FALL MIGRATION of HERRING GULLS from KNIFE ISLAND, MINNESOTA by P

104] HOFSLUND,K,ni/e Island Herring Gulls l•ird-Bandin•April

WOODBURY,A.M., W. H. BEHLE, and J. W. SUGDEN. 1946. .Color-banding California Gulls at Great Salt Lake, Utah. Bull. Univ. Utah, Biol. Ser. 10: 1-15. WOODBURY,A.M. and HOWARDKNXCH?. 1951. Results of the Pacific Gull color-bandingproject. Condor53: 57-77. WOODCOCI%A. H. 1940. Herring Gull soaring. Auk 57: 219-224. • 1942. Soaring ,over the open sea. Scientific Monthly 55: 226-232. Biological Laboratories, Harvard University, Cambridge 38, Mass.

FALL MIGRATION OF HERRING GULLS FROM KNIFE ISLAND, MINNESOTA By P. B. HOFSLUND In June, 1950 the Duluth Bird Club began a banding study of a colonyof Herring Gulls (Larus argentatus).This studyhas continued eachsummer since, the summerof 1957 marking the eighthyear. The presentpaper dealswith the recoveriesof bandedgulls up to May 15, 1958.

120 - I[0 • • • c.• )00 90 McKNIGHT 6, McKNIGHT, BLOOMINGTON, ILLINOIS .Map 1. Recoveriesof Herring Gulls banded as nestlings on Knife Island, Minnesota. 37ol.1959 XXX HOFSLUND,Knife Island Herring Gulls []05

DRAINAGE NORTH AMERICA

120 I10 • • • c• •D0 McKNiGHT •* McKNIGHT, BLOOMINGTON, ILLINOIS Map 2. Recoveriesof first-yearHerring Gulls.

The breedingsite is a smallisland in Lake Superior,roughly 800 x 300 feet at its longestand widest spots. It lies less than a half mile from the mouth of Knife River in Lake County, Minnesota, and is known, at least locally, as Knife Island. Much of the island is covered with Mountain Ash (Sorbus americana) with a scatteringof Balsam Fir (Abies balsamea),Balsam Poplar (Populus balsami/era), and a tangle of undergrowth. The periphery, however,is largely bare rock, and it is here that most of the nestsare located. One notableexception was a nestplaced approximately 18 feet up in a BalsamFir. Censusreports (Hofslund, 1952:162) for the years 1948-'52 showed nest countsvarying between150 and 312. Though no actual counts have been made since 1952, it is likely that the island now supports in excessof 250 breeding pairs each year. In recent years the only other nestingbirds found on the island have been a pair of Red- breastedMergansers (Mergus serrator). Passerinebirds are con- spicuousby their absence. Banding operationshave been confinedto an afternoon,usually on the secondweekend in June. There has been an attemptto work 106] HOFS,.U•D,Knife Island Herring Gulls Bird-BandingApril

•c

DRAINAGE NOR'r'H AMERICA

• i•4 •i

co•m• BY

,• • 100 90 McKNIGHT•. McKNIGHT,BLOOMINGTON, ILLINO•

Map. 3. Recoveriesof second-yearHerrir•g Gulls.

as swiftly and efficientlyas possiblewith a general limit of 500 bandsso as to minimizethe disturbanceof the colony. Despitethis, someyoung birds have been killed by adults or have met death in other ways associatedwith the disturbance,and more than likely, most of the birds recoveredin June of the year they were banded never survivedthe day of banding. A total of 3,028 nestlinggulls have been bandedsince the project started. We have received81 reports of recoveriesor 2.67% of the total banded. While not a large number,the recoveriesreveal certain patternsin the movementsof the gulls after they leave Knife Island. Lincoln (1928:59) and Gross (1940:153) describedthe dispersal of gulls from their breedinggrounds as "explosive,"a conclusion well-foundedon the basisof recoveriesfrom bandingsites in Lake Michiganand at KentIsland. The presentstudy was started in partto determinewhether this explosivenature of the dispersalwas apparent only whereready access to largewaterways was available. The followingpoints are pertinentto the discussionof the results of the study: Vol. 1959XXX HOFSLUND,K,nije Island Herring Gulls [107

McKblIGHT•, McKblIGHT,BLOOMINGTON, ILUblOIS Map 4. Roeoveriesof adult Herring Gulls banded ,as nestlingson Knife Island. 1. The majority of the recoveriesof Knife Island Gulls come from the Great Lakes. Of the 81 recoveriesonly 12 have not been from areas intimately associatedwith the Great Lakes-St.Lawrence Water- way. 2. Only two of the 12 recordsare from areasnorth of this waterway, the rest are south. 3. Of the 10 southeilyrecords all but three are associatedwith the Atlantic, Gulf of Mexico, or Caribbean Sea coastlines. 4. The three records not associated with coastal waters are on or near major tributaries of the Mississippi River, and the dates of recovery (Table 1) suggestthat these birds were continuing their wanderingnorthward toward the breedinggrounds rather than having reachedthese points through a southerlydispersal over land. 5. If we now turn our attentionto the distributionmaps (Maps 2, 3, and 4), it is apparent that the wanderingtendency is lessenedas the gulls reach breedingage, a conclusionpreviously brought out by others (Gross,1940:136), and in the caseof Knife Island gulls their travels seemto be confinedlargely to the Great Lakes. 6. There is a lack of uniformity in distancestravelled by gulls of ]lOS] HOFSLUND,K,nije Island Herring Gulls Bird-BandingApril

IOO

2oo 3oo

40o 5oo

6oo

70o 8oo 9oo

I ooo •oo • ,•oo i

• 4o0 I$OO

1800 19oo

2ooo •,lOO 2 2o•

24oo

25oo 26oo

29oo

5 Ioo Figure 1. Scattergramshowing distancesfrom banding point to place of recoveryof 6rst-year birds. the sameage as is shownby the scattergrams(Figures 1, 2, and 3). An extreme examplewould be the two first-year birds recoveredin February,one of whichhad wanderedabout 450 mileswhile the other had coverednearly 2,950. 7. Gulls banded at Kent Island oi• the coast of New Brunswick (Gross, 1940:137) were largely recoveredalong the Atlantic coast. Only a very few were found away from coastalwaters. 8. Gullsbanded by Lyon an• •ilson (•ross, 1940:139) an• •hose •iscusse••y Lincoln (lP28) were •an•e• mostlyin Lake Michigan. Theseshowe• a •ispersalin all •irections. Vol. XXX 1959 HOFSLUND,Knife Island Herring Gulls [109

IOO

200

500 400

500

&o o

70 o

800 900

I ooo

IlOO

i zoo

1300 1400

1500

1600

I 7'00 I 8oo

1900

2000 2 I00 Figure 2. Scattergram showing distancesfrom banding point to place of recovery of second-yearbirds. 9. The report of the Pacific gull-bandingproject (Woodbury and Knight, 1951) showedalmost a strict adherenceof Pacific gulls to coastal waters. These points bring me to the conclusionthat the term "explosive dispersal" is not suitable to describethe fall migration of Herring Gulls, but rather shouldbe called a wanderingwhose directional ten- denciesare determinedby the availability of large waterways. The explosivequality is effectivelyblocked by the lack of shoreor coastlines. The tendency to move southward after the first of the year as suggestedby Lincoln (1928:56) and Gross (1940:153) also seems to me as not being an applicablegeneral assumption.From the distribution of the recordsalong the coastlinesit appearsthat the only major southwardmovement comes from gulls that have reached the coastline.Once the coastlinewas reachedthe availability of food and the major water traffic would exert a positive force in a southerly direction.Major shippinglines might havesome effect on this and also might determineto some extent the distancesthat individual gulls might travel. The concentrationof recordsalong the southernshores of the Great Lakes indicatesthat there are factors favoring such a distribution,one of which might be that the major portsof call of the ]_]_0] HOFSLUND,Knije Island Herring Gulls Itircl-•ndingApril

ioo • oo 300

400 5 oo

6 oo

O* • YEAR BIROS 0.4 YEAR BIRDS •.5 YEAR BIRDS Figure 3. Scattergramshowing distances from banding point to place of recovery of those birds that reached t'heir •hird, fourth, or fifth year of age. iron ore shipswould be in theseareas. An observeron Lake Superior can seeconcentrations of younggulls followingsome of theseships as they move away from and toward the shippingcenters at Duluth and Two Harborsin late August. Causesof deathand the mortalityrate are muchas havebeen found in othergull studies.At leasttwo-thirds of the mortality occursbefore the Herring Gull has reachedone year of age. The usualreport is that the bird was "found dead," but several more detailed records, such as "broughthome by the cat," "killed goingover Niagara Falls," "landed in an oil pit," etc., are given in Table 1.

TABLE

RECOVERIES OF GULLS BANDED ON KNIFE ISLAND Date Banded Recovered At How Found Date Recovered June 17, 1950 Burlington, Ontario Brought ,home by Dee. 9, 1950 cat mile from Duluth Found .dead on April 23, 1951 highway June 16, 1951 Marinette, Wisconsin Landed in oil pit Aug. 28, 1951 Big MuskegoLake, Brought to museum Wisc. for identification Nov. 23, 1951 Tampico, Tamaulipas, Captured alive April 5, 1952 Mexico Pneuforbee Creek, 8 Found dead M,ay 3, 1952 miles N. Tuskegee,Ala. Coatzacoalcos,Vera Found, probably Dec. 15, 1952 Cruz, Mexico dead West River, shore of Found dead in Feb. 8, 1953 Beaver Island State water Park, Grand Island, New York Lake Kagawong, Found dead, may Oct. 5• 1953 Manitoulin Island, •have been shot Lake Huron Duluth, Minnesota Forrod dead June 7,1954 Vol. XXX 1959 HOFSLUND,Knife Island Herring Gulls [ ] ] ]

June 14,1952 Lake Superior,Duluth Found ,dead June 14, 1952 Duluth, Minnesota Drifted to shore June 24, 1952 Duluth, Minnesota Found dead• badly July 13, 1952 decomposed Knife River, Minn. Found dead Aug. 24, 1952 Manitoulin Island, Found Sept. 8, 1952 Meldrum Bay, Altoma, East County,Ontario M•assawippiLake, Killed Sept. 22, 1952 Ayer's Cliff, Start- stead County, Quebec Elmvale, Simcoe, Ont. White .diarrhea Oct. 9, 1952 Manistee,Michigan Found dead Oct. 18, 1952 West Twin River, Found dead Nov. 16, 1952 T.wo Rivers, Wisc. St. Maurice River, Caught in fox trap Blov.20, 1952 Laviolette County, Que. June 14,1952 Wilson, Niagara Found dead Mar. 31, 1953 County, New York Jay Township, Dunnell, Found with a April 24, 1953 Martin County, Minn. broken wing. Died Pentwater, Oceans Found dead July 10, 1953 County, Michigan Pur•gnivik, 10 rail.es Four•d Oct. 1953 So. of Nain, Labrador Marble Lake, Lake Found dead with May 15, 1954 County, M'innesota fishplug in mouth Mouth of Brule River, Found in fisher- May 28, 1954 Douglas County, Wisc. men's net 4 miles West of Sand Caug.ht in float nets, Nov. 18, 1954 Island off Bayfield releasedin good County, Wisconsin health Niagara Falls, Killed going over Feb. 27• 1955 Welland County, Ont. Niagara Falls Two Harbors, Minn. Found dead June IS, 1955 Mouth of French River• Found dead July 7, 1955 St. Louis County,Minn. Point Bea.eh, Two Found dead April 28, 1956 Rivers, Wisconsin Knife Island, Lake Found dead June 8, 1957 County, Minnesota June 29,1953 St. Joseph,Michigan Founddea.d (5 weeks from) Sept. 28, 1957 June 12,1954 Os.coda•losca County, Found dead Sept. 19, 1954 Miohigan BetsieBay, Frankfort, Founddea. d Sept. 23, 1954 BenzieCounty, Mich. Lake Erie, Ashtabula, Founddead Sept. 25, 1954 Ohio Des Prairies, Bizard Found Sept. 2.7,1954 Island, Laval County, Quebec I mile S.E. Woodstock, Found dead Dec. 2, 1954 M'cHenry County, Ill. Beachat Michigan Founddead Feb. 9, 1955 City, Indiana Bird-Banding 112] HOFSLUND,t•nife Island Herring Gulls April

June 12, 1954 Oswego,N.Y. Found dead Mar. 14, 1955 Corpus,Christi, Texas Foundsick, band M'ar. 22, 1955 removed Knife Island, Minn. Leg {ound June 11, 1955 Bad River, 10 m. E. of Found dead Oct. 1956 Ashland, •Wisc. Superior,Wise. Found dead Dec. 10, 1956 Superior,Wise. Found dead Dec. 10, 1956 West Kewaunee, Trapped and Jan. 17• 1957 Kewaunee, Wise. released Burlington Beach, Two Found dead June 16, 1957 Harbors, Minn. East Shore of Kewaunee Found dead July 8, 1957 Peninsula,Mich. Lake Su.perior,between Found dead Aug. 24, 1957 Munising and Mar- quette, Mich.

June 11,1955 Between Two Harbors Found dead June 22, 1955 and Duluth, M'inn. ¾• mile east Knife Found dead July 18, 1955 River, Lake Cty., Minn. Lester River, Duluth, Found dead July 19, 1955 Minn. 5% m. S. Two Harbors, Found dead Aug. 10, 1955 Minn. Turkey Island, Detroit Found dead Oct. 2041955 River, Ontario Black Cape•New Caught in f,ox trap Dec. 14, 1955 Richmort.d, ,Bonaventure Cty., GaspePeninsula, Quebec Lake Manawa, Lewis Found dead May 27, 1956 Twp., Potta•rattamie County, Iowa 12 m. E. Duluth, Minn. Hit by a car A,ug.21, 1956 3 ,m. from Ford R., Found dead May 3, 1957 Mich. East Beaver Bay, Silver Found dead May 24, 1957 Bay, Lake Cty., Minn.

June 9, 1956 17 m. N. Duluth, Minn. Found dead Aug. 8, 1956 SleepingBear Bay, Found.dead Sept. 12, 1956 Leelanau Cty., M'ieh. Riviere du Loup, Found dead Oct. 5, 1956 Quebec Shore of Ottawa l•iver, Found ,dead Oct. 10, 1956 4 .m.W. Ottawa, Quebec Grand Beach, Mich. Found dead Oct. 12•1956 Ste. Flavic• « way Found .dead Oct. 18, 1956 between Grand .Metis Bay and Mont Gali (Mr. Joli?) airport, Que. AttawapiskatRiver, 50 Shot Nov. 1956 m. N. of Ogoki, Ont. FloodBay, Two Har- Founddying Nov. 10, 1956 bors, Minn. Vol. XXX 1959 HOFSLUND,Knife Island Herring Gulls [113

Duluth, Minn. Forrod d.ead July 16, 1957 1 m. N. Camp Huan, 8 Forroddead July 16, 1957 m. N. Sheboygan,Wise. Danville, Arkansas Found dead Nov. 28, 1957 (21/• m. W.) June 8, 1957 Duluth, Minnesota Found dead July 24, 1957 Duluth, Minnesota Found ,dead (shot) July 25, 1957 Big Caribou Island, Found dead Aug. 8, 1957 Ontario Eseanaba, Mich. Found .dead Sept. 14, 1957 Park Point, Duluth, Found dead Sept. 21, 1957 Minn. S. Shore Lake Winne- Found ,dead Oct. 1957 bago, • m. N. E. Fond du Lac, Wisc. Black Bay, Lake Noted Oct. 25, 1957 Snperior• Ontario Lockport, IlL Found dead Dec. 3, 1957 Macon,Georgia Found dead J.an. 12, 1958 Bay at Bocasdel Toro, Caughton fish hook Feb. 7, 1958 Panama State Fish Hat&eries, Shot Mar. 23, 1958 Hackettstown, N.J.

Greatestwandering, as has been mentionedbefore, takes place during the first year, and there seemsto be a tendencyeven among thesefirst-year birds to turn towardtheir hatchingplace as the breeding seasonapproaches. Very few apparentlywinter near the breeding grounds;we have only one recoveryfrom lessthan 100 miles dfiring the winter, and despitethe fact that severalthousand gulls winter along the North Shoreand thoughwe have examinedflocks very carefully, we have yet to find more than one bandedbird in the flocks. Whether this was a native bird we do not know.

SUMMARY 3028 nestlinggulls have been bandedon Knife Island, Lake County, Minnesota, from 1950 to 1957. 81 recoverieshave been received. The generaltendency is for the gulls to moveto the east at the end of the breedingseason, following the shorelinesof the Great Lakes- St. LawrenceWaterway. Explosivedispersal seems to be effectively stoppedby the lack of suitablewaterways. The tendencyto move south seemsto be limited to those birds which reach the coast. Wanderingis greatestin non-breedingbirds, but there is a tendency even amongthem to return to the breedinggrounds at the beginning of the nextbreeding season. Mortality is highest in first-year birds. Causesof death where detailed are listed.

LITERATURE CITED GROSS,ALrR•;D O. 1940 The migration of Kent Island Herring Gulls. Bird Banding 11: 129-155. 114] EKLUND,IGY /lntarctica Bird Studies ]•ird-BandApril

HOFSLUND,P. ]•. 1952 1952 census of Knife Island. Flicker 24: 162-63. LINCOLN, FaEDEalCK C. 1928 The migration of young North American Herring Gulls. •luk 45: 45-59. WOODBURY,ANGUS M., and Howaa• KNIGHT 1951 Resultsof the Pacific •Gull color-bandingproject. Condor 53: 57-77. Biology Department, University o./ Minnesota, Duluth.

ANTARCTIC ORNITHOLOGICAL STUDIES DURING THE IGY

By CARLR. EKLUND Explorationand researchstimulated by the InternationalGeophysical Year have beenprimarily concernedwith the physicalsciences. The United States National Committee for IGY and similar committees of other countriesrecognized, however, the unique opportunityfor re- searchin the life sciencesin the remotepolar regions. BecauseIGY personnelwere encouragedto make suchstudies at Antarcticastations, it wasmy privilegeto do ornithologicalwork whiIe servingas Scientific Leaderat Wilkes Stationduring 1957-58. None of the countriesparticipating in the IGY employedornitholo- gists. Except for a zoologistand an ichthyologistassigned to the USSR Mirny Station,no full-timebiologists of any type were at work in Antarctica.Fortunately, most stations had personnelwith training in the biologicalsciences who were sufticientlyinterested to conduct studiesincidental to their primary duties. As we enter the post-IGY period, it is gratifying that the Polar ResearchCommittee of the NationalAcademy of Scienceshas recog- nized a continuing need for Antarctic research in the life sciences. A Panelon Biologyand MedicalScience has been functioning as part of that Committee;on its recommendation,funds have been granted for ornithologicalwork. My purposein this paper is to report ornithologicalstudies con- ductedduring the IGY at sevenUnited Statesstations and, uponthe basisof recentcontact with foreigncountries having Antarctica stations, to list their studies in this field. In the fall of 1956, throughthe USNC-IGY, I initiated a study on the distributionand life history of the South Polar Skua (Catharacta maccormicki). Relatively little has been known of the habits and movementsof this mostsoutherly of all birds, which in January, 1958, wasobserved by Sir EdmundHillary within 80 milesof the SouthPole. The main phaseof the studyrequired the bandingof a representative numberof skuas. Eight nations,including Argentina, Australia, Japan, New Zealand, Norway, USSR, the United Kingdom, and the United States,participated in making this one of the most internationallyco- operative bird studies ever conducted. The normal numbered metal bands were used and, for the first time, a l•/(2-inch-wide colored thermoplasticband also was used. A different color denoted each of eighteen banding stations in Antarctica. This enabled ready field