Adoption is a widespread phenomenon in birds and generally occurs at a low frequency within a species (reviewed in Rohwer 1986, Meek and Robertson 1991). Adoption of fledglings is common in seabirds due to the mobility of chicks soon after hatch (e.g. Hebert 1988, Morris et al. 1991, Pierotti 1991, Brown et al. 1995). In songbirds, studies of adoption have focussed on the behavior of replacement males during the nestling stage (Meek and Robertson 1991). Little is known about fledgling adoption in songbirds because the fledgling period is poorly described for many species (Smith 1978, Moreno 1984). We report two cases of fledgling adoption in the Hooded Warbler (Wilsonia citrina) observed in the course of an intensive study of fledgling care (Evans Ogden 1994). In passerines, the main hypotheses for the adaptive significance of adoption by males are: (1) increased opportunities for fathering offspring with the offspring's mother in the future (Power 1975, Rohwer 1986); and, (2) as a result of extrapair matings, the adopting bird may be the genetic parent of the young it adopts (Meek and Robertson 1991). The first hypothesis is supported by comparative evidence that adoption tends to occur in species where females are likely to renest with the adoptive male (Rohwer 1986). Recent reports that extrapair fertilizations occur at high frequency in many Temperate Zone passerines (e.g. Westneat 1987, 1993, Morton et al. 1990, Stutchbury et al. 1994) indicate that the extrapair-paternity hypothesis for adoption may be more important than previously recognized. We used DNA fingerprinting of Hooded Warblers to examine whether extrapair paternity played a role in adoption. Our study was conducted from May through August 1991-1993 in Crawford Co., Pennsylvania (41?N, 79?W). The study site is a 150-ha continuous hardwood forest that supported about 40 breeding pairs of Hooded Warblers each year. Adults were captured with mist nets, banded with U.S. Fish and Wildlife aluminum bands, and individually color banded. Each brood member that fledged also was banded with an aluminum band, and all nestlings within a brood were given color-band combinations different from those of other broods. For most adults and nestlings, we collected 30 to 100 uL of blood for use in parentage analysis. Multilocus DNA fingerprinting (with Jeffrey's probe 33.16) was used to determine actual paternity of the adopted fledgling (for detailed methods, see Stutchbury et al. 1994). Unrelated adults had a band-sharing coefficient of 0.301 ? 0.017 (n = 22), so young were considered unrelated to their social father if they had a band-sharing coefficient of less than 0.42 and had more than two novel bands when compared with their social parents. Territories of color-banded males were mapped by following singing individuals and noting border disputes. The social father was the male that defended the territory at the time of egg laying and fed the young at nests on the territory. The social mother was the female that incubated the eggs and fed the young. Each year, we systematically attempted to locate family groups after fledging to determine the period of fledgling care and food-delivery rates to fledglings. In Hooded Warblers we rarely saw both parents feeding a given fledgling. Instead, the brood was usually divided and each parent assumed full care of one or two fledglings (Evans Ogden 1994). When the social mother attempted a second brood, the male assumed full care of all fledglings from the first brood. In the forested habitat, it was difficult to see young fledglings and determine color-band combinations to confirm the nest from which they fledged. To identify whether or not a parent had adopted fledglings, we used only those cases where adults were seen feeding color-banded fledglings on more than two occasions. Using these criteria, we observed fledgling care by 17 parents from 13 different families. Two of these 17 parents (12%) were observed to adopt young (i.e. they fed young on their own territory that had fledged from a neighbor's nest). Below we describe the two case histories in detail. The cases were similar in that the social father of the fledglings was no longer present on his territory at the time of adoption, and the social mother had renested to attempt a second brood. The second case was perhaps unnatural, because the social father died accidentally during handling early in the nestling stage. In the first case, a fledgling was fed by a neighboring male on the adjacent territory, beginning when the fledgling was 28 days old postfledging. At this time, the adoptive male was also feeding two of his own young (nine days postfledgling) on his territory.