.- Apart from a few anecdotal reports (Bent, Life Histories of North American Wildlife, U.S. Natl. Mus. Bull. 130, 1925; Cleare, Birds, The Argosy Co., Georgetown, Guyana, 1938; Haverschmidt, Field Notes on the Black-bellied Tree Duck in Dutch Guyana, Wilson Bull. 59:209, 1947; Giglioli, Crop Histories and Field Investigations, 1951-1957, Br. Guiana Rice Development Co., Georgetown, Guyana, 1959; Haverschmidt, Birds of Surinam, Livingston Publishing Co., Wynnewood, Pennsylvania, 1968) and two studies in southern Texas (Bolen and Forsyth, Foods of the Black-bellied Tree Duck in south Texas, Wilson Bull. 79:43--49, 1967; Bolen and Beecham, Notes on the foods of juvenile Black-bellied Tree Ducks, Wilson Bull. 82:325-326, 1970), the food habits of Black-bellied Whistling Ducks (Dendrocygna autum- nalis) have been little studied. In 1973, I initiated an ecological study of Black-bellied Whis- tling Ducks to evaluate the magnitude of their foraging activity in ricefields. Preliminary results (Bourne and Osborne, Black-bellied Whistling Duck utilization of a rice culture hab- itat, Intercencia 3:152-159, 1978) indicate that overall depredation levels are low even though the ducks ingest newly sown, pregerminated paddy or seed rice (Oryza sativa). The purpose of this paper is to present data on the food habits of Black-bellied Whistling Ducks in Guyana, South America, when they occupied rice culture habitats during crop sowing in June 1973 and October-December 1974. Materials and methods.--I conducted fieldwork in Burma (6ø28'N, 57ø45'W) at the Mahai- cony and Abary Rice Development Scheme (MARDS). Detailed descriptions of the study area and its flora and fauna are available in Giglioli (1959) and Osborne and Bourne (Breeding behavior and food habits of the Wattled Jacana, Condor 79:98-105, 1977). In 1973, two methods were used for obtaining specimens: 15 ducks were shot between 05:00 and 07:16 with the aid of playback vocalizations and 15 were mist-netted between 20:00 and 20:55. Adults and juveniles collected in 1974 were shot between 05:00 and 07:16 with the aid of playback vocalizations. Two ducklings were hand caught in a fallow field on 8 August 1974, at 08:10 and 08:25. Specimens were dissected within 30 rain and the entire 552 THE WILSON BULLETIN ß Vol. 93, No. 4, December 1981 alimentary tract of each duck was placed in a separate large-mouth bottle containing 10% formalin. The contents of the esophagus, proventriculus and ventriculus were identified to the lowest taxon possible, and their volumes were determined by water displacement. Results 1973.--Food items identified from 30 adult Black-bellied Whistling Ducks (15 males, 15 females) indicate that plant material accounted for 97% of the food consumed (Table 1). Most (86%) of this plant food consisted of pregerminated paddy. Other plant foods accounted for 11% of the ducks' diet while animal foods made up only 3% of the birds' fare (Table 1). Most of the animals consumed were invertebrates; the majority were aquatic insects and snails (Table 1). Two ducks (7%) consumed young tadpoles of the marine toad (Bufo marinus). Of the 30 ducks examined for their food preferences, 63% ingested animal foods; 37% of these were males and 63% were females. 1974.--Plant foods accounted for 90% of the adults' and juveniles' diets (Table 1). Paddy dominated the adults' diet, constituting 74%, but was less important in the juveniles' diet, accounting for 15% of their food (Table 1). In 1974, adults consumed the same genera and species as the adults did in 1973 (Table 1), but juveniles had eaten the seeds of a grass, Paspalum sp., which accounted for 71% of their diet and occurred in 95% of the juveniles sampled (Table 1). Young apple snails (Pomacea sp.) were the most important animal foods consumed by juveniles as they accounted for 8.5% of their diet, and occurred in 70% of the sample (Table 1). Two ducklings analyzed for their food preferences consumed 54% animal foods and 46% plant foods. The plant foods consisted of the seeds of a millet, Echinochloa sp. and Paspalum sp.; they accounted for 31% and 15% of the diet, respectively. Shorefly (Scatella stagnails) larvae and pupae were found in trace amounts, but the bulk of the animal food consisted of unidentified terrestrial spiders. Discussion.--Black-bellied Whistling Ducks, like other dendrocygnids, are basically her- bivorous (Johnsgard, Waterfowl of North America, Indiana Univ. Press, Bloomington, In- diana, 1975). But at sowing time, cultivated cereals dominate the plant food preferences in Black-bellied Whistling Ducks' diets. For example, corn (Zea mays) was the most important constituent in the species' diet in Mexico (Bent 1925), and Sorgum vulgare constituted 48% of the species' diet in south Texas (Bolen and Forsyth 1967). Paddy accounted for 86% and 74% of the adult Black-bellied Whistling Duck's diet in Guyana during the early rice growing season. Observations suggest that paddy would become less important in the duck's diet as the growing season progressed because fewer suitable water-planted ricefields would be available as foraging sites. This may explain why paddy accounted for only 15% of the juveniles' diet, since they were collected after the adults were from 22 July-9 August 1974, when planting was almost completed. Animal foods do not appear to be important in adult and juvenile Black-bellied Whistling Ducks' diets. Even though adults were in breeding condition (as evidenced by gonadal mea- surements, males, N = 18; left testes mean 24 x 12 mm [10 x 5-32 x 16 mm]; females, N = 17, largest follicle mean 40 mm [3-56 mm]), they only consumed 3% animal food in 1973 and 10% in 1974, while juveniles also ingested 10% animal food in 1974. However, these data could be biased downwards due to the faster digestion of soft-bodied invertebrates in the proventriculus and ventriculus. Ducklings consumed 54% animal foods in this study, suggesting that younger whistling ducks need the higher protein content found in animal foods for growth and development. Acknowledgments.---Financial support of this investigation was provided by the Frank M. Chapman Memorial Fund, the Rob and Bessie Welder Wildlife Foundation, and Mr. Fred- crick N. Stevens. I thank the following for contributing ideas and criticisms: David R. Osborne, W. Hardy Eshbaugh, Gary W. Barrett, Michael P. Farrell and Fay M. Edwards. I am particularly indebted to Charles P. Kennard, George Hughes and their staffs of the Guyana Rice Board for providing facilities at MARDS. Finally, I am thankful to Dick L. GENERAL NOTES 553 554 THE WILSON BULLETIN ß Vol. 93, No. 4, December 1981 Deonier for identifying insects, to my wife Carol, Patrick Dougal and the Green brothers for assistance in the field. Data are taken from a M.Sc. thesis submitted to Miami University. This is Welder Contribution 195.--GODFREY R. BOURNE, Institute of Environmental Sci- ences, Miami Univ., Oxford, Ohio 45056. (Present address: School of Natural Resources, Univ. Michigan, Ann Arbor, Michigan 48109.) Accepted 3 Nov. 1980.