ARTÍCULOS Y NOTAS

The extinction of some south american camelids: the case of Lama (Vicugna) Gracilis

Jorge Cajal¹, Eduardo P. Tonni², and Viviana Tartarini²

1 Fundación para la Conservación de las Especies y el Medio Ambiente (FUCEMA), Av. de Mayo 1190, 2º piso, 1085 Ciudad Autónoma de Buenos Aires, Argentina [Corresponding author: Jorge Cajal <j_cajal@hotmail.com>].
2 División Paleontología Vertebrados, Museo de La Plata, Paseo del Bosque s/n°, B1900FWA La Plata, Argentina.

ABSTRACT. Lama (Vicugna) gracilis (Gervais and Ameghino, 1881) is a Camelidae that inhabited the low plains of Argentina and Uruguay during the Pleistocene and early Holocene. Considering that Lama gracilis was a vicariant of L. vicugna in the low plains, sympatry cases between Lama guanicoe and Lama vicugna are analyzed to test the hypothesis that considers the competence for the same resource as the cause of extinction of Lama gracilis. It is concluded that the reduced populations of L. gracilis were forced to extinction especially by hunting pressure of paleoindian groups.

RESUMEN. La extinción de algunos camélidos sudamericanos: el caso de Lama (Vicugna) gracilis. Lama (Vicugna) gracilis (Gervais y Ameghino, 1881) es un camélido que habitó las planicies bajas de la Argentina y Uruguay durante el Pleistoceno y el Holoceno temprano. Considerando que Lama gracilis fue un vicariante de L. vicugna en las planicies bajas, se analizan los casos de simpatría entre Lama guanicoe y Lama vigugna para poner a prueba la hipótesis que considera la competencia por el mismo recurso, como la causa de la extinción de Lama gracilis. Se concluye que las poblaciones reducidas de L. gracilis fueron forzadas a la extinción por la presión de caza de grupos paleoindios.

Key words. Camelidae; Extinction; Pleistocene; South America.

Palabras clave. América del Sur; Camelidae; Extinción; Pleistoceno.

Although there is no consensus among authors regarding the taxonomic status of living South American camelids, there is agreement in recognizing two wild species: the guanaco, Lama guanicoe Müller, 1776, and the vicuña Vicugna vicugna (Molina) Miller, 1924 (or Lama vicugna Molina, 1782; for a recent discussion from a molecular and chromosomal point of view, see Marín et al., 2007).
Menegaz et al. (1989) reframed the taxonomy of the genus Lama through a multivariate morphological and morphometric study, including the modern South American Camelidae and those of the Pleistocene of the Patagonian and Pampean regions. These authors consider the guanaco as a single species included in the subgenus Lama: Lama (Lama) guanicoe, and the vicugna as a subgenus of Lama including two species, the living one Lama (Vicugna) vicugna (Molina, 1782) and the extinct Lama (Vicugna) gracilis (Gervais and Ameghino, 1881). They conclude that Lama (V.) gracilis and Lama (V.) vicugna are more closely related to each other than with Lama (Lama) guanicoe.
Recently, Weinstock et al. (2009) sustained that "In light of the combined genetic and morphological arguments discussed above, we argue that the fossil remains assigned to L. gracilis probably belong to V. vicugna…" In any case, it should be noted that molecular analysis, based on few genetic markers (e.g. from mt DNA) not always match fossil records (see Cajal, 2006; Marin et al., 2006).
During the late Pleistocene and early Holocene, Lama gracilis inhabited the Patagonian region and east of the Pampean region. Outside Argentina, there are records in Uruguay for the late Pleistocene before 40 000 radiocarbon years BP (Ubilla, 2004) and for the early Holocene (10 480 to 11 090 radiocarbon years BP; cited as Lama sp. by Ubilla et al., 2007; M. Ubilla personal communication, 2008).
In Argentina, L. gracilis is stratigraphically associated to L. guanicoe at least in four sites locations of the Santa Cruz Province: La María (10 967 ± 55 radiocarbon years BP, Paunero, 2002), Los Toldos (12 600 ± 650 radiocarbon years BP; Cardich et al., 1973; Cardich, 1987), El Ceibo (ca. 11 000 radiocarbon years BP; Cardich, 1987) and Piedra Museo (units 6 to 4, dated between 12 890 ± 90 and 9230 ± 105 radiocarbon years BP; Miotti and Salemme, 2003). In the Chilean sector of the island of Tierra del Fuego, in the site Tres Arroyos 1, one mandibular symphysis referred to Vicugna sp. by Prieto and Canto (1997) probably pertains to L. gracilis; it is stratigraphically associated to L. guanicoe in a level dated 10 630 ± 90 radiocarbon years BP (Massone and Prieto, 2004). This radiocarbon date is a taxondate obtained on bone collagen, and it is similar to other three taxon-dates obtained on L. guanicoe bones from the site Cueva del Medio (10 450 ± 100, 10 710 ± 190, and 10 850 ± 130 radiocarbon years BP; see Nami and Nakamura, 1995). In the great majority of cases, in which not taxon-dates proved the coexistence of both species, the stratigraphical association averaged a short period of time. Consequently, it is considered as coeval in a broad sense. In all these sites, the remains of L. guanicoe are always dominant over those of L. gracilis.
According to the information given by Menegaz et al. (1989) and new findings, at the east of the Pampean region (Buenos Aires province) there is a stratigraphic association between L. guanicoe and L. gracilis in the middle Pleistocene (< 0.78 Ma and > 0.126 Ma) of the Reconquista River. In the late Pleistocene, this association is verified in Paso Otero (Necochea County), in the La Chumbiada and Guerrero members of the Luján Formation. The Guerrero Member was dated between ca. 24 000 and ca. 10 000 radiocarbon years BP (Tonni et al., 2003), whereas the La Chumbiada Member seems to be older than 35 000 radiocarbon years BP (L. Pomi, personal communication 2008), being its temporal extension unknown.
The vicuña has a poor palaeontological record, restricted to the areas of its modern distribution; the oldest records are that of the Atacama dessert, Chile, with a radiocarbon dating of 11 700 radiocarbon years BP (Kuch et al., 2002) and between 12 600 and 10 200 cal. year BP (Grosjean et al., 2005).
In this paper, the cases of sympatry between L. guanicoe and L. vicugna are taken as modern analogues to test the hypothesis which considers the competition for the same resource as the cause for the extinction of L. gracilis.
Today, the vicugna inhabits the high plains or Andean altipampas from northern Peru, Western Bolivia, North-Eastern and Central Chile, and North-Western Argentina down to the north region of San Juan Province (Cajal, 1991). In Argentina, it inhabits the Puna and high-Andean areas contiguous with the Puna. The vicuña has three levels of occupation in the territory with respect to the guanaco: a) altitudinal segregation, where there is no overlap between both species (in Peru, northeast of Bolivia, north of Chile -from 8º to 20º S; b) they share the altitudinal level but with incipient spatial segregation, in Tarija (Bolivia), Sierras de Zenta, Santa Victoria, Chañi and Lipán, in Jujuy, Abra de Acay in Salta and high-Andean areas (excluded de Puna stricto sensu) in Catamarca; c) high superposition degree (sympatry), in the altipampas of a fringe of the Cordillera Frontal between 27º 56' S and 29º 35' S.
Studies performed during four consecutive years (1978-1984) in the Cordillera Frontal, San Juan Province, demonstrated that both the vicuña and the guanaco showed a constant pattern relative to the mean size of their respective family groups, social structure, territorial behaviour (intra and interspecific), and population increase (census number per years). These species also display a high degree of diet overlap, and to some extent of spatial use, regardless of the differences in the pattern of habitat use and preference of sectors according to the social group and time of the year (Cajal, 1989, Cajal and Bonaventura, 1998).
In the sympatry area, both guanacos and vicuñas are clearly grazers (although they also browse on bushes and cactaceae). According to Cajal (1989), in the case of the vicugna the grass represents 88% and shrub / cacti 12% of diet, while in guanaco the grass represents 87% and shrub / cacti 13% of diet. Coexisting wild camelids respond in the same way when facing the available food at a given place and time (e.g. Stipa grass). Certainly, the genus Stipa is the main component of the diet of both species. A test of multiple comparisons among plant species ordered according to the amount of their mean frequency in the diet showed that Stipa is in both camelid species, the most important group, clearly distinguished from other plants (Cajal, 1989). This larger presence of Stipa in the respective diets is related to supply. Within perennial plants, this genus is outstanding by its larger percentile presence (> 40 %) when analyzing the plant cover of the high plain (altipampas) or steppes of the region.
The physiography where the vicuñas and guanacos are sympatric is like a high plains archipelago originated by ravines, creeks, rivers, and mountains. This configuration defines limited areas or sectors. A comparative analysis of the use of these sectors by both species (test of equality of means with heterogeneous variances using the Games and Howell method, MCHETV, see Sokal and Rohlf, 1981), revealed that for both, the guanaco and the vicuña, some sectors were significantly different to others, while others did not show such differences. Considering both species as a block, some sectors were used in a significantly different way, while in others, differences were not observed. According to Q values of this method, the vicuña turned out more sensitive than the guanaco with respect to the different sectors (Cajal, 1989). It should be noted that two species by definition have always differences, regardless how close related they are.
The body mass of L. gracilis estimated with regression equations for artiodactyls (Scott, 1990), based on the transverse width of the head of the humerus and maximum width of the distal epiphysis, is between 50.36 and 64.32 kg. Hence, its mass is similar to that of the vicuña (45-55 kg) and quite different from that of the guanaco (80-120 kg).
Both L. gracilis and the vicuña show dental specializations for abrasive grasses (i.e., Stipa). Functionally, the dental specialization of incisors in both species (see Menegaz et al., 1989) tends to optimize the intake of abrasive grasses (pastures) with high silica content. In this regard, a positive correlation has been mentioned between silica and palatability (Adler et al., 2004).
L. gracilis inhabited, during the late Pleistocene and early Holocene, the low plains from the Pampean region to Patagonia. Considering that the oldest records belong probably to the middle Pleistocene of the Pampean region, Menegaz et al (1989:170-171) postulated a southward retraction of L. gracilis during the late Pleistocene. There are no evidences to confirm this hypothesis, since in Patagonia there is no paleontological record of the late Pleistocene prior to ca. 12 000 years BP. L. gracilis seems to have followed the expansion of the Pampean fauna toward the end of the Pleistocene -around 12 000 years BP, as verified with several species present in both regions during that period (Borrero, 1999, 2005; Tonni and Carlini, 2008).
Menegaz et al (1989) pointed out that postglacial climatic changes affected the distribution of different groups of Grammineae and influenced deeply in the retraction and extinction of L. gracilis. In this sense, they suggest that the present distribution of L. vicugna is related to that of microthermal grammineae groups, especially to those best adapted to arid environments, rather than to those specialized to life in mountains. On the contrary, the guanaco is a generalist herbivore, adapted to a wide range of vegetation and habitats, which places it, according to some authors, out of the competence with other herbivores of this size (Raedeke and Simonetti, 1988).
According to Menegaz et al. (1989) the extinction of L. gracilis was caused by an unfavourable competence with the guanaco in the exploitation on similar niches (Menegaz et al., 1989:171), being the competence a causal source of extinction. However, using a modern analogue, in sympatry, both guanacos and vicugnas feed on the same food resources. If resources are scarce, competition may be the outcome of shared limiting resources. It does not matter who ate the food if it is gone. The shortage can affect the individuals of both species as a whole. There is no evidence for competition at that time and place in the Cordillera Frontal. However, it does not indicate that competition between individuals of the same species and between those of both species occurs when food resources are scarce. Neither does it mean that the competition caused by food scarcity necessarily implies the displacement (or extinction) of one species by the other; nor that it is a factor involved in future evolutionary changes.
Other causes that led to extinction must be considered, beyond the interspecific competence. One of them is the hunting burden exerted by paleoindians, a hypothesis taken to an extreme by Martin (1984) in his model of blitzkrieg, as the cause of massive extinction of the end of the Pleistocene in North America, and by extension, in South America.
The zooarchaeological records demonstrate the effects of hunting pressure on camelids, both in guanaco and L. gracilis. However, the relative frequency of the remains indicates a bias towards the first species with respect to the second. This may result from the reduced populations of L. gracilis by the end of the Pleistocene. In this regard, Cione et al. (2009) considered that the temperature and humidity increase of the interglacials, during the Quaternary, caused a decrease of the biomass of the mammals adapted to open areas. These authors indicate that the studies based on geochemical and dust proxies in Greenland, Antarctica, and South America glaciers attest that the present interglacial period is not substantially different from the preceding ones. The arrival of humans in South America was the sole new biological or geological event that occurred in the present interglacial period. In consequence, the entrance of men and their pressure on a numerically reduced population would have been the factor that led to local extinction (extirpation) or total (global) extinction. The different extinction of related species (e.g., between Cervidae and Camelidae, cited by Cione et al., 2009), may correspond to subtle differences in ecological plasticity.
If L. gracilis -unlike the guanaco-, had environmental sensitivity and strict trophic and habitat requirements, similar to those of the vicuña in modern high plains, the climatic change in low plains (Tonni et al., 1999) could have reduced their populations up to levels in which hunting pressure of paleoindian groups led them to extinction.
Alternatively, if it is proved that L. gracilis and L. vicugna (or Vicugna vicugna) are the same species (see Weinstock et al., 2009), the local extinction in the low areas and their retraction to regions of the high plains or Andean altipampas, can be attributed also to the anthropic action that not only includes the hunt burden but also other factors (i.e., environmental change and etological changes caused by the human presence).

ACKNOWLEDGEMENTS

The authors are grateful to Agencia Nacional de Promoción Científica y Tecnológica, Comisión de Investigaciones Científicas de la provincia de Buenos Aires, and Universidad Nacional de La Plata for financial support.

LITERATURE CITED

1. ADLER PB, DG MILCHUNAS, WK LAUENROTH, OE SALA, and IC BURKE. 2004. Functional traits of graminoids in semi-arid steppes: a test of grazing histories. Journal of Applied Ecology 41:653-663.         [ Links ]

2. BORRERO LA. 1999. The faunas of the Pleistocene/Holocene boundary in the Seno de la Ultima Esperanza, Chile. Pp. 59-62, in: Zooarchaeology of the Pleistocene/Holocene Boundary (JC Driver, ed.). BAR Internacional Series 800.         [ Links ]

3. BORRERO LA. 2005. Taphonomy of late Pleistocene faunas at Tierra del Fuego-Patagonia. Journal of South America Earth Sciences 20:115-120.         [ Links ]

4. CARDICH A. 1987. Arqueología de Los Toldos y El Ceibo (Provincia de Santa Cruz, Argentina). Investigaciones paleoindias al sur de la línea ecuatorial. Estudios Atacameños 8:98-117.         [ Links ]

5. CARDICH A, L CARDICH, and A HAJDUK. 1973. Secuencia arqueológica y cronológica radiocarbónica de la cueva 3 de Los Toldos (Santa Cruz, Argentina). Relaciones Sociedad Argentina de Antropología 7:85-123.         [ Links ]

6. CAJAL JL. 1989. Uso de hábitat por vicuñas y guanacos en la Reserva San Guillermo, Argentina. Vida Silvestre Neotropical 2:21-31.         [ Links ]

7. CAJAL JL. 1991. An integrated approach toward the management of wild camelids in Argentina. Pp. 305-321, in: Latin American Mammalogy: History Biodiversity and Conservation (M Mares and D Schimidly, eds). University of Oklahoma Press.         [ Links ]

8. CAJAL JL. 2006. Revisión de Libros: Investigación, Conservación y Manejo de Vicuñas. B Vila (ed). 2006. Proyecto MACS-Argentina-INCO-Unión Europea. Mastozoología Neotropical 13:283-287.         [ Links ]

9. CAJAL JL and S BONAVENTURA. 1998. Densidad poblacional y dinámica de los grupos familiares de guanacos y vicuñas en la Reserva de Biosfera de San Guillermo. Pp. 161-167, in: Bases para la conservación y manejo de la puna y cordillera frontal de Argentina: El rol de las reservas de biosfera (JL Cajal, JJ García Fernández, and R Tecchi, eds.). UNESCO (Uruguay) - FUCEMA, Montevideo.         [ Links ]

10. CIONE AL, EP TONNI, and LH SOIBELZON. 2009. Did humans cause the late Pleistocene-early Holocene mammalian extinctions in South America in a context of shrinking open areas? Pp. 125-144, in: American Megafaunal extinctions at the end of the Pleistocene (G Haynes, ed.) . Springer Publ.         [ Links ]

11. GROSJEAN M, L NUÑEZ, and I CARTAJENA. 2005. Palaeoindian occupation of the Atacama Desert, northern Chile. Journal of Quaternary Science 20:643-653.         [ Links ]

12. KUCH M, N ROHLAND, JL BETANCOURT, C LATORRE, S STEPPAN, and HN POINAR. 2002 Molecular analysis of an 11,700-year-old rodent midden from the Atacama Desert, Chile. Molecular Ecology 11:913-924.         [ Links ]

13. MARÍN JC, AE SPOTORNO, and JC WHEELER. 2006. Sistemática molecular y filogeografía de camélidos sudamericanos: Implicancias para su conservación y manejo. Pp. 85-100, in: Investigación, Conservación y Manejo de Vicuñas (B Vilá, ed.). Proyecto MACSArgentina-INCO-Unión Europea, Buenos Aires.         [ Links ]

14. MARÍN JC, C ZAPATA, BA GONZÁLEZ, C BONACIC, JC WHEELER, C CASEY, M BRUFORD, RE PALMA, E POLIN, MA ALLIENDE, and AE SPOTORNO. 2007. Sistemática, taxonomía y domesticación de alpacas y llamas: nueva evidencia cromosómica y molecular. Revista Chilena de Historia Natural 80:121-140.         [ Links ]

15. MARTIN P. 1984. Prehistoric overkill: the global model. Pp 354-403, in: Quaternary Extinctions: A Prehistoric Revolution (PS Martin and RG Klein, eds). University of Arizona Press.         [ Links ]

16. MASSONE M and A PRIETO. 2004. Evaluación de la modalidad cultural Fell 1 en Magallanes. Chungara, vol. especial, pp. 303-315, Arica, Chile.         [ Links ]

17. MENEGAZ AN, FJ GOIN, and E ORTIZ JAUREGUIZAR. 1989. Análisis morfológico y morfométrico multivariado de los representantes fósiles y vivientes del género Lama (Artiodactyla, Camelidae). Sus implicancias sistemáticas, biogeográficas, ecológicas y biocronológicas. Ameghiniana 26:153-172.         [ Links ]

18. MIOTTI LL and MC SALEMME. 2003. When Patagonia was colonized: people mobility at high latitudes during Pleistocene/Holocene transition. Quaternary International 109-110:95-111.         [ Links ]

19. MIOTTI L, M VÁZQUEZ, and D HERMO. 1999. Piedra Museo: Un yamnagoo pleistocénico de los colonizadores de la Meseta de Santa Cruz: El estudio de la arqueofauna. Actas de las III Jornadas de Arqueología de la Patagonia, pp. 113-136. Bariloche.         [ Links ]

20. NAMI HG. and T NAKAMURA. 1995. Cronología radiocarbónica con AMS sobre muestras de hueso procedentes del sitio Cueva del Medio (Ultima Esperanza, Chile). Anales del Instituto de la Patagonia 23:125-134.         [ Links ]

21. PAUNERO RS. 2002. The presence of a Pleistocenic colonizing culture in La Maria archaeological locality: Casa del Minero 1. Pp. 127-132, in: Ancient Evidences for Paleo South Americans: From Where The South Winds Blow (L Miotti, M Salemme, and N Flegenheimer, ed.). Texas A&M University Press.         [ Links ]

22. PRIETO A and J CANTO. 1997. Presencia de un lamoide atípico en la cueva Lago Sofía 4 (Última Esperanza) y Tres Arroyos (Tierra del Fuego) Región de Magallanes, Chile. Anales del Instituto de la Patagonia (Serie Ciencias Humanas) 25:147-150.         [ Links ]

23. RAEDEKE KJ and JA SIMONETTI. 1988. Food habits of Lama guanicoe in the Atacama desert of northern Chile. Journal of Mammalogy 69:198-201.         [ Links ]

24. SCOTT KM. 1990. Postcraneal dimensions of ungulates as predictors of body mass. Pp. 301-336, in: Body Size in Mammalian Paleobiology: Estimation and Biological Implications (J Damuth and BJ MacFadden, eds.). Cambridge University Press.         [ Links ]

25. SOKAL R. and F ROHLF 1981. Biometry: The principles and practice of statistics in biology research. Second edition, 859 pp.W. H. Freeman & Co.         [ Links ]

26. TONNI EP, AL CIONE, and AJ FIGINI. 1999. Predominance of arid climates indicated by mammals in the pampas of Argentina during the late Pleistocene and Holocene. Palaeogeography, Palaeoclimatology, Palaeoecology 147:257-281.         [ Links ]

27. TONNI EP, R HUARTE, JE CARBONARI, and A FIGINI. 2003. New radiocarbon chronology for the Guerrero Member of the Luján Formation (Buenos Aires, Argentina): palaeoclimatic significance. Quaternary International 109-110:45-48.         [ Links ]

28. TONNI EP and AA CARLINI. 2008. Neogene vertebrates from Patagonia (Argentina): their relationship with the most significant climatic changes. Pp. 269-284, in: The Late Cenozoic of Patagonia and Tierra del Fuego (J Rabassa, ed.). Development in Quaternary Science 11        [ Links ]

29. UBILLA M. 2004. Mammalian biostratigraphy of Pleistocene fluvial deposits in northern Uruguay, South America. Proceedings of the Geologists' Association 115:347-357.         [ Links ]

30. UBILLA M, D PEREA, A RINDERKNECHT, and A. CORONA. 2007. Mamíferos del cuaternario de Uruguay. Bioestratigrafía y paleoambientes. Workshop "Quaternario RS: integrando conocimentos", pp. 45-47, Canoas, Rio Grande do Sul, Brasil.         [ Links ]

31. WEINSTOCK J, B SHAPIRO, A PRIETO, JC MARÍN, BA GONZÁLEZ, PGM THOMAS, and E WILLERSLEV.2009. The Late Pleistocene distribution of vicuñas (Vicugna vicugna) and the "extinction" of the gracile llama ("Lama gracilis"): New molecular data. Quaternary Science Reviews 28:1369-1373.         [ Links ]

Recibido 7 noviembre 2008.
Aceptado 27 junio 2009.

Editor asociado: L Borrero