Mid–late Holocene cultural and environmental dynamics in Eastern Uruguay

ARTICLE IN PRESS

Quaternary International 132 (2005) 37–45

Mid–late Holocene cultural and environmental dynamics in Eastern Uruguay Roberto Braccoa,b,, Laura del Puertoa,b, Hugo Indaa,b, Carola Castin˜eiraa,b a

14

C, Comisio´n Nacional de Arqueologı´a (MEC), Facultad de Quı´mica (UDELAR), Ca´tedra de Radioquı´mica, Gral Flores 2124, Montevideo, Uruguay b UNCIEP, Facultad de Ciencias, Universidad de la Repu´blica. Igua´ 4225, piso 11, CP 11400, Montevideo, Uruguay

Laboratorio

Abstract This paper reviews the relationship between prehistoric mounds (5000 years BP to 17th century) with main landscape features, particularly wetlands, in the east of Uruguay. Former archeological researchers did recognize this relationship, but it was interpreted by extrapolating current environments to 5000 years of cultural behavior. In an effort to better understand how environmental dynamics were linked to the cultural record, a set of proxy records (diatoms, opal phytoliths, gastropods, shells) were introduced to take into account local environmental evolution data according to global and regional environmental evolution models. The main environmental events during the mid and late Holocene show a high correspondence with mound builders’ spatial arrangements. r 2004 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction The Merin lagoon basin in eastern Uruguay was, from 5000 to 200 BP, the site of one of the largest human landscape modifications in the marginal oceanic areas of the South American plains. Throughout the mid and late Holocene, thousands of mounds were built in well-defined areas, particularly in those that today are swamp or marsh areas. Even mounds that were built at the top of hills are linked to lowlands, because the selected places were located in middle and low plains, or in ‘‘sierras’’ with projections into the plains (Bracco et al., 2000a). 1.1. Eastern Uruguay The Merin lagoon basin is located between 311 and 341S and 521 and 541W, in the eastern part of the South American central plains. The basin area covers 54,000 km2 (24,000 km2 in southernmost Brazil and the Corresponding author. Laboratorio 14C, Comisio´n Nacional de Arqueologı´ a (MEC), Facultad de Quı´ mica (UDELAR), Ca´tedra de Radioquı´ mica, Gral Flores 2124, Montevideo, Uruguay.

remaining 30,000 km2 in Uruguay). The main geographical feature is the Merin lagoon itself, with an average area of 6000 km2. This important fresh water body drains in to Los Patos lagoon, which is connected to the Atlantic Ocean. The southern portion of the basin comprises the east region of Uruguay. This region exhibits different environmental units that are defined according to geomorphological, edaphological and biological criteria (PROBIDES, 1999):










Sierras. This unit comprises the landscapes with altitudes between 200 and 500 m asl that defined the basin limits. The landscapes consist of surficial soils with grasslands and forest developed on the rocky outcrops. Hills (‘‘Colinas y lomadas’’). These represent the transition between the sierras and the higher plains, with gentle undulating landscapes, deep soils and grassland vegetation. Plains. Within this unit, high, middle and low plains are distinguished, according to their relative altitudes and also by the period of the year in which they are covered by river water and stream flow. The main vegetation types are composed of grasslands and palm

1040-6182/$ - see front matter r 2004 Elsevier Ltd and INQUA. All rights reserved. doi:10.1016/j.quaint.2004.07.017

ARTICLE IN PRESS 38

R. Bracco et al. / Quaternary International 132 (2005) 37–45

groves in the middle and high plains, and also by vegetation adapted to permanent or ephemeral aquatic environments, particularly in the low plains affected by river and stream flow. 1.2. Archeological record of mounds The mounds are ground structures of about 30 m in diameter, reaching currently a height up to 7 m, with circular or elliptical shapes. Mounds occur both isolated and in clusters, sometimes with over 50 structures at the same site. In the southern part of the basin (Uruguayan territory), where the total population was estimated near 1500 mounds (Bracco et al., 2000a, b), the major clusters were located in the middle and the low plains linked to the middle tributaries of the Merin lagoon. This spatial arrangement also appears in neighboring areas with similar landscapes. In the southernmost end of the basin (Rocha county) the major clusters are located in the India Muerta and San Miguel swamps or neighboring areas (Fig. 1).

Fig. 1. Southern area of Merin basin. Sampling localities: (1) Saglia, Can˜o 1 and Isla del Tala; (2) Castro; (3) LNB2 core. IM S: India Muerta swamps; SM S: San Miguel swamp; LI S: Los Indios swamp; ST S: Santa Teresa swamp.

Radiocarbon dates indicate that mounds were constructed throughout extensive periods, involving in some cases 3000 years. Brazilian researchers of the 1960s and 1970s inferred that mounds were some kind of ‘‘platform’’ built to inhabit flooded areas by groups specialized in the exploitation of such environments (Naue, 1971, 1973; Schmitz, 1973, 1976; Cope`, 1991, among others). On the other hand, Uruguayan archeologists inferred that they were funerary structures of complex hunter–gatherers, taking into account the recurrence of burials at mounds (Lo´pez and Bracco, 1992, 1994;Cabrera, 2000; Lo´pez, 2001, among others). This explanation of mound functionality is currently under debate (Bracco and Ures, 1999). 1.3. Mounds and environment The coincidence between the spatial arrangement of mounds and lowlands settings distribution (e.g., swamps, marshes) was recognized early in the development of regional archeology. Since the 19th century, mounds were functionally linked to such environments (Arechavaleta, 1892; Ferre´s, 1927; Naue, 1971, 1973; Schmitz, 1976, among others). At the beginning of the academic study of Uruguayan archeology, in the 1980s, the linkage was proposed in adaptive terms, which were in turn the basis to propose a ‘‘mound builders tradition’’ developed from a ‘‘high efficiency strategy in highly productive environments’’ (Lo´pez and Bracco, 1992, 1994). However, this interpretation had two weak points: (1) empirical evidence was not enough to develop a model; and (2) it assumed an actualistic perception of cultural and environmental record (Bracco, 1993). Archeologists were dealing with current environments and cultural records, without taking into account past changes in both categories into an interpretative model. When new data about the chronology of mounds were available, the beginning of such ‘‘constructive behavior’’ was relocated (from 2500 to 5000 BP) (Bracco and Ures, 1998, 1999). Subsequently, the former interpretations were weakened again, and a strong look at cultural dynamics and especially environmental evolution was the only path to follow. This paper presents a synthesis of the available information about regional sea level evolution, and also proxy records of rainfall and humidity variations since the mid-Holocene. Finally, the major effort is placed in a conjugation of environmental periods from new data and the spatial and chronological arrangement of mounds. A harmonious synthesis of environmental and cultural data is the answer that leads to a better interpretation of the relationship between human adaptive effort and environmental changes.

ARTICLE IN PRESS R. Bracco et al. / Quaternary International 132 (2005) 37–45

2. Sea level evolution According to the topography of the basin, sea level and rainfall were the main factors to explain the origin and evolution of lowland environments (Bracco et al., 2000b). Previous researches on Holocene sea levels for the Uruguayan coast (Bracco and Ures, 1998) were able to demonstrate a good correspondence between such works and sea level models for the southern Brazilian coast (Martin and Suguio, 1989; Martı´ n et al., 1996). That research not only confirms the accuracy of the global sea level evolution model (Fig. 2) but also records dramatic regressive events (Martin and Suguio, 1989) occurring ca.r 3800–3600 BP and 2800–2500 BP (Fig. 3). Following the Brazilian model for the Uruguayan coasts, the sea level reached the present level ca. 7000 years BP. After that, the sea level reached +5 m ca. 6000 years BP and then started a progressive fall to the present sea level. During these events, the average sea level was appropriate to generate (in the study area)

39

wide swamp and marsh environments that were and are the main landscape units. The environmental dynamics are not so easy to understand, however, as other factors need to be taken into account to build an explanation. Rainfall plays a decisive role, and was not a static agent during the last 5000 years. With rainfall changes, the extension and quality of lowland environments varied in the basin. A sea level higher than the present one only permitted wetland formation and development below +5 m, when those wetlands were receiving water from the lagoons by tidal action. Wetlands at higher plains (more than 5 m) would need rainfall similar to present times, at least, to become or sustain a wet environment.

3. Climate during the past 5000 years The first available chronologies of mounds in the Merin lagoon basin at the beginning of the 1990s, placed

Fig. 2. Sea level evolution curve during the Holocene, with 14C ages from Uruguayan coasts and curve rectification according to these data (Bracco and Ures, 1998).

Fig. 3. Sea level evolution curve for the last 7000 years, with 14C ages from Castillos lagoon (Martin and Suguio, 1989). Adapted from Bracco and Ures (1998).

ARTICLE IN PRESS R. Bracco et al. / Quaternary International 132 (2005) 37–45

40

the starting point of mound builders tradition close to 2500 BP (Bracco, 1993). At that time, Bombin and Klamt (1976) have indicated (specifically in the basin) the end of a dry or strongly seasonal period, in accordance with macro-regional models that were later proposed (especially Iriondo and Garcia, 1993). This period was followed by the establishment of current conditions, wetter than the former, or at least without a clear rainy season. Chronologies and climate models were responsible, at least in part, for weakness in archeological interpretation. Researchers were seeing an environment and a climate as at present at the beginnings of the cultural record under study. However, the expansion of mound chronologies to 5000 year BP (Bracco and Ures, 1998, 1999) put those interpretations into a paradox: the ‘‘mound builders tradition’’ strongly linked to wet/low lands as we know today, started when such environments were substantially smaller and strongly different from what we currently know (Bracco et al., 2000a, b).




differences were assigned to an increase in continental water after 2000 BP, as evidence of the end of the dry period, followed by the beginning of a wetter climate, as in present times. Martinez (1989), using the composition of shell deposits in the Merin lagoon, was able to demonstrate a transgressive level around 5000 BP. The faunal composition indicates hyaline or mix-hyaline conditions (Table 2 and Fig. 5). Regarding the lagoon dynamics, this could indicate an evaporation volume higher than the influx of continental water. This was responsible for a deficit that implies an oceanic water ingression. According to historical records, at the driest periods of the 20th century, the lagoon response was only a partial and brief ingression of saline water. Data from Martinez (1989) suggests that the composition of shell deposits indicated a rainfall deficit, much larger than historical dry periods that could explain the mid-Holocene record at the lagoon.

3.1. Local evidence During the last four decades, many authors were able to recognize evidence of a dry or highly seasonal period in the area, which began during the mid-Holocene (Table 1). In eastern Uruguay, some proxies record that phenomenon in restricted but well-studied localities (Bracco et al., 2000b) (Fig. 4):







Gonza´lez (1992) recognized the evidence of recent wind erosion in rocks from the Sierra de San Miguel and Cabo Polonio localities, a situation differing from current conditions. Isotopic variations in a single species shell sequence, which records the level fluctuations of the Castillos lagoon from 5500 to 1500 BP, were interpreted by Bracco et al. (2000b) as the consequence of changes between continental and oceanic dynamics. The

Table 1 Beginning and end of mid-Holocene dry period, according to different authors Author

Beginning (years BP)

End (years BP)

Bigarela Vanzolini and Ab’Saber Bombin and Klamt D’anton Markgraf Gonzalez and Se´ller Flegenheimer and Zarate Iriondo and Garcı´ a

3500 3500 3500 3200 3000 5000 45407550 3500

2400 2400 2400 2800 — — — 1000

Average

3700

2200

Extracted from Bracco et al. (2000a) and adapted from Iriondo and Garcia (1993).

4. Understanding local environmental dynamics: proxy record New investigations are under way to obtain paleoclimatic data from lagoon cores in the eastern region (Garcı´ a-Rodrı´ guez et al., 2001). One of these studies is focused on the Negra lagoon, where preliminary results from a core sample provide new data for paleoenvironmental reconstruction in the area. 4.1. Negra lagoon record 4.1.1. Location The Negra lagoon is located in SE Uruguay (331560 –341060 S, 531330 –531420 W), at 8 m asl. The water body is 4 km inland from the Atlantic Ocean, separated by strands of Pleistocene barriers islands, dune fields, and hills. The lagoon area is 142.25 km2, with an average depth of 2.95 m. The basin extension is 720 km2, placed in a depression with granitic and metamorphic intrusions, with peat swamps (about 90 km2) that do not exist in the rest of the region. This lagoon was selected due to its particular features. Negra lagoon is quite different from the other water bodies in the Merin-Los Patos system. The connection with the ocean was lost ca. 120,000 years ago, because of the formation of sandy deposits corresponding to the Barrier System III (Villwock and Tomazelli, 1995). Currently, the Negra lagoon is a fresh water body that drains to the Merin lagoon by Los Indios and San Miguel streams. According to Montan˜a and Bossi (1995), such streams were the only drainage way during late Pleistocene and Holocene times.

ARTICLE IN PRESS R. Bracco et al. / Quaternary International 132 (2005) 37–45

41

Fig. 4. 13C values in Castillos lagoon shells, ordered by their ages. The strong inflection since 2000 BP could be a response to a rainfall increase (Bracco et al., 2000a, b). Table 2 Species composition for shell deposits from paleo shorelines of Merin lagoon (Castro locality) Sample

14

1

4410750

2

5220760

3

4870760

C

Taxa

Environment

Erodona mactroides Mactra Isabelleana Noche subrostrata Olivella puelcha Cytopleura sp. Erodona Mactroides Helobila sp. Mactra sp. Nioche subrostrata Noetia bisricata Plicatula bibbosa Erodona Mactroides Helobila sp. Mactra Isabelleana Noche subrostrata Tagelus plebeius

Fluvial–Marine

Fluvial–Marine

Fluvial–Marine

4.1.2. Methods A 187 cm long and 5 cm internal diameter core was taken in the north coast of the Negra lagoon, at the west end of Santa Teresa swamp (Fig. 1). The core was subdivided according to differences in texture, color, and preservation of biogenic material (Fig. 6). Samples from the subdivisions were processed for grain size analysis, counting and identification of gastropods, opal phytoliths, diatoms, sponge spicules, and chrysophyte cysts (Garcı´ a-Rodrı´ guez et al., 2001). In order to establish the age of the core, four samples were radiocarbon dated (two from the peat layer and two from biogenic carbonates, Littoridina australis) (Table 3). 4.1.3. Results Gastropod (L. australis) and diatom assemblages (mainly Campylodiscus spp., Surirella striatula and

Fig. 5. Species distribution, grouped by environment, for shell assemblages at three mid-Holocene deposits from Merin lagoon coasts, near the mouth of Cebollati river (source Martinez, 1989). Saglia shows an age of 48107140 BP (URU 0006) (Bracco and Ures, 1998).

Tabularia sp.) show surprisingly a brackish water body before 2000 BP, particularly in the samples dated at 3820 +70 and 3820 +120 BP. A high concentration of volcanic ashes at the bottom of the sediment sequence supports the 14C dates, considering the mid-Holocene volcanic ash layer widely recorded in NW Uruguay (Castin˜eira, 1999). Hyaline conditions in the Negra lagoon could be interpreted as salt concentration due to evaporation or brackish water ingression. If salt levels were a consequence of evaporation, then the basin was not receiving a great amount of water from rains. If salt came from the ocean, regarding the topographic position of the water body (8 m asl), the situation could be explained by an indirect connection with the ocean through the Merin lagoon, with the streams running from Merin lagoon southwards to Negra lagoon. That

ARTICLE IN PRESS R. Bracco et al. / Quaternary International 132 (2005) 37–45

42

Chloridoid phytoliths are produced in C4 grasses that dominate in warm, arid, semiarid or highly seasonal regions. A higher concentration of these phytoliths indicates an arid or highly seasonal environment. Pooid phytoliths, produced in C3 grasses, which dominate in high latitudes and elevations, are considered as temperature indicators. A high pooid phytolith concentration suggests lower temperatures (Twiss, 1992). Data from the core are suggesting two different and well-defined moments (Fig. 7):







A first period that ended near 2000 years BP, where the chloridoid percentages indicate arid or highly seasonal conditions. The low concentration of pooid phytoliths, in comparison with the following period, indicates colder temperatures. The second period, after 2000 years BP, was warm and wet, followed by a new dry and cold event. The record is highly consistent with global and regional paleoclimatic models. The warm and wet conditions of the first part fits well with the ‘‘Middle Age Warm Period’’; the following cold and dry conditions could be recording the ‘‘Little Ice Age’’ (Clapperton, 1993; Iriondo and Garcia, 1993).

Fig. 6. Lithological units of core LNB2.

Table 3 Negra lagoon core (LNB2)

14

C chronology

Sample

Depth (cm)

14

Peat Peat Litoridina australis Litoridina australis

49–52/53–58 58–63/64–68 93–108 135–140

1760750 years BP (URU 0224) 1810780 years BP (URU 0225) 3820770 years BP (URU 0222) 38207160 years BP (URU 0226)

C date

dynamic inversion was possible only at the maximum sea level of the Holocene transgression. In any case, both hypotheses need low rainfall conditions in a dry or seasonal regime to fit properly with 2000 years of brackish record. Only ca. 2000 years BP did the Negra lagoon turn to fresh water conditions, and all brackish gastropods and diatoms disappeared. Opal phytolith analysis, particularly the relationship between phytoliths from short cells of grasses, provides an index of humidity and temperature (Twiss, 1992).

Fig. 7. Temperature and humidity indexes from grass short cells opal phytoliths, according to Twiss (1992).

ARTICLE IN PRESS R. Bracco et al. / Quaternary International 132 (2005) 37–45

5. Dealing with landscape: mound builder adaptation and expansion Taking into account the set of 14C dates from mounds in the Merin basin from the top, middle and bottom levels (Schmitz, 1973; Bracco and Ures, 1999), and separating them according to the mounds’ emplacement (Fig. 8), the ‘‘behavior’’ is easily interpretable in terms of adaptation and expansion of mound builder groups. Data from 14C dates suggest that different zones were colonized at different times, and those periods fit harmoniously with the environmental periods inferred in this work. Older dates come from the India Muerta–Paso Barranca zone, shortly after the maximum sea level of the Holocene transgression, with dry or highly seasonal conditions. In the second period, ca. 4000 years BP, that ‘‘constructive behavior’’ seems to be highly intensified in the area with an expansion to the ‘‘sierra’’ zone (Los Ajos and San Miguel). That cultural phase is linked to a regressive event, which was associated at the same time to drier or seasonally dry conditions. If intensification and expansion of mounds was a cultural response to a fluctuating environment, then the mound builders did not know the concept of stress. Far from a negative response, such as wars or loss of group identity, they just were amplifying the cultural response: ‘‘build mounds’’. The period ended ca. 2500 years BP, shortly after the second regressive event, preceding the beginning of current, wetter conditions. Near the end of the period, mound construction in India Muerta and Paso Barranca decreased. 14C dates from the ‘‘Sierra’’ show the same trend. Radiocarbon dates from San Miguel swamp and Los Patos lagoon mounds point to an expansion toward

43

swamp areas near Merin lagoon. Those swamps surely received brackish water from the Merin lagoon since the maximum sea level Holocene transgression. When sea level began to fall, swamps turned progressively to fresh water conditions, particularly after 2000 BP, when rainfall became abundant. This third period (2500 BP to 17th century) shows a new cultural behavior. Mounds became the most frequent burial place throughout the region (Bracco and Ures 1999). By 1000 years BP, rainfall seems to have been more abundant, with warm temperatures, followed by a decrease of both indicators during the Little Ice Age. The first warmer and wetter phase apparently was not significant in terms of spatial response from mound builders. Former contributions (Bracco, 1993) suggest that Little Ice Age conditions could also be another factor to explain the early disappearance of mound builders in historical times, besides European arrival (particularly slavery, genocides and new diseases) (Cabrera, 2000).

6. Discussion The environmental dynamics of the southern region of the Merin basin can now be analyzed as a sequence of different environmental stages, from 5000 BP to the present, indicated by a set of proxy records. That sequence, even considering the local scale of some studies, has a high correspondence with models of regional environmental reconstruction, allowing recreation of at least the main environmental features that prehistoric inhabitants of the area dealt with.

Fig. 8. 14C ages from mounds in the Merin basin, separated according to their geographical location. Sea level information (left) and inferred climate (right) for the period.

ARTICLE IN PRESS 44

R. Bracco et al. / Quaternary International 132 (2005) 37–45

The paleoenvironmental evolution scenario shows that mound builders were established at plains 10 m above current sea level in the southern area of the basin when climate was dry or highly seasonal. This seasonal or dry period doubtlessly affected the extension of swampy areas, with seasonal differences highlighted. That is the significant difference with former interpretations, where environment was considered as a static landscape. In this period, wetlands probably were buffering the water deficit, explaining mound builders settlement preferences. The dry or highly seasonal conditions could be intensified by the regressive events (negative base level) ca. 3800 and 2800 BP. The first regression is coincident with the intensification of mound building at India Muerta–Paso Barranca, suggesting a positive cultural response to negative environmental conditions. The second regression marks the decline of the building behavior for the same area. Perhaps this negative response was linked with different social moments: when mound building started, people had more adaptive ways to solve environmental constraints, but a more developed group (1000 years of social reinforcement) had social requirements and techno-environmental constraints that could place it near the limits of carrying capacity. Shortly after the second regressive event, proxy records point to a rainfall increase. The Negra lagoon and the lower swamps started losing their salt concentrations, probably originated at the maximum sea level of the Holocene transgression and/or by the following rainfall deficit. Since ca. 2000 years BP, environments close to the Merin lagoon became intensely occupied. That final period was characterized by changes in rainfalls and temperature, but mound builders were able to face changes without spatial relocations. If that is correct, the Little Ice Age could be the exception. The development of this cold event is coincident with the mound builders’ disappearance. At this time, the negative environmental situation might amplify the ethnocide resulting from European action.

Acknowledgemnts This research was developed in the framework of the ‘‘Arqueologı´ a de las Tierras Bajas’’ (MEC-CONICYT, No. 5096) project. The authors wish to thank Dr. Felipe Garcı´ a-Rodriguez, who helped them with diatom identification and in particular the limnological point of view, and Prof. Daniel Panario, for his comments and suggestions at every step of this research. References Arechavaleta, J., 1892. Viaje a San Luis. In: Figueira, J.H. (Ed.), El Uruguay en la Exposicio´n Histo´rica Americana de Madrid. Memoria, Montevideo, pp. 65–91.

Bombin, M., Klamt, F., 1976. Evidencias paleoclima´ticas em solos do Rio Grande do Sul. Comunicac- oes do Museo de Ciencias da PUCRGS. Reimpressao dos Anais do XXVIII Congreso Brasileiro de Geologı´ a, vol. 3, 183–193. Bracco, R., 1993. Desarrollo cultural y evolucio´n ambiental en la regio´n este del Uruguay. Ediciones del Quinto Centenario, Universidad de la Repu´blica, Montevideo, pp. 45–69. Bracco, R., Ures, C., 1998. Las variaciones del nivel del mar y el desarrollo de las culturas prehisto´ricas del Uruguay. Revista do Museu de Arqueologia e Etnologia 8, 109–115. Bracco, R., Ures, C., 1999. Ritmos y dina´mica constructiva de las estructuras monticulares. Sector sur de la cuenca de la Laguna Merı´ n, Uruguay. In: Lo´pez, J.M., Sanz, M. (Eds.), Arqueologı´ a y Bioantropologı´ a de las Tierras Bajas. Universidad de la Repu´blica, Facultad de Humanidades y Ciencias, Montevideo, Uruguay, pp. 13–33. Bracco, R., Cabrera, L., Lo´pez, J., 2000a. La prehistoria de las tierras bajas de la cuenca de la Laguna Merı´ n. In: Dura´n, A., Bracco Boksar, R. (Eds.), Arqueologı´ a de las Tierras Bajas. Ministerio de Educacio´n y Cultura, Montevideo, Uruguay, pp. 13–38. Bracco, R., Montan˜a, J.R., Bossi, J., Panarello, H., Ures, C., 2000b. Evolucio´n del humedal y ocupaciones humanas en el sector sur de la cuenca de la Laguna Merı´ n. In: Dura´n, A., Bracco Boksar, R. (Eds.), Arqueologı´ a de las Tierras Bajas. Ministerio de Educacio´n y Cultura, Montevideo, Uruguay, pp. 99–116. Cabrera, L., 2000. Los niveles de desarrollo sociocultural alcanzados por los grupos constructores del este uruguayo. In: Dura´n, A., Bracco Boksar, R. (Eds.), Arqueologı´ a de las Tierras Bajas. Ministerio de Educacio´n y Cultura, Montevideo, Uruguay, pp. 169–183. Castin˜eira, C., 1999. Caracterizacio´n y cronoestratigrafı´ a regional de cenizas volca´nicas. Primeras Jornadas del Cenozoico en Uruguay. INGEPA-UNCIEP, Facultad de Ciencias, Montevideo,, pp. 11–12. Clapperton, Ch., 1993. Quaternary Geology and Geomorphology of South America. Elsevier, London. Cope`, S.M., 1991. A ocupac- a˜o pre´-colonial do sul e sudeste do Rio Grande do Sul. In: Kern, A. (Ed.), Arqueologia Pre´-Histo´rica do Rio Grande do Sul. Mercado Aberto, Porto Alegre, pp. 190–219. Garcı´ a-Rodrı´ guez, F., del Puerto, L., Castin˜eira, C., Inda, H., Bracco, R., Sprechmann, P., Scharf, B., 2001. Preliminary paleolimnological study of Rocha Lagoon, SE Uruguay. Limnologica 31, 221–228. Gonza´lez, M., 1992. Evolucio´n ambiental de la cuenca del arroyo San Miguel. Departamento de Rocha, Republica Orientale del Uruguay, Montevideo, DINATEN, pp. 1–19. Ferre´s, C., 1927. Los terremotos de los Indios. Revista de la Sociedad Amigos de la Arqueologı´ a 1, 139–151. Iriondo, M., Garcia, N., 1993. Climatic variations in the Argentine plain during the last 18000 years. Paleogeography, Paleoclimatology, Paleoecology 101, 209–220. Lo´pez, J.M., 2001. Las estructuras tumulares (cerritos) del litoral atla´ntico uruguayo. Latin American Antiquity 12, 231–255. Lo´pez, J.M., Bracco, R., 1992. Relacio´n Hombre—Medio Ambiente en las Poblaciones Prehisto´ricas de la zona Este del Uruguay. In: Ortiz Troncoso, O.R., Van der Hammen, T. (Eds.), Archaeology and Environment in Latin America. University of Amsterdam, Amsterdam, pp. 259–282. Lo´pez, J.M., Bracco, R., 1994. Cazadores-recolectores de la Cuenca de la Laguna Merı´ n: aproximaciones teo´ricas y modelos arqueolo´gicos. In: Lanata, J., Borrero, L. (Eds.), Arqueologı´ a de CazadoresRecolectores. Lı´ mites, Casos y Aperturas. Arqueologı´ a Contempora´nea, Edicio´n Especial, Buenos Aires, pp. 51–64. Martin, L., Suguio, K., 1989. Excursion Route along the Brazilian coast between Santos (SP) and Campos (RJ) (North of State of Rio de Janeiro). International Symposium on Global Changes in South

ARTICLE IN PRESS R. Bracco et al. / Quaternary International 132 (2005) 37–45 American during the Quaternary. Special Publication, No. 2. ABEQUA-INQUA, Sao Paulo, 136pp. Martı´ n, M., Suguio, K., Flexor, J.M., Dominguez, J., Bittencourt, A.C., 1996. Quaternary seal level history and variation in dynamics along the Central Brazilian coast: consequences on coastal plain construction. Anais da Academia Brasileira de Ciencias 68, 303–354. Martinez, S., 1989. Taphonomy and paleoecology of holocene molluscs from the western margin of the Merin Lagoon (Villa Soriano Fm, Uruguay). Quaternary of South America and Antarctic Peninsula 7, 121–135. Montan˜a, J., Bossi, J., 1995. Geomorfologı´ a de los humedales de la cuenca de la laguna Merı´ n, en el departamento de Rocha. Programa de Conservacio´n de la Biodiversidad y Desarrollo Sustentable (PROBIDES)-Ca´tedra de Geologı´ a, Facultad de Agronomı´ a, Montevideo, Uruguay, 98p. Naue, G., 1971. Novas perspectivas sobre a arqueologia do Rio Grande, R.S. In: O Homem Antigo na Ame´rica. Institution de Pre´Histo´ria da Universidade de Sao Paulo, Sao Paulo, pp. 91–122.

45

Naue, G., 1973. Dados sobre o estudo dos cerritos na a´rea meridional da Lagoa dos Patos, Rio Grande, R.S. Sep. Revista Veritas, vol. 27. Pontifical Catholic University of Rio Grande do Sul, Porto Alegre, pp. 1–73. PROBIDES, 1999. Plan Director Reserva de Biosfera Ban˜ados del Este del Uruguay. Programa de Conservacio´n de la Biodiversidad y Desarrollo Sustentable, Montevideo, 159pp. Schmitz, P., 1973. Cronologı´ a de las culturas do suleste do Rio Grande do Sul—Brasil. Gabinete de Arqueologia. Publicac- o˜es 4, Universidade Federal do Rio Grande do Sul, Porto Alegre. Schmitz, P., 1976. Sitios de pesca lacustre em Rio Grande. Instituto Anchietano de Pesquisas, Sa¨o Leopoldo, Brasil 273pp. Twiss, P.C., 1992. Predicted world distribution of C3 and C4 grass phytoliths. In: Rapp Jr., G., Mulholland, S.C. (Eds.), Phytolith Systematics. Emerging Issues, Advances in Archaeology and Museum Science, vol. 1, pp. 113–128. Villwock, J., Tomazelli, L., 1995. Holocene coastal evolution in Rio Grande do Sul, Brazil. Quaternary of South America and Antarctic Peninsula 2, 283–295.