An AI-assisted contemporary historical adaptation based on the expedition journals and checked sources. This is third-person narration, not a submission by either captain. The opening returns to a reconnaissance preceding the horse negotiations in the previous episode, then follows explicitly dated crossings and changes of route. It does not combine two crossings into one day.

McNeal placed one foot on either side of the stream.

First Taste of Columbia Water

On August 12, 1805, Lewis and three companions were following an Indigenous path toward the headwaters of the Missouri system. After so much labor against the current, the watercourse had become small enough for one man to straddle it. McNeal thanked God for the experience. Lewis recorded the gesture and continued uphill. They eventually reached a spring, rested, drank, and approached the dividing ridge.

The rest of the expedition was having a different morning. Clark and the main party were still hauling canoes. Shallow water and rapids kept the men struggling in the river for much of the day. Some preferred the prospect of traveling overland. The reconnaissance party seeking an outlet and the main group still tied to its boats occupied different terrain. Their experiences together explain the importance of the ridge.

Lewis climbed a gentle slope to the crest without finding an open way to his destination. Great mountain ranges stood farther west, with patches of snow on their summits. Descending roughly three quarters of a mile, he reached a clear, cold creek and drank from water he identified with the Columbia system. This was not the Columbia's main channel at his feet, nor an ocean a short paddle away. He had crossed the boundary between great drainage basins. The direction of the water did not settle the route of his expedition.

The crossing is now called Lemhi Pass, on the Montana–Idaho boundary. The name came later; Lewis could not have known it. Nor had he created the path. His party followed human tracks, passed old camps, and saw places where plants had been dug. They carried national instructions and journals through a landscape already traveled by people who knew its routes.

A Dangerous River Drawn on the Ground

That night they camped west of the divide and boiled their remaining pork. The following day brought contact with Shoshone people and the possibility of discussing horses and guides. Cameahwait drew rivers on the ground, showing their junctions and the difficult passage below. His account offered obstacles requiring investigation, not an easy waterway to the sea. He also described a northern route used by Indigenous travelers, difficult and short of food.

Lewis decided to investigate the river. After rejoining Clark on August 17, he assigned that task to his colleague: travel down, examine it, and, if both water and timber were suitable, build canoes. Otherwise, report the difficulty. Horse negotiations and river reconnaissance consequently proceeded together. The captains neither dismissed a whole route from a distant view nor assumed that a connected drainage basin guaranteed a navigable passage.

Clark took tools for building canoes. That practical preparation makes the eventual retreat more significant. His party was ready to begin work downstream, hoping to return loads carried on shoulders and horses to the water. A Shoshone guide led the way, while Sacagawea and Charbonneau assisted communication during the earlier stage. The guide's knowledge and Clark's observations together would determine which preparations proved useful.

The River in the Canyon

On August 22 they advanced along steep ground in the present Salmon River drainage. Loose rock covered the slopes, and the horses had to find footing among stones and ledges. Near a tributary's mouth they encountered families drying fish and berries. Their guide had fallen behind. Armed strangers arriving directly at the camp frightened its occupants; some hid in the brush.

Clark described offers of ornaments from children's necks, salmon, and other scarce possessions. The guide arrived and explained; Clark distributed small gifts. These measures helped without completely dispelling fear. Some women and children continued crying during his visit. For the expedition, people along the river offered information and food. For the families, the unexpected visitors first presented an uncertain danger. An exchange of gifts did not instantly erase that difference.

The terrain tightened further the next day. After about four miles, the horses had to enter the river to proceed. Farther down, mountains crowded the water and the horse trail ended. Clark left part of his party to hunt and fish, continuing with his guide and three men. Sometimes following a narrow animal track, sometimes clambering over rock, they explored roughly twelve miles farther to another creek mouth.

The obstruction was not one isolated rapid that could readily be bypassed. Much of the reach was difficult, including several major rapids where water forced between enormous rocks struck violently against the banks. Even empty boats could not safely be sent through without precautions. Persisting would require unloading and carrying goods in some places and lowering empty canoes by ropes in others. At the worst obstruction, a bypass might demand cutting into the slope and moving great stones.

Timber presented another problem. Clark watched for suitable trees and had specifically noted one pine the previous day that might yield a small canoe. Boats did not appear simply because axes had reached a riverbank. A suitable tree had to be located, felled, and worked; the finished canoe would still have to pass the rapids. More abundant pines farther away did not amount to a usable fleet at hand.

Nor did the bank offer an independent alternative. Clark estimated a width of about a hundred yards and found few places where the river could be forded. At many bends the mountains reached the water, requiring repeated changes of bank, while the current made crossing dangerous. Water and land were not entirely separate choices: the canyon obstructing one also squeezed out the other. The northern route mattered because it offered a way to leave this reach of the valley.

Such work was not impossible in the abstract. The question was whether this expedition possessed the time and provisions to undertake it. Clark considered the two together. Every unloading, portage, and reloading consumed effort; the men still had to eat while they worked. Their food supply was insecure. Salmon received earlier had already been consumed, and small catches and berries could not support a prolonged engineering operation. Delay depleted food, while hunger reduced the capacity to clear routes and carry boats.

The guide took Clark along an Indigenous trail around a bend and onto a ridge, pointing out the river farther below. Cliffs continued to restrict the channel. Clark did not personally walk every distant canyon. Combining his observations with his guide's information, he concluded that the expedition could not proceed by this water route. The limits of his direct knowledge remained, but the practical decision was clear enough.

The return was also difficult. On August 24 Clark fell from a rock and badly injured a leg. Men left behind had subsisted on small quantities of fish, birds, and berries. Clark wrote Lewis an account of the river, surrounding country, and prospects, sending Colter on horseback with the letter. An alternative was to travel north overland into another valley and seek a passage west from there. Routes drawn by his Shoshone guide supplied a practical next possibility.

The Same Spring

The letter arrived as Lewis was transferring the main party west of the mountains.

On August 26 he again passed the headwater spring with people, animals, and baggage, stopping for the men to drink and rest. Compared with August 12, the same place now carried a different weight. The earlier crossing had involved a few scouts. This one involved tools, arms, trade goods, and the equipment required to keep the whole company moving and alive. Animals needed pasture, men needed rest, and Shoshone carriers needed food. A drainage divide might look like a narrow line on a map. In a journey, it was a succession of loads to be moved.

After reaching camp that evening and reading Clark's letter, Lewis abandoned the hope of building boats and descending that river. He began acquiring horses for sustained overland travel and sought the elderly Shoshone guide who knew the northern route. Later expedition accounts commonly call this man Old Toby. That English designation does not give us his complete biography. What can be established here is the importance of his geographical knowledge to the expedition's choice of direction.

On August 30 the party departed the Lemhi Valley with twenty-nine horses. Purchasing them had not removed its difficulties. Some had sore backs; others were thin or inexperienced pack animals. Loads had to be distributed, ropes secured, and saddles adjusted. After traveling beside the river, the company left the direction of the canyon and followed a mountain creek toward a northern route. Work they had hoped to entrust to flowing water now fell to animals and people.

The Last Thermometer

On September 2 they left an existing trail leading toward the Missouri and followed a western branch through ground without a usable path. Brush had to be cut and footing was precarious. Horses slipped, rolled, and sustained injuries. Clark recorded one crippled horse and two that gave out. One load remained approximately two miles behind. After nightfall the rear of the company was still arriving.

The abandoned baggage required an actual recovery, not merely an entry in a tally. The following morning two men took Lewis's riding horse back for it, delaying departure until about eight. As mountains again narrowed around the creek, the party had to climb steep slopes. A slipping animal endangered both its load and the people following it. A geographically plausible route had to be made practical, one secure foothold at a time.

Snow fell that evening, followed by rain and sleet. Meanwhile the expedition's last thermometer had been broken. Clark left no conveniently precise temperature for the day, only an account of falling horses, terrible travel, and scarce food. Birds killed along the way and a little corn provided supper. The season was not simply a date at the top of a page. It altered the footing, available food, and distance the company could sustain.

White Skin Robes

On September 4 the coverings used for baggage were frozen. The men stopped to thaw them before repacking and departing at about eight. They climbed a ridge and descended toward a valley. A hunter killed a deer, providing a substantial meal. Continuing downward, they reached a Salish camp in the valley now known as Ross' Hole, near present-day Sula, Montana.

The residents welcomed the tired travelers, placed white skin robes over the captains' shoulders, smoked with them, and shared berries and other available food. The diarists' estimates of lodges and people do not agree exactly; no single approximation needs to be promoted into a census. What is clear is that the expedition had entered an inhabited place with horses, food preparations, and its own practices of hospitality.

Epilogue

On August 12, McNeal's stance over a little stream had reduced the great Missouri to the distance between two feet. By September 4 the travelers had learned more clearly that crossing a watershed and finding a practicable road were different achievements. Reconnaissance had closed one water route. Local knowledge and assistance had connected them to another route overland. The pass was behind them, but mountains still lay ahead. Reaching a camp where they were received gave them a chance to prepare again.

Sources, editions, and narrative boundaries

1. Actually consulted: Project Gutenberg ebook 8419, The Journals of Lewis and Clark, 1804–1806: Lewis, August 12, 14, 17, and 26, 1805; Clark, August 12, 22, 23, 24, 30, and 31 and September 1–4. https://www.gutenberg.org/cache/epub/8419/pg8419.txt . This edited electronic transcription preserves original spellings and omits some survey and astronomical materials. It is not a complete manuscript facsimile. 2. Cross-checked the UNL journal project's dated transcriptions and separately identified modern notes for August 12, 23, and 26 and September 4. The latter also provided Ordway, Gass, and Whitehouse accounts for comparison of attribution and estimates: https://lewisandclarkjournals.unl.edu/item/lc.jrn.1805-08-12 ; https://lewisandclarkjournals.unl.edu/item/lc.jrn.1805-08-23 ; https://lewisandclarkjournals.unl.edu/item/lc.jrn.1805-08-26 ; https://lewisandclarkjournals.unl.edu/item/lc.jrn.1805-09-04 . Gass's account is transmitted through an early edited publication, not treated as identical in form to the other manuscript sources. 3. The National Park Service's Lemhi Pass page confirms the scouting crossing on August 12, the main expedition crossing on August 26, and the modern county and state setting: https://www.nps.gov/places/lemhi-pass-mt-id.htm . Its Ross' Hole page identifies the September 4 Salish encounter near modern Sula: https://www.nps.gov/places/ross-hole.htm . Both pages identify November 27, 2020, as their update date and were actually read for this draft. Lewis's reference to Columbia water does not mean he had reached the Columbia's main channel. 4. The August 22–24 reconnaissance belongs to the Salmon River drainage in present-day Idaho. The early-September mountain crossing retains regional precision near Lost Trail Pass; disputed fine route details are not converted into a certain track. Lemhi Pass, the Salmon reconnaissance area, and Ross' Hole require separate links to their respective dates rather than a blanket association with the entire episode. 5. This episode and 011 concern overlapping parts of the journey but follow the established separate themes of horses and routes. Its return to August 12 is explicit. No dialogue or private thoughts are invented. Clark's downstream assessment distinguishes his observations from the guide's information. Indigenous lives and fears largely reach us through outside observers; derogatory journal labels are not adopted as objective descriptions. 6. Original journals are public-domain sources; both language versions are original adaptations. Modern research and annotation are not copied at length. Dates refer to narrated activities, and candidate places remain at the regional precision supported by the sources, without invented exact coordinates.