One Wild Bird at a Time

by Bernd Heinrich

  Chickadees usually forage in loose, species-mixed flocks of from two or three to over a dozen individuals. As a routine exercise in my walks and stays in the woods over the last twenty years I had collected data on winter flock size, composition, and behavior. Since 1980 when I started deer hunting in the fall and teaching winter ecology in Maine, I have conducted most of this late-season birdwatching while sitting in a tree, usually the same one, a large balsam fir. Over the last thirty-four years I have spent 1,700 or more hours in that tree, opportunistically watching chickadees plus the other birds that sometimes accompanied them: several golden-crowned kinglets, a brown creeper or two, a pair of red-breasted nuthatches, and occasionally a single downy woodpecker (but never any other kind of woodpecker). I didn’t and don’t know what brought the various species together or who was attracted to whom and when, why, and for how long. But I did discover a good reason why chickadees are at times found closely associated with red-breasted nuthatches.

  Red-breasted nuthatches live in conifers, and in the winter they feed on the trees’ seeds, which they pull out of the cones hanging from the topmost twigs. In Maine, a variety of conifers—red and white spruce, white pine, hemlock, and tamarack—have seeds available in winter, though not every winter. For reasons that are not understood but may involve subtle weather cues, trees of any one species tend to flower (and later fruit) synchronously, but not all species produce seeds at the same time. Chickadees are associated most conspicuously with red-breasted nuthatches in years of huge winter seed crops. A reason for this association is not hard to guess: when the nuthatches pick at the cones in the treetops they cause seeds to spill, and the chickadees forage for the seeds scattered on the snow below. It’s an example not only of having a broad diet but also of the same alertness that allows chickadees to find and adapt to a birdfeeder stocked with seed.

  Chickadee on a cedar bough, with microlepidopteran moth caterpillars inside the evergreen leaves.

  The winter of 2010–2011 was unusual in my part of Maine in that no trees of any kind were in seed. It was thus a natural experiment. One of the first, perhaps predictable, results was that the woods were empty of red-breasted nuthatches, which had presumably migrated to look for food elsewhere. Another result, perhaps not so predictable, was that while the chickadees were still there, they appeared in smaller flocks. I saw flocks in October, but by November most chickadees I encountered were alone or in groups of only two or three. Had they previously gathered in groups near the nuthatches simply because they were attracted by the dropped food? In addition, what were the chickadees feeding on now?

  I already knew that chickadees are quick learners and flexible in behavior. They build their nests mostly in the deep woods in cavities they hammer out of rotting wood. Most of the cavities I had seen them make were in dead birch stubs. It is a mystery to me how they penetrate the solid leathery bark of these trees, as if knowing that under the thick bark they may find soft, rotting wood. How can they know, given that a live tree’s outer bark is identical to that of a dead one but the living wood underneath is far too solid for them to excavate? Since they start holes and then leave them, may hack out a nest cavity fifteen meters up in a sugar maple or poplar stub, and accept almost any kind of bird box, they are likely experimenters able to adjust to a wide range of options. Whatever cavity they use, inside it they build a deep and cozy nest—of moss, animal hair, plant fiber (such as the fuzz on opening fiddlehead ferns), and down feathers—to harbor their clutch of six to eight eggs and young.

  Finding their nests is always fun. The challenge is the unpredictability. My latest discovery was especially so. As I walked up the path toward my cabin, I saw a chickadee with a green caterpillar in its bill and stopped to watch it, hoping to see it slip into a nest hole. But it stayed near me and kept flitting from branch to branch, so I knew the nest was close. I spotted a likely site, a dead gray birch stub, investigated it closely, and was surprised not to find the expected nest entrance. Seeing no other potential nest site nearby, I drew back to gain more distance from the bird—and saw it fly to the birch stub I had just investigated. I had been fooled: the nest entrance these birds had hammered out was at ground level. My surprise at their choice made me smile, and the memory was precious, so I made a drawing of it. (The next year, in 2015, a pair nested in a less picturesque location, a bird box in the open on a post alongside the garden. Red ants raided the nest the day the young hatched, but three days later the parents were nesting in another box near it.)

  How the chickadee had found this and presumably many other little green caterpillars to feed its babies nestled at the foot of the birch stump was not the result of a simple hard-wired response. Caterpillars such as those prized by birds have a bag of tricks that make them hard to locate. They roll themselves up in leaves, or camouflage themselves by aligning themselves along the leaf centers so they look like the midrib, or mimic the leaf edges, to which they attach themselves to fill in for the part of the leaf they remove. Sometimes the main clue to a caterpillar’s presence is the feeding damage it causes on a leaf. And even then, the caterpillars that are hunted by birds have evolved tricks that help foil their predators. After feeding they distance themselves from the evidence of that activity left on leaves, or they clip off the leaf remnant and then hide while digesting their food. Still, feed they must, and leaving some evidence is unavoidable. I wondered what clues chickadees used to locate the almost invisible prey. And in the summer of 1981, in the woods next to my cabin in Maine, a colleague and I discovered the answer by experiment.

  My fellow biologist Scott Collins and I captured chickadees and housed them in a screened enclosure that we built in the woods. It had two separate parts, each 2.5 meters high and covering an area of 30 square meters. We placed tarps on the bottom of both compartments to discourage the birds from trying to find food on the ground. One compartment, used as the holding area, enclosed naturally growing small conifers and other perches and had a closeable flap to and from the other compartment, which was used as the training and testing area. We held confined only one bird at a time (and released it immediately after a series of tests), except for one mated pair along with the birch stump containing their nest with five nearly grown young. The other birds in the experiments, which we knew were all males because they sang, we individually identified as Ralph, Frank, Fernald, and Duke.

  Each day we cut and placed in the training/testing compartment twenty leafy trees (chokecherry or birch) one to two meters high. In two of the trees (placed at a different location each day) we made holes in some of their 700–1,100 leaves with a paper punch, or (in some tests) the leaves had naturally occurring leaf damage made by feeding caterpillars. For the training procedure we tied segments of rose twigs onto the experimental trees (those with holes) and impaled half a mealworm on each thorn. For the tests we removed all remaining bits of mealworm to find out if and when the birds might use leaf damage as a hunting cue.

  The chickadees immediately examined leaves closely when we released them into the training/testing arena, but we found large individual differences in their behavior. The mated pair and Duke showed highly significant preference for hopping to and foraging on the trees with damaged leaves on the very first trial, before having received any rewards from our training, suggesting that they had already self-trained in the wild. Fernald, who did little inspection of individual leaves, usually flew directly to the tree with the damaged leaves when let into the foraging arena. Ralph, on the other hand, who at the beginning showed a preference for foraging on the ground and a slight avoidance of the trees with damaged leaves, after ten learning trials switched to show a highly significant preference for trees with damaged leaves.

  A chickadee observing leaf damage.

  Since our chickadees learned to hunt the caterpillars by favoring trees with fake caterpillar feeding tracks, and subsequently were also able to differentiate between different kinds of trees, different kinds of caterpillars, and even different kinds of leaf
damage, perhaps they could also learn about novel sources of food from one another.

  Learning involves the ability to remember, a proven faculty for chickadees, which is useful in the winter when they store food in caches for later retrieval. And this faculty is not irrelevant to humans. Fernando Nottebohm of Rockefeller University and his co-workers discovered that when birds learn to sing, and also when adult chickadees in the wild start to store seeds and exercise memory to retrieve them, their growth of brain cells accelerates, and when they stop those activities, nerve cell deaths follow. We had been taught that we humans start losing brain cells in our twenties and keep losing them all our lives. But these findings about birds suggest that exercising our minds may create more brain cells at any time in life. (I’m not sure, though, if researchers who study humans make a distinction between those who maintain a chickadee lifestyle and those who don’t.)

  Following chickadees in the winter over the years, when the trees had shed their leaves, I often wondered what they fed on after they ran out of stored sunflower seeds, or memory storage space. It seemed there was no obvious focus on specific objects in their foraging; they were pretty much everywhere examining pretty much everything, just as our tested chickadees did when we released them from the holding area into the compartment with twenty trees with potential food. While following a chickadee flock in the forest in the winter, I usually saw individuals alternately visiting the bare tops of tall maples; the crowns of pines, fir, and spruce; and the vegetation close to or even on the ground. They picked at lichens, dead leaves, little twigs, bark, large branches, and tree trunks of almost any species. I suspected that a large part of their foraging involved continuous exploration of everything so they might identify any potential food. They did not specialize in any specific kind or part of the tree.

  From left to right: Chickadee egg, just-hatched young, and half-grown young.

  But then I encountered a glaring exception that would prove the rule. Late in the winter of 2010–2011 I began to see flocks of chickadees foraging daily in a white cedar swamp, where I had not seen them linger before (or since). What were they finding near the tops of the cedar trees day after day?

  The snow under the cedars was strewn with bits of dead twig tips, and it was easy to see that the chickadees were picking at such tips in the trees. I followed one flock of seven as it traveled to, and worked on, one cedar after another. These birds eventually left the cedars and went to the tops of red maple trees, where they continued to pick and dislodge debris onto the snow, then left the maples and went to the understory of some balsam fir trees. All the while, they stayed together in a loose flock.

  Wondering what attracted the birds to the cedar tips, I took some cedar branches back to the cabin for a close look. Examining the discolored cedar tips under the microscope, I found minute caterpillars inside almost every browned leaf. The small leaf-mining caterpillars were those of a tiny gray moth, Argyresthia thuiella (of the microlepidopteran moth family Yponomeutidae). They overwinter at the tips of branches inside the scalelike leaves of the eastern white cedar, killing the twig tip and making it prone to dislodge from the tree when birds peck at it. They pupate in spring and the moths emerge in June. Might the chickadees have been feeding on them? Given that chickadees in other years had concentrated much of their foraging effort on nuthatch-spilled conifer seed, it seemed unlikely that they were now eating frozen caterpillars (recent temperatures had dipped to −25°C). And I had no way to see what the birds high in the cedar trees were ingesting. But later that month, thanks to a northern shrike and an accident, I found out.

  The shrike had visited my clearing when a group of chickadees was near the birdfeeder. I noticed it only after seeing the chickadees either panicked and fleeing or perched frozen in place on a twig. One hit a window and was killed at once. Not wanting to waste a bird, I skinned it, noting that it was fully muscled and had small amounts of fat under the skin near the throat and tail areas. Instead of sunflower seeds, its bean-sized gizzard held a mass of stringy matter mixed with dark bits and pieces that could have been of either plant or animal origin. I could not tell what this material was, but when I spread out the jumble of bits and pieces in alcohol in a petri dish under the microscope and rummaged around in the partially digested “stew” for a half-hour, I began to have an inkling of what I was looking at.

  The first clue came from what looked like an insect spiracle (an external respiratory opening) embedded in transparent skin-like tissue. A detached jointed leg confirmed that it was from an arthropod. Perhaps the bird had eaten spiders. Then I found unmistakable proof of a caterpillar: a head capsule. Eventually I also found a caterpillar leg attached to something long, fibrous, and transparent that had the spiracle and looked like a collapsed plastic bag. Eureka! I knew then that they were caterpillar exoskeletons emptied of their contents. There were many of them: I counted twenty-three. I also found other tiny, hard, serrated objects of at least three different kinds, then one of them attached to a caterpillar head capsule. Eureka again—they were mandibles of very tiny caterpillars.

  Caterpillar mandibles, like those of other insects that chew hard material, have serrated edges that are analogous to our teeth. The dentition pattern is specific to a species. I found three kinds of it, so the bird had consumed at least three species of caterpillars. I immediately reexamined the cedar tip caterpillars: their mandibles matched one of the three kinds I had found in the bird’s gizzard. Here was near proof that the chickadees were feeding on at least three species of caterpillars, one of which was the microlepidopteran caterpillars from the cedar leaves. There was only one local source for that species: the trees in the white cedar swamp where I had seen the chickadees routinely.

  Remnants of one chickadee's stomach contents, showing caterpillar hard parts.

  It seemed possible that a chickadee could have happened upon these caterpillars. But flocks of them? I was now left with another question: How had the many chickadees discovered this food hidden within the cedar leaves? How had they learned to associate the brown leaf tips with caterpillars that were not visible?

  I wondered if chickadees automatically examined anything that they had not seen before. To find out, I did a quick experiment with some chickadees that had been coming to the suet I had put out near the cabin door. After they had come to feed for two days I removed the food and provided instead a variety of items, including a red coffee mug, a beer bottle, a red-handled corkscrew, a marshmallow, a piece of wax, a matchbook, some cooked squash, a piece of a polypore fungus, and some coffee grounds. To my surprise the chickadees did not approach the visually most striking objects. Instead, a couple of them hopped close to and briefly examined a pile of coffee grounds, a marshmallow, and a squash rind. But the items that drew them immediately were some grease I poured onto the snow, a piece of red meat, and a bacon rind hanging from a twig.

  Three chickadees that had been coming to the suet were hopping and flying about in a perpetual motion, as is typical of their species. However, on one occasion while I was watching them from the window they stopped, frozen in place, and stayed so for several minutes. I suspected that a hawk had come near, but instead it was the aforementioned great northern shrike. It was perched on the top of a maple tree about two hundred meters from the cabin. At that distance I at first mistook it for a blue jay: jays and chickadees routinely come to the same feeder and ignore each other even when relatively close together. The chickadees obviously had not made that mistake; shrikes routinely hunt songbirds. I seldom see more than one shrike in any one winter. I therefore suspected that it was not long experience that enabled each chickadee to recognize a shrike from a long distance and distinguish it from a blue jay, and that it was unlikely that all of them would have spotted and recognized it nearly at once. The chickadees’ reaction must have been in part a social response. And if chickadees could learn about a dangerous enemy from one another, why not also the often quite specific appropriate food? It seemed credible that a chick
adee or two had discovered the caterpillars within the leaf tissue of the cedar leaves and that new knowledge had spread to the rest of at least one flock.

  It was not until three years later, on December 15, 2014, that I saw the chickadees’ reliance on each other in action. Next to my south window a patch of vegetation was sticking up out of the snow. There was a spruce tree with a young pin cherry coming up beside it. There were dried stems of fireweed, grass, and goldenrod. And directly in front of my desk through the window, I saw a chickadee pecking, pulling, and tearing into something held within a dry curly brown leaf. If not for this chickadee, I would never have noticed that leaf, because it looked to me just like the others attached to the goldenrod and fireweed stalks. The other chickadees, over ten meters away at the feeder, seemed not to have identified it as of interest either. But quickly one came within a meter of the first bird, hopped around near it, and in a few seconds was joined by four others. I had, in weeks of watching, never seen even one at that location. The birds’ heads were swiveling in all directions and they were picking at asymmetric shapes on leaves and twigs. For a time they left the first chickadee alone as it continued to peck at the debris, but after pecking here and there and finding nothing, they came over to the one that had found something. I too wanted to know what it had found, and rushed out as soon as the bird had left, and picked up a now totally empty spider egg case. The others could not have come near this bird because they saw food—they had come because they saw the other’s behavior: acting as though it had found food.

  Two days later I again was at the window. I had replaced the remnant of the spider egg case, and there were no bird tracks on the snow next to it. But now I noticed a similar silk package on the same vegetation-clad bank—because I saw a chickadee fly to it. This bird spent only three or four seconds pecking at it and then flew on, but a second bird immediately took its place and stayed a minute, and then another, another, and still another. The first bird that had pecked at it had not managed to penetrate it and had found no food. After the second bird departed I went outside to gather the object of interest, which turned out to be an egg case of a garden spider, Argiope aurantia. There were only a few tatters on it; it was not penetrated.