Winter World: The Ingenuity of Animal Survival - Bernd Heinrich (2003)
THE KINGLET’S FEATHERS
The wind whipping through the northern spruce-fir forest sounds like pounding surf, even as the thermometer routinely reads -20°C—and sometimes -30°C. I wear wool pants, two sweaters, a windbreaker, a woolen cap, gloves with liners, wool stockings, and insulated boots. My fingers become stiff in minutes when I take off my gloves. Clothing is essential to staying alive, even in the daytime. The cold is no mere abstraction. How do kinglets, who are out there day and night and who are no bigger than the end of my thumb, maintain their body temperature near 43° to 44°C? They maintain a body temperature some 3°C higher than that of most birds. For added perspective, that’s 6° to 7°C higher than that of a healthy human—a temperature at which most of us would die of heat stroke.
Puffed out in its drab-olive garb, a hummingbird-size kinglet looks like a fuzzy little ball. The physics of heating and cooling dictate that small objects cool quickly, since every point within them is close to the surface where heat is lost. The smaller the animal, the proportionally larger is its surface area, which is the drain whereby it loses heat. How can these tiny birds possibly survive even five minutes on a winter day? I know they are there.
After searching in the woods only an hour or so on any morning, I can usually hear the thin tsees of a nearby kinglet. I feel wonder, and even after exploring in the woods on hundreds of cold days, I’m still awestruck that anything as small and as dependent on keeping warm can survive. Their calls reassure me that they survived yet another night, and ultimately I must believe my senses more than my rationality.
Golden-crowned kinglets, like humans, are colloquially called “warm-blooded” animals. For us, the problem of surviving winter is similar: how to keep from freezing and have enough energy left after paying heating costs. But for the kinglet, the problem is much more severe, because the greater the difference between body and air temperature, the more energy must be expended to keep warm. Furthermore, the smaller the animal, the higher the energy cost per given body mass.
Good insulation reduces energy costs, but there are limits on the amount of insulation that a small creature can carry. Large mammals, such as musk oxen, wolves, and arctic foxes, put on thick winter coats that insulate them so well that they may not need to shiver on the coldest nights. Snowshoe hares and red squirrels also put on a thicker, though comparably modest, insulating coat in winter (underground hibernators do not have seasonal changes of fur thickness), while still smaller animals generally do not become better insulated in winter.
Birds vary their insulation less by exchanging their garb than by changing how they use it. To conserve heat they fluff out, thereby increasing the depth of the insulating air layer that surrounds them. Foot and leg temperatures stay low, regulated just above the freezing point. (We try to keep our feet warm, and pay a high energy cost for doing so.) When sleeping, kinglets insulate themselves even more by tucking their heads and feet into their inch-thick layer of feathers which, from inside to outside, can maintain the astounding difference between body and air temperature of up to 78°C.
To find out how quickly a fully feathered kinglet loses body heat, I experimentally heated a dead kinglet and then measured its cooling rate. In still air the body temperature of the heated bird dropped 0.037°C per minute for every 1°C difference between body and air temperature. Thus at an air temperature of -34°C a kinglet that maintains a steady 78°C difference between air and body temperature at its normal body temperature of 44°C during activity would have a passive cooling rate of 78 × 0.037°C/min. = 2.89°C/minute. The heat production in a live bird that is required to oppose such cooling can be calculated by multiplying cooling rate by body weight and by the specific heat of flesh (0.8 calories/gram/°C). This calculation showed that a kinglet (with feathers) must expend at least 13 calories per minute to stay warm at -34°C. This is a conservative estimate because a normally active bird would experience moving air, or wind, that would greatly increase the rate of heat loss. Given the above data I could now find out how large a role the kinglet’s feathers served in insulation—how much energy they save the bird.
I measured the plucked kinglet’s cooling rate. My naked kinglet had a 250 percent greater rapid cooling rate than fully feathered ones. That is, a naked bird would experience air temperature at least two to three times colder than a feathered one. Due to its small size, a kinglet would also cool approximately sixty times faster than a naked 150-pound pig. (I was once asked to testify in a murder case—involving a naked human body that was found chilled—on how long the victim could have been dead. Not wishing to experiment on the human body, I used an appropriate animal substitute. I bought a freshly killed, still-warm pig, shaved it, and cooled it to get the appropriate data. Perhaps not coincidentally, I was not called in as a witness when I, an empiricist, told the lawyer my results and conclusions.)
Because of their habit of fluffing out in the winter, kinglets look especially plump. But a plucked one looks like a pink cherry on spindly legs. The naked female that I examined weighed 5.43 grams, and she had shed 0.403 gram body feathers and 0.095 gram wing and tail feathers. Thus, she had about 4 times greater feather mass committed to insulation than to flight.
Golden-crowned kinglet, close to life-size.
At the time, my naked body weight was 155 pounds. Fully clothed (in a light synthetic L. L. Bean jacket and boots) I came to 166 pounds. Six of these pounds were attributable to the boots alone. The bird had added 7.4 percent to her body weight for insulation, which was about twice the percentage that I had added to insulate myself. However, more than half (55 percent) of the 11 pounds of insulation I wore were designated for my feet. The kinglet, by comparison, used none of its insulation for that purpose.
My little plucked kinglet reminded me of a scaled-down dinosaur. That’s not coincidental, as the affinities between some dinosaurs and birds have long been noted. What could not so easily be compared was body temperature, which is, unfortunately, a long-presumed defining difference between birds and reptiles. Since birds are increasingly considered evolutionary descendants of ancient reptiles, it has seemed logical, in evolutionary biology circles, to presume that their ability to regulate body temperature is also a recent, more highly evolved trait. This I dispute. Regulation of a high body temperature is no new thing for us presumably superior warm-blooded vertebrates. It is routine in all large flying insects, ancient animals whose line predates the dinosaurs. Recent discoveries in insects prove that regulation of a high body temperature with the aid of internally generated heat by shivering, which is maintained within the body with the aid of insulation, is not as newly evolved a capacity as was commonly thought. The most parsimonious conclusion, given their size and flight capacity, is that some dragonflies or dragonfly-like insects were endotherms—able to store heat—at least 300 million years ago. And at the present time many other insects of a great taxonomic diversity still regulate body temperature, using internally generated heat, insulation, counter-and alternate-current mechanisms of heat retention and loss, evaporative cooling, basking, and alternating of activity patterns. Others don’t regulate their body temperature at all. Whether or not any one does depends almost trivially on body size and lifestyle. What this suggests then is that thermoregulation can be altered through evolution to suit specific circumstances with little respect to ancient ancestors, as such. Insects also tell us that in small animals, endothermy (the process by which internally generated heat is stored to maintain body temperature) is associated with insulation. In many bees and some other insects, the insulation is created from hairlike projections. In moths it is a thick pile created from modified scales instead, and in dragonflies it is a layer of air sacs surrounding the body. In mammals it is fur, and in birds, of course, it is what we call feathers, structures that hold the key to bird body temperature regulation and associated lifestyle. But of course feathers are of special importance, because some feathers also serve another, totally different function—they are the birds’ passport to flight.
THERE ARE LONG DEBATES in the scientific literature about endothermy in dinosaurs and about the evolution of bird flight. The original rationales were constrained by a multiplicity of simplistic presumptions: Dinosaurs are “reptiles” and present-day reptiles are not insulated and therefore dinosaurs were presumed to be cold-blooded. It was a shock, when in 1861, just two years after Charles Darwin published On the Origin of Species by Natural Selection, workmen in a limestone quarry in southern Bavaria discovered a fossil that was clearly a little dinosaur complete with teeth and a long tail, and yet was also a bird, because it showed imprints of feathers exquisitely preserved in the 150-million-year-old Cretaceous limestone. Archaeopteryx, or ancient bird, as it was dubbed, was and still is the oldest bird fossil ever found. Having large feathers on its front limbs, this ancient bird was likely capable of at least rudimentary flight. But what were the precursors of those feathers? If they were derived from insulation, then even more ancient birds than this one were endothermic. (Unfortunately, subsequently discovered bird fossils have been younger, and nobody has ever found the “missing link” feathers to resolve this question.)
In the 1990s, however, spectacular bird fossils, aged 124 to 147 million years, were uncovered from volcanic ash sediments in China’s Liaoning Province. These did not provide direct-line-of-decent type evidence, but they are providing new insights into the evolution of feathers (though not necessarily into the timing and branching points of various possible scenarios of dinosaur-bird relationships). One of the first Liaoning protobirds recovered, named Confuciornis, also had the same type of feathers as fully modern birds and it was even more closely related to modern birds than was Archaeopteryx. In 1996, however, the same deposits also yielded a small bipedal birdlike theropod dinosaur named Sinosauropteryx. Like its relative the Velociraptor dinosaur, Sinosauropteryx had sharp teeth, a long tail, and sharp claws. It was a predator with hind limbs built for fast running. Most significantly, though, the body of this nonbird had featherlike structures as shown by clear imprint in the fine-grained volcanic ash within which it was exquisitely preserved. Of special note: the feathers on the front limbs of this fossil were totally inadequate to support flight. That is, none of the feathers of this primitive protobird were likely used for flight. Sinosauropteryx (and other subsequently discovered dinosaurs with similar feathery structures) strongly suggests, therefore, that feathers for flight originated from insulation. The big question was and is: How can downy insulation evolve into flight feathers? I had not given this question much thought but one night in early March 2002, when I was at my cabin in the Maine woods and felt it shake in the wind and heard the pounding of the rain on the roof at night, an idea just popped up out of nowhere, and I present it here.
It had already been a warm winter—by Maine standards, and on the night in question, temperatures were only a degree or two above freezing. So instead of blizzarding, it was pouring rain. After a day of hiking in the woods, my boots had soaked up moisture, and I still had cold feet. My thoughts turned from my feet to Jack London’s story of the cheechako who died because of wet feet. The trite and obvious was suddenly steeped with meaning: The kinglet’s insulation can do wonders, but if that insulation is wetted then the bird might as well be naked for all of the good it does. That is why a blizzard and subzero temperatures are preferable to being subjected to a cold rain. Almost any rain is a cold, potentially lethal rain. Rain must be a severe test for a small endothermic bird in particular, because the fluffier (and generally the better) the insulation, the more it could act like a sponge to suck out the heat and the life. (Insects simply cool down, and insulation is then irrelevant.) Enduring and surviving wetness must have been a large selective pressure in the evolution of birds when they became continuously or nearly continuously endothermic. How did they evolve to meet this challenge? By the use of wing feathers? Is it a coincidence that survival of grouse chicks in Maine depends on rain-free weather in the time before they can fly?
Wing as “raincoat” to protect insulating down.
In a number of the flightless dinosaur fossils we can observe patches of featherlike structures on the arms and shoulders, and these feathers are more tightly organized into flat and regular patterns. They are attached to the posterior surface of the ulna, a bone in the arm, rather than over the rest of the body. However, these already modified feathers could not possibly have been used for flight, because they were only 5 to 7 centimeters long. Could they have served instead analogously to how I had seen banana leaves used by other, more latter-day bipedal hikers in the rain in the tropics, who held them over their backs while walking to help shed water? Could such long flat feathers act like a rain guard to retain the insulating capacity of the feathers underneath? Can feathers that one second may be used in flight help channel water off the back in the next? Can the wing feathers act like thatch roofing by helping water slide off along the tight regular patterns of the feather veins?
Down feathers (enlarged).
Water-shedding, shading, and flight feather.
Fundamentally, it’s a problem of finding several likely functions of feathers that could have been a bridge between two opposite operations. A prerequisite for any such crossover of function in the bird’s evolution of flight is the original fluffy downlike insulating feathers of an endothermic dinosaurlike creature must have become even more useful as they became less insulative (by becoming longer and flattened), long before their utility in supporting flight became possible. Several ideas for the function of stiffened, flattened, and elongated feathers on forelimbs have been proposed. These include using such feathers on the wings as enlarged fingers to help capture prey, to enhance sexual signaling, to boost running speed or maneuverability, and to aid tree-climbing capacity to escape predators.
Shielding downy young under mother’s wing from rain, cold, and sun.
I’m not claiming that adding another hypothesis—that birds evolved flat feathers on their forelimbs as parasols to reduce wetting of the insulation—adds clarification. It’s hard to know now in retrospect, after a hundred million years. On the other hand, it is perhaps almost too obvious that when feather venules are hooked together into a sheetlike structure (as they are in flight feathers and body contour feathers), then they do provide a barrier to both air and water simultaneously. When the forelimbs with such strategically placed feathers are held dorsally over the back, then almost any elaboration of feathers in this direction provides some parasol effect that would help shed water and reduce the wetting of insulative down feathers underneath. If elaborated on further, then eventually such a parasol would need relatively little additional development before becoming useful in several others of the aforementioned functions, including flight itself. It is analogous to a flying squirrel’s fluffy, insulating tail having the hairs arranged into a flat aerofoil. As I thought about the kinglets perched somewhere in the cold with rainwater flowing off their long wing feathers to protect their down underneath, it seemed to me that regardless of whether the parasol theory could resolve a long-standing evolutionary puzzle, it might at least help explain how kinglets survive a stormy night.
Newly hatched chicken.
Ten days old—now less under hen’s wings.