Discussion of the Assemblage in Korea, with special reference to Yésou (2001 & 2002).

Although the extent of phenotypic variation of taxa belonging to the Assemblage occurring in South Korea, and elsewhere in East Asia (*- see note below) is not well-known, it is evident that at least three major forms, with structural characteristics and distinctive juvenile, first winter and winter adult plumages can be described and identified. Furthermore, it is apparent, based on fieldwork over several years, that each of these major forms also show some significant differences in behaviour and distribution.

This is somewhat in contrast to the findings of Yésou (2002), who, in adopting a more conservative stance than in 2001, suggests that while nominate vegae is a polymorphic taxon showing little variation in mantle colour, but either yellow or pink legs, both ’birulai’ and ’taimyrensis’ should be considered invalid taxa and Mongolian Gull should be "provisionally" considered a subspecies of vegae.

Considering ’taimyrensis’ as an invalid taxon, Yésou, for the reasons given below, presumably considers that the dark-mantled, yellow-legged gulls found in East Asia in winter are either (1) all heuglini, (which he regards as a monotypic species), are (2) birds produced in a hybrid zone that Yésou himself speculates no longer exists, or (3) that many are so-called yellow-legged Vega ("from the range of ’birulai’").

Yésou states that the only gulls in the Assemblage with mantles darker than vegae are heuglini (though see the "Busan Gull" above), strongly suggesting that the darkest birds described in Kennerley et al (1995) and Hoogendoorn et al (1996) in Hong Kong and Japan are nominate (western) heuglini (Yésou, 2001). This then gives a wintering range for heuglini stretching from East Africa, through the Middle East to Far East Asia (presumably the largest non-breeding range of any taxon in the Assemblage).

Some descriptions of tentatively identified nominate heuglini have included reference to its long-legged look, large size and aggressive behaviour. In Korea, however, many dark-mantled, yellow-legged (or sometimes yellow-pink and pink-legged) birds appear rather small and short-legged, and perhaps not as dark-mantled as the heuglini depicted in plates 240 and 241 in Yésou (2002).

Yésou appears to rule out the possibility of these individuals being either taimyrensis or of hybrid origin, suggesting that there was a "limited hybridization zone which possibly existed up to 30 years ago", and presumably no longer exists, and that ’taimyrensis’ refers to this zone".

A consideration of information presented by Yésou (2002) does however provide further very useful insight into the naming and possible origin of the gulls that winter in Korea. He gives the following distribution of "northern" gulls west to east from the Kola Peninsula: (1) Kola Peninsula to the Gydan Peninsula and possibly the north-western reaches of the Yenisey: Darkest-mantled, yellow-legged gulls, becoming paler eastward; (2) western Taimyr: dark-mantled yellow-legged gulls, intermediate between birds west and east in mantle coloration (where colonies contain "neither dark nor intermediate breeding birds, but only yellow-legged birds with a mantle colour matching vegae" [presumably of the darkest-mantled form]; (3) Taimyr east to the Lena, and extending to northern islands, such as the New Siberian Islands: paler-mantled with variable leg colour, "the frequency of yellow legs seems to decrease eastwards"; (4) Lena east to the Pacific: darker-mantled with pink legs.

Although this arrangement largely is in accordance with historical division of the forms (into (1) heuglini; (2) taimyrensis; (3) birulai; and (4) vegae), Yésou suggests a simplification of the arrangement, so that the division between heuglini and vegae is drawn at western Taimyr, and all gulls east of nominate Heuglin’s then become part of Vega, or East Siberian Gull, without subspecific division. Vegae, which is consistently dark-mantled and pink-legged east of the Lena, then becomes a highly polymorphic taxon with a vastly expanded breeding range. This single species then ranges from dark-mantled to pale-mantled to dark-mantled again west-east, with leg colour ranging from all yellow, to yellow or pink, to all pink along the same west-east axis. With the addition of mongolicus as a subspecies of East Siberian Gull, the range of biometrics, phenotypic appearance, nesting ecology and behaviour in this newly-coined East Siberian Gull becomes remarkably comprehensive.

Expectedly, this choice is disputed, and although based on extensive and valuable research appears rather flawed for several reasons: it appears to be inconsistent with the recognition of other distinct populations; it runs counter to the gradual fine-tuning of lariidae taxonomy in other regions of the world as the knowledge base has expanded; and it does not include existing information on e.g. the appearance of juveniles and first winters that could have been accessed by Yésou.

Yésou correctly asserts elsewhere in the paper, that "important phenotypic divergence" needs to be recognized, and that taxonomy needs to "describe the species’ phenotypic variation: a useful descriptive tool" (Yésou 2002). The three historically recognized "northern" taxa nesting closest to Korea are taimyrensis, birulai and vegae. Considering all three as part of a single taxon seems rather retrogressive and unhelpful, especially as they show discernible differences in mantle coloration of adults, and as this paper demonstrates, differences in juvenile/first winter plumages (at least between presumed taimyrensis and vegae/birulai).

In the understanding that primary patterning and molt timings are inconclusive at best, and that separation of the northern taxa on phenotypic elements is based largely on mantle and leg colour, it also appears surprising that the dark-mantled, yellow-legged birds breeding in the western Taimyr (and eastward?) should be considered by Yésou to be closer to vegae than heuglini. As he states, research at the west and the center of its range already shows that Heuglin’s Gull mantle-colour becomes slightly paler eastward, evidently approaching vegae (reaching 4,5 on Munsell’s 37-step scale of colour refraction, with vegae at 5: Yésou, 2001). The Taimyr birds appear on the basis of mantle coloration at least to fit in predictably with this tendency. Presumably as they show some significant but unspecified differences from heuglini (as they are treated as part of vegae by Yésou), the conformity in appearance described at those colonies that were investigated by Filchagov also strongly suggest that Amadon’s 1949 75% rule for subspecific variation was met.

There is therefore an identifiable yellow-legged subspecies, geographically located east of nominate heuglini and west of birulai.

It would be most interesting to discover on which basis it can be separated from heuglini, over how great an area it is found nesting, and its abundance.

Considering that these west Taimyr nesting, dark-mantled yellow-legged birds seem to have already been contained within the wide series originally collected and named as ’taimyrensis’ it also seems appropriate to maintain that name, and to describe such birds as taimyrensis wherever they occur. Although this might appear to be a misapplication of the name (as it perhaps covers only part of the original series), it appears that it is the most fitting one, being used for birds contained within the original series; that nest in the Taimyr; and that are already largely known as taimyrensis. It seems more important then to better define the phenotypic and geographical range of taimyrensis, through more extensive mtDNA studies and through examination of juveniles within these "pure" colonies.

Without access to clear biometric data, descriptions of juveniles from the breeding colonies, or banding recoveries, it is difficult to state with certainty where these taimyrensis winter. However, in South Korea and other parts of coastal East Asia (most especially Hong Kong and the south China coast) large numbers of gulls apparently matching descriptions of adult taimyrensis occur in winter. Such birds are often small, short-legged and not strikingly dark-mantled when compared directly with eastern vegae. They clearly do not belong to the eastern pink-legged form of vegae, and juveniles which show a similar structure to the adults they are often found with (especially during the peak of migration) show clear similarities to juvenile plumages of Lesser Blacked Gull. Such birds are therefore either a still undescribed but widespread taxon, or, more likely heuglini, taimyrensis, or perhaps barabensis, which might well have occurred as a vagrant to both South Korea and Hong Kong (Kennerley et al, 1995) but is a form that appears significantly different from the vast majority of birds in coastal east Asia both in structure and primary patterning. Many also show rather less strong yellow tones to the legs.

It is surely appropriate therefore, in the apparent absence of improved information, to continue to refer to the paler and perhaps smaller of these birds as taimyrensis, in accordance with Kennerley et al (1995) and Hoogendoorn et al (1996).

It is clearly difficult (and beyond the scope of this paper) to confidently describe the taxon taimyrensis as a full species, or as a subspecies of either L. heuglini or L. vegae (or even perhaps of southern-nesting barabensis), but as the juvenile plumages, bare parts and mantle coloration are closer to heuglini, this does strongly argue against it being part of Vega Gull. Provisionally at least, again following Kennerley et al. (1995) it should be maintained as part of Heuglin’s Gull.

A very few larger, more long-legged and powerful individuals also occur in winter, which appear to more closely resemble nominate heuglini. Considering the greater distance from their breeding grounds to Korea, their relative scarcity to the other forms could be expected.

Their scarcity also suggests that while many suspected taimyrensis appear to migrate into coastal East Asia during the non-breeding season, most nominate heuglini have a different migration strategy, taking them e.g. into East Africa and the Middle East - a situation perhaps somewhat analogous to the differing migration strategies of fuscus and more western populations of Lesser Black-backed Gull.

In addition, large numbers of paler, less distinctive/characteristic individuals also appear in winter in Korea: they show variable leg colour, and paler upperparts coloration than typical taimyrensis and eastern Vega; "atypical moult"; and a structure often closer to taimyrensis than the darker Vega. Although variable leg colour could be a product of seasonal changes, diet, and even it has been suggested age, from descriptions in Yésou (2002), many of these individuals most likely derive from populations that breed between the Taimyr and the Lena. Such populations "show a highly variable leg colour…otherwise (are) similar in mantle colour", which can be deduced is also paler than that invariably found in the pink-legged vegae east of the Lena (Yésou, 2002). As they are obviously distinct from Vega Gull east of the Lena, and within the original range described for birulai, it also appears best that these should continue to be simply considered as birulai: thus again maintaining a valuable descriptive tool. Perhaps more variable than presently suspected, including yellow-legged individuals (contra Kennerley et al 1995) they might explain most "intermediate-looking" gulls.

Accepting this existing taxonomic status quo also means that the term vegae then needs to be used only for describing invariably pink-legged, dark-mantled gulls that nest east of the Lena.

Again, it is beyond the scope of this paper to suggest changes to informed taxonomic opinion prevalent in Far East Asia. Indeed, the presence of pale juveniles and first winters that appear lighter in build than many vegae might well refer to more northern populations of birulai (that show the lightest mantle coloration as adults). That such first winters tend to show heavily notched tertials, barring across the greater coverts, and extensive pink on the bill by January suggest a distance from taimyrensis and heuglini, and a greater closeness to vegae. Again following Kennerley et al (1995) and Hoogendoorn et al (1996), it seems appropriate therefore, in the absence of new information, to retain birulai as the western (and northwestern) subspecies of the Vega Gull, thus L. vegae birulai.

If such a taxonomic arrangement is maintained, the assertion in Yésou that "differences in the mtDNA of smithsonianus and vegae…concur to show that these taxa are not closely related (Crochet 1998, Crochet et al 2002)" (Yésou, 2002), becomes somewhat less certain: the vegae studied were from the range of birulai, phenotypically quite distinct from most vegae. Indeed, if identification is correct, vegae can show a number of features that are often cited for smithsonianus. In the case of juvenile/first winters for example they can appear very dark below, with mostly dark tails, and become paler-faced by mid-winter, with extensive pink to the base of the bill.

Yésou (2001) outlines the distinctiveness of mongolicus from cachinnans and barabensis, and concludes that it should be given species status under the Phylogenetic Species Concept (PSC) on the basis of wing-tip pattern and winter-plumage marks, and under the Biological Species Concept, as it does not interbreed freely to any significant degree.

By 2002, he had adopted a more conservative stance, provisionally including it in the newly coined East Siberian Gull, as L. vegae mongolicus.

It can be stated confidently, based on extensive field work within the region that the vast majority of Type A Mongolian Gull (considered to be mongolicus on phenotypic details outlined in the descriptions section above) can be readily separated from vegae in all non-breeding plumages, from juvenile to adult. Moreover, mongolicusshows a clearly different distribution in South Korea, being most numerous in freshwater habitats (where it is often both dominant in terms of abundance and in its relations with other gulls of the Assemblage), while it is much scarcer in open estuarine habitats preferred by vegae: in mid-winter in the Nakdong estuary for example, it typically is outnumbered 10 to 1 or more by that taxon. This differing ratio of occurrence/habitat selection is further reinforced by differing tendencies in feeding behaviour (which nevertheless require much greater quantification). Even when both taxa are together, mongolicus is usually much more aggressive in its harassing of other waterbirds, either actively killing small wildfowl, or stealing fish from sawbills. Such differences clearly support the description of Mongolian Gull under the PSC at least as a distinctive species. Following Yésou (2001), this should then be referred to as L.mongolicus.

However, there are many questions that remain to be answered.

Significant is the suggestion outlined above that there are several recognizable types that exhibit many mongolicus-type elements including: overall structure; white-headedness in mid-winter; long, parallel-sided bills; narrow and long-winged profile in flight, and extensive black on outer primaries.

Such forms, if correctly attached to Mongolian Gull, appear to considerably widen the phenotypic description of Mongolian Gull. As such birds appear not to have been noted by Yésou in colonies visited in Lake Baikal or by birders in Hong Kong (where interest in wintering gulls is comparatively long and deep) it seems that they do not appear randomly, and are therefore not part of expected phenotypic variation within given known populations. Considering the Mongolian Gull’s "patchy breeding distribution, ranging from south-eastern Altai to north-eastern Mongolia and the western part of north-eastern China" (Yésou 2001), extending perhaps to Lake Khanka at the border Far East Russia and north-easternmost China and islands of the Yellow Sea, it might be considered plausible that several phenotypically distinct populations, either subspecies or even hybrid populations, have evolved, some with mantle coloration darker than vegae, and others with structure intermediate between mongolicus and vegae.

There are three clearly recognizable main forms of the "Herring Gull Assemblage" which are widely but not uniformly distributed throughout South Korea, between October and March.

Based on structure, plumage details and available published and unpublished information, these can be identified as Taimyr, Vega and Mongolian Gulls. It is asserted that each of these three main groups show consistent differences in juvenile and first winter plumages, and can also be identified as non-breeding adults.

In the absence of any new information to the contrary, this paper therefore follows Kennerley et al (1995) and Hoogendoorn et al (1996) in maintaining taimyrensis as a subspecies of Heuglin’s Gull; of recognizing the validity of birulai; of recognizing Vega Gull L. vegae as a distinct species; and of incorporating birulai as a subspecies of Vega, L.vegae birulai.

In addition, it follows Yésou (2001) in recognizing mongolicus as a distinct species, L. mongolicus, but considers that this taxon might not be monotypic.

Note:

* It is important that discussions on the non-breeding distribution of these gulls is not confused by misuse of terms describing geographical regions. South Korea and Japan are best considered as North-east or Far East Asia, and not South-east Asia as stated in Yésou (2002), where most of the Assemblage are scarce or not reliably recorded: Robson, (2000)

Acknowledgements:

Many thanks to all those who have shared time, opinions and photographs, either directly in Korea, or indirectly by the establishment of excellent websites focused on gulls (most especially, Kim Hyun-tae’s, Martin Reid’s and Osao and Michiaki Ujiharas’); also to Ujihara san, and Peter Kennerley, for their review of an original draft; Arnoud van Den Berg and Magnus Robb for access to many of their images; and to Geoff Carey for various comments on gulls in Hong Kong and South Korea.

The Ujiharas’ at: larus.hp.infoseek.co.jp/gullidentifi_.htm (Authors of the exceptionally well-illustrated Birder Special: Gull Species Guide [in Japanese] ISBN 4-8299-3018-7 c0045 p2500E)

References:

• Dierschke J., & F. Heintzenberg. (1994). Happy Island and Beidaihe Bird Report. Unpublished.
• Grant, P. (1986). Gulls. A Guide to Identification. T & AD Poyser.
• P. Hoogendoorn, T. & M. Chalmers. (1995). Identification and Systematics of Large White-headed Gulls in Hong Kong. Hong Kong Bird Report 1994: 127=156. Dec 1995.
• Hoogendoorn W., Moores N., & T. Morioka. (1996). The occurrence and field identification of adult "Herring Gulls" with yellow legs in Japan. Birder 4: 64-73 (in Japanese).
• Lee, W., Koo, T. & J-Y Park.(2000). A Field Guide To the Birds of South Korea. Evergreen Foundation.
• Moores C. & N. Moores.(2002). Birds of South Korea. Video. Charlie Moores Video Productions.
• Robson, C. (2000). A Field Guide to the Birds of South-east Asia. New Holland Publishers (UK).
• Yésou, P. (2001). Phenotypic variation and systematics of Mongolian Gull. Dutch Birding 23: 65-82.
• Yésou, P. (2002). Systematics of Larus argentatus-cachinnans-fuscus complex revisited. Dutch Birding 24 (5): 271-298.