Across the Great Divide

After a few minutes’ rest he and his three companions—George Drouillard, Private John Shields and Private Hugh McNeal—strode up to the pass where, instead of the Great River of the West flowing invitingly at their feet, they saw “immence ranges of high mountains still to the West of us with their tops partially covered with snow.” Three-quarters of a mile down the western slope he paused at “a handsome bold running Creek of cold Clear water” where he “first tasted the water of the great Columbia river.” It may have momentarily soothed his anxiety over the discouraging prospects, but we hear none of his previously exultant tone.

The photo of Lemhi Pass (Fig. 1), marked by the convergence of both modern roads and traces of older travel ways, bears northwest (310° TN) toward the west (Idaho) side of the Continental Divide.^[2]The awareness of a more or less continuous ridge, or continental divide, separating the waters of the Gulf of Mexico from those flowing into the Pacific Ocean began to emerge in the 1820s. Part of … Continue reading The east (Montana) side is on the lower third of the picture. The area is now a National Historic Landmark under the jurisdiction of the Salmon-Challis and Beaverhead National Forests.

In June 1855 a group of Latter-day Saints settled at a site they called Fort Limhi, after a prominent figure in the Book of Mormon. Only three years later, hostilities with the natives forced the abandonment of Fort Limhi, but subsequent efforts were more successful, and the place-name, re-spelled Lemhi, was eventually applied to the valley and its river (a tributary of the Salmon River), as well as a mountain range, a county, and a national forest.

Having crested the divide in the afternoon of 12 August, and “first tasted the water of the great Columbia river” from the stream eventually named Horseshoe Bend Creek, Lewis, Drouillard, Shields and McNeal followed the Indian road “over steep hills and deep hollows to a spring on the side of a mountain” where they camped for the night.

Setting out early the next morning, they soon saw the narrow, green floor of the The Lemhi Valley and the pine-clad Lemhi Mountains, their tops trimmed with snow. “At the distance of five miles the road after leading us down a long decending valley . . . brought us to a large creek about 10 yds wide,” now called Pattee Creek.^[3]Pattee Creek was named for Fred Benjamin Pattee, who arrived in the valley as a youth in 1867 and began ranching at the mouth of the creek five years later. He died in 1955. Merrill D. Beal & … Continue reading They crossed it and ascended a hill (below the picture), then proceeded about four miles through a treeless, undulating plain of shallow gullies where they spied, at a distance of about a mile, “two women, a man and some dogs on an eminence,” who fled back down the dusty road toward the river.

On the 19th, two days after arriving at the Forks of the Jefferson, Clark followed the same road across the divide and down to Pattee Creek as he began his reconnaissance of the Salmon River. En route, he was the beneficiary of the Lemhi Shoshones‘ characteristic charity when a man loaned him “a mule & Spanish Saddle to ride.” Clark rewarded the fellow with a colorful waistcoat as a gratuity.

President Jefferson had specifically directed Lewis to record “observations of latitude & longitude, at all remarkeable points.” Now, the captains crossed Lemhi Pass a total of five times in August 1805, yet not once did either of them mention the need to take observations, much less stop to set up the navigation instruments when they had them at hand on the final crossing. Here are the crossings in chronological order:

12 August 1805, Westbound—Captain Lewis and 3 companions

15 August 1805, Eastbound—Lewis and 3 men plus about 60 Shoshones on horseback

19 August 1805, Westbound—Captain Clark, 11 men, Sacagawea, Charbonneau, five Indians

21 August 1805, Eastbound—Sacagawea, Charbonneau, 50 Indian men plus some women and children, with many horses

26 August 1805, Westbound—Lewis and the rest of the Corps, plus Cameahwait and other Indians. Baggage carried by 11 horses, 1 mule, and some Shoshone women. Sacagawea rode a horse. This was the only crossing of Lemhi Pass on which they had their instruments with them.

Twice more, in four different places, the captains were to cross the Continental Divide. Clark would cross at today’s Gibbons Pass (6941 ft) on 6 July 1806, en route back to Camp Fortunate, and Lewis would walk across Lewis and Clark Pass, 160 miles northeast of Lemhi Pass, on 7 July 1806. Both would cross the Lost Trail divide, 80 miles south of Travelers’ Rest, that separated two major sources of the Columbia, the Snake River basin and the Clark Fork-Kootenai River basin. They also surmounted Lolo Pass (See Bitterroot Crossings), 30 miles west of Travelers’ Rest, going to and coming from the Pacific.

They knew which of those “dividing ridges” separated the headwaters of the Missouri from those of the Columbia, and which did not. President Jefferson had made himself perfectly clear when it came to places as “remarkeable” as these: “The interesting points of the portage between the heads of the Missouri, & of the water offering the best communication with the Pacific ocean, should also be fixed by observations.” Why didn’t either of them ever take the time to follow the commander-in-chief’s instructions regarding these particulars? That is likely to remain forever one of the foremost among the unanswerable questions left to us by Captains Lewis and Clark.

Measuring Elevation

The most enlightened theory–conjecture, rather–about the Rocky Mountains at the beginning of the 19th century predicted that they would be found to reach only about 3250 feet above their base, which might lie 3000 feet above sea level.^[5]John L. Allen, “Geographical Knowledge and American Images of the Louisiana Territory,” Western Historical Quarterly, II (April 1971), 49. Thomas Jefferson, in Notes on the State of … Continue reading It was based on the estimates of the Canadian surveyor, Peter Fidler, who explored the southern Canadian Rockies in 1792. According to the popular principle of bilateral symmetry, which then seemed to be evident everywhere else in nature, it was believed they would average no higher than the Appalachians or the Blue Ridge in Virginia and the Carolinas, say 4,000 feet above their base. Lewis had no way of measuring precisely how high he stood above sea level, yet sufficient evidence was before his eyes–snow on the surrounding mountain peaks even in August^[6]It is unusual today for snow to remain the year around on any but the northern aspects of the highest peaks. In Lewis and Clark’s time, many parts of the northern hemisphere were still under … Continue reading –as he reasoned his way to a logical conclusion about what lay before him.

the mountains do not appear very high in any direction tho’ the tops of some of them are partially covered with snow. this convinces me that we have ascended to a great hight since we have entered the rocky Mountains, yet the ascent has been so gradual along the vallies that it was scarcely perceptable by land. I do not beleive that the world can furnish an example of a river runing to the extent which the Missouri and Jefferson’s rivers do through such a mountainous country and at the same time [be] so navigable as they are. if the Columbia furnishes us such another example, a communication across the continent by water will be practicable and safe. but this I can scarcely hope from a knoweldge of its having in it comparitively short course to the ocean the same number of feet to decend which the Missouri and Mississippi have from this point to the Gulph of Mexico.—10 August 1805

He had only three sets of figures on which to base his conclusion. First, he knew the longitudes of the mouth of the Missouri, the mouth of the Columbia, and the mouth of the Knife River at the Mandan and Hidatsa villages in North Dakota, which had previously been calculated. Second, he had his own records of the latitudes he had measured at the northernmost and southernmost points in their itinerary to date. He and Clark made celestial observations to secure data from which the longitudes of 111 different locations along the trail were to have been calculated after the expedition was over, but those solutions were never worked out (see Deciphering the Celestial Data). Third, he had at hand Clark’s daily estimates of distances traveled.^[7]On 11 August 1805, Clark passed the 3,000-mile mark from the mouth of the Missouri, according to his daily estimates of distances traveled, and named a large island upstream from Beaverhead Rock … Continue reading

Why didn’t the expedition carry an altimeter? The barometer had been invented in Europe about the middle of the 17th century, and the impetus for its further development emerged with the recognition of two practical applications, the study of the relationship between atmospheric pressure and changes in the weather, and the measurement of the heights of mountains.

About 1772 the Swiss physicist Jean Andre Deluc (1727–1817) invented a portable barometer expressly for the latter purpose. But despite his and a number of others’ refinements and gimmicks,^[8]Owen’s Dictionary, which the expedition carried, described and illustrated several earlier 18th-century inventions—a “horizontal or rectangular barometer,” a “diagonal … Continue reading in 1803 the most accurate and dependable barometer still basically consisted of a 34-inch glass tube, a bowl, and a vial of mercury, and was far too fragile to be practical on an expedition such as this one. It was another forty years before Andrew Pritchard of London invented the truly portable and considerably more durable aneroid barometer.^[9]W. E. Knowles Middleton, The History of the Barometer (Baltimore: Johns Hopkins Press, 1974), 85, 132.

Even if Lewis had packed a barometer, the measurements of mountain elevations would have been of dubious significance. The concept of “mean sea level” (MSL), a hypothetical point of interface between air and the surface of an imaginary “global ocean,” did not begin to emerge until the early 1840s.^[10]J. C. Ross, A Voyage of Discovery and Research in the Southern and Antarctic Regions, During the Years 1839–43 (London: John Murray, 1847), 22–32. Ross designated an arbitrary “Zero Point … Continue reading However, a worldwide standard for the measurement of elevations on land required a long-term, systematic record of the rise and fall of tides and studies of the effects of winds and currents. The oldest continuously operating tide gauge in the world was established at Aberdeen, Scotland, in 1862, and the accumulation of significant amounts of global data was complete by the early 20th century–just in time for the nascent Aeronautical Age. (Meanwhile, the building of railroads relied not upon measured altitudes but upon angles of incline–gradients.) But MSL is an arbitrary standard, and is difficult to apply globally. For instance, MSL on the Pacific Coast of North America ranges from 20 to 30 feet higher than on the Atlantic Coast.

From the arrival of the first fur traders in the 1820s and the first settlers in the fifties, people had to make their way through the Lemhi Valley on foot or horseback. After gold was discovered on Orofino Creek in north-central Idaho in 1860, prospectors began to scour the mountains for many miles around. When gold was found six years later in the Lemhi Mountains, a supply point quickly grew up in the valley ten miles east, opposite the mouth of the Lemhi River, where George L. Shoup platted a townsite and named it Salmon City.

Little by little, the ancient Indian roads were “improved”—leveled and smoothed—to accommodate wagonloads of freight drawn by horses, mules and oxen from as far away as Fort Benton, Montana, 300 miles to the north, or Corinne, Utah, more than 250 miles south of the forks of the Jefferson. In the late 1870s a narrow-gauge railroad was built from Corinne to Red Rock, a few miles south of the Forks of the Jefferson, which sped up the long-distance movement of freight and passengers to and from the mining camps on both sides of Lemhi Pass. Stagecoach service was established between Red Rock and the Lemhi Valley in 1884, with terminals at Tendoy on the western end of the Agency Creek road, Salmon at the mouth of the Lemhi, and Leadore fifty miles to the south near the river’s source, at 7000 feet. The stagecoach route connected the valleys across the Continental Divide for 25 years.

Lemhi Pass is 7378 feet above mean sea level; the settlements of Armstead (site of Fortunate Camp) and Red Rock, 30 miles east, lie at about 5500 feet, so the average grade over the eastern leg of the journey between the Beaverhead and Salmon City was only a little over two percent. However, the stage road from the pass down the west slope to the settlement of Tendoy, Idaho, dropped 2628 feet in ten miles, an average of 6.4 percent. Obviously, the climb out of the Lemhi Valley was slow and exhausting for the animals, and the descent could be terrifying if a stagecoach or freight wagon weren’t braked with a heavy “drag” of logs cut from timber on the ridge.

The Railroad Route

In 1910 the stagecoach service was supplanted by the Gilmore and Pittsburgh Railroad—the G. & P. for short, sometimes understood as “Get out and Push.” It reached Leadore by a more direct route from upper Horse Prairie Creek, and chugged up and down the Lemhi Valley as far north as Salmon. The railroad was built primarily to haul ore from the mines at Leadore to the Union Pacific junction at Armstead, east of the pass. By the late 1920s the Leadore mines were exhausted, and that, along with the advent of automobiles and trucks, dealt a death blow to the G. & P. By that time traffic was flowing principally north and south through Salmon down the Salmon River Valley to Challis and Pocatello on Highway 93, which was commissioned in 1926, and through Dillon on U.S. 91 (now Interstate 15) to Idaho Falls and beyond.

Salmon is still the seat of Lemhi County, the principal agricultural and mining service center for the Salmon River Valley, with local offices of the Bureau of Land Management and the USDA Forest Service. Leadore (pop. 90), “somewhat isolated” beside U.S. 28 down toward Idaho Falls, basks in its historic ghost-town status. Tendoy (pop. 187) is a farm-and-ranch community, and a crossroads quick-stop for motorists retracing the Indian roads Lewis and Clark followed across the divide.

The approaches to the gap in the “deviding mountain” have devolved into comparatively forgotten byways.