism: its paleogeographic and struc- tural setting in the Central Part (30°-35°) of the Southern Andes. Pacific Geology 8:1-38. Aubouin, J., A. Borrello, G. Cecioni, R. Charrier, P. Chotin, J. Frutos, R. Thiele, and J. -C. Vicente. 1973. Esquisse paleogeographique et structurale des Andes meridionales. Revue de Geographie Physique et Geologie Dynamique 15(l/2):ll-72. Barrio, C. A. 1990. Late Cretaceous-early Tertiary sedimentation in a semi-arid foreland basin (Neuquen Basin, western Argentina). Sedi- mentary Geology 66:255-275. Biro-Bagoczky, L. 1964. El limite Titoniano- Neocomiano de Lo Valdes. Unpublished The- sis, Departamento de Geologia, Universidad de Chile, Santiago. 259 pp. Biro-Bagoczky, L. 1984. Nuevos antecedentes paleontologicos y estratigraficos de algunos afloramientos Titono-Neocomianos en la Cor- dillera Chilena, entre los 33° 45' y 35° lat. S. International Geological Programme, Project 171; Circum-Pacific Jurassic, Report 2, Special Paper 3. Briiggen, J. 1950. Fundamentos de la Geologia de Chile. Instituto Geografico Militar, Santiago. 374 pp. Charrier, R. 1973. Interruptions of spreading and the compressive tectonic phases of the Meridi- onal Andes. Earth and Planetary Science Let- ters 20:242-249. Charrier, R. 1981a. Mesozoic and Cenozoic stratig- raphy of the Central Argentinian-Chilean Andes (32° - 35° S) and chronology of their tectonic evolution. Zentralblatt fur Geologie und Palaontologie. Teil 11981 (3/4):344-355. Charrier, R. 1981b. Geologie der chilenischen Hauptkordillere zwischen 34° und 34° 30' siidlicher Breite und ihre tektonische, magmatische und palaogeographische Entwicklung. Berliner Geowissenschaftliche Abhandlungen, Berlin, A, (36) 270 pp. Charrier, R. and F. Munizaga. 1979. Edades K-Ar de vulcanitas cenozoicas del sector cordillerano del rio Cachapoal (34° 15' de latitud Sur). Revista Geologica de Chile 7:41- 51. Charrier, R. and N. Munoz. 1994. Jurassic-Creta- ceous paleogeographic evolution of the Chil- ean Andes at 23°-24° S latitude and 34°-35° S latitude: A comparative analysis. Pages 233- 242 in K.-J. Reutter, E. Scheuber, and P. Wigger (eds.). Tectonics of the Southern Central Andes. Springer Verlag, Berlin. Charrier, R., A. R. Wyss, M. A. Norell, J. J. Flynn, M. J. Novacek, M. C. McKenna, C. C. Swisher III, D. Frassinetti, and P. Salinas. 1990. Hallazgo de mamiferos fosiles del Terciario inferior en el sector de Termas del Flaco, Cordillera Prin- cipal, Chile central: implicaciones paleontologicas, estratigraficas y tectonicas. Segundo Simposio Sobre el Terciario de Chile, Actas 1990:73-84. Wyss et al. Fossil Mammals in Andean Stratigraphy Page 25 Charrier, R., A. R. Wyss, J. J. Flynn, C. C. Swisher III, S. Spichiger, and F. Zapatta. 1994. Nuevos antecedentes estratigraficos y estructurales para las formaciones Coya-Machali y Abanico, entre 33° 50' y 35° S, Cordillera Principal, Chilena. VII Congreso Geologico Chileno, Conception 2:1316-1319. Charrier, R., A. R. Wyss, J. J. Flynn, C. C. Swisher III, M. A. Norell, F. Zapatta, M. C. McKenna, and M. J. Novacek. (in press). New evidence for late Mesozoic-early Cenozoic evolution of the Chilean Andes in the upper Tinguiririca Valley (35° S), Central Chile. Journal of South American Earth Sciences. Coira, B. L., J. Davidson, C. Mpodozis, and V. Ramos. 1982. Tectonic and magmatic evolu- tion of the Andes of northern Argentina and Chile. Earth and Planetary Science Letters 18:303-332. Davidson, J. 1971. Tectonica y paleogeografia de la Cordillera Principal en el area de la Nacientes del Teno Curico, Chile. Memoria, Departamento de Geologia, Universidad de Chile, Santiago. 160 pp. Davidson, J. and J. -C. Vicente. 1973. Caracteristicas paleogeograficas y estructurales del area fronteriza de las Nacientes del Teno (Chile) y Santa Elena (Argentina) (Cordillera Principal, 35° a 35°15' de latitud sur). Actas V Congreso Geologico Argentino 5:11-55. Drake, R. E., R. Charrier, R. Thiele, F. Munizaga, H. Padilla, and M. Vergara. 1982. Distribution y Edades K/Ar de Volcanitas post- Neocomianas en la Cordillera Principal entre 32° y 36° L. S: Implicaciones estratigraficas y tectonicas para el Meso-Cenozoico de Chile Central. Actas III Congreso Geologico Chileno 2:D41-D78. Flynn, J. J., M. A. Norell, C. C. Swisher III, and A. R. Wyss. 1991. Pre-Deseadan, Post-Mustersan mammals from Central Chile: an update. Journal of Vertebrate Paleontology 11 (Supple- ment to 3):29A. Flynn, J. J. A. R. Wyss, R. Charrier, and C. C. Swisher. 1995. An Early Miocene anthropoid skull from the Chilean Andes. Nature 373:603- 607. Frutos, J. 1981. Andean tectonics as a consequence of sea-floor spreading. Tectonophysics 72:T21-T32. Godoy, E. 1988. Y Klohn tenia razon: La Formation Colimapu recupera sus miembros basales. Actas V Congreso Geologico Chileno, Santiago 3:H101-H120. Godoy, E. 1991. El corrimiento El Fierro reemplaza a la discordancia intrasenoniana en el Rio Cachapoal, Chile central. Actas VI Congreso Geologico Chileno 515-519. Groeber, P. 1951. La Alta Cordillera entre las lati- tudes 34° y 29° 30'. Museo Argentino de Ciencias Naturales "Bernardino Rivadavia," Revista Geologica 1: 235-351. Groeber, P. 1952. Mesozoico. Geografia de la Republica Argentina, Sociedad Argentina de Estudios Geograficos 2:355-510. Jordan, T. 1981. Thrust loads and foreland basin evolution, Cretaceous, Western United States. American Association of Petroleum Geolo- gists Bulletin 65:2506-2550. Klohn, C. 1960. Geologia de la Cordillera de los Andes de Chile Central; Provincias de Santiago, O'Higgins, Colchagua y Curico. Publicaciones del Instituto de Investigaciones Geologicas, Boletin 8:1-95. Legarreta, L., D. A. Kokogian, and D. A. Boggetti. 1989. Depositional sequences of the Malargiie Group (Upper Cretaceous-lower Tertiary), Neuquen Basin, Argentina. Cretaceous Re- search 10:337-356. Malumian, N. and V. A. Ramos. 1984. Magmatic intervals, transgression-regression cycles and oceanic events in the Cretaceous-Tertiary of southern South America. Earth and Planetary Science Letters 67:228-237. Malumian, N., F. E. Nullo, and V. A. Ramos. 1983. The Cretaceous of Argentina, Chile, Paraguay, and Uruguay. Pages 265-304 in B. M. Moullade and A. E. M. Nairn (eds.). The Phan- erozoic Geology of the World II. The Meso- zoic, B. Elsevier, Amsterdam. Marshall L. G. and T. Sempere. 1993. Evolution of the neotropical Cenozoic land mammal fauna in its geochronologic, stratigraphic and tec- tonic context. Pages 329-392 in P. Goldblatt (ed.). Biological relationships between Africa and South America. Yale University Press. Martinez, R. and R. Osorio. 1963. Consideraciones preliminares sobre la presencia de Carofitas fosiles en la Formation Colimapu. Revista Minerales 82:1-42. Montetinos, P. 1979. Plutonismo durante el Ciclo Tectonico Andino en el Norte de Chile, entre 18° - 29° Lat. Sur. Actas II Congreso Geologico Chileno 2:89-103. Mpodozis, C. and R. Allmendinger. 1991. Extension cretacica a gran escala en el Norte de Chile (Puquios-Sierra de Fraga, 27° S): Significado para la evolution tectonica de los Page 26 Fossil Mammals in Andean Stratigraphy Wyss et al. Andes. Revista Geologica de Chile 19(2):167- 197. Munizaga, F. and J.- C. Vicente. 1982. Acerca de la zonacion plutonica y del volcanismo miocenico en los Andes de Aconcagua (lat. 32- 33° S): Datos radiometricos K-Ar. Revista Geologica de Chile 16:3-21. Parada, M. A. 1990. Granitoid plutonism in central Chile and its geodynamic implications: A re- view. Pages 51-66 in: S. M. Kay and C. W. Rapela (eds.). Plutonism from Antarctica to Alaska. Geological Society of America Special Paper 241. Parada, M. A., S. Rivano, P. Sepulveda, M. Herve, H. Herve, A. Puig, F. Munizaga, M. Brook, and N. Snelling. 1988. Mesozoic and Ceno- zoic plutonic development in the Andes of central Chile (30° 30' - 32° 30' S). Journal of South American Earth Sciences 1:249-260. Pardo-Casas, F., and P. Molnar. 1987. Relative motion of the Naze a (Farallon) and South American Plates since Late Cretaceous time. Tectonics 6:233-248. Pilger, R. H., Jr. 1984. Cenozoic plate kinematics, subduction and magmatism: South American Andes. Journal of the Geological Society of London 141:793-802. Ramos, V. 1988. The tectonics of the Central Andes; 30° to 33° S latitude. Geological Society of America Special Paper 218:31- 54. Sempere, T., L. G. Marshall, S. Rivano, and E. Godoy. 1994. Late Oligocene-Early Miocene compressional tectosedimentary episode and associated land-mammal faunas in the Andes of central Chile and adjacent Argentina (32- 37°S). Tectonophysics 229:251-264. Spichiger, S. 1993. Geologia y estilo estructural de un perfil de los Andes chilenos a 35° de L. S. con entasis en la Cordillera de la Costa, VII Region. Memoria, Departamento de Geologia, Universidad de Chile, Santiago. 112 pp. Thiele, R. 1980. Carta Geologica de Chile, Escala 1:250,000, Hoja de Santiago. Instituto de Investigaciones Geologicas 39:1-21. Thomas, H. 1958. Geologia de la Cordillera de las Costa entre el Valle de La Ligua y la Cuesta Barriga. Instituto de Investigaciones Geologicas (Chile), Boletin 2 Santiago. 86 pp. Uliana, M. A., and K. T. Biddle. 1988. Mesozoic- Cenozoic paleogeographic and geodynamic evolution of southern South America. Revista Brasileira de Geociencias 18:172-190. Vergara, M., and R. Drake. 1978. Edades potasio- argon y su implicancia en la geologia regional de Chile. Revista Comunicaciones 23:1-11. Vergara, M., R. Charrier, F. Munizaga, S. Rivano, P. Sepulveda, R. Thiele, and R. Drake. 1988. Miocene volcanism in the central Chilean Andes. Journal of South American Earth Sci- ences 1:199-209. Vicente, J.-C. 1972. Apercu sur l'organisation et revolution de Andes argentino-chiliennes centrales au parallel de l'Aconcagua. Interna- tional Geological Congress, Proceedings of the 24th Session, Section 3:423-436. Wyss, A. R., M. A. Norell, J. J. Flynn, M. J. Novacek, R. Charrier, M. C. McKenna, C. C. Swisher III, D. Frassinetti, P. Salinas, and J. Meng. 1990. A new early Tertiary mammal fauna from cen- tral Chile: implications for Andean stratigra- phy and tectonics. Journal of Vertebrate Pale- ontology 10(4)518-522. Wyss, A. R., J. J. Flynn, C. C. Swisher III, R. Charrier, and M. A. Norell. 1992a. Fossil mammals from the central Chilean Andes: a new inter- val in the South American land mammal suc- cession, and implications for Eocene-Oli- gocene boundary events and Andean tecton- ics. Fifth North American Paleontological Convention, Knoxville, Abstracts and Pro- gram:318. Wyss, A. R., M. A. Norell, M. J. Novacek, J. J. Flynn. 1992b. New ?early Tertiary mammal locali- ties from the Chilean Andes. Journal of Verte- brate Paleontology 12(Supplement to 3):61A. Wyss, A. R., J. J. Flynn, M. A. Norell, C. C. Swisher III, R. Charrier, M. J. Novacek, and M. C. McKenna. 1993. South America's earliest ro- dent and recognition of a new interval of mammalian evolution. Nature 365:434-437. Wyss, A. R., J. J. Flynn, M. A. Norell, C. C. Swisher III, M. J. Novacek, M. C. McKenna, R. Charrier. 1994a. Paleogene mammals from the Andes of central Chile: a preliminary taxonomic, bio- stratigraphic, and geochronologic assessment. American Museum Novitates 3098:1-31. Wyss, A. R., J. J. Flynn, R. Charrier. 1994b. Yet two additional fossil mammal faunas from the Abanico Formation of the Andean Main Range. PaleoBios 16(Supplement to 1):15. Wyss, A. R., J. J. Flynn, R. Charrier, C. C. Swisher III. 1994c. The Coya-Machali Formation of the central Chilean Andes yields still two more ?Paleogene mammal f aunas. Journal of Verte- brate Paleontology 14(Supplement to 3):54A. Yrigoyen, M.R. 1979. Cordillera Principal. Pages 651-694 in J. C. M. Turner (ed.). Segundo Simposio de Geologia Regional Argentina, Wyssetal. Fossil Mammals in Andean Stratigraphy Page 27 vol. 1 no. 2. Academia Nacional de Ciencias, Cordoba. Yrigoyen, M.R. 1993. Los depositos sinorogenicos Terciarios. Pages 123-148 in V. A. Ramos (ed.). Geologia y Recursos Naturales de Mendoza, XII Congreso Geologico Argentino y II Congreso de Exploracion de Hidrocarburos, Mendoza. Zapatta, F. 1995. Nuevos antecedentes estratigraficos y estructura del area de Termas del Flaco, Valle del Rio Tinguiririca, VI Re- gion, Chile. Unpublished Thesis, Departamento de Geologia, Universidad de Chile. 122 pp. Zeil. W. 1964. Geologie von Chile. Borntrager, Berlin. 233 pp.