School of Biosciences

As biologists, we pride ourselves that we understand why plants and animals are the way they are. Natural selection is the process that ensures that organisms with the most effective functional designs survive. For example, we believe that poorly camouflaged female Mallards perish because they attract the attention of predators, especially when they are most vulnerable e.g. when incubating eggs and caring for young. By contrast, females with genes for camouflaged plumage survive better. In the case of skunks, individuals with genes for the conspicuous black and white pelage pattern survive better because predators learn to associate the pattern with an unpleasant, foul-smelling secretion, and therefore avoid individuals with such a pattern.

Drake Shelduck with a quite immaculate breast band, though there is some irregularity at the top of the band towards the front of the body. The bill knob is very small in this individual.

Yet natural selection explains only a minority of the colours and patterns we observe in wild animals. Many have bright and conspicuous patterns that do not provide camouflage, and are not associated with distastefulness, poisons or other dangerous properties. Neither do they mimic other animals that are dangerous. An examination of any field guide shows that a high proportion of male birds falls into this category. There must be a good reason why the plumage of the drake Mallard is divided into nicely separated blocks of brightly coloured green, purple, black and vermiculated feathers. When drakes are flightless during the moult, they develop a special eclipse plumage, which resembles that of females, presumably in order to escape detection by predators. But why do they have such a conspicuous and characteristic plumage pattern for the rest of the year? All we know is that sexual selection has at least something to do with it.

Recent research has shown that some colours (including those produced by both melanin and carotenoid pigments) are favoured by sexual selection since they may indicate good genes or good parenting skills. Much less attention has been paid to the patterns of plumage and pelage in this respect. I believe that many patterns are the result of sexual selection and that it is possible to explain the form such patterns take, and their precise location on the body.

Female Shelduck with a breast band of poor quality — the colour is not uniform and the border is irregular.

One example of this sort of pattern is that formed by the border between two differently coloured areas of fur or feathers. Such borders may be sharply defined and form a straight line or smooth curve, which tends to deteriorate in regularity in individuals in poor condition. It is convenient to describe individuals with a regular border of this type as being more ‘immaculate’ (see the Shelduck and Great Tit photographs).

We have shown that the red chest-band of both male and female Shelducks is more immaculate in the very small proportion of individuals (about 5%) in the Severn Estuary that manage to produce chicks. Such individuals frequent the mudflats with the richest feeding resources during spring when the females are forming eggs, and nest at the largest breeding colonies, such as that on Flat Holm island (Ferns and Lang 2003). We have also found that the drakes with the largest bill knobs have the best feeding territories, as indicated by the fact that their female partners are able to achieve optimal body condition despite devoting less time to feeding (Ferns et al. 2005).

The breast band borders of most individual Shelducks are immaculate immediately after the moult in the autumn. It deteriorates throughout the winter both as a consequence of wear during feeding, and as result of fighting between individuals. It thus provides both ducks and drakes with an indication of whether a potential breeding partner will be able to obtain and hold on to a good quality feeding site and a safe breeding burrow in the face of competition from other individuals.

Male Great Tit of high quality with an immaculate right cheek patch.

The white cheek patches provide a similar signal of quality in the Great Tit. In this case, we know the situation is more complicated because, as well as assortative mating and more successful breeding amongst immaculate individuals, we have experimental evidence that natural selection plays a role too. This role is a consequence of intraspecific competition in which less immaculate individuals (both those already in the population, and those rendered less immaculate by dyeing them) were forced to use a feeding site subject to greater predation by Sparrowhawks in winter (Ferns and Hinsley, 2004). The situation is probably very similar in principle to that in the Shelduck, it is just that we don’t have the direct evidence of the importance of intraspecific competition in the latter species. Late moulting is one of several factors that reduces feather quality and therefore accelerates wear, leading to a loss of immaculateness (Hinsley et al. 2003).

Intraspecific competition is probably a major factor involved in the evolution of plumage patterns in birds, and in many cases it operates through both natural selection and sexual selection, thus blurring the distinction between them. These forms of selection lead to immaculate borders becoming located in the most revelatory positions on the body i.e. those parts subjected to the greatest amount of damage. During intraspecific competition, animals will use whatever weapons are available to them, and this makes it likely that the nature of the weapons determines the nature and location of the patterns. For example, the delicate bills of most sandpipers make unsuitable weapons, and thus sandpipers are largely devoid of immaculate plumage borders. Many plovers, by contrast, have powerful short bills and wing spurs or clubs that they use in intraspecific contests. They also have sharply-defined borders between different areas of plumage that reveal injuries incurred in such contests (Ferns 2003).

Male Great Tit of low quality with an irregular right cheek patch.

Factors other than intraspecific competition may also account for the loss of immaculateness. For example, colours and patterns that render parasitised individuals less immaculate may be favoured by selection in a social species, if they allow conspecifics to avoid parasitism by not associating with such individuals. In this case, it is disadvantageous to parasitised individuals to display such plumage, but they may have no choice if they want to join a conspecific flock. Similarly, only careless or unlucky individuals have close encounters with predators, but in both cases they are best avoided as mates and companions. Consequently, plumage and pelage patterns that bear permanent witness to such encounters, e.g. zebra stripes, may evolve as a consequence of the social choices of other individuals. Regular patterns of stripes show up the slightest displacement, such as that caused by tearing of the skin during a close encounter with a predator. Once again, individuals with immaculate patterns should be favoured by other herd members, whilst those with disrupted stripe patterns and other scars should be avoided, and thus forced to the herd periphery or excluded altogether, and avoided as mates.

Quite a lot of recent research has demonstrated that the yellow colour of the chest and belly plumage of Great Tits and Blue Tits is brighter in high quality habitats and individuals. However, our work has been the first to show that this yellow colour reveals two sorts of information about quality. Firstly, the average colour saturation (chroma) can provide information on the local timing of breeding and moulting, such as that associated with woodland type e.g. parkland versus deciduous woodland, and the saturation of particular individuals reflects their likely parental quality within the local area (especially the lack of breeding experience in juveniles). Secondly, the colour hue (wavelength) can provide information on the quality of the local food supply e.g. caterpillar abundance, and each individual’s capacity to exploit such food efficiently (Ferns and Hinsley, 2007).

Grants

NERC Centre for Ecology and Hydrology
review of bird species and habitat interactions in Welsh terrestrial lowland environments, review of birds attracted to gardens in North America.
Cardiff Bay Harbour Authority
encouragement of chironomid midge predators to Cardiff Bay.
Cardiff County Council
monitoring the colonisation of Cardiff Bay wetlands.

Collaborations

With NERC’s Centre for Ecology and Hydrology we have carried out literature reviews of bird feeder usage in North American gardens (for an expanding international company), and of Welsh lowland birds (for the Welsh Assembly Government).

With the Centre for Ecology and Hydrology, the University of Aberdeen and Anglia Polytechnic University we have measured the foraging costs of Great Tits and Blue Tits breeding in different habitats. Doubly labelled water was used to measure the energy expenditure of females feeding approximately 10-day old young in woodland of different quality. Birds at our nestbox colony in Bute Park, Cardiff have a low breeding success due to the lack of caterpillars. This means that they expend about twice as much energy per chick as birds in good quality woodland habitat. Consequently they only manage to rear two or three chicks per nest in poor years, and seven to nine in good ones. This is despite the fact that they make an early start to egg-laying because of the greater warmth afforded by the city centre environment.

Our monitoring of the changes following the closure of the Cardiff Bay barrage are drawing to a close. James Reed has just finished a program of sampling in the wetland reserve within the bay, and is writing up an account of the changes in birds, invertebrates and vegetation brought about by barrage closure. Two studies with the British Trust for Ornithology and the Centre for Ecology and Hydrology of some of the implications of the barrage for Redshank populations have been published. Chironomid populations no longer pose the acute problems they did immediately after barrage closure, but many of the House Martin nestbox colonies we set up to ameliorate the problem continue to thrive.

Collaboration with Mark Jervis continues on a comparative study of critical life history parameters in parasitoids and Lepidoptera. My input comprises the application of phylogenetic regression, independent contrasts and other modern comparative analytical methods to the data. Other collaborators in this work include C.L. Boggs (Stanford University, California, USA), and G.E. Heimpel (University of Minnesota, USA).

A study with the University of Oxford and the Centre for Ecology and Hydrology has just started into the well established link between pesticide residues and eggshell thinning in Sparrowhawks. It turns out that the colour of the eggs, which we measured in Cardiff, tends to be greener when they have higher DDE levels, offering a potentially quick, easy and non-destructive way of monitoring environmental contamination.

Affiliated Staff

Research students

Lang, N.A.
USA external
2003
Behaviour of the Bald Eagles during the breeding season with an emphasis on captive management. San Francisco Zoo funded.

Reed, J.P.
Current
Mitigation and compensation for the loss of intertidal wetlands: a case study of Cardiff Bay.

Ross-Smith, V.
Current
Learning to forage in young Lesser Black-backed Gulls.

Others

Rae Vernon, an amateur ornithologist who carried out pioneering research on gulls in the Bristol Channel, has donated his reprint collection to the School. He has asked that they be made available to any researcher, amateur or professional, who wishes to borrow them.