Mortality rates of bird nests near the Brazilian equatorial localities of Belem and Manaus were between 1.5 and 4.5 percent per day for different types of nests and habitats, except for 6.8 percent per day in cup-shaped nests in zones at Belem. Only the last percentage is as high as rates commonly found by earlier workers in neotropical areas further from the equator. As in previous studies, cup nests were less successful than enclosed nests, but, in contrast, nests were as successful as or more successful than open-area nests. Further data are needed to verify the ideas that nest mortality rates near the equator and in tropical forests are higher than in areas, subequatorial zones, or temperate zones. Earlier work was mostly in isolated habitats in which large predators no longer occurred, and thus could not limit nest-robbing small predators; such habitats could be abnormal, as are habitats near houses. NEST SUCCESS IN BIRDS is generally said to be lower in the tropics than in temperate zones (Ricklefs 1969; Skutch 1966; Snow and Snow 1964; and Willis 1961, 1974). These examples of low success have been used to explain why tropical birds have low clutch sizes (Skutch 1967, Foster 1974), and even formed part of the basis for general ecological theories that cycling is faster in the tropics (Connell and Orias 1964). However, the cited studies of nest success are from Central America and Trinidad, from 9? to 170 north of the equator. Studies of Marchant (1960, 1961) in dry regions of western Ecuador indicate moderate nest mortalities; but these data are not comparable with studies from forested habitats. Fogden (1972) found high nest mortalities in equatorial forests of Sarawak, but Sarawak is on an island of lowered species diversity more comparable to subequatorial Trinidad than to the Amazon. There are many studies of bird nests in the equatorial forests of the Amazon (Carvalho 1957a, b, c, d, 1958a, b, c, 1960a, b, 1962, 1963; Novaes and Carvalho 1957; Oniki 1970; Snethlage 1935a, b), but studies of nest success are limited to scattered examples in these studies. Snethlage, without giving quantitative data, suggests that nest success is fairly high in the lower Amazon. Koepcke (1972) indicated that birds in eastern Peru have many adaptations against predation, but did not present data on nest success. The question is, do equatorial birds have low nest success despite their adaptations against predators? Studies of adaptations of nests and eggs against predation will be considered in a separate paper. Here data are presented on mortality rates for two equatorial Brazilian areas. STUDY AREAS At Belem (1? 28' S, 48? 27' W, altitude 13 m) nests were studied mainly in the Area de Pesquisas Ecolgicas do Guama (APEG) of the Instituto de Pesquisas Agropecuarias do ,Norte (IPEAN). The most characteristic habitats of APEG are upland forest, second growth, igapo or swamp, and v'arzea or tidally flooded (Pires, Dobzhansky and Black 1953; Novaes 1970; Oniki 1972). These types of forests form an irregular patchwork at the eastern end df IPEAN. Other areas studied at IPEAN included several types of man-modified habitats, including roadsides and yards of houses. In the present study, two areas were especially important for nests: Serraria, an area of experimental plantations of rubber trees, where there were half-burned logs overgrown with the leguminous vine Pueraria phaseoloides (Roxb.) Benth.; and Pupunhal, a plantation of spiny pupunha palms, Guilielma gasipaes ( H. B. K.) Bailey. At Manaus, nests were studied in two areas: in the Reserva Florestal Ducke, of the Instituto Nacional de Pesquisas da Amazonia (INPA) and at Sitio I-iru,anga, across the highway from the Reserva entrance at 2? 55' S, 590 59' W. Reserva Ducke (60-120 m elevation) is an extensive upland bordered by experimental plantations of native and introduced trees, mostly in the shade of tall trees of the original forest. Laboratories are in a clearing next to a small creek, the Igarape do Barro Branco. Sitio I-iruqanga is a private weekend villa, with many fruit trees and palms, bordered on two sides by forest, and set beside the paved Manaus-Itacoatiara road at km 26.3 on the Igarape Mariano. Some nests were studied in mancreated savannas and in plantations or orchards near km 26. Nests in forest, plus a few nests at the 60 BIOTROPICA 11(1): 60-69 1979 This content downloaded from 40.77.167.36 on Fri, 21 Oct 2016 04:37:10 UTC All use subject to http://about.jstor.org/terms edge and in second growth, are considered forest nests; others are nests of open habitats.