Westward Through the Gorge

A Series of Waterfalls and Rapids

William Clark and his advance detachment reached Weippe Prairie in present Idaho on September 20, 1805, starving from their Bitterroot Range crossing, and planning to hunt and send food back to Lewis and the main party, who were still on the mountainous Nez Perce Trail. The next day, while his men were out hunting, Clark sat down with an unnamed Nez Perce "Cheif" and solicited information about what lay ahead.

The Cheif drew me a kind of chart of the river... below a large forks [the Snake River joining the mainstem Columbia]... the river passed thro'gh the mountains at which place was a great fall of the water passing through the rocks.1

This was the first information they had gotten about the existence of Celilo Falls,

Clark's River: Still "fomeing"

Sherars Falls

It is unlikely that Lewis and Clark saw this very waterfall, for it would have been a 12-mile hike from the river's mouth. The one they might have seen is about 5 miles from the Columbia. But this is the first one that can be reached by highway today. The Deschutes was crossed by westering travelers just below this place beginning as early as 1826. Since 1863 it has been known as "Sherar's Falls." The salmon-fishing platforms are used by Indians from the nearby Warm Springs Reservation. The concrete wall in the foreground is part of a raceway built to facilitate seasonal salmon migrations upriver.

and it proved to be as misleading as the description they had been given of the Great Falls of the Missouri–an ostensibly singular "great fall of water," "a most tremendious Cataract." The captains carried Nicholas King's map of the lower Columbia (based on a map by explorer George Vancouver), but William Broughton of Vancouver's party, traveling up the Columbia, had failed to reach the Gorge by two miles. No other white men had ever been there when the expedition arrived. The Nez Perce informant may have condensed Celilo Falls2 and The Dalles and Cascades beyond them into one location, or Clark's translation chain may have failed. (Some Nez Perce traveled beyond Celilo and also could have described The Dalles.) So, when the Corps reached Celilo Falls on October 22, they unknowingly faced a series of four major river obstacles in fifty-some river miles: Celilo Falls, the Short Narrows and Long Narrows that comprise The Dalles and, farther on, the Cascades of the Columbia.

Meanwhile, on October 21 their attention was drawn first to a wide river entering the Columbia from the south, which they eventually learned was called Towarnehiooks by the Indians. Their word meant "enemies," which was appropriate because their mortal enemies, the Paiutes, lived along its banks. Lewis and Clark explored it a few miles up, soon encountering "a verry Considerable rapid" which Clark described as "an emence body of water Compressd in a narrow Chanel of about 200 yds in width, fomeing over rocks many of which presented their tops above the water." On their return upriver in April of 1806 the captains named it "Clark's River," but soon reconsidered that. On May 6 Lewis decided to reserve Clark's name for the long tributary of the Columbia that now bears three names–Bitterroot, Clark Fork, and Pend Oreille—and designated this one by its Indian name. But that one didn't last long either. French-Canadian engagés and traders who began arriving a decade or so later dubbed it Des Chutes—literally, "The Falls," now spelled Deschutes. The oft-repeated explanation was that the French name referred not to the numerous falls on the lower Towarnehiooks/Clark River, but to the Great Falls five miles below its mouth on the Columbia.

Clark's Map of the Great Falls of the Columbia
Pass cursor over map to read details.

From [Nicholas Biddle, editor,] History of an Expedition Under the Command of Captains Lewis and Clark, 2 vols. (Philadelphia: Bradford and Inskeep, 1814), Volume 2, facing p. 31. Original size, 4-1/8 by 7-1/8 inches.

Compare this engraving with William Clark's original map of the Falls. The text displayed above is from Biddle's edition, Volume 2, pp. 31-32. The summary statement quoted at the bottom of the map does not appear in the journals, but was perhaps dictated to Biddle by either Clark or George Shannon during their conversations with him in 1810.

The arrows on the map indicate the direction of the current. Those that point upriver indicate major eddies where the Indians would have caught the most salmon. No doubt there were many more smaller eddies as well.

The fall salmon run was ending when the Corps arrived at the Great Falls of the Columbia, several miles below the mouth of Towarnehiooks, with some native people still at the river, fishing with gigs and nets and processing their salmon harvest.

Wishram Women Preparing Salmon

Wishram woman cleaning salmon
Cleaning Salmon
Edward S. Curtis (1906)

Photogravure print from The North American Indian, (1907-1930), Vol. 8, Plate facing page 146.


pounding salmon

Pounding Salmon
Edward S. Curtis (1909)

Photogravure print from The North American Indian, (1907-1930), Vol. 8, Plate facing page 94. Curtis's narrative is from the same volume, pages 94-95.

Photographer Edward Curtis himself described the processes pictured here: "The staple of barter was . . . (pounded salmon). In the preparation of this article of food the salmon was beheaded and gutted, and with a sharp knife the two halves were separated from the backbone and the skin. The clear strips thus obtained were laid on a platform in the hot sun for a day. The next morning the flesh was soft, and the women squeezed it through their hands into shreds, placing the mass in large dishes or in a pit lined with grass or matting, and mixing with it the large quantities of roe taken from the fish. It was then thoroughly worked over with the hands and spread on a piece of matting. After a thorough drying, for two days, or more if the sun was not hot, it was pounded fine in a maple mortar with a stone pestle [or else, as Clark put it, "between two Stones"], and then rammed tightly into baskets of split cattails lined with fish-skins. Such a package weighed from one hundred to two hundred and fifty pounds, and was the product of about one hundred salmon. Fish prepared in this way kept for months. No salt was used, and to the white man's palate it was rather insipid.

"Another method of preparing salmon was to split the fish, roast it on one side slightly, then squeeze the meat out of the skin into a pit, mix it, dry it for about three days, then place it in a wooden bowl and mix it with fish-oil. It was eaten so, or mixed with berries, or with camas or other roots. Still another method, no longer in use, was to split the fish, thoroughly dry them in the sun, and tie them up in bundles, which were stored in pits lined with grass and matting. Fish so prepared could be kept for several months without spoiling. Fish-heads were dried over a slow fire and used in making soup."

Landing on the Columbia's north side, the expedition had to settle for dog-meat because "the Indians [were] not verry fond of Selling their good fish." A better trade was made when Lewis exchanged one dugout, a hatchet, and some smaller items, for a high-prowed Columbia River model. The captains instantly recognized how well designed the Indian canoes were for local river conditions. They were "neeter made than any I have ever Seen," Clark wrote, "and Calculated to ride the waves, and carry emence burthens." They were "dug thin and are suported by cross pieces of about 1 inch diamuter tied with Strong bark thro' holes in the sides."

The next day, October 23, amid clouds of fleas attracted by fish skins, men of the Corps crossed the canoes to the Oregon side. They stripped off their clothes so they were able to brush fleas from their bodies, and hauled the empty canoes over a 475-yard portage around Celilo Falls. After paddling down the stretch later known as the "Short Narrows," they eased the boats down the 8-foot "Great Rapid" with elk-hide ropes prepared especially for just such a purpose. The passage was completed by three o'clock in the afternoon, and camp established below the falls.

Ever since leaving Nez Perce country, the Corps had been accompanied by chiefs Twisted Hair and Tetoharsky, who had volunteered to introduce them to Columbia River tribes. On the 23rd, the chiefs had warned that they "over herd the Indians from below Say they would try to kill us." The captains ordered the men to check their weapons, and distributed enough ammunition to top off each man's supply at 100 rounds. The next morning, the chiefs said that, past Celilo Falls, they could no longer speak the languages and, further, were at war with local tribes, so they would be leaving. The captains convinced them to stay for two days, intending to help negotiate a peace treaty.

"bad whorl & Suck"

Bad Whorl

Edward S. Curtis (1912)
Photogravure print from The North American Indian, Vol. 9, Plate 302.
Original size, 36 x 44 cm (c. 14 x 17-1/8 in).

Clark's description of this spot, written October 25, 1805, scarcely betrays any anxiety about the prospects of getting through it safely.

Capt Lewis and my Self walked down to See the place the Indians pointed out as the worst place in passing through the gut, which we found difficuelt of passing without great danger, but as the portage was impractiable with our large Canoes, we Concluded to Make a portage of our most valuable articles and run the canoes thro accordingly on our return divided the party Some to take over the Canoes, and others to take our Stores across a portage of a mile to a place on the Chanel below this bad whorl & Suck, with Some others I had fixed on the Chanel with roapes to throw out to any who Should unfortunately meet with difficuelty in passing through; great number of Indians viewing us from the high rocks under which we had to pass, the 3 first Canoes passed thro very well, the 4th nearly filled with water, the last passed through by takeing in a little water, thus Safely below what I conceved to be the worst part of this Chanel, felt my Self extreamly gratified and pleased.

One imagines that Pierre Cruzatte, the Corps' professional river pilot, would have been in the bow of the lead canoe, reading the rapids like an instruction manual. Photographer Edward Curtis, who descended the Columbia in the spring of 1910 for the express purpose of seeing it from Lewis and Clark's perspective and went through this fearsome whirlpool at the bow of his own canoe, listed the basic qualifications for the role. "In swift water the man in the bow should have something in his head, a good deal in his back, be quick on his feet and quite indifferent as to the future."3 What is more, judging from Clark's final exultant expression, one suspects he really had had fun, and before an appreciative audience to boot.

The white line on the rocks at right was the annual high-water mark.4 Indeed, later white travelers who arrived here during the peak of the spring freshet often observed that the falls and rapids were completely submerged and consequently all but invisible. Hidden at all times were the vertical holes from one to twelve feet in diameter that pebble-and-sand-laden whirlpools had drilled deep into the bedrock in the narrow channel.5

October 24 dawned "fare after a beautifull night" with no hint of attack, and the Corps passed the Short Narrows. Here the Columbia was constricted from 200 yards wide, by Clark's estimate, to only 45 yards; the black basalt rocks stretched for a quarter of a mile. Compressed into that space, the Columbia was "an agitated gut Swelling, boiling & whirling in every direction." Even so, Cruzatte and Clark agreed that they could run the canoes through, first putting ashore "all the men who could not Swim and Such articles as was most valuable to us Such as papers Guns & ammunition." With Indians watching from the banks, two canoes were sent through successfully, and then the other two followed. Clark wrote, "we passed Safe to the astonishment of all the Inds."

"Every Civility"

Landing at what is now Columbia Hills State Park, Clark noted that here they found the "first wooden houses in which Indians have lived Since we left those in the vicinity of the Illinois." Sunken six feet into the earth, their cedar-bark roofs rose four feet above ground. Each lodge measured about 20 by 30 feet, and housed about eight men and twenty or so women and children. Sgt. Patrick Gass, the expedition's carpenter, pronounced that "This village has better lodges than any on the river above . . . tolerably comfortable houses."

"the first wooden houses...Since we left those in...the Illinois."

Ceremonial Lodge

Paul Kane (1810-1871)
Interior of a Ceremonial Lodge, Columbia River, 1846

Oil on paper, 9½" x 11½" (24.1 x 29.1 cm)
Stark Museum of Art, Orange, Texas
31.78/210, WOP 13

Clark understood the native people's name as "Echelute," which instead was likely the name of the speaker's village in reply to his inquiry. The Corps of Discovery made camp near the Wishram village. Ordway said, "we bought a number of fat dogs, crambries and white cakes of root bread"—made of either wapato or cous—for their meal. In the evening, according to Clark, "the principal Chief from the nation below visited, along with several of his tribesmen, and the captains "accomplished" a peace treaty between the Chinookans and the Nez Perce that put them "on the most friendly terms with each other." Following that the Corps smoked with men of the nearby village, Cruzatte fiddled and the men danced "which delighted the natives, who Shew every civility towards us."

On the 25th, men were stationed along the shore with ropes to throw out "in Case an accidence happened at the Whirl &c," and the canoes ran the Long Narrows one at a time. Only the third boat had trouble, and "nearly filled with water," but "we got her Safe to shore." After traveling two more miles before passing out of the rapid Long Narrows and back onto the wide Columbia, the Corps beached its canoes at Mill Creek on the Oregon side, and established the camp they called variously Fort Rock or Rock Fort: "we took possession of a high Point of rocks to defend our Selves in Case the threts of those Indians below Should be put in execution against us." But the day's single meeting with natives resulted only in smoking together, and the captains' awarding a small medal to one chief. From them or others that day, Twisted Hair and Tetoharsky "purchased a horse each with 2 robes" and headed home.

Staying at Fort Rock through October 26 and 27, some members of the Corps patched the canoes, collecting rosin from pines in the Cascade Range's nearby foothills. Cargo from the soaked canoe was put out to dry, and six men hunted for meat. The captains met with Tenino Indians (Sahaptian speakers), who they thought called themselves "E-nee-shur," and took vocabularies. Compared with the Wishrams (upper Chinookans), Clark noted, "not withstanding those people live only 6 miles apart, [they share] but fiew words of each others language." Here, too, "all the Bands flatten the heads of the female children, and maney of the male children also." Music and dancing filled the evenings, entertaining both the Corps and the Indians.

Despite comfortable dealings, one event at Fort Rock foreshadowed what the captains would consider a problem over the coming months: perceived thefts by the Indians, especially of tools. Clark wrote, on the 27th, that several Indians returned to their downstream home "in a bad humer being prevented [from] doeing as they wished with our articles which was then exposed to dry." Historian James P. Ronda explains that, to Columbia River natives, such taking was claiming "proper payment for services rendered," and also "to remind the white men of the need to offer respect and attention to the trading lords of the Columbia."6 That the Indians had a concept of private property was seen in the way each family's racks of drying salmon were left untouched. Further, when Clark shot a goose that plummeted into the river just above the Cascades' first falls, a native man swam into the river at great danger to himself, retrieved the bird, and delivered it to Clark, who in turn "suffered him to keep it."

Leaving Fort Rock on October 28, the expedition canoed downriver for three days until they reached what Clark called the "Great Shute,"7 in the Cascades of the Columbia River (see map, next page). They made camp on an island opposite modern Cascade Locks, Oregon, and Clark, Joseph Field and Cruzatte walked down to assess the Cascades on the 31st.

In the Indian villages along the way, they had been seeing many more European goods and clothing, and some guns—although most men still used bows and arrows. Clark also tried to inquire about the symbolic figures carved in low relief on wood, which fronted homes and adorned graves, and also appeared on canoes. Translation proved too much of an obstacle for him to get a full answer, but Clark went from thinking that these were idols to be worshipped, to believing that they were "treated more like ornaments than objects of aderation." They were in fact clan heraldry, representing spirit totems inherited from the ancestors who had received them in visions, and a family applied its own to any object, including bone fishhooks. For example, when Clark tried to ask one man about one of his carvings, the man reached behind it and removed his bow and arrows to exhibit.

Down the Cascades

Clark described the Cascades as having a "Great Shute" extending about one half mile, where the Columbia was 150 yards wide and dropped about twenty feet over both large and small rocks, "water passing with great velocity forming & boiling in a most horriable manner." Then the river continued among fully and partially exposed boulders for another two miles of extreme rapids. Over the entire two-and-a-half-mile length of the Cascades, he estimated that the Columbia fell 150 feet.

Even after watching local Indians portage the Cascades' entire length, the captains decided they could get the canoes through by ropes and a temporary log "road." On November 1, the men carried all their baggage over a 940-yard portage, a "bad way over rocks & on Slipery hill Sides." Then they laid poles across the exposed rocks, and let the canoes down the falls and dragged them over the rest of the rapids. Battered and needing repairs, the four canoes survived this passage.

On November 2, the Corps "Examined the rapid below us more pertcelarly" and decided to empty the canoes and have the non-swimmers walk around. They portaged their baggage for a mile and a half, then ran the Cascades' lower rapids in the canoes: "one Struck a rock & Split a little, and 3 others took in Some water[.]" They then safely passed two more sets of rapids in the next four miles before—downstream from today's Bonneville Dam—the Columbia became "a Smoth gentle Stream of about 2 miles wide, in which the tide has its effect as high as the Beaten [Beacon] rock."

Of as much interest to Clark as describing the Cascades was writing about the people who lived along them. The men had pierced noses; men and women had flattened foreheads; male and female clothing comprised skimpy garments of leather (they were "nearly necked"; their clothing was labeled "indecent" on March 29, 1806); blue and white beads decorated everything. To him, these people were "durty in the extreme both in their Coockery and in their houses," and he believed the adults' many missing and extremely worn-down teeth came largely from eating uncleaned roots just as they came from the soil.8 Had he had the time to observe the whole process of drying salmon to make fish pemmican, he would have seen that the violent Gorge winds continually blew sand into the hanging fillets, which also abraded the Indians' teeth.9

Clark failed to clarify the extent of the trade between Indians and whites lower down on the Columbia, but he concluded it was of minor importance, since the Indians' "knowledge of the white people appears to be verry imperfect, and the articles which they appear to trade mostly . . . Pounded fish, Beargrass, and roots...cannot be an object of comerce with furin merchants." Pounded fish sold especially well among the coastal tribes, since the climate there was not conducive to the fish-drying processes.

The Corps continued down the Columbia in peace, having experienced no major problems during their first visit to the area. Ahead were the miserable stormy weather of the Columbia estuary, and the winter at Fort Clatsop.

1. Unless otherwise indicated, journal quotations for this portion of the trip are from Clark; either Lewis was not keeping a journal from September 21 through December 17, 1805, or it has been lost.

2. Origins of this name (which the captains apparently never heard) have been lost, but suggestions about its meaning include "echo of falling water," "sound of water falling on rocks," and also "blowing sand clouds" after the region's strong river winds.

3. Mike Gidley, ed., Edward S. Curtis and the North American Indian Project in the Field (Lincoln: University of Nebraska Press, 2003), 89.

4. Emory and Ruth Strong, "Seeking Western Waters; The Lewis and Clark Trail from the Rockies to the Pacific (Portland: Oregon Historical Society Press, 1995), 98.

5. Emory Strong, Stone Age on the Columbia River (Portland: Binford & Mort, 1959), 44.

6. James P. Ronda, Lewis and Clark Among the Indians (Lincoln: University of Nebraska Press, 1984), 172.

7. "Shute," and "shoot" were phonetic spellings of the French word chute, denoting either a rapid, or else a fast-moving narrow channel of a river. Alan H. Hartley, Lewis & Clark Lexicon of Discovery (Pullman: Washington State University Press, 2004), 160-61.

8. Clark made a good inference. Teeth ground down from decades of eating stone-ground grains is one way that anthropologists and archaeologists age-date ancient human remains.

9. Emory Strong, Stone Age, 46.

Funded in part by a grant from the National Park Service, Challenge Cost Share Program.