Jump to ContentJump to Main Navigation
The Birds at My TableWhy We Feed Wild Birds and Why It Matters$

Darryl Jones

Print publication date: 2018

Print ISBN-13: 9781501710780

Published to University Press Scholarship Online: September 2018

DOI: 10.7591/cornell/9781501710780.001.0001

Show Summary Details
Page of
Subscriber: null; date: 06 April 2020

What Happens When We Feed?

What Happens When We Feed?

Insights from Supplementary Feeding Studies

(p.137) 5 What Happens When We Feed?
The Birds at My Table

Darryl Jones

Cornell University Press

Abstract and Keywords

The importance and influence of food in the lives of animals has been studied is great detail in a vast number of species. This chapter outlines the many findings of this critical research that are directly relevant to understanding how the provisioning of food for garden birds may be affecting their lives.

Keywords:   Supplementary feeding, Bird feeding, Experiments

“Maggie” is the inevitable nickname given to innumerable Australian Magpies, even though many are likely to be males. They are big, bold birds, abundant throughout the country, especially within the suburbs where the landscape of scattered, tall trees and endless well-watered lawns provides ideal habitat. These birds are very well known to Australians for two main reasons: their complex and evocative territorial song (“caroling”), which seems to capture the essence of an antipodean spring morning; and, in violent contrast, their aggressive aerial assaults on humans trespassing too close to the nests. It may sound far-fetched, but these attacks are extremely common during the breeding season and lead to lots of injuries and anguish each year.1 Yet somehow this vigorous combination of art and aggression seems to appeal to Australian sensibilities as the species is often declared our favorite bird. Half the sporting teams in the country seem to be called the Magpies.

Australian Magpies are also well known because they are the most frequent—and favored—visitors to the feeding tables of Australia. This (p.138) may come as a surprise because while magpies are certainly broad in their dietary tastes (they will eat almost anything), they are primarily insectivores, specializing in worms and grubs gleaned from just below the surface of lawns or grassy fields.2 They have a similar foraging style to that of American Robins and the Common Blackbirds of Europe, although they are about twice the size. Although Australian Magpies will eat seed and bread and many other things, to attract magpies most people place out a much more carnivorous selection: chopped sausage, ham, pieces of bacon, diced heart, and especially beef mince (ground beef) are all commonly used, as well as cheese and pet foods. The contrast with a typical Northern Hemisphere feeding station could hardly be greater: seed and tiny tits or chickadees versus meat and massive magpies.

The discovery of these veritable butcher’s shop smorgasbords throughout the suburbs was an unexpected and significant event in the development of my interest in bird feeding. For some time I have been interested in the wildlife of urban environments and particularly the features that enabled some species to prosper in this strange contrived landscape while many others could not. I was also fascinated by the interactions between people and the otherwise wild creatures they shared the suburbs with. Magpies were an obvious choice for an urban study species: they were abundant, relatively approachable, and most people liked them. And although they had been well studied, to my great surprise this common and familiar bird had not been investigated in its favored habitat, the suburbs. Despite being filled with people (and their cars, cats, and kids), the suburban environment offers plentiful opportunities to an intelligent and resourceful bird. Vast areas of lawns and garden beds for foraging, tall, well-spaced trees for nests and surveillance, and generally fewer aerial predators; no wonder the density of magpies in the suburbs was far higher than in rural landscapes close by.3 But there was more to the story than simply lawn and trees and fewer hawks.

Food and space are two essentials for survival, and they are closely linked in the lives of many animals. For species capable of defending an area—a territory—against others, a home patch needs to contain all the resources required for both day-to-day survival and reproduction. For the majority of birds, the effort required in defending a patch against determined trespassers and intruders potentially seeking to pilfer your resources or make off with your partner is so demanding that territoriality (p.139) is typically limited to a few weeks of the year. The process of finding a suitable area, claiming it as your own (usually through vigorous singing and displays), repelling competitors, attracting a mate, building a nest, producing and incubating eggs, and protecting nestlings from weather and predators, among other duties, is exhausting. Most birds, sensibly, are actively territorial only briefly, ceasing their boundary patrols and moving away from the breeding site once the young are mobile.

Not so the Australian Magpie. Unlike virtually all other songbirds, these birds are permanently territorial. This means remaining continuously vigilant for potential intruders and patrolling the boundaries of their territories every day of the year. This highly unusual feature of magpie life has numerous implications for the birds, including the reality that finding a life partner means really settling down: remaining in the same patch for the rest of their lives. For such a long-lived species (Australian Magpies frequently live for more than 20 years, much longer than a robin, for example), this makes the resources provided within the territory of critical importance: everything needed for survival and reproduction—and shelter and everything else—must be suitable and sufficient more or less permanently. Although we now know that both males and females do make the occasional furtive and apparently secret sojourns out of their territories,4 these escapades are almost always brief; in general, most mated magpies spend most of their lives in the same patch. Given the obvious necessities required, it is not surprising that magpie territories in the suburbs are relatively small compared to that of an out-of-town magpie.5

This was the background to one of our first studies of the ecology of suburban magpies. There were plenty of worthy questions to pursue, but we chose to investigate what was fundamental to them all: What was the food supply of an average territory in the suburban landscape? With my colleagues Tom Nealson and Dan Rollinson, we began the usual process of modern wildlife research: the careful capture of birds so as to measure, weigh, and attach colored leg bands to allow us recognize individuals, and the methodical observations of the birds as they went about their normal lives. These are activities familiar to field biologists studying birds anywhere in the world, except for the fact that we were attempting to conduct our research in the urban environment. And “urban” means people. Plenty of folks came over to see what we were up to. From simple curiosity and genuine interest to skepticism and outright alarm at our actual (p.140) or imagined methods,—we spent a lot of time patiently explaining and listening. But we also managed to obtain some important observations of our magpies.

As has been understood for some time, Australian Magpies (like Common Blackbirds and American Robins) use their ears as much as their eyes to hunt for the grubs and worms that make up the bulk of their diet. Through some remarkable experiments using recordings of these invertebrates moving through soil,6 we now know that those characteristic side-to-side movements of the bird’s head as they concentrate on a spot on the ground immediately in front allow them to accurately locate their prey beneath the surface with pinpoint accuracy. A dramatic pause, a sudden thrust deep into the sward, and the next moment a large worm is thrashing at the end of the bill. It’s an approach to foraging that predisposes birds such as magpies to success in a landscape dominated by lawns and grassy fields. Magpies do very well because these well-watered, fertilized, mowed, bright-green landscapes, so characteristic of suburbs throughout the world, are also ideal for the invertebrates they thrive on.

Our careful observations of the foraging behavior quickly allowed us to form a clear impression of the richness of lawns and sports fields as a food source for magpies. We were able to compare their feeding activities on lawns versus fields, in suburbs versus on farms, at dry times versus wet, and their foraging behavior for their own consumption compared to when they were feeding hungry young. We felt we were beginning to understand the relationship between the availability of natural foods and the success of the species in this human-dominated environment. It was when we started to focus on the chicks in the nest, however, that we began to realize that we were dealing with only part of the story.

Baby birds still in the nest (“nestlings”) are, obviously, entirely reliant on their parents for all their food. The arrival of an adult magpie at the nest was always accompanied by pandemonium among the chicks, as is entirely normal with most birds. The level of noisy confusion in the magpie nests we watched seemed especially pronounced, although this was probably due simply to the larger size of the species and the offspring’s corresponding loudness. We could often hear the magpie nestlings long before we were close to the nest tree. This vocal onslaught must have been deafening for the adults, and grew ever more cacophonous as the chicks grew. Nonetheless, the parents labored away valiantly, returning again (p.141) and again to the nest with their bills full of grubs, worms, caterpillars, moths, and skinks for the three or four ravenous craws in the nest. Watching from a safe distance through a powerful telescope, we were able to observe the continuous circuit of foraging on the ground until the bill was stuffed with squirming baby food, the direct flight back to the nest, a brief pause to unload into one of the noisy gapes, then a long swoop back to earth to start all over again. Male and female worked at almost the same rate, arriving at the nest in turns, on a punishing schedule that continued almost all day and lasted for about five weeks.

Breaks in this relentless routine of the adult magpies were infrequent but noticeable because they usually involved the birds flying out of sight from our particular vantage point. We simply assumed that they were somewhere within the territory that we did not have a direct view of. Subsequent investigation, however, revealed an unexpected discovery: these birds were partaking of the hospitality of some of the feeding stations nearby. We soon found that a large proportion of the pairs we were studying in the suburbs had the same arrangements: a conveniently positioned and well-provisioned larder. Not only that, this was a food source replenished daily, and because it was almost always within the boundaries of the territory, it was the exclusive domain of a local pair of magpies. And some pairs had at least three different stations to call on within their patch—although the people didn’t realize that.

This was a major revelation for us and had both exciting and sobering consequences for our scientific investigations into the foraging ecology of suburban magpies. Our first important—and entirely unexpected—finding came through those careful observations of nestling feeding. As the chicks grew and their apparent hunger became audibly and visibly apparent even to us on the ground beneath, we assumed that the frantic and exhausted parents would take advantage of all that free food available nearby. Certainly the adults were visiting the feeding tables more frequently as the breeding season wore on. As explained previously, however, we found that almost all the food supplied to the nestling was natural invertebrate foods.7 Even though they could so easily have ferried bill-loads of sausage and ground beef from the closest feeding tray, most of the baby food was the natural stuff. The human-provided foods were obviously being consumed by the adults, but even for them, it typically made up only a small proportion of their overall diet. This observation (p.142) too, was unexpected: with so much easily accessible human food around, why didn’t the birds simply forsake foraging for natural items altogether? These were important questions to our limited understanding of how the species was utilizing the available resources.

And our understanding was undoubtedly limited. Although we had made excellent progress in quantifying the productivity (in terms of measures such as worms per square meter) of different lawn types, sports fields, and farm pastures, and had gathered an enormous amount of data on foraging efficiency and chick provisioning,8 the discovery that many of our birds were exploiting a considerable but unknown source of food entirely additional to their regular diet meant that our estimates of feeding resources were clearly incomplete and inaccurate. Attempting to understand the role of this new—“supplementary”—food source in the lives of our birds was now unavoidable. I had only been superficially aware of the many experimental studies into these issues, but now I really needed to find out more. Indeed, for reasons that will become progressively clearer, understanding the influence of bird feeding really starts with this type of research.

Food Is Fundamental

There are few requirements for life as undeniably essential or as influential as food in the lives of all animals. Food is the key to both survival and reproduction. Changes in its availability, quality, and the balance of its components can alter every aspect of their lives. The centrality of food can be seen in the extraordinarily detailed research being conducted into the nutrition of the animals entirely dependent on humans for all their needs. The food requirements of our pets and domesticated species are now understood in minute detail, allowing sophisticated and somewhat disturbing manipulation of such things as muscle development, milk composition, and fertility rates, all via their feed trays.

For wild animals, we know far less. This is unremarkable: there are a lot of species and they all have different diets. Unlike the economic motivation that propels commercial animal production, the reasons for attempting to learn more about the role of food for wildlife are more likely to be driven by scientific curiosity. Pragmatically, that usually means a lot less funding. (p.143) Nonetheless, researchers throughout the world have undertaken a vast number of studies on a bewildering array of species from tadpoles to polar bears, though a large effort has been focused on birds. The cumulative findings of these studies have fundamentally changed our knowledge of the roles that food plays, providing many insights into its function in the ecology and behavior of wild animals. As we will see, these apparently basic biological findings have been applied to many practical problems such as wildlife management and the conservation of threatened species.

Adding a Little Something

In one of the first comprehensive reviews of the research on supplementary feeding, the Canadian biologist Stan Boutin gets straight to the point: “No one would question that individuals and populations are ultimately limited by food supply.”9 The obvious difficulty is how to assess this food supply accurately. Even if we know what a species is supposed to eat, dealing with the daily and seasonal changes in availability, while accounting for competition and interactions between individuals, makes the task extremely complicated and logistically challenging. Most of the observational studies that attempt to account for overall food supply for a species readily acknowledge the limitations and unavoidable compromises.

The main alternative approach to straightforward observations is to intentionally manipulate the food available to an animal, usually by the provision of a suitable food that is extra to what would naturally be eaten. The key point is that this provisioning is additional, or “supplementary” to the background food resource. Some describe this as a feeding “subsidy,” something supplied that, by definition, shifts the food supply to above that of natural levels. From a scientific perspective, such manipulations are often part of experiments designed to test hypotheses and provide potential explanations about the way food affects populations of wild animals. One of the most fruitful and enduring sources of such hypotheses was the research conducted by the famous British ornithologist David Lack, which he collated in two now classic books, The Natural Regulation of Animal Populations (1954) and Ecological Adaptations for Breeding in Birds (1968). Among many influential explanations for the cycles and dynamics seen in wild species was his suggestion (p.144) that the fluctuations in the sizes of populations of many small animals were strongly reliant on the food supply available during the winter, the most taxing season. Lack’s ideas have influenced generations of ornithologists, but the impact of Adaptations was especially pronounced. In 1969, the British Ecological Society brought together researchers to discuss the role of food for animal populations, with one outcome being an increasing awareness of the need for manipulation of food supplies in order to test ideas.10 One of the powerful conclusions of this important meeting was that observations alone, no matter how detailed, cannot provide the explanatory power of a well-designed experiment.

Lack’s ideas were especially linked to two of the most familiar, abundant, and well-studied species in Europe, the Great and Blue Tit. Along with the very similar chickadees of North America, tits (from “titmice,” a general name for all members of the family Paridae) usually do not migrate and so are most vulnerable during the cold months in the temperate zone when winter foods are hard to find. Usefully, these small birds readily take to artificial nest boxes, which greatly assists researchers studying their reproduction. Both tits and chickadees also happily partake of human-supplied foods, another attribute of an ideal study species. Unsurprisingly, almost every aspect of the lives of these confident little birds has been carefully studied, including the role and influence of food. While a great array of species has been included in supplementary feeding experiments around the world, the prominence of tits and chickadees within this research field argues that we should pay them careful attention. That and the fact that these are probably the commonest visitors to bird feeders in the world. I think that we can legitimately nominate the “tit-adees” group as being eminently suitable representatives of the world’s garden birds and spend some time reviewing what researchers have discovered.

Feeding Tits for Science

In and around the town of Ghent in Belgium, researchers have been studying the breeding of Great and Blue Tits using nest boxes first erected in 1959. Since those early days the number and diversity of locations with nest boxes have steadily increased until ten study sites containing over 800 nest boxes were being monitored. These sites spanned the general habitats (p.145) typical of Western Europe: parklands in urban areas, suburbs with various deciduous tree species, and rural lands with mixtures of oak, beech, and pine. In the late 1980s, a group of Belgian scientists led by a young André Dhondt (remember the name) compiled almost two decades of detailed breeding information from these tits.11 The researchers were especially interested in following up an observation made by other European ornithologists: that Great Tits living in city gardens tended to lay their first eggs of the season somewhat earlier than those nesting in nearby woods. The long-term information from Ghent allowed Dhondt and his colleagues to quantify these differences with some precision: across the sites, Great Tits in the urban parks started breeding an average of more than 10 days earlier than those in nearby rural sites. This was a remarkable difference. The effect was far less clear for Blue Tits, however, with the smaller species being later in the countryside but less pronounced in the other habitat types. The difference between the species and the habitats, the researchers speculated, was almost certainly due to the relative amounts of food being supplied by people; feeders were more often associated with urban gardens than were found outside the towns. Understanding whether this was the case would require experiments. As we will see, these were already under way.

At about the same time, ecologists working along the much drier and warmer Mediterranean coast of southern France were interested in the reasons influencing the start of breeding in the same two species, Great and Blue Tits.12 Starting the breeding cycle early would appear to have obvious benefits to the parents: it allows more time to raise and fledge the nestlings, and possibly even start another brood. Earlier broods are also more likely to survive compared to later hatchlings, an outcome associated with having more time to find their main baby food of caterpillars, the most important source of natural protein for tits and many other small birds. Counting your chicks before the eggs are even laid is, however, risky. Females must be in a suitable physiological condition following winter before they can start the extremely taxing task of producing a clutch of eggs, and that would appear to be closely tied to the availability of suitable food resources leading up to the breeding period. To ensure that their females were in the best possible condition the French researchers decided to provide additional—supplementary—supplies from midwinter, long before the actual period in which the females started to develop their eggs. In (p.146) addition, they altered the diet over time: in January, the feeding stations were provisioned with sunflowers and margarine, and a month later, dried insects were added. Then, about a month before the typical laying times, live insect food (mealworms, commonly fed and commercially available beetle larvae) was also supplied, to provide additional protein and energy for females during egg formation. All of these dietary additions stopped with the appearance of the first eggs so as not to influence other phases of the breeding cycle. Another (control) population within the same habitat but without the benefits of all that additional food was also carefully monitored.

By the end of this 2-year experiment, the local tits had consumed a total of 120 kilograms (265 pounds) of sunflowers, 8 kilograms (17 pounds) of margarine, and 19 kilograms (42 pounds) of mealworms. The main result of all that additional, high-quality food was an advance of about 6 days in the earliest eggs being laid by the Blue Tits in the study population. Not so for the Great Tits, however; there was a slight advance, but this was not statistically different from the dates of the unfed birds nearby. These results from the Great Tits are possibly even more surprising than those from the expected effect seen in the smaller species: despite all the additional foods (and both species really did consume most of what was supplied), the Great Tits did not respond in terms of when to lay. The reasons for such perhaps mixed findings were not understood with clarity but seem to indicate that food resources are only some of the cues that stimulate these birds to start breeding. Ornithologists have long argued that a complex suite of triggers may be involved, including changing day length (photoperiod), air temperature, female body condition, innate genetic predispositions, and even the availability of the artificial nest boxes. While these arguments have continued, the general importance of food in the overall breeding equation is clearly acknowledged as one of the main actors in a play with a sizeable cast.

Although similarly mixed results are standard fare for experiments involving birds in nature, it is still possible to claim that for the majority of supplementary feeding studies of titmice (and indeed, most birds),13 some advance in laying date is a typical and expected outcome. This may suggest that birds breed earlier if they have extra food at critical phases of their lives (such as when they are engaged in egg production). However, it is also important to point out that almost all experimental (p.147) studies report variable responses with some fed birds breeding much earlier than others. In addition, there is often considerable overlap in the timing of laying dates among the fed and unfed birds.14 This rather typical finding may be due to the supplied food being insufficient in quality or possibly of little importance relative to the feeding opportunities available to the birds naturally. Don’t forget that the foods being provisioned are, by definition, supplementary to the bird’s regular diet. If the supplies of natural foods that the birds use are already high, the additional supplies may be simply less attractive—and less important in the broader scheme of things. This would indicate that there may be some important threshold value of food resources for many birds, which will result in a strong reaction (such as breeding much earlier) for those living below the level, but a much more muted reaction for those living above. In other words, if there is already enough food, birds may use different cues to begin breeding.15

Does Feeding Change the Timing of Breeding?

When to start breeding is among the most critical decisions any animal can make, but this is especially important for small birds like tits, which are only likely to have a few reproductive opportunities during their brief lives. As already mentioned, starting early has crucial benefits. For example, hatchlings born earlier in the season tend to grow more rapidly, and being in better condition than later-hatched chicks, they are more likely to survive and breed in the following year.16 In contrast, chicks hatching later in the season are often less vigorous and already disadvantaged as they must compete with well-developed young that fledged before they did. The advantages of early hatching are, however, strongly reliant on their arrival in the nest from the egg coinciding with the peak in caterpillar supply, the primary source of nestling nutrition. This peak is typically measured in days, and so any situation that causes the birds to miss it, such as a period of poor weather conditions—or mistiming by the parents—can have catastrophic consequences. If birds respond to the manipulation of their food resources by breeding too many days earlier than normal, they could find themselves out of synch with the baby food supplies. This may be a legitimate concern for bird feeders: missing the peak by laying too early because the parents have responded to an enhanced food (p.148) supply could potentially mean that feeding stations actually disadvantage the local birds. If so, this is an issue of fundamental importance.

Although attitudes—and organizational promotions—as to when to feed are certainly changing (as explored in Chapter 3), most wild bird feeding across the Northern Hemisphere has been typically limited to the winter months. Traditionally, this is probably based on the perception that birds may need some help to survive the tough times but that “it is better that they look after themselves” at other times. Because of the emphasis on investigating the role of food on breeding, a large majority of supplementary feeding experiments have supplied additional food during the early breeding season, and then assessed the effects on the birds’ reproduction that followed. But most of these experiments do not replicate what happens when people feed birds. Does winter feeding actually influence the breeding activities of birds in the following breeding season? Do the effects of feeding carry over into the future?

Gillian Robb, under the guidance of Stuart Bearhop and Jon Blount, explicitly addressed this question in an important field study of Blue Tits in Northern Ireland.17 This study was also significant in that it took a broad landscape-level approach in an attempt to more closely resemble the way feeders are distributed throughout the landscape. Ten separate woodland blocks already established with nest boxes were liberally supplied with peanuts (a commonly used bird food in this region) throughout the winter in one year but not the following year. Crucially, all feeding stopped at least 6 weeks before the first eggs were laid. This ensured that none of the female tits—fed or unfed—was influenced by food supply as they began to form their eggs, although their body condition was undoubtedly enhanced during the year of the study. The results were unexpectedly clear: feeding advanced the average laying date by 2.5 days compared to the nonfeeding sites, and, of particular significance, feeding resulted in more offspring being successfully fledged. Overall, feeding led to about one additional fledgling per nest, a remarkable outcome because this was not associated with any increase in the number of eggs layed or of hatchlings raised, both of which were unaffected. The extra fledglings resulted from their better survival compared to the unfed sites, yet this was obviously not related directly to additional food supply. The researchers put this effect down to the relatively higher body condition of the parents; healthier adults raise more kids. Winter feeding appeared, therefore, to lead to both (p.149) earlier breeding and better survival of chicks. Given that we already know that feeding also improves the likelihood of survival of adults through the winter, these researchers also mention some potential implications of their work.18 For example, places where winter feeding occurs may be producing both more new birds (offspring) and better surviving older birds (their parents) than places without feeders. This in turn may mean that the areas with reliable feeders are likely to support higher densities of longer-lived birds. This may be regarded as a welcome and possibly unexpected byproduct of feeding in winter. For many species, being able to increase their reproductive output is certainly welcome news. But what about the still widespread perspective that we should be leaving the birds to themselves at other times?

As we have already discussed, the “winter-only” attitude is being challenged vigorously. Proponents of year-round feeding point to the apparent conservation benefits of studies such as that described above as why feeding should be continuous. Here we need to ask whether supplementary feeding experiments can shed light on this particular issue, given the proliferation of year-round feeding.

The best study to address this question was published under a title of commendable clarity: “Does food supplementation really enhance productivity of breeding birds?”19 The question is fundamental to the conservation claims at the heart of continuous-feeding arguments and a test of the largely accepted assumption that more feeding leads to more chicks. Focusing on Great and Blue Tits using nest boxes, the experiment offered three feeding menus over a 3-year cycle but with the foods used swapping between sites so that the birds had a different diet each year. The key element of this experiment was 500-gram (17-ounce) blocks of protein-rich peanut cake continuously available from about a month before egg laying until up to 2 months after the end of fledging. Depending on the site and year, the birds received either peanut cake, peanut cake with mealworms added after the nestlings had hatched, or nothing (the control). While not strictly “year-round,” the food was supplied during all phases of the breeding cycle, far longer than any other study on tits. Conducted by Tim Harrison, Jim Reynolds, and Graham Martin, the study took place in woodland sites in Worcestershire, central England.20 Given the quantity and quality of the additional food being provided, the researchers were bold but justified in expecting big things: they anticipated that the fed (p.150) birds would have earlier breeding, larger clutches, shorter incubation periods, enhanced hatching successes, and more chicks per nest. While these predictions tended to be a little optimistic (the effect of additional food has been quite mixed for clutch size, incubation period, hatching success, and brood size in a wide variety of species), very few studies had been as ambitious in the scale of provisioning. As a corollary, few studies were of such potential significance.

As would be reasonably expected by now, the date of laying was earlier and the length of incubation lower in the fed populations of Great and Blue Tits. Advanced laying dates are now a fairly common finding in experiments of this kind, although reduced incubation times are less so. What was not at all expected was the impact of prolonged supplementary feeding on the key components of reproduction: eggs, hatching success, and resulting brood size. Entirely contrary to the predictions, clutch size and brood size in both species were actually less for the fed birds compared to the unfed. Similarly, the proportion of eggs successfully hatching was lower in the fed Blue Tits, although there was no difference in the Great Tits. Thus, in relation to the question raised in the title of this study, the unavoidable answer was actually: “No, supplementation did not really enhance the productivity of the tits,” although this conclusion relates only to this particular experiment. Nonetheless, the findings are reliable and sound and require careful consideration.

The results are also potentially alarming and sobering. They were certainly not anticipated by the experienced research team that conducted the experiment. Although one other supplementary feeding study also reported a decline in the number of eggs laid, it was a very different situation: a far larger, long-lived waterbird (the American Coot)—hardly an ideal comparison.21 Where additional foods have affected clutch size, the result tended to be more eggs, but this is one component of breeding that appears to be difficult to alter. Most feeding studies found no change in clutch size, a finding that is even more pronounced for tits, with well over half of supplementary feeding experiments reporting no change in the number of eggs laid.22

What was far more significant in the Harrison study was the decline in hatching success in Blue Tits and brood size in both species associated with the provision of additional foods. These are unique findings among hundreds of supplementary feeding studies. This means that the extra (p.151) peanut cake somehow led to a reduction in the proportion of eggs in the clutch producing live chicks compared to the unfed nests. The productivity of bird nests may be assessed at two critical intervals: first, at the time the hatchlings emerge from their eggs (“hatching success”), and finally, as young leave the nest as fledglings (“fledging success”). Between these two key milestones, plenty can go wrong. Starvation, predation, weather, and other factors will normally lead to a difference in the number of chicks that hatch and the number of fledglings that leave the nest. What is important for overall breeding productivity is the proportion of fledglings that survive, and that starts with the size of the brood. To have lower numbers of eggs and then fewer hatchlings is a major early blow, especially if this is due to the parents partaking of what should be of clear benefit to breeding.

It is important to remind ourselves that these are findings so far limited to one study in a single location over a limited time period. Nonetheless, the research was intentionally designed to assess the possible impact of more or less continuous feeding, as is increasingly the case in urban landscapes. It is striking, therefore, that some of the key findings of the Harrison study—earlier laying dates, smaller clutches, and reduced brood sizes—are remarkably similar to the breeding attributes of Great and Blue Tits living in typical urban environments in Britain. Compared to birds in largely rural landscapes, urban tits lay fewer eggs earlier and raise significantly smaller broods.23 Furthermore, in a comprehensive review of the reproductive parameters of many bird species—including tits—living in urban areas, advanced laying and lower clutch sizes were found to be typical compared to birds living in nonurban areas,24 a result attributed primarily to the greater availability of human-provided foods. This adds just that much more substance to the significance of the study we have been discussing, as well as raising some real concerns.

What about the Composition of the Food?

An obvious next step toward understanding the way supplementary foods may be influencing the breeding of birds is to see how characteristics of the food itself may play a role. At a rather basic level, the two fundamental components of foods are fats and proteins. Although both are essential, protein is especially important in preparing the female’s body for the (p.152) substantial demands of producing a clutch of eggs. A lack of protein—or, more specifically, particular molecules known as essential amino acids—during the early spring can, potentially, greatly limit the capacity of females to form their eggs, possibly resulting in fewer, smaller, or even poor quality eggs. Although these are common effects associated with seasons with low caterpillar numbers (a main source of natural protein), the same response has been attributed to the consumption of fat-rich but protein-poor foods such as peanut cake.25 Indeed, the relative proportion of fat and protein in many commonly used bird feeder foods is decidedly fat biased; the percentage of fat and protein respectively for peanut cake is 70.5% and 17.1%, peanuts 44.5% and 28.7%, and black sunflower seeds 44.4% and 18.0%.26 If birds are consuming more of these human-provided foods than natural insects, their bodies could be misreading the nutritional cues, with serious implications for breeding activities.

Fats are crucial as short-term energy sources and are almost certainly an important reason for the improved survival over winter for birds using supplementary foods. Most small songbirds, such as tits, however, are unable to store large, complex molecules such as fat in their bodies for any extended times. Instead, the fat-derived energy from all those suet balls and peanuts is utilized fairly promptly. And while winter feeding really does seem to enhance short-term survival, there is also some evidence that fatty diets generally can have longer-term health impacts. We know this from our own waistlines and heart disease statistics. What if this type of provisioning of birds for the tough times was actually detrimental for the birds we are trying to aid? Before we tackle that big question, a more immediate issue is whether supplying fats in winter affects things later in the year when breeding starts. After all, if fats aren’t stored, perhaps there is little or no “carry over.”27

These issues are, as usual, more complicated than they appear. For one thing, while macronutrients can’t be stored for later use, a number of much simpler molecules known as micronutrients can be sequestered away and released when needed. This is especially important for reproduction in small birds because they are strongly reliant on the nutrients and energy immediately available as they start preparing for breeding. The most important of these micronutrients are calcium (needed for shells and bones), vitamin E, and a group called “carotenoids” (the source of most reds, yellows, and orange colors in animals and plants that can be stored (p.153) in the fat reserves).28 These molecules are antioxidants and play important roles in protecting the body from various physiological stresses associated with metabolism. The protective functions are increasingly valuable as the demands on the female’s body build up during egg production. Vitamin E and carotenoids are also known to play critical roles as antioxidants in the developing embryos within the eggs, and the deposition of both micronutrients in the yolk has clear benefits for the subsequent body condition of the hatchlings.

These somewhat technical details are needed as background to discussing the next crucial supplementary feeding experiment. Undertaken by Kate Plummer and colleagues, this particular study is of fundamental importance because it explored the potential carry-over influences of winter feeding on breeding in Blue Tits, with careful attention to the way that the macronutrients (fat) and micronutrients (vitamin E and other carotenoids) affected egg production.29 In other words, the experiment replicated typical winter feeding of small birds throughout the Northern Hemisphere.30 Again, the strength of this research stems from the clever but simple study design employed. The researchers set up three nest box sites in deciduous woodland in the beautiful countryside of Cornwall in the far southwest of England. Three different menus of supplementary foods were used: fat alone (handmade balls of vegetable fat); fat plus vitamin E (added to the ball at the same concentration as that found in peanuts); and, of course, no additional treats at all (the control) as a comparison. These courses were supplied for the winter period only and were stopped a month before the start of the breeding season, well before the birds had even started to think about breeding. Because the key goal was to assess whether the effects of the different winter-feeding regimes carried over into the following seasons, the diet offered to the birds in each study site was swapped each winter over the 3-year study. If there were clear outcomes, these should show up in the eggs laid in each of the sites according to the diet.

Following on from our discussion of the ability of birds to store nutrients needed for breeding, we would probably expect that the provision of fat alone—so early in the cycle—would not affect egg production, while the addition of vitamin E would probably be useful. The findings of the experiment were, again, unexpected. First, contrary to other similar studies, there was no clear change in laying date. Nor was the number of eggs produced or their relative size affected by any of the diets. What was (p.154) thoroughly unexpected, however, was that both the amount of yolk per egg and the level of stored carotenoid were significantly lower for the birds consuming the fat-only diet. These components of the eggs are critically important to the growth and development of the embryo and the health of the chick. The yolk contains all the ingredients needed for the construction and maintenance of the growing offspring within the egg. For this reason, the body condition of the mother bird at the time the eggs are formed appears to be of fundamental importance to the subsequent wellbeing of her young. Therefore, to find that partaking of a fat-heavy diet in winter appears to result in female tits producing impaired yolks—and potentially leading to lower-quality chicks—is of considerable concern.

But what happened when vitamin E was added to the fat balls? Although theoretically some benefit of the addition of this antioxidant was expected, the outcome observed was still striking: vitamin E appeared to cancel out the effect of the fat, with yolk size in the groups of birds receiving this diet being no different from that of birds eating a natural diet.31 How might this come about? Although the actual biochemical interactions involved were not studied, enough is known about fat and antioxidants to be able to offer plausible explanations as to how this might work.

As explained earlier, small birds cannot store macronutrients such as fat for long periods: it is a useful form of energy that enhances the physiological challenges of surviving the cold months. Nonetheless, access to readily available, high-fat foods in winter may mean the birds are less likely to be obtaining a more diverse natural diet. This may also mean they have fewer antioxidants, at a time when they are internalizing plenty of polyunsaturated fats, just the sort of harmful molecules the vitamin E and other carotenoids can neutralize. Although a fatty diet in winter would not have directly influenced the formation of yolk much later, the birds’ bodies appeared to be still dealing with the oxidative stress, which affected their ability to form egg yolks. By adding an antioxidant, this stress seemed to be reduced.

Yes, we certainly need to be careful in extrapolating generalization from a single experiment, however elegant. These are results specific to Blue Tits in Cornwall, and the study has yet to be replicated anywhere else or with other species. Equally, however, it would be sensible to learn from such carefully planned and relevant research that not only looked at a popular feeding practice (providing fat in winter) but may also offer a (p.155) solution (adding an antioxidant). For tits, at least, this may be as simple as throwing some peanuts in with the suet (although, of course, this still needs verification).

Feeding and Survival

Finally, we return to the primary question of the relationship between food and survival in these tiny birds, the key to population regulation as suggested by David Lack, and perhaps the fundamental issue associated with supplementary feeding experiments. Indeed, virtually all the features found to be influenced by the addition of food may be evident in the number of birds remaining in the population in the year following the experiment. If, because of the availability of additional food, more eggs are being laid by healthier females, and if the resulting hatchlings are slightly more likely to make it through to fledgling, and if these are more likely to survive the winter, then it follows that there should be more birds than before the food was provided. Yes, theoretically, but there is plenty that can go wrong too.

As portrayed by Lack, the harsh realities of winter in the Northern Hemisphere’s temperate zone provide the ultimate test of the population that survives to breed in the following spring. As conditions become increasingly tough and foods scarce, competition for what little is available is inevitable; food supplies will pretty much determine survival. Numerous supplementary feeding studies have assessed this directly, providing additional food and monitoring the changes in bird numbers before and after. In an early study conducted near Lund, Sweden, Great Tits at several sites were either provisioned with lots of sunflower seeds (dispensed in large hoppers) or had to make do with completely natural resources.32 The number of pairs were counted in each site over the two following winters and showed dramatic yet dissimilar results. During the first winter of the study, which happened to be particularly severe, populations of tits without additional foods decreased by at least 10%, whereas the number of birds at the fed sites increased by between 20% and 60%, depending on the site. Clearly, supplementary foods were greatly enhancing the ability of the birds to survive the winter. It also greatly improved the chances of young birds hatched the previous spring to remain through the winter and go on to breed. But the story gets more interesting. In the following year, (p.156) all populations, whether fed or not, increased dramatically, with numbers of tits being between 35% and 94% higher than the previous winter. This spectacular result clearly had little to do with all the seed laboriously provided by the researchers. Rather, natural events that year entirely overwhelmed the influence of careful experimental design. The winter months in southern Sweden that year were distinctly mild with considerably less snow, enabling the birds much more opportunity to forage naturally. But far more important, that year the local beech mast crop was large, providing an abundance of easily obtained natural food. In such circumstances, to paraphrase the author of this study: “In a good mast year, there was no effect of supplementary food.”33

By this stage the attentive reader will have noticed, possibly with some affront, that this discussion of feeding experiments, ostensibly covering the titmice group (the parids), has been entirely preoccupied with European tit species. But what of their close North American relatives, the chickadees, and especially the ubiquitous Black-capped Chickadee? Unfortunately, far less supplementary feeding research has been conducted on these extremely popular little birds, a somewhat surprising situation given that they are almost certainly the most abundant species using feeders throughout the United States and Canada. Black-capped Chickadees are especially conspicuous during winter in the more northern parts of their distribution as they are among the smallest birds to remain during winter and consequently flock to feeders in large numbers. The potential value of human-supplied foods during winter for chickadees has long been recognized, but was most famously studied by Susan Smith in the Massachusetts woods during the 1960s. Professor Smith would continue to investigate every aspect of the behavior and ecology of these birds for the next 30 years,34 but it was her first study of overwinter survival that is of particular importance to this discussion.35

To estimate the impact of winter feeding on birds, it is important to compare the number of birds before and after. For most studies, this involves attempting to count birds visually using standard methods that ensure that the place, duration, and searching effort are the same each time. For anyone who has watched flocks of chickadees (or most small, active birds) swarming around feeders, then sweeping into the nearby foliage before trickling back in dribs and drabs, making a reliable “count” is, shall we say, challenging. One of the main problems is that we don’t know how (p.157) often the same birds are being recounted. This inevitably makes comparisons rather sketchy, undermining our ability to make sound claims. The significance of Susan Smith’s remarkable research is that her estimates of numbers before and after winter were unusually robust. She was able to identify individuals among the clouds of otherwise anonymous chickadees because most of the birds had been marked with individually colored leg bands (or “rings”). Although color-banding is thoroughly standard practice for much bird research today, Susan Smith’s ability to recognize specific birds was fairly revolutionary at the time.

Far more prosaically, however, individually marking her chickadees allowed Susan Smith to show that survival over the winter was largely determined by the reliability of the local food supply. Being able to identify known birds confidently at regular intervals throughout the winter and again in the following year allowed careful accounting of the ongoing presence—or sudden disappearance—of each bird. But while Smith could claim that providing additional foods aided chickadee survival through the winter, unfortunately the lack of a “control” site where food was not supplied meant that little could be stated about the relative influence of supplementary foods.

Smith’s studies provided a solid foundation for the next phase of American chickadee supplementary feeding studies. During the 1980s, Margaret Clark Brittingham and Stanley Temple from the University of Wisconsin explicitly addressed the limitations of the earlier work by thoughtful yet straightforward experimental designs.36 In the first of several studies of great importance to our understanding of wild bird feeding, Brittingham and Temple color-banded 576 Black-capped Chickadees in five well-separated study sites in rural Wisconsin.37 Although chickadees are known to be fairly sedentary, remaining close to home year round, it was still remarkable to learn that only two of the almost 600 marked birds were detected more than 2 kilometers from where they were first captured.

Using techniques now familiar to us, chickadees at three of the sites were supplied with a feeder regularly filled with sunflower seeds from October to April, while those at two other sites were offered nothing. This continued for 3 consecutive years except for the crucial variation of stopping the food supply at one feeding site (although, tantalizingly, the empty feeder remained in place) and starting feeding at a site previously without a feeder. As a powerful attempt to understand the impact that the (p.158) provisioning—and withdrawal—of additional foods may have on the survival of tiny nonmobile birds, this should do it. Their persistence as winter progressed was determined by thorough weekly searches of each site for marked birds and regular captures using mist nets. The detailed information collected allowed monthly survival rates to be calculated as well as year-to-year survival. In addition, astonishingly precise data (to within one-hundredth of a gram) on the body masses of the birds were obtained at the time of capture. Though the procedure is not simple, the painstaking efforts required to handle such a small yet spirited creature were well worth it. Perhaps surprisingly, small birds such a chickadees gain weight over the course of a day as they forage and then lose the weight as they use up their meager stores of fat during the night. The difference in mass at dawn may be a much as 10% less than the night before. For birds weighing only about 12 grams (0.4 ounces), these differences may mean the difference between life and death on a long, freezing night.

It was in the body mass data that the first significant result was noted. At the start of the experiment, males and females were on average slightly lighter (0.17 grams, or 0.006 ounces) than later in the winter, but birds with access to supplementary food were just a little (0.13 grams or 0.005 ounces) heavier than the unfed birds. The differences may seem small, but they appeared to translate into significant survival statistics. Chickadees supplied with additional food had, on average, a 95% chance of surviving through to the next month and a 70% chance of making it all the way through the winter. For the unfed birds, this compares reasonably well, with a monthly survival probability of 87%, but contrasts starkly with an overwinter survival of only 37%. The benefits of having access to supplementary food is even more obvious when monthly survival for fed and unfed chickadees is compared during severe winter conditions (periods of more than five days per month with minimum temperatures less than –18 °C [–0.4 °F]). Without the supplements, about a third of the chickadees disappeared (presumably died) during these cold snaps, while only 7% of the fed birds succumbed.

While these stark numbers may convey something of the scientific significance of the experiment, it is the author’s observations of the behavior of the birds that we find especially compelling. Recall that chickadees are tiny, continually active birds, quite unable to store sufficient fat to last more than a full day without foraging. Yet Brittingham and Temple observed the birds during a period of prolonged extreme cold when the (p.159) maximum temperature did not rise above –18 °C (–0.4 °F) and the minimum reached –29 °C (–20.2 °F) for over five consecutive days. Although writing in the objective prose required by scientific publications, the authors cannot disguise the reality of what they observed:

During such periods of extreme cold, the behavior of the chickadees on control sites [the nonfed birds] changed. We could not find most individuals, and the few we located were extremely lethargic. They sat motionless, with feathers fluffed out, facing the sun.38

The researchers speculated that this marked inactivity, so unlike the usual demeanor of the species, was their only remaining physical response to the severe cold. To move would expend energy they did not have; by remaining motionless they were conserving the little reserves they retained. Whether this was a strategy that worked would be measured by the brutal calculus of each bird’s fat reserves, the duration of the extreme conditions, and the availability of suitable nutrition once the weather improved. In the frigid midwinter woods of Wisconsin, the odds would not seem favorable.

Unless there was a feeder nearby. With access to an unending supply of sunflower seeds, the behavior of the chickadees lucky enough to live near such a bonanza was shockingly different from that of their hard-pressed and unfed colleagues. Despite the severity of the prolonged bitter conditions, “[fed] birds continued to use the feeders normally.”39 The contrast could hardly be clearer: the additional food made a very real life and death difference to these birds.

These are strong findings, but are not dissimilar to other winter-feeding studies on similar species. For example, overwinter survival of two species we have not yet mentioned, Crested Tits and Willow Tits, studied in Sweden was almost double that of birds without access to supplementary foods.40 It should be noted, however, that some studies that have investigated the effect of additional foods on winter survival have not reported such clear findings.41 Indeed, some experiments report no influence on survival or on any of a long list of the other parameters mentioned above. These studies may be just as important as those reporting dramatic effects and point to the realities of scientific field studies: the results rarely play out as simply as expected. This is a theme we will return to at the close of this chapter.

(p.160) Changes in Behavior

To conclude this selective review of supplementary feeding experiments on tits and chickadees, we turn to the possible influences on perhaps the most malleable and immediate features of their lives, their behavior. It is one thing to measure physical attributes such as egg size and body mass; it’s a very different matter to ask questions about what the birds are actually doing in response to the additional food.

Given the importance of food to the day-to-day survival of these small birds, particularly during tough conditions, it would hardly be surprising to find feeding can change their social interactions. Although an obviously social group, spending almost all their time in loose groups, tits and chickadees tend to forage alone when seeking their natural diet of insects and seeds. They may be close to the rest of their flock, but the type of food they glean means that it is best obtained without the interference of other birds. During periods of natural food abundance, when there is little competition between individuals, the birds are fairly tolerant of one another’s presence. As conditions become more difficult, as occurs with the start of cooler weather, relationships often become less amicable. Wilson looked at such interactions among her Black-capped Chickadees during the harsh winters in Maine when feeders were experimentally introduced.42 In natural situations, the onset of cold conditions led to increasingly more assertive behaviors by individuals with smaller foraging areas being defended against other chickadees. These patches were established throughout the site, with each bird actively patrolling what was now a small but private “foraging territory.” Trespassing was strictly policed, each bird the master of its own modest domain.

The introduction of feeders abruptly changed all that. With the sudden arrival of a super-abundant food bonanza at just a few localized spots, the former system of territorial enforcement broke down entirely. For the birds living close to the feeders, whatever perceptions of outrageous good fortune they may have had soon dissolved in the reality of the inevitable influx of all and sundry to the feeder. Although many of the birds initially attempted to maintain their boundaries, the constant crossing by birds drawn to the feeders soon led to territoriality being given up completely. For these small birds, there was simply no way that an individual could successfully defend “their” feeder. In other situations (discussed elsewhere (p.161) in this book), usually where the species are much larger and more capable of effective defense, the introduction of feeders can lead to a serious escalation of territorial behavior.

Another influence of supplementary foods on foraging behavior relates to joining feeding flocks made up of different species. While it is generally rare for most birds to deliberately seek the company of other species, the formation of groups of similarly sized birds of a variety of species has been reported from around the world. These flocks are temporary and are usually made up of groups of insectivorous species that occur together in the same woodland or forest habitat. Although there is much we don’t know about these strange amalgamations, most ornithologists believe that birds join them as a way of improving their ability to find food in times of scarcity. This idea has been tested by researchers working on tits in the UK,43 as well as yet another parid species, the Varied Tit, in Japan.44 In both cases, providing additional foods led to a significant decline in the tendency of birds to join these mixed-species flocks. Interestingly, birds partaking of the supplementary foods were more likely to forage alone when away from the feeder.

One of the most conspicuous behaviors of songbirds such as tits and chickadees is their vigorous singing, particularly during the early stages of the breeding season. As pleasant as these sounds may be to our ears, the functions of these impassioned displays are complex and deadly serious for the (usually) male participants. While female songbirds also vocalize, it is almost always the males that produce most of the early morning noise. While we can never be entirely certain of what these vocalizations mean to the birds themselves, countless studies have established that birds use song for a range of purposes but especially to improve their chances of reproduction. Songs can be about territorial defense, the proclamation of identity and occupation of a space, as well as an invitation to mate. Such sweet melodies (to us) may be adamantly “Keep away” (intended for the ears of other males) as well as a vocal résumé of prowess to be assessed by potential female partners. Yes, both of these general functions can operate at the same time, although once the male singer has successfully attracted a mate, the song’s territorial function becomes his key modus operandi.

The introduction of a feeder into this scenario is likely to have a number of consequences. As we have already described, birds near a rich food source may become far more assertive in their territorial defense as they (p.162) attempt to repel the inevitable increase in intruders. In plenty of studies, males with territories near feeders physically attacked trespassers and sang more often.45 Fighting and singing are both activities requiring considerable energetic resources and normally take up important time that could be otherwise spent looking for food. With a regular supply close by, males near feeders have the fuel to sing more than those forced to forage naturally. This food-enhanced singing may have real implications because females of both Blue Tits and Black-capped Chickadees have been shown to prefer males that sing earlier in the day and who sing more.46 Indeed, this attraction continues even after pairs have been formed, with Blue Tit offspring not fathered by a female’s mate most likely to be the apparently sexy singer up the road.47 Unsurprisingly then, supplementary feeding has resulted in earlier dawn singing and greater song output in a number of species (including Common Blackbirds in the UK48 and Silvereyes in New Zealand49).

Working within the suburbs of Oslo in Norway, Katja Saggese and colleagues were interested in the possible changes to the vocal behavior of Great Tits associated with supplementary feeding and the possible ongoing long-term effects of feeding.50 These researchers provided ad libitum sunflower seeds and fat balls to their birds and recorded the songs of males either side of dawn, noting the exact time of all relevant events. Great Tits typically start to sing vigorously at least six weeks before egg laying, well before dawn. Would the fed birds start even earlier? Or would the extra food lead to more singing as predicted by previous studies? To the researchers’ considerable surprise, male tits with access to additional foods actually started to sing later than unfed birds. And they did not make up for sleeping in by singing more; their song output was no different from their hungrier colleagues. Again, these are important findings that deserve some consideration. The people who conducted this study were refreshingly candid about their results being unexpected, but they did offer some plausible explanations.

Waking well before dawn, by definition, means moving about in virtual darkness, and it has been suggested that attempting to find their insect foods at this time would be very difficult.51 Singing, with the various benefits mentioned above, may actually be a more sensible activity. Birds with access to feeders, on the other hand, would probably be able to feed even in poor light and may be feeding instead of singing. In addition, what of the (p.163) apparent reproductive advantages of early singing mentioned previously? To ignore aspects such as attracting the attentions of other females suggests that males may be weighing up their options. For example, having access to such a rich and reliable food resource may itself be a powerful inducement. Alternatively, females paired with feeder-owning males may be less inclined to wander. In reality, these are all simply speculations requiring further careful study. Yet again, this nicely executed feeding experiment has delivered intriguing findings that hint at the complexity of the situation. Again we have to conclude that feeding changes things in unexpected ways.

The bulk of information presented in this chapter has focused intentionally on supplementary feeding experiments conducted on tits and chickadees. This somewhat justifies our focus on this group, but there is obviously more to the field than these selective summaries suggest. For example, the significance of supplementary feeding experiments to wildlife management and conservation are such that we devote another chapter to these major themes. To conclude this exploration of supplementary feeding research on garden birds, however, we turn to some key studies of species other than titmice.

Scrutinizing Scrub Jays

One of the more unusual species visiting bird feeders in North America is the Florida Scrub Jay. Although a member of the superabundant and familiar corvid group—the crows, ravens, and (true) magpies—the Florida Scrub Jay is in serious trouble. At home in the sandy, dry scrublands of central Florida, this exceptionally intelligent and highly social species faces habitat losses and degradation due to expanding housing developments, hotels, golf courses, and citrus plantations. As a result, the species has been steadily declining over the past few decades and currently exists in a small number of isolated populations—always a situation that raises the chance of local extinction. Plenty of other species face similar predicaments, and our understanding of their situation is limited. By contrast, we know a lot about the behavior and ecology of Florida Scrub Jays, and our knowledge provides an unusually detailed background for developing sound conservation plans. A critical component of this information is based on numerous supplementary feeding studies.

(p.164) I described Florida Scrub Jays as “unusual” feeder visitors not just because they are so obviously different from the more typical garden species and their dire conservation status but also because of their remarkable social lives. There cannot be many places in the world where people living in the suburbs play host to a threatened species that also lives in permanent extended families, with all the associated dramas. It was the intriguing communal breeding arrangement of Florida Scrub Jays that led to the pioneering field studies by the late Glen Woolfenden and John Fitzpatrick (now at the Cornell Lab of Ornithology) conducted at the Archbold Biological Station in central Florida. These now famed studies, which commenced in the 1970s, described the remarkably complex social relationships based around the presence and active involvement of individuals other than the breeding pair.52 So-called helpers at the nest, these birds were discovered to be the young from previous years who remained with their parents to assist in the raising of the next batch of young. Woolfenden and Fitzpatrick’s influential research was instrumental in kicking off worldwide interest in the phenomenon now known as “cooperative breeding.”53

Stephan Schoech, Reed Bowman, and Jim Seymour are three researchers engaged in long-term studies of Florida Scrub Jays who have taken a particular interest in the influence of food and nutrition in the lives of these birds. In an initial study in the 1990s, Schoech supplied dog food, peanuts, and mealworms (a diet fairly similar to that available to the Scrub Jays in nearby suburban developments in Highlands County) to some groups and not to others, all birds living near the Archbold Biological Station.54 After decades of continuous interaction with researchers, the birds were only too willing to participate in the study. As Schoech explained: “Group members quickly learned where the feeding station was and if they were not already waiting for me, rapidly responded to my whistle.”55 The results were spectacular: breeding was advanced by an average of 16 days. (Recall that we were excited about differences of just a few days in fed tits.) Obviously, the provision of additional foods is of primary importance to decisions of when to start breeding in this species. Interestingly, the study also tested one of the key hypotheses about the causes of cooperative breeding: helpers don’t breed because there is insufficient food. However, although two of the fed female helpers did forsake the life of an au pair, found a partner, and went on to breed themselves, (p.165) the others (thirteen of fifteen) did not. It would appear that among Scrub Jays, there is more to such a fundamental change in behavior than simply food supply.

These findings have obvious implications for Scrub Jays with access to the well-filled suburban feeders, increasingly available as suburban developments expand in central Florida. Generally, these birds do not live in suburban areas but visit periodically from their territories in the scrublands nearby. This proximity to a year-round supply of foods such as pet food, peanuts, and food waste led to about a third of the diet of breeding female Scrub Jays being provided by people.56 As expected, these birds started to breed earlier compared to their scrubland relatives.57 Presumably, this was due to the relative reliability and higher fat content of the human-provided foods, but teasing this apart requires more control over the variable; that is, a proper supplementary feeding experiment.

Jim Reynolds from the University of Birmingham in the UK (whom we met earlier) is a long-term collaborator with extensive experience working with the Scrub Jays. An experiment he conducted with Stephan Schoech and Reed Bowman was designed to assess the influence of different fat and protein levels within the foods provided.58 Of course, these issues had been previously studied by other researchers, using the usual commercial seeds as supplements and assuming standard nutrition content. What Reynolds and his colleagues wanted was far greater control of the actual levels of the fats, proteins, and carbohydrates present in the foods provided. To do so, they had their own made, arranging for a commercial operator to produce vast amounts of little cylindrical pellets to a very specific recipe. These were of two “flavors”: high fat and high protein, or high fat and low protein. Both types had, however, identical energy content. This was supplementary feeding science on a new level altogether.

Clever, yes, but only if the Scrub Jays agreed that the odd-looking, gray-brown pellets appearing in their feeders were edible. After all, they looked nothing like the dog food, peanuts, or mealworms they were used to. This particularly adventurous species had no hesitation, however; they loved the pellets and the experiment was off to a good start.

As usual, Scrub Jay territories were randomly assigned to either high fat, low fat, or no fat (that is, the “control” group, who were offered welcoming but empty feeders). Supplementary feeding started in midwinter and continued until just before the start of laying. In the second year, everything (p.166) was again randomized and no group received the same diet two years in a row. Again, the results were remarkable and, because of the sophistication of the experimental design, the particular dietary influences could be teased apart. First, as we would now expect, the start of breeding was considerably earlier for fed birds compared to the unfed, although again the extent was extreme: on average almost two weeks earlier. What was unexpected, and to my knowledge has not been shown in any other supplementary feeding study, were similarly strong effects on three other fundamental breeding characteristics: clutch size, egg mass, and egg composition. Unfed Scrub Jays produced an average of 2.8 eggs while those on the low-protein diet laid 3.8 eggs and those on high protein laid 3.5 eggs, although this difference only occurred during the first year. This is especially remarkable given that the normal clutch size for this species is only three.

But simply laying more eggs does not necessarily translate into more chicks: the survival of hatchlings is closely associated with both the size of the egg and the way it is provisioned for subsequent embryo development by the female. In most birds, egg size declines steadily with each additional egg produced, with the last laid being the least likely to survive. While a steady decline in the mass of eggs as each was laid did occur in each of the three dietary treatments, Scrub Jays on the high-protein diet were able to maintain the mass so that the third egg produced was actually much heavier than the equivalent egg laid by each of the unfed females.

Studies such as this Scrub Jay experiment indicate an increasing interest by researchers in some of the finer details of the relationship between food and a wide range of aspects of animals lives. Carefully designed and well-conducted experiments on wild species have been essential to uncovering the links in this complex chain of influence and consequence. Additional food almost always changes something. As we have discovered here, supplementary feeding experiments have also revealed unexpected and unwelcome findings. Some of these results may well alter the way we think about and practice our own wild bird feeding.

Maggie’s Meat

As a final example of the way that our feeding can affect garden birds we return to the unusual scene at a typical feeding station in an Australian back yard. Instead of sunflowers, peanuts, and suet there is often (p.167) salami, ground beef, and cheese. And that’s because rather than birds like tits, chickadees, or even Florida Scrub Jays, the most likely visitors are all large, assertive, and often—strangely enough—black and white, with a distinct taste for meat (though just about anything vaguely edible will do). These pied picnickers may include (Pied and Gray) Butcherbirds and (Pied) Currawongs, as well as (the non-pied) Laughing Kookaburras. But far more probable are the Australian Magpies introduced at the beginning of the chapter. Not only are these big, bold birds the commonest species at Australia feeders, they are also the most welcome. More people try to attract “Maggies” than any other species: hence all that meat.

As demonstrated by numerous examples described already, researchers are often intrigued, if not concerned, by the physiological impacts of anthropogenic foods on wild birds. Such was the case with the overly meaty diet of Australian Magpies being fed throughout the suburbs of the country. In an important experiment conducted by Go Ishigame and colleagues from the University of Queensland, three types of common foods—minced meat (ground beef), pressed dog food (“dog sausage”), and shredded cheese—were supplied to groups of free-ranging magpies.59 These were typical foods available at feeders used by local magpies. Experimental provisioning occurred over a series of several month-long sessions, during which each group received one food type or none. Rather than the usual breeding parameters we are now familiar with, the focus of this experiment was on blood chemistry. Critics of wild bird feeding have suggested that some of the foods being consumed by urban birds may be harmful to their health. Certainly, the foods offered—and accepted—by magpies could be included in this category. To assess this, at the end of each monthly feeding session, Ishigame and company caught the birds and obtained tiny blood samples, which were tested for several parameters.

Interestingly, both cholesterol and uric acids levels were found to have increased significantly but only for the dog sausage. Both parameters are well known as indicators of negative health conditions. The result is also noteworthy because the birds consumed similar amounts (about 40 grams [1.4 ounces] per day) of each food type, yet the mince and the cheese had no discernible effect. This finding may be associated with the relative level of commercial processing involved, with both the meat and cheese being considerably simpler foods compared to the dog food. Indeed, the fact that this item is manufactured and marketed as a balanced pet food—albeit obviously for dogs—but produced a marked spike in cholesterol (p.168) levels in birds may be of some concern. And while plenty of magpies are known to partake of such pet foods—both as supplied by feeders as well as by sneaking bits from pet bowls—there are also vast numbers of birds being provisioned daily with processed meats such as salami, German sausage, and other cholesterol and fat-rich fare. And we already know what happens when these foods are eaten to excess in human diets.

Food for Thought

That’s probably enough examples of supplementary feeding experiments for now. Those featured here are only a fraction of the many such studies that have been published, but they represent a fair selection of those that relate directly to urban birds. As we are obviously interested in understanding the influences of providing additional food to wild garden birds, those dealing with other bird groups or topics associated with conservation and wildlife management were not included here. Furthermore, the studies featured have tended to be those with the more marked results. That is hardly unusual as it’s always easier to write about—and more interesting to read—the research where the findings were spectacular or unexpected. Nonetheless, it is also important to be aware that plenty of well-designed and carefully conducted experiments did not find an effect. I suspect that many such studies simply do not get published, leading to a definite bias toward positive and significant findings in the available literature. Some, however, have made it through the rigors of the peer-review process and these also need to be acknowledged.

Although there are several excellent reviews of supplementary feeding experiments available (see Boutin’s 199060 review for many examples), a particularly readable summary of relevance to this discussion is that of Gillian Robb and colleagues (whose study of winter feeding of tits in Ireland we described earlier).61 In their article, titled “Food for thought,” the authors reviewed as many supplementary feeding studies on birds as they could find and determined whether the addition of food resulted in either a positive or negative or no response on each of seven commonly measured breeding parameters. Although these researchers looked at a number of different bird groups, we will concentrate on their “small passerines” category, as the most representative of the species that come to feeders.

(p.169) The breeding parameters featured in “Food for thought” were as follows: laying date, clutch size, egg size or quality, incubation time, hatching success, chick growth rate, and fledging success. Considering the direction (positive, negative, or none) of the response of all these features together, two conclusions are immediately clear. First, negative responses to supplementary feeding were very rare; only a single study, reporting a later laying date with feeding, was cited. Second, the most common response, by far, to supplementary feeding among small passerines was none. Only for laying date and fledging success did positive responses outnumber no response. While this was most pronounced for laying date effects, with eighteen out of twenty-eight studies reporting earlier egg production due to feeding, almost a third of studies found no effect. Bear in mind that an advance in the timing of breeding is certainly the most publicized effect associated with the provision of additional food. These studies suggest that this commonly reported outcome is far from ubiquitous. Indeed, when all the bird groups included in the review are considered, thirty-four studies reported earlier breeding while twenty-four—that’s 40%—found no change at all. For the only other breeding parameter where a majority of studies found a positive response—fledging success—the ratio was even closer: ten were positive compared to nine with no response. Food for thought indeed.

Good News and Otherwise

We have traversed a lot of territory in this chapter, providing perhaps too much detail to remember. The main message to be gained from these experiments is that feeding garden birds often really does change things. For example, feeding does assist survival through the winter and especially during periods of prolonged severe conditions and lean years when natural foods may be scarce. Winter feeding also leads to better survival of chicks, due in part to the elevated body condition of the breeding females. Providing additional food may therefore lead to the presence of more birds locally, through the attraction of visitors to an abundant foraging resource as well as through the enhanced breeding of the residents. All of this is surely good news. However, these changes may also include unwelcome outcomes we did not expect.

(p.170) One of the roles of scientific rigor is to ensure that we see beyond what we hope is happening or what we might logically expect. Certainly, some of the researchers conducting these supplementary feeding studies were surprised—and sometimes alarmed—at their own findings. For example, to discover that a study replicating typical continuous (“year-round”) feeding practices actually led to fewer offspring is of considerable concern. This is because the experimental results appear to mirror what is now widely appreciated more generally: that suburban birds tend to lay fewer eggs and have smaller broods than rural populations of the same species. Feeding is directly implicated. Similarly, studies mimicking the fat-heavy diets of many typical feeders found that this can impair yolk and embryo quality, with real risks to the survival chances of the hatchlings. Again, a worrying result, although this important study may also have come up with a possible solution (the addition of an antioxidant such as vitamin E).

This is a suitably sobering note on which to conclude our discussion of the science of supplementary feeding. As we have seen repeatedly in the examples described here, a proper scientific approach requires a genuine willingness by researchers to be critical, clever, and open-minded. When conducted with appropriate rigor and attention to detail, careful experimentation may be essential for progressing ideas beyond what is possible through observations alone. This is certainly the case for anyone seeking to understand the apparently obvious, actually complex interaction between birds and the foods we provide for them.

So, in conclusion: feeding almost always changes things although sometimes in ways that are different from what we were expecting.


(1.) A summary of our research on Australian Magpies and especially their aggression toward humans is provided in Jones 2002.

(2.) See Jones 2002.

(3.) See Jones 2002.

(4.) This surprising finding was reported in Hughes et al. 2003.

(5.) Details provided in Jones 2002.

(7.) Our findings are presented in O’Leary and Jones 2006.

(8.) These studies are summarized in Jones 2002.

(9.) See Boutin 1990.

(11.) This is described in detail by Dhondt et al. 1984.

(13.) An excellent synthesis of this large body of work is provided by Robb et al. 2008a.

(15.) Two foundational papers on these topics are Drent and Daan 1980 and Meijer and Drent 1999. Genuinely essential readings.

(16.) A key reference for this topic is Robb et al. 2008a.

(17.) This important study was published as Robb et al. 2008b.

(21.) See Hill 1988.

(p.296) (24.) This key reference is Chamberlain et al. 2009.

(28.) The standard reference for all aspects of nutrition in wildlife generally is still Robbins 1983.

(34.) See Smith 1991.

(35.) See Smith 1967.

(43.) See Grubb 1987.

(53.) A comprehensive review of the issues is given in Hatchwell 2009.