Читать книгу Gulls - Professor John C. Coulson - Страница 8
CHAPTER 1 An Overview of Gulls
ОглавлениеGULLS ARE CONSPICUOUS WEB-FOOTED, long-winged, medium or large seabirds that are readily recognised by the public. Adults are mainly white with shades of grey or black on the mantle and wings. Most species have black wing-tips, some have white ‘mirrors’ within the black areas, but a few species – mainly those restricted to an Arctic breeding distribution – have entirely white wing-tips. In the breeding season, adults of different species either have entirely pure white or very dark (black or brown) heads, and all revert to white heads in the autumn and winter, often with small grey marks behind the eye or grey streaking on the neck.
Gulls are widely known to the public because of their size and the habit of many species to frequent harbours, follow ships, visit landfill sites and visit outdoor areas also frequented by humans, such as seaside resorts, sports fields, beaches, rivers, picnic areas and large car parks at shopping complexes. In recent years, they have become even more familiar in parts of Europe and North America because their numbers have increased and several species have taken to nesting on buildings in urban areas. This habit of urban breeding has developed independently several times in different countries during the twentieth century and has spread rapidly. Urban nesting is now occurring in several species, and has almost certainly arisen through the marked increases in the size of gull populations, coupled with the increased protection given to them over the last century. The presence of gulls in urban areas has been given considerable adverse publicity, including reported cases of adults protecting their unfledged young by diving close to people’s heads, or of gulls snatching food from unsuspecting members of the public. Such reports have resulted in gulls, and particularly the large species, acquiring pest status in certain areas.
EVOLUTION OF GULLS
The Charadriiformes constitutes a single large and distinctive lineage of modern birds, and includes waders, skuas, auks, terns, gulls and a few apparently aberrant species such as jacanas. Although the skuas appear to be similar to gulls, current evidence – including DNA studies – suggests that their ancestry may be nearer to the auks than to gulls.
The lightly built bones of birds associated with flight are fragile and therefore do not often produce good fossil remains. As a result, the evolution of present-day birds is poorly known, and much less so than that of reptiles, fish or mammals. Fossil remains attributable to gulls are particularly scarce. Many of those that have been found have not or cannot be attributed to the presently recognised genera, and certainly not to present-day species. More recently, several fossil bones initially attributed to gulls have been found to belong to other avian groups.
Fossil bones attributed to gulls and possibly members of the genus Larus have been reported both in Europe and the USA from deposits from the Middle Miocene, 20–15 million years ago. The relationship of these fossils to modern-day gulls is unclear; fossilised bird bones are often, and perhaps uncritically, given different specific names to those of currently existing species, ignoring the fact that the bones of a present-day species vary considerably in size according to sex and locality.
TAXONOMY OF GULLS
Initially, gull species were separated and identified on the basis of plumage, skeletal structure and size. In many species, specimens from different geographical areas held in museums and private collections were often named and given subspecies status based on minor differences in size and plumage, but all too frequently this relied on small numbers collected from only a few localities. Many of these named subspecies are still used today and, while the majority are probably justified, others that were described and named many years ago should be re-evaluated using modern techniques and larger samples. Some subspecies have already been rejected on this basis, and it is likely that others will not stand critical re-examination and will also be rejected.
In other cases, some existing subspecies have been promoted to the status of a full species, as has occurred recently within the Herring Gull complex in Europe and Asia. Still others may show only gradual changes in size, structure or plumage shades over their geographical range, a concept not recognised by early taxonomists until Julian Huxley applied the term clines to these groups in 1942. Such clines have already been demonstrated for the Black-legged Kittiwake (Rissa tridactyla), the Puffin (Fratercula arctica) and the Common Guillemot (Uria aalge) breeding in the North Atlantic. Questionable subspecies of gulls still exist, and some are discussed in more detail later in this chapter.
Initially, the eighteenth-century taxonomist Carl Linnaeus placed all gulls in the genus Larus, and most species remained there in what became a very large taxon. Eventually, the two species of kittiwake were removed from Larus and placed in the genus Rissa, while the Ivory Gull was moved to the genus Pagophila (P. eburnea), Sabine’s Gull was transferred to the genus Xema (X. sabini), and Ross’s Gull was placed in the genus Rhodostethia and then, more recently, to Hydrocoloeus (H. rosea), alongside the Little Gull (H. minutus). The Swallow-tailed Gull became the sole species in the genus Creagrus (C. furcatus). These separations were not unreservedly accepted, however, and as late as 1998, Philip Chu proposed returning all gulls to a single genus, Larus. Seven years later, the intensive study of the mitochondrial DNA of many gull species made by Jean-Marc Pons, Alexandre Hassanin and Pierre-Andre Crochet (2005) moved in the opposite direction and separated gulls into nine genera, and in so doing created the new genera Chroicocephalus (with 10 species worldwide), Hydrocoloeus (with two species) and Saundersilarus (comprising only Saunders’ Gull, S. saundersi, in China). Worldwide, at least 24 gull species, especially those with white heads in the breeding season, are still retained in the large genus Larus. Aside from Saundersilarus, three other genera are composed of only one species: Creagrus, containing the Galapagos Islands’ Swallow-tailed Gull; Xema, containing the High Arctic Sabine’s Gull; and Pagophila, including the Ivory Gull, also breeding in the High Arctic. Like Hydrocoloeus, the genus Rissa also contains two species.
One of the major findings made during the in-depth investigation by Pons et al. (2005) was that the gulls that had dark heads as adults did not form a single taxonomic group, as had been suggested by studies made in the second half of the twentieth century, but were composed of three distinct groups of species. These groups were called the ‘black-headed gulls’ and placed in the genus Ichthyaetus, while ‘hooded gulls’ were separated into another new genus, Leucophaeus. The third group, including the Black-headed Gull, were called ‘masked gulls’ and were placed in the genus Chroicocephalus. To an extent, this separation of gulls with dark heads in the breeding season is supported by similar courtship behaviour within each group, as originally suggested by Niko Tinbergen and his co-workers in the 1950s and supported by more extensive recent studies.
GULL SPECIES WORLDWIDE
Currently, there are about 50 species of gulls in the world. This total has increased in recent years and will probably be increased further as improved molecular techniques are used to revise their status; even the definition of a species may be modified or revised. The uncertainty about the precise number of species reflects the fact that the gulls as a group present a taxonomic nightmare, and this has resulted in years of confusion and disagreement. For example, the American Ornithologists’ Union (AOU) considers the Herring Gull breeding in North America to be a subspecies of the European Herring Gull (Larus argentatus smithsonianus), while the British Ornithologists’ Union (BOU) regards it as a separate species, Larus smithsonianus. There is still much confusion within the extensive Herring Gull and Lesser Black-backed Gull complex of species and subspecies, particularly those occurring in Asia (see box).
Speciation concepts
The decision as to whether and under what circumstances two populations of animals that occur in different geographical areas can be considered distinct species remains a taxonomic problem, because the level of genetic difference between the two that justifies specific status is often an arbitrary decision and one that is not always universally accepted.
One major taxonomic problem relates to the Herring Gull and Lesser Black-backed Gull complex of subspecies. In 1942, the evolutionary biologist Ernst Mayr suggested that the subspecies formed a chain around the northern hemisphere, starting with the Lesser Black-backed Gull in Europe, and then further subspecies occurring eastwards through Asia, each having progressively lighter-coloured wings and leading to the American Herring Gull in North America, and finally completing the ring with the Herring Gull of western Europe. The theory is that, by the time this chain of subspecies has spread eastwards around the northern hemisphere and the ends meet up again in Europe, the Lesser Black-backed Gull and the Herring Gull are obviously separate species and interbreed only very rarely. This beautiful explanation of a series of subspecies first spreading, then part of each becoming isolated, allowing the formation of further subspecies around the northern hemisphere and eventually producing two distinct species, was widely acclaimed and has been frequently quoted in books, scientific papers and lectures on genetics and speciation.
However, the recent development and application of mitochondrial DNA techniques has shown that the American Herring Gull is not the closest relative of the European Herring Gull as was previously thought, nor did it spread eastwards historically from North America to Europe to evolve into the European Herring Gull. That said, while the fascinating concept of a ring of gull subspecies spreading around the northern hemisphere and ending with two distinct species has been discredited, it may soon, with some minor modifications, become viable again. This is because the Lesser Black-backed Gull is currently spreading from Europe to North America via Iceland and Greenland, and is beginning to breed in Canada. As such, it is establishing new end points of the chain of subspecies, this time in North America and involving the American Herring Gull.
The Iceland Gull (Larus glaucoides), which breeds in Greenland and parts of arctic Canada, has entirely white primaries and is a well-established species, but there is conflict over the status of two similar gulls, Thayer’s and Kumlien’s gulls, both of which show some black on the tips of the primaries. The AOU recognises Thayer’s Gull as a distinct species (L. thayeri), but regards Kumlien’s Gull as a subspecies of Thayer’s Gull (L. thayeri kumlieni). Within Europe, there is considerable disagreement about the status of the three forms, with some national bodies (such as those in Ireland) agreeing with the AOU classification, and others (including the BOU) considering both Thayer’s Gull and Kumlien’s gull as subspecies of the Iceland Gull (L. glaucoides thayeri and L. g. kumlieni). Yet other bodies believe that they represent three distinct species. In this book and without strong conviction on the matter, I have treated Thayer’s Gull as a distinct species but have followed the BOU and regarded Kumlien’s Gull as a subspecies of the Iceland Gull. Fortunately, most individuals that visit Britain are typical Iceland Gulls and lack any black or brown on the wing-tips.
Elsewhere in the world, birds in the Kelp Gull and Dominican Gull complex (currently all known as Larus dominicanus) are similar and obviously related to the Great Black-backed Gull (L. marinus) of the North Atlantic. They are also a taxonomic problem and currently are regarded as consisting of five geographically separated subspecies. Just as some of the former subspecies of the Herring Gull have been recognised as distinct species, some of the L. dominicanus subspecies may also be elevated to species status when more intensive genetic and ecological investigations have been completed.
CURRENT GEOGRAPHICAL RANGES
Gulls breed on all continents, with Kelp Gulls extending their southern range into Antarctica and several gull species breeding in the High Arctic. The number of gull species breeding in each 10-degree zone of latitude varies considerably, with two peaks of abundance, one in each hemisphere (Fig. 1). In the northern hemisphere, the number of species peaks between 40°N and 60°N, and in the southern hemisphere, a smaller peak occurs between 20°S and 40°S. Few gull species breed in the tropics, Antarctica or the High Arctic regions. This variation in species abundance, particularly between the two hemispheres, correlates reasonably closely with the amount of land within each latitude zone. This may offer a partial explanation as to why appreciably fewer species of gulls are found and breed in the southern hemisphere.
Few individual gull species breed over a wide range, with about 80 per cent spread over less than 20 degrees of latitude, and very few breed in both hemispheres. These patterns differ markedly from the terns, where many species breed in both the northern and southern hemispheres. A comparison of the ratio of gull and tern species breeding in different latitude zones throughout the world is shown in Fig. 2. There are more gull than tern species in only the zones north of 40°N, which begs the question as to why fewer gull species occur in the other zones. Could this be the result of competition between gulls on the one hand, and with petrels and shearwaters in the southern hemisphere?
FIG 1. The number of gull species (upper graph) and tern species (lower graph) breeding throughout the world in each zone of 10-degrees latitude. The distribution of gull species is clearly bimodal, while that for tern species peaks in the tropical zone between 30°N and 30°S. Tern data from Cabot & Nisbet (2013).
FIG 2. The ratio of the number of breeding gull species to the number of tern species in relation to zones of latitude. Gull species are more numerous than tern species only north of 40°N. The dashed line indicates equality.
The dominance of gulls over terns in temperate and arctic regions of the northern hemisphere is even greater when numbers of individuals are considered. For example, the average numbers of gulls per species breeding in Britain are very much greater than for terns. Using the figures from the national census in Britain and Ireland in 2000, there was a total of 1,810,000 breeding gulls (of seven species), but only 176,000 breeding terns (of five species), indicating a ratio of 10 gulls for every tern. Each gull species was represented by an average of seven times the average numbers for each tern species. Similar large differences are evident elsewhere in Europe and in North America.
The numerical dominance of gull species in the northern hemisphere suggests that much of their speciation occurred there. However, species of the genera Larus, Leucophaeus and Chroicocephalus breed in both the northern and southern hemispheres, indicating that in the past at least one species belonging to each genus must have spread, as breeding birds, across the equator on at least one occasion.
GULL HABITATS
The majority of gull species frequent coastal areas, marshes, rivers, estuaries and large inland lakes. Many occupy the same habitat zones used by marsh and sea terns, and in this respect they contrast markedly with shearwaters and petrels, which are pelagic. The smaller species often feed and breed inland, while the larger gulls breed mainly at coastal sites. Within the last hundred years, several species of large gulls have bred inland more frequently, a change in behaviour that has coincided with their overall increase in abundance.
Gulls breeding on the coast move only moderate distances from the shore. They are tied by relatively short incubation shifts and the need to feed their young frequently and regularly. In general, the density of gulls at sea tends to decline rapidly as the distance from shore increases, although the Kittiwake does not show this tendency in winter. Outside the breeding season, most gulls remain within daily flying distance of the shore, preferring to roost overnight on land or on sheltered coastal waters. The exception to this is when they are migrating. Only the two kittiwake species, Sabine’s Gull and Ross’s Gull, occur regularly in oceanic waters far from land throughout the long non-breeding season.
GULL SPECIES RECORDED IN BRITAIN AND IRELAND
The box is the current list of 26 species recorded in Britain and Ireland as breeding species, regular visitors or occasional vagrants. The list represents about half of all gull species in the world. Kumlien’s Gull is listed, but is retained as a subspecies of the Iceland Gull.
An approximate phylogenetic tree of the evolution of gull species recorded in Britain and Ireland (mainly based on the research by Pons and colleagues) is shown in Fig. 3 and involves eight genera. Such a representation can be only approximate, as their evolution has most likely been multi-dimensional and so cannot be presented accurately in two dimensions. There is still considerable uncertainty about the relationships between the species in the genus Larus, and no attempt is made in the order shown in Fig. 3 to indicate these, other than to suggest that the species with totally white wing-tips probably represent a distinct group.
Audouin’s Gull, which lacks a black head at any time, is placed in the same genus (Ichthyaetus) as two black-headed species on the British list (Mediterranean Gull and Great Black-headed Gull), along with three other black-headed species that occur elsewhere in the world, so its inclusion is surprising. Similarly, the Slender-billed Gull, which has a white head in all seasons, is included with the dark-headed Black-headed and Bonaparte’s gulls. However, Jean-Marc Pons in response to my query believes that ‘the dark hood is not a good character to construct evolutionary relationships because it has repeatedly been lost during the evolution of gulls’. In addition, he confirms that there is additional evidence indicating that the Black-headed and Bonaparte’s gulls should be included in the genera Ichthyaetus and Chroicocephalus, respectively.
The national censuses of gulls and other seabirds have been incredibly important and informative, and at last we have a sound knowledge of both the distribution and numbers of adults. However, we do not have a census value for numbers of immature individuals that have never bred for any of the gull species. Since immature birds may include up to five year classes (varying according to species), the numbers involved are appreciable and can be estimated only from a life table formed from survival rates obtained from detailed marking studies. Table 1 gives rough estimates of the proportion of immature gulls of the six commonest species occurring in Britain and Ireland. The figures are approximations but indicate that, by autumn, there is a large proportion (probably about 40 per cent) of individuals of every species listed that have not yet matured and bred. The proportions of immature individuals will have decreased by early spring because the mortality rate of young birds in their first year of life is usually markedly higher than that of adults, but they will still form an appreciable minority of the numbers of each species.
Gulls recorded in Britain and Ireland
| Genus Hydrocoloeus | |
| Little Gull (H. minutus) | Regular visitor, very occasional breeder |
| Ross’s Gull (H. roseus) | Vagrant |
| Genus Xema | |
| Sabine’s Gull (X. sabini) | Regular visitor, usually in small numbers |
| Genus Pagophila | |
| Ivory Gull (P. eburnea) | Vagrant |
| Genus Chroicocephalus | |
| Slender-billed Gull (C. genei) | Rare vagrant |
| Bonaparte’s Gull (C. philadelphia) | Vagrant |
| Black-headed Gull (C. ridibundus) | Abundant breeder and winter visitor |
| Genus Larus | |
| Common Gull (L. canus) | Common breeder and winter visitor |
| Ring-billed Gull (L. delawarensis) | Vagrant |
| Great Black-backed Gull (L. marinus) | Common breeder and winter visitor |
| Glaucous-winged Gull (L. glaucescens) | Vagrant |
| Glaucous Gull (L. hyperboreus) | Regular winter visitor |
| Iceland Gull (L. glaucoides) | Regular winter visitor |
| Kumlien’s Gull (L. glaucoides kumlieni) | Vagrant subspecies |
| Thayer’s Gull (L. thayeri) | Vagrant |
| European Herring Gull (L. argentatus) | Abundant breeder and winter visitor |
| American Herring Gull (L. smithsonianus) | Vagrant |
| Caspian Gull (L. cachinnans) | Currently vagrant but increasingly recorded |
| Yellow-legged Gull (L. michahellis) | Visitor and now an occasional breeder |
| Lesser Black-backed Gull (L. fuscus) | Abundant breeder |
| Slaty-backed Gull (L. schistisagus) | Vagrant |
| Genus Ichthyaetus | |
| Great Black-headed Gull or Pallas’s Gull (I. ichthyaetus) | Vagrant |
| Mediterranean Gull (I. melanocephalus) | Rapidly increasing breeder |
| Audouin’s Gull (I. audouinii) | Vagrant |
| Genus Leucophaeus | |
| Laughing Gull (L. atricilla) | Vagrant |
| Franklin’s Gull (L. pipixcan) | Vagrant |
| Genus Rissa | |
| Black-legged Kittiwake (R. tridactyla) | Abundant breeder |
FIG 3. A phylogenetic tree of the gull species that occur in Britain and Ireland. In general, the shorter the line leading from each species, the more recently that species is presumed to have arisen. This does not apply to the large genus Larus, where more work is necessary to elucidate the affinities of the different species. This diagram includes all of the genera in the world, with the exception of the genus Saundersilarus (where Saunders’s Gull, S. saundersi, is the only species) and the genus Creagrus (where the Swallow-tailed Gull, C. furcatus, is the only species), neither of which occur in Europe. The Vega Gull (L. vegae) has not been included, and its presence in Britain has yet to be confirmed, while Kumlein’s Gull is regarded as a subspecies of the Iceland Gull. Based in part on the work of Pons et al. (2005).
Table 1 also shows estimates of numbers of the six most abundant gulls in Britain and Ireland in about 2000. Again, these figures are estimates, but they indicate that for all six species combined there are about 3.2 million gulls of all ages in Britain and Ireland in the early autumn. This total is probably about 3.3 million in winter, when the numbers of departing Kittiwakes are replaced by immigrant Black-headed, Herring and Common gulls. By late spring, the winter visitors have departed and Kittiwakes have returned to their colonies, maintaining numbers at about 2.6 million just before breeding begins.
SUBSPECIES IN WESTERN EUROPE
There are only a few western European gull species that are represented by more than one subspecies. As mentioned earlier in the chapter, the Herring Gull was previously split into several named subspecies, some of which have now been elevated to species (Yellow-legged Gull, Caspian Gull). Two subspecies occur in Britain and Ireland: Larus argentatus argenteus, which breeds here; and the larger, darker L. a. argentatus, which breeds in northern Scandinavia and north-west Russia, and winters in Britain and the North Sea region. The nominate Common Gull subspecies, L. canus canus, occurs widely in western Europe, while the larger, darker subspecies L. c. heinei, which breeds in Russia, probably (based on ringing data) occurs occasionally in Britain, but is difficult to identify. In the Atlantic, the Kittiwake is represented by a single subspecies (Rissa tridactyla tridactyla); individuals are progressively larger towards the north of its range, but this gradual change does not justify separate subspecies status and is described as a cline. There are probably several other clines among gulls that are yet to be recognised, including the Black-headed Gull, Glaucous Gull and Lesser Black-backed Gull.
SURVIVAL AND LONGEVITY
Gulls are medium- to long-lived birds, with an average expectation of adult life and number of breeding years of different species ranging between four and 12 or more years. Annual adult survival rates for different species vary between 80 per cent and 92 per cent, and often vary appreciably from year to year and over longer periods of time. The annual survival rate tends to be higher in the larger gulls, which also have a longer period of immaturity. This delay in reaching breeding age in gulls appears to be associated with the time that is necessary to acquire competence in obtaining food, but why this should be longer in the large species is not evident. Ringed Herring Gulls that are nearly 35 years old have been recorded, but these represent a few extreme individuals comprising less than 1 per cent of those that reached maturity. In several gull species, the peak of mortality occurs during and just after the breeding season, when the adults are at their lowest weights during the year, suggesting that breeding is a significant stress. Data suggest that survival of gulls is usually high in winter, but the Kittiwake may be an exception, with most mortality occurring while the species is in its pelagic wintering range.
Less is known of the survival of immature gulls, but there is usually a lower survival rate in the 12 months following fledging, after which the survival rate approaches that of the adults.
The longevity records based on birds ringed as nestlings and living under natural conditions are given below, although several individuals are known to have lived longer in captivity.
| Mediterranean Gull | 22 years 1 month |
| Little Gull | 20 years 11 months |
| Black-headed Gull | 30 years 7 months |
| Common Gull | 33 years 8 months |
| Lesser Black-backed Gull | 34 years 10 months |
| European Herring Gull | 34 years 9 months |
| Great Black-backed Gull | 29 years 2 months |
| Black-legged Kittiwake | 28 years 6 months |
| Ivory Gull | 23 years 11 months |
| Laughing Gull | 22 years 1 month |
| Ring-billed Gull | 27 years 6 months |
| Glaucous-winged Gull | 23 years 10 months |
The species with the longest recorded lifespans are mainly those that have been ringed in large numbers, and therefore have more chance of an exceptional record. It should be kept in mind that these lifespans are reached by exceptional individuals – perhaps one in several thousand – and so it is very likely that the maximum known age of many gull species in the wild will increase in future years as more recoveries of marked birds accumulate.
SIZE DIFFERENCES BETWEEN SPECIES
Gull species vary considerably in size. The Little Gull is the smallest and weighs about 100 g (the weight of an Arctic Tern, Sterna paradisaea), while the largest is the Great Black-backed Gull, with males averaging 1,800 g and some individuals exceeding 2,000 g. Fig. 4 shows the average weights of adult females of 19 gull species on the British list. The weights of females of nine of these species are less than 400 g on average and overlap with terns, of which the adult females of all except one species on the British breeding list weigh under 400 g. The distribution chart for male weights is similar, but is shifted to the right because of their slightly greater size.
FIG 4. The average weight of the females of 19 gull species on the British list.
INDIVIDUAL VARIATION
Like all animals, individuals of each gull species show variation in many characters, including size, colour and age at first breeding. Males are larger than females and tend to have a more substantial bill, and size within a species can also vary geographically.
Variation in the immature plumage is widespread in gulls of the same age and this is frequently overlooked in the field identification of species, particularly within the genus Larus. The occurrence of hybrid individuals further adds to plumage variation and typical examples of hybrids can often be identified in the field by experienced observers; the characteristics used often overlap with those of other species. Consequently a proportion of immature and even adult birds that are infrequently recorded in Britain and Ireland may fail to be identified because of potential confusion with other species.
Immature plumages
The plumage, leg and bill colours of recently fledged chicks are very different from those of their parents, to such an extent that, many years ago, a first-year Kittiwake was claimed as a species new to science, despite the adult having already been described and named some years earlier. The first plumage of the young of most gull species is made up of feathers of varying shades of brown and grey, producing a cryptic pattern that helps to conceal them in vegetation in a colony and also appears to reduce aggression from adults. In the smaller gull species, the plumage is replaced by one that resembles the adult at the first annual moult; that is, when the bird is 13 months or so old. In the larger species, all feathers are replaced each year, but only some of the new ones resemble those of the adults and the full adult plumage pattern is not achieved until four years after hatching. These progressive changes in plumage at successive annual moults can vary between individuals and produce a series of different plumage patterns that make it a challenge to identify both the species and the age of immature individuals.
The slow and progressive acquisition of the adult plumage through successive moults contrasts with the rapid growth of bones and wing feathers, which reach full size within a few weeks after hatching because they are necessary before flight can be achieved. Why the acquisition of the adult plumage takes longer in the larger species of gull than in the smaller ones is not clear. The mechanism determining plumage patterns is obviously controlled by hormones and is linked with the greater length of immaturity in the larger gull species, but it is not evident why the large species delay reaching maturity for so long.
Differences between the sexes
The sexes of gulls have identical plumage and differ only in that females are usually smaller and tend to have slightly less substantial bills. Fig. 5 shows the relationship between the differences in weight of male and female gulls of seven species, using data from different parts of their geographical ranges where available. The extent of the difference between sexes is not constant from species to species, but increases with the weight of the species, ranging from a 5 per cent increase in the Black-headed Gull to more than 20–25 per cent in large species.
Standard measurements of wing length also tend to be longer in males than in females, but the magnitude of the difference is much smaller, ranging from 1 per cent in some small species to 6 per cent in the largest species (Fig. 6). Even when this difference is converted to wing area, it still results in the wing loading being higher in the large gulls, which explains why these species typically have a more laboured flight, with a slower, more powerful wing-beat. The small gull species, which are of similar weight to many tern species, have a characteristic buoyant flight similar to that of terns.
FIG 5. The percentage by which male gulls of several species are heavier than females, based on data for seven well-studied species. It is evident that there is a much greater difference between the size of males and females in the larger species of gulls.
FIG 6. The relationship between adult weight and the extent to which the male has a longer wing than females. The percentage difference in wing length between the sexes increases in heavier species, but is much less than the difference of body weight shown in Fig. 5.
The reason why there is a greater size difference between the sexes in large gull species is not known, and currently it is possible only to speculate. Perhaps there is a greater need in the large species to reduce competition for food between the sexes, or perhaps the dimorphism is related to the greater need for males of large species to defend nesting territories. The reader might speculate further, bearing in mind that in the skuas, females are invariably larger than males, while male terns are only 1–3 per cent heavier than females.
Despite the average size differences, there is an overlap in the range of sizes of male and female gulls. Niko Tinbergen claimed that, despite the overlap in size between the sexes of Herring Gulls, invariably the male is larger in every pair. Because of the average difference in size between the sexes, by chance the male will be larger than the female in many pairs, and more so in the larger gull species, but I have not found evidence that the male is invariably larger than the female. Size, and particularly the size of the bill, may play a part in individual birds recognising the sex of other gulls, but it is more likely that behaviour – particularly during courtship – plays the major role in sex recognition in gulls, especially in smaller species.
Sexing gulls
As male and female gulls have identical plumage features, distinguishing them in the field can be very difficult. The most reliable way to determine the sex of an individual bird – without killing it and then dissecting it to examine the gonads – is by carrying out a DNA analysis on samples obtained from feathers or blood. While this method is highly efficient, it is time consuming and it is expensive when large numbers of birds are being studied. In the field, biometric measurements taken while a bird is temporarily captured for ringing can also be used for sexing the individual. I found that the best measure was the head and bill length (from the back of the head to the bill tip), which also had the advantage of showing the highest degree of consistency when measured by different people. Further, the proportionate difference in head and bill length between the sexes is almost twice that for wing length (for example, 9.6 per cent compared to 5 per cent in the Great Black-backed Gull). The only disadvantage of this measure is that in some museum specimens part of the back of the skull was removed during preparation, which prevents it being used in these cases. As shown in Table 2, the head and bill measurement is satisfactory in sexing 92–98 per cent of individuals of several gull species. Including two other body measurements (wing length and bill depth) in a discriminant analysis increased the accuracy of sexing only by less than 2 per cent points.
TABLE 2. The head and bill lengths of adults of six species of gulls in Britain, the measure separating the sexes and the proportions sexed correctly by this single measurement. The data are based on samples of at least 80 individuals of each sex breeding in Britain, except for Common Gulls (Larus canus), which were captured in winter and so were from unknown breeding areas. Based on Coulson et al. (1983a) and additional data.
When a group of breeding gulls is being studied, the behaviour of marked individuals can be used as a reliable method of sexing. Copulation is totally reliable in this respect, as is courtship feeding of the female by the male and intensive food begging by the female. More details on the methodology used to sex gulls are given in Chapter 12.
Adult plumages
There is considerable variation in the shade of grey on the wing and mantle in adult gulls of the same species, which is evident in birds nesting in the same colony. This is illustrated in Fig. 7, which shows the extent of such variation in Herring Gulls breeding on the Isle of May in Scotland (subspecies Larus argentatus argenteus) and in northern Norway (subspecies L. a. argentatus). Because of the variation, there is overlap in wing shades between the two subspecies of Herring Gulls and most, but not all, individuals can be identified on this basis alone (see also box). Even using more measurements of body size does not completely separate all argenteus males from argentatus females.
In Lesser Black-backed Gulls breeding in the Netherlands (Fig. 7), there is also considerable variation in wing shade, with the palest approaching the darkest shade of Herring Gulls breeding in northern Norway. The darkest shade reported in Lesser Black-backs in the Netherlands is said to fall within the shade range of the subspecies Larus fuscus fuscus, which breeds in eastern Scandinavia and typically has a black mantle and wings very similar to those of the Great Black-backed Gull. The majority of Lesser Black-backed Gulls breeding in the Netherlands have a range of shades found in both the subspecies L. f. intermedius (breeding in north-west Europe) and L. f. graellsii (breeding in Britain).
Identification of the Lesser Black-backed Gull subspecies intermedius and graellsii in the field is further complicated by whether the individual is seen in bright sunlight or under dull conditions, and also by the direction of the light, all of which affect the apparent shade of grey of the same individual recorded in photographs or observed in the field. Reliable records of shade need to be measured with the bird in the hand, using standard lighting and comparing it against a reliable shade chart, but even this would not identify the subspecies of all individuals.
FIG 7. The shade of the wings of adult (a) Herring Gulls (Larus argentatus) breeding in Britain (n = 1,591) and (b) northern Norway (n = 140), and (c) Lesser Black-backed Gulls (L. fuscus) breeding in the Netherlands (n = 899). The shades increase in darkness from left to right and correspond approximately to shades of grey, which range from 1 (white) to 20 (black). The data for Lesser Black-backed Gulls are taken from Muusse et al. (2011).
Differences in wing pigmentation
A major source of plumage variation within gull species is the pattern of black or brown pigmentation on the wings of immature individuals. These show progressive changes at each annual moult, until adult plumage is eventually achieved. In addition, this patterning varies appreciably between individuals of the same age, even within a single colony. For example, fully adult Herring Gulls in the same colony showed a range in the number of primaries that are tipped with black pigment (Table 3), and they also showed variation in the extent of white on the tip of the longest (10th) primary. Part of this variation is linked to the age of the birds (Table 4), with change continuing for several years after individuals reach breeding age, but the patterning is not related to gender.
TABLE 3. The percentage of fully adult Herring Gulls (Larus argentatus) examined in breeding colonies with different numbers of black-tipped outer primaries. Dutch data from Muusse et al. (2011), Norwegian data mainly from Barth (1975).
TABLE 4. The wing-tip pattern in Herring Gulls (Larus argentatus) of known age breeding in colonies in Britain. The extent of white on the tip of the 10th primary tends to increase with age.
Part of the variation is also geographical, as seen in adult Herring Gulls breeding on the east and west sides of England and Scotland (Table 5). This difference between east and west in the black-and-white patterning on the primaries is maintained in the winter, presumably because the north–south dispersive movements of British Herring Gulls mainly follow either the eastern or the western coastlines, with relatively few individuals crossing between the two coasts. Figs 8 and 9 illustrate the variation in two adult Herring Gulls captured in north-east England, with one showing the thayeri-type pattern on the ninth primary (second from left), where the black does not spread across the whole of the width of the feather.
TABLE 5. Comparison of wing-tip patterns of adult British Herring Gulls (Larus argentatus) breeding on the east and west coasts of England and Scotland. The differences in both characters are significant.
FIG 8. The wing-tip pattern of an adult Herring Gull (Larus argentatus) captured in north-east England, showing what is known as the thayeri-type pattern on the ninth primary. The bird was subsequently found breeding in northern Norway. (John Coulson)
FIG 9. A typical wing-tip pattern of a Herring Gull (Larus argentatus) breeding in Scotland, with the 10th primary still growing. (John Coulson)
Differences caused by hybrids
One definition of a biological species is that the individuals can form a group of interbreeding or potentially interbreeding organisms that produce viable offspring. If hybrids occur between two species, such as in the classical case of a horse and a donkey, the hybrid offspring of the two (a mule or hinny) are usually sterile, probably because the two parent species have different numbers of chromosomes. In the case of gulls, hybrids are not uncommon and are reported far more frequently than in major taxa, such as terns. Studies have revealed that Larus gulls have the same number of chromosomes (72) and as a result, hybrids are usually fertile and they have been reported breeding successfully. There are now numerous records of gulls of different species and even different genera pairing and rearing hybrid offspring, as listed below:
Mediterranean Gull × Black-headed Gull
Herring Gull × Lesser Black-backed Gull
Herring Gull × Yellow-legged Gull
Lesser Black-backed Gull × Yellow-legged Gull
Western Gull (Larus occidentalis) × Glaucous-winged Gull
Great Black-backed Gull × American Herring Gull
Herring Gull × Glaucous Gull
Glaucous-winged Gull × Glaucous Gull
American Herring Gull × Kelp Gull
Iceland Gull × Thayer’s Gull
Common Gull × Ring-billed Gull
Mediterranean Gull × Common Gull
Laughing Gull × Black-headed Gull
Laughing Gull × Ring-billed Gull
Herring Gull × Caspian Gull
In most cases, adults that are believed to be hybrids have been recognised by the intermediate nature of their plumage and the colouring of their legs and bill, but in only a very few instances has the plumage been described for adults that are known to be hybrids and were ringed as such before they fledged. It is usually believed that hybrid gulls show intermediate characters of their parents in terms of plumage, bill colour and leg colour, but this is not always the case, and in several instances they display minor characteristics not evident in either parent.
There is little doubt that some hybrids can share similarities with, and resemble, other gull species. As a result, it is sometimes difficult to accept a new sight record of a species from a geographical area where it has not been previously or convincingly been recorded before, and to confirm that it is not a hybrid between species that breed nearby. Rarity committees have a particularly difficult job with gulls, and ideally need DNA samples obtained from feathers of the presumed rarity to be certain of the record.
Some hybrid gulls, when adult, have been known to pair and mate with an individual of one of their parent species, producing offspring known as back-crosses. Even less is known about the plumage of these offspring, but it is likely that they differ both from the original species and from the hybrid parent. Breeding between pairs of hybrid gulls has not been recorded. However, hybridisation and subsequent breeding is likely to produce at least three different types of individuals, all of which vary in some respect from the original parent species as well as from each other. The immature plumages of hybrid gulls are poorly known and in many cases their origin has been assumed only because of their intermediate characteristics.
Eventually, after several generations of breeding, a particular gene can be transferred via the offspring of a hybrid from one of the parent species to the other. This has been recorded in the American Herring Gull, which appears to have acquired a gene from the Great Black-backed Gull in North America, presumably through hybrids between the two species. To date, this gene has not been recorded in the European Herring Gull.
In Belgium and the Netherlands, mixed pairs of Yellow-legged Gulls and either Lesser Black-backed Gulls or Herring Gulls have occurred particularly frequently. For example, more than 15 mixed pairs were reported in Rotterdam annually from 1986 to 1998 (van Swelm, 1998) and more in more recent years, and others have been frequently identified in at least five other colonies in the Netherlands and Belgium. Hybrid individuals that have reached adulthood and that are presumed to be crosses between Yellow-legged and Lesser Black-backed gulls have also been recorded in Belgium breeding with Lesser Black-backed Gulls, producing back-crosses.
Inter-species breeding is more frequent when one of the gull species is rare and spreading into the main range of the other. For example, when Lesser Black-backed Gulls first started to breed in the Netherlands in the 1930s, a few individuals joined large colonies of Herring Gulls and several mixed breeding pairs were recorded. Despite the fact that both species are now numerous and breed in the same colonies, hybrid pairs still occur, although they are infrequent. Very few pairings between these two species have been reported in Britain, except when experimentally induced (see below).
When Herring Gulls spread to Iceland in the 1920s, individuals formed mixed pairs with Glaucous Gulls, and by 1966 about half of the adults there were considered to be hybrids. These were distinctive in showing small but variable amounts of dark pigment on the tips of the primaries (Ingolfsson, 1970).
When Mediterranean Gulls first started to breed in Britain, early pioneers frequently paired with Black-headed Gulls (as they have done so elsewhere). In fact, this is ongoing, as a few individuals continue to spread north from the south coast of England. The recent arrival of a few adult Yellow-legged Gulls in Britain has seen them join both Herring Gull and Lesser Black-backed Gull colonies. Again, they have formed mixed pairs that on some occasions have managed to fledge hybrid chicks. Perhaps this inter-species breeding occurs because individuals arriving in new areas are mainly of the same sex and fail to find a mate of the same species.
In a study carried out on the island of Skokholm in south-west Wales, Mike Harris (1970) switched large numbers of eggs between Lesser Black-backed Gull and Herring Gull nests. The chicks that subsequently hatched imprinted on their foster parents and apparently considered that they were the same species, so that when they matured they chose a mate of that species, forming a series of mixed-species pairs. The young produced and reared by these mixed pairs were hybrids between the parent species and usually (but not always) showed plumage and leg colour intermediate between the two. At least 40 of these hybrids later returned to breed on Skokholm and on nearby Skomer, and most paired with adult Herring or Lesser Black-backed gulls. The chicks they produced were back-crosses and, when adult, were more similar to one of the parent species than the first generation of hybrids. While some of these hybrids reared chicks, it is not known whether they and their offspring were less viable. However, as each generation was produced, presumably both parent species incorporated small amounts of the genetic material belonging to the other species into their make-up despite appearing to be ‘pure’ Herring or Lesser Black-backed gulls (as discussed above for the American Herring and Great Black-backed gulls in North America).
BREEDING
Gulls are monogamous, although a few cases of male Kittiwakes breeding simultaneously with two females at different nest sites have been recorded. Pairs of gulls produce only one brood each breeding season, but if their eggs are lost, many will lay a replacement clutch. While most gulls breed annually during a well-defined breeding season, some individuals skip breeding for a year. The exception is the Swallow-tailed Gull on the Galapagos Islands, which does not have a clear-cut breeding season and nests throughout the year, with individuals breeding at nine- to 10-month intervals.
Breeding sites
Gulls typically favour bare ground and areas with short vegetation for nesting, or floating vegetation on lakes or marshes. The main exceptions are Bonaparte’s Gull, which regularly nests in trees; Common, Black-headed and American Herring Gulls, which occasionally nest in low trees at a small number of localities; Kittiwakes, which favour narrow ledges on steep sea cliffs; and Herring, Glaucous and Ivory gulls, which sometimes use larger cliff ledges.
Ground nesting makes gulls particularly susceptible to mammalian predators, and most species nest only at sites where these predators are usually unable to reach the colonies, such as small islands or isolated peninsulas. Gulls vary in their ability to deter avian predators. Adults will attack birds of prey and corvids, but in parts of northern Scandinavia White-tailed Eagles (Haliaeetus albicilla) are having an increasing impact on breeding gulls – this is a future risk for Britain, since the species has been reintroduced here and its numbers are increasing. In addition, adult gulls at breeding sites suffer occasional predation from Peregrine Falcons (Falco peregrinus). Ravens and crows are a problem for some smaller gulls, but in general they are attacked and prevented from entering dense gull colonies. Individual Herring, Lesser Black-backed and Great Black-backed gulls, as well as Great Skuas (Stercorarius skua), have developed the ability to reach and prey on eggs and young gulls at otherwise well-protected nesting sites (sometimes even attacking their own species).
Mammalian predators such as Red Foxes (Vulpes vulpes) and Badgers (Meles meles) have now reached some gull colonies in Britain after being absent for many years, and these and the spread of American Mink (Neovison vison) has often resulted in the sudden desertion of sites used by breeding gulls. This desertion may be immediate, while in other cases the decreases in numbers of adults are spread over several years, apparently because new recruits to the colony are deterred by the presence and activity of predators. In particular, Foxes have become very much more abundant in Britain in recent years, and have captured and killed many incubating gulls at night. Six gull species now also nest on buildings in urban areas (here); these have the same characteristics as natural sites, in that mammalian predators cannot normally reach them and they are generally given public protection.
Humans entering gull colonies are not usually attacked by the smaller gulls, which instead tend to fly overhead giving alarm calls. Large gulls do frequently dive at human intruders, however, usually from behind. While they pass closely overhead, they seldom actually strike. I have been struck only once by a large gull, although one of my students was knocked to the ground by a particularly aggressive Lesser Black-backed Gull defending its unfledged chicks.
Colony and nest site fidelity
Adults gulls are highly site faithful, provided that the nesting site remains safe and is not subject to high levels of predation on eggs, young or adults. Males – particularly those that were successful in rearing young in the previous breeding season – tend to return to the same nesting sites. In contrast, young birds are much less likely to return to the place where they were reared as chicks. A proportion – usually dominated by males – does so, and these birds often return to the same part of a large colony in which they were reared. This behaviour is called philopatry, the extent of which is influenced by many factors, including competition for nesting sites and food availability. In the past, the proportion of birds moving elsewhere has often been underestimated because of the much greater difficulty in locating those that have moved. In some species, the majority of the young that survive to breed move to other colonies, with some moving 100 km or more away. In several species, young individuals have been recorded visiting a number of colonies while approaching maturity, including their natal colony, only then to move and breed elsewhere. Such movements are, of course, necessary to form new colonies.
Colonial breeding
Colonial breeding is widespread among gulls, and only a few species regularly breed both as isolated pairs and in colonies. Some gull colonies are composed of mixed species, and the smaller species frequently nest in or alongside colonies of terns. In many gull species, it appears that single pairs cannot breed in isolation and the presence of a group of gulls of the same or even different species is necessary before egg-laying is possible, thus making colonial breeding essential. The main exceptions are Common and Great Black-backed gulls, where a proportion of breeding pairs nest in isolation from others. It is obvious that colonial nesting is not forced upon gull species as a result of shortage of suitable nesting sites, and it is usually regarded that there is an advantage to breeding close together. An obvious reason is that it improves defence of eggs and unfledged young against predation, but while this is evident in deterring Ravens (Corvus corax) and crows, it is less evident that colonial nesting prevents mammalian predators from raiding colonies and consuming eggs, young and even adults.
The reliance upon a group to ensure breeding suggests that stimulation from other individuals is necessary. This was first suggested in 1938 by Frank Fraser Darling, who noted that the display from neighbours within the colony stimulated pairs to breed and in larger groups or colonies to lay eggs earlier, and that it resulted in greater synchrony with neighbouring pairs. He suggested that the effect of this synchrony was that eggs or chicks were available over a shorter time period, and therefore fewer were predated and breeding success was enhanced. At the time, Fraser Darling’s idea appealed to some, but others were critical of the concept. Unfortunately, his original data on Herring and Lesser Black-backed gulls only hinted at this effect, and others later showed that the variations he found between colonies of different sizes could have been produced by chance and were not statistically significant.
In the 1950s, Edward White and I collected information on the timing of breeding in a series of colonies of Kittiwakes. We found that the laying season in large colonies was more spread out than in small ones, the opposite of Fraser Darling’s contention. While some pairs in the larger colonies started laying earlier than those in the smaller ones, pairs of late-breeding birds (which we now know were mainly young, first-time breeders) were recorded in all the colonies studied.
However, we did discover that breeding in smaller groups of 20–50 Kittiwake pairs within a single colony was more synchronous than in the colony as a whole, and that the average breeding date of each of these groups was earlier as the density of pairs increased. This finding suggested that social stimulation was occurring at the more local level, among groups of breeding pairs, and not in the colony as a whole.
I know of no single pair of Kittiwakes that has been recorded breeding in isolation. As a result, it is difficult for members of the species to form new colonies because this requires a group of individuals to be attracted to, and display at, a new potential breeding site before egg-laying can occur. Up to 20 pairs will collect at the site of a potential new colony for a year or more, and before a few of the pairs succeed in building a nest and laying eggs. However, in subsequent years the numbers breeding increase rapidly. In contrast, Herring Gulls sometimes nest in isolation, but these are birds that have bred before. Young Herring Gulls attempting to breed for the first time do not seem to be able to breed without joining a group already breeding together.
Colonial nesting allows gull species to benefit from the stimulation of their neighbours engaging in courtship activity, and probably encourages earlier nesting in some individuals. This was clearly evident in Kittiwakes when observing a pair reuniting at the colony after one member had been away on a fishing trip (Coulson & Dixon, 1979; Coulson, 2001). The birds engaged in a vocal display of calling, which immediately stimulated similar mutual calling in neighbouring pairs. Where Kittiwake nests were denser, the frequency of this stimulation from neighbours was more frequent. Anyone who has visited a Kittiwake colony will recognise the loud, periodic outburst of calling by groups of pairs as a major feature of the pre-laying period.
New colonies of Kittiwakes and Herring Gulls are highly attractive to potential breeders, and as a result they increase rapidly in size. Part of this effect is a result of the low density of breeding birds in a new colony, with relatively large areas available where recruits can settle without aggression from neighbours. As colonies grow, the area available for new recruits within, and at the immediate edge, decreases relative to their size, and their rate of growth decreases over time. Whether this eventually determines the maximum size for colonies of some species is unclear, but remarkably large colonies are known, numbering tens of thousands of pairs. In contrast, Great Black-backed Gull colonies rarely exceed 200 pairs in Britain (although much larger colonies are known in North America), while large colonies of Common Gulls do occur but are exceptional.
Mixed colonies of Herring Gulls and Lesser Black-backed Gulls are common, but the two species show a tendency to select slightly different areas within the colony. On natural sites, they often nest side by side, but the Herring Gull more frequently nests on open ground with little or short vegetation, while the Lesser Black-backed Gull shows a partial preference for areas with taller vegetation. At urban sites, both species frequently nest side by side and there is no obvious or consistent difference in the sites they select. Great Black-backed Gulls often nest in isolation or in small numbers of pairs in colonies of Herring Gulls or Lesser Black-backed Gulls, where they frequently select sites on raised ground.
Terns often nest in association with Black-headed Gulls, with the latter arriving and laying earlier and taking the first choice of sites. However, this is not always the case – the first Black-headed Gulls to nest on Coquet Island in Northumberland did so alongside Sandwich Terns (Thalasseus sandvicensis), and only arrived after the terns had laid.
Age at first breeding and mate fidelity
Within a species, the age at first breeding is variable and can span three or more years. Smaller gulls tend to breed when they are two or three years old, while many of the large gulls do not normally start breeding until they are four or five years old. This long period of immaturity often allows individuals to visit colonies for one or more years before they breed themselves.
The claim that gulls pair for life is not justified. While the same individuals frequently re-form as a pair in successive years, ‘divorce’ often occurs, with both individuals taking new partners. Change of mate also occurs when one partner dies.
Nests
In general, gulls’ nests are often substantial structures composed of plant material collected by the adults nearby or, in some cases, picked up either from the sea surface, on the tideline or from cliff tops. Gulls that breed on marshes and ponds probably start nest-building earlier than those on drier sites, and they produce a platform above the water level on which most of the displaying and courtship takes place. In situations where the water level rises after rain, there is often an attempt by the pair of gulls to build up the height of the nest to keep the eggs dry. However, this is effective only when relatively small increases in the water level occur.
On several occasions I have seen Kittiwakes flying along the coast carrying nest material as far as 15 km from the nearest colony. This contrasts with Herring Gulls, which usually collect plant material for the nest from within their territory, although if it is not available, such as on roofs of buildings, material is collected from further away. Having collected material, adults shape a central hollow in the centre of the nest by compressing the material with their breasts. This hollow is large enough to hold the usual two or three eggs of the clutch together and prevents the eggs from rolling away by accident.
While nests of most gulls are formed by an accumulation of plant material, the exception is the Kittiwake. The adults collect marine algae and mud to form a base for the nest, and as this dries, it becomes firmly cemented to the rock surface, forming a base on which the rest of the nest is then built. The nest is completed by bringing in grass to form the top. This species is exceptional in that the adults at the nest defecate onto and over the edge of the structure, leaving obvious white streaks on both the nest and the cliff face below, and thus clearly make no effort to disguise the nest. Other gull species move away from the nest to defecate. The Kittiwake also differs from other gulls in that nest-building is often highly social, with a hundred or more individuals repeatedly flying back and forth between a particular source of mud and vegetation and the nest sites. On one occasion when in spring a local farmer spread straw and dung onto a cliff-top field, within a week the nearby Kittiwake colony was entirely draped with long strands of golden straw.
Recently fledged young and immature one- and two-year-old Kittiwakes often pick up potential nest material from the tideline and fly about with it in their beaks, but eventually drop it without taking it into a colony. I have not seen immature individuals of other gull species doing this. Non-breeding three-year-old Kittiwakes often occupy sites in a colony and bring in small quantities of nesting material very late in the breeding season, when most pairs of Kittiwakes have half- or full-grown chicks. In some cases, this forms a base for a nest in the following year. The chicks of most gulls leave the nest when a few days old, seeking cover and protection, but the Kittiwake is again exceptional in that the chicks remain on the nest until they fledge five or more weeks after hatching.
Depending on the position of the colony, some Kittiwake nests are washed off in winter storms. However, those in more sheltered positions remain over winter and until the following breeding season when they are used as the base of new nests. One nest on the Farne Islands was supported in a vertical cleft in the rock face, and as material accumulated year after year, it reached a height of more than 100 cm.
Eggs
All gull eggs have a brown or grey background with numerous spots or streaks of darker pigment, offering a degree of camouflage if they are left uncovered in the nest. The extent to which camouflage prevents or reduces predation of gull eggs has not been investigated, but it can only be one of several factors affecting successful breeding.
The size of egg laid by gulls varies with the average weight of each species, but not in a simple linear manner (Fig. 10). At one extreme, the Little Gull lays an egg that, on average, weighs 18.6 per cent of the normal body weight of females, while the egg of the largest species, the Great Black-backed Gull, weighs only 7.3 per cent of the female’s body weight. The change in egg size as a proportion of the adult weight is a common effect in many bird groups and not just a peculiarity of gulls. Large eggs, irrespective of the species laying them, usually require a longer incubation period, although the reason for this is not immediately obvious. Presumably the body of a chick of a large gull species has more cells and requires more cell divisions before it is ready to hatch. However, that is not the only factor involved, since the eggs of birds in some other taxa, such as the Procellariiformes (petrels and shearwaters), require a much longer incubation period even when the eggs are of the same size.
FIG 10. The relationship between the average weight of adult females of 13 gull species and the average weight of the eggs they lay. The gull species that normally lay two rather than three eggs (Kittiwake, Rissa tridactyla, indicated by the lower of the two filled squares; and the Ivory Gull, Pagophila eburnea, indicated by the upper filled square) sit exactly on the curve calculated for the other species, which normally lay three-egg clutches, indicating that despite having smaller clutches they do not lay bigger eggs.
FIG 11. The relationship between the average weight of adult females of 13 gull species on the British list and the average weight of one of their eggs, expressed as a percentage of the female’s weight.
While the most frequent clutch size in gulls is three eggs, two species that occur in Britain typically lay two-egg clutches, namely the Kittiwake and Ivory Gull. This raises the question as to whether they respond by laying larger eggs. At first inspection, Kittiwake eggs look large, but this is an apparent effect of their paler background in comparison to the eggs of most other gulls (see here). Fig. 11 shows the relationship between female weight and egg weight for a series of species. The relationship is curvilinear, but the Ivory Gull and Kittiwake sit almost exactly on the curve, fitting the data based on the other gull species.
Each egg is laid at about two-day intervals. Usually, the last egg in the clutch is the smallest and tends to differ in shape, being longer and narrower than the rest. The infrequent cases of four eggs in a clutch are probably the result of a second female contributing one or more eggs. This can occur when female–female pairs are formed, caused by a shortage of males. Incubation lasts for more than three weeks and is longer in the larger species. Both members of the pair contribute to the incubation and feeding of the young.
Incubation period
Ideally, the precise measure of the incubation period is the length of time that the eggs are maintained at a temperature just below the adult’s body temperature of about 40 °C. There is a lower temperature (which may be as high as 25 °C) below which the development of the embryo ceases; in temperate regions, this is usually above the environmental temperature. As a result, development soon stops when the eggs are not covered by an adult and the temperature falls below the critical temperature for development. If the incubation of the eggs is frequently interrupted, the time taken for the eggs to complete development to hatching is lengthened.
In practical terms, the incubation period is usually measured as the number of days taken from the start of incubation to the time the egg hatches, but hatching takes several hours and confusion has existed as to whether the end of incubation is when the shell is first pierced by the chick or when it becomes entirely free of the shell. Covering the eggs by an adult gull often takes place before the final egg of the clutch is laid, but in some cases the vascularisation of the brood patches on the adult may not have become fully developed with supplementary blood vessels and the optimal temperature for embryo development may not be reached. Consequently, there may be a delay of a day or so until the eggs reach the critical temperature for development. As a result, the recorded incubation period for a species will vary. The incubation period of birds, and in gulls in particular, is slightly longer in the larger species, but the differences are small and the variation recorded between individual pairs is considerable (Table 6).
In gulls, the incubation of the eggs is often initiated by the female, but overall the sexes tend to share incubation equally in a shift system. Either sex may incubate overnight. In most cases, the incubating individual remains on the eggs until relieved by its partner, so that once incubation starts, the eggs are covered by an adult for about 90–95 per cent of the time.
TABLE 6. Estimates of the incubation period in eight well-studied gull species, presented in order of adult size (smallest first). Data mainly from Cramp & Simmons (1983), Fisher & Lockley (1954) and other sources where appropriate, including personal data for the Black-legged Kittiwake.
Periodically – perhaps five to eight times a day – the incubating adult will stand up and roll the eggs. Egg-turning is believed to facilitate the absorption of the albumen by the embryo and prevents the embryo from adhering to the shell membrane. In domestic hens, the failure to roll eggs in an incubator does not necessarily prevent chicks from hatching, although the success rate is reduced and the chicks often have a lower weight. Egg-turning is unique to birds and the reason why reptile eggs do not need to be rolled is as yet unclear.
In gulls that lay three-egg clutches, it is common for the third egg – which is laid two or three days later than the second egg – to hatch one or sometimes two days after the first two. This asynchronous hatching suggests that effective incubation in these cases starts during the normal two-day period between the laying of the second and third eggs.
Chicks
At hatching, chicks of most gull species have a patterned down that offers some cryptic protection; this is replaced by a pattern of dark feathers that helps to conceal the young birds when they attempt to hide under vegetation. The dark patterning at hatching is missing in Kittiwake chicks, which hatch with a pale grey down and then acquire white, grey and black feathering when older, all of which could also be regarded as aiding camouflage among the whitewashed nests and rocks on the cliffs where they nest.
An effect of the relationship between egg size and adult weight is that, at hatching, the chicks of larger gulls are smaller in proportion to the adults than those of small gull species, and they therefore must achieve more growth in order to reach their adult size and weight. This is presumably one of several reasons why the period of chick growth (and the fledging period) tends to be longer in large gull species than in small ones.
The time between the chick hatching and first flying is known as the fledging period and also shows much individual variation (Table 7). There is a weak trend for it to be longer in the larger species of gulls. The fledging period of the Kittiwake tends to be longer than in other gulls of a similar size. In the case of the Kittiwake, the first flight has to be totally successful from the cliff-nesting site and so is an all-or-nothing event, with the young bird either flying well enough to reach and return to the nest, or to achieve sustained flight immediately, leave the colony and start an independent life.
Fledging is often a gradual process and tends to take longer in the larger species. It is longer in Kittiwakes, where the young cannot fly until they are at least five weeks old and many are at least a week older before their first flight. Unlike most terns, parental feeding of young gulls often ceases at the time of fledging, but sometimes family parties of large gulls stay together for some weeks. During this time they visit feeding sites together and make loud contact calls while moving between areas, and they may even return to the colony in the evening. Exceptionally, a Herring Gull has been observed feeding chicks three months after they fledged.
TABLE 7. Estimates of the fledging period in eight well-studied gull species, presented in order of adult size (smallest first). Data mainly taken from Cramp & Simmons (1983), Fisher & Lockley (1954) and modified from other sources where appropriate. Note that the considerable variation in several species is genuine, and depends on the rate of growth and development of the chicks and whether there had been a stimulus to fly, for example, by the presence of a predator or the observer.
Breeding success
Under favourable conditions, breeding success of gulls is high, with about half the eggs laid producing young that fledged. However, lower success rates occur when there are food shortages or predation occurs. No gulls are known to breed as one-year-olds and they do so only when they have acquired almost adult plumage.
WALKING AND FLYING
Gulls have relatively short legs, and this is particularly true of Kittiwakes and Ivory Gulls. Gulls typically walk or run only short distances, and rapid movement beyond a few metres is usually achieved by flying. The exception to this occurs when gulls are searching for food on grassland and mudflats, when walking – with stops to pick up food – is the norm. Of all gull species, Kittiwakes walk less than any other. During an appreciable part of the year they are oceanic, either flying over or floating on the sea. Immature Kittiwakes that visit land during moult typically remain where they alight, and the adults fly directly onto their nests on precipitous sea cliffs without needing to take more than a step or two as they land. This behaviour is associated with the species having short legs (an adaptation to produce a low centre of gravity to enable the birds to withstand strong gusts of wind while nesting on sea cliffs), while the Kittiwake’s hind toe, which in other birds – including gulls – assists with walking, is reduced to a small protrusion.
The long, narrow wings of gulls, petrels, albatrosses and shearwaters are an adaptation for flight over water, particularly where the sea surface has peaks and troughs that affect wind movement. Many gulls fly with steady wing-beats, while at other times they introduce periods of gliding, between wing-beats. They utilise rising air currents, particularly those over the sea produced by waves or swell, although it is rare for them to fly more than a few metres above the surface of the open sea. In addition, they glide in thermals over land and cliffs, and those produced by buildings in urban areas. Several species of gulls use thermals to rise to considerable heights over land without the need for flapping flight, although the smaller species more rarely soar. Favourable conditions for soaring occur on warm, sunny days. Such apparent effortless flight can be seen in Herring Gulls that breed on the coast but visit the highest ground on Snowdon in north Wales, where they search for food discarded by people walking and climbing here.
MOVEMENTS
Gulls breeding in extreme northern and southern areas of the world are migratory, but this is usually less pronounced or even absent among those living in temperate or equatorial regions. In general, the distances gulls migrate are much shorter than those made by terns, with only Sabine’s and Franklin’s gulls making regular large trans-equatorial migrations. In other gull species, a few individuals may reach or cross the equator, but most remain in the hemisphere in which they breed or were reared. Whether gulls migrate mainly depends on whether food is readily available in winter in their breeding areas.
The seasonal movements of many gulls are better described as dispersal rather than migration, with individuals from the same area or even a single colony moving in different directions and for variable yet often short distances. Immature individuals tend to move further than adults and often do not visit their natal or other colonies where they will ultimately breed for several years.
MOULTING
In all gulls, the 11 primary feathers on the wing are moulted in sequence, with the inner primary being replaced first. The remaining primaries are then progressively replaced outward towards the longest, the 10th, and then finally the small 11th primary on the bastard wing (Fig. 13). With the exception of one species, the flight and tail feathers are moulted and replaced once a year, but some of the body feathers are moulted twice each year.
The exception to this is Franklin’s Gull, which moults all its feathers – including the primaries – twice a year, and as a result adults are missing primaries somewhere in their wing for about two-thirds of the year. This is a surprising pattern, because the efficiency of flight is affected by the loss of some of the primary feathers, particularly the longer outer ones. Most adults will have just completed the post-breeding moult when they start the next primary moult in December, which continues for several months. The information on moult in Franklin’s Gull is based on a small number of specimens and requires further investigation, but the available evidence suggests that the post-breeding primary moult is spread over four months and the pre-breeding moult over a slightly shorter period. Carrying out two primary moults in a year is energetically and nutritionally costly, and no convincing explanation has been offered for it. It could be that exposure to more intense sunlight throughout the year (Franklin’s Gulls winter south of the equator) causes the structure of the primary feathers to deteriorate quicker. However, this is not a convincing explanation, because other gull (and tern) species living in similar areas and making migrations across the equator appear to manage with a single annual primary moult (see also here).
FIG 12. Herring Gull (Larus argentatus) presumably starting its third year of life, still lacking white mirrors and with the new outer primaries and the outer tail feathers still growing. (Mike Osborne)
FIG 13. Above: adult Lesser Black-backed Gull (Larus fuscus) in primary wing moult, with only five of the longest primaries remaining to be moulted and several new, inner primaries in growth. Below: adult Black-headed Gull (Chroicocephalus ridibundus) in primary moult, with the two longest primaries (P9 and P10) not yet dropped and replaced, P8 missing and P7 half-grown. (Nicholas Aebischer)
FOOD AND FEEDING
Gulls are usually classified as omnivores and scavengers, and animal material greatly dominates the diets of all species. The spectrum of food sources taken by gulls is much wider than in terns. Fish and invertebrates are the main food of many species, and small mammals are also taken by the large species. In addition, the larger gulls are often predators on eggs, chicks and even adults of other birds, and several consume eggs and chicks of their own species, although they usually avoid killing their own broods.
Typically, gulls obtain their food from the ground or near the surface of water. Most species can plunge from flight into water in an attempt to capture live food organisms within the top metre or so, but they lack the penetrating diving thrust of Northern Gannets (Morus bassanus) and some terns. Small gulls often pick food items from the water’s surface, either while in flight or phalarope-like when floating on the surface. Gulls breeding inland feed on insects emerging from freshwater bodies and from flat terrestrial areas, and some even catch flying insects such as swarming ants and mayflies.
No gull species regularly and predominantly feeds on vegetable material, although some take berries and grain – for example, Herring and Lesser Black-backed gulls regularly search for animal feed on farmland and also consume grain in the spring and autumn. In some arctic areas, berries have been reported as a component of the diet of gulls, but this source of food is uncommon in more southern areas. That said, there are reports of berries being plucked off Hawthorn (Crataegus monogyna) bushes and even acorns still on oak trees being consumed by Black-headed Gulls. Many stomach analyses of gulls report small quantities of plant material, but these could have been accidentally ingested when picking up animal prey from the ground or even originate from the guts of the animals the gulls consumed. Many gulls will eat bread and other vegetable matter when offered it or when it is discarded by humans, and they often compete for food offered to waterfowl. The late Max Nicholson, who wrote Birds and Men (New Naturalist 17, 1951), was one of many who frequently and regularly fed gulls in winter at lakes, in parks and on the riverside embankment in London in the first half of the twentieth century.
In the last century or so, several species of gull have started to utilise waste materials dumped by humans and they also follow behind tractors during the ploughing of agricultural land. Availability of foods associated with human habitation has attracted gulls to places such as landfill sites and fishing boats, particularly when nets are being hauled in or fish are being gutted at sea. In many areas across Britain, such sites have become a major food source for several species and have played a role in the growth of gull populations. As a consequence, some species are now considered pests and their numbers are managed. As refuse is increasingly being managed in ways that gulls can no longer exploit, and landfill sites previously used by gulls are being closed, this is likely to have an effect on the size of gull populations. Evidence of the magnitude of such changes is yet to be clearly identified, however, and the extent to which gulls can change their feeding habits and find new food sources remains unknown.
Scavenging on carrion or stealing food from other birds (kleptoparasitism) are other commonly employed methods of obtaining food. Kleptoparasitism in gulls is sometimes directed towards their own species and seems more frequent among male gulls. Smaller species of gulls join flock of Lapwings (Vanellus vanellus) and Golden Plover (Pluvialis apricaria), and steal prey items captured by the plovers, particularly when these are too large for them to swallow immediately. Sea ducks, particularly Eiders (Somateria mollissima), are attacked by large gulls when they surface with prey items in their bill, such as crabs, that need to be handled and processed before they can be swallowed. Indigestible items that are eaten by gulls as part of their diet are regurgitated as pellets, and offer a method of identifying some of the food consumed.
DRINKING
All animals need to replace fluid in their bodies; while some obtain sufficient water in their food, others will travel some distance to obtain it. Those species that spend long periods of time at sea or nest on small islands without available fresh water have no option but to drink seawater, although this causes physiological problems owing to its appreciable salt content.
A sailor crossing a large expanse of ocean in a small boat can use a solar still to produce drinking water, and gulls (and many other birds) have the equivalent of this in the form of nasal glands, which are situated between the eyes. When they drink seawater, gulls use these glands to excrete a saline solution that has twice the concentration of the seawater, and by doing so reduce the concentration of salt in their bodies. Not infrequently, gulls are seen with water dripping from the end of their bill. These drops do not necessarily indicate that the bird has recently dipped its beak into water, but are the hypertonic saline solution excreted from the nasal glands. Using this method to remove excess salt from the body involves an energy cost for the birds, so it is not surprising that gulls show a strong preference for accessing and drinking fresh water wherever possible and are frequently seen congregating at freshwater pools and lakes. Herring Gulls at some urban sites even use small pools of fresh water when larger freshwater bodies are not available (Fig. 14). While Kittiwakes depend entirely on seawater and the healthy functioning of their nasal glands in their pelagic wintering areas, in the breeding season some will fly up to 2 km inland from their colonies to drink and bathe in fresh water. Those gulls that breed in colonies where no fresh water is locally available must depend on their nasal glands throughout the year.
FIG 14. Urban Herring Gull (Larus argentatus) attracted to drink at a small source of fresh water. (Mike Osborne)
VOICE
Most gulls are vocal, particularly in the breeding season, and this plays a part in their breeding behaviour. Calls are usually harsh and loud, and are used in territorial defence, as an alarm when birds are disturbed by predators or in greeting a mate. Quieter calls are used during courtship, and specific calls are also uttered, such as the wack-wack call of the Kittiwake during breeding, which tells its mate that it is not moving locally but is leaving and will be away some time on a feeding trip. Other specific calls are used during copulation, to maintain contact in flocks in flight, and to indicate that a food source has been found and that the calling individual is about to feed.
Studies using sonographs to analyse calls for several gull species have shown that there is a wide range of variation in the same call made by different individuals. Playbacks of the calls indicate that birds can recognise the call of their own mate but fail to respond to the same call when it is made by other individuals, suggesting that voice plays a part in individual recognition in addition to visual cues. Chicks of many gull species are able to recognise their parents at a relatively young age, and likewise the parents can identify their own chicks, apparently by their calls and actions – they respond differently to strange individuals. This is particularly important in species where the young wander or group together in a crèche. The Kittiwake is the exception in that adults appear to be unable to recognise their own chicks, although the usual isolated position of the nest on a vertical cliff face forces most chicks to remain in the same nest until they fledge. Occasionally, when two nests are built side by side, a Kittiwake chick will move across to the neighbouring nest and join chicks already there. If the wandering chick is placed back in its own nest, it soon returns to the neighbouring nest again, but such moves seem to be accepted by both pairs of adults.
ROOSTING
Gulls are inactive in the darkness of the night, spending these hours passively, although some species feed locally where there is strong artificial illumination, such as around working trawlers and in town centres. Most gull species spend the night in large flocks, with many roosting on water in sheltered bays at the coast, on reservoirs and on lakes, or on land in places that are free of predators, such as small islands.
Gull roosts are composed of single or mixed species, and may number thousands of individuals. Roost sites are often used year after year if their position excludes mammalian predators. Most species avoid roosting on water exposed to severe wave action, presumably because they do not possess the behavioural adaptations needed to cope with resting and sleeping while sitting on rough or breaking water. Sabine’s Gull, Ross’s Gull and the Kittiwake are totally oceanic outside the breeding season and are the only gulls that regularly spend the night far from shore. Kittiwakes will also roost offshore during the breeding season if they are not incubating or protecting small young in the nest. Before egg-laying, they leave the colony at, or long after, sunset and fly many kilometres out to sea in the dark.
