“I'd forgotten how trees full of bird sounds made you sense the world differently.” (Barbara Kingsolver, The Bean Trees)

No one I have ever met was more attuned to bird sounds, or to birds in general, than was Ted Parker. It was from Ted that many of us, who thought that we already knew a thing or two about birds, truly learned how to sense the world.

No matter when one first met, or heard of, Ted Parker, the name was already a legend. I first met Ted over 20 years ago, and like many people at the time I was already somewhat in awe of him. Then he was known for his birding exploits in the United States and in Mexico. Although I did not realize it at the time, he was already studying the birds of South America, and soon would be transforming our understanding of the avifauna of that continent. By the end of the 1970s, within just a few years of his first trip to the Andes, and without benefit of field guides or commerical bird tapes, he was already the field ornithologist of South America.

It is now two years since the death of Ted Parker, and I still find it extremely difficult to talk or write about him in the past tense. Indeed, I doubt whether anyone linked with him through ornithology or conservation has yet come to terms with the enormity of the loss. At one level, it is simply that he knew so much of Neotropical birds and their habitats, in so much depth and so much breadth: no-one was his match, and no-one expected ever to be so. At another, it is simply that he was so distinct and vivid a personality, much more than merely “charismatic”: superhumanly energetic, trenchantly assertive, irrepressibly good-natured, subversively frank; and always, inexhaustibly, himself. This mix of features made him larger than life — legendary, even — and awesomely enriching to know, even to someone such as myself who, although corresponding with him for years, met him only twice.

Because museum scientists and conservationists are natural allies in the struggle to preserve biodiversity, conflict over the legality, morality, and value of collecting scientific specimens is counterproductive. Modern bird specimens contain a variety of data, summarized briefly herein, that are applied to numerous questions concerning the biology of birds, many of which have direct and often critical relevance to conservation. In particular, continued collecting of specimens has been shown to be critical in determining species-level classification in birds; unless species limits are established correctly, conservation priorities cannot be established reliably. Objections to collecting specimens are summarized and discussed. Calculations are presented to show that the effect of collecting specimens on most bird populations is insignificant. Moral objections t o collecting specimens seem to reflect a lack of awareness of the extent and causes of natural mortality, as well as a failure to recognize the magnitude of unintentional mortality inflicted on bird populations by routine human activities. The reason why more specimens are needed than currently exist in museum collections is that most existing specimens lack the data needed for most kinds of modern analyses, and even common species are represented by inadequate samples for research. Reasons are given for why equivalent data cannot be obtained solely from living birds that are subsequently released. Objecting to collecting specimens because it sets a bad example for developing countries trying to establish an environmental ethic is counterproductive in that it draws attention away from the fundamental units of concern for conservation biology: the population, and the habitat that supports it. Biological specimens differ from some other scientific specimens (e.g. archaeological) in that they are renewable resources whose removal does not deplete a country's national heritage. Misconceptions about museum scientists and their motives are discussed. Regarding collecting permits, recommendations are presented concerning (1) numbers of specimens, (2) percentage of specimens left in the host country, (3) species composition, (4) deposition of specimens, and (5) processing permit applications. Regulating agencies are often overly enthusiastic i n restricting scientific collecting, which is the only kind of mortality that is so highly controlled and yet from which bird species might derive benefit, whereas the same or sister agencies often permit and even encourage activities that are responsible for massive mortality in bird populations. Given that (1) the goal of scientists, conservation agencies, and governments is protection of populations, not individual birds; (2) scientific collecting has no measurable impact on the vast majority of bird populations; (3) scientific specimens represent an important source of information on bird biology and conservation; and (4) existing scientific collections are largely inadequate for answering many questions that could be answered with greater numerical, seasonal, or geographic representation, then it follows that continued scientific collecting will benefit ornithology and conservation and should, therefore, be encouraged by conservation and government agencies.

Establishing a reserve of approximately 10,000 km at a strategic location in depto. La Paz, northern Bolivia, would create the planet's richest park for birds and presumably all other forest-dwelling biota that reach peak diversity at tropical latitudes. By connecting puna and montane forest habitats of the Andes with lowland tropical forest and savannas of the Amazon basin, we predict that the proposed reserve would contain at least 1,088 bird species, or roughly 11% of all bird species on the planet. Among these are many threatened species and species with relatively small geographical ranges. The proposed reserve would also include threatened habitats, such as lower montane forest, dry forest, and grassland.

El establecimiento de una reserva de aproximadamente 10.000 km en un lugar estratégico en el departamento de La Paz, Bolivia, crearía el parque de major riqueza de aves en el planeta. Así es de suponer que la flora y fauna que habitan en la área, llegan a su grado máximo de diversidad en las latitudes tropicales. Conectando puna, los hábitats de bosque tropical (áreas bajas) y sabanas de la cuenca del Amazonas, predecimos que la reserva que se propone tendría a lo menos 1.088 especies de aves, o aproximadamente 11% de todas las especies de aves del planeta. Dentro de estas hay muchas especies amenazadas y especies con áreas de rango geográfico muy restringido. Además la reserva que se propone también incluiría hábitats amenazados, como bosque montano, bosque seco y pajonales tropicales.

Roughly 30% of the avifauna of the dry tropical forests of north-western Peru and adjacent south-western Ecuador is endemic to this small region. Large areas in this region, especially on the Ecuadorian side of the border, have been deforested, further heightening the threatened conservation status of this fauna. The largest remaining tracts of forest in this region are in the Bosque Nacional de Tumbes, Peru. This forest, which was visited in 1986 and 1988, holds significant populations of 12 threatened or near-threatened species, and smaller populations of at least three more endangered species. Six species (Aramides axillaris, Thalurania colombica, Picumnus olivaceus, Attila torridus, Arremonops conirostris, Carduelis siemiradzkii) are reported in Peru for the first time. Evidence is presented for seasonal movements of some species, including the threatened Leptotila ochraceiventris and in the small Peruvian population of Pionus chalcopterus. Based primarily on vocal evidence, it is proposed that Thamnophilus zarumae and Grallaria watkinsi be recognized as full species, and that “Empidonax” griseipectus be transferred to Lathrotriccus; further work is needed on the species status of endemic forms in other polytypic species such as Sittasomus griseicapillus and Pyriglena leuconota.

Aproximadamente 30% de la avifauna de los bosques secos tropicales del noroeste del Perú, y del suroeste de Ecuador, es endémica a esta pequeña región. Extensas areas en esta región, especialmente del lado de la frontera ecuatoriana, han sido deforestadas, aumentando así la amenaza para la conservación de la fauna. Los más grandes vestigios de bosque seco tropical que subsisten en esta región están localizados en el Bosque Nacional de Tumbes en el Perú. Este bosque, que fue visitado en 1986 y 1988, tiene poblaciones significativas de 12 especies amenazadas o con posible amenaza de extinción, y poblaciones pequeñas de 3 especies más que están en peligro de desaparecer. Notas sobre la historia natural son presentadas para 28 especies, incluyendo 11 especies que están amenazadas o en peligro de extinción. Seis especies (Aramides axillaris, Thalurania colombica, Picumnus olivaceus, Attila torridus, Arremonops conirostris, Carduelis siemiradzkii) están reportados en el Perú por primera vez. Evidencia es presentada para movimientos estacionales de algunas especies, incluyendo la amenazada Leptotila ochraceiventris y la población pequeña peruana de Pionus chalcopterus. Basandose principalmente en evidencia vocal, se propone que Thamnophilus zarumae y Grallaria watkinsi pueden ser reconocidas como especies y que “Empidonax” griseipectus puede ser transferido a Lathrotriccus. Se necesita más estudios para determinar el estatus de las formas endémicas de especies politípicas como Sittasomus griseicapillus y Pyriglena leuconota.

In order to determine whether Aratinga erythrogenys and Brotogeris pyrrhopterus can formally classified as globally threatened species, their distribution, population sizes and ecological requirements are reviewed, together with national and international trade data from the last decade. The study concludes that, on current data, it is not possible to classify the two species formally, but there are strong indications that, at least locally, the populations of the two species are in decline. The causal factors appear to be a combination of high trade levels and habitat loss. It is essential that trade moratoria are implemented, pending the outcome of further research into the species' population sizes, ecology and breeding success, and further work to determine the extent of trade. Interim recommendations are given and it is proposed that future exploitation of the species should continue only when it has been determined that it has a sustainable basis.

The distribution of 335 species of birds in 33 islands of humid montane forest in Mesoamerica is summarized, and patterns of distribution, diversity and endemism are analysed.The montane forests of Costa Rica and western Panama far exceed other habitat islands considered for species-richness, richness of species endemic to Mesoamerica, and richness of species ecologically restricted to humid montane forests. Other regions, such as the Sierra Madre del Sur of Guerrero and Oaxaca, the Los Tuxtlas region of southern Veracruz and the mountains of Chiapas and Guatemala, also hold rich and endemic avifaunas. Based on patterns of similarity of avifaunas, the region can be divided into seven regions holding distinctive avifaunas (Costa Rica and western Panama; northern Central America and northern Chiapas; southern Chiapas; eastern Mexico north of the Isthmus of Tehuantepec; Sierra Madre del Sur; interior Oaxaca; and Transvolcanic Belt and Sierra Madre Occidental), which serve as useful guides for the setting of priorities for conservation action.

Se resumen las distribuciones de 335 especies de aves en 33 islas de bosque húmedo de montaña en Mesoamerica, y se analizan patrones de distributión, diversidad y endemismo. Los bosques montanos de Costa Rica y del oeste de Panamá tienen la más alta riqueza de especies, riqueza de especies endémicas a Mesoamerica, y riqueza de especies ecologicamente restringidas a bosque húmedo de montana. Otras regiones, tales como la Sierra Madre del Sur de Guerrero y Oaxaca, la región de Los Tuxtlas y las montanas de Chiapas y Guatemala, también tienen avifaunas ricas en especies y en endémicas. Basado en patrones de similitud de avifaunas, se puede dividir Mesoamerica en siete regiones que tienen avifaunas distintas (Costa Rica y el oeste de Panamá; el norte de Centroamérica y el norte de Chiapas; el sur de Chiapas; el este de México; la Sierra Madre del Sur; el interior de Oaxaca; y el Eje Neovolcánico y la Sierra Madre Occidental), las cuales pueden servir como guias en el establecimiento de prioridades para la conservatión.

Recent ornithological surveys of humid forest remnants in the Serra da Ouricana near Boa Nova, southern Bahia, Brazil, revealed the existence of an as yet unnoticed montane Atlantic forest avifauna. Among the 220 species recorded, 27 represent range extensions of several hundred kilometres, several others were preceded only by single or a few old, disputed records from Bahia, and at least two are undescribed taxa. A portion of this avifauna has been recorded in other mountainous areas of interior Bahia (Chapada Diamantina, plateau of Maracás), where a complex vegetation mosaic that includes patches of humid forest is found.

In addition to the two undescribed passerines, nine other threatened and thirteen near-threatened species have been recorded in the study area. Considered together with the adjoining and also highly endangered “mata-de-cipo”, forest habitats in the Boa Nova area are probably the most neglected habitats in Brazil concerning bird conservation.

Levantamentos ornitológicos recentemente feitos em remanescentes florestais úmidos na Serra da Ouricana perto de Boa Nova, sul da Bahia, Brasil, revelaram a existência nessa região de uma avifauna montícola da mata Atlântica. Entre as 220 espécies registradas, estão 27 que tiveram sua área de distribuição estendida por centenas de quilômetros, muitas outras conhecidas da Bahia apenas através de registros antigos e contestados, além de ao menos duas aves não descritas. Parte dessa avifauna foi também registrada em outras áreas montanhosas do interior da Bahia (Chapada Diamantina, Chapadão Maracás), onde se encontra um complexo mosaico de vegetação que inclui manchas de floresta úmida. Além dos dois pássaros não descritos, outras nove espécies ameaçadas e treze quase-ameaçadas foram registradas na área de estudo. Considerando-se em conjunto essa área com a vizinha “mata-de-cipo”, também muito ameaçada, pode-se dizer que os ambientes florestais na região de Boa Nova são os hábitats mais negligenciados em relação à conservação de aves no Brasil.

The distribution of the ornithological localities in the cerrado region of South America is analysed. When plotted out, the distribution of total sampling localities appears to be even. However, when only those localities at which birds have been sampled in any depth (> 80 specimens collected or >100 species recorded) are considered a different pattern emerges: c.70% of the cerrado region has never been satisfactorily sampled for birds, and similar or worse situations have been reported for other groups of organisms. Priority areas for future ornithological exploration in the cerrado region are most of the states of Tocantins, Maranhão, and Mato Grosso do Sul, southern Goiás, and western Minas Gerais. In these areas, inventory efforts could be focused mainly on the avifaunas of tropical dry forests, gallery forests and “campos rupestres”. Because the biological diversity in the cerrado region is poorly known and most of this region has already been modified by human activity, it is suggested that the most feasible way to define both a set of priority areas for conservation and a coherent reserve system is to focus analyses on ecosystems and landscapes rather than on species, subspecies or populations.

A distribuição das localidades ornitólogicas na regiao do cerrado, América do Sul, é analisada. Quando todas as localidades são plotadas em um mapa, o inventário das aves nessa região parece ser bem distribuido. Entretanto, quando somente aquelas localidades que tern sido amostradas com certa intensidade (>80 espécimes coletados ou >100 espécies registradas) são plotadas, um padrão bastante distinto é encontrado: cerca de 70% da região do cerrado tem nunca sido amostrada satisfatoriamente para aves, e situações similares ou piores tern sido também descritas para o inventário de outros grupos de organismos. Áreas prioritárias para futuras explorações ornitológicas na região do cerrado são grande parte dos estados do Tocantins, Maranhão e Mato Grosso do Sul, sul de Goiás e oeste de Minas Gerais. Nestas áreas, inventários poderiam ser direcionados principalmente para as avifaunas de florestas secas tropicais, florestas de galeria e campos rupestres. Em virtude da diversidade biológica na região do cerrado ser pouco conhecida e de grande parte dessa região já ter sido modificada pelas atividades humanas, é sugerido que a melhor estratégia para definir um conjunto de áreas prioritárias para conservação e um coerente sistema de reservas seria concentrar análises sobre ecossistemas e paisagens ao invés de espécies, subespécies ou populações.

We documented the geographical distributions and habitat selection of Neotropical migrants in South America along a successional gradient in the lowlands of Amazonian Peru, and along elevational gradients in the Andes of south-eastern Peru and of eastern and western Ecuador. Most of the 30 species of northern migrants that regularly winter in South America appear to be concentrated in the western edge of the Amazon basin and on the lower slopes (2,000 m) of the Andes. Migrants in a lowland site were documented more often in early successional habitats than in primary forest, and no species were confined to mature forest habitats. The number of species and relative abundance of migrants in primary forest, however, increased with elevation up to about 1,200 m and decreased above that elevation. Several species (Contopus borealis, Dendroica cerulea and Wilsonia canadensis) were largely confined to primary forest in the 1,000-2,000 m elevational zones in both Peru and Ecuador. Migrants on the western slope of the Ecuadorean Andes included several species that primarily winter further north. In general, the species richness of migrants and residents was inversely correlated, both on a biogeographical and a local scale. Migratory birds are most likely to be adversely affected by deforestation of the lower slopes of the Andes, which is proceeding at a rapid pace. The impact of human alterations of Amazonian forests will be greater on resident than on migratory birds. The loss of mid-successional lowland forests, however, might have a negative effect on several species.

Based on an extensive literature review for the 81 species of Falconiformes (excluding vultures) that breed primarily in Central and South America, I summarize the current state of our knowledge of the biology and conservation status of these birds. Most of what is known about the diurnal raptors in the region is based on studies carried out in the southern U.S.A., the Petén of Guatemala, extreme north-eastern South America (Venezuela through the Guianas) and Chile. The least-known species are residents of primary forest, especially in the genera Accipiter, Leucoptemis and Micrastur. Nests remain undescribed for 19 species, and less than five nests have been described for an additional 12 species. No prey data are available for six species and only anecdotal data have been published for a further 25 species. Breeding behaviour is unknown for 27 species and known only anecdotally for an additional 18. The migratory habits of 28 species are poorly understood. Available data permit a reasonable assessment of the conservation status for 39 species and a well-educated guess for another 31, but any assessment for the remaining 11 would be speculation at best.

Basado en un estudio muy extensivo de la literatura para 81 especies de Falconiformes(excluyendo Cathartidae) que se aparean principalmente en Centro y Sur América, hago un resumen de nuestro conocimiento más actual de la biologia y el estado de conservación en que se encuentran estas aves. La mayor parte de la información que se tiene sobre raptores diurnos de la región está basada en estudios que fueron llevados acabo en el sur de los Estados Unidos, el Petén en Guatemala, la parte noreste de América del Sur (Venezuela hasta las Guianas) y Chile. Las especies menos conocidos son residentes de bosque primario, especialmente de los géneros Accipiter, Leucoptemis y Micrastur. Falta describir nidos para 19 especies y menos de cinco nidos han sido descritos para 12 especies adicionales. No hay informacion sobre la alimentatión para seis especies y solamente información anecdotal ha sido publicada para 25 especies adicionales. Comportamiento de apareo se desconoce para 27 especies y solamente es conocido por anécdotas par 18 especies adicionales, Las costumbres migratorias de 28 especies se conocen muy poco. Información disponible permite juzgar razonablemente el estado de conservación para 39 especies y se puede estimarlo para 31 más, pero el juzgar de los 11 que restan seria solamente especulación.

Historical and anecdotal data indicate that out of 14 original species of resident diurnal raptors, six have become extinct and two more are considered endangered in the middle Cauca Valley, Colombia. While the decline of several species is related to a drastic reduction of original habitats such as forests and wetlands, three open-habitat species have also suffered population declines. Possible causes include the recent expansion of intensive agriculture (at the expense of cattle-ranching areas), agrochemical pollution, hunting, and factors such as body size, the proximity of the valley to the upper altitudinal limit of most of the affected species, and the relative isolation of the valley from the Caribbean region and the savannas east of the Andes. The need for monitoring projects s i stressed for the vertebrate fauna, including those species considered common in open and disturbed habitats.

Informaciones históricas y anecdóticas indican que de una fauna original de 14 especies de aves rapaces diurnas residentes en el valle medio del rio Cauca, Colombia, se han extinguido seis especies y dos mas se consideran amenazadas. Mientras que la disminucion de algunas especies se relaciona con la drástica reductión de hábitats originales tales como bosques y humedales, tres especies propias de hábitats abiertos también se han visto afectadas. Se identifican como probables factores la expansión de la agricultura intensiva (a expensas de las areas previamente dedicadas a la ganaderia), la contaminatión por agroquimicos, los conflictos con los intereses humanos y ciertos factores como el tamano corporal, la proximidad del valle medio del Cauca al limite superior de distributión altitudinal de estas especies y el relativo aislamiento del valle con respecto de la región del Caribe y de las sabanas al oriente de los Andes. Se relieva la necesidad de establecer programas de monitoreo de las faunas de vertebrados, inclufdas las especies consideradas como comunes en hábitats abiertos y perturbados.

The reproductive ecology and behaviour of the endangered Puerto Rican Nightjar Caprimulgus noctitherus was studied at the Guanica Forest, located in south-western Puerto Rico. From 1985 to 1987 a total of 23 nightjar nests were located. Nests were initiated between 24 February and 2 July. Nightjar males were mostly responsible for parental duties. An elaborate nest-relief ceremony was discovered for this species. Courtship and laying activities were most common during the new moon and last quarter phases. Hatching dates were centred 3-5 days around the first quarter and during full moon. A total of 87% of the nests were successful in producing at least one fledgling. Locomotory activity of radio-tagged nightjar males was higher during periods of twilight, and averaged 61.5 m/movement. Area of primary utilization (i.e. home range) averaged 5.2 ha/nightjar during the period of transmission. A large portion of the species's current range is privately owned forestland that could be subjected to changes (i.e. deforestation), rendering it useless for nightjars. The sedentary nature and nesting habits of this species make it particularly sensitive to the physical and ecological alteration of mature dry forest. Conservation of existing reserves, reforestation of disturbed areas with native and selected plantation species, and acquisition of privately owned tracts of mature dry limestone forest will help to insure the long-term survival of the species throughout its range.

La ecologia reproductiva del Guabairo Pequeno de Puerto Rico Caprimulgus noctitherus, conocido comunmente como guabairo, se estudió en el bosque de Guánica, localizado en el suroeste de Puerto Rico. Del 1985-1987 se localizaron un total de 23 nidos. La temporada reproductiva se extendió desde el 24 de febrero hasta el 2 de julio. El macho de la especie es mayormente responsable por el cuidado parental. Se descubrió una ceremonia altamente elaborada para el relevo en el nido. La actividad de cortejeo y el comienzo de los nidos era mas común durante las fases de luna nueva y cuarto menguante. Las fechas de eclosion se encontraban de 3–5 dfas alrededor del cuarto creciente y la luna llena. Un 87% de los nidos fueron exitosos en producir por lo menos un volantón. La actividad locomotora de machos de guabairo equipados con radiotransmisores fue más alta durante el perfodo crepuscular y promedió 61.5 m/ movimiento. El área de utilizatión primaria promedio fue de 5.2 ha/guabairo durante el perfodo de transmisión. Una parte considerable de la distributión del guabairo aún se encuentra en áreas bajo propiedad privada, podiendo sufrir cambios dramáticos en la compositión del paisaje (i.e. desmonte). La naturaleza sedentaria de esta especie y sus hábitos reproductivos la hace particularmente vulnerable a la alteratión fisica y ecológica del bosque seco. La conservatión de las unidades de conservatión existentes, la reforestatión con especies de árboles nativos y algunas especies de plantatión, y la adquisición de áreas privadas de bosque seco maduro ayudará a asegurar la sobrevivencia a largo plazo de la especie a través de su distribution geográfica.

The Bearded Tachuri Polystictus pectoralis occupies lowland grasslands with scrubby vegetation, generally near water, in the Andean grasslands of Colombia at two sites (threatened race bogotensis), savannas in eastern Colombia and the lowland and tepui grasslands of mainly southern Venezuela, Guyana, Surinam, French Guiana and northern Brazil (race brevipennis), reappearing south of the Amazon in central-southern Brazil, eastern Bolivia (no recent records), Paraguay, Uruguay and northern and central-eastern Argentina (nominate pectoralis). It is an austral summer visitor (October/November to February/April) to central-east Argentina, nesting (commonly in thistles) around December, clutch-size three. It feeds on insects by perch-gleaning, sallying, hover-gleaning and still-hunting. It is unobtrusive and must be commonly overlooked, and in some localities may be moderately well represented. Overall, however, it is scarce and appears to be very patchy in occurrence; grassland habitats within its range have been converted wholesale to farming. New quantitative criteria support earlier qualitative judgement that the species is probably not (yet) threatened, but that it merits near-threatened status. Suggestions that one or all of its three subspecies may be good species are premature; it is not even clear how distinct these forms are as subspecies.

Habitat preferences of the Rufous-fronted Antthrush Formicarius rufifrons were studied along the Rio Tambopata in south-eastern Peru. All territories of this threatened species, endemic to lowland forests in the department of Madre de Dios, Peru, contained the near-juxtaposition of two distinct vegetation types: tall floodplain forest with shaded understorey, and more open, second-growth vegetation with a very dense layer of vegetation close to the ground. The latter habitat type was closely associated with disturbances, such as river erosion, tree falls, or human clearings. Two measures of antthrush density were calculated for the study area. A satellite photograph was used to identify appropriate habitat along the upper Rio Tambopata, and using the antthrush density figures, a population size of 100-500 pairs was estimated for this area. This estimate was then used to calculate a global population of 700-3,500 pairs. Approximately 207c of the global population occurs within currently protected areas. The habitat and diet of this species are discussed in relation to two othercongeners.

Las preferencias de hábitat de Formicarius rufifrons fueron estudiadas a lo largo del Tambopata en el sureste del Perú. Los territorios de esta amenazada especie, endémica de los bosques tropicales en el depto. de Madre de Dios, están compuestos por dos formaciones vegetales distintas: bosque alto de llanura y vegetación secundaria con un estrato inferior tupido. La vegetación secundaria se debe a los trastornos ecológicos, tales como la erosión fluvial, las caidas de árboles, y también a la tala de bosques. Los limites inferior y superior de densidad (tasa de población) fueron calculados en el área de estudio; los habitats apropriados se identificaron a lo largo del rio a través de imágenes de satélite y se pudo calcular un total de 100-500 parejas para la zona. Se empleó la tasa poblacional, entonces, para calcular la supuesta población global de 700-3,500 parejas. Hoy en dia, aproximadamente 20% de la población global de la especie ocurre dentro de áreas protegidas. Se trata, ademas, el hábitat y la dieta de F. rufifrons en relatión a dos especies congenéricas.

The Red-tailed Amazon Amazona brasiliensis was found to be restricted to a complex mosaic of forests growing on the narrow coastal plain of eastern Brazil. The species depends on habitat heterogeneity for both food and breeding. In Sào Paulo state the 1,550 individuals are divided into 16 populations. The global total of the species may be around 3,600 birds. They feed mainly on fruits, flowers and nectar, also occasionally insects. Most nests are found in permanently flooded forest, apparently because of greater cavity availability. Poaching has had a great impact and is the most immediate threat to the species.

O Papagaio de Cara-Roxa Amazona brasiliensis é restrito a um complexo mosaico de florestas na estreita planicie costeirano leste do Brasil. Os papagaios dependem da heterogeneidade do ambiente tanto para sua alimentaçāo como reproducao. No estado de São Paulo existem 1,550 papagaios, divididos em 16 grupos. A populaçāo total da espécie pode ser de 3,600 individuos. Os papagaios se alimentam principalmente de frutos, flores e néctar, com alguns casos de insetivoria. A maioria dos ninhos foi encontrada em florestas permanentemente inundadas (caxetais), aparentemente devido a maior disponibilidade de cavidades. A captura para o comércio ilegal tern tido um grande impacto sobre a espécie e é a maior ameaça à sua sobrevivéncia.

We review the distribution and habitat of four Myrmotherula antwrens endemic to the Atlantic Forest of eastern Brazil, and present data to correct and update the information available in the literature. Based primarily on field experience over the past several years, we show that these birds are highly threatened despite having beenrecorded in several reserves. The chief issue affecting the continuedsurvival of these Myrmotherulas is that very little lowland (i.e. below 300 m) forest is included in the reserve system, and almostnone exists outside the protected areas. Those lowland forests that do exist, both within and outside of reserves, are invariably under greater human pressure than the much more extensive forests of the adjacent uplands, and are disappearing at a much greater rate. With the aim of protecting not only these Myrmotherulas but also the entire lowland fauna of the Atlantic Forest, we recommend that funding for a significant and monitored increase in the guard system be made available to help ensure the integrity of the existing forest reservesinto the future. As probably the single most important lowland forestpatch in Espirito Santo/Rio de Janeiro that does not have official reserve status, we highly recommend that negotiation to establish “Fazenda Uniāo” in northern Rio de Janeiro as a carefully maintained protection area be initiated immediately, before thepossibility to preserve it passes.

Apresenta-se uma resenha da distribuição e habitat de quatro espécies endêmicas do gênero Myrmotherula da Mata Atlântica e também dados para corrigir e atualizar a informaçao disponivel em literatura. Baseando-se principalmente na experiência de campo dos autores durante os ultimos anos, evidências são fornecidas para demonstrar que a sobrevivência destas espécies está claramente ameaaçcada a despeito da ocorrência destas em diversas unidades de conservaaçcão existentes ao longo de suasdistribuiaçcōes. O dado mais relevante que interfere diretamente na continuada sobrevivência destas formas é quemuito pouco das florestas baixo-montanas estao inseridas no sistema de unidades de conservaaçcão ao longo da Serra do Mar e mesmo quase nada existe destes ambientes fora das reservas. Estas poraçcões de floresta nas baixadas ou em cotas inferiores a 300 m quando existem, fora ou dentro das unidades de conservaaçcão, estão invariavelmente sob maior pressão humana e assim desaparecendo em muito maior proporaçcāo do que as mais extensas florestas montanas adjacentes. Dentro de uma perspectiva conservacionista n???amacr;o somente destas quatro especies de Myrmotherula mas de toda fauna da Mata Atlântica de baixas altitudes recomenda-se o investimento monitorado a longo prazo de um expressivo sistema de guarda-parques para garantir a integridade das unidades de conservaaçcāo existentes permanentemente. Recomenda-se especialmente tambem a negociaaçcão imediata para o estabelecimento da Fazenda Uniao no norte do Estado de Rio de Janeiro como unidade de conservaaçc‘atilde;o para que esta area, provavelmente o mais importante trecho de floresta nao preservada legalmente dos estados Espirito Santo/Rio de Janeiro, desapareaçca antes da sua possibilidade de preservaaçcāo.