FIVE
COLUMBIA

SOUTH OF CRAB CREEK and Wanapum Dam, the Columbia River plunges through the deep and narrow canyon known as Sentinel Gap, then flattens out. As the major route in and out of the interior Northwest for everything from salmon, geese, and loons to Lewis and Clark and barges, the Great River of the West conducts the butterflies of passage to and from the basinland that bears its name. In postflood, predam centuries, human settlements sprinkled these productive shores. They still do, but they are very different from the Indian camps and from one another. A nice big shore habitat looked inviting from a distance, but pulling in, I found it a linear, litter-strewn homeless camp built mostly of torn and flapping plastic tarps. Men sat on car seats in the dust, smoking. I caught a young woman’s eye, and it seemed to carry a rainbow of regrets.

Then came Desert Aire, a planned community, mostly trailers and prefabs tucked tidily above a sandbar willow shore. People worked on their lawns in the hot wind. A woman looked visibly relieved to see me go. On a third flat, trucks and trailers (no cars) congregated on uneven cobbles and deep sand. Daughters and wives comprised a temporary women’s village where meals were made, games played, and interloping males and fish duly admired. The fish were accepted, the men sent out to fish some more or to trailer the boat for the haul back to the Tri-Cities, Yakima, or Goldendale. In these latter-day encampments of aluminum, plastic, and lawns, I saw wrenching reflections of other villages belonging to people and fish now absent except in shrunken, diminished versions of, as the treaties put it, their “usual and accustomed places.”

The broad span of Vernita Bridge took me over to Riverside, where I turned back upstream. A scratch of a lava track passed below dramatic basalt formations pocked with thousands of cliff swallows. The lava flow crumbled up into caves, hoodoos, pillars, and columns in a palette of browns and a confusion of shadows, spattered with bird lime. Stands of an autumn-blooming buckwheat looked like snowdrifts, as its name, Eriogonum niveum, implies it should. Down by the water, mulberries and Chinese elms overhung the beach. I noticed the green hearts of Viola odorata leaves among the milkweed and other native plants. These sweet-smelling garden violets and the exotic trees told of a homestead that once flowered here. Originally the homesteaders would have confronted annual floods in spring. Once the big dams were built for hydropower, like the Priest Rapids Dam that lay just upstream, the regular, renewing floods were controlled. Now the violets, milkweed, and all the creatures that feed on them fluctuate with the engineered outflow and the thin silt on the river cobbles.

In a milkweed leaf rolled with silk, I watched a gray-and-black hunting spider with a yellow blaze on her abdomen, squat and alert. I’d noticed many species of spiders living in and around milkweed, a mark of abundant traffic in insect life. A tan, russet, and beige silver-marked moth with a fancy thoracic furpiece hid under another leaf, head down near the base, and a black, yellow-saddled, tussocky tiger moth larva browsed nearby. Monarchs share the floppy leaves with these and other insects adapted to feed on milkweed in spite of its toxins. Even humans consume milkweed, boiling the young pods to remove their bitter taste. I doubted that this was done now in any of the settlements I’d seen across the river, but both Indians and white pioneers did it over much of the continent.

Three big monarch larvae seemed to leap out at me, then a fourth. I wandered along the bank, and an adult monarch sailed into view from upstream. It perched in an elm, sunned, flew down to a grove of little mulberries on the beach, turned and floated back upslope toward me. I caught her on the wing, another very fresh, perfect female, tagged her, and put her back on the elm where she had alighted briefly. She crawled up a bare limb, basked, pumped, then sailed up toward the road and the slope. Just then, vast swarming clouds of caddis flies arose, thousands in each congealing nimbus. Both caddises and the sun were in my eyes as the monarch, magicianlike, disappeared behind a smokescreen of insects and sagebrush.

Sweeping a netful of the little caddis flies and nabbing one for a look, I saw that the translucent gray wings were about a centimeter long, the fine antennae a little less. The aquatic larvae of Trichoptera construct protective cases of pebbles, sticks, or other detritus in which to live on streambeds and pond bottoms. I wondered what the larval cases of this species were made of, and whether the entire bed of the Columbia was covered with them before the hatch. In their mating swarms they shift together like sandpipers in flight, responding to one another’s minute movements. Backlit in the sun they looked like swirling snow, as if all of the buckwheat petals had taken off together and come alive. They boiled over the black slopes, recently burned even blacker than the basalt, in uncounted millions or maybe billions. It might be that I had never seen an animal this abundant, even the impossible mobs of monarchs in Mexico.

A huge darner dragonfly rested in a nearby sage, her abdominal spots the same pale turquoise as the pungent sage buds. A great blue heron stood on a point silhouetted against the river. But the monarch had vanished. She was clearly working her way southward along the shoreline when I first saw her, and I guess she continued. I turned around to follow that way. Starlings burst out of the basalt caves as I started the engine. A chukar squawked as the still-warm sun melted behind cold lava, reminding me of a local confection called Chukar Cherries. The track ran south through cherry orchards. You could tell without seeing the trees that it was cherry country from the abundant coyote scat full of cherry pits. I was hungry, hot, and thirsty, and wished I had a bag of chocolate-covered Chukar Cherries and a big icy glass of cherry cider.

Yet I also felt renewed, recovered from the hot wind, the sad riverside camps, the partings. A monarch a day keeps the blues away. The Columbia rolled on, a broad, sluggish thing here, waddling between rimrock and gravel flats, orchard and dam and powerline. Below Vernita it stirred with the suggestion of a real current as it lolled toward Hanford Reach: the only undammed stretch, the real river, and my next stop. Virga, those gray wisps of evaporating rain cloud, stroked the southeastern sky like the eyelashes of the gods.

No highway runs along Hanford Reach, one of its glories. A bleak drive over the plateau and back down a grade brought me to Ringold in the dark, where Powdermilk just fit on a knoll beneath a sheltering alder. Swarms of midges and caddis flies completely covered the windows. At four in the morning, the sounds of guys in trucks and boats pulling in woke me. I turned over in my comfortably reclined car seat, shifting weight to the other cheek and tugging my sleeping bag around my ears.

Butterflies being quite civilized, there is no need to rise at break of dawn to see them, as one must for birds. Cloud and cold wind met my next awakening, but on the third I saw cabbage butterflies and their shadows, and I stirred. I was perched beside a rushing canal, a wasteway for Potholes water from the plateau into the Columbia, mainlining agricultural chemical runoff into the Hanford Nuclear Reservation’s radionucleide brew. Hanford has dominated the Big Bend of the Columbia River for more than half a century and has left parts of it as polluted with radioactive wastes as anywhere east or west of Chernobyl. Reactors and their fell smokes and red steam hulked on the far shore, but I was not interested in thinking about it much that morning.

What I was interested in were Hanford’s inadvertent treasures. Rattlesnake Mountain rose across the river to the west, showing its long, shady blue side. Lupines had painted much of the mountain indigo when I first saw it the previous spring, the summit abuzz with hilltopping swallowtails of four species, about the time north-flying monarchs were returning. I was one of a group of ecologists visiting the Arid Lands Ecology unit (ALE), a research reserve that has become a great unlooked-for beneficial effect of the government’s occupation of the region for the atom-bomb-building Manhattan Project. ALE, the largest piece of eastern Washington shortgrass steppe left unplowed and ungrazed for the past fifty years plus, is a site of enormous ecological significance.

The other treasure is Hanford Reach. Had it not been for the reactors and their appropriated hinterlands, the Ben Franklin Dam might have gone in long ago. Now that the age of giant porky water projects has largely passed, and the removal of certain dams has been officially mooted in the debate over salmon survival, it seems unlikely that a dam joining Benton and Franklin counties will ever be built. Now, except for the tidal stretch where I live, this is the one place where the Columbia runs wild in Washington. I’d guessed that it would offer prime monarch hunting grounds.

The dirt road north from the Ringold hatchery follows the river below Wahluke Slope through public wildlife lands. Gravel spurs lead to the sites of old homesteads above the shore. I took the second one down to a shady grove of honey locusts, and, climbing out of the car, I beheld a great big beautiful monarch flying about a locust, then another down among the knapweed. For the first time I was tracking two monarchs at once! The first sailed to another locust, and I missed a high shot. The other rose out of the knapweed, flew south, then back. I lost it but caught a third. Tagged, she flew off fast at one hundred feet plus, southeasterly, toward Wahluke Slope, as a swallow made two abortive passes at her.

Surely some monarchs must be sacrificed to the swallows’ education. After demonstrating that blue jays really do avoid monarchs once they’ve been subjected to their bad taste, Lincoln Brower went on to show how certain grosbeaks and orioles have taken to preying on the massed monarchs in Mexico. The grosbeaks simply scarf them down, buffering the cardiac glycosides with other foods. But the orioles actually pluck out the untainted, nutritious fat deposits, dissecting the prey with their sharp bills. I have watched them in the forests above El Rosario in Michoacán, four species of orioles in large flocks the colors of the monarchs themselves, stripping butterflies out of the firs like kids in a berry patch. When that much food is concentrated in one place, some organism will usually find a way to make use of it. Orioles figuring out how to feed on monarchs are not unlike the monarchs themselves evolving a way to use the vast milkweed resource of North America. In both cases, animals have come up with means of getting around protective toxins in order to exploit the goods within. With monarchs more dispersed in the North (especially the Northwest), birds have less incentive to find a way to eat them, so they generally do as they are taught by parents or by direct experience. But it does take a few monarchs to do the teaching.

The sun came out and meadowlarks sang as I set up my stove for a bite and coffee. Then another Danaus appeared on the upstream side of the grove, and I lost it down the bank into the blue willows. These butterflies had evidently spent the night in the beige-leafed old farm locusts, which stretched to fifty or sixty feet in height. Still another appeared by the windbreak, and I caught her on the wing and tagged her. Released onto a yellow coreopsis, she flew below the bank, south into the willows before I lost her downstream on the slough behind Savage Island.

The riverside vegetation was composed chiefly of sandbar willows (Salix exigua), which I call blue willows for their long skinny leaves, glaucous and redolent of sandy, damp places. A small fresh male viceroy leapt up from the same thicket. Blue willows being one of the favored food plants of Limenitis archippus, that’s where he was likely to find a female. On the ground, an incredibly brilliant blue-green hunting wasp prowled for a cricket, which she would sting and stuff into a burrow for her single larva. Her shimmer recalled that of the metallic blue Vespa I drove among the blue willows of my youth, hunting viceroys and monarchs and great gray coppers.

What is it with metallicism in Nature — like the gold spots on monarch pupae or the wholly gold chrysalides of the black-and-blue milkweed butterflies, called crows, that we had seen in the butterfly house in Kelowna? Chrysalis means a gilded box, more or less. But is all such brilliance solely for the benefit of educable birds? Many beetles are metallic shades of blue, green, gold, and silver, including flower, flat-headed, blister, and tortoise beetles. This hunting wasp has numerous iridescent and metallic kin. Certain species even effect the candy-apple red I once tried (very poorly) to paint my Vespa. It can’t be for purposes of sexual selection, since both males and females shine.

At least for fritillary butterflies, with their quicksilver spots, I have heard a plausible theory — that they resemble dewdrops sparkling in the grass, thus conferring cryptic protection upon butterflies warming up to fly in morning meadows. Perhaps this could account too for otherwise camouflaged brown and green pupae bearing dramatic medallions of seemingly molten gold and silver. But I have heard no overall theory of intense metallicism in insects. It may be that warning, of either bad taste or sting, drives such flagrant coloration; or it may be a byproduct of other processes, with no special function. Sometimes we are too quick to seek a purpose for every feature in nature, when many traits, if they confer no liability for survival, may be fixed simply as side effects of genes with entirely unrelated tasks to perform.

By noon it was clear, in the seventies, the wind blowing only off and on. The river here was as fine a corridor as I had hoped it might be, full of nectar — mauve aster, yellow coreopsis, purple loosestrife, goldenrod — and milkweed, too. A dragonfly basked on a stone: a solid pale blue libellulid with black vein marks at the tips of all four wings. Its eyes were a deep-spotted turquoise. Just as birders are discovering butterflies, several butterfly watchers I know have taken a keen fancy to odonates — dragonflies and damselflies. I am falling for their charms myself after years of dumb admiration. There were hundreds of meadowhawks, darners, skimmers, and bluets here. The colors of odonates are maddeningly fugitive; to appreciate their hues, you must see them alive; unlike butterflies, whose colors remain intact on the pin.

I was examining a yellow composite I didn’t know in the damp swale above the shore when an untagged monarch popped up and paused to nectar on goldenrod. When I stroked long and bricked it, it took off up and east, toward Wahluke Slope. Then, admiring golden sulphurs on lavender asters, I saw a big male monarch nectaring on the same kind of aster, closer to the ground than usual. This time I took better care and netted him cleanly. After affixing a tag to his left forewing, I compared him to the little viceroy I had safely tucked, alive, into a paper envelope earlier for the purpose of photographing it next to a monarch. This monarch was immense, fully four inches in wingspan, the viceroy less than half his size. The difference is not always so great. An orange butterfly crossed my bow sailing NW, low. I called it a big female viceroy, then almost convinced myself it was a modest monarch; but she flew to the swamp, perched among the peachleaf willows, and I saw I was right in the first place — a big, intensely colored L. archippus. When I released my captives, the small one darted right into the willows after her, while the big one flew strong and high toward the southeast.

This was the first male I had tagged since the Similkameen. Unlike many butterflies, male monarchs tend to be larger and paler than females. But the real giveaway is the set of androconial (or sex) patches, black velvety pads of specialized scales also called alar (wing) pockets, located along a vein on the male’s hind wings. Specialized cells within the pockets produce male pheromones. In most danaiines (milkweed butterflies), the males possess brushes called “hair pencils” that they can extrude from the end of their abdomen. They wipe these brushes in the alar pockets, then waft the air with molecules of pheromones that help induce the females to mate. Both the hair pencils and their use are much abbreviated in monarchs, but they retain the black patches. Female danaiines produce their own pheromones in their terminal abdominal segments. We know them by the absence of alar pockets and by the broader black scaling along their veins. Since I was finding females almost exclusively, I began to wonder whether male monarchs migrate earlier, on average, than females. Certainly in many species, notably among the fritillaries, males emerge days or even weeks prior to females. And since the southward migrants are undeveloped sexually, there is no reason for fall males and females to stick together.

Eating a honey and raisin sandwich for lunch, I thought of bees. There was a great deal of nectar here, but not many honeybees, whether because of the mites that have devastated them lately or simply a lack of local hives. There was an abundance of native pollinators, such as solitary and carpenter bees and an array of wasps and flies. As Stephen Buchmann and Gary Nabhan have shown in The Forgotten Pollinators, when natives decline, plants come to rely on the pollination services of European honeybees, and when they drop out, the plants are left high, dry, and unpollinated. Agricultural sprays, crop monocultures, paving of habitats, and other causes of pollinator loss do not intrude on Hanford Reach, so its native pollinators are largely intact, including the monarchs. You wouldn’t know it here, but the availability of nectar is clearly a limiting factor for migrating monarchs in many areas. They serve as effective pollinators all along their routes.

At parking area no. 7, another old farm site, I walked along a natural shoreline of cobbles and mud. In a lush little aster-and-daisy swale, I thought I saw a checkered white, but in Marsha’s gentle folds it turned out to be a great white skipper, the first I’d seen on the trip. A male, its silky wings shone with the same mother-of-pearl as the opened valves of the green-backed river clams that lined the water’s edge. Actual whites, a pair of copulating cabbages, flew by for comparison. The smaller male carried the closed female, and they were harried by other, mateless, males.

The foreshore came down in an asparagus-mulberry savannah. It looked promising, and as I strode into it, a moose of a monarch glided past. I followed at a trot, but it pumped south in a hurry, so I watched it with my binoculars for several hundred yards farther along the bench before it blended into the background. Savannah sparrows owned the benchland, where deer bedded down and red flickers clicked. I avoided the rising wind by walking below the rim of the old river terrace. Now and then, when it seemed there were no more monarchs to be seen, one popped up out of the asters over the bank or materialized beside me on the wind. Depending on my alertness and the gusts, not to mention the animals’ keen reactive grace, some I netted and some I didn’t.

The final side track lay opposite the one-time village of Hanford, which gave its name, land, and identity to the nearby nuclear establishment at the time of the Manhattan Project. From there I could see the famous White Bluffs stretching away to the north like a regiment of Sphinxes’ paws. As the sun settled toward Rattlesnake Mountain, they deepened from chalky white to old gold. On the way back south, I pulled again into parking area no. 5 because its two big elms looked like a monarch spot to me. I walked upriver to a high bank and sat on a mat of cheatgrass straw, scanning the sunflowers below. A green, flowery sward opened out toward the river through the backwater outlet that, during high water, makes Savage Island an island. Life had settled down out there, getting ready for evening. But when I got back to the car, I looked up and saw two monarchs still in action.

Circling the tall elms, they acted like a premature male pursuing an uninterested female for two or three minutes. Then they both began poking about the foliage for suitable roost sites. The female perched low, and I was able to net her. The swoosh put up the male, which took a higher perch where I could not make him out. Instead of rejoining him after tagging, the female flew high to the east-southeast. I watched her until she had nearly reached the rim of Wahluke Slope, then lost sight. I was sorry she had forsaken the communal roost, for that is what it surely was, even if they were the only two there. I hoped she found another suitable bivouac uphill, or downriver.

Weary, hungry, and well satisfied with Hanford Reach, I headed in search of a roost of my own. The monarchs I’d found might seem like a miserly few in Minnesota, Maine, or Mississippi, but they were a respectable total in the intermountain West. I’d found the first group roosts I’d ever seen in the Northwest. The reach was a good place, for all it had been through, a rich and volatile place where signs beside great klaxons warn “Listen for siren sounding for about three minutes followed by an audible message: LEAVE THE RIVER IMMEDIATELY.” The only siren I heard was the high-pitched, tiny one of the rising caddis swarm. I quit the river reluctantly, deliberately.

Two cities often arise where rivers meet, so at the merger of three rivers, three cities make sense. Kennewick, Richland, and Pasco (the Tri-Cities, Washington’s fifth largest conurbation) straddle the deltas of the Yakima and the Snake where they debouch into the Columbia. Monarchs flowing out of the milkweed-rich breeding grounds in the Yakima Valley would come through here, as would those of the lower Snake and the ones I’d been following from the north.

At the Yakima Delta Wildlife Park, a maze of trails wove through Russian olive thickets to the Yakima River proper, where the milkweed was well chewed. I have found monarch larvae up the Yakima Valley as far as Naches, at the eastern base of the Cascades passes that climb past Mount Rainier. Autumn emergents there travel a valley of vineyards and hop fields that smells now like Concord grape juice, now like a brewery. The river route delivers butterfly arrivals into a gauntlet of sprawl and a hydra of freeways that must be negotiated to reach the Columbia. It is good they can fly high when they feel like it. The delta park might offer an oasis before the city passage.

Each path led to a fishing hole marked with blue chicory and fishing litter. The “park” was completely undeveloped, but it was full of wonderfully woodsy glades and glens for a kid to swing a net in without the restrictions of a formal preserve. It seemed like a fine place for fishing and fooling around. Whites and skippers danced across open fields surrounding a lone cottonwood, whose huge bole was spray-painted with crosses, peace signs, “Anarchy,” and so on, each vandal in his own twisted way recognizing a holy site and marking it as such. The place went on and on, thick and thrashed, with little nectar. All the common, weedy butterflies of late summer showed up, but not the King.

At noon I crossed a causeway, lined with black people fishing, to Bateman Island, a thumb stuck right into the confluence of rivers. Ten thousand cabbages, a big mourning cloak, a viceroy; goldenrod, milkweed, and limitless roost trees. Could monarchs be far away? Great old Russian olives, red-trunked like cypress, tangled like vine maples, made loops and roofs of interlaced branches, caves and grottoes that would have thrilled me as a kid. They gave me new respect for a plant I knew in childhood as a backyard hedge. My brothers and I shagged the hard berries at each other but had no mature spinneys like this for forts. Russian olive, the Eurasian Eleagnus angustifolia, was introduced to the West for windbreaks. Escaped and naturalized, the silvery-leaved trees have had several decades to get this big. The banyanlike interiors of the groves struck me as perfect for hobo camps in this railroad town. Indeed, signs of a big empty homeless camp sprawled nearby.

A man was poking about for Indian artifacts, despite a sign warning of up to five years in prison and a $100,000 fine for taking or disturbing cultural materials: “surveillance is Conducted.” Scott showed me two rough-worked scrapers, then put them back. Thin, in a blue shirt, blue jeans, and a blue billed cap that said “Alaska,” he had parents nearby but was living out of his truck just now. “I’m what I call a naturalist-preservationist,” he said, and told me about a “whole culture” he’d discovered near Lake Owyhee in southeastern Oregon, complete with animistic carved agates, temple floors, and so on. “I have an ethics problem with what to do with it,” he said. “Maybe I should just leave it be.” He’d taken an agate effigy home for safekeeping: “It was just layin’ there on the ground. Somebody would’ve just took it and sold it.”

Noticing my thong necklace with a catlinite turtle and beads, Scott said, “I used to have a necklace like that, but I can’t wear it anymore. The spirit world is so close to me now all the time.” He was distressed to think of the original village of the island now displaced by the immense middens of plastic, fast-food packaging, and decayed bedding of the camp. Across the water, fancy homes faced the homeless slum: cold comfort either way, I thought, when the folks are home and regarding one another’s situation. Five white pelicans sat on delta mud with gulls and a logful of cormorants. I walked off the tip of the island and stood in the cool mingled waters of the Yakima and Columbia rivers, ignoring the radioisotopes and insecticides seeping into my knapweed-scratched feet as I toed a four-inch river clam shell.

A particularly pretty viceroy cruised the sandbar willow edge and settled with its cinnamon wings spread flat on reed canary grass, then flew out to investigate a bright hoop of mating darners. Viceroys and other admirals assume posts, such as sunny branch tips, from which they conduct sorties at anything that might be a potential mate or a competing male. Scott saw the viceroy too. “You know, they’re pretty similar to monarchs,” he said, “at least the birds’re s’posed to think so.” My eyebrows must have gone up a notch.

“Do you ever see any monarchs around here?” I asked.

“Oh, yeah — all over. I saw some monarch caterpillars a while back over at Two Rivers Park, on that patch of milkweed near the boat ramp. And I saw a woolly worm yesterday,” he said. “Time for trout fishing!”

“Do you use woolly bears for bait?”

“Nah, I move them off the road. But when I see ’em out, I use a woolly worm fly. By the way, I saw something in the paper about some lady doing a study on a little yellow butterfly that’s coming north, like killer bees. Doing her Ph.D., I think it said.”

Really?” I asked, both surprised and skeptical.

“That’s right,” he said, “in the Tri-Cities Herald, two or three weeks ago.” But at the thought of someone doing a Ph.D. thesis, he grew pensive. “If only I could get the pros interested, I could bring that Owyhee culture to light. But they’d probably trash it. I’m going to write a book on it, if I can just get started.” He invited me to visit his folks’ house to see the agate head, which he was keeping there until he could decide on a proper repository. I was tempted. But the woolly worms were out. It was time to be looking for monarchs.

On the way back to my car, I watched a red admiral alight on a willow branch, suck sap, bat yellowjackets, and fly off with quick flicks of wing. As various brush-footed butterflies often do, it had flown directly into the right part of the tree, searched for a few seconds to find the exact spot, and proceeded to drink at the sap flow. Sap is an important source of sugars for hibernating butterflies awakening in flowerless February or March and for migrants like the red admiral, but I have never seen a monarch visiting sap.

That sight made me thirsty and I stopped for a V-8 at a mom&pop, opposite an interpretive sign for the Lewis and Clark Expedition. Bateman Island was as far as they got up the Columbia River, on October 17, 1805, on their way home. They found Wanapum Indians drying salmon: “multitudes of this fish,” recorded Clark, “almost inconceivable.” The water was so clear they could see them at fifteen to twenty feet of depth. Clark shot a sage grouse, forty-two inches in wingspread. Absent from the journals is any mention of Russian olive, mulberry, locust, knapweed, purple loosestrife, freeways, nuclear reactors, dams, etc., etc. Almost inconceivable is right. There might have been monarchs, but the journals are silent on the subject.

My next stop was in familiar territory. In the late sixties, my former wife JoAnne and I took a winter field trip to Columbia Park in Richland with a collector friend, Dan Carney. It was a cold day, with the sycamore balls blowing on their stems, and, in the willows, the hibernacula of viceroys quivering in the river’s bitter breath. Dan showed us how to find the little sleeping bags, bits of leaves rolled about the center, the midvein poking out and giving them away to the sharp eye. The minute larvae lash their leaves to the branch with silk before retreating into their leaf rolls to hibernate. In spring, with the new growth of willow, they creep out to continue feeding.

Now Columbia Park is all things to all people, with all manner of recreation and far fewer willows. Fortunately, a natural area remains, managed by the Lower Columbia Basin Audubon Society. I found the nature trail and took the path designated as the Monarch Loop. Here, I thought, if anywhere. Sandhill and woodland skippers visited the goldenrod among robust, six-foot-tall milkweed. The underside of one big floppy leaf was covered with brilliant yellow-orange aphids, like bee’s pollen baskets. Deep blue berries of Virginia creeper hung on red stems among ghostly Russian olives. Eastern fox squirrels and California quail skittered and whistled in the underbrush.

A blue-green graffito on an observation blind announced simply “Butterfly.” From its tower I could see sachem skippers nectaring on thistle, making this a three-hesperiine walk — three species of grass-feeding tawny skippers. The thumbnail-sized golden triangles darted about the glade — skipping. In male tawny skippers, the androconial patch is called a brand or stigma. I can see how the impressive stigmata of this skipper inspired the name sachem, which means chief. Its rectangular black sex brands, bigger than a male monarch’s alar patches, are truly chieflike regalia.

As I was thinking this, as if to illustrate the point, a fresh male monarch descended from the sky onto the goldenrod patch, surrounded by tall wild rose and nettle. I climbed down from the tower, inched past nettles into the thicket, and caught him. As I was tagging him, he reached out with his front tarsi, grabbed my mustache, and hung on hard. So as not to hurt the hooked legs, I allowed him to launch from his perch at will. For a few moments my face was the habitat of a monarch at large. It tickled like crazy, and I tried not to twitch. Finally, with a lift-off flutter so light I wasn’t quite sure I felt it, he sailed off easterly into still air, rose above the trail and the trees, and headed downriver.

I prowled south on the Columbia River dike, solid with blooming rabbitbrush, and arrived at Two Rivers Park, opposite the mouth of the Snake River. This was where a lepidopterist friend, Patti Ensor, had once known a roost tree, and where Scott said he’d seen larvae earlier in the season. I quickly found milkweed beside the boat ramp, where a man was trying to start his outboard motor. The plants were caterpillar-free. Nor were there any milkweed longhorn beetles, but some beautiful yellow-on-black, chevron-striped longhorns occupied the goldenrod. I surveyed suitable locusts for roosting monarchs, but Patti’s tree did not reveal itself. The only orange butterfly was a slick little west coast lady, bagging some beams as the sun came out for the end of the afternoon before sinking ruddy in a cloud bank. Backlit caddis flies swarmed and swirled over blue willow going gold. There were five days of summer left.

An hour later, the same guy was still trying to start his boat motor. He was skinny and young, with an obese, red-haired wife and two cute little daughters, one redhead, one brunette. The mom was eating chocolate chip cookies while the girls played beside the “Caution: Barge Waves — Do Not Leave Children Unattended” signs. They kept saying “Dodo! Dodo!” The man’s wife walked off the dock, tripped on the ramp, and almost went into the drink, dropping a cookie in the action. The little redhead, already pudgy, went for it. As his females drifted off to their truck, the sad dad struggled on into the dusk with the recalcitrant engine.

Curious about the article that Scott had mentioned about “a little yellow butterfly coming north, like killer bees,” I sought out the Kennewick library when I left the park. The only little yellow migrant butterfly recorded in Washington was the dainty sulphur, also known as the dwarf yellow or Nathalis iole irene. I discovered it in the southeast corner of the state in 1975, and it hasn’t been found in the region since; it couldn’t be that. No, the article turned out to concern the very butterfly I’d been watching earlier when the monarch had come along, the sachem skipper (Atalopedes campestris). It is neither very yellow nor anything like the Africanized bees, but it is at least moving northward.

The sachem is another species that I had the privilege and thrill of finding for the first time in Washington State. It moved up the Willamette Valley of Oregon in the seventies, and my friend Maurita Smyth had spotted one in her Portland butterfly garden in 1986. So each fall I scanned open spaces and marigold gardens along the Columbia in search of the first Washington pioneers. When I finally found them, it was in a mall.

Thea had business in an office building on the edge of Vancouver Mall, across the Columbia from Portland, on August 28, 1990. As I waited for her, I was watching the many woodland skippers in the garden around the building’s flagpole when I thought I saw a female sachem — bigger, longer-winged, more richly colored and strongly marked than the others. But I had no net with which to confirm the state record, Marsha having stayed home. I borrowed a length of stiff insulated wire from a handy electrician and quickly fashioned a mini-Marsha from a little cotton pecan bag. On my hands and knees in the garden, all eyes at the windows above upon me, woodland skippers chasing away the sachems every time they landed, I finally succeeded in netting both a female and a male to establish A. campestris as an official member of the Washington fauna.

Since then, sachems have spread along the Columbia to the Tri-Cities, where they are now common. A University of Washington graduate student, Lisa Reed, had been investigating their range expansion in hopes of catching clues about how organisms adjust rapidly to colder climates. Animals extend their ranges northward as the climate ameliorates or as they adapt to colder winters. It remains unclear just how far north, and in what stage, sachems are able to overwinter, but they are expanding from what was once a Sunbelt residence with only summer excursions northward. So Scott was partly right. He was more accurate than the reporter, who called the sachem skipper “rather drab . . . mundane in appearance: brown, with a dab of white.” In fact, it is a delicate and elegant blend of golds, olives, wheats, and mauves, along with those startling black brands, like epaulettes of rich velvet.

In any case, it was nice to know that someone else in the area cared about such things. In a town whose high school athletic teams are called the Bombers, where the microbrewery makes Atomic Ales, and where nuclear waste disposal is the hot topic among the morning latté crowd in Jennifer’s Bakery, you don’t have to dig deep to find most folks’ agenda, and it isn’t butterflies. Still, I heard some wise words about monarchs before I left the Tri-Cities. The maids at the Vagabond Motel wanted to know what I was up to. Living out of a Honda Civic, I found that I needed now and then to empty everything and resort. Viewing my gear spread about the room, they couldn’t imagine my business, and they weren’t too shy to ask. When I explained that I was following monarch butterflies, Serena asked, “How much patience do you have? When I see a butterfly, I say, Bye-bye!”

“Yes,” I laughed, “for me, it’s mostly the same.”

Pasco is a city of bridges. By early afternoon I had crossed all three meant for automobiles: a basic one, a handsome blue one, and a soaring suspension bridge with a single pair of central masts and a magnificent webwork of white cables. Cruising the east-side dike on the blue-collar side of the river, I found a series of old, dried-up “sanitary lagoons” at the south end of the railyards and truck docks. Sewage farms have long been revered as prime birding sites. I remember watching a stoat jump on the back of a moorhen in a sewage farm outside Cambridge, England. Tossed into the fetid muck by the tough, cootlike fowl, that was one disgusted stoat. Now I wondered if such places, dried out and gone fallow, can be good for butterflies, too, which frequently visit scat and manure for the nutrients they provide.

The pits were full of fresh milkweed and blooming Canada thistle and goldenrod. Right off, I spied a bright monarch over goldenrod on the raised border of the field. I waited too long to swing at it and was obliged to track it down into the pit. Once tagged and released on a thistle head, she waited a moment, then flew off strong and high, direct to the southwest. After a couple of football fields, she glided down toward the Snake River. Rounding the former lagoon I saw another monarch, standing out like an orange billboard among the lavender thistles. We repeated the same dance: the netting, the tagging, the release onto thistle; a two-minute pause, then the flight up and away for two hundred yards and the descent, but this time southeast.

Number 83 got a pass from a dragonfly on the wing; no. 84 got the once-over from a barn swallow, twice. The first time the swallow looked and veered. When it came back closer, the monarch avoided it by agilely dropping, very much as I’ve observed moths behave under bat attack. Then, falling toward the river, she crossed the glide path of a white pelican. Back in the pit, whites, sulphurs, and painted ladies lingered over the sweet-smelling thistles on this sour ground where shit had become first flowers, then butterflies.

At three I arrived at Sacajawea State Park and the confluence of two of the great western rivers, the Snake and the Columbia. I grew up, and first learned how closely butterflies are wedded to watercourses, along an irrigation ditch on the high plains of Colorado. In the hyperbolic language of its Victorian builders, a simple forking of the High Line Canal was called the Great Bifurcation. Here was a truly great bifurcation, though since it is where the rivers merge, it could equally well be called the Great Convocation of Northwest Waters.

I had pictured myself following the Snake out of the state in the southeast. But the river is hard to reach over most of its run to Idaho, and its upstream course actually curves north before heading east and finally south, where it emerges from Hell’s Canyon. It was hard to imagine a migrant getting to this point from, say, Hanford Reach, then turning back north just so it could eventually travel south again. But I could imagine them flying southeast from here between the Blue and Wallowa Mountains to hit the Snake later, or continuing on down the Columbia and eventually abandoning it to fly south. Not that they all necessarily go the same way. Butterflies are wedded to watercourses, not welded to them. They employ rivers when rivers work for them, abandon them when they do not. The challenge now was to decide which river worked best: the Snake, the Columbia, or some overland average between the two. But I didn’t want to figure it out; I wanted to be shown.

In the muggy afternoon, I made a river crossing on foot over the railroad bridge that no. 83 had flown toward from the sewage lagoon. Snake River Bridge no. 2 was a rusty old drawbridge that looked to be up for good. The tracks appeared unused. I strode out to the middle, where the immense counterweight reposed overhead. Scoping a similar bridge downstream on the Columbia I could see that the bridge tender had to climb sixty feet of caged ladder and sit in a hot tin box on top, above the raised section. Such a lot of iron to shift about at the whim of river commerce. It was spooky being out here — the river so big, so far below, the massive stone supports bereft of ladders. No migrants plied the river — just the motor barge Prospector from Vancouver, Washington, an immense, rumbling green slab, slightly raked. On shore, oil tanks and grain docks; rabbitbrush and asters beside the oxidizing tracks; bulrush banks and red osier dogwood, in unseasonable bloom. Cottontails and redtails, trying to get together or not, depending on their points of view. Someone shot at doves. Nothing stirred but me on my high perch.

Once back in the park I stepped onto a big navigational aid on the point of the peninsula between the rivers. Day-Glo pink and chartreuse, with a big letter S in green and a green light on top, the beacon stood on a concrete, beam-bound platform littered with goose poop full of little striped rugby balls — Russian olive seeds — and otter spraints full of fishy, clammy, crawdaddy bits. Geese cackled, magpies purled. From up on the otter’s platform, I could see any monarchs that might be coming down either river.

Instead I saw the river tug Sundial motoring down the Columbia pushing a grain barge, its wake plashing onto the riprap by a bit of milkweed as a train whistled between the two railroad bridges. Five barges stacked with wood chips, grain, and containers grumbled down the Snake, shoved by one big Tidewater tug out of Portland, and made a tortured turn into the Columbia. And then, even as the tug lumbered beneath the downstream bridge, ye gods! the train rumbled over MY bridge. The horn blew, the counterweights swung, the center span of the bridge rose once more. It still worked after all. If I had still been out there, it would have been very awkward.

And then something truly amazing happens. As I walk around the point in this park at the mouth of the Snake River on this clear, sparkling, warm afternoon, an inviting stonework summerhouse lures me with its shade. Leaning against its low wall, my palms out on the cool basalt blocks, I ponder the courses of nos. 83 and 84 after they dropped toward the river. I say to myself, “I need to see a monarch come beating its way up or down the river through here.”

As the thought settles, a monarch comes beating its way down the broad Snake, some twenty yards offshore. It wears no tag. I watch it fly right out of the Snake River mouth and over the Columbia. Then, flapping very close to the water, just inches from the barge wakes, it crosses the half-mile of open river before my eyes. It has nearly made landfall on the far shore at Two Rivers Park before I lose it in the complex pattern of wave-lap reflections. That amazing monarch is flying strong, pointing me down the Columbia — to the sun! I have my answer.

A fingernail moon came up, and my hands went pink in the lowering sun. Two white pelicans passed the moon, crossed the river, hit the sun, ignited, merged, separated, fizzled in the purple haze so only their cinder-black wingtips remained. Sixteen flaps and then a glide. I followed them for a mile as the sun slopped into an ocher pool. Some big fish splashed, probably not salmon, though the sky and the Columbia were the color of salmon flesh with blue streaks, as if taking on the substance of the fish that were no longer there. Here, after all, is where the nearly extinct Snake River salmon meet their bifurcation, too.

It seemed to me that Sacajawea Park should be a “power point” for monarchs, like Point Pelee on Lake Erie, Lighthouse Point in New Haven and in Santa Cruz, and Cape May in New Jersey. And so it was. Once the Shoshone woman got Bill and Meriwether to this point, the river did most of the rest, because they were bound for the Pacific. It was just the opposite for me. The river route had been clear so far, but the river must soon be left behind. Unlike the salmon, river-running monarchs travel cross-country as well. We would both need to find another route south.

I spent the night under sharp stars. A brisk, cool wind brought the morning, sunshine, and a monarch flying or blowing out of a locust next to my campsite on the Walla Walla River. The only others in Madame Dorian Park, a rough Army Corps of Engineers facility, were a pair of old hippies settled in for the full fourteen-day limit. Next to their trailer they’d laid down an Astroturf dooryard and set up a wrought-iron stand and hanger for bird feeders. I was just wondering whether monarchs use the invasive alien star thistle, when a pale one came off a clump of it and whirled around to a Russian olive — and I saw I had been fooled again. It was a big female viceroy, three inches plus. She basked, slowly fanning in the wind-blown sun, an orange advertisement of coming autumn, brightly backlit, spread in full view of birds that paid her no notice. Then a real monarch popped off the rabbitbrush between the highway and the old road. After I missed my wing shot on the stiff breeze, the monarch let the wind sweep it easterly over the yellow rabbitbrush plateau. Watching it go, I could easily imagine, as Robert Frost did, that “those great careless wings” were made “for the pleasure of the wind.”

I followed, but it was all birds. Unhunted dowitchers safely grazed, and fifteen killdeer cut past a backwater slough, making their sweet racket. Over the water, barn swallows chattered their impatience to leave. A female northern harrier, honey-brown and white-rumped, hunted the flats. I drove east along the Walla Walla for a few miles. Along the edges of the river road bloomed Russian thistle — pretty up close, like a million little pink hibiscuses; but Salsola sticks to your skin and later grows into tumbleweeds, which stick to fences and culverts and the wind itself. A fellow drove up and scanned the river, probably a working man on a break from the nearby Georgia Pacific pulp mill, wishing he were fishing. We nodded but did not ask each other’s business. The riparian border was a weedy, stickery jungle bounded by cottonwoods and willows. Forging through a floodplain summer-baked into six-foot blocks of clay with three-inch cracks, I entered a dense tangle of goldenrod bound together by ten-foot nettle stalks. It was so thick I had real trouble getting out again, and nothing flashed orange for my trouble.

I filled my water bottle in a mini-mart at the junction of the Columbia highway and the road to Walla Walla. Between green neon beer signs depicting a jumping fish and a cactus, a root beer readerboard announced: “Sturgeon Candy Engelhart 65″ 7-7-96.” I think it meant that Candy had caught a big fish, and that it, not she, was five foot five. As much as I love fresh sturgeon, I would no more eat the bottom-feeding beast so near downstream from both Hanford and the paper mill than I would willingly make a meal of monarchs.

As it turns into the gorge, the Columbia River rounds its Great Bend and wends west through Wallula Gap, a major choke point for the floods from Glacial Lake Missoula. At peak postglacial flow, Wallula discharged close to fifty times the amount of water released by the Mississippi in high flood. The Two Sisters stand in the breach, dramatic pillars of Frenchman Springs basalt graffitized with “96” and peace signs. Having been two of Coyote’s three wives, the sisters were turned into rock by the jealous trickster. I put up an osprey with a big fish in its talons. It spiraled off upstream for miles.

For most of its westerly run, the Columbia serves as the state line between Washington and Oregon. East of the Great Bend the border is just a straight line on a map. As I rounded the gap, I crossed that line into Oregon. Masses of rabbitbrush and sunflowers, suitable fodder for my quarry but sharply wind-tossed, stained the roadsides pee-yellow where the river became the frontier. But the next monarch I saw was miles downstream, just above McNary Dam.

In the late overcast afternoon I set out on the McNary Beach Trail to stretch my road-rubber legs. I’d gone no more than fifty yards when a monarch flew up over the bank and dove directly into a Russian olive to roost, too high to net. But instead of settling right in below her until dark and getting a solid fix on her final clinging position, I continued my walk east along the huge, flaccid pool of the Columbia on the former railroad bed, hoping for others. A few minutes later I saw a cabbage white go to roost some twenty feet up in another Russian olive. The air was still now, but both butterflies had chosen the eastern sides of the trees against the morning westerlies. Returning to the monarch’s roost tree at 5:30, I couldn’t find the butterfly. It must have crawled farther in. But afraid that it had left, disturbed by a jogger, I tossed small stones. One actually struck the right limb, and the monarch fluttered around from the southeast side of the tree to the northeast. I couldn’t spot it again and decided I’d just have to try to see it leave in the morning.

I bedded down in the car in the parking lot a few yards from where the monarch was sleeping. Downstream, McNary Dam and locks hummed. A big paddle wheeler joined a trim tour boat in the lock, both of them lit up like a carnival, for passage down to the next pool. The green starboard lights of a big ship slipped past in the dark. It was good to be there, with a fine, hoppy Grant’s Scottish Ale to drink, Steinbeck to read, the river, and a monarch roosting nearby. When I turned out the light, I tried to compress my being into the butterfly’s poppyseed brain and imagine what it was like to be clinging to a leaf in the dark, with a hundred miles to fly against tomorrow’s wind.

First thing in the morning, louder than the trains and barges in the night, louder than the turbines of the dam itself, even louder than a jetski, was the industrial-strength riding leaf blower driven around and around the parking lot by an industrial-sized worker. I got up and went to stand knee-deep in the Columbia River, the water about the same temp as the air, a little below 60. My tree-watching vigil lasted all morning. The chief entertainment was a pine siskin trying over and over to get a drink. Each time it hopped to the riverside, a wavelet struck, causing it to fly a few feet back. Working along the gravel shore, the yellow-barred, brown-streaked finchlet did its little dance right past my feet. I had to laugh, though the poor siskin seemed doomed to go thirsty.

I also watched a furry tiger moth caterpillar walk right into the Columbia River. This got me thinking of the adaptive value of random dispersal. Rather than a sharp unidirectional pulse, the migrations of monarchs and many other animals actually describe a shotgun pattern, with much of the population getting to the most desirable destinations while some others go wildly astray (like the monarchs that arrive annually in Britain). For the northward migration this makes particular sense, since the milkweed resource is widespread. For the southward, this pattern might assure that the migrants spread across the full spectrum of acceptable winter quarters. Many other species, not necessarily migratory, spread out in all directions, seemingly to optimize their survival chances overall. So it is with woolly bears. By either strategy, some will necessarily drown in the process, or be crushed in the road, sacrificed to the experiment. I was going to save the damp larva, but a sneaker wave got it. A little after noon, in good full sun with slight haze, I decided my monarch had probably slipped out the back door, and I did the same.

From the mouth of the Umatilla River west, monarchs were moving along the Columbia shores. In a wildlife area east of Irrigon, Oregon, a swale between the roadside olives and riverside willows was chromium with rabbitbrush, fleecy with milkweed puffs on the wind. Checking some chewed-up milkweed, I spotted a mint-fresh monarch hanging from it. In fact, the small male had only recently eclosed. Nearby, the chrysalis shell hung colorless by its cremaster, hooked into a button of silk on the midvein of an unchewed leaf. I was not sure I should tag him, as his wings were still soft, but I managed it with great care. Placed on a Russian olive bough just above his birthplace, he walked around a bit, then lined up cleverly on a twig so that the westerly breeze would not bend his petal-soft copper wings, which felt like silk to my fingers.

I prowled the rabbitbrush upswale, where monarchs were attending to nectar. As the day waned, I revisited the soft, silky one, and his wings seemed firm and fine. Another monarch flew in from the east, swung around an olive two times, and alighted on the northwest side about twelve feet up. I considered camping here between the two monarchs, tagging the new one when it grew cold, and watching them both in the morning. I bent to the task of making a stick marker beneath the untagged one, looked up, and found it gone. Big skeins of ducks sutured the sky, on the move. I declined to remain after all.

So far I’d been obliged to follow the south (Oregon) shore, since no road continued down the river on the Washington side from where my Sacajawea had crossed the river before my eyes. Now I crossed over the Umatilla Bridge back into Washington, continued downstream, and camped to the lights and whistles of freight trains at Crow Butte, almost an island in the reservoir behind John Day Dam. Morning came on a cold wind that pinned a pale monarch to the marsh in the crook of the island causeway. Far downstream and a few years before, a fisherman friend named Dan Pentilla had introduced me to a term for a concentration of whatever one happened to be hunting. When we found a spot in the river where the silver salmon schooled, Dan called it a fish patch. Now when I implored the butterfly gods for a mess of monarchs, I decided “fish patch” was as good a term as any. Today, the last day of summer, the gods were good to me.

Mount Hood loomed off to the southwest, and beyond it lay the edge of the world as far as monarchs were concerned, since milkweed promptly drops out farther west. The highway traced a low shoreline between dust-dry hills, punctuated by little bars bearing milkweed, locust, and licorice. The first trees for many miles appeared in a thick grove at a place called North Roosevelt. The trees, tall elms and locusts, defined a horseshoe embracing a bunchgrass mini-prairie. Within the tree ring, twenty-four basalt boulders stood on concrete plinths, ranging in size from a microwave to a small fridge. Each boulder bore Indian petroglyphs — inscribed relief patterns of lizard, bighorn, and other motifs. Most had colors leftover from pastel rubbings people had done; others had been painted over. At the base of one, letters crudely inscribed in concrete read “East Klickitat Gem & Mineral Club, 1964.” I rightly surmised that the club had transferred the rock art from cliffs along the Columbia to save them from flooding by the John Day Dam.

Between the park and the highway ran a low swale of burdock and milkweed. Checking it out, I saw one, then two, monarchs fiddling around their host plant. A formidable fence stood between us, and when I got to them they sailed over a hedgerow of Russian olives. This play went on for quite a while, as I played their fool repeatedly. Then I spied a very tatty female nectaring on a four-foot goldenrod. I netted her, but she got out when the bag flipped over, and laboriously flew off across the entry road to the eastern hollow, a hundred yards away. A few minutes later, to my astonishment, I saw that she had returned to the same inflorescence of goldenrod. I carefully caught her this time, handled her gently, tagged her, replaced her, and she resumed nectaring without hesitation. She knew exactly where she wanted to be, and no one would keep her from it.

This venerable, ragged mother had only a third to a half of her wings left. Her body and wings were greasy. Also, some of her veins were squiggly, showing developmental difficulties. I softly palped her abdomen but could feel no spermatophore — normally, an older female would have one or several of these seed packets deposited by the male, unless she was of the migratory generation that would not mate until spring. She might have been old enough to have absorbed any spermatophores she once possessed. It was remotely possible that she was a survivor from the spring’s northbound immigration, or even, conceivably, from the previous autumn’s emigration. If so, she would be a full year old, though this is unlikely; I have heard of only one monarch living a year, and very few of the fall departers are thought to make it back north the next spring. She was still tough, and it is possible that she belonged to a recent generation, was no more than five or six weeks old, and had simply lived hard. She wasn’t telling. But no matter her age, she taught me the valuable fact that a monarch can learn, leave, and regain the location of a particular nectar plant. I had not known that.

Soon I saw a big male and caught him. Quite possibly a son of the sedentary old gal, he had only recently eclosed nearby. After being adorned, he perched on an olive, spread, quivered, flicked his tagwing two or three times, and lifted off to the west. A gust drove him up and back, and he dropped down into the brush. When I looked for him, I fell into a treacherous maze of deep badger diggings and tried to break an ankle. And there, my face near the ground, I found Washington’s other species of milkweed: Asclepias fascicularis, narrow-leaved milkweed. Its thin, willowy leaves make it much harder to spot than showy milkweed, which was here too, along with half a dozen monarch larvae, feeding side by side on both species, just a few feet apart.

There were others, and finally one big monarch floated over the swale, evading me, eventually alighting in a tree in the center of the rock horseshoe, then rising again and drifting away. It was the last I would see in this veritable swarm, the largest one-time concentration so far on the expedition, quite justifiably known ever after as the fish patch.

After Roosevelt, everything changed. Hackberry trees appeared, then Oregon oaks. Showy milkweed dropped out in favor of narrowleaved (also called Mexican whorled) milkweed, which I learned to recognize by its slender, bursting pods. It wasn’t very far west to where the Cascade crest runs down to the river at the Skamania County line, the maritime green and Douglas-fir reassert themselves, and milkweed disappears. Very soon the low hills over in Oregon rose up to form the high-cliffed, waterfall-laced canyon for which the Columbia Gorge National Scenic Area was designated. I knew that monarchs must soon strike south: they have to cross the Columbia. When I finally witnessed such a crossing, it was in one of the strangest places in the Pacific Northwest.

The railroad maintenance road under the cliffs beyond John Day Dam was rocky and slow but passable all the way to the apricot and peach orchards and whitewashed church steeple of Maryhill. Sam Hill, a remarkable public figure, intended Maryhill (named for his daughter) to be a model Quaker farming community. Above, perched on cliffs like a Rhineland fortress, stands the stately home he built for his wife, daughter of the (unrelated) railroad magnate James Hill. When she wouldn’t come to the windy, lonely mansion, Hill invited the exiled Queen Sophia of Romania; Alma Spreckles, the sugar heiress; and Loie Fuller, the popular dancer and film star, to visit Maryhill House. They helped to assemble the important collections, including Rodin sketches and sculptures, that make Maryhill a superb art museum today. Hill began the building of the revolutionary, reinforced concrete mansion in 1914, but it was not completed until 1940, well after his death.

Maryhill also has splendid gardens, shielded from the ever-present gorge gales by high windbreaks of trees. There I found monarchs dropping from the sky to take the nectar of purple and yellow butterfly bushes, pink cosmos, lavender asters, and blue salvias. It was easy to catch and tag them as long as the sun lasted. Each time I finished one, a new one appeared. If the Roosevelt swale was a fish patch, this was like shooting fish in a barrel compared to the chases I was used to. One had a broken costa, the stiff leading-edge vein that supports the wing like the strut of a kite, and I was able to splint it successfully with the tag so that it could lift off again.

Perhaps they had come on high from the Yakima country to the north or had worked their way along the Columbia. Some had arisen on the premises no doubt. Those that left sailed south. The mansion blocked my view, so before the day grew too old, I dropped down to the rabbitbrush bench just above the Columbia shore. There I found still more monarchs, harder to catch, visiting the wild plants. As I released a female, she fulfilled my urgent hope and headed out over the open river. I followed her with my binoculars as she danced with the whitecaps, brighter and more adept than any of the colorful windsurfers that throng the river. The channel narrowed here to just half a mile, and she aimed southeast, directly at the grain elevator of Biggs, Oregon.

Above and a little east of Maryhill stands Sam Hill’s memorial to peace, a concrete replica of Stonehenge, now an eerie echo of the Roosevelt rocks. I had camped there one summer, on the very lip of the basalt flow, and had awakened to the boundless river and gorge far below. Now, sunset was near, and I considered bedding down there again. But I made a different decision. I crossed the bridge to Biggs, on the track of that last monarch.