The role of grit selection and retention on mortality rates of captive mallards (Anas platyrhynchos) that had ingested lead shot was evaluated in a series of experiments. The calculated rate of grit retention was determined from the negative exponential relationship between the number of marked grit particles in the gizzard and time. Grit consumption, the primary rate-determining step of grit retention, was significantly (P < 0.01) affected by diet, sex, season, mineral composition, and size of particles. Peak grit ingestion by males occurred in the fall, and in females in spring. Ducks maintained on a corn diet consumed twice as much grit as ducks maintained on commercial duck pellets. When offered grit particles between 0.42 and 4.75 mm in diameter, both sexes selected grit particles between 0.84 and 2.00 mm. In spring, females selected oyster shell and sandstone grits over granite, calcium, and quartzite, whereas males preferred sandstone to all other grit types. J. WILDL. MANAGE. 45(1):64-73 Avian species require grit for food digestion (McIlhenny 1932, Leopold 1933) and as a source of trace minerals (Sadler 1961, Harper and Labisky 1964, Kopishke 1966). Bellrose (1959) reported that the time of lead-shot retention was important in determining the degree of mortality suffered by a group of lead-ingesting individuals. Beer and Stanley (1965) postulated that lead and grit interacted, and when excess grit was ingested it was passed and in some instances took the lead shot with it. Bellrose (1975) stated that the size of the lead particle influenced its rate of passage. The kind of grit (calcium) has been shown to lower the mortality of experimentally dosed mallards (Godin 1967, Longcore et al. 1974). The objectives of this study were to identify the factors affecting grit selection and retention in mallards and the role of grit in shot ingestion and retention. I am indebted to the Max McGraw Wildlife Foundation and the College of Agriculture and Life Sciences, University of Wisconsin, Madison, for financial support. The assistance of the foundation staff and G. V. Burger's help with the manuscript are appreciated. I also thank the faculty and graduate students of the Department of Wildlife Ecology, University of Wisconsin, Madison, for their help and suggestions. In particular I acknowledge the guidance and support of R. A. McCabe. METHODS AND MATERIALS Ten experiments were conducted, 9 of which took place within specially constructed pens designed to prevent access to grit. The pens were 2.4 x 2.4 x 1.5 m, with a floor of woven hardware cloth (1-cm mesh) raised 0.6 m above the ground. Six pens were constructed within 1 large outdoor pen. Sheets of corrugated fiberglass were placed 5 cm beneath the floor and inclined toward the front of each pen so that runoff drained into a collection trough. All pens were equipped with a plastic pool 38 cm deep and 75 cm in diameter. A grit feeder (Sadler 1961) was provided inside each pen. A preliminary study was conducted to obtain variance estimates for the computation of sample sizes and types of marker grit. A sample of 10 ducks/experime tal group was determined to be adequate. The best marker grit was green quartzite. Four grit-particle sizes were used: 10 (2.01-4.76 mm), 20 (0.85-2.00 64 J. Wildl. Manage. 45(1):1981 This content downloaded from 157.55.39.56 on Tue, 18 Oct 2016 04:35:10 UTC All use subject to http://about.jstor.org/terms GRIT SELECTION IN MALLARDS * Trost 65 mm), 30 (0.60-0.84 mm), and 40 (0.43-0.59 mm); size was measured as the largest diameter of the smallest cross-section of each particle. Ducks were banded and randomly assigned to treatment groups. At 48-hour intervals, measurements were made of grit and feed consumed by each experimental group, and fecal material was collected. The pens were then washed to prevent reingestion of any marker grit remaining. Fecal materials were washed into large metal containers and agitated to liquefy the fecal pellets. The effluent was poured off and the sediment dried overnight at 120 C. The dried sediment was then crushed lightly using a mortar and pestle and rewashed. After a final drying the grit was sorted and weighed. During the preliminary experiment, a minimum acclimatization period of 2 weeks was necessary to attain a steady consumption of food and grit. In all subsequent experiments, ducks were acclimated to their pens for 2 weeks prior to