One hundred seventy-nine hand-reared mallards (Anas platyrhynchos) of two genetic strains were banded and released in two areas of Minnesota. One area usually had high and the other low natural productivity of waterfowl. Eighty birds were marked with miniature radio transmitters prior to release. Mortality, movements, and habitat use were determined by radio-tracking. Forty of the 56 radio-marked birds that were monitored ied within 21 days of release. Mink (Mustela vison) killed 21, avian predators killed 7, and unknown mammalian predation accounted for 9 deaths. Starvation poaching, and unknown predation were each responsible for one death. Twelve carcasses of banded birds were found. Mink predation was the cause of death in all these cases. In addition, 9 radio-marked and 16 banded birds were shot by hunters. There was no measurable difference in survival between genetic strains or sexes. Ducklings released in the area of high natural productivity had significantly ( P < 0.01 ) lower mortality than birds released in the area of low natural productivity. This lower mortality was attributed to differences in fertility and vegetation on the two areas. No significant difference (P < 0.05) in mortality between radio-marked and control birds was noted. Most mortality occurred within a 2-week period after release. Released birds gathered in large groups. Our findings are in close agreement with previous tudies and indicate low survival and a high vulnerability o the gurl. Lack of wariness, tendency of birds to gather in large groups, and releases in habitat of poor quality svere the main factors contributing to the low survival of the hand-reared mallards. Interest in artificial propagation and release of mallards to supplement low natural populations has increased in recent years because of declines in the continental mallard population. The present study was made to determine survival rates and causes of mortality of 6-week-old handreared mallards liberated in Minnesota. Bellrose and Chase (19SO:3) stated7 is evident that a large proportion of any game population will disappear each year from natural causes; it is a responsibility of management to see that the greatest possible use is made by man of the annual osses that normally occur. Little is known of the nature of these natural losses. Stout (1967) reviewed the nature and cause of mortality in wild fledged waterfowl; however, since his review was based almost entirely upon banding studies and obser1 Research supported by the Max McCraw Wildlife Foundation, Dundee, Illinois. 2 Present address: Department of Natural Resources, Madison, Minnesota. 1118 vations of wild birds, there are still many questions concerning the nature of nonhunting deaths of hand-reared waterfowl. During the 1930's, there was speculation as to what happens to hand-reared mallards once they are released in the wild. Lincoln (1934:313) suggested, <'Possibly, because of their semidomesticationa they are merely leading lives of indolence in the marshes7 refusing to migrate or to fly for the hunter; or else, untrained irl the rigors of natural environments, and being solely on their own resources, they may be unable to cope with living conditions and so succumb rapidly to the elements and to natural enemies.' Most analyses of liberated hand-reared mallards have been through banding returns. It is possible from these data to determine what happens to the birds in terms of gross movement, survival, and lnortality through the hunting season. However, banding returns do rlot reveal the types and extent of mortality occurring prior to HAND-REARED MALLARDS Schladweiter and Tester 1119 the first hunting season. The development of biotelemetry has given researehers the neans to study movements, with a miniInum of disturbanee and interferenee with the normal behavior patterns of an animal. With the use of telemetry, it is also possible to quiekly find remains of dead animals and thereby determine the eause of death. Both telemetry and banding data were used in this study. G. V. Burger and the staff of the Max MeCraw Wildlife Foundation offered eneouragement and suggestions during the entire projeet. Dr. Burger reviewed the manuseript and the Foundation provided the dueks for release and all finaneial support. The Minnesota Division of Game and Fish and the U.S. Bureau of Sport Fisheries and Wildlife assisted in planning and earrying out the projeet and provided some equipment. We are grateful to M. Nelson, R. Jessen, M. Wesloh, and H. Shepperd, Minnesota Division of Game and Fish; G. Brakhage, U.S. Bureau of Sport Fisheries and Wildlife; W. J. Kortesmake, Minnesota Future Farmers of Ameriea; nd to the many personnel of the Cedar Creek radio traeking projeet who assisted in both field and laboratory aspects. E. Hessler, D. B. Siniff, and R. J. Mackie offered editorial assistanee.