October 17, 1805 found the Corps of Discovery encamped at the junction of the Snake and Columbia rivers, near present-day Wallula Gap. The three surviving journal entries from this date — by William Clark, Patrick Gass, and John Ordway — offer a striking case study in how the same day could be recorded with radically different scope, accuracy, and editorial layering.
Clark’s Reconnaissance Upriver
Clark’s entry is by far the most substantial and observationally rich. After morning lunar observations and the purchase of dogs from a visiting chief, he set out at 2 o’clock with two men in a small canoe to ascend the Columbia. His prose is dense with bearings, distances, and ethnographic detail: N. 83° W 6 miles to the lower point of a Island on the Lard. Side
, passing lodges with emenc quantites of dried fish
and large Scaffols of fish drying at every lodge.
Most arresting is Clark’s puzzlement over the salmon mortality:
I observe in assending great numbers of Salmon dead on the Shores, floating on the water and in the Bottoms which can be seen at the debth of 20 feet. the Cause of the emence numbers of dead Salmon I can’t account for So it is I must have seen 3 or 400 dead and maney living
Clark was witnessing the natural post-spawning die-off of Pacific salmon — a phenomenon entirely outside his eastern experience. He notes too that local Indians retrieved fish that were not long dead
, recording an observation he made by experiment: he struck a near-dead salmon, left it floating, and watched a trailing canoe collect it. This empirical curiosity — testing rather than merely recording — is characteristic of Clark’s field practice.
He also describes a Fowl of the Pheasent kind as large as a turkey
with measurements to the quarter-inch (the sage grouse), and notes its diet of grasshoppers and the Seed of wild Isoop
(sagebrush).
Gass and the Editorial Hand
Patrick Gass’s entry, as it appears in the published 1807 edition, illustrates the heavy editorial intervention David McKeehan applied to the sergeant’s manuscript. The lengthy footnote concerning Alexander Mackenzie’s Tacoutche Tesse
and its conjectured identity with the Columbia is plainly McKeehan’s interpolation, not Gass’s field observation — Gass on October 17 had no access to Mackenzie’s published narrative or to longitude-from-London calculations.
Stripped of the footnote, Gass’s actual content overlaps with Clark’s: the dog purchases, the abundance and poor condition of salmon (very plenty but poor and dying, and therefore not fit for provisions
), and the grouse, which Gass calls heath hens or grous
and pronounces very good eating.
Gass also supplies the river measurements that would become standard expedition data: The Columbia here is 860 yards wide, and the Ki-moo-ee-nem (called Lewis’s river from its junction with the Koos-koos-ke) 475 yards.
These figures align with Clark’s surveying work and likely originate with Clark or Lewis, transmitted to Gass through camp conversation or later editorial harmonization.
Ordway’s Anomalous Entry
John Ordway’s entry for this date presents a genuine puzzle. He writes that hunters returned with deer, geese, and brants, and then states:
Cap* Lewis and party returned to Camp also, and informed us that they had been about 30 miles down which took them on the Sea Shore and a verry bad road the most of the way. they Saw the harbour where the vessells had lain but they were all gone.
This is geographically impossible for October 17, 1805. The expedition was at the Columbia–Snake confluence, hundreds of miles inland; they would not reach the Pacific until mid-November. The passage almost certainly belongs to a later date — likely in November or December 1805 — and reflects either a misdated entry, a transcription error, or a later interpolation by Ordway when fair-copying his journal. The reference to vessels having departed the harbor matches the expedition’s well-documented disappointment at finding no trading ship anchored at the river’s mouth.
The contrast among the three narrators is instructive: Clark provides the day’s primary field observation; Gass’s published text fuses authentic camp detail with later cartographic commentary imposed by his editor; and Ordway’s entry, whatever its true date, reminds researchers that fair-copy journals are not always reliable chronological records. Only by reading the three side by side does the texture of October 17 — and the limits of each source — come into focus.
