Migration of Little Gull (Larus minutus) 1n the North Atlantic Region

By BERTEL BRUUN

(Med et dansk resume: Dværgmågens (Larus minutus) træk i det Nordatlantiske område.)

INTRODUCTION While on vacation in U.S.A. I observed led to a complete reevaluation of our a Little Gull (Larus minutus) in Maryland knowledge of the distribution and migra­ in 1963. This observation spurred my tion of this species in the entire North interest in the frequent occurence of this Atlantic region. The results of this re­ species in N orth America. evaluation is presented here. The Little Gull is one of the most In my work I have received much help numerous palaearctic species occuring on from 0. L. AusTIN, S. CHRISTOFFERSEN, R. the western side of the North Atlantic. EDBERG, L. FERDINAND, F. GuDMUNDssoN, E. HANSEN,]. V. HANSEN, N. HESSELBJERG­ The research done in regard to this CHRISTENSEN, H. JOHANSEN, J. JØRGENSEN, phenomenom has led me far afield and B. NEATH, N. 0. PREUSS, c. s. ROBBINS, has included all other palaearctic species N. RosENBERG and J.-P. VAUDE WEGHE. which have occured in North America. The U. S. Wildlife Service and the Danish The results of these more involved studies Meteorological Institute have furnished will be published elsewhere. me with information essential for my work. My investigations of the puzzling oc­ I take the opportunity to thank all for their curence of the Little Gull in North America contribution and patience.

METHODS. In my research I have accumulated all Most of the literature used in accumulating the records of Little Gulls from North records is not cited as it includes several America, Greenland, Northern Germany, hundred different references. These re­ Balticum and the Scandinavian countries cords have been supplemented by banding up untill 1965 as they have occured in the records and breeding records from other literature or have been put at my disposal parts of Europe. by ornithologists from these countries. Migration of Little Gull (Larus minutus) 127

BREEDING. The Little Gull has a rather restricted area is: Holland c: a 20 (ERARD 1960), breeding range in Europe. It stretches Denmark c:a 20, Sweden 20-100 (SALO­ from the eastern part of Central Russia MONSEN 1963), Finland 200 (MERIKALLIO to the N orth Sea and from the Botnic Bay 1958), and at the most 5000 along the to an isolated population along the north­ southern coast of the Baltic Sea (BERZIUS ern coast of the Black Sea. It is also found 194 7' LEPIKSAAR and ZASTROV 1963' breeding in Central and Eastern Siberia N IETHAMMER et al. 1964, T1sHLER 1941) . (VAURIE 1962). In the westernmost part of its range it is very scarce, and the main Considering this very small number (in part of the population is found in the relation to the numbers of other species Baltic countries and Western Russia. of gulls in this region) it becomes even Only the hirds breeding in Western more puzzling that this species should occur Europe and along the Baltic Sea reach on the western side of the North Atlantic as the Atlantic in any numbers in winter. A often as it does. This implies the existence few wintering in the Mediterranean reach of factors favouring Atlantic crossings of the Atlantic through the Strait of Gibral­ this species compared to the other gulls tar. The number of breeding pairs in this found in this area.

Fig. I. Recoveries of Little Gulls in western Europe. •: recovery si te, O: banding site. Numeral at recovery si te: mon th of recovery. Fig. I. Genmeldinger af Dværgmåger fra Vesteuropa. •: genmeldingssted, 0 : ringmærkningssted. Tal ved genmel­ dingsstedet: genmeldingsmåned. 128 Migration of Little Gull (Larus minutus)

BAND ING. From Western Europe I have been able the Baltic Sea rather long after the main to accumulate a total of 18 recoveries of exodus from this region (July and August - banded Little Gulls. These are shown on see below). fig. 1. The recoveries indicate a westerly Another noteworthy return is that of a course of migration towards the N orth Sea, Finnish bird from Norway in late October amore southern (probably via the Central of the same year. Also of interest is the re­ European rivers) towards the Mediter­ covery from the Atlantic coast of France ranean and a southeastern course towards of a bird banded on the coast of the Black the Black Sea. This corresponds complete­ Sea, showing at least some birds of ly with ERARD's ( 1960) findings, and does southeastern origin can reach the Atlantic, not throw light on the problem of the probably via the Strait of Gibraltar. It occurrence in N orth America. must be pointed out though that banding Of particular interest is the recovery in has not clarified the migratory pattern of October of a bird close to the place of this species and further banding is neces­ banding in Letland. This shows that some sary to get a more detailed picture. birds can remain in the eastern part of

FALL MIGRATION. The records from N orth America seem new to elucidate the phenomenom (BRUUN, completely incompatible with the migra­ MS). tory pattern of the Little Gull as previous­ A reevaluation of our knowledge of the ly known (ERARD 1960). The areas tra­ migratory habits in Western Europe is versed, the time of migration and the therefore necessary to elicit the special winterquarters of the main bulk of the factors favouring Atlantic crossings of this population show none of the character­ species. istics necessary for a transatlantic crossing A possible place of origin is the Atlantic on a larger scale ( either westerly directed off the Iberian peninsula and N orth course and/or little adherence to coast­ Africa, but the Little Gull is but a rare lines and/or wintering grounds in the vagrant to these parts. Although some northern North Atlantic) except for indi­ crossings might take place from here, they cations of a more pelagic winterexistence cannot possibly account for the large than is the case of most other gulls (see majority ofbirds recorded from the western below). As the faet of the magnitude of side of the N orth Atlantic. this crossing remains it indicates 1) either our knowledge of the conditions for such It seems more likely a crossing could be a crossing is inadequate or 2) our know­ made at the northern part of the North ledge of the migration of Little Gulls in Atlantic as is the case of most other euro­ Western Europe is insufficient. pean species regularly encountered in An investigation of and comparison with North America. In the following an at­ other species in regard to transatlantic tempt is made to clarify the migration in crossings from east to west reveals nothing this more northern area. Migration of Little Gull (Larus minutus) 129

FALL MIGRATION ALONG THE SOUTHERN COAST OF THE BALTIC SEA. The figures from Kurischer N ehrung of Schleswig-Holstein and it is known from clearly show the main exodus from the several observations they turn south to breeding grounds in the Baltic countries continue through Central Europe at the taking place in July and August (fig.2). Oder and other rivers which they follow Only very small numbers pass in Septem­ southward (ERARD 1960). There are no ber and October. The figures from Schles­ records of Little Gulls from Rugen indi­ wig-Holstein show a clear maximum in cating the southward turn is made before September indicating a continuation of this islands is reached. A small proportion the migration along the southern coast of does continue along the coast though. The the Baltic Sea (fig. 2). It must be noted direction is westward and the birds pass that a much smaller number of birds are Schleswig-Holstein to reach the N orth Sea, involved in this western continuation. where their continued course can be foll­ Most of the birds passing Kurischer Neh­ owed to England and the coast of France. rung disappear before they reach the coast

FALL MIGRATION ALONG THE NORTHERN COAST OF THE BALTIC SEA. An analysis of the Swedish records show is not involved. Whatever the ongm, it a picture quite different from that found is evident that some birds do not follow along the southern coast of the Baltic Sea. the regular pattern, but migrate at a later The fall migration culminates two month time and on a more northern latitude. later (October), but a much smaller The fall records from N orway are too few number of birds are involved (fig. 3). to be significant, but the maximum is 5 The origin of these birds is obscure. records from September. There are two The return of a Finnish bird from Norway records from October and one from No­ in October implies birds from the north­ vember. This indicates that these northern ernmost part of the Baltic breeding migrants do not to any extend pass Nor­ grounds. The three returns ofbirds banded way, but probably pursue their westerly in Sweden indicate that this population course just south of this country.

Kurischer Nehrung Schleswig-Holstein

3000 150 Fig. 2. All published records of Little Gull at Kurischer Nehrung and Schleswig-Holstein by 2000 100 months. Note number of birds compared to other diagrams.

1000 50 Fzg. 2. Alle publicerede rekorder af Dværgmåge fra K urischer Nehrung og Schleswig-Holstein månedsvis op­ delt. Bemærk skalaen i forhold til 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 andre figurer. 130 Migration of Little Gull (Larus minutus)

THE FALL MIGRATION INDENMARK. The Danish records peak in August and by these birds passing over Denmark, September reflecting the migration along which country thus enjoies part of both the southern coast of the Baltic Sea also migratory routes in the Baltic Sea. The passing Schleswig-Holstein (fig. 3). Of Schleswig-Holstein records do not show great importance though is the increase any rise in November indicating the area in records in November reflecting the traversed by the late migrants is narrow. migration at that time along the northern The birds probably pass rather fast com­ coast of the Baltic Sea (Sweden). This pared to the more leisurely pace of the diphasic pattern is significant although a early migrants. The late migration is prob­ rather small number of birds are involved. ably spurred by cold weather in the The lack of N orwegian records of the late eastern part of the Baltic Sea. migration is at least partially explained

THE FALL MIGRATION OF THE BRITISH ISLES. Numbers slowly build up through August detected (fig. 4). The northern latitude of and September, culminate in October and these islands is significant. taper off in November in England and The Irish records are very few indeed, Wales (fig. 4). In recent years there have but most are from December (fig. 5). It been an increasing number observed in is not possible to determine if these birds Scotland late in the season (NEATH in are true migrants or wintering birds (see litt.), numbers which are not large enough below). though to be reflected in the collected ERARD ( 1960) finds evidence of an area material (fig. 4). The pattern seems to of concentration found in the Bay of Biscay reflect the early migration (in which a which is frequented by many other pelagic larger number of birds participate), and birds during the winter. The concentration only in the records from the Orkney and here starts building up already in July. Shetland Isles can the late migration be That a crossing to North America can

Sweden

Den mark 300

Fig. 3. All published records of 100 Little Gulls observed in Sweden 80 and Denmark by months. Note 20 number of birds compared to 60 other diagrams. 100 40 Fig. 3. Alle publicerede rekorder af 20 Dværgmåge fra Sverige og Danmark månedsvis opdelt. Bemærk skalaen i forhold til andre figurer. l 2 3 4 5 6 7 8 9 10 11 12 l 2 3 4 5 6 7 8 9 10 11 12 Migration of Little Gull (Larus minutus) 131

England and Wales

Shetland and Orkney (sies 600 Scotland 10

400 800 8 600 6.'

200 400 4

200 2

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 .4 5 6 7 8 9 10 11 12

Fig. 4. All published records of Little Gull from England and Wales, Scotland, and Shetland and Orkney by months. Note number of hirds compared to other diagrams.

Fig. 4. Alle publicerede rekorder af Dværgmåge fra England og Wales, Skotland og Shetland-Orkney månedsvis opdelt. Bemærk skalaen i forhold til andre figurer. take place from this area is likely. An known to traverse the area in question indication of this is the recovery in south­ and have prohahly started their crossing ernmost North America of a Black-headed at this point. Crossings starting from this Gull ( Larus ridibundus) and two Gull-billed region are quite rare though in comparison Tems (Gelochelidon nilotica), all three with the large numher of hirds of different handed in northwestern Europe (Den­ species crossing further north (BRUUN, mark and Germany). These hirds are MS).

FALL RECORDS FROM ICELAND AND THE FAROE ISLES. Records are very few, but the largest Shetland Isles (fig. 5). Although com­ numher is from N ovemher reflecting the paratively few hirds are involved in this late northern migration evident from late migration, the western direction is Sweden, Denmark and the Orkney and significant in respect to Atlantic crossings.

lceland and the Faeroe lsles lreland 6 Fig. 5. All published records of 5 Little Gull from Ireland and 4 4 Iceland-Faroes by months. Note the number of birds compared to 3 other diagrams. 2 2 Fig. 5. Alle publicerede rekorder af Dværgmåge fra Irland og Island­ Færøerne månedsvis opdelt. Bemærk 1 2 3 4 5 6 7 8 9 10 11 i2 1 2 3 4 5 6 7 8 9 10 11 12 skalaen i forhold til andre figurer. 132 Migration of Little Gull (Larus minutus)

FALL RECORDS FROM NORTH AMERICA. The Little Gull is one of the most regular in its direction. Further faciliating the visitors from Europe to the western side crossings of these late hirds is the increased of the North Atlantic, superceded in frequency of eastern storms in the eastern numbers only by Teal ( Anascrecca), Wigeon part of the northern North Atlantic at (Anas penelope), Curlew Sandpiper (Calidris this time (from 5% in October to 11 % firruginea), Ruff (Philomacus pugnax) and in November, Danish Meteorological In­ Black-headed Gull. stitute). Studies (BRUUN MS, ERSKINE One of the hig events in American 1963) have shown many other European ornithology in this century was the dis­ covery of three pairs of Little Gulls nesting in a swamp in Ontario in 1962 (ScoTT 1963). This discovery is indicative of the North America extent of the transatlantic crossings per­ 10 formed by this species. I find it highly in­ 8 conceivable that the records of Little Gulls from N orth America stem from a small 6 local population. First of all we have no 4 indication that such a breeding population 2 exists but infrequently, and secondly it would be most unlikely that such a small 1 2 3 4 5 6 7 8 9 10 11 12 population could account for the many Fig. 6. All published records of Little Gull from records. North America by months. Note the number of Some numbers arrive already in J uly, birds compared with other diagrams. August and September, but the majority Fig. 6. Alle publicerede rekorder af Dværgmåge fra Nord­ are observed in November and December amerika månedsvis opdelt. Bemærk skalaen i forhold til (fig. 6). andre diagrammer. Several of the records are from the Great Lakes Region. Apparently these hirds have reached these very large lakes visitors to North America originate in the via the St. Lawrence. The early records area between Iceland and the British probably stem from the small westbound Isles i.e. Black-headed Gull, Barnacle branch of the route along the southern Goose ( Branta leucopsis), Wigeon and T eal. coast of the Baltic Sea, whereas the later It is also of interest to note the welldocu­ and much larger part is derived from the mented invasion of Greenland by the later and more northern route along the Fieldfare (Turdus pilaris) originated in the northern coast of the Baltic Sea. This same area and at the same late season of latter route is also more directly western the year (SALOMONSEN 1950).

SUMMARY OF FALL MIGRATION. The greatest part of the westernmost Little of the Baltic Sea. Here most turn south to Gull population leave the breeding follow the great European rivers to the grounds in the Baltic in July and August, Mediterranean, while some continue on travelling west along the southern shore a westerly course, pass over Denmark and Migration of Little Gull (Larus minutus) 133 northern Germany to reach the North Sea, in November. From here they, aided by the Channeland the Bay of Biscay. They rather frequent easterly gales, reach the arrive here fromjuly to September. A few North American coast in November and of these birds continue their western course December in considerable numbers. and reach N orth America, but these are few and rather insignificant in numbers. The reason for the preference for this A part of the breeding population stays more northern route by the late migrants in the eastern part of the Baltic Sea untill is obscure. The rivers of central and east­ October when they migrate west along ern Europe are not yet frozen at this time the northern coast of the Baltic Sea, con­ (Danish Meteorological Institute) al­ tinue their western course past Denmark though the weather certainly is colder and reach the northernmost part of the here than along the seashores of N orth­ N orth Sea and the open sea beyond that western Europe.

WINTER EXISTENCE. During the nineteenfifties there were in­ that the birds which reach the North creasing evidence the Little Gull led a far Atlantic, either as the small western branch more pelagic winterexistence than was of the early route or as the later and more previously known (i. e. sightings from northern migration, lead a more pelagic lightships (HANSEN 1962), sightings of life than is the case of most other gulls Little Gulls in the company of true pelagic breeding in the same area. species (BouRNE 195 7)). This was pointed Some of the Little Gulls wintering in out by V oous ( 1960) who compared the the Mediterranean do stray into the North Little Gull with Sabines Gull ( Xema sabini) Atlantic (see page 127) and there is evi­ and by ERARD (1960). Analysis of earlier dence (BouRNE 195 7) from N orth Africa, records show that this is not a new phe­ that the Little Gulls wintering in the Medi­ nomenom though. Thus the observation terranean often stay far from shore. Some of c :a I 00 Little Gulls near Larkallen, of the winter records from the southern Norway in late January 1930 probably part of North America and Bermuda are of birds blown ashore from their usual could be interpreted as representing hirds wintering grounds further out to sea of this origin. (LøvENSKIOLD 194 7). The records from the North Atlantic (fig. 3-5) show the The birds of the open sea are more sub­ species present in small numbers only in jected to the plays of the elements, and the North Sea where it is usually observed some Little Gulls probably make landfall after storms. There are proportionally on the western side of the Atlantic during many records from Ireland, Iceland and the winter months although they do not the Faeroe Isles and N orth America. From compare in numbers to the late fall mi­ these observations it can be concluded grants reaching this far.

SPRING MIGRATION. The pattern of migration in spring is Sweden are passed during the month of simpler than that of the fall (fig. 2-5). May, late in which month the birds arrive The main migration in Scotland is in at their breeding grounds. April, whereas England, Denmark and In Iceland there is an increase in num- 134 Migration of Little Gull (Larus minutus)

hers already in March, whereas most of to breed on the North American continent, the North American spring records are at least on one occasion. from May with a fair number seen in The Little Gulls observed in spring in April. This latter pattern is probably North America are recruited from a) early caused by the faet that the hirds are more fall migrants (few), b) late fall migrants vulnerable to displacement at the time of (many) and c) wintering hirds starting migration as it is known from many other spring migration over the open sea (few). species (DoRsT 1960). The American hirds With our present knowledge of the Ameri­ move northward along the coast even can breeding population there is no reason reaching as far north as Greenland (SALO­ to believe many of the records are of pure MONSEN 1950). The number involved is so American origin. great that the species has even been able

DISCUSSION AND SUMMARY. From the above some new facts about the holds true for the vast majority of hirds migratory habits of the Little Gull have but a small propotion displays a com­ emerged. Our previous knowledge still pletely differeni, and for the occurence of

Fig. 7. Summary of fall migration of Little Gull. The latter half of fig. 2 to 6 superimposed on map of the North Atlantic. The scale of each diagram corresponds to that of fig. 2 to 6. Open arrow: direction of the late, northern migration. Shaded arrow: direction of the early, southern migration. Fig. 7. Dværgmågens efterårstræk. Sidste halvdel af figurerne 2 til 6 oveifort på kort over Nordatlanten. Bemærk at skalaerne svarer til de ved fig. 2 til 6 benyttede. Aben pil: det sene, nordlige træks retning. Skraveret pil: det tidlige, sydlige træks retning. Migration of Little Gull (Lerus minutus) 135 the species in the North Atlantic, highly of the majority of Little Gulls in North significant pattern (see fig. 7): America and is supported by the occurrence When the main group leaves in early of the species in the Northwestern Euro­ fall the !argest number follows inland pean countries. Earlier arrivals in America routes to southern Europe to spend the are probably recruited from the popu­ winter there. A smaller number goes west lation wintering further south in the Bay to reach the British Isles and the Bay of of Biscay and off the English coast, but Biscay where at least a part ofthem pursue these birds are few in numbers. a pelagic type of life throughout the winter During the winter and spring several months. A small proportion of the birds birds from both these areas reach the of the Bal tic region stayes in the Baltic Sea western North Atlantic during easterly untill October-November, when they de­ storms. Some Mediterranean birds, pass­ part on a directly western course bringing ing through the Strait of Gibraltar, like­ them into the northern North Atlantic. wise may reach this far west although Here they stay well off shore probably their numbers are insignificant. It is now mixing with Kittiwakes ( Rissa tridactyla) known that a very small breeding popu­ and other truly pelagic species. They are lation exists in North America, at least in subjected to strong and rather frequent some years. This population is too small easterly storms aiding them in reaching though (3 nests found) to be responsible the western shores of the N orth Atlantic. for all the records, and most of the birds The pattern outlined above corresponds seen in U. S.A. and Canada have crossed well with the time and place of the arrival the Atlantic.

REFERENCES

BENT, A. C., 1921 : Life Histories of N orth American MERIKALLIO, E., 1958: Finnish Birds: Their distri­ Gulls and Terns. - U. S. Nat. Mus. Bull. 13. bution and numbers. Fauna Fennica. BERZIUS, B., 1946: Något om Lettlands Måsfåglar. NmTHAMMER, G., H. KRAMER and H. E. WALTERS, - Vår Fågelviirld 3: 119-125. 1964: Die Vogel Deutschlands. - Frankfurt BouRNE, W. R. P., 1957: Manx Shearwater, Little am Main. Gulls and other Sea-birds wintering off the SALOMONSEN, F., 1950: Grønlands Fugle. - Kø­ Algerian coast. Ibis 99: 117-18. benhavn. BuLL, 1964: Birds of New York Area. - New J., - 1950: The Immigration and Breeding of the York. Fieldfare (Turdus pilaris L.) in Greenland. - DoRST, 1956: Les Migrations des Oiseaux. - J., Proc. of Xth Int. Orn. Congress. Paris. - 1963: Nordens Fugle i Farver vol. VII pp. ERARD, C., 1960: Sur l'aire de Reproduction, les 120-22. København. zones d'hivernage et les migrations de la Mou­ ette Pygmee ( Larus minutus) Pallas. - Alauda SCOTT, G. A., 1963: First Nesting af the Little Gull 28: 196-224. ( Larus minutus) in Ontario and in the New ERSKINE, A., 1963: The Black-headed Gull ( Larus World. Auk 80: 548. ridibundus) in Eastern North America. Audu­ TrscHLER, F., 1941: Die Vogel Ostpreussens. Ber­ bon Field Notes 17. lin. HANSEN, L., 1962: Fugle på Lolland-Falster. - TRANSCHE, N. v., 1965: Die Vogelwelt Lettlands. - Dansk Ornith. Foren. Tidsskr. 56: 164-65. Hannover Dohren. LEPIKZAAR, J. and M. ZAsTRov, 1963: Die Vogel VAuRm, C., 1965: The Birds of the Palearctic Estlands. - Ann. Soc. Tartuensis ad Res Na­ Fauna, vol. II. - London. turae Investigandas Constitutae. Lund. LøVENSKIOLD, H. L., 194 7: Handbok over Norges Voous, K. H., 1960: Atlas of European Birds. - Fugler. - Oslo. London. 136 Migration of Little Gull (Larus minutus)

DANSK RESUME

Dværgmågens (Larus minutus) træk i det Nordatlantiske område.

Dværgmågen er en af de palæarktiske fugle, som består af fugle som er forblevet i Østersøen i læn• hyppigst er observeret i Nordamerika. Årsagen til gere tid, og først i oktober og november starter denne optræden er uklar, specielt når artens for­ deres træk. I modsætning til de tidligere fugle træk• holdsvise sjældenhed i Vesteuropa tages i betragt­ ker disse sidste langs Østersøens nordkyst i en di­ ning. Med vor traditionelle viden om artens ud­ rekte vestlig retning, og når ud i Atlanten i om­ bredelse, træk og trækvaner lader forholdet sig rådet mellem Island, de Britiske øer og Norge. ikke forklare. Her er de på denne årstid udsat for ret hyppige I et forsøg på at finde til de faktorer, som er af­ østlige storme, som fører dem over Atlanten til gørende for de hyppige atlantiske krydsninger, har Nordamerikas østkyst. Det er netop fra denne jeg dels sammenlignet Dværgmågen med andre i årstid, de fleste observationer af Dværgmåger i Nordamerika hyppigt optrædende arter (Hætte• U. S.A. og Canada stammer. Det er desuden fast­ måge ( Larus ridibundus), Sildemåge ( Larus Juscus), slået at Dværgmågen i vinterhalvåret fører en me­ Bramgås (Branta leucopsis), Krumnæbbet Ryle (Ca­ get pelagisk tilværelse i selskab med arter som Sa­ lidris ferruginea), Brushane ( Philomachus pugnax) Pibe­ binemåge ( Xema sabini) og Sule ( Morus bassanus). and (Anas penelope), Krikand (Anas crecca), etc.), De amerikanske iagttagelser anses derfor hoved­ dels foretaget en nøjere analyse af artens optræden sageligt at stamme fra dette sene, vestligt rettede og trækveje og -vaner i Vesteuropa. Den først­ træk, mens kun få stammer fra det tidligere og nævnte undersøgelses resultater vil blive offentlig­ mere sydvestligt rettede træk og fugle overvintrende gjort andetsteds. Undersøgelsen gav ikke nogen til havs. direkte oplysninger om Dværgmågen, omend nok antydninger om hvori disse faktorer måtte ligge. Forårstrækket Således er det klart, at hovedparten af de i Amerika Forårstrækket i Europa begynder allerede i marts, iagttagne europæiske fugle stammer fra atlanter­ men finder hovedsageligt sted i april og maj. Dette havsområdet mellem Island, de Britiske øer og træk er til dels reflekteret i det amerikanske materi­ Norge. ale, idet de til havs overvintrende fugle i perioden med øget trækaktivitet er mere udsat for afdrift. Efterårstrækket Dog er den største part sikkert fugle, som allerede Dværgmågens hovedudbredelsesområde er at finde tidligere har krydset Atlanterhavet, og som ved i Østeuropa, og den vestlige population er kun af deres træk langs kysten udsætter sig mere for iagt­ ringe størrelse, næppe overstigende 6000 par. Af tagende ornithologer. disse trækker en meget stor del via de europæiske floder til middelhavsområdet, hvor de overvintrer. Slutning Nogle af disse fugle kan nå Atlanterhavet gennem Et af århundredets store ornithologiske begiven­ Gibraltarstrædet. heder i Amerika var opdagelsen af tre par Dværg• En del trækker imidlertid gennem Østersøen måger ynglende i Ontario 1962. Jeg tror ikke, langs dennes sydkyst til deres overvintringsom­ denne enkelte yngleiagttagelse er udtryk for andet råder langs Atlanterhavets kyster, specielt de Bri­ end det store antal fugle, som regelmæssigt krydser tiske øer og Biscayabugten. Atlanterhavet, og der er ikke med vor nuværende En nøjere undersøgelse af dette vestgående træk viden grund til at tro, at flertallet af de i Amerika (fig. 2-6) viser, at det foregår i to etaper. Første optrædende Dværgmåger skulle være klækket der. etape involverer det største antal fugle og foregår Ringmærkninger (fig. 1) støtter den ovenstående allerede i juli, august og begyndelsen af september. udredning af trækket, men giver i sig selv ikke Fuglene trækker i sydvestlig retning og når vinter­ noget væsentligt bidrag til vor forståelse af de­ kvartererne' allerede i disse måneder. Anden etape taillerne; dertil er materialet for lille.

Manuscript received 2nd oct. 1967. Authors address: 317 West 83 Street, New York, N.Y. 10024, U.S.A.