100 years of behaviour studies Angela Turner

‘The courtship habits of the Great Crested Grebe Podiceps cristatus; with an addition to the theory of sexual selection’ (Huxley 1914); this article was a good example of single-species studies of bird behaviour published in Britain early in the twentieth century. Dan Cole

ABSTRACT In the first half of the twentieth century, studies of bird species were influential in the evolution of the science of behaviour. Following the development of key theoretical ideas, behavioural research diversified in the second half of the century, with studies using as model species providing important examples and insights in such areas as optimal foraging, cognition, sexual selection, mating systems, co-operation, communication and navigation. Such single-species studies offer insights not possible from comparative studies.They have also had more practical applications, for example in wildlife management and in monitoring pollution and climate change.

he observation and recording of the captive species, but also those species that lived behaviour of birds has a long history. in and around human habitations and farms. TAristotle was the first important chronic- Johann Pernauer, in , kept birds in ler of the lives of birds and other , aviaries and trained them to fly out for a day or recording his own and others’ observations in more into the wild. His book Agreeable Country the Historia Animalium. Much of his informa- Pleasures or Taming and Training Birds, first tion came from hunters and bird catchers who published in 1702, includes detailed observa- came to know the habits of their prey, the better tions of a variety of common species. In to catch them. Observations in later centuries England, Gilbert White kept detailed and often concerned not only domesticated or other careful records of the local birds of Selbourne.

© British Birds 100 • November 2007 • 665–679 665 100 years of bird behaviour studies

By the late nineteenth century some mainly habits of Brown Noddies Anous stolidus and observational studies of bird behaviour were Sooty Terns Onychoprion fuscata in the Dry being made by, for example, Edmund Selous in Tortugas and devised experiments to test their England and John James Audubon in America, orientation ability, how they recognise their ter- but it is really only since the mid twentieth ritory and their behaviour in mazes (e.g. century that descriptions of natural history Watson 1908); in Britain, Julian Huxley have changed to a search for explanations of recorded the courtship displays of birds, behaviour. The study of various species of birds including Common Redshank Tringa totanus has been at the forefront of this recent growth and Great Crested Grebe Podiceps cristatus (e.g. in biological research, and my intention in this Huxley 1912, 1914); H. Eliot Howard wrote essay is to give some examples of how such about territorial behaviour in birds (Howard studies have contributed to the science of 1920); F. H. A. Marshall studied the effects of animal behaviour over the past 100 years. male bird displays on the breeding behaviour of Inevitably, given the space available, I shall females (e.g. Marshall 1936); and in Germany touch on relatively few particular studies and Oskar Heinroth compared the anatomy and those chosen are a personal selection to illus- behaviour of several species of ducks and geese trate the developments in this field. (Anatidae) (e.g. Heinroth 1911). At the beginning of the twentieth century, Studies and observations of the biology of the emphasis was still mainly on describing the bird species proliferated in the twentieth typical behaviour and biology of a species. Little century, and were often published in the experimental or theoretical work was being various local and national specialist bird jour- done and, in the behavioural field, what there nals such as British Birds that had appeared was primarily concerned mechanisms of behav- since the middle of the nineteenth century. iour. For example, the American psychologist British Birds has published numerous articles J. B. Watson recorded the feeding and breeding on bird biology and behaviour, for example the Lacks’ study of Common Swifts Apus apus (Lack & Lack 1952) and Bryan Nelson’s study of Northern Gannets Morus bas- sanus (Nelson 1965), as well as shorter observations and descriptions in the behaviour notes section. Notable early con- tributions to BB include Lack & Lockley’s (1938) study showing that displaced European Storm- petrels Hydrobates pelagicus and Manx Shearwaters Puffinus puffinus do not need familiar landmarks to return home, and the articles by Fisher & Hinde (1949) and Hinde & Fisher (1951) reporting the widespread ability of some birds, mostly Blue Tits Cyanistes caeruleus and Great Tits Parus major,to open the tops of milk bottles to get at the cream, which has been widely cited as an example of avian innovation and imitation.

Richard Chandler Many descriptive studies of 310. When BB was first published, descriptive accounts of the typical the basic biology of bird species behaviour and biology of a species were the norm; for example, in Britain, Julian Huxley recorded the courtship displays of the Common Redshank have long been done by both Tringa totanus (Huxley 1912). amateur and professional

666 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies ornithologists because of an interest in the the only models available for studying behav- species itself. Biologists, however, often study a ioural science – other taxonomic groups may be particular species to explore fundamental scien- more suitable on practical or welfare grounds, tific questions in, for example, behaviour or or because of particular characteristics of the ecology. One (or several related) species is thus group – but they have been, and still are, very seen as a model for animals in general with popular subjects of scientific research. About a behaviour that is representative of other taxo- quarter of papers published in the journal nomic groups. This inductive approach based Animal Behaviour between 2000 and 2006, for on studies of single species, and of how and example, were on birds. why species differ, can thus lead to general Ideally a model species needs to be abun- underlying scientific principles. Detailed studies dant, easy to catch and observe, and amenable of a species’ biology, in contrast, often provide to humans interfering with their everyday lives background information, inspiration and ideas to some extent. It also helps if a species has a for further questions in these hypotheses-driven relatively short life span and is sedentary or reg- studies. As an example from BB, the observa- ularly breeds in the same place so that individ- tion of bottle opening by tits referred to above uals can be followed over their own and their has been followed by other studies on how offspring’s lifetimes. If the species lives in novel behaviours are passed between individ- colonies or at high densities with individuals uals (see e.g. Galef & Laland 2005) and by an breeding at about the same time of year, experimental examination of the original phe- researchers can easily conduct experiments such nomenon, which suggests that were as manipulating brood size or numbers of nest learning the technique themselves after finding parasites. Birds often fit the bill in these respects opened and partially opened bottles rather than and, consequently, researchers frequently use copying the behaviour of other birds (Sherry & them as models. Nestbox nesters such as Great Galef 1984, 1990). Tits or Tree Swallows Tachycineta bicolor (in A variety of species have been used as North America) are particularly popular sub- models of animal behaviour, from dung-flies jects because they are easy to catch and their (Scathophagidae) and social spiders (Araneae) breeding attempts can be readily recorded, to dolphins (Odontoceti) and chimpanzees whether by checking the nest contents manually (Pongidae), and good models have become the or (nowadays) remotely by video or webcam. focus of long-term studies, with the input of In the early twentieth century in North numerous researchers over the years. Some of America, psychologists such as B. F. Skinner these studies concern charismatic large used mainly rats Rattus and pigeons mammals. For instance the Chimpanzees Pan (Columbidae) as model animals to study topics troglodytes at Gombe Stream National Park, such as learning in the laboratory. In contrast, Tanzania, have been studied since 1960, and in scientists were turning their attention both the African Elephants Loxodonta africana to animals in the wild. Studies of birds were in Amboseli National Park, Kenya, and the Red influential in the development of ethology, the Deer Cervus elaphus on the island of Rum, Scot- science of animal behaviour, in Europe at this land, since 1972. Smaller creatures are no less time, particularly studies by the Dutch zoologist important, however; the mating system of Niko Tinbergen and the Austrian-born zoolo- Yellow Dung-flies Scathophaga stercoraria,for gist Konrad Lorenz. Both Tinbergen and Lorenz example, has been a topic of investigation since were interested in searching for explanations of 1965. Studies of birds have also continued over behaviour rather than just describing it. They many years: Anders Møller, for example, has shared the Nobel Prize in Physiology or Medi- observed and studied Barn Swallows Hirundo cine in 1973 with Karl von Frisch (unfortu- rustica at Kraghede, Denmark, since 1971. nately there is no Nobel Prize for studying Researchers usually select a model species behaviour or ecology). Tinbergen, together with that is suitable for answering the questions that students and colleagues, carried out many they want to ask, although a childhood interest important experiments on the causation and in a particular taxon, whether it be dung-flies function of behaviour in a variety of animals, cavorting on a cowpat or swallows skimming including digger wasps (Sphecidae), stickle- around the cows, can also influence what a biol- backs (Gasterosteidae) and gulls (Laridae) (e.g. ogist studies as an adult. Birds are by no means Tinbergen 1951, 1953). He showed, for

British Birds 100 • November 2007 • 665–679 667 100 years of bird behaviour studies

example, that animals will respond to certain more realistic Robin model without the red specific cues in their environment, called ‘sign breast. stimuli’. Herring Gull Larus argentatus chicks As well as external cues, early topics of peck at the tip of their parent’s bill to stimulate research also included internal, motivational feeding. By experimenting with model gull factors that cause an animal to behave in a heads and bills with different sizes and colours certain way. In a detailed analysis of the behav- of spots, Tinbergen and his students found that iour of Herring Gulls, Baerends (1970) it is the contrast between the bill and the red described four major motivational systems con- spot at the end, rather than the bill itself, that trolling preening, nesting, incubation and stimulates the chick to peck (Tinbergen & responses to predators. One system would Perdeck 1950). A long red rod with three white inhibit the others, so an incubating bird would circles painted on the end proved to be better at remain sitting on the nest until another system eliciting pecking by chicks than a model of a took control in response to, say, a fox Vulpes normal gull’s head; Tinbergen called this a approaching, when escaping would become ‘supernormal’ stimulus. Another familiar more important than brooding. Sometimes two example of a sign stimulus is the red breast of such factors are relevant at the same time and the Robin Erithacus rubecula.David Lack there is a conflict between them. The ‘upright’ (1943) carried out experiments with stuffed posture assumed by two Herring Gulls dis- models and found that Robins gave a threat playing to each other across their territory display to a tuft of red feathers but ignored a boundary shows aspects of typical fleeing behaviour – i.e. neck stretched up and plumage sleeked – and aggressive behaviour – i.e. bill pointing down and carpal joints raised. In such situations aggres- sive behaviour is commonly redirected to another object rather than to the rival; thus Herring Gulls will peck and pull at a tuft of grass as if gathering material for a nest instead of attacking the other gull (Tin- bergen 1953). In the 1960s and 1970s, behavioural research grew rapidly and diversified. A number of seminal theoretical scientific papers and books in the fields of behavioural and evolutionary ecology were pub- lished which led to numerous experiments to test hypotheses. The emphasis then changed from mechanistic explanations of behaviour to more functional and evolutionary ones and from asking what a species typically Graham Catley Graham does to why individuals in a 311. In the middle of the twentieth century, Niko Tinbergen and Konrad Lorenz were among those interested in searching for explanations of population behave differently behaviour rather than just describing it.Tinbergen found that animals will from each other. New tech- respond to certain specific cues in their environment – sign stimuli – and niques such as the use of satel- his research experiments showed that it is the contrast between the bill lite tracking and DNA and the red spot at the end, rather than the bill itself, that stimulates a Herring Gull Larus argentatus chick to peck at the spot on its parent’s beak fingerprinting have also influ- (Tinbergen & Perdeck 1950). enced what could be studied.

668 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies

Optimal behaviour guishing between some prey types and take the Studies of single species of birds have been wrong one; they may deliberately sample dif- important in answering questions such as ferent food patches to check whether it is worth whether animals behave optimally, for example switching to another one; or, in the case of the whether they behave so as to maximise the ratio Barn Swallows, with a bill full of large insects it of benefits gained to costs incurred by their may be more profitable to catch a few small behaviour. Optimal foraging was a topic of par- ones rather than another large one (Krebs & ticular interest in the 1970s. Foraging birds are McCleery 1984; Houston 1985). Further studies often easy to observe, especially when they are on birds have shown that foraging decisions are bringing food to chicks, and to conduct experi- based on several factors. For example, birds take ments on. They are thus useful for investigating into account whether potential feeding sites are the costs and benefits of behaviour, and safe from predators and competitors (e.g. Lima answering questions such as whether animals 1985). In addition, birds sometimes avoid risky forage in the most profitable places and what feeding situations and at other times take risks: sort of prey animals they should take to max- a well-fed bird may prefer to feed at a site with a imise the net rate of return, in terms of time small but constant food supply whereas one and energy spent foraging and energy gained whose energy reserves are low may choose to from the prey (Stephens & Krebs 1986). In feed at a site where prey abundance is variable studies of captive Great Tits, for example, as but that offers the possibility of a high reward predicted, birds distinguished between patches (e.g. Caraco 1981). of food of different quality and chose to forage The concept of optimal behaviour has been in the better patch (e.g. Smith & Sweatman applied to areas other than foraging, such as 1974). Moreover, in a test of a model known as how large a territory should be. Rufous Hum- the ‘marginal value theorem’ (Charnov 1976), mingbirds Selasphorus rufus defend patches of the more time and energy Great Tits spent in flowers on stopovers during their migration; travelling to a patch, the longer they stayed in it these territories are likely to be large enough for to reap the most benefit (Cowie 1977). When the birds to maximise the amount of fat they given prey of different sizes at high densities, put on, but too large a territory may be costly to again as predicted by theory, Great Tits also defend from other birds. Carpenter et al. (1983) selected the largest prey, which are the most showed that this was so by setting up perches energetically profitable, even when smaller prey attached to balances to weigh hummingbirds were more abundant (Krebs 1978). Studies of and recording how much weight they put on in wild birds bear out the findings from experi- relation to the size of territory they had. The ments on captive ones. Northern Lapwings hummingbirds maintained intermediate-sized Vanellus vanellus, for example, catch worms of territories at which their rate of weight increase different sizes but concentrate on the smaller was greatest. worms because very large ones take a long time to get out of the ground. The birds take more of Cognition them as the density of small worms in the An aspect of foraging behaviour of current ground increases, whereas their intake of large research interest is caching, as it provides worms does not depend on the density of these insights into the mental world of birds. Some less profitable prey (Thompson & Barnard birds have remarkable spatial memories; Clark’s 1984). Evidence from studies on birds therefore Nutcracker Nucifraga columbiana, for example, supports models of optimal behaviour. stores up to 33,000 seeds in thousands of dif- However, both Great Tits and Lapwings take ferent sites each autumn, retrieving them peri- some unprofitable prey when theory suggests odically until the following spring (Vander Wall that they should ignore them. Barn Swallows & Balda 1977). In addition, individuals making collecting insects for their chicks also bring caches have to contend with thieves trying to back small aphids (Aphidoidea) as well as the steal them. Experiments on corvids show that more profitable large hover-flies (Syrphidae) they take the presence of would-be thieves into and blow-flies (Calliphoridae) (Turner 1982). account when making caches. Common Ravens There may be a number of reasons why birds do Corvus corax and Western Scrub Jays Aphelo- not forage in a perfectly optimal manner: they coma californica hide food in places that are out may, for example, have difficulty in distin- of sight of competitors when these are present,

British Birds 100 • November 2007 • 665–679 669 100 years of bird behaviour studies

such as behind a rock or in shady sites (Bugnyar Wild crows in New Caledonia make hooked & Kotrschal 2002; Dally et al. 2005). If a com- tools by stripping leaves and bark from twigs petitor had seen the initial caching, Western and ‘stepped-cut’ tools by tapering the ends of Scrub Jays also move food from a cache to a Pandanus leaves; they use the tools to extricate new site that is unknown to the observer, but insects from under leaves and from holes in only if they have experience of stealing from trees (Hunt 1996). Captive New Caledonian caches themselves, suggesting that they are Crows have demonstrated their skill at choosing using their experience to judge what the appropriate tools to obtain food in experi- observer will do (Emery & Clayton 2001). mental tasks and, in a possible instance of a bird However, it is not yet known whether corvids showing insight, one female, Betty, made a have a ‘theory of mind’, that is whether they hooked tool by bending a straight piece of wire, truly understand what other individuals can an unfamiliar object to her, and used it to pull and cannot see and know. up a small bucket of food in a pipe (Weir et al. Humans have long been thought to be the 2002). only animal to plan for the future, but Nicky This section also warrants mention of Irene Clayton and her colleagues have shown that a Pepperberg’s Grey Parrot Psittacus erithacus, bird can do this too (Raby et al. 2007). For six called Alex. Pepperberg trained Alex to identify consecutive mornings they allowed Western specific objects and to categorise objects by Scrub Jays to visit either a compartment in their their colour, shape and number; he also cage where they could feed or one without any- appeared to understand abstract concepts such thing to eat; in the afternoons the birds had as ‘same or different’ (e.g. Pepperberg 1999). powdered pine nuts to eat. On the sixth evening Whether Alex actually understood such con- the researchers provided some whole pine nuts cepts or whether this is an impressive example and some sand trays in which the birds could of learning to discriminate categories is not cache the nuts; the scrub jays showed foresight clear, however. and stored the nuts in the compartment without food, so that breakfast would be avail- Sexual selection able in the morning. Sexual selection, the evolutionary process Corvids also rival the great apes in their use whereby some males gain a reproductive advan- of tools. The most spectacular example is the tage over other males, for example by being New Caledonian Crow Corvus moneduloides. more attractive to females seeking a mate, is another area of research where studies of birds have made important contribu- tions. Many exaggerated characters such as long tails are thought to arise from sexual selection, such as by females actively choosing males with these characters as mates. Malte Andersson (1982) con- ducted an elegant experi- ment on Long-tailed Widowbirds Euplectes progne, a species in which the males have extremely long tails. He changed the length of the tails of males,

Markus Varesvuo shortening some, length- 312. One particular aspect of foraging behaviour which has attracted much ening others, and main- recent research effort is that of food caching. Common Ravens Corvus corax,for taining others at their example, have been shown to take account of potential robbers when making original length. Long- food caches and will hide food only when out of sight of competitors (Bugnyar & Kotrschal 2002). tailed males subsequently

670 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies had more females breeding in their territories, and generally invest less time and energy in suggesting that females prefer males with long their offspring. Females and males therefore tails. tend to have different mating strategies: males Hamilton and Zuk (1982) suggested that invest in mating with many females and females sexually selected characters might reflect the invest in parental care, although paternal care is bearer’s resistance to disease and thus its suit- still frequent among birds in contrast to other ability as a mate. Females may well spurn a male taxonomic groups such as mammals. One area with dull plumage indicative of a parasite infes- of disagreement between the sexes is their tation, for example, and instead choose as a mating system: many birds are socially monoga- mate a bright, vigorous male that is clearly free mous, probably because some paternal care is of such problems. If the resistance to disease is needed for successful rearing of the brood, but heritable, the female will benefit by having some are polygynous or polyandrous or even healthy offspring. Several researchers, including polygynandrous. For most of the twentieth Anders Møller, have attempted to test this century, monogamy was assumed to be the hypothesis. The forked tail of male Barn Swal- norm among birds and males and females were lows varies markedly in length between individ- assumed to be faithful to their mates. Biologists uals. By manipulating the length of the tail with were concerned with such topics as the effect of pieces of feather glued on or cut out, Møller ecological and social conditions on mating (1988) showed that female Barn Swallows prefer systems and particularly the contribution of males with long tails, mating sooner with them paternal care. An influential model in the 1970s than with short-tailed males. Møller (1990) also was the polygyny threshold model (Orians found that long-tailed males have offspring that 1969), which suggested that whether females harbour few blood-sucking mites, i.e. they pass chose to mate with an already mated male on a genetic resistance to these parasites. When depended on the quality of the male’s territory he cross-fostered chicks between nests so that and his parental contribution. That is, a female some chicks were reared by adoptive parents, could do as well by sharing a high-quality terri- the number of mites that the chicks had tory and a male with another female as she depended on the tail length of the biological would having exclusive help from a male in a parent, not the adoptive one. Females mating poor territory. Other factors affecting the with these long-tailed males therefore benefit by mating decisions of individuals have come to having healthier chicks. As well as fewer para- light, however. Females may choose an area sites, long-tailed males also have a stronger with good food resources rather than a specific immune system and a better survival rate than do short-tailed males, so females can get good information about the health and viability of males from the length of their tails (Saino et al. 1997).

Mating systems Conflicts between individ- uals are commonplace in nature and birds are no exception. Female birds generally produce rela- tively few eggs and, depending on the species, do most or all of the incu- Roger Tidman Roger bation and chick rearing, 313. The long tail-streamers of male Barn Swallows Hirundo rustica vary whereas males produce markedly in length between individuals. Experimental studies (e.g. Møller 1988) huge numbers of sperm have shown that females prefer males with long tails.

British Birds 100 • November 2007 • 665–679 671 100 years of bird behaviour studies

male territory. Among Dunnocks Prunella mod- the pair bond occurs in around 90% of so- ularis, for example, females compete with each called monogamous avian species (Griffith et al. other to settle in with abundant food 2002). Males are unfaithful in order to sire as and the males compete among themselves to many offspring as possible; why females do it is pair with the females (Davies 1991). Female less clear but they may thereby get direct bene- Dunnocks fledge more offspring in a monoga- fits such as access to resources, or indirect bene- mous relationship than when they share a male fits such as better quality fathers for their chicks with another female, and do even better if a and thus have healthier or better quality off- second male is present and helps to feed the spring. This is still a controversial area and the chicks (co-operative polyandry), whereas males reasons for infidelity are likely to vary among do best with two females and least well if they species. Among Barn Swallows, females with share females with another male. These con- short-tailed social partners apparently benefit flicts of interest can lead to one female in a from mating with other, long-tailed males polygynous relationship trying to expel the because they then have sons that have attractive other one and to monogamous females solic- long tails as adults (Saino et al. 2003). iting matings from other males, which may then help to rear their broods. Co-operation Early ideas about mating systems were over- The evolution of co-operation has been a major turned in the 1980s and 1990s when genetic issue in biology. Until the 1960s, it was not analysis, especially using DNA fingerprinting, thought surprising that animals would co- revolutionised our understanding of mating operate with each other for the common good systems. Gowaty & Karlin (1984), in a study of and even for the good of the species. Such ideas Eastern Bluebirds Sialia sialis, were the first to fell into disfavour as the concept of selfish indi- report that broods of an apparently monoga- viduals and selfish genes came to the fore but mous contained some nestlings sired where it did occur, co-operation still needed an by males other than the female’s social mate. explanation. Hamilton (1964) put forward a Since then, multiple paternity has been found highly influential theory of inclusive fitness, in the broods of numerous species. We now that individuals help those with whom they know that infidelity is the norm: mating outside share genes, not only their own offspring but also, for example, their sib- lings, nephews and nieces. Birds provide numerous examples of this; around 96% of bird species that live in family groups breed co-operatively, helping to rear another individual’s offspring (Emlen 1995) and some studies have shown that helpers prefer to help their close kin. Among White-fronted Bee-eaters Merops bullock- oides, for example, pairs live in extended family groups; depending on how good the environmental conditions are, some birds Markus Varesvuo 314. For most of the twentieth century, it was thought that monogamy was the do not breed or they start norm among birds and that males and females were faithful to their mates. Our nesting and are then ideas about mating systems were turned upside-down in the 1980s and 1990s unsuccessful and about when genetic analysis, especially using DNA fingerprinting, revolutionised our half of these help other understanding and revealed that commonplace, soberly plumaged and pairs by incubating unobtrusive species like the Dunnock Prunella modularis could have a colourful and complex mating system (Burke et al. 1989). clutches and feeding

672 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies nestlings and fledglings (Emlen 1990, 1997). between species, with some species learning They are most likely to help their closest genetic into adulthood (see Marler 1997 for a review). relatives and show preferences even between In addition, birds such as White-crowns are not first cousins and half siblings. Similarly, failed hard-wired to learn only their own species’ song breeders in European Bee-eaters Merops apiaster and will acquire those of other species (e.g. and Long-tailed Tits Aegithalos caudatus choose Baptista & Morton 1988). to help their close relatives (Lessells 1990; Studies in the 1970s and 1980s showed that Russell & Hatchwell 2001). Groups, and fam- birdsong has two main functions: it repels indi- ilies, of birds have their conflicts, however, and viduals intruding into a territory and it attracts individuals sometimes manipulate others. females (e.g. Krebs 1977, Eriksson & Wallin When subordinate pairs of European Bee-eaters 1986). John Krebs (1977), for example, found try to breed, dominant individuals sometimes that when male Great Tits were removed from harass them, forcing them to stop and to their territories, replacement males took over become helpers. It is usually fathers that do this the territory within ten daylight hours when no to their youngest sons; these young males may male was present but within 30 hours when be the easiest victims and the ones that benefit loudspeakers broadcast song simulating the most by staying in the group and so they presence of a territorial male. Great Tits sing submit to the coercion. only a few song types but some species such as Common Nightingales megarhynchos Communication have hundreds in their repertoire. The size of In the field of communication, birds and their the repertoire also varies among males within amazing variety of songs and visual displays species; repertoire size is thought to be sexually have been an obvious choice for studying the selected, with females preferring large ones (e.g. development, nature and function of signals. MacDougall-Shackleton 1997). Repertoires may The modern study of song learning started with also be involved in territory defence, allowing W. H. Thorpe’s (1958) work on the Common neighbouring males to interact by replying to Chaffinch Fringilla coelebs and his pioneering each other with song types that they share. use of the sound spectrograph to analyse songs. Males may thus indicate their aggressive intent Thorpe’s student Peter Marler also studied by matching their rival’s song. For example, in a Chaffinch song at Cambridge and then the song study of Song Sparrows Melospiza melodia, of White-crowned Sparrows Zonotrichia leu- Beecher et al. (2000) found that the proportion cophrys at Berkeley, Cali- fornia. He showed that young White-crowns have an innate preference to learn the song of their own species and that the song is learnt during a brief period early in life (Marler 1970). Further studies by Marler and other researchers showed that young birds practise singing, producing first a subsong, matching their output with the mem- orised song and refining it until they produce the full crystallised song, and that the details of learning, such as the duration and Simon Stirrup timing of the sensitive 315. A male Common Nightingale Luscinia megarhynchos will have hundreds of period when songs are song types in its repertoire.The size of this repertoire will vary among males, however. Female birds are thought to prefer males with a large repertoire (e.g. learnt, vary considerably MacDougall-Shackleton 1997).

British Birds 100 • November 2007 • 665–679 673 100 years of bird behaviour studies

of songs shared with neighbours was related to found that female Great Tits compare the per- how many years that males were able to keep a formance of their mates and their neighbours territory. during singing contests; females visit the neigh- As well as song matching, when rival males bouring males if these appear to be superior to sing at each other, one male may start singing their own mate. before the other one has finished; this overlap- ping of song may be a signal of status or will- Navigation ingness to fight on the part of the overlapping Birds make conspicuous migrations, so it is not singer (e.g. Hultsch & Todt 1982). Signals such surprising that they have been the focus of as this can be picked up not only by the many studies on the subject. In the first half of intended receiver but also by other animals in the last century, several experiments with wild the vicinity. These ‘eavesdroppers’ can obtain birds were carried out to investigate homing information about the signaller and the abilities (such as Lack & Lockley’s study receiver, such as their readiness to fight, that referred to above; see Wiltschko & Wiltschko they may use to their own advantage. To test 2003 for a review). These involved displacing this idea, Naguib & Todt (1997) played songs the birds, which included terns (Sternidae), from two loudspeakers, imitating interacting Common Starlings Sturnus vulgaris, Barn Swal- rivals, to male Nightingales; in some cases the lows and House Martins Delichon urbicum,as songs from one speaker overlapped those from well as petrels and shearwaters (Procellariidae the other one, whereas in others they did not. and Hydrobatidae), many miles away from their Naguib and Todt then recorded these males’ breeding site and recording whether they responses to the two types of song. The males returned. Although these experiments revealed appeared to have eavesdropped on the loud- the superb homing abilities of birds, they failed speakers’ interaction; they remembered which to shed light on how the birds homed success- speaker’s songs overlapped the other and fully, and it was not until the 1950s that responded more intensely to this apparently research on orientation took off following more aggressive individual. In a similar play- methodological and theoretical insights. Much back experiment carried out early in the season of the research has been done on homing when females were looking for mating partners pigeons or other species in captivity, often using outside their pair bond, Otter et al. (1999) Emlen funnels, which are cages in which the preferred direction of movement of birds can be recorded from the marks they make on inked or typing correction paper. These studies have shown that birds use a number of cues to navigate, including a sun or star compass, magnetic cues, familiar landmarks and odours. Modern techniques such as tracking tagged birds by satellite have pro- vided more detailed infor- mation about migration in free-living animals. Aler- stam et al. (2006), for

Markus Varesvuo example, tracked Ospreys 316. Blackcaps Sylvia atricapilla from the Continent have been wintering in Pandion haliaetus on Britain in increasing numbers since the 1950s. Studies have shown that there are spring and autumn migra- two distinct, inherited, migratory patterns within central European Blackcap tions over a number of populations (e.g. Berthold et al. 1992). In recent decades, increasingly mild winters may be encouraging more individuals to make the relatively short years between Europe and westward journey to winter in Britain. Africa. Individual birds did

674 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies not use the same route for each migration and Stoats but learnt to do so after being exposed to their flight paths were up to around 400 km them, suggesting that training with predators apart. However, an individual’s different flight will improve the ability of animals to avoid paths converged in some areas which the birds them. visited each year. Thus Ospreys do not seem to According to life-history theory, individuals use landmarks and may instead use a form of have a finite amount of resources at their dis- map-based navigation to reach these interme- posal for the various components of their life diate areas. histories, which they are thought to trade off A study on Blackcaps Sylvia atricapilla indi- one against another (Stearns 1992). For cates that migratory behaviour is under genetic example, producing and rearing more offspring control (Berthold et al. 1992). Blackcaps from in one year may be so demanding that the indi- the Continent have been wintering in Britain in vidual’s chances of survival are reduced. This increasing numbers since the 1950s. When effect has been recorded in experiments, carried taken to Germany, individuals from this out on wild birds such as Great Tits (Visser & migrant British population and their offspring, Lessells 2001), in which brood sizes were bred in a German laboratory, showed a prefer- changed and parental survival recorded. This ence in orientation tests for a heading a little trade-off has implications for conservation and north of west. In contrast, German Blackcaps, management plans that involve exploiting a which normally head for the Mediterranean, particular life-history stage such as eggs, as favoured a south-westerly direction. There are shown by a recent study on the Mauritius thus two distinct, inherited, migratory patterns Kestrel Falco punctatus (Nicoll et al. 2006). This within central European Blackcap populations. once-endangered species has been reintroduced Milder British winters may have led to a higher to areas where it was extinct and as part of the frequency of individuals taking the shorter recovery programme first clutches were route westward in the last few decades. removed, so that chicks could be reared and fos- tered in other nests when 10–12 days old. Applied studies Females will produce a replacement clutch More applied areas of research in the last when their first is removed. Although they pro- century have often been the management of duced more eggs, managed birds thus spent less species for pest control, sport or other commer- time incubating and rearing chicks than cial reasons. Increasingly, however, studies of unmanaged ones and Nicoll et al. (2006) birds and other animals are concerned with recorded greater improvements in clutch size understanding not just the natural world but and survival of both females and males that also the impact that humans are having on it were managed in this way, as predicted if there and how we can mitigate the harm we do, or is a trade-off between these life-history compo- how animals in turn affect human populations. nents. Many studies now deal with topics such as pol- Bird species can be good indicators of lution, habitat loss, transmission of diseases environmental contamination, especially those such as West Nile virus and avian influenza, that eat insects and other animals because they biodiversity, conservation and particularly are at the top of their food chains: although a climate change. single aquatic larval insect may pick up a small Studies on bird species can have direct bene- amount of a pollutant from sediment, an fits to conservation and wildlife management. insectivore feeding on large quantities of these Learning influences appropriate responses to prey when they emerge as adults will ingest a predators; hence relocation of animals or substantial amount. Any pollutants in the release of captive individuals without experi- ecosystem thus accumulate in predators until ence of predators can have fatal consequences. they reach the top of the food chain, such as a Maloney & McLean (1995) found that New hirundine that eats insects or a raptor that eats Zealand Robins Petroica australis from the fish. The health or conservation status of these mainland, where small introduced carnivores indicator species provides information about were present, immediately recognised Stoats the condition of the ecosystem they inhabit and Mustela erminea as predators. In contrast, of other species in that ecosystem. robins from an island offshore, with no experi- Organochlorine pesticides such as DDT and ence of these predators, failed to respond to dieldrin were widely used in North America

British Birds 100 • November 2007 • 665–679 675 100 years of bird behaviour studies and Europe from the late 1940s to the 1960s, elsewhere, whereas it is usual for a small causing dramatic declines in populations of percentage of males not to breed because of an raptors such as the Peregrine Falcon Falco excess of males in the population. Birds that did peregrinus and Eurasian Sparrowhawk Accipiter breed at Chernobyl had smaller clutches, nisus; the latter was almost wiped out from averaging 4.3 eggs compared to 4.6 at Kanev, some eastern areas of Britain where the and lower hatching success (94% versus over pesticide cyclodiene was used on arable 98%); broods were markedly (14%) smaller at farmland. The application of these chemicals Chernobyl. Annual adult survival was only 28% was associated with the thinning of eggshells, in the contaminated area, much less than the egg breakages and breeding failures in raptors, 40% for birds at Kanev, and less than survival as well as with direct poisoning of individuals rates for other populations. The Chernobyl (see Newton 1979 for a review). Populations of population cannot sustain itself and draws in Sparrowhawks and other raptors recovered birds from elsewhere (Møller et al. 2006). when the use of these pesticides declined. Bird Møller has recorded basic breeding data of species such as Tree Swallows are still often used Barn Swallows such as laying dates, clutch sizes, as indicators of environmental pollution, brood sizes and fledging success since the 1970s through the monitoring of, for example, their and this has proved valuable in understanding reproductive success, nestling growth and some of the effects of a recent man-made mortality, endocrine function, immune phenomenon, global warming, on wildlife. response and behaviour. In the Great Lakes–St About two-thirds of Barn Swallows have two Lawrence Basin, for example, numerous health broods a year. In 1971 Barn Swallows started effects of pollutants such as polychlorinated their second broods about 45 days after starting biphenyls (PCBs) have been recorded in studies their first but the time between broods has of various bird species including Bald Eagles increased by an average of 8 days since then Haliaeetus leucocephalus, Herring Gulls, Night (Møller 2007). The birds are laying their first Herons Nycticorax nycticorax and Tree Swallows clutch earlier and their second clutch later. The as well as of other animals such as fish and change in laying schedule is related to the whales (Cetacea) (Fox 2001). These effects average temperature during April, which has included enlarged thyroids in adult Herring risen by more than 2.2°C in the study area in Gulls, feminised reproductive tracts in Herring Denmark since 1971. The longer interval Gull embryos and chicks, and abnormally between clutches is advantageous for Barn developed bills in Bald Eagles (Fox 2001). Swallows, as they can then rear more fledglings. Behaviour is also affected. High levels of PCBs, Climate change has adverse effects too, for example, disrupt nest building in Tree however; survival rates of Barn Swallows have Swallows and lead to egg mortality and females declined as environmental conditions in North deserting nests (McCarty & Secord 1999a,b, Africa have become poorer (Møller & Szép 2000); PCBs are also associated with high levels 2005). of porphyrins in the livers of nestling Tree Swallows, indicating the inhibition of certain Concluding remarks enzymes (Bishop et al. 1999). In this essay I have touched on only a few Animals are also affected by radioactive selected topics and these not in any depth; contamination, such as at the Chernobyl site, readers may also discern a slight bias towards although little is known compared with our my favourite species, the . The knowledge of the effects of radioactivity on examples used were all based on a single humans. Møller et al. (2005) looked at how species, however, to illustrate that long-term Barn Swallows fared just outside the exclusion study of one species provides a lot of data that zone around the Chernobyl reactor. During six can be interesting in its own right and that can years of study between 1991 and 2004, nearly a also lead to fruitful avenues of investigation in quarter of Barn Swallow pairs there did not ways that were never considered when the study breed; in contrast almost no pairs failed to was started. Anders Møller, for example, breed at another site, Kanev, 220 km southwest initially studied the effect of farming on birds of Kiev and in other European populations that and then why Barn Swallows breed in groups. have been well studied. Most strikingly, females He later became interested in sexual selection, were not breeding, which is almost unheard of and investigated why the length of the tail

676 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies varied markedly between male Barn Swallows References (Møller 2001). Studies of Barn Swallows now Alerstam,T., Hake, M., & Kjellen, N. 2006.Temporal and cover many aspects of behavioural research, and spatial patterns of repeated migratory journeys by ospreys. Anim. Behav. 71: 555–566. the data obtained over the years have proved Andersson, M. 1982. Female choice selects for extreme tail important for investigating more recent length in a widowbird. Nature 299: 818–820. problems such as the effects of a nuclear Baerends, G. P.1970. A model of the functional accident and of climate change. organization of incubation behaviour. Behav. Suppl., XVII: 263–312. Despite the eclectic nature of the examples Baptista, L. F., & Morton, M. L. 1988. Song learning in given in this essay, it is clear that single-species montane White-crowned Sparrows: from whom and studies have helped to inform many aspects of when. Anim. Behav. 36: 1753–1764. Beecher, M. D., Campbell, S. E., & Nordby, J. C. 2000. behavioural research over the past 100 years. Territory tenure in Song Sparrows is related to song Although I have given examples for birds, the sharing with neighbours, but not to repertoire size. concepts discussed apply across many taxo- Anim. Behav. 59: 29–37. nomic groups. Studies of model species have Berthold, P., Helbig, A. J., Mohr, G., & Querner, U. 1992. Rapid microevolution of migratory behaviour in a wild provided examples of general phenomena such bird species. Nature 360: 668–670. as sign stimuli and conflict behaviour which Bishop, C. A., Mahony, N. A.,Trudeau, S., & Pettit, K. E. 1999. occur throughout the animal kingdom. They Reproductive success and biochemical effects in Tree Swallows (Tachycineta bicolor) exposed to chlorinated also provide empirical support for theoretical hydrocarbon contaminants in wetlands of the Great ideas such as optimality theory, Hamilton- Lakes and St. Lawrence River Basin, USA and Canada. Zuk’s hypothesis that sexually selected charac- Environ.Toxicol. Chem. 18: 263–271. ters reflect the bearer’s health, and Hamilton’s Bugnyar,T., & Kotrschal, K. 2002. Observational learning and the raiding of food caches in Ravens, Corvus corax: is it concept of inclusive fitness, and they help to tactical deception? Anim. Behav. 64: 185–195. answer basic questions such as why females Burke,T., Davies, N. B., Bruford, M.W., & Hatchwell, B. J. sometimes mate with just one male, and males 1989. Parental care and mating behaviour of with one female, and sometimes with more polyandrous Dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature 338: 249–251. than one, and how animals navigate. Studies Caraco,T. 1981. Risk-sensitivity and foraging groups. Ecology following individuals of single species over 62: 527–531. time, such as studies of song development and Carpenter, F. L., Paton, D. C., & Hixon, M. A. 1983.Weight gain and adjustment of feeding territory size in migrant song structure, have provided additional hummingbirds. Proc. Nat.Acad. Sci., USA 80: 7259–7263. insights, and song learning is an excellent Charnov, E. L. 1976. Optimal foraging: the marginal value example of how nature and nurture interact in theorem. Theoretical Population Biology 9: 129–136. the development of a motor skill. Studies of Cowie, R. J. 1977. Optimal foraging in Great Tits (Parus major). Nature 268: 137–139. species such as the Western Scrub Jay suggest Dally, J. M., Emery, N. J., & Clayton, N. S. 2005. Cache that they have cognitive skills, including tool protection strategies by Western Scrub-jays, use, mental attribution and mental time travel, Aphelocoma californica: implications for social cognition. Anim. Behav. 70: 1251–1263. once thought to be uniquely human. That Davies, N. B. 1991. Dunnock Behaviour and Social Evolution. these skills are evident in birds implies that OUP,Oxford. there is not just an intellectual continuum Emery, N. J., & Clayton, N. S. 2001. Effects of experience from apes to humans, but that complex cogni- and social context on prospective caching strategies by scrub jays. Nature 414: 443–446. tion has evolved more than once, in the Emlen, S.T. 1990.The White-fronted Bee-eater: helping in a corvid’s case because of the need to store food, colonially nesting species. In: Stacey, P.,& Koenig,W. out of reach of other individuals, for periods (eds.), Cooperative Breeding in Birds: long-term studies of when it is scarce. Finally, although much ecology and behavior, 301–337. Blackwell Science, Oxford. behavioural research on birds has been driven — 1995. An evolutionary theory of the family. Proc. Natn by curiosity, the theoretical ideas and knowl- Acad. Sci., USA 92: 8092–8099. edge gained, particularly in long-term studies — 1997. Predicting family dynamics in social vertebrates. In: Krebs, J. R., & Davies, N. B. (eds.), Behavioural Ecology: of single species, increasingly have practical an evolutionary approach, 4th edn, 228–253. Blackwell, applications, for example in wildlife manage- Oxford. ment and for monitoring the impact of Eriksson, D., & Wallin, L. 1986. Male bird song attracts humans on the environment. females: a field experiment. Behav. Ecol. Sociobiol. 19: 297–299. Fisher, J., & Hinde, R. A. 1949.The opening of milk bottles Acknowledgment by birds. Brit. Birds 42: 347–357. Thanks to Patti Loesche for valuable comments on the Fox, G. A. 2001.Wildlife as sentinels of human health manuscript. effects in the Great Lakes – St. Lawrence Basin. Environ. Health Perspect. 109: 853–861.

British Birds 100 • November 2007 • 665–679 677 100 years of bird behaviour studies

Galef, B. G., Jr, & Laland, K. 2005. Social learning in animals: Marshall, F. H. A. 1936 Sexual periodicity and the causes empirical studies and theoretical models. Bioscience 55: which determine it. Phil.Trans. Roy. Soc. Lond. B 226: 489–499. 423–456. Gowaty, P.A., & Karlin, A. A. 1984. Multiple maternity and McCarty, J. P.,& Secord, A. L. 1999a. Reproductive ecology paternity in single broods of apparently monogamous of Tree Swallows (Tachycineta bicolor) with high levels of Eastern Bluebirds (Sialia sialis). Behav. Ecol. Sociobiol. 15: polychlorinated biphenyl contamination. Environ.Toxicol. 91–95. Chem. 18: 1433–1439. Griffith, S. C., Owens, I. P.F., & Thuman, K. A. 2002. Extra- — & — 1999b. Nest-building behaviour in PCB- pair paternity in birds: a review of interspecific variation contaminated Tree Swallows. Auk 116: 55–63. and adaptive function. Molec. Ecol. 11: 2195–2212. — & — 2000. Possible effects of PCB contamination on Hamilton,W. D. 1964.The genetical evolution of social female plumage color and reproductive success in behaviour: I & II. J.Theor. Biol. 7: 1–16, 17–52. Hudson River Tree Swallows. Auk 117: 987–995. — & Zuk, M. 1982. Heritable true fitness and bright birds. Møller,A. P.1988. Female choice selects for male sexual tail A role for parasites? Science 218: 384–387. ornaments in a swallow. Nature 332: 640–642. Heinroth, O. 1911. Beitrage zur Biologie, namentlich — 1990. Effects of a haematophagous mite on the Barn Ethologie und Psychologie der Anatiden. Proc. Fifth Swallow Hirundo rustica: a test of the Hamilton and Zuk International Ornithological Congress, Berlin, 1910. hypothesis. Evolution 44: 771–784. Hinde, R. A., & Fisher, J. 1951. Further observations on the — 2001. Sexual selection in the Barn Swallow. In: Dugatkin, opening of milk bottles by birds. Brit. Birds 44: 392–396. L. A. (ed.), Model Systems in Behavioral Ecology: integrating Houston, A. 1985. Central-place foraging: some aspects of conceptual, theoretical, and empirical approaches, prey choice for multiple-prey loaders. Am. Nat. 125: 359–377. University Press, Princeton, New Jersey. 811–826. — 2007. Interval between clutches, fitness, and climate Howard, H. E. 1920. Territory in Bird Life. John Murray, change. Behav. Ecol. 18: 62–70. London. — & Szép,T. 2005. Rapid evolutionary change in a Hultsch, H., & Todt,D. 1982.Temporal performance roles secondary sexual character linked to climatic change. during vocal interactions in Nightingales (Luscinia J. Evol. Biol. 18: 481–495. megarhynchos). Behav. Ecol. Sociobiol. 11: 253–260. —, Hobson, K. A., Mousseau,T.A., & Peklo, A. M. 2006. Hunt, G. R. 1996. Manufacture and use of hook-tools by Chernobyl as a population sink for Barn Swallows: New Caledonian crows. Nature 379: 249–251. tracking dispersal using stable-isotope profiles. Ecol. Huxley, J. 1912. A first account of the courtship of the Applic. 16: 1696–1705. Redshank (Totanus calidris, L.). Proc. Zool. Soc. Lond. —, Mousseau,T.A., Milinevsky, G., Peklo, A., Pysanets, E., & 43: 647–655. Szép,T. 2005. Condition, reproduction and survival of — 1914.The courtship habits of the Great Crested Grebe Barn Swallows from Chernobyl. J.Anim. Ecol. 74: (Podiceps cristatus); with an addition to the theory of 1102–1111. sexual selection. Proc. Zool. Soc. Lond. 45: 491–562. Naguib, M., & Todt,D. 1997. Effects of dyadic vocal Krebs, J. R. 1977. Song and territory in the Great Tit, Parus interaction on other conspecific receivers in major. In: Stonehouse, B., & Perrins, C. M. (eds.), Nightingales. Anim. Behav. 54: 1535–1543. Evolutionary Ecology, 47–62. Macmillan, London. Nelson, J. B. 1965.The behaviour of the Gannet. Brit. Birds — 1978. Optimal foraging: decision rules for predators. In: 58: 223–288, 313–336. Krebs, J. R., & Davies, N. B. (eds.), Behavioural Ecology: an Newton, I. 1979. Population Ecology of Raptors. Poyser, evolutionary approach, 23–63. Blackwell, Oxford. Berkhamsted. — & McCleery, R. H. 1984. Optimization in behavioural Nicoll, M. A. C., Jones, C. G., & Norris, K. 2006.The impact ecology. In: Krebs, J. R., & Davies, N. B. Behavioural of harvesting on a formerly endangered tropical bird: Ecology: an evolutionary approach, 91–121. Blackwell, insights from life-history theory. J.Appl. Ecol. Oxford. 43: 567–575. Lack, D. 1943. The Life of the Robin.Witherby, London. Orians, G. H. 1969. On the evolution of mating systems in — & Lack, E. 1952.The breeding behaviour of the Swift. birds and mammals. Amer. Nat. 103: 589–603. Brit. Birds 45: 186–215. Otter, K., McGregor, P.K.,Terry,A. M. R., Burford, F. R. L., — & Lockley, R. M. 1938. Skokholm bird homing Peake,T. M., & Dabelsteen,T. 1999. Do female Great Tits experiments. I. 1936–37: Puffins, Storm-Petrels and (Parus major) assess males by eavesdropping? A field Manx Shearwater. Brit. Birds 31: 242–248. study using interactive song playback. Proc. Roy. Soc. Lessells, C. M. 1990. Helping at the nest in European Bee- Lond. B 266: 1305–1309. eaters: who helps and why? In: Blondel, J., Gosler,A., Pepperberg, I. 1999. The Alex Studies: communication and Lebreton, J.-D., & McCleery, R. (eds.), Population Biology cognitive capacities of an African Grey Parrot. Harvard of Passerine Birds: an integrated approach, 357–368. University Press, Cambridge, MA. Springer, Berlin. Raby, C. R., Alexis, D. M., Dickinson, A., & Clayton, N. S. Lima, S. L. 1985. Maximizing efficiency and minimizing time 2007. Planning for the future by Western Scrub-jays. exposed to predators: a trade-off in the Black-capped Nature 445: 919–921. Chickadee. Oecologia 66: 60–67. Roff, D. A. 1992. The Evolution of Life Histories. Chapman & MacDougall-Shackleton, S. A. 1997. Sexual selection and Hall, London. the evolution of song repertoires. Curr. Ornithol. Russell, A. F., & Hatchwell, B. J. 2001. Experimental evidence 14: 81–124. for kin-biased helping in a cooperatively breeding Maloney, R. F., & McLean, I. G. 1995. Historical and vertebrate. Proc. Roy. Soc. Lond. B 268: 2169–2174 experimental learned predator recognition in free-living Saino, N., Bolzern, A. M., & Møller,A. P.1997. New Zealand robins. Anim. Behav. 50: 1193–1201. Immunocompetence, ornamentation and viability of Marler, P.1970. A comparative approach to vocal learning: male Barn Swallows (Hirundo rustica). Proc. Nat.Acad. song development in White-crowned Sparrows. Sci., USA 94: 549–552. J. Comp. Physiol. Psychol. 71 (Suppl.): 1–25. —, Martinelli, R., Romano, M., & Møller,A. P.2003. High –– 1997.Three models of song learning: evidence from heritable variation of a male secondary sexual character behavior. J. Neurobiol. 33: 501–516. revealed by extra-pair fertilization in the Barn Swallow.

678 British Birds 100 • November 2007 • 665–679 100 years of bird behaviour studies

Ital. J. Zool. 70: 167–174. — 1953. The Herring Gull’s World. Collins, London. Sherry, D. F., & Galef, B. G., Jr. 1984. Cultural transmission — & Perdeck, A. C. 1950. On the stimulus situation without imitation: milk bottle opening by birds. Anim. releasing the begging response in the newly hatched Behav. 32: 937–938. Herring Gull chick (Larus a. argentatus Pontopp.). — & –– 1990. Social learning without imitation: more Behaviour 3: 1–38. about milk bottle opening by birds. Anim. Behav. Turner,A. K. 1982. Optimal foraging by the Swallow 40: 987–989. (Hirundo rustica, L.): prey size selection. Anim. Behav. Smith, J. N. M., & Sweatman, H. P.1974. Food searching 30: 862–872. behaviour of titmice in patchy environments. Ecology Vander Wall, S. B., & Balda, R. P.1977. Coadaptations of the 55: 1216–1232. Clark’s Nutcracker and the Piñon Pine for efficient seed Stearns, S. C. 1992. The Evolution of Life Histories. OUP, harvest and dispersal. Ecol. Monogr. 47: 89–111. Oxford. Visser, M. E., & Lessells, C. M. 2001.The costs of egg Stephens, D.W., & Krebs, J. R. 1986. Foraging Theory. production and incubation in Great Tits (Parus major). University Press, Princeton, NJ. Proc. Roy. Soc. Lond. B 268: 1271–1277. Thompson, D. B. A., & Barnard, C. J. 1984. Prey selection by Watson, J. B. 1908.The behavior of Noddy and Sooty plovers: optimal foraging in mixed-species groups. Anim. Terns. Papers Tortugas Lab. Carnegie Inst.Washington Behav. 32: 534–563. 2: 187–255. Thorpe,W. H. 1958.The learning of song patterns by birds, Weir,A. A. S., Chappell, J., & Kacelnik, A. 2002. Shaping of with especial reference to the song of the Chaffinch hooks in New Caledonian crows. Science 297: 981. Fringilla coelebs. Ibis 100: 535–570. Wiltschko, R., & Wiltschko,W. 2003. Avian navigation: from Tinbergen, N. 1951. The Study of Instinct. Clarendon Press, historical to modern concepts. Anim. Behav. 65: Oxford. 257–272. Dr Angela Turner, School of Biology, University of Nottingham, University Park, Nottingham NG7 2RD; e-mail [email protected]

Looking back

One hundred years ago: whether they were old or young birds. I do not think ‘EARLY NESTING OF THE SHAG IN ORKNEY. THIS there are a dozen records of this bird having visited year the Green Cormorants or Shags (Phalacrocorax Scotland, and the occurrence of a flock of them in graculus) started nesting remarkably early in Orkney. Orkney is most unusual, as I believe the species has They commenced building their nests in January, and only occurred twice before in these islands, viz., a the first eggs were found on February 24th on the young bird near Stromness on September 19th, 1903, Islands of Sules Skerry [sic]. The weather during these which I mentioned in the “Field,” etc., at the time, and months was very stormy but not cold, and perhaps another near Kirkwall as long ago as September, 1857. this latter fact had something to do with their early It will be noticed that the three occurrences were all in nesting, which is much earlier than has ever been September, during the autumn migration. H. W. known in Orkney before. H. W. ROBINSON.’ (Brit. Birds ROBINSON.’ (Brit. Birds 1: 190, November 1907) 1: 189) ‘PALLAS’S SAND-GROUSE IN MIDDLESEX. AT ‘GLOSSY IBISES IN ORKNEY. ON September 24th noon on Monday, September 23rd, when sitting with last, a flock of about twenty Glossy Ibises (Plegadis Mr. Henry M. Hill on the lawn in front of his resi- falcinellus) appeared at Sandwick, a small township dence, Downage, Hendon, I saw a Pallas’s Sand- about four miles inland, and about eight miles from Grouse (Syrrhaptes paradoxus) flying towards us. It Stromness, Orkney, where they frequented some was alone and, at an altitude of not more than sixty marshy ground. The man who made the discovery did feet, passed directly over our heads, giving me a not think of shooting them at first, and so for three splendid opportunity of carefully observing and iden- days they remained in peace, but on the 27th this idea tifying it. It flew rapidly over the house, taking a seemed to strike him, and from this date to October course due north. I have never previously had the 1st he shot two or three each day as they were feeding pleasure of seeing this bird alive, but have had ample in a burn, until he had accounted for no less than ten. opportunities of studying, amongst others, the speci- They were very wary, and rose high in the air when mens which were obtained in Staffordshire, in 1863, disturbed. Most of them were sent, I believe, to Mr. by the late Mr. Samuel Yates of Eccleshall. W. WELLS Mallock, of Perth, and it would be interesting to know BLADEN.’ (Brit. Birds 1: 190, November 1907)

British Birds 100 • November 2007 • 665–679 679