Marta Lizarralde, Julio Escobar and Guillermo Deferrari

Marta Susana Lizarralde. Ph.D., University of La Plata (UNLP), Argentina. Researcher, Consejo de Investigaciones Científicas y Técnicas (CONICET), Argentina. Director, Ecogenetics Laboratory, Centro Austral de Investigaciones Científicas (CADIC), Argentina. Address: CADIC-CONICET, CC 92, 9410 Ushuaia, Tierra del Fuego, Argentina. e-mail: mlizarralde@cpsnet.com.ar

Julio Martín Escobar. Technician, Ushuaia High School, Argentina. Research Assistant, Ecogenetics Laboratory, CADIC, Argentina.

Guillermo Deferrari. M.Sc., UNLP, Argentina. Research Assistant, Ecogenetics Laboratory, CADIC, Argentina.

Los castores (Castor canadensis) fueron introducidos en la Isla Grande de Tierra del Fuego, Argentina, en 1946. La ausencia de predadores y competidores naturales y la abundancia de alimento y sitios de refugio favorecieron su rápida expansión y crecimiento poblacional. Este artículo muestra el estado poblacional y las modificaciones al paisaje producidas por los castores en estos ecosistemas australes. Actualmente los castores son encontrados en todos los ríos de las áreas andinas y extra-andina y en casi la totalidad de hábitats acuáticos de la Isla Grande como también en otras islas chilenas del Archipiélago de Tierra del Fuego (70000 km²). Áreas con orden de gradiente bajo en cursos de agua pequeños son ocupados más densamente que aquellos de valles en pendientes. Las densidades son similares a las del Hemisferio Norte. Los hábitats más colonizados mostraron una abundancia de 0,7 colonias activas por km². Cuatro clases (A, B, C y D) para usar en manejo y planificación de recursos naturales, fueron determinadas sobre la frecuencia de los sitios de colonia y el patrón de ocupación del castor en las cuencas. Las altas densidades en las clases C y D (4,7 y 5,8 sitios de colonia por km de cuenca) indican que, potencialmente, ambas áreas tuvieron la mayor productividad. Los sitios alterados por castor tuvieron niveles de N orgánico e inorgánico altos sugiriendo que los cambios hidrológicos estacionales podrían afectar la nitrificación y desnitrificación, permitiendo también la acumulación de C y P en el curso de agua. Los diques de castor pueden ser considerados fuentes de nutrientes esenciales (N y P) y C. Los análisis cromosómicos no mostraron diferencias con el cariotipo de Norteamérica. En la actualidad, se está analizando la estructura y variabilidad genética de la población introducida.

Beavers (Castor canadensis) were introduced at Isla Grande, Tierra del Fuego, Argentina, in 1946. Suitable feeding and lodging sites, coupled with lack of natural predators or competitors favored rapid population growth and range expansion. This paper shows the current population status and landscape modifications induced by beavers in these southern ecosystems. Beavers are now found in all streams in the Andean and extra-Andean areas, and in nearly all aquatic habitats on Isla Grande as well as other Chilean islands of the Tierra del Fuego archipelago (70000km²). Areas with low gradient in small streams are more densely occupied than those in slope valleys. Densities are similar to those in the Northern Hemisphere. Extensively colonized habitats showed 0.7 active beaver colonies per km². Based on their different beaver occupancy patterns and frequency of colony sites, four land capability classes (A, B, C and D) were established to use in planning and resource management. The highest densities were found in classes C and D (4.7 and 5.6 colony sites per km²) indicating that both these areas had the greatest potential for beaver production. Beaver-altered sites had higher levels of organic and inorganic N, suggesting that seasonal hydrological changes could be affecting nitrification and denitrification, also resulting in accumulated organic C and P in the stream channel. Beaver ponds may be considered sources of essential nutrients (P and N) and C. Chromosome analysis showed no differences with the North American karyotype. Genetic structure and variability of the beaver population are analyzed.

Os castores (Castor canadensis) foram introduzidos na Ilha Grande da "Tierra del Fuego", Argentina, em 1946. A ausência de predadores e competidores naturais e a abundância de alimento e lugares de refugio favoreceram sua rápida expansão e crescimento populacional. Este artigo mostra o estado populacional e as modificações na paisagem produzidas pelos castores nestes ecossistemas austrais. Atualmente os castores são encontrados em todos os rios das áreas andinas e extra-andina e em quase a totalidade de habitats aquáticos da Ilha Grande como também em outras ilhas chilenas do Arquipélago da "Tierra del Fuego" (70000km²). Áreas com ordem de gradiente baixo em cursos de água pequenos são ocupados mais densamente que aqueles de vales em pendentes. As densidades são similares às do Hemisfério Norte. Os habitats mais colonizados mostraram uma abundância de 0,7 colônias ativas por km². Quatro classes (A, B, C e D) para usar em direção e planificação de recursos naturais, foram determinadas sobre a freqüência das zonas de colônia e o padrão de ocupação do castor nas bacias. As altas densidades nas classes C e D (4,7 e 5,8 zonas de colônia por km de bacia) indicam que, potencialmente, ambas áreas tiveram a maior produtividade. As zonas alteradas por castor tiveram níveis de N orgânico e inorgânico altos sugerindo que as mudanças hidrológicas estacionais poderiam afetar a nitrificação e desnitrificação, permitindo também a acumulação de C e P no curso de água. Os diques de castor podem ser considerados fontes de nutrientes essenciais (N e P) e C. As análises cromossômicas não mostraram diferenças com o cariótipo de Norte América. Na atualidade, se está analisando a estrutura e variabilidade genética da população introduzida.

KEYWORDS / Beaver / Castor canadensis / Colonization / Ecosystem / Invader / Tierra del Fuego /

Received: 02/19/2004. Accepted: 05/25/2004.

Wild furbearers have aesthetic, economical, and cultural value. On the southernmost tip of Patagonia, native furbearing species are not abundant (Lizarralde and Escobar, 2000; Lizarralde and Elisetch, 2001). For that reason, in 1946, 25 pairs of beaver (Castor canadensis) were introduced on the Río Claro of Isla Grande (Tierra del Fuego Archipelago, Argentina) by the federal government (Figure 1). The absence of most natural predators and human hunting, and the availability of suitable feeding and lodging conditions favored rapid population growth and range expansion (Lizarralde, 1993).

Beavers are considered as one of the most important natural agents of landscape alteration as well as a keystone species (Paine, 1966). Beavers modify stream morphology and hydrology by felling trees and building dams. These activities retain sediments and organic matter in the channel, create and maintain wetlands, modify nutrient cycling and decomposition dynamics, modify the structure and dynamics of the riparian zone, influence the character of water and materials transported downstream, and ultimately influence biotic community composition and diversity (Naiman et al., 1988). Areas abandoned by beavers may produce fertile grasslands. However, the influence of beavers on undisturbed areas such as the southern ecosystems had not been intensively studied.

Tierra del Fuego ecosystems are pristine, fragile, and located in an extreme climatic zone that provides only marginal possibilities for the development of many species. Therefore, introduction of new fauna, particularly beavers, may result in changes in the regional ecology, causing effects on the biotic composition and the physical structure of forests and bogs (Lizarralde and Venegas, 2002).

The beaver introduced in Tierra del Fuego is considered as an invader because it modifies the local ecological interactions. Therefore, intensive studies have been carried out to understand the role of beavers on southern ecosystems and to compare this with boreal ecosystems (Lizarralde, 1993; Lizarralde and Escobar, 1999; Lizarralde et al., 1989, 1996a, b, 2002; Coronato et al., 2003).

Our research group has studied during 15 years the problems induced by beavers in Tierra del Fuego. This paper is a review of the main data produced about 1) beaver abundance, distribution, and spatial occupancy patterns, 2) habitat preference, 3) landscape alteration effects, 4) management and control, and 5) genetic structure and variability to detect possible genetic bottlenecks and founder effects in this invader species.

The Tierra del Fuego Archipelago consists of a large number of islands located between the Pacific and Atlantic Oceans, the largest being Isla Grande (Figure 1). Insular conditions and the Antarctica influence have resulted in a humid climate that is classified as temperate-cold (Pisano, 1981). Isla Grande has three ecological areas with distinctive geographical relief (Lizarralde, 1993): 1) The Andean region is a mountainous area with vast basins and broad stream courses of glacial origin. 2) The ecotone or transitional area, with a flat aspect and gradual undulations influenced by glaciation. 3) The steppe or extra-Andean area.

In the Andean area, the vegetation cover is dominated by an evergreen forest of Nothofagus betuloides alternating with Sphagnum bogs along the Beagle Channel coasts, on the south side of Isla Grande. The area surrounding Lake Fagnano is an ecotone with a vegetation cover of deciduous forest (Nothofagus pumilio, N. antarctica) that alternates with Sphagnum bogs and flat lowlands containing Carex. The extra-Andean area has a vegetation pattern typical of the Patagonia steppe.

Colonization patterns were determined by analyzing aerial photographs, aerial and field surveys, and beaver census carried out on the Isla Grande according to published procedures (Lizarralde, 1993).

All watersheds of Isla Grande (n=53) were systematically flown over at altitudes of 250-900m to determine beaver abundance and spatial occupancy patterns. Surveys were performed between fall and freeze-up, when food caches were most visible. Census were obtained during seasonal periods from 1988 to 1992, 1994 to 1998 and 2000 to 2002. The population size was estimated from counts of beaver colony sites on stream sections. A colony site was defined as a pond or series of ponds used by a colony of beavers throughout the year or years.

Field surveys were conducted to analyze the areas where beaver colonization, as detected from aerial survey, was high. All indirect evidence of habitat utilization was assessed, including tracks, food caches, scent mounds, wood debris and den banks. The number of active colonies was determined at study sites of Tierra del Fuego National Park to estimate beaver abundance and to compare these data with those from boreal forest areas. The average number of animals per colony was 5 and was determined from the data provided by hunters who trapped all the animals in each colony (Lizarralde, 1993). Height, width and length of beaver dams were measured to estimate pond surfaces and potential volume of sediment accumulation. By mapping pond locations in the streams, it was possible to determine spatial occupancy patterns. Data on foraging and diameters of trees felled by the beavers were recorded from linear transects from the pond to the last felled tree.

The different habitats were classified according to vegetation coverage and geomorphology. The physical characteristics of the streams colonized by beavers were also determined through geomorphological analysis and mapping techniques (Coronato et al., 2003).

Animals (n=50) were caught with live (Tomahawks™) and kill (Conibear™) traps. Blood, reproductive tracts, skulls and fresh tissue samples were obtained for cytogenetic, biological and molecular studies. Additional skulls (n=240) and tissue samples (n=20) were collected in the field from animals that died naturally or were killed by commercial hunters. Reproductive performance, age, and sex were determined for each individual. Skull voucher specimens of all of the animals studied were deposited in the Collection at the Ecogenetics Laboratory (CADIC-CONICET, Tierra del Fuego, Argentina).

The contents of the principal organic elements and inorganic compounds of sediments (n=50), and pond-riffle water samples from beaver altered (n=39) and unaltered control (n=21) sites were examined over a three-year period, from 1991 to 1994 (Lizarralde et al., 1996a). Altered sites were chosen according to habitat type, stored organic material, seasonal hydrologic regime, dominant vegetation and lack of evidence of human disturbance. Unaltered control sites were riffles and watercourses without a history of beaver presence. Sediment and pond water were collected from each of the sites during the ice free period (October to May). Sediment cores of 1000-2000g were taken in the ponds behind dams or, in unaltered sites, riparian zones 10 to 20cm in depth. Samples were stored in plastic bags at -20ºC. Concentration of total organic C, organic N, organic P, nitrite (NO₂^-), nitrate (NO₃^-), ammonium, and phosphate were determined by standard methods (Lizarralde et al., 1996a).

In water samples, concentrations of carbonate and bicarbonate, Cl, sulfate, Ca, Mg, Na, and K were also measured. Sediment and water pH were also determined (Lizzarralde et al., 1996a).

Beavers were found in all streams from the North of Isla Grande to the Beagle Channel coast, covering about 40000km². Beavers also occupy Isla Navarino, Isla Hoste, and Península Dumas, with a geographical range of about 70000km² (Lizarralde, 1993; Figure 1).

Colony-site densities were estimated to be between 0.2 and 5.8 per km of stream. Extensively colonized habitats (e.g. National Park) showed 0.7 active beaver colonies per km². Populations have been increasing at a rate of 20-23% per year (K= 0.20-0.23; Lizarralde et al., 1996b). The reproductive season was determined to be from June to November, with an average of four kits per litter and a sex ratio of 1:1 (Lizarralde and Escobar, 1999; Lizarralde, unpublished data).

Chromosome analysis of Giemsa-stained preparations showed a karyotype 2n=40 (n=10 individuals) with 16 pairs of metacentric biarmed chromosomes and four pairs of submetacentrics. The X sexual chromosome is a large metacentric one, and Y is one of the smallest from the complement (Figure 2).

Beavers primarily altered uplands stream valleys in mountainous areas and wetlands. The formation of ponds in streams with flat valley floor increases the area of wetland as wide, flooded zones are formed (Figure 3). In these flat valleys, ponds 120000 to 160000m² and adjacent wet areas of about 10,000m² influenced by beavers, were observed. Mean surface area of ponds in mountainous areas was 2660m² (300-16000m², std=3587.89, n=22) depending on the geomorphology of the basins. For steep slopes, riparian areas influenced by beavers were estimated to be about 400m².

Beavers colonized all streams and occurred in nearly all aquatic habitats. Low gradient (0-6º), first and second order streams were selected by beavers. Higher-order streams were seasonally occupied (Coronato et al., 2003). The rivers of the lateral hanging valleys were more densely occupied than those of the slope units. Clear-cutting by beavers in riparian forests dominated by deciduous (Notophagus pumilio) or evergreen (N. betuloides) species promoted the formation of grasslands with high productivity, probably due to increased light, organic material, and humidity (Lizarralde, 1993).

Submerged vegetation and algae (Diatoma, Cyanophita and Clorophyta), which indicate high N content, were also identified in beaver sites. It was also noted that these environments are suitable for introduced fish species (Salmo trusttafario, Salvelinus fontinalis and Onchorrychus mybis), and that they sustain invertebrate communities typical of slow-water habitats. In addition, they became the nesting places of a number of migratory bird species (Lizarralde et al., 1996a).

Austral beavers forage on trees of the genus Nothofagus (N. betuloides, N. pumilio and N. antarctica) and Drymis winteri ("canelo"), shrubs such as Pernettia mucronata ("chaura"), Berberis ilicifolia ("michay"), and Chillotricum diffusum ("mata negra"), and grass of the genera Marsipospernum and Iuncus.

The immediate effect of beaver-induced deforestation was increased erosion because slopes were directly exposed to the winds. Riparian grasslands with high productivity due to increased organic material and humidity were observed in Class A basins.

Wood debris from fallen trees causes an accumulation of organic material that modifies the biochemical composition of waters, sediments, soils and adjacent riparian areas. Data indicated that beaver dams were able to retain sediment levels from 684 to 120000m³. Nutrients are probably released to riparian vegetation during seasonal periods of high floods when the ponds are flushed. Standing stocks of principal elements differed greatly between unaltered and beaver-altered sites. Mean organic C, organic N, organic P, and inorganic N (NO₂ and NO₃) were significantly greater in beaver-altered sites. Beaver pond water had significantly greater concentrations of nitrate, nitrite, and K than control water. Beaver ponds stored 7 times more C, 3.5 times more N and 1.85 times more P in sediments than did control ponds. Nitrites (24.75x) and nitrates (11.15x) were also more prevalent in sediments and waters (7–10x) of beaver-altered ponds. (Lizarralde et al., 1996a).

Plant and animal introductions have genetic and ecological effects on the invader species, as well as on the native ecosystems where they are introduced. North American beavers were introduced in the Isla Grande of Tierra del Fuego about 57 years ago. Several factors favored rapid population growth and range expansion. In 1981 the Government of Tierra del Fuego authorized hunting this species in order to control its increasing population.

Beavers are obligate residents of riparian areas and their construction of dams, channels and lodges produces noticeable effects on the ecosystems. As in boreal forests, we have reported increases in the accumulation of organic compounds and changes in decomposition processes and nutrient cycles as a result of beaver activities. Beavers have also created conflict in areas where timber exploitation is important.

From 1946 up to several years ago, there was no information on the beaver population in Tierra del Fuego. For many years it was not known whether beavers had adapted successfully to the Fueguian ecosystem. Beavers modify stream morphology and hydrology by felling trees and building dams; the latter retain sediment and organic matter in the channel, create and maintain wetlands, modify nutrient cycling and decomposition dynamics, alter the structure and dynamics of the riparian zone, influence the character of water and materials transported downstream, and ultimately influence biotic community composition and diversity (Naiman et al., 1986; Naiman, 1988). Areas abandoned by beavers produced fertile grasslands in Tierra del Fuego. Our results indicate that beaver-altered sites had higher levels of organic and inorganic N (nitrites and nitrates), suggesting that seasonal hydrological changes could be affecting nitrification and denitrification, and also resulting in accumulated organic C and P in the stream channel. Consequently, by changing the hydrological regime, beavers alter the character of stream channels. Our data also suggest that beaver ponds may be considered sources of essential nutrients (P and N) and C. Several characteristics of rivers used by beavers in Tierra del Fuego were found to be similar to those in the Northern Hemisphere (Naiman et al., 1988; Smith et al., 1991; Coronato et al., 2003).

Beavers can be considered major factors influencing nutrient transformation processes in southern beech (Nothofagus spp.) forests. They greatly influence microbiological processes such as N mineralization, organic C decomposition, and P availability. Beaver impoundments caused less alteration of ecosystem level processes in wetlands than in uplands, because the soil and vegetation was already in under-saturated conditions before impoundment (Naiman et al., 1988; Lizarralde et al., 1996a).

Modifications of the aquatic and terrestrial biotic composition were also observed in the southern beech forest. Our studies also found that beavers forage more selectively at increasing distances from the central place (pond). Data obtained from study transects could be related to several predictions of the theory of central place foraging (Orians and Pearson, 1979; Jenkins 1980; Lizarralde et al., unpublished data). Basey et al. (1988) reported that beavers often select trees with large trunks which probably contain low concentrations of a specific phenolic compound. This phenolic compound, which may act as a deterrent, was found in aspen bark with a significant negative regression of relative concentrations and tree diameter. Future studies should attempt to determine whether similar phenomena occur in Nothofagus spp.

Chromosome analysis of Giemsa-stained preparations showed a similar karyotype to that of Castor canadensis in the Northern Hemisphere (Genest et al., 1979; Hsu and Benirschke, 1968; Ward et al., 1991; Lizarralde et al., 1996b). However, the austral karyotype differs only in the X sexual chromosome because it was more variable and totally metacentric in all the animals analyzed. C. fiber and C. canadensis karyotypes share a fundamental number (FN=80), but differ by 8 acrocentric pairs present in the C. fiber karyotype. Several authors assumed that the C. canadensis karyotype has originated by Robertsonian fusions of C. fiber karyotype (Lavrov and Orlov, 1973; Sysa and Zurowsky, 1980).

Data revised herein were obtained during an intensive 15-year study. This led to a management plan to control the population through trapping (Lizarralde and Escobar, 1999). Beaver trapping in Tierra del Fuego began to be regulated 3 years ago by the Natural Resources Administration. Kill traps (Conibear model 330™) are used.

The austral beaver is an excellent example of an introduced invader species that rapidly exploited the environment in the new habitat. Twenty-five mated pairs from Canada have grown to a population of 35000-50000 animals. What is the magnitude of the genetic drift that has occurred in this isolated population? This, and many other questions, remain to be answered and require extensive molecular studies. Beavers in Tierra del Fuego also represent a model to study the process of invasion.

At present, molecular genetic markers are being analyzed to detect levels of genetic variation, and to identify DNA polymorphisms and mitochondrial linkages to understand the effects of severe reduction of genetic variability on the survival ability and reproductive success of Argentine beavers. Low levels of variability also have been reported in C. fiber and Ondatra zibheticus (Ellegren et al., 1993; Marinelli et al., 1992).

The results of this study will have important implications for biodiversity conservation and wildlife management.

The authors express their recognition for the financial support provided by CONICET, Argentina (PID-BID 50/91, PIP 4306/96), International Foundation for Sciences, Sweden (grants awarded 1988-1994), and the Tierra del Fuego Government.

1. Basey J, Jenkins S, Busher P (1988) Optimal central place foraging by beavers: tree-size selection in relation to defensive chemicals of quaking aspen. Oecologia 76: 278-282.        [ Links ]

2. Collins TC (1976) Population Characteristics and Habitat Relationships of Beavers Castor canadensis in Northwest Wyoming. Thesis. University of Wyoming. Laramie, WY, USA. 188 pp.        [ Links ]

3. Coronato A, Escobar JM, Mallea C, Roig C, Lizarralde MS (2003) Geomorphological characteristics of mountain watershed rivers colonized by Castor canadensis in Tierra del Fuego. (In Spanish). Ecol. Aust. 13: 15-26.        [ Links ]

4. Ellegren H, Hartaman G, Johansson M, Andersson L (1993) Major histocompatibility complex monomorphism and low levels of DNA fingerprinting variability in a reintroduced and rapidly expanding population of beavers. PNAS 90: 8150-8153        [ Links ]

5. Genest FB, Morisset P, Patenaude RP (1979) The chromosomes of the canadian beaver Castor canadensis. Can. J. Genet. Cytol. 21: 37-42.        [ Links ]

6. Hsu TC, Benirschke K (1968) Castor canadensis (beaver). In Atlas of mammalian chromosomes. Vol 2. Springer. New York, USA. p. 59.        [ Links ]

7. Jenkins SH (1980) A size distance relation in food selection by beavers. Ecology 61: 740-746.        [ Links ]

8. Lavrov LS, Orlov VN (1973) Karyotypes and taxonomy of modern beavers (Castor, Castoridae, Mammalia). Zool. Zh. 53: 734-742.        [ Links ]

9. Lavrov LS, Orlov VN (1973) Karyotypes and taxonomy of modern beavers (Castor, Castoridae, Mammalia). Zool. Zh. 53: 734-742        [ Links ]

10. Lizarralde MS (1993) Current status of the beaver (Castor canadensis) introduced in Tierra del Fuego (Argentina). AMBIO 22: 351-358.        [ Links ]

11. Lizarralde MS, Elisetch M (2001) Economic Significance and Standard Development of Mammal Trapping in Argentina. In Field R, Warren RJ, Okarma H, Sievert PR (Eds.) Wildlife, Land and People: Priorities for the 21st. Century. The Wildlife Society. Bethesda, USA. pp. 340-342.        [ Links ]

12. Lizarralde MS, Escobar JM (1999) Management of beaver Castor canadensis population in Tierra del Fuego (Argentina). (In Spanish). Gobierno Tierra del Fuego, Argentina. 10 pp.        [ Links ]

13. Lizarralde MS, Escobar JM (2000) Exotic mammals in Tierra del Fuego, Ciencia Hoy 10: 52-63.        [ Links ]

14. Lizarralde MS, Venegas C (2002) El castor un ingeniero exótico en las tierrras más australes del planeta (In Spanish). In Primack, Rossi, Feinsinger, Dirzo, Massardo (Eds.) Fundamentos de Conservación Biológica. Fondo de Cultura Económica. México. pp. 231-232.        [ Links ]

15. Lizarralde MS, Escobar J, Bianciotto O (1989) Beaver (Castor canadensis) in Tierra Del Fuego: Alteration effects and its exploitation as natural resource. (In Spanish). Reports of Museo del Fin del Mundo. Tierra del Fuego, Ushuaia, Argentina. 25 pp.        [ Links ]

16. Lizarralde MS, Deferrari G, Escobar JM, Alvarez SE (1996a) Nutrient dynamic alterations induced by Beaver (Castor canadensis) on southern forest ecosystem. Ecol. Aust. 6: 101-105.        [ Links ]

17. Lizarralde MS, Deferrari G, Escobar JM, Alvarez SE (1996b) Current status of the introduced Castor canadensis population in Tierra del Fuego: its chromosome analysis. (In Spanish). CADIC. Gobierno Tierra del Fuego, Argentina. 15 pp.        [ Links ]

18. Lizarralde MS, Deferrari G, Álvarez SE, Escobar JM (2002) Effects of beaver (Castor canadensis) impoundments on the nutrient dynamics of streams of Tierra del Fuego (Argentina) (In Spanish). Proc. VI Int. Symp. Patagonia Watersheds. pp. 69-72.        [ Links ]

19. Marinelli L, Messier F, Plante Y (1992) Use of DNA fingerprinting to determine parentage in muskrats (Ondatra zibethicus). J. Hered. 83: 356-360.        [ Links ]

20. Naiman R, Melillo J, Hobbie J (1986) Ecosystem alteration of boreal forest stream by beaver (Castor canadensis). Ecology 67: 1254-1269.        [ Links ]

21. Naiman R (1988) Animal influences on ecosystem dynamics. Bioscience 38: 750-752.        [ Links ]

22. Naiman R, Descamps H, Pastor J, Johnston C (1988) The potential importance of boundaries to fluvial ecosystems. J. North Am. Benthol. Soc. 7: 289-306.        [ Links ]

23. Nordstrom W (1972) Comparison of trapped and untrapped beaver population in New Brunswick. Thesis. University of New Brunswick. Canada. 104 pp.        [ Links ]

24. Orians GH, Pearson NE (1979) On the theory of central place foraging. In Horn DJ, Mitchell RD, Stairs GR (Eds.) Analysis of ecological systems. Ohio State University Press. Columbus, OH, USA. pp. 155-177.        [ Links ]

25. Paine RT (1966) Food web complexity and species diversity. Amer. Naturalist 100: 65-75.        [ Links ]

26. Pisano E (1981) Phytogeography of the Region Fuego Patagonia. (In Spanish) Anales del Instituto de la Patagonia (Pta. Arenas, Chile) 12: 159-171.        [ Links ]

27. Smith M, Driscoll C, Wyskowski B, Brooks C, Cosentini CH (1991) Modification of stream ecosystem structure and function by beaver (Castor canadensis) in the Adrirondack Mountains, New York. Cand. J. Zool. 69: 5-61.        [ Links ]

28. Sysa P, Zurowski W (1980) The chromosomes of Eurasian beaver (Castor fiber L.) from Pasleka River (Poland). 4th Europ. Coll. Cytogen. Domest. Anim. (Uppsala) pp. 432-436.        [ Links ]

29. Ward OG, Graphodatsky AS, Wurster-Hill DH, Eremina VR, Park JP, Yu Q (1991) Cytogenetics of beavers: a case of speciation by monobrachial centric fusions. Genome 34: 324-328.        [ Links ] ]]> 1 1988 76 278-282 2 1976 188 3 2003 13 15-26 4 1993 90 8150-8153 5 1979 21 37-42 6 1968 2 59 7 1980 61 740-746 8 1973 53 734-742 9 1973 53 734-742 10 1993 22 351-358 11 2001 340-342 12 1999 10 13 2000 10 52-63 14 2002 231-232 15 1989 25 16 1996 a 6 101-105 17 1996 b 15 18 2002 69-72 19 1992 83 356-360 20 1986 67 1254-1269 21 1988 38 750-752 22 1988 7 289-306 23 1972 104 24 1979 155-177 25 1966 100 65-75 26 1981 12 159-171 27 1991 69 5-61 28 1980 432-436 29 1991 34 324-328