July]19611 327 THE MIGRATION OF LAPLAND LONGSPURS TO ALASKA LAURENCE IRVING E^RL¾in May Lapland Longspurs,Calcarius lapponicus• arrive at Anaktuvuk Passin the centerof the BrooksRange of arctic Alaska in large flocksthat fly swiftly northward in compactformations or stop to feed among the hummocksof sedgethat are exposedthrough the snow (Irving, 1960a). Judgingfrom my observations,verbal reports, and compilationsby Gabrielsonand Lincoln (1959), theselongspurs are generallythe commonestnesting species of the Alaskanarctic tundra. They also nest in westernAlaska over the grassyland beyondthe forestsas far southas the AlaskaPeninsula, on St. Lawrence(Fay and Cade, 1959) and other islandsin the Bering Sea (Preble and Atee, 1923), and alongthe AleutianIslands to Attu (Murie, 1959). Publishedreports (Friedmmm,1935) showthe occurrenceof LaplandLong- spurson Kodiak. Unpublishedreports (Fish and Wildlife Service) indicatethat they are commonsummer residentsin the Kodiak National Wildlife Refuge, which comprisesa large part of Kodiak Island. The extent of their presenceon Kenai Peninsulais not clearly shownin publications.Comments of residentson the northwesternshore indicatethat longspurspass that part of Kenai in migra- tion. Nestingprobably occurs on Kodiak and possiblyon Kenai, but I have not been able to verify this. On Middleton Island in the Gulf of Alaska (Rausch, 1958), the nestingof longspurs is detached from their arctic and western Alaskan nesting grounds. They have also been reported nesting in the interior of Alaska at Mt. McKinley National Park (Dixon, 1938); and, accordingto Gabrielsonand Lincoln (1959), George Schaller found them nesting at Black Lake in the Talkeetna Mountains. Except for the reported nesting of longspurs in these mountainous areas of the interior, the nesting populationof the mainland of Alaska is settled in a peripheral band about its arctic and western perimeter (Figure 1). South of Barrow this heavily populated band extends from the coast about 320 kilometers (200 miles) inland to the forest. Elsewhere on the malnland the densely occupiednesting area is no wider, but it extends to the islands in the Bering Sea, southwestwardalong the Alaska Peninsula and for some 2,200 kilometers (1,400 miles) into the Pacific Ocean over the Aleutian Islands. Since longspurs are the commonest birds nesting over this great area, they are one of the most numerousarian populatlons in northwestern America. The breedingLapland Longspursof Alaska are ascribedto the sub- speciesC. 1.alascensis Ridgway (A.O.U. Check-list,1957) and are said to winter in interior western states and as far north as southern British Columbia(Munro and Cowan,1947). Taking latitude 50ø and longi- tude 134ø as referencepoints for the northwesternpart of this wintering area, their migrationwould extend 20 ø of latitudenorthward and 14ø of longitudewestward to the eastern nestinglimits of the race in 328 IRw•,'(;,Migration of LaplandLongspurs [Vol.[ Auk78 160ø 150ø 140e 130e 120• 70 e 60* 160• 150• 140 ß Figure l. Northwesternrange of Calcariuslapponicus (diagonal hatched) and location of well-recorded directional migration in spring (heavy arrows). 3,lackenzie(A.O.U. Check-list,1957). In the westernmostAleutian Islandsthe migrationwould extend their summerrange only 4 ø north- ward but 67ø westward. During severalyears I have been gathering evidencefor the courseand programof this extendedand large migra- tion in Alaska. MIGRATION THROUGH THE BROOKS RANGE Froin 1948 to 1960 Simon Paneak, a resident of Anaktuvuk Pass (latitude 68ø 19' N., longitude 151ø 26' W. in the central part of the Brooks Range), recordedfor me his observationsof birds and thoseof his family and neighbors. In 1949 ThomasBrower kept recordsthere during April, May, and June. Theseobservations are supportedby many specimens.I have beenat Anaktuvukduring all or part of five spring migrations. In 10 years of suitablerecords the first Lapland Longspurswere recordedat Anaktuvuk on 1 May 1949, 29 April 1951, 9 May 1952,4 May 1953,13 May 1954,5 May 1956,8 May 1957,3 May 19'58,11 May 1959, and 29 April 1960; the averagedate of these ar- rivals being 5 May. Julyl1961] IRWN½,Migration of LaphindLongspurs 329 The dates of first recordedarrivals of longspursare more variable than for many arctic species.The daily recordsof migrationsduring eight yearsare completeenough to showthat after their first arrival on the averagedate of 5 May only a few longspurs,usually identified as males,were seenbefore the averagedate of 17 May (11 to 20 May). On that averagedate flocksof 40 or "many" first appeared.Thereafter, for 10 days,numbers recorded by Tom Browerwere as largeas 2,000 a day in flocksthat commonlyincluded from 50 to 150 longspurs. Until about 20 May the flocks included mainly males, but thereafter the proportionof femalesincreased. The less conspicuousappearance and behaviorof femalelongspurs may have biasedreports, but we have not noticeda late excessof femalesthat would offset the obvionsearly preponderanceof males. Like pipits, ,4nthus spinoletta (Irving, 1960b), and several other species,the first male longspursarrive at Anaktuvuk fat and with partially developedtestes (Irving, 1960a, p. 123). After 15 May the testeshad reachedfull breedingsize, and flight songsbegan to be noticed among coherent flocks that were apparently still migrating northward. Pairing was recordedas early as 20 May, nest buildinga few days later, and first eggsjust before1 June in most years. In the confusionof so many migrating flocksI did not discernwhether entire flocks separated and settled at Anaktuvuk or whether individuals droppedout as they paired and nestedthere while the remainderof the flocks continued northward. The great northwardmigration of longspursthrough Anaktuvuk con- tinues for about 10 days after the averagedate of 17 May and inter- mittently for a few more days. The large numbersthat remain to nest are an insignificantpart of the many flocksthat passswiftly northward or occasionallyalight to feedand drink busilyas if in hasteto continue. Occasionallya flock has beenwatched while groundedfor severalhours, but I have not searchedto find if the migrants roost overnight in the valley. Sometimesa flock circlesor makes a short flight, and on a few occasionsflights have been seen to pass southwardout of view. In most of the flocksthe movementis well coordinated,closely organized, and pursuesits coursenorthward. I have gained the impressionthat the northward movementof longspursat Anaktuvuk is as pronounced a directionalmigration as is the tremendousmovement of eider ducks flying eastwardin springalong the arctic coastpast Barrow. The long- spur is the most numerousmigrant to passAnaktuvuk in spring, an(1 many thousandsare transferredtoward the arctic slope. Althoughthe NunamiutEskimos now are basedat AnaktuvukPass, 330 IRVING,Migration of LaplandLongspurx [Vol.[ Auk78 their olderpeople have lived in manyvalleys of the BrooksRang• betweenlongitude 150ø and 158ø , which their ancestorsoccupied as nomadic caribou hunters. These Eskimos named six other valleys through which longspursmigrate northward in numberscomparable with or exceedingthe migration at Anaktuvuk and indicatedthat all valleyswere utilized accordingto the extent of their grassyland. I have seentheir migration through the Killik Valley and Howard Pass. Recordsfrom Anaktuvuk show that the earliestreports of longspurs in springdiffer from year to year morethan is the casefor other arctic migrants,each species of which commonlyarrives within'a few days of its regular date. Further, as a result of close observationby resi- dent people,it is apparentthat after the earliestlongspur was noticed, on the average 12 days elapsedwhile only a few individuals,mostly males,were seenuntil numericallysignificant migration in flocksbegan suddenlyand continuedfor 10 or 15 days. This rather unusualprogram of migration may be peculiarto the longspursat Anaktuvuk; or it may be that the observationof migration by many Eskimos resident in open tundra, where visibility is especiallygood and migration is con- strictedin a narrow valley, detectsthe few early migrantsthat escape the view of singleobservers at lessfavorable positions. MIGRATION ELSEWHERE IN THE WESTERN ARCTIC Longspursnest near Kittigazuitin the MackenzieDelta '(Porsild, 1943). Severalreports of migrationand nestingon the arctic coastof Yukon Territory have been compiledby Rand (1946), and Kesseland Schaller (1960) recentlyfound a few nestingnear the headwatersof the Sheenjek River in the northeasternAlaskan Brooks Range. We did not find them in the valley or on the tundra within 16 kilometers (10 miles) north of Old Crow, Yukon Territory (Irving, 1960a). While watching the spring arrivals at Demarcation Point and Humphrey Point, respectively,in 1914, Brooks (1915) and Dixon (1943) reportedthat a few longspursarrived there to nest. From their accountsand from the reportsof Eskimoswho have lived on the eastern arctic coast of Alaska, I have concludedthat there is no evidencefor a westward migration passing from Mackenzie Valley coastwiseinto Alaska (Irving, 1960a). Alfred Bailey (1948) reportedthat longspursseemed suddenly to arrive at Barrow and all alongthe westernarctic coastin the third week of May. During the 1st International Polar Year John Murdoch (1885)reported their arrival at Barrowon 20 May 1880. Sincemigra- tioncoastwise along the northeasternAlaskan coast has n6t beenre- July}1961J IRVING,Migration of LaplandLongspurs ported,I havetaken for grantedthat migrantsto the arcticslope