Amazonian Savanna Research Articles

Geographical Variation in the Ecology of Populations of Some Brazilian Species of Cnemidophorus (Squamata, Teiidae)

We investigated geographical variation in ecological parameters among populations of Cnemidophorus cryptus, Cnemidophorus gramivagus, Cnemidophorus lemniscatus, Cnemidophorus ocellifer, and Cnemidophorus parecis, from three Brazilian biomes (Cerrado, Caatinga, and Amazonian Savannas). Lizards used mainly the open ground, with a high similarity in microhabitat use among populations. Differences in microhabitat use probably resulted from the availability of microhabitats and not from microhabitat preferences. Body temperatures were high and little influenced by environmental temperatures, there being no differences among populations. There were significant differences in diet among populations, with C. ocellifer from Caatinga and Cerrado consuming large quantities of termites, whereas Amazonian Savanna species used primarily ants and insect larvae. The data on reproductive seasonality indicated cyclical reproduction in seasonal biomes and continuous reproduction in unpredictable climate regions. We found significant differences in mean clutch size among populations, independent of body size, with C. lemniscatus having the smallest clutch size (1.50) and C. ocellifer from Cerrado the largest (2.10). There were fewer differences in clutch size among species from Amazonian Savannas, than between populations of C. ocellifer from Caatinga and Cerrado. Apparently, populations under seasonal climates concentrate their reproductive effort during the short reproductive season, producing larger clutches, whereas those under more stable or unpredictable climates reproduce continuously, yielding smaller clutches, corroborating the hypothesis that environmental conditions exert an important influence upon life-history parameters. There were significant differences in body shape among populations, but most of the variation was related to sex. We also observed significant differences in body size among populations, seemingly unrelated to differences in community structure, but the highly conservative morphology of Cnemidophorus species suggests the presence of historical constraints.

ANALYSIS OF THE FLORISTIC COMPOSITION OF THE BRAZILIAN CERRADO VEGETATION III: COMPARISON OF THE WOODY VEGETATION OF 376 AREAS

An analysis was made of the floristic composition of 376 areas of cerrado and Amazonian savanna, encompassing most of the area of such vegetation in Brazil. A total of 951 species of trees and large shrubs was recorded for these areas, of which 334 (35%) occurred at only a single locality (unicates).

Size and Carbon Acquisition in Lizards from Amazonian Savanna: Evidence from Isotope Analysis

We studied the effects of lizard size, and availability of C3 and C4 plants, on stable carbon-isotope (δ13C) ratios of lizards in an Amazonian savanna. Availability of C3 and C4 plants was estimated from 500-point quadrats in each of 38 plots of 3.75 ha. Ratios of δ13C in muscle tissue were used to estimate the contributions of C3- and C4-based food chains to the diets of four species of lizards: Cnemidophorus lemniscatus, Kentropyx striata, Ameiva ameiva, and Anolis auratus. The relative contributions of C3 and C4 plants to the diets of the lizards were affected by the extent of shrub cover in C. lemniscatus, A. auratus, and possibly in K. striata. However, there was a strong effect of lizard size among and within lizard species. These results are at variance with a previous evaluation of diet overlap among size classes of these species that was based on overlap in prey taxonomic categories. This indicates that species and/or size classes of predators on arthropods may obtain their energy from food chains based on different autotrophic sources, despite high overlap in the taxonomic categories of prey.

SIZE AND CARBON ACQUISITION IN LIZARDS FROM AMAZONIAN SAVANNA: EVIDENCE FROM ISOTOPE ANALYSIS

We studied the effects of lizard size, and availability of C3 and C4 plants, on stable carbon-isotope (δ13C) ratios of lizards in an Amazonian savanna. Availability of C3 and C4 plants was estimated from 500-point quadrats in each of 38 plots of 3.75 ha. Ratios of δ13C in muscle tissue were used to estimate the contributions of C3- and C4-based food chains to the diets of four species of lizards: Cnemidophorus lemniscatus, Kentropyx striata, Ameiva ameiva, and Anolis auratus. The relative contributions of C3 and C4 plants to the diets of the lizards were affected by the extent of shrub cover in C. lemniscatus, A. auratus, and possibly in K. striata. However, there was a strong effect of lizard size among and within lizard species. These results are at variance with a previous evaluation of diet overlap among size classes of these species that was based on overlap in prey taxonomic categories. This indicates that species and/or size classes of predators on arthropods may obtain their energy from food chains based on different autotrophic sources, despite high overlap in the taxonomic categories of prey.

Breeding and Migrating Birds in an Amazonian Savanna

Seasonal occurrence, breeding and migration records are presented for about 150 species of birds, monitored during a four years survey in an Amazonian savanna near Alter do Chão, Pará state, Brazil, by means of mist netting and observations along transects and on random walks. About half of the recorded species were Passeriformes. The local avifauna was mostly comprised of residents, with major breeding activities during the dry season, August through February. Thirteen non-resident species were classified as migrants which is considered a rather small fraction. Calculations of the total number of species expected to occur at the study site are presented. The composition of this savanna bird community is discussed under aspects of seasonality and migrational activity. The data are compared with records from other cerrado habitats in Amazonia and elsewhere, especially in Central Brazil. Resumo Registros de ocorrência sazonal, reprodução e migração são apresentados para cerca de 150 espécies de aves, monitoradas durante quatro anos de levantamentos em uma savana Amazônica perto de Alter do Chão, Estado do Pará, Brasil, através de redes ornitológicas, observações ao longo de transectos e caminhadas ao acaso. Perto da metade das espécies registradas foram passeriformes. A avifauna local foi constituída principalmente de espécies residentes, com o pico de atividades reprodutivas durante a estação seca, ou seja 80% das espécies reproduzindo de Agosto até Fevereiro. Este padrão se repetiu em 1988 e 1989. Juvenis foram capturados durante todo o ano com picos durante a estação chuvosa. Treze espécies não residentes foram consideradas migrantes os quais são uma pequena fração da comunidade. Estimativas do número total esperado de espécies ocorrendo na área de estudo são apresentados. As capturas com redes ornitológicas em 4 ha resultaram em 41 espécies e para 13 ha, 64 espécies. Quando a área amostrada foi duplicada para 24 ha, o número de espécies aumentou em somente 4 e correspondeu à metade do total registrado para a área. O número assintótico estimado para 13 ha foi de 82 espécies o qual foi similar ao número estimado para 24 ha (80 espécies). A composição desta comunidade de aves de savana é discutida sob os aspectos de sasonalidade e atividades migratórias. Os resultados são comparados com registros de outros habitats de cerrado na Amazônia e outros locais especialmente no Brasil central.

Diversity and distribution of frogs in an Amazonian savanna in Brazil

AbstractWe studied the distribution of anurans in savanna, regrowth and mature forest at the mouth of the Tapajós river in Brazilian Amazonia. We sampled 31 sites on the right bank of the Tapajós river and on the shores of lakes nearby, and recorded all acoustic and visual observations of frogs. We found 18 anuran species: Leptodactylidae (8), Hylidae (7), Bufonidae (2) and Pseudidae (1). The distribution of species indicated that some species are generalists and others are restricted to certain habitats. The species at Alter do Chão are a subset of those found in forested sites in Amazonia. Lakes and forested areas had the greatest anuran diversity, but are also the areas most disturbed by tourist development and agriculture. Reduction of anuran diversity in the area could be minimized by concentrating development along the main beaches of the Tapajós river.

Contributions of C3 and C4 plants to higher trophic levels in an Amazonian savanna

We studied the energy flow from C3 and C4 plants to higher trophic levels in a central Amazonian savanna by comparing the carbon stable-isotope ratios of potential food plants to the isotope ratios of species of different consumer groups. All C4 plants encountered in our study area were grasses and all C3 plants were bushes, shrubs or vines. Differences in δ13C ratios among bushes (x¯ = -30.8, SD = 1.2), vines (x¯ = -30.7, SD = 0.46) and trees (x¯ = -29.7, SD = 1.5) were small. However the mean δ13C ratio of dicotyledonous plants (x¯ = -30.4, SD = 1.3) was much more negative than that of the most common grasses (x¯ = -13.4, SD = 0.27). The insect primary consumers had δ13C ratios which ranged from a mean of -29.5 (SD = 0.47) for the grasshopper Tropidacris collaris to a mean of -14.7 (SD = 0.56) for a termite (Nasutitermes sp.), a range similar to that of the vegetation. However, the common insectivorous and omnivorous vertebrates had intermediate values for δ13C, indicating that carbon from different autotrophic sources mixes rapidly as it moves up the food chain. Despite this mixing, the frogs and lizards generally had higher values of δ13C (x¯ = -21.7, SD = 1.6; x¯ = -21.9, SD = 1.8, respectively) than the birds (x¯ = -24.8, SD = 1.8) and the only species of mammal resident in the savanna (x¯ = -25.4), indicating that they are generally more dependent on, or more able to utilise, food chains based on C4 grasses.

Tropical Tadpole Vulnerability to Predation: Association between Laboratory Results and Prey Distribution in an Amazonian Savanna

We tested concordance between patterns of tadpole assemblage structure seen in the field and in laboratory experiments to infer causal relationships. Based on a previous study that showed predators as the main proximal factor associated with tadpole distributions in an Amazonian savanna, we investigated interspecific patterns of vulnerability of six common tadpole species in relation to the distribution of predators in the field. The predators used were hydrophilid larvae (Coleoptera), libellulid naiads (Odonata), and adult Geophagus gr. altifrons (Cichlidae). Forty riverside aquatic sites were searched for predators and tadpoles during 28 months in a savanna area at the border of the Tapaj6s River, Central Amazonia. Tadpole species were tested for their vulnerability to predators in individual and choice tests. Prey unpalatable to fishes (Bufo marinus, B. granulosus, and Hyla wavrini) were palatable to invertebrates. Leptodactylus macrosternum and Scinax rubra were vulnerable to fish and invertebrate predators. Lysapsus limellus was not tested with fish, but it was vulnerable to naiads. Eggs had low vulnerability to predation, but there was generally little ontogenetic change in tadpole vulnerability. The vulnerability of tadpoles in choice tests was significantly related to vulnerability in individual tests. A tadpole vulnerability index based on tadpole survivorship with different predators was significantly related to the distribution of predators in the field, indicating that it is possible to predict the coexistence of prey and predators in the field by screening species with simple laboratory predation experiments. In this study area, both vertebrate and invertebrate predators may potentially limit the distribution and the relative abundance of tadpole species through differential predation.

Home-Range Size of the Bare-ear Marmoset ( Callithrix argentata ) at Alter do Chão, Central Amazonia, Brazil

We studied factors affecting variation in home-range size of four groups of bare-ear marmosets (Callithrix argentata) in patches of forest within a central Amazonian savanna. We determined relative use of different parts of the home range by radiotelemetry. We estimated fruit availability monthly in transects through each home range and mapped the habitats within the home range of each group. We determined the densities of gum- and fruit-producing trees in 18 50 × 50-m quadrats and related these data to the frequency of use by marmosets. Home-range size varied by a factor of 6 between groups, even though the study area covered <15 km2and included only one major biome. Marmoset activity was concentrated in areas with many gum-producing trees. Monthly range size is positively correlated with fruit availability only for the group with the largest home range; the other groups appeared to be responding to other factors. Home-range sizes appeared to be limited by the size of the main patch of contiguous forest available to each group. Our findings suggest that conservation planning that does not consider the possibility of large differences among primate home-range sizes may be unsatisfactory.

Foraging Mode and Reproductive Seasonality in Tropical Lizards

Foraging Mode and Reproductive Seasonality in Tropical Lizards

Lizards, Snakes, and Amphisbaenians from the Quaternary Sand Dunes of the Middle Rio Sao Francisco, Bahia, Brazil

Lizards, Snakes, and Amphisbaenians from the Quaternary Sand Dunes of the Middle Rio Sao Francisco, Bahia, Brazil

Analysis of the floristic composition of the Brazilian cerrado vegetation II: Comparison of the woody vegetation of 98 areas

An analysis was made of the floristic composition of 98 areas of cerrado and Amazonian savanna, encompassing most of the area of such vegetation in Brazil. A total of 534 species of trees and large shrubs were recorded for these areas, of which 158 (30%) occurred at a single site only. Such unicates and taxa without determinations to specific level were excluded from the study since they provide no basis for comparison. The data were analysed by three techniques of multivariate analysis: (a) a divisive hierarchical classification by Two-way Indicator Species Analysis (TWINSPAN). (b) an agglomerative hierarchical classification by UPGMA (Unweighted Pair-Groups Method using Arithmetic Averages) using the Sørensen Coefficient of Community (CC) as a measure of similarity, and (c) an ordination by Detrended Correspondence Analysis (DCA). The results from all three methods showed great similarity, demonstrating a strong geographic pattern in the distribution of the flora of the cerrado biome and allowing the recognition of southern (Sâo Paulo and S Minas Gerais), southeastern (largely Minas Gerais), central (Federal District, Goiás and parts of Minas Gerais), central-western (largely Mato Grosso, Goiás and Mato Grosso do Sul) and northern groups (principally Maranhão, Tocantins and Pará), as well as a disjunct group of Amazonian savannas. Soil type (mesotrophic or dystrophic) is an important factor in determining floristic composition. The study demonstrated that cerrado vegetation is extremely heterogeneous: none of the 534 species occurred at all sites and only 28 species were present at 50% or more.

Variation in growth and reproduction of Bolomys lasiurus (Rodentia: Muridae) in an Amazonian savanna

ABSTRACTReproduction by Bolomys lasiurus is strongly seasonal in the Amazonian savanna of Alter do Chāo, Pará, Brazil. No pregnant or lactating females were recorded during the early dry season (July-September). Relative growth rates showed a similar though less consistent pattern. These patterns do not appear to be due to a direct effect of rainfall as high reproduction and growth were recorded during the late dry season (October-November), even though rainfall and the availability of fruits of shrubs were low during that period. It is likely that reinitiation of reproduction during the dry season is due to an increase in the availability of invertebrates and/ or wind-dispersed seeds. Fire at the end of the dry season appears to affect the diet of the rats, causing an increase in the ingestion of invertebrates during the following wet season. However, our data do not indicate that fire affects individual reproductive output or growth.

Effects of annual fires on the production of fleshy fruits eaten by birds in a Brazilian Amazonian savanna

ABSTRACTThis study describes changes in the numbers of plant species fruiting during six years in I ha of Amazonian savanna on the right bank of the Tapajós River, Brazil. The work was based on fleshy fruits from tree and shrub-layer species eaten by birds. The area was burned in the first two years, protected during the following two years and burned again in the last two years. The number of tree species fruiting did not differ significantly between years with and without fire. The number of shrub-layer species fruiting was markedly reduced in years with fire, but the responses varied between species. Data on changes in shrub cover after two consecutive years of fire indicate that species also respond differently in terms of biomass lost due to fires. The two most abundant genera of shrubs, Miconia and Myrcia, had their cover reduced by up to 95% and 72% respectively. The area made available by the loss of shrubs was occupied mainly by the grass Paspalum carinatum. The fires and the decrease in the availability of fruits occurred during the season in which most species of birds breed. An interval of less than three years between fires is insufficient for shrub-layer species to recover.

Geographical ecology of a Neotropical lizard: Ameiva ameiva (Teiidae) in Brazil

The large-bodied teiid lizard Ameiva ameiva was studied at eight different sites in four major South American habitats of Brazil: caatinga, cerrado, Amazonian rain forest, and Amazonian savanna. We found striking similarity in ecological attributes of this lizard among very different habitats. Activity is concentrated in late morning and early afternoon; active body temperatures average 37.9 ± 0.09 °C and vary little among sites or throughout the day; the diet consists of a variety of vertebrates and invertebrates but is dominated by grasshoppers, roaches, beetles, spiders, and insect larvae; and niche breadths for prey are similar among study sites but the actual composition of the diets varies. There is minimal morphological variation among sites (mostly size); the most striking morphological variation is between the sexes. Males reach larger body sizes and have relatively larger heads than females. Juveniles have relatively larger heads than would be predicted on the basis of body size alone. Sexual selection may explain the sexual differences in head size of adults, whereas the relatively large heads of juveniles may be associated with food competition with sympatric teiid lizards. Clutch size varies from 1 to 11 eggs, is related to female body size (snout–vent length), and varied among study sites. Similar variation among sites occurs in egg size but not in relative clutch mass. An interesting positive relationship was found between body size and relative clutch mass. In a population from the state of Rondônia egg dry mass was correlated with female size, indicating that individual offspring size is, to some extent, a consequence of female size. The reproductive season is extended for all populations and it appears that predictability of rainfall may regulate the season length. Reasons for the apparent success of A. ameiva in a diversity of habitats on a large geographic scale include its large body size, foraging mode, and preferred microhabitat (ecotones and disturbed areas).

Amazonian savanna vegetation on the sandstone plateau near Araracuara, Colombia

Amazonian savanna vegetation on the sandstone plateau near Araracuara, Colombia

Relative Effects of Size, Season and Species on the Diets of Some Amazonian Savanna Lizards

The foods taken by Ameiva ameiva, Cnemidophorus lemniscatus, Kentropyx striatus, and Anolis auratus in the Amazonian savanna at Alter do Chao depend more on the species of lizard than on lizard size. Lizard size affects the size of food items but this has relatively little effect on the types of foods eaten. The effect of body size appears to be similar for both stationary vegetable foods (fruits), and for the total diet, which consist mainly of invertebrates. The occurrence of vertebrates in the diet of the teiids is much less than for the same species in a seasonally inundated Amazonian savanna. Diet may vary among species of lizards be- cause they differ in size, use different habitats, or because they differ in foraging behavior (Toft, 1985). Competition is often inferred as the ul- timate cause of these differences but evolution- ary hypotheses are difficult to test. Questions about proximal mechanisms are also difficult to analyze statistically. It is possible to test for any difference among the diets of an assemblage of lizards but this question is usually trivial. The null hypothesis is that the lizards have inde- pendently evolved exactly the same diet. This possibility is sufficiently remote that the rejec- tion of the null hypothesis is simply a question of sample size. Questions about communities are more profitably examined as the relative

Development and distribution of the epiphytic flora in an Amazonian savanna in Brazil

The abundance and distribution of epiphytes were studied in a plot of about one third hectare of Amazonian savanna close to the city of Humaitá (7°31′ S, 63°00′ W; altitude ≈ 50 m). Non-vascular epiphytes (lichens and bryophytes), apparently undergo succession on trunks and branches of their host trees. While lichens dominate in young trees, and usually diminish in quantity on older ones, bryophytes are more abundant on older trees and seem to surpress the lichens. In total, 118 juvenile and adult individuals of 7 vascular epiphyte species (Catasetum saccatum: 65 individs., Polypodium decumanum: 33 individs., P. hirsutissimum: 16 individs., Polystachya estrellensis, Polypodiaceae sp., Gesneriaceae sp. and Araceae sp.: 1 individ. each) were found on 18 of the 111 tree individuals of the plot The furrowed and persistent bark of Physocalymma scaberrimum-trees seems to offer by far the best conditions for settlement and growth (average 2 epiphytes per tree). On the other hand, the flaky deciduous bark of Curatella americana and the smooth one of Cordia bicolor (averages about 0.2 epiphytes per tree) apparently make settlement more difficult. As much as 113 epiphytes (almost 96 %) were found on 15 trees higher than 4.5 m, whereas only 5 (about 4 %) had settled on 3 lower ones. Further, there appears to exist a positive correlation between density of vegetation and number of epiphyte individuals. Since the humid Amazonian climate permits vascular epiphytes to grow even in strongly insolated open savannas, the factors which determine their development seem to be the specific surface characteristics (mainly bark) of the host trees, the age of the trees, and also the density of vegetation. However, because of the low number of epiphyte bearing trees within the investigated plot, none of these three chracters is statistically significant.

Reproductive Cycles of Teiid Lizards in Amazonian Savanna

-Ameiva ameiva, Cnemidophorus lemniscatus, and Kentropyx striatus reproduce seasonally in Amazonian savanna near Alter do Chao, Brazil. Most egg deposition occurs during the wet season but most development of gonads and growth of juveniles occur during the height of the dry season. Relative fat-body size varies little throughout the year and there is no apparent correlation between fatbody size and gonad development. Reproductive cycles of reptiles in seasonal tropical climates have been considered to be correlated with rainfall, most egg deposition occuring during the wet season (Fitch, 1982). However, recent work indicates that the reproductive cycles of tropical lizards show great variability and that reproductive cycles are not always tightly linked to rainfall patterns (Vitt, 1986). Variation in the availability of food is generally given as the reason for cycles. On the other hand, Stamps and Tanaka (1981) have shown that the availability of water may limit the growth of Anolis aeneus independent of the effects of food availability, and Rose (1982) found that supplemental feeding increased fat stores but not reproduction in Anolis acutus. Dunham and Miles (1985) suggest that phylogenetic constraints and biogeographic history may have played major roles in shaping squamate reproductive patterns. James and Shine (in press) suggest that, in northern Australia, lizard species reproduce in the season (wet or dry) that most closely simulates climatic conditions where their antecedents evolved. The data base is insufficient to define clearly the reproductive patterns of lizards from seasonal tropical environments. Here I present data on the reproductive cycles of three sympatric species of teiid lizards (Cnemidophorus lemniscatus, Ameiva ameiva, and Kentropyx striatus) which occur in Amazonian savanna. Seasonal variation in interstitial testicular tissue (Del Conte, 1972) and testis volume (Leon and Cova, 1973) has been reported for Venezuelan C. lemniscatus but female C. lemniscatus with oviductal eggs were found in the garden of the Universidad de Orient en Cumana in most months (Leon and Cova, 1973). Ameiva ameiva in rainforest habitat in Ecuador have strongly seasonal reproduction (Simmons, 1975) but A. ameiva in the arid caatinga of northeastern Brazil breed continuously and show only weak seasonality which is not directly related to rainfall (Vitt, 1982). I know of no studies of the reproductive cycle of K. striatus. Reproductive cycles can be inferred from: (1) longitudinal studies of individual animals; (2) gonadal cycles of males and females; (3) population age structure; (4) the presence of eggs or neonates in the population; or a combination of all four lines of evidence. In this study I present data on aspects (2), (3), and (4) for Cnemidophorus lemniscatus, (3) and (4) for Ameiva ameiva, and (4) only for Kentropyx striatus. During a previous study (Magnusson et al., 1985) we noted that the mean clutch size differed between C. lemniscatus collected in the savanna and on freshwater beaches. The densities of C. lemniscatus in these two habitats also differed significantly (Magnusson et al., 1986). I therefore collected data on possible population differences between habitats for that species. Fat-body cycles are often closely related to reproductive cycles in temperate lizards (Derickson, 1976). However, there may be less relationship between fat-body cycles and reproduction in tropical lizards (DerThis content downloaded from 157.55.39.104 on Sun, 19 Jun 2016 06:07:10 UTC All use subject to http://about.jstor.org/terms WILLIAM E. MAGNUSSON 7 5 7 3 2 7 4 A e. /-Reproductive A Ameiva / /GArovid

Floristic, Structural and Phytogeographical Analysis of the Savannas of Humaitá (Amazonas)

The extensive Puciari-Humaitá savannas in the southern Amazon region were partly investigated in regard to its flora, physiognomy, soils and climate. They can be defined as open tree-woodland which is neither floristically nor ecologically a cerrado. A quantitative floristic analysis revealed ca. 100 species/ha, 40 of them woody and 10 arboreous. The vegetation is usually dominated by 1 to 3 tree species, e.g. Curatella americana and Physocalymma scaberrimum . A low degree of species diversity per genus and/or family is typical. Stronger floristic relations to the surrounding forests and relations to northern and southern savannas are evident. Some species are widespread and common neotropical savanna plants, some are local endemics. The structure of the vegetation is documented by profile diagrams and various quantitative data. Leaves of all trees have a very similar rough silicious surface; 79.5 % are mesophyllic, 18.6 % macrophyllic and 1.9 % microphyllic. The soils are extremely poor in nutrients, rich in exchangeable Al and ill-drained; they are probably mainly responsible for the permanence of the open savannas within the forest region. The climate is humid-tropical with a slightly pronounced dry season. The vegetation is interpreted partly as being an impoverished relic savanna from the former large and species rich, dry pleistocene Amazonian savannas.