In Search of the Phantom Tanager
A bird obsessive chases a maddening truth: the more elusive the prize, the more tantalizing the hunt
Outside's long reads email newsletter features our strongest writing, most ambitious reporting, and award-winning storytelling about the outdoors. Sign up today.
THERE WAS NOTHING SUBTLE about the smell. When the breeze shifted, pushing the damp morning air into our faces, the stench walloped us like a physical blow. “Whew, man,” said Doug Wechsler, our expedition photographer. “Smells like something died.”
“Lots of somethings,” I said, trying to breathe through my mouth.
We were walking along a narrow, packed-dirt road in the hinterlands of Mato Grosso, in western Brazil, less than a hundred miles from the Bolivian border. On either side of us, low forest covered the rolling hills, with a wide band of tall grass fringing the road for a hundred yards or so. Just ahead, several yellow-headed vultures formed a mourning party, and as we got closer, they flapped languorously into the air.
Now we could hear the buzz of flies, and realized the smell was coming from a roadside dump full of cow hides, entrails, heads, and partial skeletons, apparently the offal of a butchering operation at one of the ranches we’d passed a few miles back. Unfortunately, the trail we needed to take was almost buried by the stinking heap.
A few minutes later, we reached the edge of the woods upwind of the dump and gasped clean air. Fabiano Olivera, a 21-year-old biology student from the university in Cuiabá, paused and nudged a white object with his foot—the shoulder blade of a cow, one of hundreds of bleached bones lying in great piles. We’d hired Fabiano, a slim young man with a long brown ponytail, to be a driver and guide, a job he often takes between semesters, but he was coming to realize we were not his typical ecotourists. The three of us were at the start of an expedition that would, over the course of several weeks, carry us across more than a thousand miles of bad road, forest, highland savanna, degraded farming land, and lush river valleys, searching for an enigmatic bird that had been seen only once, more than 60 years ago, in a place no one today can pinpoint. There were times when I felt it might make more sense to stand in the dark with a burlap sack, yelling, “Here, snipe, snipe, snipe!” But that came later.
We were looking for a songbird, one of the most famous lost species of South America. On August 25, 1938, a Frenchman named Dr. A. Vellard traveled through Mato Grosso and, like many amateur naturalists in those days, he gathered a small collection of birds typical of the region, shooting and stuffing a few dozen species he encountered. Later he showed his skins to the prominent French ornithologist Jacques Berlioz, who realized that one of them—somewhat larger than a sparrow, glossy black with a white belly and a small patch of white on each wing like the corner of a handkerchief poking out of a man’s dark suit—was a new species. Berlioz named it the cone-billed tanager.
That is all we know about the bird—the physical description of a single male specimen (the female remains unseen and undescribed). No one has reported a cone-billed tanager since 1938, and after more than 60 years, it is considered extinct, perhaps pushed over the edge by massive habitat destruction in large areas of Mato Grosso. But Brazil is a huge country, and Mato Grosso itself is vast—at 352,400 square miles, the region is almost a third larger than the state of Texas. Immense areas of it have rarely if ever been visited by ornithologists, and a six-inch bird would be easy to overlook.
NOW THAT WE WERE in the field, though, we were beginning to appreciate the sheer weight of geography that was against us. The cone-billed tanager, as near as anyone can tell, was a creature of the woodland edge, and we’d been looking for several days for just this kind of low, dry forest habitat on sandy soil. Moving in single file, we followed a meager trail through the forest, Doug hauling a voluminous blue backpack from whose top protruded a telephoto lens the size of a kitchen waste can. I’ve known Doug professionally for more than a decade, and after I began working on a book about the search for animals that might or might not be extinct, we started planning a collaborative hunt in South America. Tall and lanky, with a boyishness that belies his 49 years, Doug was the ideal partner—a trained ornithologist and a respected wildlife photographer, he directs VIREO (Visual Resources for Ornithology), an ambitious photo archive at the Academy of Natural Sciences in Philadelphia that will house photographic documentation of all the world’s 9,600 or so species of birds.
I had my heavy tape recorder slung over one shoulder and an 18-inch-long shotgun microphone hung in a leather holster at my belt. While Doug played a tape of the black-and-white tanager (the conebill’s nearest relative, which we hoped might entice the rarer bird into view), I was recording unidentified birdsongs and playing them back, usually luring the singer into view as it hunted for what it assumed was an intruder.
The conebill’s there-and-gone history is unusual, but by no means unique. There are a number of South American birds that turned up in specimen collections in the 19th and early 20th centuries, never to be seen again—like the buff-cheeked tody-flycatcher, a tiny bird with a name longer than itself, collected just once, in 1830 in Brazil. Yet despite the long stretches since their original discoveries, some birds have recently been relocated; they include the Tumaco seedeater, first collected in 1912 on a tiny island off the Colombian coast and not found there again until the mid-1990s. A case that gave us particular hope was that of the cherry-throated tanager from eastern Brazil—first recorded in the mid-19th century, it vanished for more than 100 years, was briefly sighted in 1941, and then dropped from sight again for almost six decades before it resurfaced in Espirito Santo on Brazil’s eastern coast in 1998.
The biggest hurdle we faced was simply not knowing where to look. The entire canon on the cone-billed tanager consists of two scientific journal articles, both by Jacques Berlioz, who had examined Vellard’s collection of Brazilian birds. In 1939, Berlioz published a brief, seven-paragraph note in the Bulletin of the British Ornithology Club announcing the discovery of a new species of tanager in central Brazil; after the war, he followed this up in a French ornithological journal with a somewhat longer description of the specimen and a list of the other birds Vellard had collected on the trip.
The papers indicate that Vellard shot the bird “at Juruena, northeast of Cuyaba, Matto Grosso (Central Brazil).” This is less straightforward than it may appear. “Juruena” could refer to either the 600-mile-long river by that name or to a city on its lower reaches, near the northwestern corner of Mato Grosso. But both are several hundred miles northwest of Cuiabá (as it is now spelled)—not northeast, as both papers state. No one knows if the error was made by Vellard (perhaps through poor record-keeping or unfamiliarity with the region) or by Berlioz, but attempts over the years to clarify the issue failed.
We did have one powerful clue—the other birds that Vellard collected on the same expedition. Taken together, they point to the intersection of Brazil’s three great ecosystems, which link hands in western Mato Grosso—the seasonally flooded Pantanal marshes to the south of Cuiabá, the Amazonian rainforests to the north, and the belt of cerrado grassland stretching across the midriff of the state, mostly at higher elevations. Putting our heads together with birders and ornithologists in Brazil and the United States, we decided to aim for the western fringe of Mato Grosso, beyond the Rio Paraguai and close to Bolivia, where several large chapadas, or plateaus, meet the upper Juruena, Guaporé, and other rivers. This region, we felt, offered the best chance for the right habitat—low, scrubby, dry forest at the edge of the cerrado, which we had finally found.
WE’D SETTLED into a routine—up at 4:30 in the morning, we’d load our old VW van in the dark, wash down some crackers or granola bars with packaged fruit juice, and arrive in the field as it was getting light, to take advantage of the dawn frenzy of bird activity. We spent days exploring the mountains near the Bolivian border and the floodplain of the Rio Guaporé; it was rugged, beautiful country, a green wall edged with orange cliffs like castle ramparts, above which flew flocks of scarlet macaws. Many times, our hearts would stutter when a black-and-white bird flew into view, but it was always something pedestrian. In truth, it was much too lush and wet a forest for the conebill, and so eventually we turned north, hunting for a way to the Rio Juruena.
We were following a clue, the sort of accidental lead that can change everything. We’d heard from an old man of a ghost town called Juruena, on the headwaters of the river of the same name—a place long gone from the maps, but which was a thriving community in the 1930s when Vellard passed through this region. This, we hoped, was the mysterious “Juruena” where Vellard’s notes indicated he shot the first cone-billed tanager.
The problem lay in getting there; the only access was by swinging far to the west and north, then backtracking east across one of the chapadas. After spending so much time in dense forest, it was liberating to rumble along beneath an open prairie sky. The cerrado extended to the horizon in all directions, dotted with head-high trees and bushes bearing colorful flowers, among which grazed groups of rheas, South American ostriches, one of them a male tending a convoy of new chicks the size of terriers.
The map said the river was about 50 miles across the empty chapada. At first, the going was smooth, but as the morning coolness gave way to a white sun, the road steadily deteriorated until it was no more than a dry, sandy streambed along which our van slowly lurched and fishtailed, almost bogging down time and again. Finally, as both dusk and the edge of the plateau approached, the old VW settled into a deep rut and wouldn’t budge.
For a long time, no one said much; three guys stood staring at a white van that tipped unpleasantly toward its left rear corner. Doug and I shouldered the back of the van while Fabiano gave it some gas, but the tires just splattered a rooster tail of sand against our legs, and the vehicle sank farther. So we did what anyone stuck in the middle of nowhere would do. We went for a hike.
It was an odd time to take a walk, I now realize—the sun had set, our only transportation was mired a very long way from help, and we hadn’t made camp for the night. We walked, I think, because we’d been on the go all day and we just couldn’t quit. We walked because we sensed the end of the road was close, and after weeks of scheming to get here, we wanted to see it. Finally, we walked because the crest of the hill was just ahead—and peeking over the crests of hills is what human beings have been doing for hundreds of thousands of years.
Over the last rise, in the dim twilight, the land dropped off in a series of mile-wide steps to the valley of the Rio Juruena, vague in the hazy distance. Not far away, like a line drawn across the chapada, lapped a low, scrubby forest that could not have looked more ideal as potential tanager habitat. We were stuck, but we were stuck in the right place.
ONE SODDEN MORNING, we had a teaser, the first time on the trip that our hopes really soared. We’d been making slow progress through the forest, made slower by the ever-fainter trail, which eventually dissolved entirely into featureless scrub, leaving us to backtrack to the edge of the cerrado and try a different route to the river. While we had seen tanagers of half a dozen other species, including one or two rarities, we saw nothing that even hinted at a conebill.
It was sometime after sunrise, and we were in brushy forest, listening to a very different suite of birds than we’d encountered thus far—channel-billed toucans, turkeylike Spix’s guans, mustached wrens. Many of the birds, like brown jacamars, were species that Vellard had collected on the same expedition as the cone-billed tanager, giving us hope that we might have stumbled at last on the right mix of habitat and locality.
I was recording the song of a wren when Doug spun around, pointing, and said with quiet but unmistakable excitement, “A black bird just flew across the road!” My adrenaline surged even more a moment later, when I spotted it perched in a solitary tree draped with vines—it was clearly a tanager, almost entirely black, with a spot of white on its wing. I remembered the drawings I’d seen of the conebill, with its fleck of white like a pocket handkerchief.
But then I noticed two things. This bird was black clear to its belly, without the white underparts that a cone-billed tanager would have, and the white on its wing was a thin sliver at the shoulder, not a square spot on the folded flight feathers.
“Where is it? Do you have it?” Doug asked, binoculars poised; then, before I could say anything, he saw the bird move and snapped his glasses to his eyes. “Oh,” he said, his voice flat. “That’s a white-lined tanager. It’s a good bird to find, but…” He didn’t need to finish the thought. We’d spent a couple of fruitless weeks in the field, and I was growing a little jaded about our chances. Still, I was content to putter around with my tape recorder and microphone, concentrating on the solid, substantial birds of the chapada instead of chasing after mirages. Doug and I parted company for a few hours, he to photograph some small species of spinetail he’d spotted, me to wander off along the dirt road, stopping here and there to tape new species.
At midmorning, the insect changing-of-the-guard was beginning; with the mosquitoes abating, I rolled up my shirtsleeves against the heat. The first few sweat bees were starting to appear, and I knew that within half an hour we’d be smothered beneath the small, stingless insects thirsty for the moisture on our skin. It was probably time to head back to the van, half a mile up the track, where Fabiano was taking a nap.
I turned to retrace my route, which is the only reason I saw the flock of birds flit across the road, a dozen of them in a ragged line, landing in a fruit tree about 20 feet high. Even that quick glance showed there were several sizes and colors—tropical birds often move in mixed-species flocks like this. Then one more bird flew into the outside edge of the tree—a tanager similar in size and shape to the others, but olive, with a more intense lime-green on the upper parts and upper breast, with a fair bit of streaking. The beak was typical of a tanager, somewhat conical and powerful, light at the base and dark toward the tip.
I will not be melodramatic and say that my heart stopped; the embarrassing truth is that for the first few seconds I couldn’t recall where I’d seen the bird before. But then I remembered the field-guide illustrations I’d studied of a female black-and-white tanager, the conebill’s closest relative—a bird that was a dead ringer for this one but not found within a thousand miles of us. No one knows what a female cone-billed tanager looks like, but given the similarity of the male to its black-and-white cousin (and the general propensity among female birds of the same genus to resemble one another), what I was seeing is exactly what one would expect a female conebill to look like.
Unfortunately, I had only a few seconds more to stare before the flock broke into flight and disappeared into the cerrado. I dug out my notebook, furiously scribbling down every detail that I could remember, then found myself unsure what to do next. My inclination was to charge off after the flock, but Doug and Fabiano had no idea where I’d gone; besides, if the tanager was the undescribed female of a long-lost species, I’d need all the witnesses I could get.
My indecision lasted only a minute or two, because Doug appeared at the crest of a low hill several hundred yards away, and my frantic gestures brought him at a trot. Fabiano, by another stroke of luck, followed moments later in the van, wondering what was keeping us. We grabbed gear, tapes, and fresh film and slathered ourselves with bug repellent, for the sweat bees were now bearing down in force. We sketched out a hurried plan—the featureless brush of the chapada is an easy place to get lost in, and we didn’t have a compass handy. Fabiano would stay with the van, and if we weren’t back at a prearranged time, he would start blowing the horn to guide us back to the road.
Heading into the brush, I felt the same keen edge as I have when tracking a buck during deer season—the heightened awareness of my surroundings, the way movement and sound are amplified, the reaching beyond oneself to connect with another animal. There were a few birds in the low trees, mostly sparrows and spinetails, flickers of movement cataloged and ignored. We moved quickly, trying to stay on a straight course, but the undergrowth was so thick—vines, sharp-spined agaves, stunted trees with soot-blackened bark from the last brushfire—that our progress slowed and our path meandered. I tried to keep the sun over my left shoulder, but the clouds, which had shown some breaks earlier in the morning, thickened almost as soon as we entered the cerrado, and I was forced to rely on instinct to stay on course.
All the while, we were battling ferocious clouds of sweat bees, far worse than anything we’d yet experienced. More seemed to rise from the brush with every step. My eyes were red and watering from bees that had crawled under the lids; they didn’t sting, but the little bastards bit like tiny ants, and so many of them clogged my nose that I was finding it hard to breathe without choking on them. They packed into my ears, crawled by the dozens under my hat and down to my scalp, and crept with abandon into my clothes so that my skin twitched with them.
There is a reason lost species are lost in the first place. Sometimes the reasons are weighty and formidable, like civil unrest, impenetrable mountains, or bandit warlords who use visitors for target practice. Sometimes they are more prosaic, like bad roads and worse information. And sometimes the reason is sweat bees—too many sweat bees. We finally staggered out of the brush, our eyes streaming and faces speckled with dead, smeared bugs just as Fabiano started honking the horn, never having found the tanager flock.
HAD I SEEN A FEMALE cone-billed tanager? Perhaps, but I later learned something that casts a fair bit of doubt on it. Like any birder, I’d pored over the field guides for months before going to Brazil, and had memorized most of the common species illustrated in the color plates. But I’d fallen into the trap of mostly looking at the pictures and paying scant attention to the text. When I later sat down with Bob Ridgely’s Birds of South America, I read something I’d overlooked until then, hidden in the account of the black-faced tanager, one of the species that made up the flock that day: “Sexes alike, but immature entirely different (and may not even be recognized as the same species): olive above, more yellowish on the head and with yellow eye-ring; paler yellowish-olive below.”
That’s a rough description of what I saw, but it’s not precisely the same; for instance, my tanager was brighter green than this description indicates, nor did I see a yellow eye ring—the sort of basic field mark birders usually notice quickly. But it seems far more likely that my greenish mystery bird was a young black-faced tanager traveling with its parents, rather than a never-before-seen cone-billed female.
But our failure to find the tanager did not diminish our belief that the bird was still out there somewhere—if anything, it reinforced it, by the simple experience of showing us how huge the bird’s potential world is. We had looked hard for the tanager, but when we plotted our route on the map, it was only a squiggle in an immensity of space. Even within the areas we had visited, there are many places where few but the Indians have gone, valleys and drainages that have changed not at all with the coming of the Europeans. Though the world is shrinking daily, much of it is still unknown; the blank spots are disappearing beneath the unblinking eyes of satellites and the probing fingers of chainsaws, bulldozers, and the farmer’s hoe, but great swaths of the planet remain a mystery to polite society, fit habitat for myths and monsters, a place where dreams can live. One of those dreams is six inches long and has a white spot on each wing.
Who cares if the cone-billed tanager is alive, or if it, like tens of billions of organisms since life first arose, is now extinct? What does one tanager, more or less, matter when weighed against more than 9,000 living species of birds? You’re asking the wrong person. Leaving aside the fundamental worth of every species, representing the unique harvest of nearly four billion years of natural selection—forgetting the value of a salvaged treasure in the face of global extinction—Vellard’s small black bird has come to matter a very great deal to me, out of all proportion to its size and ecological role, or even its place in a 60-year-old ornithological mystery. Obsession doesn’t need reason or rationale, it simply requires an object of desire, the less attainable the better. I lie awake some nights, thinking back to the hot breeze on the chapada, hawks wheeling overhead and tinamous hooting mournfully in the thickets. I am certain the tanager is still there, as I am certain I am the one who can find it. I’m going back to Mato Grosso to look for that bird again. Soon.