Department of Zoology North Carolina State University Campus Box 7617 Raleigh, NC 27695-7617 Introduction The American Woodcock [Scolopax minor] is a migratory bird that breeds in eastern North America and winters in the southeastern United States (Sheldon 1971}. Begin- ning in 1975, studies surveyed the sex and age structure, and determined the migratory habits, of North Carolina's wintering population of woodcock (Stamps and Doerr 1976, Connors and Doerr 1982). Connors and Doerr (1982) found (1) that juvenile males made up the largest percen- tage of the winter population and (2} that these winter- ing birds came from numerous places in the Ariantic flyway throughout the northeastern U.S. and southeastern Canada. This paper expands on the results of earlier studies which reported results from 1975 to 1978. We report additional information on the population structure and migratory habits of birds banded in Hyde County, North Carolina, during the period 1975-1982. The changes in the age and sex structure of a wintering population over time and how this population structure differs from that revealed by South Carolina data are discussed. We also present new information on the migrational habits of woodcock winter- ing in North Carolina. Finally, this paper presents infor- Present address: The Pennsylvania State University, DuBois Campus, College Place, DuBois, PA 15801. mation, based on band recoveries and radiotelemetry data, showing daily and annual site fidelity demonstrated by this wintering population. A cautionary note on the radio- transmitter harness is also included. Methods Woodcock detected by eyeshine in soybean fields at night using a 6-volt headlamp were captured with long- handled nets (Glasgow 1958). Fields were systematically searched by a pair of workers moving perpendicular to the cultivation rows. The sex and age of each captured woodcock were determined by wing plumage characteristics (Martin 1964) and its weight and field loca- tion were also recorded. Each bird was banded with a U.S. Fish and Wildlife Service band and released. Woodcock weights were analyzed as an index body con- dition using Analysis of Variance and Duncan's Multiple Range procedures. Sex and age structures of the popula- tion between years were analyzed using a chi-square con- tingency table. All analyses were conducted at the 0.05 level of significance using the Statistical Analysis System (Helwig and Council 1979). Five woodcock were outfitted with radio-transmitters to determine the distance woodcock traveled when moving between nocturnal and diurnal habitat and to determine the daily regularity of these movements on the study area. The location of each bird was determined before and after crepuscular periods each day and at midpoints between these periods. Woodcock locations were plotted on large- scale aerial photographs and later surveyed on the ground. Woodcock weights were analyzed as an index body con- dition using Analysis of Variance and Duncan's Multiple Range procedures. Sex and age structures of the popula- tion between years were analyzed using a chi-square con- tingency table. All analyses were conducted at the 0.05 level of significance using the Statistical Analysis System (Helwig and Council 1979). Results and Discussion During 39 banding attempts covering 5 seasons {December through March, 1978-83) 577 woodcock were captured on the study area near New Holland, North Carolina {Table 1). Of these, 476 were banded and releas- ed. In previous studies {1975-78) 1184 birds had been banded {Stamps and Doerr 1976, Connors and Doerr 1982); therefore, 1660 woodcock were banded and releas- ed on the study area from 1975 to 1983. Age and Sex Composition The age and sex composition of populations of wood- cock captured during the years 1978-83 remained relatively constant from year to year. Juvenile males were the predominant group in the wintering population on the study area {Table 1). The large percentage of juvenile males {42%) is consistent with reports from other studies conducted on migration routes in New Jersey (Rief- fenberger and Ferrigno 1970) and on wintering areas in Louisiana {Martin et al. 1969, Britt 1971) and North Carolina {Connors and Doerr 1982). We found approx- imately equal numbers of adult males, adult females, and juvenile females {Table 1). Table 1. Age and sex distribution of woodcock cap- tured near New Holland, Hyde County, North Carolina. Adult Juvenile Adult Juvenile Total Season Females Females Males Males (%) 1978-79 Number 9 8 17 15 49 Percent 18.4 16.3 34.7 30.6 (8.5) 1979-80 Number 34 68 41 115 258 Percent 13.2 26.4 15.9 44.6 (44.7) 1980-81 Number 39 42 57 92 230 Percent 17.0 18.3 24.8 40.0 (39.9) 1981-83 Number 4 8 6 22 40 Percent 10.0 20.0 15.0 55.0 (6.9) Total 86 126 121 244 577 (%) (15.0) (21.8) (21.0) (42.3) (100) In two South Carolina coastal studies, however, a predominance of female birds was found by Pace and Wood {1979) and Ingram and Wood {1983) (66% and 71%, respectively). This high percentage of females may reflect the fact that the South Carolina birds were from a diur- nal hunting sample, whereas the North Carolina birds were live trapped at night in agricultural fields. Dif- ferences in collection methods, habitats, and time of col- lection could account for these variations in sex ratios. Sex- ually differential migrational rates could also explain the differences between the North and South Carolina studies. Hale woodcock may winter farther north than females in order to be closer to the breeding grounds when migra- tion occurs, thus enabling them to arrive on the breeding grounds earlier than females. Hales could then compete for singing grounds and establish territories so that when the females arrive, breeding can procede. This phenomenon has been observed in many species of passerinc in North America and Europe, such as the Red- winged Blackbird [Agelaius phoeniceus}, Ortolan Bunting [Emberiza hortulanaJ {WeRy 1975 I, and Lapland Longspur {Calcarius lapponicus} {West et al. 1968 I. Band Recoveries Returns of 17 bands during the period 1975-78 were reported by Connors and Doerr {1982 I. Since 1978, 14 more returns have been received for a total of 31 of 1660 birds banded and a return rate of 1.9%. This low rate of recovery is somewhat disappointing but can be partially explained by the ages of the woodcock banded in North Carolina. The age distribution of birds banded and released on the North Carolina study area was approximately 64% juvenile and 36% adult {Table 1). However, our band recovery data indicated that the most band recoveries {82%) were from birds banded as adults. Since only 36% of all birds banded on the North Carolina study area were adults, a high percentage of adult recoveries would explain the low overall recovery rate {1.9%) we experienced. Without additional data we can- not determine reasons for the low recovery rate for birds banded as juveniles. However, the most likely explana- tions would seem to be low survival of North Carolina juvenile birds moving into northern breeding areas or dilu- tion of North Carolina juvenile birds after reaching north- ern breeding areas with a large number of adults from other southern states. The band recoveries indicated, as other authors have sug- gested {Krohn and Clark 1977, Wishart 1977, Coon et al. 1978, Connors and Doerr 1982), that birds on our study area generally use the Atlantic flyway {Fig. 1). It appears that the study area along the North Carolina coast is an important wintering area for breeding birds from the northeastern United States. Other locations along the southeastern coast of the United States may also serve as primary wintering areas. Figure 1. Number and distribution of band recoveries in North America from wood- cock banded near New Holland, North Carolina QB 1 / / /  I / /  / PA / / / '/ / / / t/ qew Holland, NC qS Connors and Doerr (1982) received no returns from points farther south than North Carolina. In the current study one band was returned from Louisiana (Fig. 1), indicating some southward movement of birds banded in North Carolina. This bird, banded in the Piedmont region of North Carolina one year before it was shot, could have migrated north for one breeding season then returned to a different wintering area in Louisiana using the Atlantic or Mississippi flyway. Recaptures Data from birds recaptured in the same year they were originally banded suggest that they tended to stay in the same area where they were first captured (Table 2). In this study, encompassing 7 fields, 90% of same-season recap- tures occurred in the same field where the bird was originally banded. This rate was significantly higher than those found in the earlier study (60-84%, Connors and Doerr 1982). Daily Movements Since the number of recaptures was low, additional in- formation on site fidelity was achieved using radiotelemetry. The radio package allowed monitoring of daily movements. We fitted 5 woodcock with radio packages: of these, 2 were never contacted again, 2 died, and 1 provided data. This woodcock was monitored for 21 radio-days. Each day was spent within a 1.25-ha area in typical diurnal woodcock habitat: wet, thickly vegetated woodlands ISheldon 1971). Each night was spent within a 4.0-ha area in the same cutover soybean field-- the closest field to the diurnal habitat. The average distance the bird moved between nocturnal and diurnal habitat was 0.5 km, with a range of 0.075 to 0.75 km. While conducting the radiotelemetry portion of this study we encountered a problem with the harness system that should be documented. The system designed by Godfrey {1970) consisted of a surgical latex rubber "backpack" harness. In 3 of 5 cases, the latex straps split after about 10 days. This happened to 3 of our 5 telemetered birds. Usually, only one side of the harness broke, leaving the transmitter package attached to only one wing. This poor distribution of weight prevented flight. The basic design as described by Godfrey is sound, but a more durable lightweight material than surgical rubber should be used for harness construction. Woodcock Weights For all ages and both sexes combined, there was no significant year-to-year variation in woodcock weights ITable 3). The average weight was lowest during the 1978-79 season (166.7) and highest during the following season 1180.4). When woodcock weights were compared by year, sex, and age, however, females were always significantly heavier than males. Pettingill (1939) first documented that female woodcock are heavier than males. Weights of juvenile and adult female woodcock did not change significantly from one season to the next (Table 3). Among male birds, only juveniles differed significantly in weight between years. Juvenile males were significantly heavier in the 1979-80 season than in the 1980-81 season. In summary, this study showed that for woodcock cap- tured in the North Carolina coastal plain, juvenile males made up the largest percentage of the wintering popula- tion. Birds banded in North Carolina used the Atlantic Flyway to move from the breeding grounds in north- eastern North America to wintering grounds in the southeastern United States. Woodcock wintering in North Carolina showed strong intraseason site fidelity. Table 2. Recaptures of woodcock in fields near New Holland, Hyde County, North Carolina. Original Same band field <0.5 km >0.5 km location Total undeter- mined Same season 10 1 0 0 11 Next season 1 1 0 3 5 Total 11 2 0 3 16 Table 3. Woodcock weights by year, sex, and age. Adult Juvenile Adult Juvenile Season Females Females Males Males 1978-79' 193.3+8.7' 191.8+3.6 154.6+2.0 157.5+7.3 (6) 2 (6) (15) (I4) 1979-80 204.9+3.6 203.9+2.3 164.1 +2.7 165.5+ 1.2 (34) (68) (41) (115) 1980-81 203.1 +3.3 201.1 +3.7 166.7+3.2 155.9+2.2 (32) (28) (47) (70) 1981-832 197.3+10.3 202.3+4.4 162.0+3.6 162.3+2.5 (4) (8) (5) (22) 'Mean weights (g) + standard error. 2Sample size. Acknowledgments This study was supported by the North Carolina Agricultural Founation, by the Accelerated Research Pro- gram of the U.S. Fish and Wildlife Service and the North Carolina Wildlife Resources Commission, by the National Wildlife Federation, and by the North Carolina Agricultural Research Service at North Carolina-State University. Particular thanks are due to John Connors, Tim Stamps, and personnel the of North Carolina Wildlife Resources Commission for their field assistance and to the land owners in Hyde County, North Carolina, who allowed us to use their land for banding activities. Special thanks are extended to Charlie Carawan and family for their hospitality. This paper is number 9157 of the ]our- hal Series of the North Carolina Agricultural Research Ser- vice, Raleigh, NC 27695-7617. Literature Cited Britt, T.L. 1971. Studies of woodcock on the Louisiana wintering ground. M.S. thesis. Louisiana State Univer- sity, Baton Rouge. 105pp. Connors, J.L., and ED. Doerr. 1982. Woodcock use of agricultural fields in coastal North Carolina. U.S. Fish and Wildl. Serv., Wildl. Res. Rep. No. 14: 139-147. Coon, R.A., T.J. Dwyer, and J.W. Artmann. 1978. Identifica- tion of potential harvest units in the United States for American woodcock. Proc. Sixth Woodcock Symp., Fredericton, N.B. 6:147-154. Glasgow, L.L. 1958. Contributions to the knowledge of the ecology of the American woodcock, Philohela minor, on the wintering range in Louisiana. Ph.D. thesis. Texas A.&M., College Station. 153pp. Godfrey, G. A. 1970. A transmitter harness for small birds. Inland Bird-Banding News. 42(1):3-5. Helwig, J.T., and K. Council. 1979. SAS users guide. SAS Institute Inc., Cary, N.C. 494pp. Ingram, R.P., and G.W. Wood. 1983. Characteristics of woodcock harvest data in coastal South Carolina. Wildl. Soc. Bull. 11:356-359. Krohn, W.B., and E.R. Clark. 1977. Band-recovery distribu- tion of eastern Maine woodcock. Wildl. Soc. Bull. 5(3):118-122. Martin, F.W. 1964. Woodcock age and sex determination from wings. J. Wildl. Manage. 28(2):287-293. Martin, F.W., S.O. Williams, III, J.D. Newsom, and L.L. Glasgow. 1969. Analysis of records of Louisiana-banded woodcock. Proc. Annu. Conf. Southeast. Assoc. Game and Fish Comm. 23:85-96. Merovka, L.J. 1939. The woodcock in Louisiana. Louisiana Conser. Review. 8(4):10-14. Pace, R.M., III, and G.W. Wood. 1979. Observations of woodcock wintering in coastal South Carolina. Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agen- cies. 33:72-80. Pettingill, O.S. 1939. The American woodcock, Philohela minor (Gmelin). Memoirs Boston Soc. Nat. Hist. 9(2):173-391. Pullen, S.B. 1973. A biotelemetric study and observations of wintering American woodcock, Philohela minor (Gmelin} in east central Alabama. M.S. thesis. Auburn U., Auburn, Alabama. 55pp. Rieffenberger, J.C., and F. Ferrigno. 1970. Woodcock ban- ding on the Cape May Penninsula, New Jersey. Bird- Banding. 41(1}:1-10. Sheldon, W.G. 1971. The book of the American woodcock. U. Mass. Press, Amherst. 227pp. Stamps, R.T., and ED. Doeft. 1976. Woodcock on North Carolina wintering grounds. Proc. Annu. Conf. Southeast. Assoc. Game and Fish Comm. 30:392-399. Welty, J.C. 1975. The life of birds. W.B. Saunders, Co., Philadelphia, PA. 623pp. West, G.C., L.J. Peyton, and L. Irving. 1968. Analysis of spring migration of Lapland Longspurs to Alaska. Auk 85:639-653. Wishart, R.A. 1977. Some features of breeding and migra- tion of woodcock in southwestern Quebec. Bird- Banding. 48(4):337-340. (Eastern) NOW AVAILABLE An index to Volumes 1-5 North American Bird Bander. To order send $2.00 to Western Bird Banding Association, 4.100 N. Romero, //108, 3acson, AZ 85705.